Modern Keratoplasty

Essentials in Ophthalmology
Series Editor: Arun D. Singh
Jorge L. Alió
Jorge L. Alió del Barrio Editors
Modern
Keratoplasty
Surgical Techniques and Indications
Series Editor
Arun D. Singh, Cleveland Clinic Foundation, Cole Eye Institute
Cleveland, OH, USA
Essentials in Ophthalmology aims to promote the rapid and efficient transfer
of medical research into clinical practice. It is published in four volumes per
year. Covering new developments and innovations in all fields of clinical
ophthalmology, it provides the clinician with a review and summary of recent
research and its implications for clinical practice. Each volume is focused on
a clinically relevant topic and explains how research results impact diagnostics,
treatment options and procedures as well as patient management.
The reader-friendly volumes are highly structured with core messages,
summaries, tables, diagrams and illustrations and are written by internationally
well-known experts in the field. A volume editor supervises the authors in
his/her field of expertise in order to ensure that each volume provides cutting-
edge information most relevant and useful for clinical ophthalmologists.
Contributions to the series are peer reviewed by an editorial board.
Jorge L. Alió • Jorge L. Alió del Barrio
Editors
Modern Keratoplasty
Surgical Techniques and Indications
Editors
Jorge L. Alió Jorge L. Alió del Barrio
Vissum Miranza Vissum Miranza
Miguel Hernández University Miguel Hernández University
Alicante, Spain Alicante, Spain
ISSN 1612-3212 ISSN 2196-890X (electronic)

ISBN 978-3-031-32407-9 ISBN 978-3-031-32408-6 (eBook)
https://doi.org/10.1007/978-3-031-32408-6
© The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature
Switzerland AG 2023
This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher,
whether the whole or part of the material is concerned, specifically the rights of translation,
reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any
other physical way, and transmission or information storage and retrieval, electronic adaptation,
computer software, or by similar or dissimilar methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specific statement, that such names are
exempt from the relevant protective laws and regulations and therefore free for general use.
The publisher, the authors, and the editors are safe to assume that the advice and information in
this book are believed to be true and accurate at the date of publication. Neither the publisher nor
the authors or the editors give a warranty, expressed or implied, with respect to the material
contained herein or for any errors or omissions that may have been made. The publisher remains
neutral with regard to jurisdictional claims in published maps and institutional affiliations.
This Springer imprint is published by the registered company Springer Nature Switzerland AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Preface
Over the last two decades, corneal surgery has experienced major and highly
significant changes.
In the recent past, most of the cases that required a corneal graft were the
subject of penetrating keratoplasty as the only repeatable, reliable, and effec-
tive corneal grafting surgical technique. Only in few anecdotical cases were
lamellar techniques used. Such lamellar techniques, introduced some time
ago by advanced corneal surgeons who anticipated their clinical value, were
performed with an incomplete knowledge about the surgical anatomy of the
cornea, its optics, and using inadequate instrumentation. With this back-
ground, such lamellar techniques were simply an art that was seldom prac-
ticed and only by a few talented surgeons, but with questionable clinical
outcomes.
However, the mind of the human being never stops and this indeed also
applies to corneal surgeons. The conceptualization of corneal surgical inno-
vations based on improvements in the level of knowledge of corneal anatomy
and its surgical management, the development of instruments such as femto-
second laser, and the general introduction of corneal topography, tomogra-
phy, and high-resolution anterior segment OCT have led to a completely new
perspective on corneal grafting that has been modified by the creation of
highly precise, highly refined, and sophisticated surgical techniques that have
evolved into the different corneal lamellar techniques that are in practice
today.
Parallel to this, and alongside the continuous improvement in the out-
comes of modern penetrating and lamellar corneal transplantation techniques
(DALK, DSAEK, DMEK, and others), and the impressive improvement in
eye banking techniques and more advanced knowledge of the immunological
processes that govern the local immune response at the cornea and ocular
surface, it is clear that with this alone, we are not able to treat and overcome
corneal blindness. Today, it has become evident that we need to develop new
and radically different methods to alleviate and treat corneal blindness. This
is why, slowly but surely, corneal regeneration procedures are becoming a
new promising group of techniques to restore the transparency and health
condition of the diseased ocular surface, stroma, and endothelium. Based on
this perspective, in this book we have tried to include all that is known at this
moment, based on evidence, about corneal transplantation techniques, as
practiced by modern corneal surgeons.
v
vi Preface
Corneal surgery has become today a highly specialized discipline in oph-

thalmology. It not only requires a high level of education in both clinical and
basic sciences, but also an extremely high level of surgical training in surger-
ies that are not frequent and should be carefully selected and performed to be
successful. It requires good knowledge about anatomy but also about immu-
nology and a comprehensive approach to the treatment of ocular surface dis-
eases. Globally speaking, corneal surgery today is a highly specialized,
highly technified, and highly qualified subspeciality that demands properly
trained eye surgeons to accomplish successful outcomes in the treatment of
corneal blinding diseases.
While editing this book, we have envisaged a future, in which cell biolo-
gists, pharmacologists, and bioengineers may all intervene in corneal surgery,
increasing its capability and efficiency in the treatment of corneal diseases.
Throughout the chapters of this book, we aim to integrate all this new surgi-
cal, clinical, and basic knowledge into what is today the modern practice of
keratoplasty and what is going to be developed to implement corneal surgery
in the coming years. The term Modern Keratoplasty really addresses this con-
cept and fundamentals that, as understood by the authors, corneal surgery
will further evolve in the coming years for a better treatment of corneal dis-
ease and corneal blindness.
Alicante, Spain Jorge L. Alio

Alicante, Spain Jorge L. Alio del Barrio
Preface
Sight is God’s most precious gift to mankind; cherish it, preserve it, donate it (HSD).
After cataract, corneal blindness is the commonest cause of curable sight

loss. It is not surprising therefore that attempts to replace the cornea are part
of recorded ancient medical history. The concept and the idea of rendering an
opaque cornea clear or by replacing the cornea preceded the actual procedure
by several years. Donor tissue was the first challenge to confront the early
surgeons and it seems natural that they resorted to animal donor tissue, her-
alding the early phase of xenogeneic corneal grafts, refining techniques from
animal to animal corneas before applying them for animal to human corneas.
Porcine cornea was reportedly the first tissue used in a xenograft to the human
eye followed by rabbit corneal tissue. Elegant accounts of the history and pre-
historic considerations are in the published literature by Charles McGhee,
Chad Rostrom, Gabriel Van Rij, and Satish Srinivasan [1–4].
The year 2005 was significant in that it represented the completion of a
century of successful full-thickness or penetrating keratoplasty (PK). Today
we live in an era where a new drug or procedure is introduced, developed,
practiced, and rendered obsolete, in its components or entirely, within a pro-
fessional career, more than once. Such is the pace of “progress.” The progress
of PK was not linear. The challenges, complications, and failures of PK com-
pared less favorably with lamellar keratoplasty (LK) but visual outcomes of
successful grafts were better with PK, leading to fluctuating popularity of one
over the other.
The advent of topical anesthesia (cocaine), improvements, innovations,
and inventions in instrumentation, suture material and microsurgical tech-
niques, eye banking, the operating microscope, and not least the realization of
the existence of the endothelium and its importance in the physiology of cor-
neal transparency constituted major landmarks on the path to successful
PK. Things that we today take for granted did not happen overnight. They
were the product of the cumulative research and efforts of numerous indi-
viduals, institutions, and industry.
Despite a century and a score years of success, PK was plagued by major
problems that defied resolution. Immune-mediated graft rejection and failure,
suture-related complications including severe infection, induced astigmatism
from unpredictable and irregular scarring along the circumference of the
graft-host junction, which remained forever weak and prone to rupture with
even trivial trauma, with devastating consequences, are long recognized and
vii
viii Preface
remain today as major issues. However, Hope, that translated to viable solu-
tions to some but not all of the above issues, came from revisiting old prin-
ciples with advances in knowledge and modern technology. PK gave way to
LK.
Lamellar keratoplasty is today the norm where indicated and where pos-
sible. The key principles of LK are “replacing like for like,” which applies to
all LK, and the “thinner the better” that applies primarily to endothelial kera-
toplasty (EK). The initial techniques introduced by Gerrit Melles and Francis
Price [5–7] for both anterior lamellar keratoplasty and EK were manual and
crude compared to modern techniques, but were game changing and the start
of a revolution in corneal transplantation. The near like-for-like replacement
in anterior lamellar keratoplasty came from Mohammed Anwar’s big bubble
technique for deep anterior lamellar (DALK) [8]. This was initially proposed
as a Descemet’s membrane baring technique but with the discovery of the
pre-Descemet’s layer (Dua’s layer, PDL/DL) [9] it became obvious that in
over 80% of cases of DALK by the big bubble technique the host PDL/DL
was retained (the type 1 big bubble); hence it was not a true like-for-like
replacement but nevertheless a desirable outcome as it conferred extra
strength to the eye and posed fewer intraoperative risks. Knowledge of the
PDL/DL improved understanding of DALK (type 1, type 2, and mixed bub-
bles) and made the operation safer. The cleavage plane offered by the PDL/
DL enabled successful completion of DALK even when air injection pro-
duced a very small or no big bubble. DALK by any technique (air bubble,
visco bubble, or manual dissection) confers two main advantages, a stronger
eye compared to PK and retention of the healthy host endothelium, obviating
the risk of endothelial rejection and associated graft failure [10]. The disad-
vantages of suture-related complications, induced astigmatism, and delayed
visual recovery associated with PK, however, remain. The recent innovation
of Bowman’s membrane transplant, by Melles’ group [11], represents the
thinnest anterior lamellar transplant, which both strengthens and flattens the
cornea in keratoconus eyes. A similar outcome has been demonstrated by
stromal lenticule transplantation.
Descemet’s stripping endothelial keratoplasty (DSEK) was the novel EK
procedure that heralded a paradigm shift in corneal transplantation. Manual
preparation of donor lenticules was rapidly replaced by automated keratomes,
adding Descemet’s stripping automated endothelial keratoplasty (DSAEK) to
the nomenclature. These procedures did away with the need for sutures, elimi-
nated induced astigmatism, led to rapid visual recovery, and as a bonus,
reduced the risk of endothelial rejection, probably related to reduced antigen
load (there is no epithelium and the posterior stroma has fewer keratocytes and
considerably fewer antigen-presenting dendritic cells), and reduced inflamma-
tion due to absence of sutures. However, the final thickness and contour
assumed by the transplanted lenticule (thick and more curved at the periphery)
caused a hyperopic refractive change with patients often attaining no better
than 6/12 or 6/9 vision. The realization that thinner donor lenticules apposed
better to the posterior curvature of the recipient cornea and gave better visual
outcomes was the driver to create thinner donor lenticules and the evolution of
the technique of ultra-thin (UT) DSAEK, pioneered by Massimo Busin [12].
Preface ix
Pre-Descemet’s endothelial keratoplasty (PDEK) wherein the donor trans-

plant tissue consisting of the PDL/DL, Descemet’s membrane, and endothe-
lium, obtained by creating a type 1 big bubble in the donor cornea, was
proposed and demonstrated ex vivo by Harminder Dua [9, 13], first intention-
ally performed by Amar Agarwal (others had inadvertently performed this
thinking that they were performing Descemet’s membrane endothelial kera-
toplasty [DMEK]—see below) and reported by Agarwal and Dua [14], can be
regarded as the thinnest UT-DSAEK. PDEK gives visual outcomes similar to
DMEK but has the advantage of being technically less challenging and pro-
vides the option of obtaining tissue from very young donors including infants.
Some of these innovations occurred contemporaneously rather than chron-
ologically. Maintaining his indelible stamp on EK, Melles introduced the
technique of Descemet’s membrane endothelial keratoplasty (DMEK), the
ultimate in like-for-like replacement [15]. It has all the advantages of any EK
procedure with considerable refinement on size of the entry wound and a
significantly better visual outcome. It is technically challenging both in terms
of harvesting donor Descemet’s membrane for transplantation and un-
scrolling, centration, and attachment in the recipient eye. Young donors have
thinner Descemet’s membrane which scrolls tightly, making them less suit-
able for DMEK. The support and tamponade afforded by the PDL/DL makes
use of young donor tissue a viable option for PDEK as a type 1 big bubble can
be produced in a cornea of any age. Nevertheless, DMEK is the gold standard
procedure in EK. DMEK and PDEK, unlike DSAEK and UT-DSAEK, do not
require expensive equipment (automated keratomes) with expensive mainte-
nance costs. Growing experience, education, and training will make these
procedures more accessible to surgeons globally.
This pace of progress has occurred at a price. As with any innovation that
offers a potential of better outcomes, there is a learning curve for the first
generations of surgeons who take it on and for the patients who they take on.
Success is built on some or more failures. Donor tissue wastage occurs as
harvesting techniques are refined and standardized, and failed grafts lead to
repeat grafts. Eye banks struggle to keep pace with surgical innovations and
surgeon’s demands compounded by global donor tissue shortage, augmented
by the pandemic. However, the worst is probably behind us. Eye banks have
taken on the challenge of providing pre-cut DSAEK tissue, pre-prepared
(pre-stained, pre-stamped, and pre-loaded) DMEK and PDEK tissue, easing
the pressure on surgeons.
The imminent future offers the exciting prospect of endothelial cell trans-
plantation, pioneered by Shigeru Kinoshita [16]. The introduction of cultured
endothelial cells into the anterior chamber through a paracentesis port is the
simplest of all techniques. The face-down position that the patient has to
adopt to allow the injected cells to gravitate to the back of the cornea is prob-
ably more inconvenient, but the results, as demonstrated by ongoing studies,
are as good as any. A taster of this concept is seen in the techniques of
Descemet’s stripping only (DSO) and Descemetorhexis without endothelial
keratoplasty (DWEK), wherein the guttata bearing central Descemet’s mem-
brane (in Fuchs’ endothelial keratoplasty) is peeled off and the surrounding
x Preface
endothelial cells encouraged to migrate on to the surface thus exposed, restor-

ing sight and function.
The story of corneal transplantation will not be complete without jumping
from the back to the front of the cornea and penning a few words on corneal
epithelial regeneration and limbal stem cell transplantation. Much of this
went through an equally exciting period of innovation and discovery, but is
now old hat. Auto limbal grafts, cadaver allo limbal grafts, and living related
allo limbal grafts, with their variations and subtle nuances, are well estab-
lished and practiced procedures. Ex vivo expansion of sheets of limbal, con-
junctival, and oral mucosal epithelial cells on substrates such as fibrin and
amnion is also well established and practiced. Systemic immunosuppression,
with all its implications in terms of drug toxicity and monitoring, constitutes
the biggest challenge with allografts. Autografts give excellent results, driv-
ing developments in the exploration of the use of autologous mesenchymal
stem cells from sources such as liver, bone marrow, and dental pulp and from
other sources such as the umbilical cord and embryonic tissue.
The holy grail of a synthetic and/or tissue engineered cornea, complete from
epithelium to endothelium, is running the roller coaster of promise and not living
up to the promise. To learn all about modern keratoplasty, and the past and future
of keratoplasty, Jorge Alio’s book on “Modern Keratoplasty” is a must read.
Nottingham, UK Harminder S. Dua
References
1. Crawford AZ, Patel DV, McGhee CNJ. A brief history of corneal transplantation: from
ancient to modern. Oman J Ophthalmol. 2013;6(Suppl 1):S12–7.
2. Rostron CK. The history of corneal transplantation. In: Hakim NS, Papalois VE,
editors. History of organ and cell transplantation; 2003. p. 274–92. https://doi.
org/10.1142/9781848160057_0013.
3. Gabriël van Rij G, van Dooren BTH. The history of corneal transplantation. In:
Hjortdal J, editor. Corneal transplantation. Cham: Springer International Publishing;
2016. p. 1–8. https://doi.org/10.1007/978-3-319-24052-7_1.
4. Srinivasan S. Evolution and revolution in corneal transplant surgery. J Cataract Refract
Surg. 2021;47(7):837–8.
5. Melles GR, Lander F, Rietveld FJ, Remeijer L, Beekhuis WH, Binder PS. A new
surgical technique for deep stromal, anterior lamellar keratoplasty. Br J Ophthalmol.
1999;83(3):327–33.
6. Melles GRJ. Landmark study on descemet stripping with endothelial keratoplasty:
where has it led us? J Cataract Refract Surg. 2021;47(5):561–2.
7. Price FW Jr, Price MO. Descemet’s stripping with endothelial keratoplasty in 200 eyes:
early challenges and techniques to enhance donor adherence. J Cataract Refract Surg.
2006;32(3):411–8.
8. Anwar M, Teichmann KD. Big-bubble technique to bare Descemet’s membrane in
anterior lamellar keratoplasty. J Cataract Refract Surg. 2002;28(3):398–403.
9. Dua HS, Faraj LA, Said DG, Gray T, Lowe J. Human corneal anatomy redefined: a
novel pre-Descemet’s layer (Dua’s layer). Ophthalmology. 2013;120(9):1778–85.
10. Dua HS, Freitas R, Mohammed I, Ting DSJ, Said DG. The pre-descemet’s layer (Dua’s
layer, also known as the Dua-Fine layer and the pre-posterior limiting lamina layer):
discovery, characterisation, clinical and surgical applications, and the controversy. Prog
Retin Eye Res:101161. https://doi.org/10.1016/j.preteyeres.2022.101161.
Preface xi
11. van Dijk K, Liarakos VS, Parker J, Ham L, Lie JT, Groeneveld-van Beek EA, Melles
GR. Bowman layer transplantation to reduce and stabilize progressive, advanced kera-
toconus. Ophthalmology. 2015;122(5):909–17.
12. Busin M, Madi S, Santorum P, Scorcia V, Beltz J. Ultrathin descemet’s stripping auto-
mated endothelial keratoplasty with the microkeratome double-pass technique: two-
year outcomes. Ophthalmology. 2013;120(6):1186–94.
13. Dua HS, Faraj LA, Said DG. Dua’s layer: discovery, characteristics, clinical appli-
cations, controversy and potential relevance to glaucoma. Exp Rev Ophthalmol.
2015;10:531–47.
14. Agarwal A, Dua HS, Narang P, et al. Pre-descemet’s endothelial keratoplasty (PDEK).
Br J Ophthalmol. 2014;98:1181–5.
15. Dapena I, Ham L, Droutsas K, van Dijk K, Moutsouris K, Melles GR. Learning curve
in descemet’s membrane endothelial keratoplasty: first series of 135 consecutive cases.
Ophthalmology. 2011;118(11):2147–54.
16. Numa K, Imai K, Ueno M, Kitazawa K, Tanaka H, Bush JD, Teramukai S, Okumura N,
Koizumi N, Hamuro J, Sotozono C, Kinoshita S. Five-year follow-up of first 11 patients
undergoing injection of cultured corneal endothelial cells for corneal endothelial fail-
ure. Ophthalmology. 2021;128(4):504–14.
Acknowledgments
We, the editors Jorge Alio and Jorge Alio del Barrio, wish to express our
gratitude to all the authors who have contributed to this book for their dedica-
tion and talent. Only with their expertise, experience, and generosity can an
endeavor like this be possible. Many thanks to all.
Jorge L. Alio
Jorge L. Alio del Barrio
xiii
Contents
1
Introduction: Current Status of Modern Corneal
Transplantation—Success, Failures and Turning Points�� 1
Jorge L. Alió, Dominika Wróbel-Dudzińska,
and Tomasz Żarnowski
Part I Preparing the Patient
2 Modern Eye Banking: Preservation, Type of Tissues,

and Selection�� 17
Loïc Hamon, Loay Daas, Adrien Quintin, Tarek Safi,
Isabel Weinstein, and Berthold Seitz
3 Risk Classification and Management of Corneal Grafts,
Human Leukocyte Antigen Matching, and Options
for Immunosuppression Therapy �� 41
Paula W. Feng and Guillermo Amescua
4 Management of the Vascularized Cornea Before Corneal
Graft Surgery: Fine-Needle Diathermy and Inhibition
of VEGF�� 59
Nadim S. Azar, Matias Soifer, and Victor L. Perez
Part II Penetrating Keratoplasty
5 Main Issues to Overcome in Modern Penetrating

Keratoplasty �� 71
Farideh Doroodgar, Sana Niazi, Hassan Hashemi,
and Mohammad Ali Javadi
6 Improvements in Surgical Techniques and Suturing
in Penetrating Keratoplasty�� 103
Abdo Karim Tourkmani and Colm McAlinden
7
Closing the Wound: Can Sutures Be Avoided?�� 117
Luca Menabuoni, Alessandra Balestrazzi, Luca Buzzonetti,
Romina Fasciani, Claudio Macaluso, Luigi Mosca,
Roberto Pini, Giada Magni, Paolo Matteini, Fulvio Ratto,
Michele Rossi, and Francesca Rossi
xv
xvi Contents
8 Femtosecond-Assisted
Penetrating Keratoplasty and Deep
Anterior Lamellar Keratoplasty�� 127
Alfredo Vega-Estrada and Jorge L. Alió
9 F
emtosecond Laser-Assisted Tuck-In Penetrating
Jorge L. Alió del Barrio, Olena Al-Shymali, and Jorge L. Alió
10 Rescuing
Failed Penetrating Keratoplasty Grafts�� 145
Jorge L. Alió del Barrio, Scott Robbie, Marcus Ang,
Andrea Montesel, and Jorge L. Alió
11 Excimer
Laser-Assisted Keratoplasty: Penetrating
Keratoplasty “Excimer-PKP” and Deep Anterior
Lamellar Keratoplasty “Excimer-DALK”�� 159
Loay Daas, Loïc Hamon, Elias Flockerzi, Shady Suffo,
and Berthold Seitz
Part III Epithelial Lamellar Keratoplasty
12 Keratolimbal
Grafts: Outcomes, Innovations
and Alternatives�� 169
Rafael I. Barraquer and Juan Alvarez de Toledo
13 Simple Limbal Epithelial Transplantation�� 191
Anahita Kate and Sayan Basu
14 Cultivated
Limbal Epithelial Transplantation (CLET)�� 203
Paolo Rama
15 Mesenchymal
Stem Cells for Regeneration of the Ocular
Surface�� 211
Marina López-Paniagua, Sara Galindo, Margarita Calonge,
Inmaculada Pérez, José M. Herreras, Ana de la Mata,
and Teresa Nieto-Miguel
16 Induced
Pluripotent Stem Cells in Epithelial Lamellar
Sanja Bojic, Francisco Figueiredo, and Majlinda Lako
Part IV Stromal Lamellar Keratoplasty
17 Bowman Layer Transplantation�� 245

Achraf Laouani, Lydia van der Star, Silke Oellerich,
Korine van Dijk, Gerrit R. J. Melles, and Viridiana Kocaba
18 Anterior
Lamellar Keratoplasty: Current State of the Art �� 257
Enrica Sarnicola, Caterina Sarnicola, and Vincenzo Sarnicola
19 Stromal
Lenticule Addition Keratoplasty (SLAK) �� 283
Leonardo Mastropasqua, Niccolò Salgari, Manuela Lanzini,
and Mario Nubile
Contents xvii
20
Lamellar Surgeries with SMILE Lenticules�� 297
Sri Ganesh and Sheetal Brar
21
Corneal Allogenic Intrastromal Ring Segments (CAIRS)�� 311
D. Sravani and Soosan Jacob
22 Advanced DALK Techniques: Mushroom Mini Big Bubble
DALK�� 319
Kunal A. Gadhvi and Bruce D. Allan
23
Large Diameter Deep Anterior Lamellar Keratoplasty �� 331
Angeli Christy Yu and Massimo Busin
24
Regenerative Surgery of the Corneal Stroma for Advanced
Keratoconus�� 341
Mona El Zarif, Jorge L. Alió del Barrio, and Jorge L. Alió
Part V Endothelial Keratoplasty
25
Endothelial Keratoplasty. Historical Review and Current
Outcomes�� 365
Farideh Doroodgar, Hassan Hashemi, Sana Niazi,
Sepehr Feizi, and Mohammad Ali Javadi
26
Endothelial Keratoplasty: Current State of the Art �� 381
Anjulie Gang, Francis W. Price, and Marianne O. Price
27 Corneal Endothelial Cell Transfer �� 395
Shigeru Kinoshita, Morio Ueno, and Chie Sotozono
28 Ultrathin DSAEK�� 407
29 Innovations in Descemet Membrane Endothelial
Keratoplasty (DMEK) �� 415
Darren S. J. Ting and Marcus Ang
30
Descemet Stripping Only (DSO)�� 431
Maryam Eslami and Greg Moloney
31 Pre-Descemets Endothelial Keratoplasty (PDEK):
Science and Surgery�� 437
Harminder Singh Dua
32 Pre-Descemet’s Endothelial Keratoplasty (Pdek):
Clinical Considerations and Surgical Details�� 451
Priya Narang and Amar Agarwal
33 Descemet Membrane Transplantation �� 461
Hon Shing Ong and Jodhbir S. Mehta
34
Femtosecond Laser-Assisted Deep Lamellar Endothelial
Jorge L. Alió del Barrio, Verónica Vargas, and Bruce D. Allan
xviii Contents
35 Femtosecond
Descemet Membrane Endothelial
Nir Sorkin, David S. Rootman, and Michael Mimouni
36 Cultured
Cells for Corneal Endothelial Therapy�� 485
M. P. De Miguel, M. Cadenas Martín, A. Moratilla,
and F. Arnalich-Montiel
37 What
Is New in Keratoprostheses�� 499
Saif Bani Oraba and Christopher Liu
38 Intraoperative
OCT in Anterior Segment Surgery�� 513
Francis W. Price Jr, Anjulie Gang, and Marianne O. Price
39 Epilogue:
Corneal Graft Surgery, a Glance to the Future �� 519
Jorge L. Alió and Jorge L. Alió del Barrio
Index�� 521
Introduction: Current Status
of Modern Corneal
1
Transplantation—Success, Failures
and Turning Points
Jorge L. Alió, Dominika Wróbel-Dudzińska,

and Tomasz Żarnowski
Key Points Introduction

• Corneal keratoplasty is a broad term that
includes all the numerous methods of surgery Corneal diseases are reported to be the second
to improve sight, relieve pain and treat severe leading cause of reversible blindness worldwide
infection or corneal damage. The indications [1]. The only effective treatment for this ailment
for transplantation have changed with the is a corneal graft, considered the most successful
development of surgical techniques and equip- organ transplantation in the human body [2]. The
ment and the possibility of performing not reasons for such success are mainly based on the
only full-thickness but also partial-thickness fact that the cornea is an immunologically privi-
grafts. leged tissue due to the absence of blood and lym-
• The success of modern keratoplasty should be phatic vessels, the blood-eye barrier, the presence
analyzed from an anatomical perspective of immunomodulatory factors in aqueous humor,
(graft survival) and functional results (refrac- the relative paucity of mature antigen-presenting
tive outcomes and regularity of the cornea). cells (APCs), and good intraocular penetration of
• In this chapter, we present a complete review topical steroids.
of the main complications and risk factors that The indication for corneal graft surgery as a
determine the success and failure rate of cor- solution for cornea-related visual handicaps and
neal grafts. blindness has changed over time. Haziness and
corneal scaring due to infection were the first
indication of corneal keratoplasty. Over time,
keratoconus, Fuchs endothelial dystrophy and
bullous keratopathy, graft failure after a previous
keratoplasty have now become the main indica-
tions [3].
Corneal transplant is classified as therapeutic
or tectonic (when removing an infective cornea
J. L. Alió (*)
Vissum Miranza, Miguel Hernández University, resistant to treatment, or affected by an impend-
Alicante, Spain ing or already established perforation, or severe
e-mail: [email protected] trauma with corneal tissue loss) and visual-
D. Wróbel-Dudzińska · T. Żarnowski optical (to restore corneal clarity and improve
Department of Diagnostics and Microsurgery of vision). Cosmetic indications in severely dis-
Glaucoma, Medical University of Lublin, torted blind eyes are also performed in cases
Lublin, Poland
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 1

J. L. Alió, J. L. A. del Barrio (eds.), Modern Keratoplasty, Essentials in Ophthalmology,
https://doi.org/10.1007/978-3-031-32408-6_1
2 J. L. Alió et al.
when the goal is to remove scars or corneal haze, secondary glaucoma, and a higher risk of graft
which is mainly an esthetic defect. rejection prompted ophthalmologists to search
According to research by the World Health for new techniques consisting of replacing only
Organization (WHO), corneal disease is one of the affected and damaged part of the cornea, con-
the most common causes of vision deterioration stituting the basis of lamellar keratoplasty. The
and blindness worldwide, in both adults and chil- new methods not only maintain eyeball integrity
dren [4]. The Vision Share Consortium of Eye but also provide the opportunity of avoiding com-
Banks in the USA estimates that the number of plications associated to open-sky surgery and
patients with corneal diseases exceeds ten mil- lead to a lower rate of graft rejection due to less
lion. Each year, injuries and ulceration of the cor-tissue being transplanted. This has been possible
nea cause blindness in 1.5–2 million people. due to a better understanding of the corneal anat-
However, in many cases, it is possible to trans- omy, advanced surgical techniques, instruments,
plant the cornea, the organ with the highest suc- microscopes and medications.
cess rate. Deep Anterior Lamellar Keratoplasty
(DALK) was developed to avoid the complica-
tions of PK [7, 8]. According to the literature, its
Modern Modalities of Corneal current main surgical indications are for kerato-
Transplantations conus or corneal ectasia (keratoglobus, pellucid
marginal degeneration) in 51.3 % of cases, fol-
Depending on the amount of corneal thickness lowed by herpetic or other infectious keratitis
tissue to exchange, corneal transplants can be (bacterial, fungal and parasitic) in 18.9% with
divided into: opacities located in the deep stroma, chronic
inflammation with stromal scarring and stromal
1. Penetrating, in which a fragment of the full- dystrophies (granular, lattice) [9].
thickness cornea is replaced, There are a variety of surgical techniques to
2. Lamellar, where only the affected part of the perform DALK including manual dissection,
cornea is exchanged. manual dissection supported by air [8] or visco-
elastic [10], or big bubble technique [11] or
assisted by femtosecond laser [12].
Penetrating Keratoplasty (PK) full-thickness Nowadays, we can offer customized treatment
corneal transplantation is still the gold standard consisting of focused surgeries. Furthermore, due
treatment for opaque or morphologically abnor- to advanced equipment, for example, femtosec-
mal corneas due to infections or immunological ond laser, we can perform perfect cuts that pro-
diseases and profound corneal defects due to vide more regular sealing, better and faster
stromal scarring with posterior corneal involve- healing and less irregular astigmatism, thus lead-
ment and corneal ulcerations or perforations and ing to better refractive outcomes. Consequently,
injuries [5]. lamellar procedures have replaced penetrating
keratoplasty.
According to Alio et al., the main causes for In comparison to PKP, the incidence of graft
PK are combined endothelial and stromal dis- rejection is lower (due to the absence of the endo-
eases like chronic pseudophakic bullous keratop- thelial rejection), DALK also has superior bio-
athy in 31.5% of cases, followed by keratoconus mechanical properties, may end in reduced
and ectatic cornea disorders in 11.6% and cornea induction of higher-order aberrations, visual
scarring in 10.4% [6]. rehabilitation is faster, and the risk of intra- and
The risk of expulsive hemorrhage and vitreous postoperative complications is significantly
body outflow that may happen during open-sky reduced. However, DALK has a difficult learning
surgery, prolonged visual rehabilitation, high curve and is more technically demanding. It is
astigmatism, unpredictable refractive outcome, worth bearing in mind that, sometimes irregulari-
1 Introduction: Current Status of Modern Corneal Transplantation—Success, Failures and Turning Points 3
ties in the interface or double anterior chamber neal transplants has increased in the last 30 years,
may appear. Nonetheless, the number of proce- which is undoubtedly related to the development
dures has increased over the last years. DALK of microsurgery, the use of nonabsorbable
has apparently better long-term anatomical sur- sutures, the introduction of better antibiotics and
vival rates (85–98% at 5 years) [6]. corticosteroids, wider immunology knowledge,
Posterior lamellar keratoplasty techniques as well as the introduction of tissue preservation/
aim to remove the Descemet membrane and banking procedures (previously, it was necessary
endothelium followed by their replacement from to immediately transplant tissues after donation).
the donor of either Descemet and endothelium As a result, the range of indications for corneal
alone (DMEK) or with posterior stroma remains transplants has also expanded, which was ini-
(DSAEK). The main causes for endothelial kera- tially limited to hopeless cases.
toplasty (EK) are posterior corneal dystrophies— According to the European Cornea and Cell
endothelial dystrophy including Fuchs Transplantation Registry (ECCTR), the cornea
endothelial dystrophy (FED) in 58.3% of cases, is the most frequently transplanted intact tis-
pseudophakic bullous keratopathy (PBK) in sue, with over 180,000 transplants performed
30.6% [7] and previous graft failure in 16.5% [6], annually worldwide [15]. As it was stated in the
iridocorneal endothelial syndrome (ICE) or other Eye Bank Association of America Report of
causes of endothelial dysfunction due to trauma, 2020, the total number of corneal grafts was
foreign body or age. 66,278. This figure decreased from 85,601 in
The advantages of EK are lower corneal post- comparison to the previous year. There were
operative astigmatism due to the absence of 15,402 penetrating keratoplasties (PK) per-
sutures, and suture-related problems such as infil- formed (a decrease of 11.6%), while endothelial
trates and astigmatism, and thus a faster visual keratoplasty (EK) numbers decreased by 14.9%
rehabilitation and a lower rejection rate. to 26,095. In 2019, the number of penetrating
Unfortunately, graft dislocation may happen, and keratoplasty grafts performed was 17,409 grafts
primary graft failure. Moreover, this technique and 30,650 endothelial keratoplasty grafts due to
requires special technical equipment (e.g., a a 23% increase in DMEK procedures. Regrafting
microkeratome to prepare the donor tissue). The was needed in 12.8% of cases (PK in 18.5%,
advantages of DMEK over DSAEK are faster ALK in 3.6% and EK in 9.9%) [16].
visual rehabilitation and a lower graft rejection The ECCTR estimated that about 30,000 cor-
rate due to the thinner graft (no stromal tissue), neal transplantations are performed annually in
but it is demanding for the surgeon and associ- Europe. But the total number of registered kera-
ated with a higher rate of rebubbling. However, it toplasties was 12,913 in 2019. Sixteen percent of
has been reported that endothelial keratoplasty these were regrafting procedures [15].
(EK) survival increases over time at the same As reported by the NHS transplant registry,
time as surgeon and center experience. It is worth there were 4580 corneas transplants performed
bearing in mind that a hyperopic shift of approxi- per million population in the UK from 1s April
mately 0.8–1.5 D [13] is observed in DSAEK 2019 to 31 March 2020. According to the indica-
depending on the lenticule thickness that is tion, there were 32% transplants for FED, 15.8%
transplanted. for PBK, 18.7% for regrafts, 16.9% for others,
According to recent studies, DMEK is defi- 8.9% for keratoconus and 7.8% for infectious.
nitely a better option than DSAEK in terms of PK grafts accounted for 30.3% of transplants in
visual outcomes achieved with a faster visual 2020–2021, 29.4% of the grafts were DSAEK
recovery and improved graft survival [14]. and 29.1% were DMEK and 7% were anterior
Currently, with the improvements in corneal lamellar keratoplasty [17].
transplantation surgical techniques and improve- In 2020, Canadian eye banks distributed 3786
ments in corneal banking, supplying more and corneas for surgical use, 1327 corneas for
better corneal donor material, the number of cor- DSAEK ad 937 for DMEK and 845 for
PK. Results relating to ocular tissues in 2020 neal graft procedure. We shall analyze in detail
showed reduced numbers compared to 2019, these outcome considerations later in this
with a 25% decrease in the number of overall chapter.
donors, which correlated to a 16% decrease in
ocular tissue produced and released for transplant
during 2020 [18]. Complications of Corneal
As reported by the Australian Corneal Graft Transplants That Affect
Registry 2020 Annual Report, there were 40,864 the Outcome
registered corneal grafts. Twenty-three percent of
them failed [19]. Despite the new surgical techniques and equip-
ment, better pre- and postoperative treatment and
diagnostics methods, there is still a risk of com-
uccess, Failures and Complications
S plications. They vary somewhat between the dif-
of Modern Corneal Transplants ferent types of corneal transplants, especially
intraoperative complications. In PKP and DALK,
Corneal transplantation is the most common, suture-related problems (breaking or loosening
widely practiced and most successful form of sutures, exposure, vascularization, infection,
solid-tissue transplantation in the human body immune infiltrates) and wound dehiscence may
due to the immune-privileged condition of the occur. As stated in the literature, persistent epi-
eye. thelial defects range from 0% to 77.2% in PKP
The success of modern keratoplasty tech- during active inflammation [22]. Filamentary
niques should be analyzed both from the anatom- keratitis is also common after PKP. In addition,
ical perspective (graft survival) and the functional elevated intraocular pressure, glaucoma, pupil-
result of the transplant, which involves not only lary block, Urrets-Zavalia syndrome, synechias,
how transparent the tissue grafted is but also how postoperative inflammation/infection, even endo-
regular the resulting cornea surface is and the phthalmitis, unpredictable refractive outcome
refractive outcome, which is frequently unac- and anisometropia, cataract formation may hap-
ceptable for an adequate visual rehabilitation pen in all types of corneal grafts.
leading to binocularity problems. The most common complication during
Concerning anatomical success, the overall DALK is perforation of the Descemet membrane,
first-year survival rate of all types of corneal which may occur in approximately 10–39.2% of
grafts appears to be as high as 91%. Unfortunately, cases, especially at the beginning of the learning
the long-term reality is that the overall anatomi- curve of the surgeon [23]. After surgery pseudo-
cal success rate diminishes to 63.7–83% in PKP, anterior chamber or double anterior chambers
85–98% in DALK and 79.4–96% in EK at 5 may appear. Endothelial keratoplasty is also tech-
years and only 62% of PKP grafts are functional nically demanding, and we can encounter graft
at 10 years [20]. These results are very impres- dislocation or irregular graft profile, detachment
sive in comparison to other solid-tissue transplant problems or even residual host Descemet mem-
outcomes, but are still far away from the success- brane, interface haze, potentially significant loss
ful outcomes of most of the modern ophthalmic of donor endothelial cells during the surgery and
surgical techniques such as cataract surgery, in pseudophakic eyes opacification of the hydro-
refractive surgery or retinal detachment proce- philic acrylic lens.
dures [21]. This is especially true when factors Corneal graft failure is another significant
such as the level of corneal irregularity or final problem characterized by the difficulty of resto-
refractive outcome are taken into consideration to ration of the visual function without any accom-
further qualify the functional success of a cor- panying diseases. Thus, corneal transplant
survival rates vary according to the primary cor- (b) Nonrejection causes: unclassifiable in
neal disease. any of the other two categories.
The long-term success of corneal graft • endothelial decompensation without
depends on the cause of the corneal damage, sur- rejection
gical technique used, the expertise of the surgeon, • IOP elevation/glaucoma
the possibility of organ rejection and other fac- • diseases of the ocular surface, espe-
tors that we are still unaware of. If we wish to cially limbal stem cell deficiency, dry
improve our outcomes in corneal transplantation, eye disease
we need to know the anatomical and functional • recurrence of the primary disease, her-
reasons for failures. petic disease
• wound dehiscence/hypotony and
trauma
Graft Failure Modalities 3. Morphological graft failure involves poor
visual function but with clear transparent graft
Generalizing corneal graft failure can be clas- (severe irregular astigmatism).
sified as
1. Anatomical failure: defined as an irreversible
decrease in graft transparency despite the cor- natomical Failure in the Different
A
rect treatment. Types of Corneal Grafts
2. Functional failure: consisting of unsatisfac-
tory visual acuity due to irregular cornea with Anatomical failure is defined as an outcome
astigmatism, high order abberations (HOA) or that limits corneal transparency leading to lack
inadequate refractive outcomes that limit the of functional recovery concerning the preopera-
binocularity outcome. The pre-existing ocular tive condition in terms of gains in best corrected
morbidities or complications that happen at visual acuity (BCVA). According to the litera-
the postoperative, especially glaucoma, may ture, the 5-year survival graft rate ranges from
also limit the visual outcome. 63.7 to 83% for PK, 90 to 98% for DALK and
79.4 to 96% in EK and 30 to 66% for PK
regrafts [6].
d. 1. Frequency and Main Causes
A
of Corneal Graft Anatomical Failure
Penetrating Keratoplasty
1. Primary graft failure (PGF) which happens
when the graft presents edema from the first As reported by Alio et al., the main reason for
postoperative day and transparency does not graft failure in primary PK was 16.4% PGF
recover within 3 months. It might be the result (primary graft failure), 28.2% immunological
of donor endothelial cell dysfunction, incor- rejection, 17.8% surface disease and 17.3%
rect tissue preservation or surgical trauma late endothelial decompensation. In PK
during the harvesting or transplantation regraft, 34% of the graft failure was due to
process. immunological rejection, 18.5% to ocular sur-
2. Secondary graft failure (SGF) occurs when a face disease and 17.3% to endothelial decom-
previously transparent graft becomes obscure pensation [7].
with deterioration of vision. Wiliams et al. observed graft failure in 28%
(a) Immunological rejection: when an due to allograft rejection, 20% to late endothelial
immune reaction occurs, causing graft failure, 11% due to infection, 5% glaucoma, 3%
decompensation. PGF and 34% due to others [20]. According to
Roozbahani, the most common complication detachment, 30.5% late endothelial failure and
after therapeutic PK, was cataract, which 2.1% PGF [26]. Aboshiha et al. compared dif-
appeared in 81.8% of phakic eyes, followed by ferential survival of penetrating and lamellar
47.1%, graft failure and 45.1% secondary transplants in the management of failed corneal
glaucoma 9.8% of cases developed infection,
grafts performed due to keratoconus or FED
7.84% had a persistent corneal epithelial defect and PBK; they showed that the most common
and unfortunately in almost 4% of patients evis- cause of failure of the first graft was endothelial
ceration was performed [24]. decompensation in 36.4% of cases, irreversible
rejection in 20.6% and primary graft failure in
15% [27].
Deep Anterior Lamellar Keratoplasty To sum up, in DSEK PGF ranged from 0 to
29%, graft rejection from 0–45.4%, respectively,
The review study of Williams et al. showed a to DMEK PGF varied from 0 to 12.5%, SGF
DALK PGF rate of 12%. Other causes of graft from 0 to 6% and graft rejection from 0.8 to 5%
failure were: 18% infection, 12% scaring, 5% [23].
late endothelial failure, 43% others, and 9% poor
functional performance due to astigmatism [20].
In accordance with Alio, in the DALK popula- Ad. 2
tion, the main reason for graft failure was in
37.8% of cases due to ocular surface diseases One of the main but underestimated reasons for
such as limbal stem cell deficiency, infectious functional graft failure is poor visual acuity.
keratitis, persistent epithelial defect and keratoly- The refractive outcomes of the corneal trans-
sis [7]. plant are determined by many factors such as
tissue distribution of the donor graft, suturing
technique and healing at the graft–host
Endothelial Keratoplasty junction.
According to the Spanish study, the main reason

for failure in EK was 31.9% endothelial decom- unctional Outcomes of Corneal
F
pensation without rejection in DSAEK and Graft Procedures
15.4% in DMEK, PGF was the main cause of
failure in 64.1% of failed DMEKs and 23.2% of Many times, it is the functional outcome, rather
failed DSAEKs [7]. than the anatomical one, that is the real limitation
A large study about DSEK reported that the of the outcome of anatomically successful cor-
causes of graft failure were: in 32% of cases neal grafts. The reasons are the uncontrollable
late endothelial failure, 28% PGF, 13% allograft astigmatic outcome, the induction of high levels
rejection, 3% infection and 24% others [20]. of regular and irregular astigmatism and the fre-
Researchers from USA revealed that in FED quent induction of anisometropia incompatible
with DSEK the leading cause of the graft fail- with adequate binocularity. Such refractive issues
ure was: in 29% of cases PGF, 19% late endo- are usually overlooked in most of the reports
thelial graft failure, 6% allograft rejection and about the clinical success of corneal grafts but in
43% unsatisfactory vision. In DMEK proce- real practice happen much more frequently than
dure, PGF was 66%, late endothelial failure anatomical failures in such a way that today they
20%, allograft rejection 2% and unsatisfactory constitute probably one of the main reasons for
vision 7% [25]. The main reason for failure in corneal graft failure, especially in PKP and
the Birbal study about DMEK was 67.4% graft DALK surgeries.
The functional outcome of corneal grafts tion for the surgery and procedures performed.
should be estimated including the following data: To sum up, 51–69% of corneal transplant proce-
dures should be classified as functional failures
1. Subjective: UCVA, BCVA, refractive out- due to the low visual acuity achieved.
come, refractive binocularity compared to the Unfortunately, the BCVA achieved in many
fellow eye and level of anisometropia cases is not satisfactory. As stated in the
2. Objective: Australian Corneal Graft Registry (ACGR)
(a) Ocular surface stability and tear film 2020, although visual acuity improved from
dynamics and corneal irregularities counting fingers to 6/12 by 1 year after the cor-
induce irregular astigmatism, corneal neal graft in keratoconus patients, more than
anterior and posterior higher-order aber- half of the recipients were still wearing specta-
rations (HOAs), corneal forward and cles. Fifty-four percent of patients after pene-
backward light scattering, pupil decentra- trating keratoplasty for keratoconus needed
tion or abnormalities. spectacles and 14% contact lenses. After DALK
(b) Level of corneal irregular astigmatism, 41% used glasses and 8% used contact lenses to
HOA and level of anisometropia when achieve functional visual acuity. The short-term
the refractive outcome is compared to that survival of penetrating keratoplasty was very
of the fellow eye. good, reaching 87% of cases, but unfortunately,
(c) Vision-related quality of life analysis, excellent visual acuity was only achieved by
which measures the impact that the out- approximately half of all the grafts. Moreover,
come of the surgery has induced on the the survival rate dropped to 46% at 15 years
patient’s daily life. This parameter is very postoperatively [20].
seldomly investigated and reported in The functional results were also shown in
studies about the outcomes of corneal another study. There was a significant improve-
graft procedures. ment between the pre- and postoperative values
of visual outcomes. 65.4% of patients with
All these variables lead to poor functional DALK achieved BCVA ≥20/40 and, 66.7% with
vision. Thus, there are only a few studies report- EK, 51.3% with PK. The results were better than
ing the functional outcomes of PK performed as in primary PK. DMEK showed better visual
a therapeutic or tectonic procedure and their results than DSEK [6].
visual acuity prognosis is poor. The achieved According to the tables above, the worst func-
BCVA was 20/276 in the Roozbahani study and tional results were observed in PKP and the best
20/100, according to Krysik [24, 28]. Such poor visual acuity was achieved in the DMEK proce-
visual acuity results should be classified as a dure. The leading type of corneal transplant with
complete functional failure. acceptable anatomical success was DALK, while
The tables (Tables 1.1 and 1.2) below show PKP has the highest percentage of anatomical
the distribution of BCVA according to the indica- failure.
Table 1.1 Distribution of BCVA following corneal transplant for the most common indication for 2 years follow-up
according to the European Corneal and Cell Transplantation Registry [29]
2-year postoperatively FED (%) KC (%) PBK (%) Regraft (%) Trauma (%) Infection (%)
Functional failure BCVA <20/32 10 12 44 44 55 43
Qualified failure BCVA <20/40 6 8 9 9 14 9
BCVA ≥20/40 84 80 47 47 31 48
Table 1.2 Distribution of BCVA according to the surgical procedure performed

No of eyes and BCVA 6 BCVA
Study procedure BCVA general months BCVA 1 year BCVA 5 years 10 years
Feizi [30] 45 PK ≥20/40 91.1%
54 DALK ≥20/40 83.3%
Yüksel [31] 38 DALK ≥20/40 76.3%
≥20/20 7.9%
38 PKP ≥20/40 47.4%
≥20/20 5.3%
Alzahrani [32] 16 DALK ≥20/40 81%
21 PKP ≥20/40 66%
Gadhvi [33] 338 DALK ≥20/40 75.9%
Amar [34] 31 DALK ≥20/40 58%
16 DSAEK ≥20/40
68.75%
Dickman [35] 2725 EK EK Fuchs 0.32 EK Fuchs 0.29 EK Fuchs 0.29
(log MAR) (log MAR) (log MAR)
EK PBK 0.58 EK PBK 0.55 EK PBK 0.51
2390 PKP PKP 0.47/0.77 PKP 0.39/0.74 PKP 0.32/0.70
Khattak [36] 108 DALK 20/142
(0.25 ± 0.28
log MAR)
99 PK 20/123
(0.28 ± 0.24)
Birbal [26] 500 DMEK <20/40 6% <20/40 2.2% <20/40 1.4%
≥20/40 94% ≥20/40 97.8% ≥20/40 98.6%
≥20/25 75.1% ≥20/25 80.1% ≥20/25 82.4%
≥20/20 41.1% ≥20/20 48.3% ≥20/20 53.6%
≥20/17 12.9% ≥20/17 14.7% ≥20/17 15.5%
Vassiliauskaite 100 DMEK ≥20/40 96% ≥20/40 98% ≥20/40
[37] 98%
≥20/25 81% ≥20/25 82% ≥20/25
89%
≥20/20 ≥20/20 53% ≥20/20
(1.0)-49% 64%
Peraza-Nieves 500 DMEK <20/40 6%
[38] ≥20/40 94%
≥20/25 75.1% ≥20/25 1%
≥20/20 41.1% ≥20/20 49%
≥20/17 13% ≥20/18 15%
Heinzelmann DMEK 450 ≥20/25 53%
[39] DSEK 89 ≥20/25 15%
PKP 329 ≥20/25 10%
Phillips [40] 100 DSEK 20/20 13%
100 DMEK 20/20 55%
Ham [41] 250 DMEK 20/25 73% 20/25 76.8%
20/20 44% 20/20 44%
Siggel [42] 120 DMEK ≥20/40 92% ≥20/40 88%
≥20/25 49% ≥20/25 50%
≥20/20 20% ≥20/20 22%
(continued)
Table 1.2 (continued)

No of eyes and BCVA 6 BCVA
Study procedure BCVA general months BCVA 1 year BCVA 5 years 10 years
Schlögl [43] 97 DMEK ≥20/40 90% ≥20/40 88%
≥20/25 40% ≥20/25 48%
Weller [44] 66 DMEK ≥20/40
91%
≥20/25
67%
Dunker [45] 29 DMEK ≥20/25 66%
≥20/20 24%
25 UT-DSAEK ≥20/25 32%
≥20/20 4%
Wacker [46] 52 DSEK ≥20/25 26% ≥20/25 56%
Weisenthal [47] 64 DSAEK ≥20/25 64%
≥20/20 41%
64 DMEK ≥20/25 73%
≥20/20 51%
Anatomical Outcomes After Regraft operatively. Graft survival varied between indica-
tions and techniques performed. In the 2-year
As reported by Srujana in his study, primary graft follow-up, the highest graft survival was 98% in
failure was observed in 40.6% of cases and sec- keratoconus, followed by 94% in corneal dystro-
ondary graft failure in 59.4%, mainly due to graft phies other than Fuchs endothelial dystrophy,
infection, graft rejection, secondary glaucoma 92% in FED, 82% in infectious keratitis, 82% in
and endothelial decompensation. After regrafting PBK, 82% in regraft and 80% in trauma. There
PGF was not observed, but graft rejection was no difference observed between PK and
occurred in 43.8%, and secondary glaucoma was DALK performed for keratoconus in the 2-year
present in 59.4% [23]. The 5-year graft failure for graft survival rate (98% vs. 99%). As far as FED
repeated PK ranged from 34% to 70% [48]. is concerned, the 2-year graft survival observed
was 97% in PK, 93% with DSAEK and 71% with
DMEK. Statistics for PBK were as follows; 87%
Functional Outcomes After Regraft with PK, 81% with DSAEK and 58% with
DMEK. Moreover, graft failure was observed in
The BCVA of the regrafts ranged from 20/60 to 4% of cases (1% due to endothelial decompensa-
20/600. BCVA of 20/200 or better was observed tion, less than 1% due to primary graft failure,
in 43.8% of cases, with 31.2% having a BCVA of infection, endothelial rejection, recurrence of
20/80 or better, 18.8% having a BCVA of 20/60 original disease and graft detachment) [30]. Total
and no case with a CDVA of 20/40 or better [23]. graft failure, according to ACGR, is around 22%,
Generalizing, a BCVA of 20/40 or better was with 4% early graft failure and 2% due to the rest
achieved in only 4.8–43.1% of clear PK regrafts [27]. Another study showed that despite the
by the last follow-up visit [48]. immunosuppressive regimen in high-risk corneal
transplants, graft rejection ranged from 30 to
60% and up to 70% within 10 years [21].
Survival Rate of Corneal Graft In the UK, 1 year graft survival after first-time
corneal transplants for keratoconus was 96–98%,
Following the ECCTR report, the overall graft for FED 87–90% and PBK 85–89%. Five-year
survival rate was 96% at 3 months, 95% at 6 graft survival for KCN 90–94%, FED 77–82%,
months, 93% at 1 year and 89% at 2 years post- PBK 58–66%. The survival rate of the first-time
Table 1.3 Five-year anatomical and functional successment, the outcomes of corneal graft surgery still
rate [4–6, 10, 20, 24, 35]
constitute a problem taking into consideration the
Type of corneal Anatomical Functional data presented in this chapter. If anatomical suc-
transplant/ outcome (graft outcome (BCVA
outcomes survival rate) (%) ≥20/40) (%)
cess is further completed with functional out-
DALK 85 65.4 come success, we must conclude that modern
PK 63.7–83 51.3 keratoplasty techniques have a wide scope and
DSAEK 79.4–96 68.75 potential to be improved by the control of immu-
DMEK 90–96 90–97.8 nity, ocular surface, and, nowadays, more evi-
dent, the functional outcome. Overall, we can
conclude from the most recently available litera-
graft in low-risk patients was around 90% at 5 ture that the success rate expectation of PKP has
years [16]. Unfortunately, a reduction in success a mean of 63.7–83% anatomical and 51.3% func-
rates is observed over time [49]. tional success, DALK 85% anatomical and
According to the Singapore Corneal Transplant 65.4% functional success, DSAEK 79.4–96%
Study (SCTS), the overall corneal graft survival anatomical and 68.75% functional success, and
rate at 1 year was 91%, at 5 years 66.8%, at 10 DMEK 90–96% anatomical and 90–97.8% func-
years 55.4%, at 15 years 52%, and 20 years 44%. tional success [5, 6, 10, 20, 24, 35].
Graft survival decreased over time from 91% at 1 Therefore, we need to optimize the anatomical
year to 44% at 20 years’ follow-up. Allograft and functional outcomes of modern corneal
rejection and late endothelial failure accounted transplantation techniques, as presented in this
for more than 60% of graft failures [50]. Similar chapter and which are related to many factors.
results were reported by the Australian Corneal Better strategies should be developed in cases
Graft Registry (ACGR). After 15 years corneal of immunological rejection, ocular surface
graft survival rates had dropped to 46% for full- inflammation and glaucoma. In addition, it is
thickness grafts and 41% for lamellar grafts [21]. worth pointing out that only 1.5% of the world-
Table 1.3 summarizes the 5-year anatomical wide demand for corneal transplants is currently
and functional success rates of different types of fulfilled. According to another source, it has been
corneal grafts. estimated that there is only 1 corneal donor avail-
The above-mentioned results are quite good in able for every 70 needed [53]. Moreover, the
comparison to other solid transplants, but we COVID-19 pandemic continues to significantly
need to bear in mind that good anatomical results impact the reduction in tissue donation and graft
and long graft survival do not guarantee good production. The gap between organ demand and
functional outcomes, which affect the patient's supply is a huge universal problem in transplan-
quality of life. In the past, the most important aim tation. Likewise, the problem is growing in spite
in corneal transplantation was graft survival. of efforts made in medical, educational, and
Nowadays, patients are more demanding; thus, social fields and mass media support. This reality
refractive outcomes such as functional/good has created the need for completely new thera-
quality of vision are now highly important. Thus, peutic alternatives for the management of end-
there is still much to be done on this topic. stage organ disease. Moreover, bearing in mind
the fact that the functional outcomes of corneal
transplants are not satisfactory, new research
Conclusion: Real Success Rate should continue in the future to aim at discover-
of Modern Corneal Transplantation ing systems and devices capable of totally replac-
Procedures ing the traditional transplantation procedures or
alternative methods that are less dependent on the
To conclude, despite the advanced surgical tech- availability of allogeneic tissue and new solu-
niques, innovative equipment, the surgeon’s tions based on modern approaches such as
skills, and proper pre- and postoperative treat- advanced therapies and stem cell corneal regen-
eration, bioengineered artificial corneas, the 8. Archila EA. Deep lamellar keratoplasty dissection
of host tissue with intrastromal air injection. Cornea.
development of natural corneal replacements and 1984;5(3):217–8.
biosynthetic matrices for host tissue 9. Gómez-Benlloch A, Montesel A, Pareja-Aricò L,
regeneration. Mingo-Botín D, Michael R, Barraquer RI, Alió
J. Causes of corneal transplant failure: a multicentric
study. Acta Ophthalmol. 2021;99(6):e922–8. https://
Take Home Notes doi.org/10.1111/aos.14708. Epub 2021 Jan 9.
• In spite of corneal grafting being one of the 10. Melles GR, Rietveld FJ, Beekhuis WH, Binder
most frequent and successful organ transplan- PS. A technique to visualize corneal incision and
tations in the human body, there is still much lamellar dissection depth during surgery. Cornea.
1999;18:80–6.
to be done to improve the outcomes. 11. Anwar M, Teichmann KD. Big-bubble technique to
• Anatomical and functional results vary in dif- bare Descemet’s membrane in anterior lamellar kera-
ferent types of cornea graft procedures, and toplasty. J Cataract Refract Surg. 2002;28:398–403.
unfortunately, in many cases, they are far 12. Alio JL, Abdelghany AA, Barraquer R, Hammouda
LM, Sabry AM. Femtosecond laser assisted deep
below patients’ expectations and significantly anterior lamellar keratoplasty outcomes and heal-
affect the quality of life. ing patterns compared to manual technique.
• Bearing in mind moderate outcomes, lack of BioMed Res Int. 2015;2015:397891. https://doi.
the tissue and highly trained surgeons to per- org/10.1155/2015/397891, 6 pages.
13. Dupps WJ Jr, Qian Y, Meisler DM. Multivariate
form this difficult and costly procedure, there model of refractive shift in Descemet-stripping auto-
is a need to develop new techniques and/or mated endothelial keratoplasty. J Cataract Refract
novel therapeutic approaches. Surg. 2008;34:578–84.
14. Singh A, Zarei-Ghanavati M, Avadhanam V, Liu
C. Systematic review and meta-analysis of clinical
outcomes of Descemet membrane endothelial kera-
Conflict of Interest None of the authors have any con- toplasty versus Descemet stripping endothelial kera-
flict of interest to disclose. toplasty/Descemet stripping automated endothelial
keratoplasty. Cornea. 2017;36(11):1437–43. https://
doi.org/10.1097/ICO.0000000000001320.
15. https://www.google.com/url?sa=t&rct=j&q=&esr
c=s&source=web&cd=&cad=rja&uact=8&ved=2
References ahUKEwj1q8D-w sv1AhXhkYsKHbpiBDkQFnoE
CBQQAQ&url=https%3A%2F%2Fec.europa.eu%
1. Oliva MS, Schottman T, Gulati M. Turning the 2Fresearch%2Fparticipants%2Fdocuments%2Fd
tide of corneal blindness. Indian J Ophthalmol. ownloadPublic%3FdocumentIds%3D080166e5c9
2012;60:423–7. dc9246%26appId%3DPPGMS&usg=AOvVaw1m
2. Singh R, Gupta N, Vanathi M, Tandon R. Corneal Xgu_EptMyZ-HN0ELt9vU.
transplantation in the modern era. Indian J Med 16. Eye banking statistical report 2019. Eye Bank
Res. 2019;150(1):7–22. https://doi.org/10.4103/ijmr. Association of America. https://restoresight.org/
IJMR_141_19. what-we-do/publications/statistical-report.
3. Trevor Roper PD. The history of corneal grafting. 17. NHS. Survival rates following transplantation. [Survival
In: Casey TA, editor. Corneal grafting. London: rates following transplantation.] 2020. https://www.
Butterwork; 1972. p. 1–5. organdonation.nhs.uk/helping-you-to-decide/about-
4. Pineda R. World corneal blindness. In: Colby organ-d onation/statistics-a bout-o rgan-d onation/
K, Dana R, editors. Foundations of corneal transplant-activity-report/.
disease. Cham: Springer; 2020. https://doi. 18. h t t p s : / / p r o f e s s i o n a l e d u c a t i o n . b l o o d .
org/10.1007/978-3-030-25335-6_25. c a / e n / o r g a n s -a n d -t i s s u e s / r e p o r t s /
5. Patel SV. Graft survival after penetrating keratoplasty. e y e -a n d -t i s s u e -r e p o r t s -a n d -s u r v e y s /
Am J Ophthalmol. 2011;151(3):397–8. https://doi. report-canadian-eye-and-tissue-data-1.
org/10.1016/j.ajo.2010.10.006. 19. https://www.flinders.edu.au/fhmri-e ye-v ision/
6. Montesel A, Alió Del Barrio JL, Yébana Rubio P, corneal-graft-registry.
Alió JL. Corneal graft surgery: a monocentric long- 20. Williams KA, Keane MC, Coffey NE, et al. The
term analysis. Eur J Ophthalmol. 2021;31(4):1700–8. Australian corneal graft registry 2018 report. Flinders
https://doi.org/10.1177/1120672120947592. Epub Unviersity. https://dspace.flinders.edu.au/xmlui/han-
2020 Aug 5. dle/2328/37917. Published 2018.
7. Castroviejo R. Keratoplasty in treatment of keratoco- 21. Lundström M, Manning S, Barry P, et al. The
nus. Arch Ophthalmol. 1949;42(6):776–800. European registry of quality outcomes for cataract and
refractive surgery (EUREQUO): a database study of Int J Ophthalmol. 2017;10(3):385–90. https://doi.

trends in volumes, surgical techniques and outcomes org/10.18240/ijo.2017.03.10. Published 2017 Mar 18.
of refractive surgery. Eye Vis. 2015;2:8. https://doi. 32. Alzahrani K, Dardin SF, Carley F, Brahma A, Morley
org/10.1186/s40662-015-0019-1. D, Hillarby MC. Corneal clarity measurements in
22. Wan S, Cheng J, Dong Y, Xie L. Epithelial defects patients with keratoconus undergoing either pen-
after penetrating keratoplasty in infectious keratitis: etrating or deep anterior lamellar keratoplasty. Clin
an analysis of characteristics and risk factors. PLoS Ophthalmol. 2018;12:577–85.
One. 2018;13(11):e0208163. https://doi.org/10.1371/ 33. Gadhvi KA, Romano V, Fernández-Vega Cueto L,
journal.pone.0208163. Published 2018 Nov 28. Aiello F, Day AC, Allan BD. Deep anterior lamellar
23. Alió Del Barrio JL, Bhogal M, Ang M, Ziaei M, keratoplasty for keratoconus: multisurgeon results.
Robbie S, Montesel A, Gore DM, Mehta JS, Alió Am J Ophthalmol. 2019;201:54–62. https://doi.
JL. Corneal transplantation after failed grafts: options org/10.1016/j.ajo.2019.01.022. Epub 2019 Feb 2.
and outcomes. Surv Ophthalmol. 2021;66(1):20–40. 34. Amar SP, Sinha R, Kalra N, Agarwal T, Sharma N,
https://doi.org/10.1016/j.survophthal.2020.10.003. Titiyal JS. Demographic and clinical profile, sur-
Epub 2020 Oct 14. gical outcome, and quality of life in patients who
24. Roozbahani M, Hammersmith KM, Nagra PK, Ma underwent bilateral lamellar corneal grafts. Indian
JF, Rapuano CJ. Therapeutic penetrating kerato- J Ophthalmol. 2021;69(7):1747–52. https://doi.
plasty: a retrospective review. Eye Contact Lens. org/10.4103/ijo.IJO_3194_20. PMID: 34146020;
2018;44(Suppl 2):S433–41. https://doi.org/10.1097/ PMCID: PMC8374770.
ICL.0000000000000522. 35. Dickman MM, Peeters JM, van den Biggelaar FJ,
25. Price DA, Kelley M, Price FW Jr, Price MO. Five- Ambergen TA, van Dongen MC, Kruit PJ, Nuijts
year graft survival of Descemet membrane endothe- RM. Changing practice patterns and long-term
lial keratoplasty (EK) versus Descemet stripping EK outcomes of endothelial versus penetrating kera-
and the effect of donor sex matching. Ophthalmology. toplasty: a Prospective Dutch Registry Study.
2018;125(10):1508–14. https://doi.org/10.1016/j. Am J Ophthalmol. 2016;170:133–42. https://doi.
ophtha.2018.03.050. Epub 2018 May 3. org/10.1016/j.ajo.2016.07.024. Epub 2016 Aug 4.
26. Birbal RS, Ni Dhubhghaill S, Bourgonje VJA, et al. 36. Khattak A, Nakhli FR, Al-Arfaj KM, Cheema
Five-year graft survival and clinical outcomes of 500 AA. Comparison of outcomes and complications of
consecutive cases after Descemet membrane endothe- deep anterior lamellar keratoplasty and penetrating
lial keratoplasty. Cornea. 2020;39(3):290–7. keratoplasty performed in a large group of patients
27. Aboshiha J, Jones MNA, Hopkinson CL, Larkin with keratoconus. Int Ophthalmol. 2018;38(3):985–
DFP. Differential survival of penetrating and lamel- 92. https://doi.org/10.1007/s10792-017-0548-9. Epub
lar transplants in management of failed corneal grafts. 2017 May 22.
JAMA Ophthalmol. 2018;136(8):859–65. https:// 37. Vasiliauskaitė I, Oellerich S, Ham L, Dapena I,
doi.org/10.1001/jamaophthalmol.2018.1515. PMID: Baydoun L, van Dijk K, Melles GRJ. Descemet
29931227; PMCID: PMC6142952. membrane endothelial keratoplasty: ten-year graft
28. Krysik K, Wroblewska-Czajka E, Lyssek-Boron A, survival and clinical outcomes. Am J Ophthalmol.
Wylegala EA, Dobrowolski D. Total penetrating kera- 2020;217:114–20. https://doi.org/10.1016/j.
toplasty: indications, therapeutic approach, and long- ajo.2020.04.005. Epub 2020 Apr 10.
term follow-up. J Ophthalmol. 2018;2018:9580292. 38. Peraza-Nieves J, Baydoun L, Dapena I, Ilyas A, Frank
https://doi.org/10.1155/2018/9580292. PMID: LE, Luceri S, Ham L, Oellerich S, Melles GRJ. Two-
29850220; PMCID: PMC5933013. year clinical outcome of 500 consecutive cases under-
29. Dunker SL, Armitage WJ, Armitage M, Brocato L, going Descemet membrane endothelial keratoplasty.
Figueiredo FC, Heemskerk MBA, Hjortdal J, Jones Cornea. 2017;36(6):655–60. https://doi.org/10.1097/
GLA, Konijn C, Nuijts RMMA, Lundström M, ICO.0000000000001176.
Dickman MM. Outcomes of corneal transplanta- 39. Heinzelmann S, Böhringer D, Eberwein P,
tion in Europe: report by the European Cornea and Reinhard T, Maier P. Outcomes of Descemet mem-
Cell Transplantation Registry. J Cataract Refract brane endothelial keratoplasty, Descemet stripping
Surg. 2021;47(6):780–5. https://doi.org/10.1097/j. automated endothelial keratoplasty and penetrating
jcrs.0000000000000520. keratoplasty from a single centre study. Graefes
30. Feizi S, Javadi MA, Karimian F, Abolhosseini M, Arch Clin Exp Ophthalmol. 2016;254(3):515–22.
Moshtaghion SM, Naderi A, Esfandiari H. Penetrating https://doi.org/10.1007/s00417-015-3248-z. Epub
keratoplasty versus deep anterior lamellar kerato- 2016 Jan 7.
plasty in children and adolescents with keratoco- 40. Phillips PM, Phillips LJ, Muthappan V, Maloney CM,
nus. Am J Ophthalmol. 2021;226:13–21. https://doi. Carver CN. Experienced DSAEK surgeon’s transi-
org/10.1016/j.ajo.2021.01.010. Epub 2021 Jan 30. tion to DMEK: outcomes comparing the last 100
31. Yüksel B, Kandemir B, Uzunel UD, Çelik O, Ceylan DSAEK surgeries with the first 100 DMEK surgeries
S, Küsbeci T. Comparison of visual and topo- exclusively using previously published techniques.
graphic outcomes of deep-anterior lamellar kerato- Cornea. 2017;36(3):275–9. https://doi.org/10.1097/
plasty and penetrating keratoplasty in keratoconus. ICO.0000000000001069.
41. Ham L, Dapena I, Liarakos VS, et al. Midterm results 47. Weisenthal RW, Yin HY, Jarstad AR, Wang D, Verdier
of Descemet membrane endothelial keratoplasty: DD. Long-term outcomes in fellow eyes comparing
4 to 7 years clinical outcome. Am J Ophthalmol. DSAEK and DMEK for treatment of Fuchs corneal
2016;171:113–21. dystrophy. Am J Ophthalmol. 2022;233:216–26.
42. Siggel R, Adler W, Stanzel TP, Cursiefen C, Heindl https://doi.org/10.1016/j.ajo.2021.06.013. Epub 2021
LM. Bilateral Descemet membrane endothelial kera- Jun 19.
toplasty: analysis of clinical outcome in first and 48. Armitage WJ, Goodchild C, Griffin MD, Gunn DJ,
fellow eye. Cornea. 2016;35(6):772–7. https://doi. Hjortdal J, Lohan P, et al. High-risk corneal trans-
org/10.1097/ICO.0000000000000811. plantation: recent developments and future possibili-
43. Schlögl A, Tourtas T, Kruse FE, Weller JM. Long-term ties. Transplantation. 2019;103:2468–78. https://doi.
clinical outcome after Descemet membrane endothe- org/10.1097/TP.0000000000002938.
lial keratoplasty. Am J Ophthalmol. 2016;169:218– 49. Anshu A, Li L, Htoon HM, de Benito-Llopis L,
26. https://doi.org/10.1016/j.ajo.2016.07.002. Epub Shuang LS, Singh MJ, et al. Long-term review of
2016 Jul 15. penetrating keratoplasty: a 20-year review in Asian
44. Weller JM, Kruse FE, Tourtas T. Descemet mem- eyes. Am J Ophthalmol. 2020;224:254–66. https://
brane endothelial keratoplasty: analysis of clini- doi.org/10.1016/j.ajo.2020.10.014.
cal outcomes of patients with 8-10 years follow-up. 50. Gain P, Jullienne R, He Z, Aldossary M, Acquart
Int Ophthalmol. 2022;42(6):1789–98. https://doi. S, Cognasse F, Thuret G. Global survey of corneal
org/10.1007/s10792-021-02176-3. transplantation and eye banking. JAMA Ophthalmol.
45. Dunker SL, Dickman MM, Wisse RPL, Nobacht S, 2016;134(2):167–73. https://doi.org/10.1001/
Wijdh RHJ, Bartels MC, Tang ML, van den Biggelaar jamaophthalmol.2015.4776.
FJHM, Kruit PJ, Nuijts RMMA. Descemet membrane
endothelial keratoplasty versus ultrathin Descemet
stripping automated endothelial keratoplasty: a
multicenter randomized controlled clinical trial. Suggested Reading
Ophthalmology. 2020;127(9):1152–9. https://doi.
org/10.1016/j.ophtha.2020.02.029. Epub 2020 Mar 2. Srujana D, Kaur M, Urkude J, Rathi A, Sharma N, Titiyal
46. Wacker K, Baratz KH, Maguire LJ, McLaren JW, JS. Long-term functional and anatomic outcomes of
Patel SV. Descemet stripping endothelial kerato- repeat graft after optically failed therapeutic kerato-
plasty for Fuchs’ endothelial corneal dystrophy: five- plasty. Am J Ophthalmol. 2018;189:166–75. https://
year results of a prospective study. Ophthalmology. doi.org/10.1016/j.ajo.2018.03.011. Epub 2018 Mar
2016;123(1):154–60. https://doi.org/10.1016/j.oph- 14.
tha.2015.09.023. Epub 2015 Oct 17.
Part I
Preparing the Patient
Modern Eye Banking: Preservation,
Type of Tissues, and Selection
2
Loïc Hamon, Loay Daas, Adrien Quintin, Tarek Safi,
Isabel Weinstein, and Berthold Seitz

• In present times, the role of eye banks has
evolved beyond simply storing corneas. Since the first penetrating keratoplasty (PKP)
• Continuous improvement of preservation was performed by Eduard Zirm in 1905 [1], the
techniques extends storage time and provides number of keratoplasties has inexorably increased
better tissue quality. worldwide [2]. According to the German kerato-
• The increasing number of keratoplasty tech- plasty register [3], 8912 keratoplasties were per-
niques and the high demand for “ready-to- formed in 2020 in Germany (Fig. 2.1). In order to
use” tissue is challenging eye banks to improve meet this need, it is necessary to work with spe-
and develop new preparation techniques. cialized and organized structures more than ever.
• Besides necessary examinations, new Eye banks fulfill this role.
approaches to tissue analysis in eye banks— In the past, the role of eye banks was limited
such as the sterile donor tomography—allow a to collection, storage and evaluation of the tissues
better selection of corneal tissues. prior to transplantation. The technique of kerato-
• These new challenges in tissue preservation, plasties has constantly evolved over the years
preparation and selection are propelling eye with the purpose to minimize immune reactions
banks into the era of “modern eye banking.”. [4] and to reduce postoperative astigmatism [5].
These improvements include new trephination
techniques, for example, the femtosecond [6] and
the excimer laser-assisted trephination [7], new
L. Hamon (*) · L. Daas · I. Weinstein · B. Seitz suture techniques, for example, the double run-
Department of Ophthalmology, Saarland University ning cross-stitch suture [8, 9] and the develop-
Medical Center (UKS), Homburg, Germany
ment of new lamellar techniques, for example,
Klaus Faber Center for Corneal Diseases, Saarland the Descemet Membrane Endothelial
University Medical Center, Homburg, Germany
Keratoplasty (DMEK) [10, 11] or the Deep
LIONS Eye Bank Saar-Lor-Lux, Trier/Westpfalz, Anterior Lamellar Keratoplasty (DALK) [12].
Saarland University Medical Center,
Homburg, Germany Eye banks have followed this evolution toward
e-mail: [email protected]; [email protected]; modern keratoplasty, extending their role to the
[email protected]; [email protected] (pre-)selection and preparation of tissues for
A. Quintin · T. Safi surgeons, entering a new Era of modern eye
Department of Ophthalmology, Saarland University banking.
Medical Center (UKS), Homburg, Germany
e-mail: [email protected]; [email protected]

https://doi.org/10.1007/978-3-031-32408-6_2
18 L. Hamon et al.
10000
9000
8000
7000
5602 5923
6000 5829
4712
4175
5000 3629
1727
3031
612 1123
2219
4000
178 269 234 240 277
241 212 232
3000 231 231 272
2000
3240 3141 3123 2979 2944 3100 3273 3042 2811
2733 2721
1000
0
2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020
Penetrating keratoplasty (PKP) (Deep) anterior lamellar keratoplasty (DALK) Posterior lamellar keratoplasty (PLK)
Fig. 2.1 Keratoplasties in Germany. Until 2013 domi- 65.4% of 8912 keratoplasties in 2020. (Source: The
nated by penetrating keratoplasties (PKP), posterior German Keratoplasty Registry listed in Homburg/Saar
lamellar keratoplasties and particularly DMEK has expe- since 2006)
rienced a substantial expansion in Germany, representing
onor’s Screening and Tissue

D diseases [15]. In order to minimize this risk,
Collection careful donor selection is necessary. Regulatory
agencies such as the European Eye Bank
Corneal transplantation safety is widely depen- Organization (EEBA) and the Eye Bank
dent upon clinical donor selection. The risk of Association of America (EBAA) are continu-
donor-to-host disease transmission through cor- ously working to minimize this risk with well-
neas has been a major concern since the begin- documented and broadly accepted regulations in
ning of keratoplasty. Regarding the first known the field. A list of contraindications for corneal
case of donor-to-host retinoblastoma transmis- donation (Table 2.1) and new testing to exclude
sion in 1939 [13] or the well-established cases potentially transmissible diseases are regularly
of Creutzfeldt-Jakob and rabies transmissions in adapted and updated. However, these tests have
the 1970s [14], with lethal consequences for the a cost (financial, but also false positivity that
recipients, many diseases have been proven or could discard transplantable corneas) and the
assumed to be transmissible from donor to decision to implement or exclude new tests must
recipient via corneal transplantation. The epi- be based on scientific evidence [17].
demic of acquired immune deficiency syndrome If the donor has no contraindications, the eye
(AIDS) in the 1980s, as late stage of the human bank staff can proceed with the collection of the
immunodeficiency virus (HIV), brought a major donor corneas [18]. There are, in principle, two
change in general awareness of transmissible ways to collect donor corneas:
2 Modern Eye Banking: Preservation, Type of Tissues, and Selection 19
Table 2.1 Contraindications for corneal donation [16] 1. Whole globe collection is a simple and quick
Unknown cause of death method of collecting donor tissue. The collec-
History of a disease of unknown etiology tion of the “whole eye globe” can be performed
Donors with following malignant diseases:
in the morgue, in a refrigerated room or at the
• Retinoblastoma
donor’s bed, depending on the possible col-
• Hematologic neoplasm (leukemia, lymphoma,
myeloma) lection process. The globe is brought forward
• Malignant tumor of the anterior segment of the eye (pulled) with an instrument and the optic
(such as conjunctival intraepithelial neoplasia, nerve is cut with scissors. It must then be pre-
squamous cell carcinoma, malignant melanoma, pared at the eye bank for the surgery. An
metastasis)
Risk of transmission caused by prions:
advantage of this method is that it saves time
• Creutzfeldt-Jakob disease or variant Creutzfeldt- during the collection process. Another advan-
Jakob disease tage is the removal—at the same time—of
• History of rapid progressive dementia or scleral tissue allowing other types of surgery
degenerative neurological disease such as such as sclerocorneoplasty [19] or scleral
Alzheimer’s disease, multiple sclerosis,
amyotrophic lateral sclerosis patches [20]. The symbolism of the removal
• Recipients of hormone-derived from human of the eye (as a whole “organ”) is unfortu-
pituitary gland (before 1987) nately perceived as “intrusive” by donor’s
• Recipient of graft of cornea, dura matter or sclera relatives, which may lead to a higher rate of
Uncontrolled systemic infection: donation refusal.
• Bacterial disease
2. Corneoscleral explantation (15 mm disc) is
• Systemic viral diseases such as rabies
• Fungal or parasitic infection
a more expensive alternative regarding time
Exception for septicemia, if the cornea is stored in and resources. It is performed in the same way
organ culture media to allow detection of any bacterial as an ophthalmic surgical procedure, in sterile
contamination of the tissue conditions and using surgical equipment
History or clinical evidence of human (Fig. 2.2). After conjunctival dissection, and
immunodeficiency virus (HIV) or acquired
immunodeficiency syndrome (AIDS) the corneoscleral disc is removed using a
Acute or chronic hepatitis B, hepatitis C and/or human 15 mm diameter round hand-trephine. An
T-lymphotropic virus (HTLV) I/II exactly concentric trephination is important in
History of chronic systemic autoimmune and/or order to guarantee that the graft can be opti-
inflammatory disease that could impact the quality of mally processed later on. After corneal collec-
collected tissue
Donors from or have traveled in zones with high-risk
tion, plastic shells are applied for aesthetic
for epidemical diseases (in accordance with the reconstruction and the eyelids are tightly
European Centre of Disease Control): closed. The deceased will not show any aes-
• Zika virus thetic restrictions. This method of procure-
• New coronavirus (2019-nCoV) ment, although more complex, has a much
• Ebola virus
higher acceptance rate by the donors’ rela-
Ingestion or exposure to substances such as cyanide,
lead, mercury, gold tives [22, 23]. The corneoscleral disc is also
Recent history of vaccination with a live attenuated already cut, avoiding further manipulations in
virus the eye bank prior to preservation.
Transplantation with xenografts
Behaviors or presence on the donor’s body of physical After collection, the whole globes or corneas
signs implying a risk of transmissible disease(s) such as
bruises, lacerations, etc. are taken to the eye bank, where they are preserved,
examined and prepared for surgical purposes.
20 L. Hamon et al.
a b c
Fig. 2.2 Cornea procurement—corneoscleral technique. in order to guarantee that the graft can be optimally pro-
The material for corneal removal is prepared in a sterile cessed later on. (c) After trephination, the corneoscleral
manner. Eye bank staff need to wash and sterilize them- disc is placed in a container with organ culture medium II
selves like for intraocular surgery. (a) Povidone-iodine (also called transport medium). (d) A plastic disc is
1.25% is applied slowly 5 min before trephination, in inserted in the place of the trephined corneoscleral disc.
order to minimize potential contamination [21]. (b) The eyelids are closed and glued or sutured. Thus, there is
Trephination is performed with a 15 mm diameter round no evidence of the procedure for the relatives
trephine. An exactly concentric trephination is important
Preservation Techniques Cryopreservation
Moist Chamber Preservation Cryopreservation has long been a goal in the

field of eye banking, promising a very long
Introduced in 1935 [24], the conservation of preservation period and a lower risk for tissue
whole globes in moist chambers directly after infections. The first protocols for cryopreserva-
enucleation was the only preservation technique tion were developed in the 1960s [27, 28]. In
until the 1960s. The explanted globes were stored 1981, Sperling [29] also developed a technique
in an airtight container. The air was humidified for cryopreservation of corneas kept under
through a wet gauze placed in the container. organ culture conditions, then frozen.
Containers were stored at low temperatures Nevertheless, despite all attempts, cryopreser-
(4–8°C) or alternatively put on ice [25] until sur- vation protocols for human corneas were not
gery. This uncomplicated method allowed the able to provide tissues with the endothelial
preservation of the globe for up to 2–3 days [26]. qualities required to ensure a transplantable
However, the sterility of the tissue could not be cornea [30, 31]. To date, donor corneas cannot
ensured and the restricted storage period did not reliably be frozen.
leave enough time for sufficient quality control With the beginning of the worldwide
by the eye banks. Nowadays, this technique has COVID- 19 pandemic in 2019 [32], some eye
been replaced by more advanced alternatives but banks have experienced difficulties in finding
still remains used in certain developing countries viable donors. This disturbance in the constancy
where the implementation of more expensive of cornea procurement has led to a renewed inter-
methods is problematic. est in cryopreservation techniques [33] and new
protocols may emerge. Nevertheless, the absence Organ Culture Preservation

of active viral structural proteins in SARS-CoV-2
positive donor corneas [34] and the negligible The organ culture technique aims at the long-
risk of transmission after proper donor selection term preservation of the isolated human cornea
[35] should allow a return to normal donation in under simulated physiological conditions.
the very near future. Summerlin et al. first investigated this storage
method in vitro in 1973 [43]. In 1976, this method
was adapted by Doughman et al. to be used for
Hypothermic Storage eye banking [44]. Since Sperling popularized this
preservation method in Europe, corneal graft
This storage technique at 2–8 °C was developed preservation is mostly done by organ culture on
by McCarey and Kaufman in 1974 [36]. The this side of the Atlantic [45]. This technique was
original McCarey–Kaufman (M–K) medium also (re)introduced in the USA as “Minnesota
consisted of tissue culture medium TC-199, dex- system corneal preservation” [46] but has not
tran (an osmotic agent preventing corneal swell- supplanted hypothermic storage in North
ing), bicarbonate and antibiotics (penicillin/ America.
streptomycin) and claimed a storage period of up The corneas are suspended in a cell culture
to 10 days. Other solutions introduced later, such container filled with an organ culture medium
as the modified M–K medium [37], K-sol [38] or (modified Minimal Essential Medium—MEM,
the very popular Optisol (GS) [39], claimed bet- so-called medium I) and supplemented with fetal
ter storage capabilities and an extended storage or newborn calf serum (2–10%), antibiotics and
period of 14–16 days [40]. antimycotics (Fig. 2.3). Cell culture containers
Hypothermic storage is by far the most are stored at 30–37 °C for a maximum recom-
widely used preservation technique for corneo- mended period of 28 days, with medium renewal
scleral discs in North America [41]. Donor cor- after 7–14 days, depending on the organ culture
neoscleral discs are stored in flat cylindrical medium composition. Some eye banks have
containers, allowing morphological inspection investigated storage periods of up to 48 days
with the slit-lamp and endothelium inspection [47], but this practice is largely not recommended
by specular microscopy, both under sterile con- [2]. At the beginning of the culture period, cor-
ditions if using special fixation devices. As the neas are kept in quarantine while microbiological
appearance of the endothelial cells vary with testing is carried out in accordance with current
temperature, an examination at room tempera- standards [48, 49]. The medium I being isotonic,
ture is recommended. Degenerative changes are the organ-cultured cornea swells to twice its nor-
not rare during hypothermic storage and may mal thickness during storage [50]. Before pene-
lead to severe endothelial cell loss. The low trating keratoplasty can be performed, the
temperature drastically reduces the metabolic organ-cultured cornea has to be transferred from
process of the corneas, according to the medium I into another hypertonic organ culture
Arrhenius relation. Wound healing is, therefore, medium (so-called medium II) for stromal
not possible, and tissue damage may progress deswelling, containing a macromolecule, mostly
[42]. dextran T500, with concentrations varying from
The technique of hypothermic storage is sim- 4 to 8% between eye banks [45]. The dextran
ple and does not require expensive equipment. cannot, due to technical limitations, be added to
Donor corneas are also directly available for cor- the isotonic medium I [51]. Deswelling time in
neal surgery. Compared to organ culture, storage medium II is usually between 1 and 7 days.
time appears shorter, but more recent storage Nevertheless, the proven toxicity of dextran on
solutions nonetheless allow for scheduled endothelial cells [52, 53] imposes a deswelling
surgery. time as short as possible, as the cornea can—in
22 L. Hamon et al.
a b
Fig. 2.3 Organ-cultured corneas in cell culture flasks. (a) Organ-cultured corneas are stored at +34 °C in an incubator
up to 28 days. (b) In cell culture flask, corneoscleral discs are maintained vertically on a plastic holder (arrow)
theory—already be implanted after only 12 h of Penetrating Keratoplasty

deswelling [54, 55]. This process of deswelling is
not necessary prior to DMEK. The penetrating keratoplasty (PKP) consists of a
Organ culture offers many advantages com- transplantation of all the layers of the cornea and
pared to hypothermic storage, such as the con- represented the only feasible keratoplasty until
trol of pathogen contamination during the the end 1950s [60]. Before transplantation, the
storage time [56], a longer storage time [45] and preserved corneas, previously transferred into
better endothelial vitality [57]. To permit these medium II for deswelling (in case of organ cul-
possibilities, the organ culture procedure is ture) [35, 54], must be appropriately dimensioned
more complicated and more expensive than in order to match the recipient’s bed.
hypothermic storage and requires well-equipped The type of trephination has a major impact on
facilities. the correct placement of the first four or eight car-
dinal sutures [61, 62] and, thus on the postopera-
tive astigmatism. Recipient trephination can be
Type of Tissues mechanical or nonmechanical. Conventional
mechanical trephination is always associated with
In recent years, a marked trend toward lamel- deformation of the recipient corneal tissue, includ-
lar surgery instead of PKP has been observed ing deformation of the incised edges, with irregu-
[58]. Anterior or posterior lamellar surgeries lar cut surfaces related to the axial and radial forces
have advantages in terms of visual rehabilita- induced by the use of the trephine [62, 63].
tion and immune response rate. Nevertheless, Nonmechanical trephination includes femtosec-
PKP still has an important place for certain ond or excimer laser cutting techniques. Significant
indications and in case of complicated surgery improvement in postoperative astigmatism can be
with a more uncertain prognosis [59]. This achieved using the Homburg/Erlangen technique
multiplicity of techniques encourages eye of nonmechanical excimer laser trephination [7,
banks to adapt and vary their procedures for 64, 65]. This induces significantly lower corneal
graft preparation and storage in order to meet astigmatism, more regular topography, and, thus,
this increasing demand. ultimately, better vision [7, 64].
For each keratoplasty, the diameter of the A major advantage of DALK is the absence of
graft should be individualized according to the allograft endothelial immune reaction, as donor
specificities of the patient. This diameter is endothelium is not transplanted. Moreover, tissue
determined before the surgery for each patient. preserved in the eye bank but presenting insuffi-
Each graft should be as large as possible (for cient endothelial cells for PKP or posterior lamel-
optical reasons) and as small as necessary (for lar surgery may still be selected and prepared for
immunological reasons) [62, 66, 67]. Excessive DALK. This allows greater flexibility in tissue
oversizing or undersizing of the graft relative to management and allocation for eye banks.
the recipient bed should be avoided to prevent
distension or retraction of the peripheral donor
tissue. Posterior Lamellar Keratoplasties
Mechanical or nonmechanical trephination of
the graft is usually performed in the operating Posterior lamellar keratoplasty techniques have
room by the surgeon. The role of eye banks prior steadily improved over the past 20 years, allow-
to PKP is limited to providing a corneoscleral ing faster visual recovery and triggering fewer
disc of the best possible quality. immune reactions compared to PKP due to the
use of very thin grafts [72]. Indications for poste-
rior lamellar keratoplasty include all diseases of
Anterior Lamellar Keratoplasties the corneal endothelium. Patients with Fuchs’
endothelial corneal dystrophy (FECD) represent
Stromal corneal pathologies and keratoconus are by far the largest group. Other endothelial dis-
typical indications for anterior lamellar kerato- eases include congenital hereditary endothelial
plasty. In these cases, deep anterior lamellar dystrophy (CHED), posterior polymorphous cor-
keratoplasty (DALK) has become an increas- neal dystrophy (PPCD), herpetic endothelitis or
ingly popular alternative compared to PKP [68]. buphthalmus [58].
In this procedure, the recipient’s corneal Descemet membrane endothelial kerato-
stroma is totally excised, leaving only the endo- plasty (DMEK) is becoming increasingly popu-
thelium and the Descemet membrane, with or lar internationally and especially in Europe and
without pre-Descemet’s layer [69]. To dissociate can now also be used in difficult conditions of the
the posterior stroma of the endothelial layer, sev- anterior segment of the eye [10, 58]. In DMEK,
eral techniques have been proposed. A popular only the Descemet membrane and the corneal
technique is the “big bubble technique” by endothelium are transplanted. The transplant is
Anwar et al. [70]. Descemet’s membrane and prepared prior to DMEK, with a risk of mem-
endothelium of the previously trephined donor brane rupture that may cause graft loss [73].
cornea are removed in a second step. The donor Several techniques for DMEK donor preparation
and recipient trephination is usually performed have been described, such as direct peeling with
manually. Still, it can also be performed with an microkeratome [74], the submerged corneas
excimer laser, thus combining the advantages in using backgrounds away (SCUBA) method,
terms of visual rehabilitation and immunological where the cornea is submerged in Optisol, bal-
“protection” of a DALK [71] and the advantages anced salt solution (BSS) or organ culture
in terms of graft regularity and low astigmatism medium to reduce surface tension during the
of an excimer-PKP for the patient [5]. The preparation [75] or pneumatic dissection [76].
donor’s full-thickness stroma is then positioned The use of artificial anterior chambers with aspi-
against the recipient’s Descemet membrane and ration or pressurization also proves to be useful
sutured using standard techniques as for PKP in facilitating the dissection [77]. Following dis-
[68]. section, the Descemet membrane with corneal
24 L. Hamon et al.
endothelium is prepared in order to be injected The use of prestripped tissues offers advan-
into the anterior chamber by the surgeon in place tages, such as the guaranteed immediate avail-
of the previously removed recipient’s affected ability of the graft, gain of time and reduced
endothelium (descemetorhexis) [11]. surgical complexity of the DMEK or DSAEK
Descemet’s stripping automated endothe- surgery [85–87]. Nevertheless, recent studies
lial keratoplasty (DSAEK) is one of the most have shown severe endothelial cell loss after pre-
performed “endothelial keratoplasties” in North stripping compared to nonpre-stripped DMEK-
America [73]. The technique consists of the tissues preserved in organ culture medium I, with
removal of endothelial tissue from the recipient endothelial cell loss reaching up to 23% for pre-
and to implant a donor posterior lenticle stripped corneas versus 4% for nonpre-stripped
(<200 μm) composed of posterior corneas after 5 days of storage [88, 89].
stroma + Descemet membrane + endothelium. Prestripped tissues have also shown decreased
Many factors contribute to visual outcomes fol- adhesion forces and elastic modulus, which may
lowing DSAEK, which are generally poorer after contribute to increased re-bubbling rates, com-
DMEK [78], including the presence of a stromal pared to nonpre-stripped tissues [90].
interface (stroma-to-stroma contact) and tech- Preloaded DMEK tissues are generally pre-
niques to prepare the donor lenticules [73]. For stripped and then preloaded in a transport car-
these reasons, the amount of DMEK (97.3% in tridge in order to be injected by the surgeon [91],
the year 2020) outweighed by far the number of comparable to a preloaded intraocular lens (IOL)
DSAEK in the past years in Germany [3]. in cataract surgery. In recent years, several non-
The donor lenticle can be prepared mechani- touch DMEK preloading techniques have been
cally using a microkeratome for intrastromal cut- developed [91, 92]. These techniques induce less
ting in corneal preparation, achieving a lenticle endothelial cell loss than previous preloading
thickness under 200 μm. The femtosecond laser- techniques, with comparable cell loss as pre-
prepared lenticle has been explored to improve stripped tissues, and demonstrate the practicality
lenticule uniformity, unfortunately resulting in of preparing injectable endothelial membranes
rougher stromal beds and increased irregularity, [93]. Nevertheless, the preparation of prestripped
without providing expected visual results [79]. and preloaded membranes must be performed in
Nowadays, ultra-thin lenticules (<130 μm) are accredited and experienced eye banks [85, 94]
preferred and used for so-called ultra-thin and the use of preloaded tissues should be lim-
DSAEK (UT-DSAEK) [58]. To achieve this ited to centers without on-site eye banks or
thinness, donor corneas undergo two passes with prompt availability of nonstripped tissues. A
a microkeratome, first with a thicker and second preparation (stripping and nontouch loading)
with a thinner pass [73]. Ultra-thin tissue can also immediately before surgery by an experienced
be prepared using low-pulse energy, high- surgeon should ensure the optimum viability of
frequency femtosecond laser [80]. the tissue. However, prestripped tissues pre-
served in medium I (without dextran) [52, 95] for
1 day, with a mean endothelial cell loss of 4.1%
restripped, Precut and Preloaded
P [96], represent a reasonable compromise
Tissues in Eye Banks between tissue quality and organizational con-
straints [11].
Advances in the field of eye banking have resulted Concerning (UT-)DSAEK, precut tissues pro-
in the preparation and validation of prepared tis- vided similar visual and refractive outcomes as
sues suitable for elective procedures [81]. Eye nonpre-cut tissues [97]. In addition, re-bubbling
banks have therefore started preparing precut and rates were also similar for precut and nonpre-cut
preloaded tissues for (UT-)DSAEK [82] and pre- transplants [98]. Donor endothelial cell loss from
stripped and preloaded tissues for DMEK [83, 6 to 12 months was also comparable after both
84]. techniques [97]. However, laboratory data on the
biomechanics of DSAEK grafts suggests that Tissue Selection

surgeon-cut DSAEK grafts present higher elastic
modulus and adhesion force compared to eye Before cultured corneas can be transplanted, they
bank-prepared DSAEK grafts [99]. have to fulfill certain quality criteria in accor-
Outcomes for using preloaded tissues for dance with current international and/or national
DMEK and (UT-)DSAEK were also compared in standards [16, 108]. These standards vary accord-
the literature. Romano et al. showed a signifi- ing to the type of transplantation (PKP, anterior
cantly higher visual acuity but also, unfortu- or posterior lamellar keratoplasty) and the elec-
nately, a higher rate of re-bubbling for tive or urgent nature of the surgery. Besides nec-
prestripped + preloaded DMEK, compared to essary examinations, new approaches to tissue
precut + prestripped UT-DSAEK [100]. analysis have been developed in recent years,
allowing an even better quality of transplanted
corneas from modern eye banks.
mniotic Membrane and Scleral
A
Patches
Microbiological Testing
Beside corneas, eye banks can also collect, pre-
serve and prepare alternative tissues such as In case of hypothermic storage, microbiological
amniotic membrane (the innermost layer of a testing with samples of the storage solution is gen-
fetal membrane) or scleral tissues. erally not performed, as the storage time is too
The amniotic membrane can be used, in short to receive the results before keratoplasty [45].
regards to its advantageous biological and immu- Moreover, the number of contaminating microbes
nological properties [101], as a biological ban- should be low and not grow at this temperature
dage (amniotic membrane patch [AMT]) for the after proper decontamination during retrieval [21]
treatment of nonhealing corneal ulcers or to sup- and before storage. Supplemented antibiotics have
port physiological wound-healing [102]. Small- a low effect during hypothermic storage but
or medium-sized perforations, leaking become active in the eye after keratoplasty as the
descemetoceles, and corneal melting may be body temperature recovers [109]. Although the rate
treated with AMT as tectonic surgery, if donor of keratitis or endophthalmitis post-keratoplasty is
cornea is not available [103]. very low [110], cases of graft-related infections—
Procurements of amniotic membrane are real- mostly fungal—have been described [111].
ized under strict aseptic conditions after elective In case of organ culture, microbiological test-
cesarean section [104]. In the eye bank, the amni- ing of the medium sample is realized during a
otic membrane is rinsed from blood remnants quarantine period at the beginning of the culture
and small segments are prepared and then cut process [48, 49]. The microbiological safety of
into multiple pieces of different dimensions (e.g., donor corneas is obtained by discarding contami-
2 × 2, 5 × 5 or 10 × 10 mm) [103]. After prepara- nated tissues prior to keratoplasty [112]. The con-
tion, these pieces are conserved using cryopreser- tamination rate widely varies between eye banks
vation at −75 to −85 °C [102] or lyophilization [45] and can be reduced through strict adherence
[105]. Prior to surgery, patches are warmed at to a quality management system (QMS) [113].
room temperature and/or rehydrated.
Scleral tissue can also be prepared in eye
banks for tectonic purposes, for example, in case orphological Examination: Slit-
M
of necrotic scleritis. The outside of the globe is Lamp Biomicroscopy
carefully cleaned and all other tissues must be
removed [106]. The sclera can then be stored dry Slit-lamp biomicroscopy has been a fundamental
or in 95% ethanol for at least 1 year until trans- method of tissue evaluation since the first insti-
plantation [107]. tuted criteria for eye bank certification [114]. The
26 L. Hamon et al.
technique has not changed in the past several and 2000 cells/mm2 is advised for anterior
decades and remains the gold standard for deter- lamellar keratoplasty or for emergency/tectonic
mining surgical suitability. Slit-lamp biomicros- surgery, but no minimum ECD is officially
copy allows an evaluation of all layers of the required [118]. Corneas with ECD lower than
cornea from both anterior and posterior perspec- 2000 cells/mm2 and corneas with endothelial
tives, using different magnifications and illumi- cell loss of more than 25% during cultivation
nation techniques, including direct illumination, have to be discarded for elective penetrating or
retro-illumination, specular reflection, and scle- posterior lamellar surgery.
rotic scatter, to focus on different layers [114]. Two other parameters are important for assess-
The examination is performed under sterile ing endothelial viability: morphology and vitality
conditions and remains very simple. In case of of the endothelial cells. Corneas with consider-
hypothermic storage, the cylindrical storage able polymegethism or pleomorphism have
plates are put on a fixation device mounted on the reduced functional reserves and an increased
slit lamp. In case of organ culture, the cell culture incidence of postoperative graft failure [119].
flasks are fixed on a horizontal support positioned Therefore, corneas with large central multicellu-
in place of the chin rest. lar necrosis, distinct endothelial cell necrosis,
Donor corneas are usually examined at the fold-associated endothelial cell necrosis, grouped
beginning of the preservation period to exclude a endothelial cell necrosis affecting more than 10%
morphological anomaly. The corneas that have of the total endothelial cell area, pronounced
passed this screening will be analyzed under polymegethism, pronounced pleomorphism, pro-
specular microscopy. A cornea presenting a scar nounced granulation/vacuolization or cornea gut-
can still be used for an emergency keratoplasty, tata (CG) also need to be discarded [2].
DSAEK, DMEK or—depending on the donor’s Inverted light microscopy is the first-choice
consent—for research projects. Donor corneas technique for organ-cultured corneas. The endo-
are mostly re-examined with a slit lamp at the thelial cells are visualized by swelling the intra-
end of the preservation period for definitive tissue cellular space using a hypotonic solution allowing
validation for patient’s allocation. endothelial layer inspection regardless of the cor-
neal hydration [120]. Intracellular border swell-
ing disappears after a couple of minutes.
ndothelial Evaluation: Inverted
E Application of vital stains such as trypan blue
Light or Specular Microscopy may help to discriminate dead or necrotic cells
[120]. To evaluate donor cornea endothelium,
The evaluation of the endothelial cell layer is of inverted light microscopy should be performed in
major importance to ensure tissue viability. the center, in the four paracentral/midperipheral
Indeed, a too low endothelial cell density (ECD) quadrants and in the periphery of the donor cor-
before keratoplasty is associated with postoper- nea [121] (Fig. 2.4).
ative graft failure [115]. Endothelial cell loss Specular microscopy can also be used and is
occurs during the keratoplasty and continues the first-choice if using hypothermic storage. The
after the surgery [116]. Up to 70% endothelial images generated are similar to in vivo specular
cell loss has been observed after 20 years [117]. microscopy images performed on patients pre-
Considering these facts, endothelial microscopy and postoperatively. In contrast to inverted light
using light or specular microscopes was added microscopy, specular microscopy is usually
in the early 2000s in eye bank protocols for tis- restricted to the center of the cornea, and visual-
sue evaluation. According to European stan- ized areas are relatively small because of a
dards, a minimum ECD of 2000 cells/mm2 is microscope- related fixed magnification.
needed for penetrating or posterior lamellar However, this technique does not require osmotic
keratoplasty in order to ensure long-term graft stimulation of the endothelial cells. Thus, the
survival [2]. A minimum ECD between 1000 donor cornea can remain in its cylindrical plate,
a c
Fig. 2.4 Evaluation of corneal endothelium vitality. The differentiate the endothelial cells gives this with spicu-
evaluation is performed with an inverted light microscope lated aspect of cell membranes. (b) Another endothelial
under sterile conditions in the eye bank. (a) Zone present- area with zones presenting necrosis (arrows). (c) A mem-
ing healthy endothelial cells. The use of hypotonic solu- ber of the eye bank staff performing endothelial cell count
tion to better perceive the intracellular spaces and thus
avoiding manipulations and associated endothe- refraction. Indeed, donor tissues with prior kera-
lial cell loss [121]. torefractive surgery or with pre-existing
After each endothelial evaluation, findings “refraction-affecting” pathologies cannot always
should be documented in detail, preferably be reliably recognized by slit-lamp examination
together with corresponding endothelial images. alone. By means of illustration, previous studies
Specific software for light or specular micro- demonstrated a false negative rate of 3.4–50.0%
scopes can help to determine parameters describ- for the identification of post-LASIK (laser-
ing the endothelial mosaic in terms of variation in assisted in situ keratomileusis) donor corneas,
cell shape and percentage of hexagonal cells in depending on whether detection was based on
addition to the assessment of the cell density [45]. slit-lamp examination, clinical history, or a com-
bination of both [126–128]. Therefore, many
authors have highlighted the need for improved
Sterile Donor Tomography screening of donor corneas in order to avoid
refractive surprises after keratoplasty [129–132].
Any corneal microsurgeon who performs kerato- Among the methods that have been proposed
plasties may have inadvertently transplanted a through the years as refractive screening devices
donor cornea with (subclinical) keratoconus or for corneal tissues in the eye bank, several authors
with prior keratorefractive surgery, typically have focused on the potential role of donor
resulting in significant postoperative refractive tomography [131, 133, 134]. A limitation of
disadvantages for the patient [122–125]. Given these devices was the difficulty of performing a
the increasing number of performed keratorefrac- tomographic measurement without having to
tive procedures in recent decades, eye banks will remove the corneoscleral disc from its container,
soon have to deal more intensively with the prob- i.e., in a sterile manner, due to the flat interfaces
lem of identifying donor corneas with abnormal of the container.
28 L. Hamon et al.
Our department succeeded, in collaboration

with our Institute of Experimental Ophthalmology,
in enabling and approving a new concept to topo-
graphically measure donor corneal tissues in
their sterile cell culture flask to detect potential
curvature and/or thickness deviation such as ker-
atoconus, prior refractive surgery or high astig-
matism [135–137].
Since 2018, 1061 donor corneas (Klaus Faber
Center for Corneal Diseases incl. LIONS Eye
Bank Saar-Lor-Lux, Trier/Westpfalz, in
Homburg/Saar) have been routinely measured
preoperatively using the swept-source anterior
segment optical coherence tomograph
(AS-OCT) CASIA 2 (Tomey Corp., Nagoya,
Japan). Because this device uses a central wave-
length of λ = 1310 nm and allows a penetration
depth of up to 13 mm in vivo [138], measure- Fig. 2.5 Sterile donor tomography in the eye bank.
ments of donor corneas are possible, achieving a Preoperative measurements of donor corneas are per-
lateral range of up to about 7 mm diameter, formed using the anterior segment optical coherence
essentially constrained by the tissue holder tomography (AS-OCT) in a sterile manner through their
cell culture flask, mounted on the chin rest of the AS-OCT
[139]. This technique enables a sterile assess- in a holder previously constructed using a three-
ment and excludes potential measurement- dimensional printer
related contamination of tissues within their
sealed cell culture flask, which is mounted on
the chin rest of the AS-OCT in a holder previ- wall and tissue holder, remove background
ously constructed by a three-dimensional (3D) noise, adjust the contrast size, reduce the bright-
printer (Ultimaker 2 Go, Ultimaker B.V., ness of the central reflex and to detect the edge
Geldermalsen, The Netherlands) (Fig. 2.5). In of the anterior and posterior surfaces of the
order to allow reliable measurements of corneal donor cornea [135, 137]. Considering the differ-
tissues, preoperative donor tomography was ent displacement of the apex of each sample
performed >12 h after transferring them into from the zero point of the coordinate system in
medium II (hypertonic medium enriched with space, the translation, rotation and 3D tilt of
6% dextran T500), i.e., after >12 h of deswell- samples were taken into account [137], thus not
ing [54, 55]. During measurement, a raster scan influencing the calculation of the donor corneas’
is performed from the posterior surface of the keratometry. Subsequently, a spherocylindrical
donor cornea, generating a 3D volume data set surface model was adapted with raytracing and,
and achieving a depth resolution of 5.621 μm/ ultimately, various geometric features such as
voxel in an aqueous medium and a lateral reso- the central corneal thickness (measured at the
lution of 6 μm/voxel [137]. Subsequently, the apex) and the anterior and posterior radii of cur-
measured raw data were imported into MATLAB vature at the steep and flat meridian were deter-
(MathWorks Inc., Natick, Massachusetts, USA) mined, from which the refractive power was
and preprocessed to extract artifacts due to flask calculated (Table 2.2).
Table 2.2 Result values of sterile donor tomography in pia correction or keratoconus in respectively
the eye bank
remarkably flat and steep donor corneas.
n = 1061 donor corneas
Mean Range
(x̅) ± SD (min.–max.)
Refractive Anterior Steep 50.5 ± 2.0 45.4–62.0
etection of Cornea Guttata
D
power (D) surface Flat 48.9 ± 1.5 42.7–56.1 in the Eye Bank
Posterior Steep −6.2 ± 0.3 −7.5 to −5.3
surface Flat −5.9 ± 0.3 −6.9 to −5.0 Cornea guttata (CG) was first described in 1921
Central corneal thickness 612 ± 81 379–1029 by Vogt as a clinical finding characterized by
(μm) droplet-like changes on the posterior surface of
Refractive power and pachymetry of the 1061 donor cor- the cornea detected using the slit-lamp examina-
neas were measured in their cell culture flask
x̅ = mean; SD = standard deviation; min. = minimum;
tion [145, 146]. CG is formed by the accumula-
max. = maximum tion of basement membrane and deposition of
collagen and fibril fibers resulting in endothelial
excrescences disrupting the endothelial mosaic
To avoid postoperative refractive surprises [147]. Since the endothelial layer regulates the
after keratoplasty, donor corneas whose refrac- inflow of aqueous humor into the stroma through
tive power deviates too much from the mean are an active transport pump, the presence of CG
discarded for penetrating or anterior lamellar leads to the loss of endothelial cells and inter-
keratoplasty. In Homburg/Saar, we have so far set feres with the function of the endothelial layer to
the threshold at curvature deviations of more than keep a transparent cornea [149, 148]. Although,
x̅ (mean) ± 3 standard deviations (eminence- in many cases, CG is considered an isolated find-
based) [140, 141] (status February 2022: donor ing without clinical significance, in many other
corneas with a total refractive power of <39.9 and cases, the disease progresses to become clini-
>47.8 D). Those donor tissues may, however, still cally significant; at that point, the disease entity
be suitable for posterior lamellar keratoplasty, is called Fuchs’ endothelial corneal dystrophy
such as DMEK or DSAEK. Furthermore, in the (FECD). As the disease progresses, the endothe-
future, sterile donor tomography could enable: lial layer can no longer maintain the transpar-
(1) the harmonization of donor and recipient ency of the cornea resulting in corneal edema
tomography [142, 143], which could potentially and visual loss [150–152]. The prevalence of CG
minimize persistent residual astigmatism after varies in the literature and increase with increas-
suture removal for a particular donor-recipient ing age reaching up to 11% in one study [153,
pair; and (2) the improvement of the IOL power 154]. CG usually appears centrally and spreads
calculation in a classical triple procedure by peripherally as the disease progresses and can be
means of regression analysis between pre- and clinically detected by its “beaten metal” appear-
postoperative total refractive power of donor cor- ance on the slit-lamp biomicroscopy [155].
neal tissues [144]. These projects are currently Using specular microscopy as a diagnostic
under research, as well as (a) the reliability [55] method, CG can be objectively detected and
and (b) the validity of donor tomography by quantified [156].
means of available keratometric measurements Unfortunately, the detection of CG in donor
during the donor’s lifetime, or alternatively using corneas is not as simple as in living patients. On
histological reprocessing of unused corneal tis- one side, the visualization of CG in the swollen
sues to detect evidence of keratorefractive myo- donor corneas stored in their culture medium
30 L. Hamon et al.
clinically using the slit lamp is extremely difficult underwent keratoplasties. These patients were
and, in most cases, not possible. On the other classified according to the postoperative CG
side, its detection using the inverted light micros- grade whereafter their corresponding 1582 pre-
copy in the eye banks is neither simple nor stan- operative endothelial pictures taken by an
dardized. Therefore, in order to assess the inverted light microscope were analyzed. The
prevalence and severity of cornea guttata in donor results of this study showed that three morpho-
corneas, two large-scale studies analyzed the logical criteria correlated with the presence of
prevalence and clinical significance of cornea cornea guttata. Those are the presence of less
guttata in 1758 patients after PKP and 664 than 50% of the cells in an endothelial picture
patients after DMEK [157, 158]. These studies having a hexagonal or a circular shape, the pres-
showed a CG prevalence of 14.9% after PKP and ence of cell membrane defects and interruptions
18.7% after DMEK, of which 13.6% and 16.9% and the presence of a small thickening of the cell
represented a low-grade CG without clinical sig- membrane “blebs.”
nificancy, respectively, whereas 1.3% and 1.9% Artificial intelligence has been a hot topic in
showed high-grade CG, respectively. Only high- recent years and has greatly contributed to the
grade CG showed a significant negative impact improvement of diagnostic techniques in the
on clinical parameters such as visual acuity and field of ophthalmology [159]. Artificial intelli-
central corneal thickness. In total, around 13% gence can also be used in the field of eye banking
and 8% of the whole population studied under- for the (semi-)automated detection of CG in
went a progression from either a healthy corneal specular microscopy images. Such a project is
transplant to one having CG, or from a transplant currently under development at the Klaus Faber
with low-grade CG to high-grade CG after PKP Center for Corneal Diseases, incl. LIONS Eye
and DMEK, respectively. These results highlight Bank Saar-Lor-Lux, Trier/Westpfalz, in collabo-
the importance of detecting CG in donor corneas ration with the German Research Center for
in the eye bank, especially high-grade CG, to pre- Artificial Intelligence (DFKI). The programmed
vent the transplantation of such diseased machine learning algorithms based on complex
corneas. neural networks allow automated endothelial
In order to characterize the features of CG cell count as well as the detection of abnormali-
seen under the inverted light microscope used in ties in specular microscopy images, including
the eye banks, a large retrospective study aimed areas of necrosis and the presence of GC
to establish semi-quantitative morphological cri- (Fig. 2.6). This emerging technology should
teria for the detection of CG in donor corneas greatly contribute to improving the cornea selec-
[59, 141]. This study included 262 patients who tion for eye banks.
Fig. 2.6 Automated detection of cornea guttata (CG) in an enhanced visualization of the endothelial layer. (b)
the eye bank using artificial intelligence (AI). Kittool: Machine learning classifiers including Case-Based
decision support tool for the detection of CG integrating Reasoning (CBR) for AI-based support enable automated
two components: (a) Graphical analytic tools, whereby CG detection in the eye bank by comparison with previ-
endothelial cells are processed to generate several cell ous endothelial cell images with known postoperative
representations such as “honeycomb” representation for classification of the graft endothelium after keratoplasty
Benefits of a Highly-Structured Organization for Standardization) 9001:2015

Quality Management System (QMS) standard can be adopted for the entire process
by Example of the LIONS Eye Bank from donation to transplantation. This standard
Saar-Lor-Lux, Trier/Westpfalz follows a process-oriented approach with the
goal of continuous improvement. The quality
A quality management system (QMS) is an management system is based on the principle
essential component of a highly functioning of good practice and provides defined instruc-
eye bank to ensure a maximum level of quality tions and standard operating procedures
and safety of human tissues [160]. While a (SOPs) for every step of the donation-
number of different quality management sys- transplantation process and documentation and
tems can be used, the ISO (International processing of data.
32 L. Hamon et al.
a b
Fig. 2.7 Sterile room in the Klaus Faber Center for B according to ISO Cleanroom Standards for the manu-
Cornea Diseases, incl. LIONS Eye Bank Saar–Lor–Lux, facture of Sterile Medicinal Products. (b) A staff member
Trier/Westpfalz. (a) Decontamination corridor. The staff sterilely dressed, renewing the organ culture medium of
dresses in several steps through this corridor (cleanroom organ-cultured corneas under maximum sterile condi-
grades D and C) before arriving in the clean rooms grade tions, under laminar flow (grade A)
The eye bank is bound by the laws, technical Upon arrival at the eye bank, an identification
norms, guidelines and legislative frameworks of code ‑is assigned to the corneal tissue. The tissue
its country which give guiding values to the pro- is quarantined and serologic and microbiologic
cess of corneal donation, handling of the donor examinations [21] in accordance with state regu-
tissue within the eye bank, transport of tissues lations as well as slit-lamp examinations and
and transplantation [2, 108]. The eye bank must evaluations of the endothelial cell density (ECD)
be located in a suitable facility and equipped with and morphology are performed before the tissue
the necessary technical apparatus, instruments can be declared suitable for transplantation [161].
and materials. The Klaus Faber Center for The traceability of the donor and the recipient
Corneal Diseases (KFZH) including the LIONS must be ensured and documented throughout the
Eye Bank Saar-Lor-Lux, Trier/Westpfalz for entire process. In addition, the LIONS Eye Bank
example, which was newly founded in 2019, pro- Saar-Lor-Lux, Trier/Westpfalz routinely uses
vides a class A clean room and all facilities which sterile donor tomography to further improve the
sets the highest quality standard for the cultiva- quality of the donor tissue and avoid transplant-
tion work-flow (Fig. 2.7). ing tissues with curvature abnormalities [140].
Managing human resources is an essential Written instructions should be available as a
part of a well-functioning eye bank. The staff documental architecture that can easily be
must successfully complete initial basic training accessed and kept up-to-date for the entire
and necessary refresher courses and demonstrate process. Instructions need to be shared with and
essential knowledge in order to carry out the followed by the personnel of the eye bank as well
expected tasks. Before retrieving human corneas, as interested external parties. Each document
personnel must be sufficiently trained and be should provide detailed instructions. These need
familiar with the necessary documentation to be written, authenticated, approved and dis-
regarding the consent of the donor family, donor tributed. In the event of changes to the organiza-
selection criteria, contraindications, and the tion, new legislative requirements, new
proper techniques for harvesting the cornea and medical-scientific knowledge or changes in the
reconstruction of the eye. process, the documents need to be updated.
Internal and external audits need to be held the future due to its economic and logistical
regularly in order to monitor, maintain and advantages. Nevertheless, these practices tend
improve the quality management system and to transform eye banks into “market places”
obtain national and international accreditations. for surgeons, a development that presents risks
Furthermore, the eye bank should continuously of unequal access to “good quality” tissue for
monitor and analyze data to ensure continuous small institutions and their patients and breaks
performance improvement. Hereby, tissue qual- the link between patients, surgeons and eye
ity can constantly be increased and the number of banks.
discarded tissues due to quality concerns can be Advances in preservation processes are being
diminished [113]. developed, with—for example—the active stor-
age machine (ASM), a device where corneas are
conserved in banks of “storage plates” (Fig. 2.8),
Future Perspectives allowing almost physiological conditions of elec-
trolytic medium and pressure for donor corneas
The number of active eye banks has decreased [163, 164].
drastically in recent years [162]. The global In terms of selection techniques, new methods
activity is refocused on large structures with of tissue analysis such as high-resolution two-
more resources, allowing optimal conditions photon imaging provide information about the
of preservation as well as a better selection cells metabolic state and structural organization
and preparation of tissues. These large struc- of the stroma, with subcellular resolution [165].
tures, in charge of the “production” of tissues The rise of artificial intelligence and convolu-
for external institutions, are developing tional neural networks should enable automation
“ready-to-use” tissues to facilitate transport and better efficiency of tissue selection processes
and use [91]. This trend is expected to grow in in eye banking.
a b
Fig. 2.8 Active Storage Machine (ASM). (a) Docking tor) from a tank. Fluid is circulated by a peristaltic pump
station with monitoring system for modular bioreactors driven by a pressure sensor, a solenoid valve and a micro-
fitted into slots. (b) “Bioreactor.” The cornea, located in controller. The pressure inside the endothelial chamber is
the heart of the bioreactor, separates the endothelial and kept at 21.5 mmHg. Used medium is removed to a waste
epithelial chambers. The endothelial chamber is filled compartment. (©Sincler. All rights reserved)
with culture medium (containing red phenol as pH indica-
34 L. Hamon et al.
Take Home Notes Hamon, Loay Daas, Adrien Quintin, Tarek Safi, Isabel
Weinstein, and Berthold Seitz.
• Hypothermic storage, simple and not expen- The authors attest that they meet the current ICMJE
sive but allowing only short-term storage, and criteria for authorship.
organ culture, more complex and expensive
but allowing better control, longer storage
and better endothelial quality, currently References
remain the two major storage methods in eye
banks. 1. Zirm E. [A successful total keratoplasty].
• The use of prestripped and preloaded tissues Albrecht Von Graefes Arch Klin Exp Ophthalmol.
1906;64:580–93.
produced by eye banks for DMEK and 2. European Eye Bank Association (EEBA). Technical
DSAEK represents a valuable option for sur- guidelines for ocular tissue (TGOT). Venice:
geons in centers without on-site eye banks or EEBA; 2020. https://www.eeba.eu/files/pdf/
prompt availability of nonstripped tissues but EEBA_Technical_Guidelines_for_Ocular_Tissue_
Revision11.pdf. Accessed 22 Feb 2022.
should be carried out by accredited and expe- 3. Flockerzi E, Maier P, Böhringer D, Reinshagen
rienced eye banks and as close as possible to H, Kruse F, Cursiefen C, Reinhard T, Geerling
the elective surgery. G, Torun N, Seitz B, All German Keratoplasty
• New selection techniques such as the sterile Registry Contributors. Trends in corneal transplan-
tation from 2001 to 2016 in Germany; a report of
donor tomography or the detection of cor- the DOG-section cornea and its keratoplasty regis-
nea guttata using artificial intelligence allow try. Am J Ophthalmol. 2018;188:91–8. https://doi.
a better selection and allocation of tissues in org/10.1016/j.ajo.2018.01.018.
the eye bank in order to avoid transplanting 4. Molter Y, Milioti G, Langenbucher A, Seitz B.
[Appearance, recurrence and prognosis of immu-
tissues with curvature or morphological nological graft rejection after penetrating kerato-
abnormalities. plasty]. Ophthalmologe. 2020;117:548–56. https://
• A well-designed quality management sys- doi.org/10.1007/s00347-019-00975-9.
tem is essential to ensure the quality and 5. Seitz B, Langenbucher A, Naumann GO.
[Perspectives of excimer laser-assisted kerato-
safety of corneal tissue and to face the mul- plasty]. Ophthalmologe. 2011;108:817–24. https://
tiple new challenges of modern eye doi.org/10.1007/s00347-011-2333-x.
banking. 6. Holzer MP, Rabsilber TM, Auffarth GU. Penetrating
• Improvement of existing techniques and new keratoplasty using femtosecond laser. Am
J Ophthalmol. 2007;143:524–6. https://doi.
advances for tissue preservation, preparation org/10.1016/j.ajo.2006.08.029.
and selection promises further developments 7. Seitz B, Langenbucher A, Kus MM, Küchle M,
in the field of eye banking in the near future. Naumann GOH. Nonmechanical corneal trephi-
nation with the excimer laser improves outcome
after penetrating keratoplasty. Ophthalmology.
1999;106:1156–65.
Acknowledgments The authors thank Prof. Dr. Gilles 8. Hoffmann F. [Suture technique for perforat-
Thuret, the team of CHU Saint-Etienne (France) and ing keratoplasty]. Klin Monatsbl Augenheilkd.
SINCLER company for their contribution to the section 1976;169:584–90.
concerning the “active storage machine (ASM)” and the 9. Suffo S, Seitz B, Daas L. [The Homburg cross-stitch
availability of Fig. 2.8. marker for double-running sutures in penetrating ker-
The authors thank Ms. Christina Turner (director’s atoplasty]. Klin Monbl Augenheilkd. 2021;238:808–
assistant and in charge of the publication and education 14. https://doi.org/10.1055/a-1275-0807.
pool, Department of Ophthalmology, Saarland University 10. Melles GR. Descemet membrane endothelial kera-
Medical Center (UKS), Homburg/Saar, Germany) who toplasty (DMEK). Cornea. 2006;25:987–90. https://
reviewed the first draft of the manuscript and ensured a doi.org/10.1097/01.ico.0000248385.16896.34.
correct English language. 11. Seitz B, Daas L, Flockerzi E, Suffo S. [Descemet
membrane endothelial keratoplasty DMEK -
Funding Funding/Support: No funding or grant support. Donor and recipient step by step]. Ophthalmologe.
2020;117:811–28. https://doi.org/10.1007/
s00347-020-01134-1.
Conflicts of Interest The following authors have no 12. Luengo-Gimeno F, Tan DT, Mehta JS. Evolution
financial disclosures and no conflict of interest: Loïc of deep anterior lamellar keratoplasty (DALK).
Ocul Surf. 2011;9:98–110. https://doi.org/10.1016/ 26. Basu PK, Hasany SM, Ranadive NS, Chipman
s1542-0124(11)70017-9. M. Damage to corneal endothelium cells by lyso-
13. Hata B. The development of glioma in the eye to somal enzymes in stored human eyes. Can J
which the cornea of a patient, who suffered from Ophthalmol. 1980;15:137–40.
glioma, was transplanted. Acta Soc Ophthalmol Jpn. 27. Capella JA, Kaufman HE, Robbins JE. Preservation
1939;43:1763–7. of viable corneal tissue. Arch Ophthalmol.
14. Armitage WJ, Tullo AB, Ironside JW. Risk of 1965;74:669–73. https://doi.org/10.1001/archo
Creutzfeldt-Jakob disease transmission by ocu- pht.1965.00970040671015. PMID: 5891917.
lar surgery and tissue transplantation. Eye. 28. O’Neill P, Mueller FO, Trevor-Roper PD. On
2009;23:1926–30. the preservation of corneae at -196 degrees C
15. Caron MJ, Wilson R. Review of the risk of HIV for full-thickness homografts in man and dog.
infection through corneal transplantation in the Br J Ophthalmol. 1967;51:13–30. https://doi.
United States. J Am Optom Assoc. 1994;65:173–8. org/10.1136/bjo.51.1.13.
16. European Eye Bank Association. Minimum Medical 29. Sperling S. Cryopreservation of human cadaver
Standards (MMS). Venice: EEBA; 2020. https:// corneas regenerated at 31 degrees C in a modi-
www.eeba.eu/files/pdf/EEBA%20Minimum%20 fied tissue culture medium. Acta Ophthalmol
Medical%20Standards%20Revision%205%20Final. (Copenh). 1981;59(1):142–4148. http://doi.
pdf. Accessed 24 Feb 2022. org/10.1111/j.1755-3768.1981.tb06722.x.
17. Heim A, Wagner D, Rothämel T, Hartman U, Flik J, 30. European Directorate for the Quality of Medicines
Verhagen W. Evaluation of serological screening of and Healthcare (EDQM). Guide to the quality and
cadaveric sera for donor selection for corneal trans- safety of tissues and cells for human application, vol.
plantation. J Med Virol. 1999;58:291–5. 2. Strasbourg: Council of Europe; 2017. p. 185–95.
18. Martin C, Tschernig T, Hamon L, Daas L, Seitz 31. Rodríguez-Fernández S, Álvarez-Portela M,
B. Corneae from body donors in anatomy depart- Rendal-Vázquez E, Piñeiro-Ramil M, Sanjurjo-
ment: valuable use for clinical transplantation Rodríguez C, Castro-Viñuelas R, Sánchez-Ibáñez
and experimental research. BMC Ophthalmol. J, Fuentes-Boquete I, Díaz-Prado S. Analysis of
2020;13(20):284. https://doi.org/10.1186/ cryopreservation protocols and their harmful effects
s12886-020-01546-2. on the endothelial integrity of human corneas. Int
19. Serfözö A, Suffo S, Viestenz A, Seitz B, Daas L. J Mol Sci. 2021;22:12564. https://doi.org/10.3390/
[Central sclerocorneoplasty à chaud in bacterial ijms222212564.
superinfected mycotic keratitis confirmed by confo- 32. Tsang HF, Chan LWC, Cho WCS, Yu ACS, Yim
cal microscopy at the base of recurrent herpetic kerati- AKY, Chan AKC, Ng LPW, Wong YKE, Pei XM,
tis]. Klin Monbl Augenheilkd. 2021;239(10):1245–7. Li MJW, Wong SC. An update on COVID-19 pan-
https://doi.org/10.1055/a-1386-5229. demic: the epidemiology, pathogenesis, prevention
20. Sangwan VS, Jain V, Gupta P. Structural and and treatment strategies. Expert Rev Anti Infect
functional outcome of scleral patch graft. Eye. Ther. 2021;19:877–88. https://doi.org/10.1080/1478
2007;21:930–05. 7210.2021.1863146.
21. Li S, Bischoff M, Schirra F, Langenbucher A, Ong 33. Chaurasia S, Das S, Roy A. A review of long-term
M, Halfmann A, Herrmann M, Seitz B. [Correlation corneal preservation techniques: relevance and
between microbial growth in conjunctival swabs of renewed interests in the COVID-19 era. Indian
corneal donors and contamination of organ culture J Ophthalmol. 2020;68:1357–63. https://doi.
media]. Ophthalmologe 2014;111:553–9. org/10.4103/ijo.IJO_1505_20.
22. Lawlor M, Kerridge I, Ankeny R, Dobbins TA, Billson 34. Casagrande M, Fitzek A, Spitzer MS, Püschel K,
F. Specific unwillingness to donate eyes: the impact Glatzel M, Krasemann S, Nörz D, Lütgehetmann
of disfigurement, knowledge and procurement on M, Pfefferle S, Schultheiss M. Presence of SARS-
corneal donation. Am J Transplant. 2010;10:657–63. CoV-2 RNA in the cornea of viremic patients with
https://doi.org/10.1111/j.1600-6143.2009.02986.x. COVID-19. JAMA Ophthalmol. 2021;139:383–8.
23. Wykrota AA, Weinstein I, Hamon L, Daas L, https://doi.org/10.1001/jamaophthalmol.2020.6339.
Flockerzi E, Suffo S, Seitz B. Approval rates for 35. Hamon L, Bayyoud T, Seitz B. Ocular findings in
corneal donation and the origin of donor tissue for patients with COVID-19: impact on eye banking
transplantation at a university-based tertiary refer- [Letter]. Clin Ophthalmol. 2021;15:2051–2. https://
ral center with corneal subspecialization hosting a doi.org/10.2147/OPTH.S317378.
LIONS Eye Bank. BMC Ophthalmol. 2022;22:17. 36. McCarey BE, Kaufman HE. Improved corneal stor-
https://doi.org/10.1186/s12886-022-02248-7. age. Invest Ophthalmol. 1974;13:165–73.
24. Filatov VP. Transplantation of the cornea. Arch 37. Basu PK. A review of methods for storage of cor-
Opthalmol. 1935;13:321–47. nea for keratoplasty. Indian J Ophthalmol. 1995;43:
25. Bourne WM. Corneal preservation. In: Kaufman 55–8.
HE, Barron BA, Waltman SR, Mc Donald MB, edi- 38. Keates RH, Rabin B. Extending corneal stor-
tors. The cornea. Oxford: Butterworth-Heinemann; age with 2.5% chondroitin sulfate (K-Sol).
1988. p. 713–24.
36 L. Hamon et al.
Ophthalmic Surg. 1988;19:817–20. https://doi. preservation on DMEK (Descemet membrane endo-

org/10.3928/0090-4481-19881101-13. thelial keratoplasty) outcome. Graefes Arch Clin
39. Lindstrom RL, Kaufman HE, Skelnik DL, Laing RA, Exp Ophthalmol. 2018;256:213–5.
Lass JH, Musch DC, Trousdale MD, Reinhart WJ, 53. Spelsberg H, Reinhard T, Sundmacher R. [Epithelial
Burris TE, Sugar A, Davis RM, Hirokawa K, Smith damage of corneal grafts after prolonged storage
T, Gordon JF. Optisol corneal storage medium. Am J in dextran-containing organ culture medium – a
Ophthalmol. 1992;114:345–56. prospective study]. Klin Monbl Augenheilkd.
40. Steinemann TL, Kaufmann HE, Lindstrom RL, 2002;219:417–21.
Beuerman RW, Varnell ED. Intermediate-term stor- 54. Hamon L, Daas L, Mäurer S, Weinstein I, Quintin
age media (K-Sol, Dexsol, Optisol). In: Brightbill A, Schulz K, Langenbucher A, Seitz B. Thickness
FS, editor. Corneal surgery. St Louis: Mosby; 1993. and curvature changes of human corneal grafts in
p. 609–13. dextran-containing organ culture medium before
41. Moshirfar M, Odayar VS, McCabe SE, Ronquillo keratoplasty. Cornea. 2021;40:733–40. https://doi.
YC. Corneal donation: current guidelines and future org/10.1097/ICO.0000000000002543.
direction. Clin Ophthalmol. 2021;15:2963–73. 55. Hamon L, Quintin A, Mäurer S, Weinstein I,
https://doi.org/10.2147/OPTH.S284617. Langenbucher A, Seitz B, Daas L. Reliability and
42. Armitage WJ. Preservation of human cornea. efficiency of corneal thickness measurements using
Transfus Med Hemother. 2011;38:143–7. https:// sterile donor tomography in the eye bank. Cell Tissue
doi.org/10.1159/000326632. Bank. 2022;23(4):695–706. https://doi.org/10.1007/
43. Summerlin WT, Miller GE, Harris JE, s10561-021-09980-2.
Good RA. The organ-cultured cornea: an 56. Borderie VM, Laroche L. Microbiologic study of
in vitro study. Invest Ophthalmol. 1973;12: organ-cultured donor corneas. Transplantation.
176–80. 1998;66:120–3.
44. Doughman DJ, Harris JE, Schmitt MK. Penetrating 57. Thuret G, Chiquet C, Bernal F, et al. Prospective,
keratoplasty using 37 C organ cultured cornea. Trans randomized, clinical and endothelial evalua-
Sect Ophthalmol Am Acad Ophthalmol Otolaryngol. tion of 2 storage times for cornea donor tissue in
1976;81:778–93. organ culture at 31 degrees C. Arch Ophthalmol.
45. Pels E, Beele H, Claerhout I. Eye bank issues: 2003;121:442–50.
II. Preservation techniques: warm versus cold stor- 58. Bachmann B, Schaub F, Cursiefen C. [Treatment
age. Int Ophthalmol. 2008;28:155–63. https://doi. of corneal endothelial disorders by DMEK and
org/10.1007/s10792-007-9086-1. UT-DSAEK. Indications, complications, results and
46. Lindstrom RL, Doughman DJ, Skelnik DL, follow-up]. Ophthalmologe. 2016;113:196–203.
Mindrup EA. Minnesota system corneal preserva- https://doi.org/10.1007/s00347-016-0221-0.
tion. Br J Ophthalmol. 1986;70:47–54. https://doi. 59. Seitz B, Daas L, Hamon L, Fraenkel D, Szentmàry
org/10.1136/bjo.70.1.47. N, Langenbucher A, Suffo S. La kératoplastie trans-
47. Frueh BE, Böhnke M. Corneal grafting of donor fixiante. In: Thuret G, Muraine M, Gain P, editors.
tissue preserved for longer than 4 weeks in organ- L’endothélium cornéen. Paris: Elsevier-Masson;
culture medium. Cornea. 1995;14:463–6. 2021. p. 121–5.
48. Thuret G, Carricajo A, Chiquet C, Vautrin AC, Celle 60. Hallerman W. [Some remarks on keratoplasty]. Klin
N, Boureille M, Acquart S, Aubert G, Maugery Monatsbl Augenheilkd. 1959;135:252–9.
J, Gain P. Sensitivity and rapidity of blood cul- 61. Olson RJ. Modulation of postkeratoplasty astig-
ture bottles in the detection of cornea organ cul- matism by surgical and suturing techniques. Int
ture media contamination by bacteria and fungi. Ophthalmol Clin. 1983;23:137–51.
Br J Ophthalmol. 2002;86:1422–7. https://doi. 62. Seitz B, Langenbucher A, Naumann
org/10.1136/bjo.86.12.1422. GOH. Trephination in penetrating keratoplasty. In:
49. Thuret G, Carricajo A, Vautrin AC, Raberin H, Reinhard T, Larkin F, editors. Essentials in ophthal-
Acquart S, Garraud O, Gain P, Aubert G. Efficiency mology - corneal and external eye disease. Berlin:
of blood culture bottles for the fungal sterility test- Springer; 2006. p. 123–52.
ing of corneal organ culture media. Br J Ophthalmol. 63. Naumann GOH. Part II: corneal transplantation in
2005;89:586–90. https://doi.org/10.1136/ anterior segment diseases. The Bowman Lecture
bjo.2004.053439. (Number 56) 1994. Eye. 1995;9:395–421.
50. Müller LJ, Pels E, Vrensen GF. The effects of organ- 64. Seitz B, Hager T, Langenbucher A, Naumann
culture on the density of keratocytes and collagen GOH. Reconsidering sequential double running
fibers in human corneas. Cornea. 2001;20:86–95. suture removal after penetrating keratoplasty – a pro-
51. Sperling S. Human corneal epithelium in organ cul- spective randomized study comparing excimer laser
ture. The influence of temperature and medium of and motor trephination. Cornea. 2018;37:301–6.
incubation. Acta Ophthalmol. 1979;57:269–76. 65. Seitz B, Daas L, Milioti G, Szentmàry N,
52. Abdin AD, Daas L, Pattmöller M, Suffo S, Langenbucher A, Suffo S. [Excimer laser-assisted
Langenbucher A, Seitz B. Negative impact of dex- penetrating keratoplasty: on 1 July 2019 excimer
tran in organ culture media for pre-stripped tissue laser penetrating keratoplasty celebrates its 30th
anniversary - Video article]. Ophthalmologe. keratoplasty versus Descemet stripping auto-

2019;116:1221–30. https://doi.org/10.1007/ mated endothelial keratoplasty. Am J Ophthalmol.
s00347-019-00990-w. 2012;153:1082–1090.e1082.
66. Seitz B, Langenbucher A, Zagrada D, Budde W, 79. Vetter JM, Butsch C, Faust M, Schmidtmann I,
Kus MM. [Corneal dimensions in various types Hoffmann EM, Sekundo W, Pfeiffer N. Irregularity
of corneal dystrophies and their effect on pen- of the posterior corneal surface after curved interface
etrating keratoplasty] Klin Monatsbl Augenheilkd. femtosecond laser-assisted versus microkeratome-
2000;217:152–8. assisted Descemet stripping automated endothelial
67. Seitz B, Langenbucher A, Küchle M, Naumann keratoplasty. Cornea. 2013;32:118–24. https://doi.
GOH. Impact of graft diameter on corneal power org/10.1097/ICO.0b013e31826ae2d8.
and the regularity of postkeratoplasty astigmatism 80. Phillips PM, Phillips LJ, Saad HA, Terry MA,
before and after suture removal. Ophthalmology. Stolz DB, Stoeger C, Franks J, Davis-Boozer D.
2003;110:2162–7. “Ultrathin” DSAEK tissue prepared with a low-
68. Muraine M, Toubeau D, Gueudry J, Brasseur pulse energy, high-frequency femtosecond laser.
G. Impact of new lamellar techniques of kera- Cornea. 2013;32:81–6. https://doi.org/10.1097/
toplasty on eye bank activity. Graefes Arch Clin ICO.0b013e31825c72dc.
Exp Ophthalmol. 2007;245:32–8. https://doi. 81. Parekh M, Ruzza A, Steger B, Willoughby CE,
org/10.1007/s00417-006-0390-7. Rehman S, Ferrari S, Ponzin D, Kaye SB, Romano
69. Dua HS, Faraj LA, Said DG, Gray T, Lowe V. Cross-country transportation efficacy and clini-
J. Human corneal anatomy redifined. A novel pre- cal outcomes of preloaded large-diameter ultra-thin
Descemet’s layer (Dua’s layer). Ophthalmology. Descemet stripping automated endothelial kerato-
2013;120:1778–85. https://doi.org/10.1016/j. plasty grafts. Cornea. 2019;38:30–4. https://doi.
ophtha.2013.01.018. org/10.1097/ICO.0000000000001777.
70. Anwar M, Teichmann KD. Big-bubble technique to 82. Bielefeld V, Vabres B, Baud’huin M, Lebranchu P,
bare Descemet’s membrane in anterior lamellar ker- Le Meur G, Orignac I. [UT-DSAEK using grafts
atoplasty. J Cataract Refract Surg. 2002;28:398–403. provided by a tissue bank, a one-year follow-up
71. Daas L, Hamon L, Ardjomand N, Safi T, Seitz study of 79 grafts]. J Fr Ophtalmol. 2021;44:176–88.
B. [Excimer laser-assisted DALK: a case report https://doi.org/10.1016/j.jfo.2020.04.060.
from the Homburg Keratoconus Center (HKC)]. 83. Newman LR, DeMill DL, Zeidenweber DA,
Ophthalmologe. 2021;118:1245–8. https://doi. Mayko ZM, Bauer AJ, Tran KD, Straiko MD,
org/10.1007/s00347-021-01342-3. Terry MA. Preloaded Descemet membrane endo-
72. Anshu A, Price MO, Price FW Jr. Risk of corneal thelial keratoplasty donor tissue: surgical technique
transplant rejection significantly reduced with and early clinical results. Cornea. 2018;37:981–6.
Descemet’s membrane endothelial keratoplasty. https://doi.org/10.1097/ICO.0000000000001646.
Ophthalmology. 2012;119:536–40. 84. Ruzza A, Salvalaio G, Bruni A, et al. Banking of
73. Boynton GE, Woodward MA. Eye-bank prepara- donor tissues for Descemet stripping automated
tion of endothelial tissue. Curr Opin Ophthalmol. endothelial keratoplasty. Cornea. 2013;32:70–5.
2014;25:319–24. https://doi.org/10.1097/ https://doi.org/10.1097/ICO.0b013e31825e8442.
ICU.0000000000000060. 85. Martin AI, Devasahayam R, Hodge C, Cooper S,
74. Yoeruek E, Bartz-Schmidt KU. Novel surgical instru- Sutton GL. Analysis of the learning curve for pre-
ments facilitating Descemet membrane dissection. cut corneal specimens in preparation for lamellar
Cornea. 2013;32:523–6. https://doi.org/10.1097/ transplantation: a prospective, single-centre, con-
ICO.0b013e3182588ae9. secutive case series prepared at the Lions New South
75. Price MO, Giebel AW, Fairchild KM, Price FW Jr. Wales Eye Bank. Clin Exp Ophthalmol. 2017;45:
Descemet’s membrane endothelial keratoplasty: pro- 689–94.
spective multicenter study of visual and refractive 86. Marty AS, Burillon C, Desanlis A, et al. Validation
outcomes and endothelial survival. Ophthalmology. of an endothelial roll preparation for Descemet
2009;116:2361–8. https://doi.org/10.1016/j. membrane endothelial keratoplasty by a cornea bank
ophtha.2009.07.010. using “no touch” dissection technique. Cell Tissue
76. Venzano D, Pagani P, Randazzo N, Cabiddu F, Bank. 2016;17:225–32.
Traverso CE. Descemet membrane air-bubble sepa- 87. Terry MA, Straiko MD, Veldman PB, et al. Standardized
ration in donor corneas. J Cataract Refract Surg. DMEK technique: reducing complications using pre-
2010;36:2022–7. stripped tissue, novel glass injector, and sulfur hexa-
77. Muraine M, Gueudry J, He Z, Piselli S, Lefevre S, fluoride (SF6) gas. Cornea. 2015;34:845–52.
Toubeau D. Novel technique for the preparation of 88. Safi T, Daas L, Kiefer GL, Sharma M, Ndiaye A,
corneal grafts for descemet membrane endothelial Deru M, Alexandersson J, Seitz B. Semi-quantitative
keratoplasty. Am J Ophthalmol. 2013;156:851–9. criteria in the eye bank that correlate with cornea
https://doi.org/10.1016/j.ajo.2013.05.041. guttata in donor corneas. Klin Monbl Augenheilkd.
78. Tourtas T, Laaser K, Bachmann BO, Cursiefen 2021;238:680–7.
C, Kruse FE. Descemet membrane endothelial
38 L. Hamon et al.
89. Safi T, Seitz B, Berg K, Schulz K, Langenbucher A, surgeon prepared DMEK tissues: influence on
Daas L. Reproducibility of non-invasive endothelial adhesion and re-bubbling rate. Br J Ophthalmol.
cell loss assessment of the pre-stripped DMEK roll 2022;106:177–83. https://doi.org/10.1136/
after preparation and storage. Am J Ophthalmol. bjophthalmol-2020-317608.
2021;221:17–26. https://doi.org/10.1016/j. 100. Romano V, Pagano L, Gadhvi KA, Coco G,
ajo.2020.08.001. Titley M, Fenech MT, Ferrari S, Levis HJ, Parekh
90. Romano V, Parekh M, Kazaili A, Steger B, Akhtar M, Kaye S. Clinical outcomes of pre-loaded
R, Ferrari S, Kaye SB, Levis HJ. Eye bank versus ultra-thin DSAEK and pre-loaded DMEK. BMJ
surgeon prepared Descemet stripping automated Open Ophthalmol. 2020;5:e000546. https://doi.
endothelial keratoplasty tissues: influence on adhe- org/10.1136/bmjophth-2020-000546.
sion force in a pilot study. Indian J Ophthalmol. 101. Jirsova K, Jones GLA. Amniotic membrane in ophthal-
2022;70:523–8. https://doi.org/10.4103/ijo. mology: properties, preparation, storage and indications
IJO_3637_20. for grafting - a review. Cell Tissue Bank. 2017;18:193–
91. Rickmann A, Wahl S, Hofmann N, Knakowski J, 204. https://doi.org/10.1007/s10561-017-9618-5.
Haus A, Börgel M, Szurman P. Comparison of pre- 102. Dua HS, Gomes JA, King AJ, Maharajan VS. The
loaded grafts for Descemet membrane endothelial amniotic membrane in ophthalmology. Surv
keratoplasty (DMEK) in a novel preloaded transport Ophthalmol. 2004;49:51–77.
cartridge compared to conventional precut grafts. 103. Krysik K, Dobrowolski D, Wylegala E, Lyssek-
Cell Tissue Bank. 2020;21:205–13. https://doi. Boron A. Amniotic membrane as a main com-
org/10.1007/s10561-020-09814-7. ponent in treatments supporting healing and
92. Ighani M, Dzhaber D, Jain S, De Rojas JO, patch grafts in corneal melting and perforations.
Eghrari AO. Techniques, outcomes, and complica- J Ophthalmol. 2020;2020:4238919. https://doi.
tions of preloaded, trifolded Descemet Membrane org/10.1155/2020/4238919.
Endothelial Keratoplasty using the DMEK 104. European Directorate for the Quality of Medicines
EndoGlide. Cornea. 2021;40:669–74. https://doi. & HealthCare (EDQM). Guide to the quality and
org/10.1097/ICO.0000000000002648. safety of tissues and cells for human application, vol.
93. Català P, Vermeulen W, Rademakers T, van den 1. Strasbourg: Council of Europe; 2015. p. 349–55.
Bogaerdt A, Kruijt PJ, Nuijts RMMA, LaPointe 105. Burgos H, Sergeant RJ. Lyophilized human amniotic
VLS, Dickman MM. Transport and preservation membranes used in reconstruction of the ear. J R Soc
comparison of preloaded and prestripped-only Med. 1983;76:433.
DMEK grafts. Cornea. 2020;39:1407–14. https:// 106. Hodge C, Sutton G, Devasahayam R, Georges
doi.org/10.1097/ICO.0000000000002391. P, Treloggen J, Cooper S, Petsoglou C. The use
94. Terry MA. Precut tissue for descemet stripping auto- of donor scleral patch in ophthalmic surgery.
mated endothelial keratoplasty: complications are Cell Tissue Bank. 2017;18:119–28. https://doi.
from technique, not tissue. Cornea. 2008;27:627–9. org/10.1007/s10561-016-9603-4.
https://doi.org/10.1097/QAI.0b013e3181775e55. 107. Mehta JS, Franks WA. The sclera, the prion, and the
95. Menzel-Severing J, Walter P, Plum WJ, Kruse FE, Salla ophthalmologist. Br J Ophthalmol. 2002;86:587–92.
S. Assessment of corneal endothelium during contin- 108. Eye Bank Association of America (EBAA). Medical
ued organ culture of pre-stripped human donor tissue standards. Washington, DC: EBAA; 2020. https://
for DMEK surgery. Curr Eye Res. 2018;43:1439–44. restoresight.org/wp-content/uploads/2020/07/Med-
96. Bayyoud T, Röck D, Hofmann J, Bartz-Schmidt Standards-June-20-2020_7_23.pdf. Accessed 22
KU, Yoeruek E. [Precut technique for Descemet’s Feb 2022.
membrane endothelial keratoplasty, preparation and 109. Barza M, Baum JL, Kane A. Comparing radioactive
storage in organ culture]. Klin Monbl Augenheilkd. and trephine-disk bioassays of dicloxacillin and gen-
2012;229:621–3. tamicin in ocular tissues in vitro. Am J Ophthalmol.
97. Terry MA, Shamie N, Chen ES, Phillips PM, Hoar 1977;83:530–9.
KL, Friend DJ. Precut tissue for Descemet’s stripping 110. Antonios SR, Cameron JA, Badr IA, Habash NR,
automated endothelial keratoplasty: vision, astig- Cotter JB. Contamination of donor corneas: post-
matism, and endothelial survival. Ophthalmology. penetrating keratoplasty endophthalmitis. Cornea.
2009;116:248–56. https://doi.org/10.1016/j. 1991;10:217–20.
ophtha.2008.09.017. 111. Lau N, Hajjar Sesé A, Augustin VA, Kuit G, Wilkins
98. Pagano L, Gadhvi KA, Coco G, Fenech M, MR, Tourtas T, Kruse FE, Højgaard-Olsen K,
Titley M, Levis HJ, Ruzza A, Ferrari S, Kaye SB, Manuel R, Armitage WJ, Larkin DF, Tuft SJ. Fungal
Parekh M, Romano V. Rebubbling rate in pre- infection after endothelial keratoplasty: association
loaded versus surgeon prepared DSAEK. Eur J with hypothermic corneal storage. Br J Ophthalmol.
Ophthalmol. 2021:11206721211014380. https://doi. 2019;103:1487–90. https://doi.org/10.1136/
org/10.1177/11206721211014380. bjophthalmol-2018-312709.
99. Romano V, Kazaili A, Pagano L, Gadhvi KA, Titley 112. Zanetti E, Bruni A, Mucignat G, Camposampiero
M, Steger B, Fernández-Vega-Cueto L, Meana A, D, Frigo AC, Ponzin D. Bacterial contamina-
Merayo-Lloves J, Diego P, Akhtar R, Levis HJ, tion of human organ-cultured corneas. Cornea.
Ferrari S, Kaye SB, Parekh M. Eye bank versus 2005;24:603–7.
113. Laun D, Suffo S, Kramp K, Bischoff M, Huber 126. Kang SJ, Schmack I, Edelhauser HF, Grossniklaus
M, Langenbucher A, Seitz B. How implementing HE. Donor corneas misidentified with prior laser in
a quality management system at the LIONS Eye situ keratomileusis. Cornea. 2010;29:670–3. https://
Bank Saar-Lor-Lux, Trier/Western Palatinate from doi.org/10.1097/ICO.0b013e3181c325cc.
2006 to 2016 impacted the rate and reasons for dis- 127. Mootha VV, Dawson D, Kumar A, Gleiser J, Qualls
carding human organ-cultured corneas. Klin Monbl C, Albert DM. Slitlamp, specular, and light micro-
Augenheilkd. 2021;239:717–23. scopic findings of human donor corneas after laser-
114. Clover J. Slit-lamp biomicroscopy. Cornea. assisted in situ keratomileusis. Arch Ophthalmol.
2018;37(Suppl 1):5–6. https://doi.org/10.1097/ 2004;122:686–92. https://doi.org/10.1001/
ICO.0000000000001641. archopht.122.5.686.
115. Borderie VM, Scheer S, Touzeau O, Vedie F, 128. Vavra DE, Enzenauer RW. Predictive value of slit
Carvajal-Gonzalez S, Laroche L. Donor organ cul- lamp examinations in screening donor corneas
tured corneal tissue selection before penetrating ker- for prior refractive surgery. Arch Ophthalmol.
atoplasty. Br J Ophthalmol. 1998;82:382–8. https:// 2005;123:707–8. https://doi.org/10.1001/
doi.org/10.1136/bjo.82.4.382. archopht.123.5.707-c; author reply 708–709.
116. Langenbucher A, Seitz B, Nguyen NX, Naumann 129. Ousley PJ, Terry MA. Objective screening
GO. Corneal endothelial cell loss after nonme- methods for prior refractive surgery in donor
chanical penetrating keratoplasty depends on diag- tissue. Cornea. 2002;21:181–8. https://doi.
nosis: a regression analysis. Graefes Arch Clin org/10.1097/00003226-200203000-00011.
Exp Ophthalmol. 2002;240:387–92. https://doi. 130. Ousley PJ, Terry MA. Use of a portable topography
org/10.1007/s00417-002-0470-2. machine for screening donor tissue for prior refrac-
117. Bourne WM. Cellular changes in transplanted tive surgery. Cornea. 2002;21:745–50. https://doi.
human corneas. Cornea. 2001;20:560–9. https://doi. org/10.1097/00003226-200211000-00002.
org/10.1097/00003226-200108000-00002. 131. Stoiber J, Ruckhofer J, Hitzl W, Grabner G. Evaluation
118. Schaub F, Enders P, Adler W, Bachmann BO, of donor tissue with a new videokeratoscope: the
Cursiefen C, Heindl LM. Impact of donor graft qual- Keratron Scout. Cornea. 2001;20:859–63. https://
ity on deep anterior lamellar keratoplasty (DALK). doi.org/10.1097/00003226-200111000-00016.
BMC Ophthalmol. 2017;17:204. https://doi. 132. Terry MA, Ousley PJ. New screening methods for donor
org/10.1186/s12886-017-0600-6. eye-bank eyes. Cornea. 1999;18(4):430–6. https://
119. Shaw EI, Rao GN, Arthur EJ, Aquavella VJ. The doi.org/10.1097/00003226-199907000-00007.
functional reserve of corneal endothelium. Trans 133. Lin RC, Li Y, Tang M, McLain M, Rollins AM, Izatt
Ophthalmol. 1978;85:640–9. JA, Huang D. Screening for previous refractive sur-
120. Sperling S. Evaluation of the endothelium of human gery in eye bank corneas by using optical coherence
donor corneas by induced dilation of intracellu- tomography. Cornea. 2007;26:594–9. https://doi.
lar spaces and trypan blue. Graefes Arch Clin Exp org/10.1097/ICO.0b013e31803c5535.
Ophthalmol. 1986;224:428–34. 134. Priglinger SG, Neubauer AS, May CA, Alge
121. Schroeter J, Rieck P. Endothelial evaluation in the CS, Wolf AH, Mueller A, Ludwig K, Kampik A,
Cornea Bank. In: Bredehorn-Mayr T, Duncker GIW, Welge-Luessen U. Optical coherence tomography
Armitage WJ, editors. Eye banking. Developments for the detection of laser in situ keratomileusis in
in ophthalmology, vol. 43. Basel: Karger; 2009. donor corneas. Cornea. 2003;22:46–50. https://doi.
p. 47–62. org/10.1097/00003226-200301000-00011.
122. Fargione RA, Channa P. Cornea donors who have 135. Damian A, Seitz B, Langenbucher A, Eppig
had prior refractive surgery; data from the Eye Bank T. Optical coherence tomography-based topography
Association of America. Curr Opin Ophthalmol. determination of corneal grafts in eye bank culti-
2016;27:323–6. https://doi.org/10.1097/ vation. J Biomed Opt. 2017;22:16001. https://doi.
ICU.0000000000000278. org/10.1117/1.JBO.22.1.016001.
123. de Mello Farias RJ, Parolim A, de Sousa LB. 136. Janunts E, Langenbucher A, Seitz B. In vitro cor-
[Corneal transplant utilizing a corneal graft that had neal tomography of donor cornea using anterior
undergone laser surgery - case report]. Arq Bras segment OCT. Cornea. 2016;35:647–53. https://doi.
Oftalmol. 2005;68:266–9. https://doi.org/10.1590/ org/10.1097/ICO.0000000000000761.
s0004-27492005000200022. 137. Mäurer S, Asi F, Rawer A, Damian A, Seitz B,
124. Michaeli-Cohen A, Lambert AC, Coloma F, Langenbucher A, Eppig T. [Concept for 3D mea-
Rootman DS. Two cases of a penetrating kerato- surement of corneal donor tissue using a clinical
plasty with tissue from a donor who had undergone OCT]. Ophthalmologe. 2019;116:640–6. https://doi.
LASIK surgery. Cornea. 2002;21:111–3. https://doi. org/10.1007/s00347-018-0801-2.
org/10.1097/00003226-200201000-00023. 138. Tomey GmbH. CASIA2. 2022. https://assets.
125. Mifflin M, Kim M. Penetrating keratoplasty website-files.com/60c345b08518d0bd95024abf/60
using tissue from a donor with previous LASIK ddc04f9c8b48347261217a_CASIA2_brochure.pdf.
surgery. Cornea. 2002;21:537–8. https://doi. Accessed 3 Feb 2022.
org/10.1097/00003226-200207000-00024; author 139. Eppig T, Mäurer S, Daas L, Seitz B, Langenbucher
reply 538–9. A. Imaging the cornea, anterior chamber and lens in
40 L. Hamon et al.
corneal and refractive surgery. In: Lanza M, e ditor. factors for cornea guttata in the Reykjavik Eye
Clinical application of OCT in ophthalmology. Study. Ophthalmology. 2006;113:565–9. https://doi.
London: IntechOpen; 2018. p. 133–52. https://doi. org/10.1016/j.ophtha.2005.12.014.
org/10.5772/intechopen.74129. 155. Giasson CJ, Solomon LD, Polse KA. Morphometry
140. Quintin A, Hamon L, Mäurer S, Langenbucher A, of corneal endothelium in patients with corneal gut-
Seitz B. OCT application for sterile cor- tata. Ophthalmology. 2007;114:1469–75.
neal graft screening in the eye bank. Monbl 156. McCarey BE, Edelhauser HF, Lynn MJ. Review of
Augenheilkd. 2021;238:688–92. https://doi. corneal endothelial specular microscopy for FDA
org/10.1055/a-1443-5451. clinical trials of refractive procedures, surgical
141. Seitz B, Asi F, Mäurer S, Hamon L, Quintin A, devices and new intraocular drugs and solutions.
Langenbucher A. Anterior segment OCT: applica- Cornea. 2008;27:1–16.
tion to improve graft selection for corneal transplan- 157. Schmitz LM, Safi T, Munteanu C, Seitz B, Daas
tation. In: Alió JL, Alió del Barrio JL, editors. Atlas L. Prevalence and severity of cornea guttata in the
of anterior segment optical coherence tomography. graft following Descemet Membrane Endothelial
Cham: Springer; 2021. p. 223–36. Keratoplasty (DMEK). Acta Ophthalmol.
142. Mäurer S, Seitz B, Langenbucher A. “Harmonization” 2022;100(8):e1737–45. https://doi.org/10.1111/
of donor and recipient tomography in corneal trans- aos.15195.
plantation. Z Med Phys. 2021;31:73–7. https://doi. 158. Schönit S, Maamri A, Zemova E, Munteanu C, Safi
org/10.1016/j.zemedi.2020.05.006. T, Daas L, Seitz B. Prevalence and impact of cor-
143. Seitz B, Langenbucher A, Naumann nea guttata in the graft following penetrating kerato-
GOH. Astigmatismus bei Keratoplastik. In: Seiler T, plasty in Germany. Cornea. 2022;41(12):1495–502.
editor. Refraktive Chirurgie der Hornhaut. Stuttgart: https://doi.org/10.1097/ICO.0000000000002971.
Enke; 2000. p. 197–252. 159. Nuzzi R, Boscia G, Marolo P, Ricardi F. The impact
144. Quintin A, Hamon L, Mäurer S, Langenbucher A, of artificial intelligence and deep learning in eye dis-
Seitz B. [Comparison of sterile donor tomography in eases: a review. Front Med. 2021;8:710329. https://
the eye bank and graft tomography after penetrating doi.org/10.3389/fmed.2021.710329.
keratoplasty]. Ophthalmologe 2021;118:1038–44. 160. Pels E, Rijneveld WJ. Organ culture preservation
https://doi.org/10.1007/s00347-020-01256-6. for corneal tissue. Technical and quality aspects.
145. Son H-S, Villarreal G, Meng H, Eberhart CG, Jun Dev Ophthalmol. 2009;43:31–46. https://doi.
AS. On the origin of ‘guttae’. Br J Ophthalmol. org/10.1159/000223837.
2014;98:1308–10. 161. Kramp K, Suffo S, Laun D, Bischoff-Jung M, Huber
146. Vogt A. [Other results of slit lamp microscopy of the M, Langenbucher A, Seitz B. Analysis of factors
anterior segment of the bulb: (cornea, anterior cham- influencing the suitability of donor corneas in the
ber, iris, lens, anterior vitreous, conjunctiva, eyelid LIONS Cornea Bank Saar-Lor-Lux, Trier/Westpfalz
margins.) I. Section: cornea]. Graefes Arch Clin Exp from 2006 to 2016. Klin Monbl Augenheilkd.
Ophthalmol. 1921;106:63–103. 2020;237:1334–42.
147. Waring GO III, Rodrigues MM, Laibson PR. Corneal 162. Eye Bank Association of America (EBAA). 2019
dystrophies. II. Endothelial dystrophies. Surv eye bank statistical report. Washington, DC:
Ophthalmol. 1978;23:147–68. EBAA; 2019. https://restoresight.org/wp-content/
148. Adamis AP, Filatov V, Tripathi BJ, Tripathi uploads/2020/04/2019-EBAA-Stat-Report-FINAL.
RC. Fuchs’ endo thelial dystrophy of the cornea. pdf. Accessed 24 Feb 2022.
Surv Ophthalmol. 1993;38:149–68. 163. Garcin T, Gauthier AS, Crouzet E, He Z, Herbepin
149. Weiss JS, Møller HU, Aldave AJ, Seitz B, Bredrup P, Perrache C, Acquart S, Cognasse F, Forest F,
C, Kivelä T, Munier FL, Rapuano CJ, Nischal KK, Thuret G, Gain P. Innovative corneal active stor-
Kim EK, Sutphin J, Busin M, Labbé A, Kenyon KR, age machine for long-term eye banking. Am
Kinoshita S, Lisch W. IC3D classification of corneal J Transplant. 2019;19:1641–51. https://doi.
dystrophies – edition 2. Cornea. 2015;34:117–59. org/10.1111/ajt.15238.
150. Borboli S, Colby K. Mechanisms of disease: Fuchs’ 164. Garcin T, Gauthier AS, Crouzet E, He Z, Herbepin
endothelial dystrophy. Ophthalmol Clin North Am. P, Perrache C, Acquart S, Cognasse F, Forest F, Gain
2002;15:17–25. P, Thuret G. Three-month storage of human cor-
151. Wilson SE, Bourne WM. Fuchs’ dystrophy. Cornea. neas in an active storage machine. Transplantation.
1988;7:2–18. 2020;104:1159–65. https://doi.org/10.1097/
152. Yee RW, Matsuda M, Schultz RO, et al. Changes in TP.0000000000003109.
the normal corneal endothelial cellular pattern as a 165. Batista A, Breunig HG, König A, Schindele A, Hager
function of age. Curr Eye Res. 1985;4:671–8. T, Seitz B, Morgado AM, König K. Assessment of
153. Eghrari AO, Gottsch JD. Fuchs’ corneal dystrophy. human corneas prior to transplantation using high-
Expert Rev Ophthalmol. 2010;5:147–59. resolution two-photon imaging. Invest Ophthalmol
154. Zoega GM, Fujisawa A, Sasaki H, Kubota A, Sasaki Vis Sci. 2018;59:176–84. https://doi.org/10.1167/
K, Kitagawa K, Jonasson F. Prevalence and risk iovs.17-22002.
Risk Classification
and Management of Corneal
3
Grafts, Human Leukocyte Antigen
Matching, and Options
for Immunosuppression Therapy
Paula W. Feng and Guillermo Amescua
Key Points been considered to have immune privilege, and

This chapter discusses: although survival rates vary significantly depend-
ing on associated risk factors, low-risk trans-
• Risk factors affecting graft success in corneal plants generally portend a high allograft survival
transplantation rate. Proposed immune mechanisms responsible
• The role of lymphangiogenesis in mediating for this immune privilege include the cornea’s
graft rejection and associated mitigation relative absence of corneal lymph and blood ves-
strategies sels, relative lack of lymphatic drainage, dimin-
• The role of Human leukocyte antigen (HLA) ished T-cell proliferation, and natural killer cell
matching in donor selection and graft activation, allograft inducement of regulatory
survival T-cells which inhibit the function and induction
• Topical and systemic strategies for prevention of alloimmune T-cells, protection from
of corneal graft rejection complement- mediated cytolysis, and induced
• Special considerations for immunosuppres- apoptosis of neutrophils and T-cells at the graft-
sion limbal stem cell transplantation host interface [3, 4]. During implantation of cor-
neal allografts, endothelial cells from allografts
are theoretically sloughed off, enter the anterior
Introduction chamber, and while there exposed to anti-
inflammatory and immunosuppressive mole-
Corneal transplantation is the most common cules. Antigen-specific suppression of
solid tissue transplant performed in humans. delayed-type hypersensitivity responses, exclu-
Over 180,000 transplants are performed annually sion of complement-fixing antibodies, and pref-
worldwide [1, 2]. The cornea has traditionally erential production of noncomplement-fixing
antibody isotypes result from this exposure, pro-
moting graft survival via a spectrum of immune
Supplementary Information The online version con- responses known as anterior chamber autoim-
tains supplementary material available at https://doi. mune deviation (ACAID) [4].
org/10.1007/978-3-031-32408-6_3.
However, immune privilege alone is insuffi-
cient to completely prevent corneal transplant
P. W. Feng · G. Amescua (*) rejection. Certain host risk factors increase the
Bascom Palmer Eye Institute, University of Miami, risk profile for corneal transplant rejection. For
Miami, FL, USA example, patients with ongoing ocular inflam-
e-mail: [email protected]

https://doi.org/10.1007/978-3-031-32408-6_3
42 P. W. Feng and G. Amescua
mation, preexisting corneal neovascularization, Table 3.1 Risk factors for corneal transplantation
according to indication for transplantation and patient
or a history of previous corneal graft rejection,
history
among other risk factors, experience loss of
Minor Major
immune privilege and have a high rate of
Fuchs dystrophy Infectious leukoma
immune rejection [4–6]. In the setting of ongo- Keratoconus, corneal ectasia Chemical burn
ing inflammation, vascularization of the corneal Pseudophakic and aphakic Trauma
bed alters multiple immune aspects, including corneal edema
lymphangiogenesis, antigen presentation, and HSV/VZV
anterior chamber-associated immune deviation. Previous
transplantation
In solid organs, allospecific primed effector
Active inflammation
T-cells are recruited into the site of transplanta- Atopy
tion, a critical step in the rejection cascade. Corneal
Early inflammation after surgery of vascular- neovascularization
ized organs causes more production of chemo-
kines and late production of T-cell
chemoattractants [7]. Among patients with such Primary Diagnosis
risk factors, the risk of acute rejection or graft
failure can be similar to that seen in solid organ One large retrospective study of 895 penetrating
transplants. High-risk grafts have poor survival keratoplasties found that patients with keratoco-
despite the use of systemic and topical immuno- nus had the best 10-year survival outcome (95%),
suppressive therapy [7, 8]. In the Collaborative followed by endothelial and stromal dystrophies
Corneal Transplantation Study (CCTS), corneas (55%), infectious leukomas (49%), trauma
with two or more quadrants of deep stromal vas- (33%), and chemical burns (14%) (Table 3.1)
cularization or previous graft rejection were [13]. Given these widely varying results, it should
considered to be “high-risk” transplants [9]. In be emphasized that graft survival varies widely
addition, we consider grafts in eyes with a his- based on the indication for transplantation. Other
tory of HSV, VZV, uveitis, or atopy to be high- large cohorts of penetrating keratoplasties have
risk (see “Risk Factors” section below). Optimal found similar results, with the best survival for
management strategies for mitigating the risk of keratoconus and endothelial dystrophies and
graft rejection must be taken into consideration worse survival for patients with a history of ante-
in light of each patient’s risk profile. In this rior segment inflammation, herpetic infection,
chapter, we discuss the risk stratification of cor- corneal neovascularization, and raised intraocu-
neal transplants and consider strategies for lar pressure [5, 10, 12, 14].
immune suppression.
Previous Transplantation
Risk Stratification
This same large retrospective study demonstrated
Risk Factors that the 10-year survival rate for primary pene-
trating keratoplasties was highest at 81%, fol-
Risk factors for an immune rejection or graft fail- lowed by second grafts (33%), followed by third
ure include primary diagnosis, increasing quad- or more grafts (16%) [13]. The comparative
rants of host neovascularization, previous immune privilege of the cornea and anterior
glaucoma, ocular inflammation or history of uve- chamber decrease with successive corneal grafts
itis, herpetic infection, and previous graft fail- [4]. Pathogenic mechanisms include sensitization
ures, among others [10–12]. due to previous grafts and greater previous expo-
3 Risk Classification and Management of Corneal Grafts, Human Leukocyte Antigen Matching… 43
sure of the recipient immune system to cornea- glaucoma, iridocorneal touch may cause endo-
derived antigens, a heightened alloimmune thelial trauma, disrupt the aqueous flow, induce
response, and subsequent loss of immunological hypoxia, and affect nutritional support to the
privilege, preceding recipient corneal neovascu- endothelium [21–24]. It has additionally been
larization, and lymphangiogenesis, loss of avas- hypothesized that toxicity from the chronic use
cularity, along with the higher likelihood of the of topical glaucoma medication and congenital
presence of other risk factors predisposing to alteration of the endothelium and trabecular
graft failure [15, 16]. See the sections below on meshwork may contribute to endothelial cell
active inflammation, the role of lymphangiogen- loss at baseline in glaucoma patients [21].
esis, and anti-angiogenic and lymphangiogenic Additionally, glaucoma drainage implantation or
activity. filtration surgery is associated with significant
loss of endothelial density. Studies have reported
between 3 and 23% endothelial cell loss over 3
Glaucoma months to 3 years after glaucoma drainage
implantations or filtration surgery [21]. The less
The Cornea Donor Study was a multicenter pro- the distance between a glaucoma shunt tube and
spective, double-masked, controlled clinical trial the corneal endothelium, the greater the rate of
conducted at 80 clinical sites which followed endothelial cell attrition [25, 26]. Altered aque-
1090 moderate-risk corneal transplants, princi- ous flow in the setting of glaucoma drainage
pally performed for pseudophakic or aphakic implantation or trabeculectomy may theoreti-
corneal edema or Fuchs dystrophy. History of cally also affect the nutritional support of the
glaucoma before penetrating keratoplasty was a endothelium.
significant risk factor for graft failure within 10
years. In one study, 58% of patients with previ-
ous glaucoma surgery and 22% without a history Active Inflammation
of glaucoma surgery or medication sustained
graft failure within 10 years [6]. In the Australian In patients who subsequently developed graft
Corneal Graft Registry (ACGR), a study of 1035 failure, histopathologic analysis of excised host
penetrating corneal grafts in Australia, elevated corneas showed higher counts of leukocyte-
intraocular pressure was a common cause of graft common antigen, class II major histocompatibil-
failure [10]. In the Collaborative Corneal ity complex antigens, myeloid-lineage markers,
Transplantation Studies, glaucoma was also asso- and peripheral T-cell markers. Higher leukocyte
ciated with an elevated chance of graft failure counts in recipient graft beds were correlated
[17]. Glaucoma and ocular inflammation were with subsequent graft failure [27]. In the
both found to be independent risk factors for Australian Corneal Graft Registry, patients with
graft failure at 1 year in multivariate analysis by inflammation at the site of the graft or in the past
Tourkmani et al. as well [12]. at the graft site had a 9.6 relative risk of graft fail-
Mechanisms for the association between ure compared to those who had never been
glaucoma and graft failure are multifactorial. inflamed (P < 0.001) [10]. Lymphatic vessels are
Elevated intraocular pressure after penetrating thought to regress earlier than blood vessels after
keratoplasty has been associated with acceler- acute inflammatory insults, suggesting that
ated chronic endothelial cell loss [18]. Direct delaying transplantation after the resolution of
endothelial compression due to elevated intra- inflammation may be of benefit [28]. In ideal cir-
ocular pressure has been proposed as a cause of cumstances, corneal transplantation should be
endothelial cell loss [19–21]. In addition, in avoided when the recipient eyes are actively
patients with acute or chronic angle closure inflamed [4].
Herpetic Infection Atopy
Past Herpes simplex eye and Herpes zoster dis- Patients with a history of atopy have a higher risk
ease are major risk factors for graft failure, and of rejection, perhaps due to the loss of immune
patients are at risk of disease reactivation after privilege related to corneal inflammation at or
transplantation. In addition to having a higher before the time of transplantation unrelated to the
rate of graft rejection, patients with a history of allergy itself [ 4, 37–40]. Given the frequent co-
herpetic eye disease also demonstrate a higher occurrence of atopy with keratoconus, this is an
rate of postkeratoplasty corneal neovasculariza- especially important consideration in this group
tion and corneal epithelial defects [29, 30]. In [41]. Patients with a history of atopic dermatitis
addition, immunosuppression with topical and have a higher rate of graft rejection, suggesting
oral steroids and steroid-sparing immunosup- that systemic atopy may contribute to graft rejec-
pressive therapy may lead to the reactivation of tion [38]. In mouse models, exposure to airway
herpetic eye disease [31, 32]. Long-term main- allergens increases the likelihood of allograft
tenance high doses of oral antivirals for graft rejection [42]. Mouse models have also demon-
failure prophylaxis postkeratoplasty is recom- strated accelerated rejection of corneal trans-
mended [29, 31, 33, 34]. In patients who do not plants in mice with active allergic conjunctivitis
receive antiviral prophylaxis, more than half of at the time of transplantation [43]. Postkeratoplasty
postpenetrating keratoplasty patients develop atopic sclerokeratitis can potentially have severe
recurrence of herpetic stromal keratitis or graft consequences, including loosening of running
rejection. Long-term antiviral therapy has been sutures and wound leakage, microbial suture
shown to reduce the rate of graft failure at 5 abscess, persistent epithelial defects, and corneal
years by more than two-thirds compared to no perforation [44, 45]. Atopic rhinitis has been
prophylaxis [29, 35]. associated with spontaneous wound dehiscence
after keratoplasty [46]. Although the reasons for
this association are unclear, eye rubbing could be
Neovascularization a potential contributor. Careful pre- and postop-
erative immunosuppression as well as control of
Increasing quadrants of stromal neovasculariza- atopic meibomian gland dysfunction and blepha-
tion due to lymphangiogenesis and hemangio- ritis with eyelid hygiene, warm compresses, and
genesis is associated with an increased likelihood oral tetracyclines may lead to success in these
of graft failure. In the ACGR, neovascularization cases [41, 47, 48]. Because of the risk of post-
of the graft was associated with a 6.8 relative risk keroplasty atopic sclerokeratitis, patients should
of graft failure compared to no neovasculariza- be closely monitored for control of atopy both
tion (P < 0.001) [10]. This causes loss of the before and after keratoplasty.
angiogenic privilege state of the host cornea.
Blood vessels carry immune effector cells into
the graft, and lymphatic vessels carry graft anti- Aphakia
gens that interact with host T-cells [16]. That
removal of lymph nodes in mice enhances cor- Data conflict on whether aphakia affects graft
neal graft survival suggests a role of lymphangio- survival. The Australian Corneal Graft Registry
genesis in mediating graft rejection [36]. See found that aphakia before or after penetrating
“Role of Lymphangiogenesis” and “Anti- keratoplasty was associated with decreased sur-
angiogenic and Lymphangiogenic Therapy” vival [10]. However, no association between
section. aphakia and graft survival was found in the
American Cornea Donor Study [6]. A retrospec- long-term success rates of penetrating kerato-
tive study of penetrating keratoplasty found that plasty have been reported [51, 52, 54]. In one
aphakia was a significant risk factor for rejection such series, the majority of eyes developed glau-
and graft failure in univariate, but not multivari- coma or ocular hypertension, and only half of the
ate analysis, after accounting for other risk fac- patients achieved any visual acuity gain after a
tors [12]. mean follow-up time of 24 months [52]. Some
suggest that lamellar keratoplasty after kerato-
limbal allograft transplantation may have more
Age success than penetrating keratoplasty, but avail-
able data are limited to small case series [54]. In
The Cornea Donor Study found an increased risk eyes with only partial limbal stem cell deficiency,
of graft failure in patient recipients over 70 com- keratoplasty may be carefully considered on a
pared to under 60 years of age, though the effect case-by-case basis.
was moderate (29% vs. 19%, P = 0.04) [6]. The
ACGR corroborated this trend; recipients over 50
years of age had a 3.0 relative risk of graft failure Additional Factors
compared to those less than or equal to 50 years
of age at transplantation (P < 0.001) [10]. History of smoking was found to be associated
with increased risk, as was African American
background [6]. Most studies on graft failure are
Limbal Stem Cell Deficiency limited by short-term outcomes. One study exam-
ined at 500 consecutive PKs by a single surgeon.
In eyes with total limbal stem cell deficiency, sec- Over 20 years, the most common reason for graft
ondary keratoplasty after keratolimbal allograft failure over the long term was progressive endo-
is associated with poor outcomes [16, 49–52]. A thelial cell loss and low endothelial cell count [5].
retrospective, noncomparative case series com-
pared the long-term outcome of keratolimbal
allograft and amniotic membrane transplantation itigation Strategies for Corneal
M
with or without subsequent secondary penetrat- Graft Rejection
ing keratoplasty. Keratolimbal allograft with
amniotic membrane alone had significantly better Role of Lymphangiogenesis
visual outcomes than keratolimbal allograft with
secondary penetrating keratoplasty; at 2 years, Lymphangiogenesis is a crucial role in the patho-
86% versus 47% retained ambulatory vision in genesis of graft rejection, as the outgrowth of
these groups, respectively [50, 53]. In the same new lymphatic vessels enables and hastens the
study, central corneal graft survival was only exit of antigen-presenting cells, immunomodula-
14% at 3 years in the secondary penetrating kera- tory cytokines, and memory T-cells from the
toplasty group. Given the poor prognosis and graft to regional lymph nodes, leading to alloim-
less-than-ideal visual outcomes, penetrating ker- munization and subsequent rejection [55]. The
atoplasty should be avoided in cases of total lim- significance of lymphangiogenesis is demon-
bal stem cell deficiency even after a successful strated by the fact that lymphadenectomy in
keratolimbal allograft. Although some studies mouse models of high-risk corneal transplanta-
have advocated waiting at least 9–12 months tion results in increased survival of grafts.
between keratolimbal allograft transplantation However, lymph node removal was ineffective in
and secondary keratoplasty, similarly dismal mice who had previous transplantation preopera-
tively, likely because they had already been allo- treatment group; confounding cannot be
sensitized by the previous graft [36, 56]. However, excluded.
such studies have not been replicated in humans, A multicenter, prospective, randomized,
and this practice is uncommonly performed out- placebo-controlled clinical trial examining the
side of experimental models. Molecular targets safety and efficacy of bevacizumab in high-risk
for inhibition of lymphangiogenesis, such as corneal transplantation is currently underway.
VEGF-receptor 3, are currently under investiga- Preliminary data show that subconjunctival injec-
tion [55]. tion of 2.5 mg/0.1 mL bevacizumab at the time of
surgery and topical 1% bevacizumab four times
daily for 4 weeks postoperatively has a modest
Anti-angiogenic effect on reducing endothelial rejection and
and Lymphangiogenic Therapy improving overall graft failure compared to pla-
cebo over 52 weeks of follow-up. However, the
Topical and subconjunctival bevacizumab effec- results did not reach significance in the small
tively reduces corneal neovascularization and sample size (n = 43) presented [66].
causes regression of corneal vessels [57, 58].
In mouse models of corneal transplantation,
the application of anti-vascular endothelial Histocompatibility Matching
growth factor agents to corneal grafts preopera-
tively as well as postoperatively, or into recipient Human leukocyte antigen (HLA) matching has
host beds, reduces hemangiognesis as well as been shown to be effective in reducing the rejec-
lymphangiognesis, recruitment of mononuclear tion rate and improving survival in solid organ
phagocytes into the graft, dendritic cell traffick- transplants [67, 68]. Modern techniques for HLA
ing to draining lymph nodes, induction of delayed typing are reliable, enabling HLA typing and
hypersensitivity reactions and improves graft sur- matching among donors and recipients. However,
vival [59–64]. there is no consensus on the cost efficacy and
In humans, one small retrospective case- benefit of HLA matching in corneal graft sur-
control study of 37 eyes with preoperative cor- vival, and the technique remains relatively
neal neovascularization covering at least underutilized.
one-quarter of the corneal surface compared The American Collaborative Corneal
patients who underwent 3 months of preoperative Transplantation Studies was a double-masked
subconjunctival and/or intrastromal 5 mg/0.2 mL antigen masking study on the basis of HLA-A,
bevacizumab injections at the limbus before pen- -B, and -DR antigen matches conducted in the
etrating keratoplasty or deep anterior lamellar 1990s. The study did not demonstrate the benefit
keratoplasty with patients who did not receive of HLA matching [9]. However, the success of
treatment. One-third of patients receiving bevaci- this study may have been limited by the poorer
zumab treatment were able to completely avoid reliability of more primitive typing techniques; in
transplantation due to regression of the patho- 55% of cases, serology-based tissue typing in this
logic process. The remaining patients who study differed from the results of molecular tech-
received bevacizumab preoperatively and under- nique typing [15]. When recipients enrolled in
went transplantation had a higher graft survival this trial were later retyped as part of a quality
rate (80%) compared to controls (64%). However, assurance program, the agreement between the
this improvement did not reach significance in original and retyping laboratories was poor, espe-
this small sample (P = 0.43) [65]. Complicating cially for HLA-DR, which had structural simi-
the interpretation of the results of this study, 80% larities among DR3, DR5, and DR6 [69].
of the failures in the bevacizumab group had ele- Subsequently, the Corneal Transplant Follow-up
vated intraocular pressure, which could have arti- Study (CTFS) examined 2777 corneal grafts in
ficially elevated the potential rejection rate in the the United Kingdom from 1987 to 1991 among
400 surgeons and yielded conflicting results. 17,000 endothelial, lamellar, and penetrating ker-
Poor HLA class I antigen matching was associ- atoplasties performed for moderate- and low-risk
ated with an increased risk of rejection, but mis- indications including Fuchs, keratoconus, infec-
matched HLA-DR grafts had improved rejection tions, PBK, posttraumatic scars in inherited dys-
rates over those with no mismatched HLA-DR trophies, demonstrated that gender-matching
grafts. The authors concluded that HLA-DR mis- between donors and recipients increased graft
matching is associated with minimal harm at survival between 20 and 40% compared to
worst and should not be considered [70]. A later nongender-matched donor-recipient pairs, likely
study suggested the utility of HLA-DR typing on on the basis of H-Y matching [77].
1681 transplants performed by a single surgeon As another barrier to widespread adoption, the
based on a typing performed by a single experi- wait time for HLA-matched corneal grafts may
enced laboratory and suggested that the negative be logistically prohibitive for at least some trans-
effect in the previous study was the result of labo- plants, including high-risk transplants necessitat-
ratory technique and the accuracy of HLA typing urgent intervention, which may most benefit
ing. Simulations demonstrated that errors in as from HLA matching [78]. A simulated model of
few as 5% of tissue types could sufficiently wait times based on data from the Cornea Lions
reduce the efficacy of HLA-DR matching [71]. Bank in Germany suggested that waiting for zero
Given the previous studies demonstrating poor mismatches would result in an estimated average
reliability of typing techniques, some authors wait time of 17 ± 14 months, whereas waiting for
suggested retyping and verifying typing results at as many as two mismatches would result in a wait
least once after initial testing [69]. time of 1 ± 3 months [78]. With increasing num-
Evidence suggests that HLA matching based bers of corneal transplants performed each year,
on modern, reliable HLA typing techniques is low donor availability and high wait times can be
likely to be of benefit, especially in high-risk expected to increase [79]. An algorithm has been
grafts [71–73]. Additionally, efficacy has been proposed that can help balance waiting times and
demonstrated in HLA typing in limbal allogeneic histocompatibility for patients with rare HLA
stem cell transplantation. One prospective study phenotypes [80]. Since it is thought that HLA
of 12 eyes undergoing allogenic conjunctival typing improves long-term graft survival, HLA
transplantation found that eyes with incompatible typing of major and minor histocompatibility
HLA donor-recipient pairs were most likely to antigens could conceivably become a highly
experience a rejection episode than those with cost-effective strategy by reducing the need for
matched HLA allotypes [74]. A retrospective, care of corneal graft rejections and repeat trans-
noncomparative case series of nine living-related plantation [81].
donors, eight recipients (ten eyes) with the ocular
surface disease found success with HLA-matched
conjunctival limbal allografts, and the only two Strategies for Immunosuppression
cases that experienced allograft rejection had two
class I HLA mismatches [54]. The Cornea Topical Therapy
Transplant Follow-Up Study II (CTFS II) is a
prospective, longitudinal clinical trial on the effi- Steroids
cacy of HLA class II typing in improving the suc- Topical steroids are a mainstay of immunosup-
cess of corneal transplantation which has pressive therapy after corneal transplantation.
completed enrollment and is currently underway The overwhelming majority of practitioners use
[75]. topical steroids for at least 6 months postopera-
In addition to the matching of major HLA tively, which likely explains the low rate of graft
antigens above, evidence has shown that match- rejection during this period [82, 83]. In addition,
ing of minor H-Y antigens is likely of benefit a survey by United States-based Cornea Society
[76]. Indeed, a large retrospective study of over revealed that 75% of cornea specialists who
responded to the survey used intraoperative sub- analysis that used cyclosporine A 0.05%
conjunctival corticosteroids for avascular corneas (Restasis, Allergan), there was no improvement
receiving penetrating keratoplasty, and 68% said on graft survival or rates of rejection [90, 91].
they did so for endothelial keratoplasty [83]. The When cyclosporine 0.05% was substituted for
overwhelming majority of cornea specialists sur- prednisolone after 13-weeks posttransplant in
veyed preferred prednisolone, likely at least in low-risk transplants, there was a higher rate of
part due to its ubiquitous availability in the rejection and rejection occurred earlier.
United States, and a minority used fluoroethyl- According to the Cornea Society survey, about
ene, loteprednol, or difluorinated. The latter was 13–14% of respondents use cyclosporine routinely
most commonly used in high-risk eyes, and for penetrating and endothelial keratoplasty,
loteprednol use increased after 6 months postop- respectively. This number increases to 48% and
eratively, likely to reduce the effect of intraocular 40% in high-risk penetrating and endothelial kera-
pressure [83]. In addition, long-term usage of toplasties, respectively [83]. Higher doses of topi-
topical steroids significantly decreases the risk of cal cyclosporine A such as cyclosporine A 2%
graft rejection and increases the cumulative sur- over 0.05% may be necessary to achieve the
vival rate of corneal grafts without any episode of desired immunosuppressive effect for graft rejec-
endothelial allograft rejection [84, 85]. However, tion prophylaxis. Based on current randomized
risks of long-term topical steroid use include the clinical trials and meta-analyses, there is insuffi-
development of cataracts and glaucoma, the latter cient evidence to support the usage of topical
of which is a risk factor for graft failure [86]. cyclosporine A at any concentration in resolving
Patients should be monitored long-term for the or reversing acute rejection episodes [87, 89, 91].
development of late graft rejection as well as
treatment side effects from prolonged use of anti- Tacrolimus
inflammatory medication. Tacrolimus (FK506) is a macrolide antibiotic. Its
mechanism of action is in suppression calcineu-
Cyclosporine rin, which, like cyclosporine A, then inhibits
Cyclosporine binds cyclophilin, which inhibits T-cell activation and subsequent T-cell signal
IL-2 transcription, which suppresses T-cell acti- transduction and IL-2 transcription, affecting the
vation and subsequent graft rejection [87]. downstream release of cytokines such as interleu-
Topical cyclosporine A 0.05% and 0.09% is com- kins, TNF-alpha, interferon-gamma [16, 93]. In
monly available for the treatment of dry eye, pos- addition, topical and systemic tacrolimus effec-
terior blepharitis, ocular rosacea, atopic tively cause regression of neovascularization and
keratoconjunctivitis, graft-versus-host disease, decreased VEGF expression in neovascularized
and herpetic stromal keratitis, among other uses murine corneas [94]. Topical tacrolimus effec-
[88]. Higher dosages such as cyclosporine A 2% tively penetrates the cornea into the aqueous
and 5% have been utilized for immunosuppres- humor [93, 95]. Topical tacrolimus 0.03% is
sion after corneal transplantation. A recent meta- FDA-approved for atopic dermatitis and has been
analysis of four small randomized clinical trials used off-label, and has been shown to be safe and
and a retrospective cohort study examined sup- effective in a variety of ophthalmic conditions
plemental topical cyclosporine treatment in addi- such as atopic and vernal keratoconjunctivitis
tion to postoperative topical corticosteroids in [96–98], necrotizing and nodular scleritis [99],
penetrating keratoplasty. Supplemental cyclospo- Mooren ulcer [99], keratoconjunctivitis sicca due
rine led to an overall benefit in rejection-free to Sjogren’s syndrome [100, 101], ocular cicatri-
graft survival at 12-months and 24-months post- cial pemphigoid [99], Stevens-Johnson syndrome
operatively in both high-risk and normal popula- [99], graft-versus-host disease [102], and supe-
tions of penetrating keratoplasties [87, 89–92]. rior limbic keratoconjunctivitis [93]. There have
Most studies in the meta-analysis utilized cyclo- been no adverse effects reported with long-term
sporine A 2%. In the two studies in this meta- use [103–105].
Topical 0.03% tacrolimus was effective in pre- ticosteroids. The tacrolimus group had a 0%
venting corneal graft rejection in a retrospective rejection rate at 12 months, compared to 16% in
cohort of 72 high-risk eyes which had undergone the control group, a small difference that did not
more than one penetrating keratoplasty in the reach significance in this small study (P = 0.9,
same eye or had a severe chemical burn (Video n = 40). Larger series may be necessary to ade-
3.1). The patients receiving topical tacrolimus quately assess the efficacy of topical tacrolimus
0.03% twice daily had half the graft rejection rate in the immunosuppression of normal-risk grafts,
compared to topical prednisolone alone [106]. given the high graft success rates in this
Compared to cyclosporine, there is greater population.
among of and stronger evidence for the usage In the author’s opinion, based on current evi-
of topical tacrolimus in preventing as well as dence, topical tacrolimus is a relatively underuti-
reversing graft rejection in high-risk kerato- lized and likely effective immunosuppressive
plasty. However, the current evidence is of therapy for a variety of eye diseases as well as in
moderate quality and limited by small sample immunosuppression of high-risk keratoplasty.
sizes and lack of blinding in randomized con-
trolled trials. A randomized, prospective clini-
cal trial of 49 eyes by Zhai and colleagues in Systemic Therapy
high-risk penetrating keratoplasty compared
the addition of tacrolimus 0.1% versus cyclo- Cyclosporine
sporine 1% to traditional therapy with tobra-
mycin-dexamethasone for 3 weeks Cyclosporine A is a protein derived from fungi.
postoperatively. Patients were followed for an Systemic cyclosporine A increases graft survival
average of 23–24 months. In the tacrolimus in high-risk corneal transplantation. Risks of this
group, the graft rejection rate was 16% and medication are significant, and include hyperten-
there were no irreversible graft rejections. In sion, renal toxicity, neurotoxicity, hepatotoxicity,
the cyclosporine group, the graft rejection rate and posttransplant lymphoproliferative disorders
was 45.8% (P = 0.02) [107]. Dhaliwal and col- [109–112].
leagues presented a prospective case series of A meta-analysis of 16 studies examining the
four eyes that underwent high-risk penetrating utilization of systemic cyclosporine A in high-
keratoplasty and had all developed steroid- risk keratoplasty studied 518 patients using sys-
induced glaucoma and had failed traditional temic cyclosporine for postkeratoplasty
immunosuppressant therapy. All four cases immunosuppression and 299 controls with a
developed acute rejection which reversed with mean postoperative follow-up period of
topical tacrolimus treatment at 0.03% twice 26.5 ± 12.9 months. In the studies with 3 years or
daily. Patients were followed for 26–48 more of follow-up, 66% of rejection episodes
months, and none developed any repeat epi- were successfully reversed in patients on sys-
sodes of graft rejection [105]. temic cyclosporine A, compared to 27.8% in con-
In a prospective, randomized clinical trial of trols (P = 0.02) [113]. The study conducted a
repeat penetrating keratoplasty comparing treat- meta-analysis of ten studies that had 1-year or
ment with topical tacrolimus versus oral myco- longer follow-up data and found that the odds
phenolate in addition to standard treatment with ratio for clear graft survival in patients receiving
topical and oral corticosteroids, topical tacroli- cyclosporine versus controls was 2.43 (95% CI:
mus was equally effective as oral mycophenolate 1.00–5.88, I2 = 37.9%) and 3.64 for rejection-free
in preventing graft rejection at 1 year (P = 0.74) episodes (95% CI: 1.48–8.91, I2 = 64.8%) [113]
[108]. Individual studies may be limited by low power,
In a prospective case series of normal-risk however, the meta-analysis provides strong evi-
penetrating keratoplasties, tacrolimus 0.06% was dence for the benefit of systemic cyclosporine
compared to traditional therapy with topical cor- administration.
A randomized control trial of 38 high-risk reversible and seven irreversible graft rejections,
eyes trialed the usage of systemic cyclosporine compared to six reversible, and two irreversible
and topical steroid in 20 patients, compared with graft rejections in the mycophenolate group.
18 patients who only received topical steroid for However, the rate of adverse events in the myco-
postkeratoplasty immunosuppression. No differ- phenolate group was considerable; the majority
ence was found in graft survival or rejection (63%) in the mycophenolate group experienced
among both groups, but the rate of graft rejection mostly reversible adverse events including gastro-
was extremely low in the study, with only one intestinal disturbances, infections such as bron-
rejection in the control group and none in the chitis, pneumonia, oral candidiasis, elevated liver
treatment group [114]. Another randomized con- enzymes, weight loss or gain, paresthesias, and
trol trial of 40 high-risk corneal transplants found others. Irreversible adverse events in this study
no differences in the rates of graft clarity loss or included myocardial infarction and malignancies
endothelial rejection among patients receiving such as prostate gland and gladder carcinoma in a
systemic cyclosporine in combination with sys- small minority of subjects [119].
temic steroids, compared to a control group that A randomized trial of patients with recurrent
received topical steroids alone [115]. herpetic keratitis who underwent penetrating ker-
atoplasty studied the effect of prophylactic acy-
clovir 200 mg five times/day for 3 weeks
Mycophenolate postoperatively, for 1 year postoperatively and
the addition of mycophenolate 1 g BID for 1 year
Mycophenolate mofetil is an antimetabolite, the in addition to acyclovir. Graft rejection was
morpholinoethylester of mycophenolenic acid. It decreased with the addition of mycophenolate
inhibits inosine monophosphate dehydrogenase mofetil [120].
and T- and B-lymphocyte proliferation. Common Another randomized multicenter trial studied
side effects include gastric discomfort, myelosup- 86 patients undergoing high-risk penetrating ker-
pression, and gingival hyperplasia. Thus, regular atoplasty, of whom 38 controls received fluocor-
monitoring of complete blood counts is recom- tolone 1 mg/kg body weight/day, tapered within
mended [116–118]. A meta-analysis of four stud- 3 weeks, and topical prednisolone 1% tapered
ies examining the usage of systemic mycophenolate within 5 months, and the treatment group addi-
mofetil in high-risk corneal transplantation found tionally received mycophenolate daily 2 × 1 g for
that mycophenolate mofetil increased the revers- the first six postoperative months. The mycophe-
ibility of rejection episodes to 91.7% compared to nolate group had higher graft survival and fewer
52.05% in the control group (P = 0.01) [113]. rejection episodes [116].
A prospective, multicenter, randomized trial
examined the efficacy of mycophenolate in high-
risk penetrating keratoplasties. In the control Mycophenolate vs. Cyclosporine
group, 41 patients received fluocortolone 1 mg/
kg/day tapered over 3 weeks, topical prednisolone In two randomized clinical trials comparing
acetate 5×/day tapered over 5 months. The myco- mycophenolate versus cyclosporine, there was no
phenolate group received all the same medica- difference in graft rejection among both groups
tions as controls in addition mycophenolate 1 g [117, 121].
twice daily for 6 months. Mycophenolate mofetil A meta-analysis assessed four studies which
improved rejection-free graft survival. Over a compared the efficacy of mycophenolate mofetil
mean follow-up time of 34.9 ± 16.3 months, and cyclosporine A in graft survival after high-
immune-reaction-free graft survival was 83% in risk penetrating keratoplasty. In the meta-
the mycophenolate group and 64.5% in the con- analysis, 278 patients received mycophenolate
trol group (P = 0.04). The control group had five mofetil, and 304 patients received cyclosporine A
for postoperative immunosuppression over a Selecting Immunosuppression

mean follow-up time of 24.3 ± 12.7 months. In According to Risk Stratification
the meta-analysis, mycophenolate appears to
have had a slightly higher rejection-free and clear Despite the large plethora of immunosuppressive
graft survival rate compared to cyclosporine therapies available, no standardized protocol has
(88.6% and 97.3%, respectively, compared to been proposed for corneal immunosuppression,
88.8% and 88.6%, respectively), but no statistical likely because much of the above evidence is of
analysis was provided. moderate quality and limited by small study
Retrospective studies comparing systemic sizes. For ocular surface limbal stem cell trans-
mycophenolate mofetil and cyclosporine A have plantation, the Cincinnati protocol was highly
shown similar results [119]. A retrospective effective in a retrospective study of 225 eyes
study of 417 high-risk keratoplasties receiving [123]. Additionally, there is scant evidence for
systemic cyclosporine at blood trough levels of the role of immunosuppression in the highest-
120–150 ng/mL or mycophenolate at a daily risk patients and those with severe graft rejection.
dose of 1 g twice daily found that patients receiv- Nevertheless, immunosuppression is widely uti-
ing mycophenolate were less likely to experi- lized in such patients in order to provide the high-
ence graft rejection (72% rejection-free graft est theoretical probability of graft survival.
survival) compared to cyclosporine (60%) at 3
years. Clear graft survival was 87% in the myco-
phenolate group and 77% in the cyclosporine pecial Considerations for Limbal
S
group at 3 years. This was statistically signifi- Stem Cell Transplantation
cant [118].
Based on these data, there is no convincing Traditionally, limbal stem cell transplantation has
evidence of the superiority of mycophenolate relied on allogeneic or, in the case of keratolim-
versus cyclosporine in high-risk penetrating bal allograft, cadaveric limbal stem cell allografts,
keratoplasty. which require systemic immunosuppression due
to highly vascularized and immunogenic donor
tissue. The need for systemic immunosuppres-
Tacrolimus sion limits recipients to those who can safely
receive immunosuppression. Elderly patients and
There is little evidence on systemic tacrolimus patients with comorbidities such as obesity, dia-
in high-risk keratoplasty, and case series are betes, cardiac, liver, or renal disease are not can-
limited by lack of controls. A noncomparative didates [50, 123–126]. Due to the need for
case series of 17 high-risk corneal and limbal long-term systemic immunosuppression, alloge-
grafts treated with oral tacrolimus found that no neic simple limbal epithelial transportation
patient had irreversible graft rejection while (SLET) should be performed with caution in
receiving tacrolimus, while three patients had children [127]. Preliminary studies suggest the
reversible graft rejection in the setting of low success of allogeneic simple limbal epithelial
levels of tacrolimus [122]. Another case series transplantation with less intensive or minimal
explored the effect of systemic tacrolimus on immunotherapy, but no long-term data are avail-
ten high-risk penetrating keratoplasties which able [128, 129].
developed graft failure while on treatment with Living-related conjunctival limbal allografts
systemic cyclosporine. After treatment with rely on conjunctival and limbal tissue derived
tacrolimus, there were significantly fewer epi- from an HLA-matched living relative and may
sodes of graft rejection (P = 0.03) and longer have a lower rate of rejection by reducing the
graft survival on Kaplan-Meier survival plots potential antigenic burden of the allograft [50,
(P = 0.04) [122]. 130, 131].
Cultivated limbal epithelial transplantation is angiogenic and lymphangiogenic therapy

a modern advancement that requires shorter-term have demonstrated promise in animal and
immunosuppression [132]. Because allogeneic human studies in reducing graft rejection rate.
donor DNA material is not present 9 months fol- • Human leukocyte antigen (HLA) matching
lowing transplantation [133], and the volume of has been shown to be effective in reducing the
tissue transplanted is smaller. Cultivated limbal rejection rate and improving survival in solid
epithelial transplantation only requires immuno- organ transplants but may be costly and logis-
suppression for a year. tically prohibitive for many transplant
Autologous limbal autografts do not carry the centers.
risk of immunoreactivity and thus bypass the • Topical therapy with steroids, cyclosporine,
need for immunosuppression [132, 134, 135]. tacrolimus, and systemic therapy with cyclo-
sporine, mycophenolate, and tacrolimus and
their efficacy in corneal graft rejection pro-
Conclusion phylaxis are reviewed.
• Limbal stem cell transplantation and the rec-
Corneal transplantation is overall a highly suc- ommended strategies for immunosuppression
cessful procedure, with a low risk of graft rejec- are discussed.
tion. Certain risk factors can alter immune
privilege and increase the risk of rejection.
Human leukocyte antigen matching has been
shown to be useful in large series in which typ- References
ing was performed by experienced laboratories,
but larger-scale studies have failed to show ben- 1. Williams KA, Brereton HM, Coster DJ. Prospects
efit, potentially due to a high error rate in sero- for genetic modulation of corneal graft survival.
logical typing. In addition to traditional Eye (Lond). 2009;23(10):1904–9. https://doi.
org/10.1038/eye.2008.378.
treatment with topical and systemic corticoste- 2. Gain P, Jullienne R, He Z, et al. Global survey of
roids, topical tacrolimus and cyclosporine, and corneal transplantation and eye banking. JAMA
systemic cyclosporine, mycophenolate, and Ophthalmol. 2016;134(2):167–73. https://doi.
tacrolimus may be useful adjuncts in improving org/10.1001/jamaophthalmol.2015.4776.
3. van Essen TH, Roelen DL, Williams KA, Jager
graft survival. Future work remains to be done MJ. Matching for human leukocyte antigens (HLA)
in exploring the benefit of these promising ther- in corneal transplantation – to do or not to do.
apies and antigen matching in improving graft Prog Retin Eye Res. 2015;46:84–110. https://doi.
survival. org/10.1016/j.preteyeres.2015.01.001.
4. Niederkorn JY, Larkin DFP. Immune privilege
of corneal allografts. Ocul Immunol Inflamm.
Take Home Notes 2010;18(3):162–71. https://doi.org/10.3109/092739
• Fuchs dystrophy, keratoconus, corneal ecta- 48.2010.486100.
sias, pseudophakic, and aphakic corneal 5. Patel SV, Diehl NN, Hodge DO, Bourne WM. Donor
risk factors for graft failure in a 20-year study
edema are primary transplant indications of penetrating keratoplasty. Arch Ophthalmol.
associated with high rates of corneal trans- 2010;128(4):418–25. https://doi.org/10.1001/
plant survival. Risk factors for corneal trans- archophthalmol.2010.27.
plant failure include a history of infectious 6. Writing Committee for the Cornea Donor Study
Research Group, Sugar A, Gal RL, et al. Factors asso-
leukoma, chemical burn, trauma, HSV/VZV, ciated with corneal graft survival in the cornea donor
previous transplantation, glaucoma, active study. JAMA Ophthalmol. 2015;133(3):246–54.
inflammation at the time of transplantation, https://doi.org/10.1001/jamaophthalmol.2014.3923.
atopy, and corneal neovascularization, among 7. Amescua G, Collings F, Sidani A, et al. Effect of
CXCL-1/KC production in high risk vascularized
others. corneal allografts on T cell recruitment and graft
• Lymphangiogenesis plays a crucial role in the rejection. Transplantation. 2008;85(4):615–25.
pathogenesis of graft rejection. Anti- https://doi.org/10.1097/TP.0b013e3181636d9d.
8. Coster DJ, Williams KA. Management of high-risk 21. Janson BJ, Alward WL, Kwon YH, et al. Glaucoma-
corneal grafts. Eye (Lond). 2003;17(8):996–1002. associated corneal endothelial cell damage: a review.
https://doi.org/10.1038/sj.eye.6700634. Surv Ophthalmol. 2018;63(4):500–6. https://doi.
9. Collaborative Corneal Transplantation Studies org/10.1016/j.survophthal.2017.11.002.
Research Group. The Collaborative Corneal 22. Ferreira TB, Ribeiro FJ, Silva D, Matos AC, Gaspar
Transplantation Studies (CCTS): effective- S, Almeida S. Comparison of refractive and visual
ness of histocompatibility matching in high- outcomes of three presbyopia-correcting intraocular
risk corneal transplantation. Arch Ophthalmol. lenses. J Cataract Refract Surg. 2022;48(3):280–7.
1992;110(10):1392–403. https://doi.org/10.1001/arc https://doi.org/10.1097/j.jcrs.0000000000000743.
hopht.1992.01080220054021. 23. Jonas JB, Aung T, Bourne RR, Bron AM,
10. Williams KA, Roder D, Esterman A, Muehlberg Ritch R, Panda-Jonas S. Glaucoma. Lancet.
SM, Coster DJ. Factors predictive of corneal graft 2017;390(10108):2183–93. https://doi.org/10.1016/
survival. Report from the Australian Corneal Graft S0140-6736(17)31469-1.
Registry. Ophthalmology. 1992;99(3):403–14. 24. Chen M-J, Liu CJ-L, Cheng C-Y, Lee S-M. Corneal
https://doi.org/10.1016/s0161-6420(92)31960-8. status in primary angle-closure glaucoma with a his-
11. Thompson RW Jr, Price MO, Bowers PJ, Price tory of acute attack. J Glaucoma. 2012;21(1):12–6.
FW. Long-term graft survival after penetrating kera- https://doi.org/10.1097/IJG.0b013e3181fc800a.
toplasty. Ophthalmology. 2003;110(7):1396–402. 25. Tan AN, Webers CAB, Berendschot TTJM, et al.
https://doi.org/10.1016/S0161-6420(03)00463-9. Corneal endothelial cell loss after Baerveldt glau-
12. Tourkmani AK, Sánchez-Huerta V, De Wit G, et al. coma drainage device implantation in the ante-
Weighing of risk factors for penetrating keratoplasty rior chamber. Acta Ophthalmol. 2017;95(1):91–6.
graft failure: application of Risk Score System. https://doi.org/10.1111/aos.13161.
Int J Ophthalmol. 2017;10(3):372–7. https://doi. 26. Koo EB, Hou J, Han Y, Keenan JD, Stamper RL,
org/10.18240/ijo.2017.03.08. Jeng BH. Effect of glaucoma tube shunt parameters
13. Barraquer RI, Pareja-Aricò L, Gómez-Benlloch A, on cornea endothelial cells in patients with ahmed
Michael R. Risk factors for graft failure after pene- valve implants. Cornea. 2015;34(1):37. https://doi.
trating keratoplasty. Medicine. 2019;98(17):e15274. org/10.1097/ICO.0000000000000301.
https://doi.org/10.1097/MD.0000000000015274. 27. Williams KA, White MA, Ash JK, Coster
14. Williams KA, Muehlberg SM, Lewis RF, Coster DJ. Leukocytes in the graft bed associated with
DJ. Long-term outcome in corneal allotransplanta- corneal graft failure. Analysis by immunohistol-
tion. Transplant Proc. 1997;29(1–2):983. https://doi. ogy and actuarial graft survival. Ophthalmology.
org/10.1016/S0041-1345(96)00335-1. 1989;96(1):38–44. https://doi.org/10.1016/
15. Armitage WJ, Goodchild C, Griffin MD, et al. s0161-6420(89)32949-6.
High-risk corneal transplantation: recent develop- 28. Cursiefen C, Maruyama K, Jackson DG, Streilein
ments and future possibilities. Transplantation. JW, Kruse FE. Time course of angiogenesis and
2019;103(12):2468–78. https://doi.org/10.1097/ lymphangiogenesis after brief corneal inflammation.
TP.0000000000002938. Cornea. 2006;25(4):443. https://doi.org/10.1097/01.
16. Di Zazzo A, Kheirkhah A, Abud TB, Goyal S, Dana ico.0000183485.85636.ff.
R. Management of high-risk corneal transplantation. 29. Goodfellow JFB, Nabili S, Jones MNA, et al.
Surv Ophthalmol. 2017;62(6):816–27. https://doi. Antiviral treatment following penetrating kerato-
org/10.1016/j.survophthal.2016.12.010. plasty for herpetic keratitis. Eye. 2011;25(4):470–4.
17. Maguire MG, Stark WJ, Gottsch JD, et al. Risk factors https://doi.org/10.1038/eye.2010.237.
for corneal graft failure and rejection in the collab- 30. Halberstadt M, Machens M, Gahlenbek KA, Böhnke
orative corneal transplantation studies. Collaborative M, Garweg JG. The outcome of corneal grafting in
Corneal Transplantation Studies Research Group. patients with stromal keratitis of herpetic and non-
Ophthalmology. 1994;101(9):1536–47. https://doi. herpetic origin. Br J Ophthalmol. 2002;86(6):646.
org/10.1016/s0161-6420(94)31138-9. https://doi.org/10.1136/bjo.86.6.646.
18. Reinhard T, Böhringer D, Sundmacher 31. Gessa-Sorroche M, Kanclerz P, Alio J. Evidence in
R. Accelerated chronic endothelial cell loss the prevention of the recurrence of herpes simplex
after penetrating keratoplasty in glaucoma eyes. and herpes zoster keratitis after eye surgery. Archivos
J Glaucoma. 2001;10(6):446–51. https://doi. De La Sociedad Española De Oftalmología Engl
org/10.1097/00061198-200112000-00002. Ed. 2022;97(3):149–60. https://doi.org/10.1016/j.
19. Cho SW, Kim JM, Choi CY, Park KH. Changes oftale.2022.02.003.
in corneal endothelial cell density in patients with 32. Lee JJ, Kim MK, Wee WR. Adverse effects of
normal-tension glaucoma. Jpn J Ophthalmol. low-dose systemic cyclosporine therapy in high-
2009;53(6):569–73. https://doi.org/10.1007/ risk penetrating keratoplasty. Graefes Arch Clin
s10384-009-0740-1. Exp Ophthalmol. 2015;253(7):1111–9. https://doi.
20. Gagnon MM, Boisjoly HM, Brunette I, Charest M, org/10.1007/s00417-015-3008-0.
Amyot M. Corneal endothelial cell density in glau- 33. Abdelmassih Y, Dubrulle P, Sitbon C, et al.
coma. Cornea. 1997;16(3):314–8. Therapeutic challenges and prognosis of Descemet’s
membrane endothelial keratoplasty in herpes sim- suture. Graefes Arch Clin Exp Ophthalmol.
plex eye disease. Cornea. 2019;38(5):553–8. https:// 2005;243(9):958–9. https://doi.org/10.1007/
doi.org/10.1097/ico.0000000000001891. s00417-005-1136-7.
34. Tambasco FP, Cohen EJ, Nguyen LH, Rapuano CJ, 47. Ghoraishi M, Akova YA, Tugal-Tutkun I, Foster
Laibson PR. oral acyclovir after penetrating kerato- CS. Penetrating keratoplasty in atopic keratocon-
plasty for herpes simplex keratitis. Arch Ophthalmol. junctivitis. Cornea. 1995;14(6):610–3.
1999;117(4):445–9. https://doi.org/10.1001/ 48. Geggel HS. Successful penetrating keratoplasty in
archopht.117.4.445. a patient with severe atopic keratoconjunctivitis and
35. Sterk CC, Jager MJ, Swart M, Berg VD. Recurrent elevated serum IgE level treated with long-term topi-
herpetic keratitis in penetrating keratoplasty. cal cyclosporin A. Cornea. 1994;13(6):543.
Doc Ophthalmol. 1995;90(1):29–33. https://doi. 49. Solomon A, Ellies P, Anderson DF, et al. Long-term
org/10.1007/bf01203291. outcome of keratolimbal allograft with or without
36. Plšková J, Holáň V, Filipec M, Forrester JV. Lymph penetrating keratoplasty for total limbal stem cell
node removal enhances corneal graft survival in deficiency. Ophthalmology. 2002;109(6):1159–66.
mice at high risk of rejection. BMC Ophthalmol. 50. Atallah MR, Palioura S, Perez VL, Amescua
2004;4:3. G. Limbal stem cell transplantation: current perspec-
37. Küchle M, Cursiefen C, Nguyen NX, et al. Risk tives. Clin Ophthalmol. 2016;10:593–602. https://
factors for corneal allograft rejection: intermedi- doi.org/10.2147/OPTH.S83676.
ate results of a prospective normal-risk kerato- 51. Ozer MD, Altinkurt E, Alparslan N. The long-term
plasty study. Graefes Arch Clin Exp Ophthalmol. surgical outcomes of conjunctival-limbal autograft
2002;240(7):580–4. https://doi.org/10.1007/ procedure with or without penetrating keratoplasty
s00417-002-0496-5. in eyes with unilateral limbal stem cell deficiency.
38. Nguyen NX, Martus P, Seitz B, Cursiefen C. Atopic Taiwan J Ophthalmol. 2020;10(1):22–8. https://doi.
dermatitis as a risk factor for graft rejection follow- org/10.4103/tjo.tjo_55_19.
ing normal-risk keratoplasty. Graefes Arch Clin 52. Krysik K, Dobrowolski D, Tarnawska D, Wylegala E,
Exp Ophthalmol. 2009;247(4):573–4. https://doi. Lyssek-Boroń A. Long-term outcomes of allogeneic
org/10.1007/s00417-008-0959-4. ocular surface reconstruction: keratolimbal allograft
39. Watson SL, Tuft SJ, Dart JKG. Patterns of rejection (KLAL) followed by penetrating keratoplasty (PK).
after deep lamellar keratoplasty. Ophthalmology. J Ophthalmol. 2020;2020:5189179. https://doi.
2006;113(4):556–60. https://doi.org/10.1016/j. org/10.1155/2020/5189179.
ophtha.2006.01.006. 53. Solomon A, Pires RT, Tseng SC. Amniotic mem-
40. Reinhard T, Möller M, Sundmacher R. Penetrating brane transplantation after extensive removal of
keratoplasty in patients with atopic derma- primary and recurrent pterygia. Ophthalmology.
titis with and without systemic cyclosporin 2001;108(3):449–60. https://doi.org/10.1016/
A. Cornea. 1999;18(6):645–51. https://doi. s0161-6420(00)00567-4.
org/10.1097/00003226-199911000-00003. 54. Daya SM, Ilari FA. Living related conjunctival lim-
41. Shimmura-Tomita M, Shimmura S, Satake Y, bal allograft for the treatment of stem cell deficiency.
et al. Keratoplasty postoperative treatment update. Ophthalmology. 2001;108(1):126–33. https://doi.
Cornea. 2013;32:S60. https://doi.org/10.1097/ org/10.1016/s0161-6420(00)00475-9; discussion
ICO.0b013e3182a2c937. 133–4.
42. Niederkorn JY, Chen PW, Mellon J, Stevens 55. Cursiefen C, Chen L, Dana MR, Streilein
C, Mayhew E. Allergic airway hyperreactivity JW. Corneal lymphangiogenesis: evidence, mecha-
increases the risk for corneal allograft rejection. Am nisms, and implications for corneal transplant immu-
J Transplant. 2009;9(5):1017–26. nology. Cornea. 2003;22(3):273.
43. Flynn TH, Ohbayashi M, Ikeda Y, Ono SJ, Larkin 56. Yamagami S, Dana MR. The critical role of lymph
DF. Effect of allergic conjunctival inflammation nodes in corneal alloimmunization and graft rejec-
on the allogeneic response to donor cornea. Invest tion. Invest Ophthalmol Vis Sci. 2001;42(6):
Ophthalmol Vis Sci. 2007;48(9):4044–9. https://doi. 1293–8.
org/10.1167/iovs.06-0973. 57. Bahar I, Kaiserman I, McAllum P, Rootman D,
44. Tomita M, Shimmura S, Tsubota K, Shimazaki Slomovic A. Subconjunctival bevacizumab injec-
J. Postkeratoplasty atopic sclerokeratitis in kerato- tion for corneal neovascularization in recur-
conus patients. Ophthalmology. 2008;115(5):851–6. rent pterygium. Curr Eye Res. 2009. https://doi.
https://doi.org/10.1016/j.ophtha.2007.07.018. org/10.1080/02713680701799101.
45. Lyons CJ, Dart JKG, Aclimandos WA, Lightman 58. Kim SW, Ha BJ, Kim EK, Tchah H, Kim T-I. The
S, Buckley RJ. Sclerokeratitis after keratoplasty in effect of topical bevacizumab on corneal neovas-
atopy. Ophthalmology. 1990;97(6):729–33. https:// cularization. Ophthalmology. 2008;115(6):e33–8.
doi.org/10.1016/S0161-6420(90)32523-X. https://doi.org/10.1016/j.ophtha.2008.02.013.
46. Jayaram H, Falcon MG. Atopic rhinitis: a risk fac- 59. Bachmann BO, Bock F, Wiegand SJ, et al. Promotion
tor for spontaneous wound dehiscence following of graft survival by vascular endothelial growth fac-
removal of a continuous penetrating keratoplasty tor a neutralization after high-risk corneal transplan-
tation. Arch Ophthalmol. 2008;126(1):71–7. https:// 72. Sundmacher R, editor. Adequate HLA matching
doi.org/10.1001/archopht.126.1.71. in keratoplasty. Dusseldorf: Karger Medical and
60. Chen L, Hamrah P, Cursiefen C, et al. Vascular endo- Scientific Publishers; 2003.
thelial growth factor receptor-3 mediates induction 73. Völker-Dieben HJ, Schreuder GMT, Claas FHJ,
of corneal alloimmunity. Nat Med. 2004;10(8):813– et al. Histocompatibility and corneal transplanta-
5. https://doi.org/10.1038/nm1078. tion. Dev Ophthalmol. 2003;36:22–41. https://doi.
61. Zhang W, Schönberg A, Hamdorf M, Georgiev T, org/10.1159/000067653.
Cursiefen C, Bock F. Preincubation of donor tis- 74. Kwitko S, Marinho D, Barcaro S, et al. Allograft
sue with a VEGF cytokine trap promotes subse- conjunctival transplantation for bilateral ocular sur-
quent high-risk corneal transplant survival. Br J face disorders. Ophthalmology. 1995;102(7):1020–
Ophthalmol. 2022;106(11):1617–26. https://doi. 5. https://doi.org/10.1016/s0161-6420(95)30918-9.
org/10.1136/bjophthalmol-2021-319745. 75. Armitage WJ, Winton HL, Jones MNA, et al.
62. Su W, Sun S, Tian B, et al. Efficacious, safe, and Corneal transplant follow-up study II (CTFS II): a
stable inhibition of corneal neovascularization by prospective clinical trial to determine the influence
AAV-vectored anti-VEGF therapeutics. Mol Ther of HLA class II matching on corneal transplant
Methods Clin Dev. 2021;22:107–21. https://doi. rejection: baseline donor and recipient characteris-
org/10.1016/j.omtm.2021.06.007. tics. Br J Ophthalmol. 2019;103(1):132–6. https://
63. Salabarria A-C, Braun G, Heykants M, et al. Local doi.org/10.1136/bjophthalmol-2017-311342.
VEGF-A blockade modulates the microenviron- 76. Böhringer D, Spierings E, Enczmann J, et al. Matching
ment of the corneal graft bed. Am J Transplant. of the minor histocompatibility antigen HLA-A1/
2019;19(9):2446–56. https://doi.org/10.1111/ H-Y may improve prognosis in corneal transplanta-
ajt.15331. tion. Transplantation. 2006;82(8):1037–41. https://
64. Dohlman TH, Omoto M, Hua J, et al. VEGF- doi.org/10.1097/01.tp.0000235908.54766.44.
trap aflibercept significantly improves long-term 77. Hopkinson CL, Romano V, Kaye RA, et al. The
graft survival in high-risk corneal transplantation. influence of donor and recipient gender incompat-
Transplantation. 2015;99(4):678–86. https://doi. ibility on corneal transplant rejection and failure.
org/10.1097/TP.0000000000000512. Am J Transplant. 2017;17(1):210–7. https://doi.
65. Fasciani R, Crincoli E, Mosca L, Guccione L, org/10.1111/ajt.13926.
Caristia A, Rizzo S. Role of pre-transplant corneal 78. Böhringer D, Reinhard T, Böhringer S, Enczmann
injective anti VEGF treatment in high risk trans- J, Godehard E, Sundmacher R. Predicting time on
plantation corneas. 2021. https://doi.org/10.21203/ the waiting list for HLA matched corneal grafts.
rs.3.rs-246503/v1. Tissue Antigens. 2002;59(5):407–11. https://doi.
66. Stodola E. Applications for bevacizumab in corneal org/10.1034/j.1399-0039.2002.590507.x.
surgery. EyeWorld. 2021. 79. Alió Del Barrio JL, et al. Corneal transplanta-
67. Opelz G, Döhler B. Effect of human leukocyte anti- tion after failed grafts: options and outcomes.
gen compatibility on kidney graft survival: com- Surv Ophthalmol. 2021;66(1):20–40. https://doi.
parative analysis of two decades. Transplantation. org/10.1016/j.survophthal.2020.10.003.
2007;84(2):137–43. https://doi.org/10.1097/01. 80. Duquesnoy RJ, Askar M. HLAMatchmaker: a molec-
tp.0000269725.74189.b9. ularly based algorithm for histocompatibility deter-
68. Loupy A, Hill GS, Jordan SC. The impact of donor- mination. V. Eplet matching for HLA-DR, HLA-DQ,
specific anti-HLA antibodies on late kidney allograft and HLA-DP. Hum Immunol. 2007;68(1):12–25.
failure. Nat Rev Nephrol. 2012;8(6):348–57. https:// https://doi.org/10.1016/j.humimm.2006.10.003.
doi.org/10.1038/nrneph.2012.81. 81. Böhringer D, Sundmacher R, Reinhard
69. Hopkins KA, Maguire MG, Fink NE, Bias T. Histocompatilibility matching in penetrat-
WB. Reproducibility of HLA-A, B, and DR typing ing keratoplasty. In: Reinhard T, Larkin F, Larkin
using peripheral blood samples: results of retyping DFP, editors. Cornea and external eye disease.
in the collaborative corneal transplantation stud- Berlin: Springer; 2006. https://books.google.com/
ies. Collaborative Corneal Transplantation Studies books/about/Cornea_and_External_Eye_Disease.
Group (corrected). Hum Immunol. 1992;33(2):122– html?id=uAfeQjUKTf4C.
8. https://doi.org/10.1016/0198-8859(92)90062-r. 82. Shimazaki J, Iseda A, Satake Y, Shimazaki-Den
70. Vail A, Gore SM, Bradley BA, Easty DL, Rogers S. Efficacy and safety of long-term corticosteroid
CA, Armitage WJ. Conclusions of the corneal eye drops after penetrating keratoplasty: a prospec-
transplant follow up study. Collaborating Surgeons. tive, randomized, clinical trial. Ophthalmology.
Br J Ophthalmol. 1997;81(8):631–6. https://doi. 2012;119(4):668–73. https://doi.org/10.1016/j.
org/10.1136/bjo.81.8.631. ophtha.2011.10.016.
71. Völker-Dieben HJ, Claas FH, Schreuder GM, 83. Kharod-Dholakia B, Randleman JB, Bromley JG,
et al. Beneficial effect of HLA-DR match- Stulting RD. Prevention and treatment of corneal
ing on the survival of corneal allografts. graft rejection: current practice patterns of the
Transplantation. 2000;70(4):640–8. https://doi. Cornea Society (2011). Cornea. 2015;34(6):609.
org/10.1097/00007890-200008270-00018. https://doi.org/10.1097/ICO.0000000000000403.
84. Nguyen NX, Seitz B, Martus P, Langenbucher 96. Taddio A, Cimaz R, Caputo R, et al. Childhood
A, Cursiefen C. Long-term topical steroid treat- chronic anterior uveitis associated with vernal
ment improves graft survival following normal- keratoconjunctivitis (VKC): successful treat-
risk penetrating keratoplasty. Am J Ophthalmol. ment with topical tacrolimus. Case series. Pediatr
2007;144(2):318–9. https://doi.org/10.1016/j. Rheumatol Online J. 2011;9(1):34. https://doi.
ajo.2007.03.028. org/10.1186/1546-0096-9-34.
85. Ross AH, Jones MNA, Nguyen DQ, et al. Long-term 97. Kheirkhah A, Zavareh MK, Farzbod F, Mahbod
topical steroid treatment after penetrating kerato- M, Behrouz MJ. Topical 0.005% tacrolimus eye
plasty in patients with pseudophakic bullous kera- drop for refractory vernal keratoconjunctivitis. Eye
topathy. Ophthalmology. 2009;116(12):2369–72. (Lond). 2011;25(7):872–80. https://doi.org/10.1038/
https://doi.org/10.1016/j.ophtha.2009.06.006. eye.2011.75.
86. Butcher JM, Austin M, McGalliard J, Bourke 98. Miyazaki D, Fukushima A, Ohashi Y, et al. Steroid-
RD. Bilateral cataracts and glaucoma induced sparing effect of 0.1% tacrolimus eye drop for treat-
by long term use of steroid eye drops. BMJ. ment of shield ulcer and corneal epitheliopathy in
1994;309(6946):43. https://doi.org/10.1136/ refractory allergic ocular diseases. Ophthalmology.
bmj.309.6946.43. 2017;124(3):287–94. https://doi.org/10.1016/j.
87. Marques RE, Leal I, Guerra PS, Barão RC, Quintas ophtha.2016.11.002.
AM, Rodrigues W. Topical corticosteroids with 99. Lee YJ, Kim SW, Seo KY. Application for tacro-
topical cyclosporine A versus topical corticosteroids limus ointment in treating refractory inflamma-
alone for immunological corneal graft rejection. tory ocular surface diseases. Am J Ophthalmol.
Eur J Ophthalmol. 2022;32(3):1469–81. https://doi. 2013;155(5):804–13. https://doi.org/10.1016/j.
org/10.1177/11206721211023320. ajo.2012.12.009.
88. Donnenfeld E, Pflugfelder SC. Topical ophthal- 100. Moscovici BK, Holzchuh R, Chiacchio BB, Santo
mic cyclosporine: pharmacology and clinical uses. RM, Shimazaki J, Hida RY. Clinical treatment
Surv Ophthalmol. 2009;54(3):321–38. https://doi. of dry eye using 0.03% tacrolimus eye drops.
org/10.1016/j.survophthal.2009.02.002. Cornea. 2012;31(8):945–9. https://doi.org/10.1097/
89. Javadi MA, Feizi S, Karbasian A, Rastegarpour ICO.0b013e31823f8c9b.
A. Efficacy of topical ciclosporin A for treatment 101. Moscovici BK, Holzchuh R, Sakassegawa-Naves
and prevention of graft rejection in corneal grafts FE, et al. Treatment of Sjögren’s syndrome dry
with previous rejection episodes. Br J Ophthalmol. eye using 0.03% tacrolimus eye drop: prospective
2010;94(11):1464–7. https://doi.org/10.1136/ double-blind randomized study. Cont Lens Anterior
bjo.2009.172577. Eye. 2015;38(5):373–8. https://doi.org/10.1016/j.
90. Ünal M, Yücel I. Evaluation of topical ciclosporin clae.2015.04.004.
0.05% for prevention of rejection in high-risk cor- 102. Abud TB, Di Zazzo A, Kheirkhah A, Dana
neal grafts. Br J Ophthalmol. 2008;92(10):1411–4. R. Systemic immunomodulatory strategies in
https://doi.org/10.1136/bjo.2008.143024. high-risk corneal transplantation. J Ophthalmic
91. Poon A, Constantinou M, Lamoureux E, Taylor Vis Res. 2017;12(1):81–92. https://doi.
HR. Topical cyclosporin A in the treatment of org/10.4103/2008-322X.200156.
acute graft rejection: a randomized controlled trial. 103. Durazzo TS, Frencher S, Gusberg R. Influence of
Clin Experiment Ophthalmol. 2008;36(5):415–21. race on the management of lower extremity isch-
https://doi.org/10.1111/j.1442-9071.2008.01808.x. emia: revascularization vs amputation. JAMA
92. Sinha R, Jhanji V, Verma K, Sharma N, Biswas Surg. 2013;148(7):617–23. https://doi.org/10.1001/
NR, Vajpayee RB. Efficacy of topical cyclosporine jamasurg.2013.1436.
A 2% in prevention of graft rejection in high-risk 104. Jung JW, Lee YJ, Yoon SC, Kim T-I, Kim EK, Seo
keratoplasty: a randomized controlled trial. Graefes KY. Long-term result of maintenance treatment with
Arch Clin Exp Ophthalmol. 2010;248(8):1167–72. tacrolimus ointment in chronic ocular graft-versus-
https://doi.org/10.1007/s00417-010-1388-8. host disease. Am J Ophthalmol. 2015;159(3):519–
93. Shoughy SS. Topical tacrolimus in ante- 27.e1. https://doi.org/10.1016/j.ajo.2014.11.035.
rior segment inflammatory disorders. Eye Vis 105. Dhaliwal JS, Mason BF, Kaufman SC. Long-term
(Lond). 2017;4(1):1–7. https://doi.org/10.1186/ use of topical tacrolimus (FK506) in high-risk pene-
s40662-017-0072-z. trating keratoplasty. Cornea. 2008;27(4):488. https://
94. Turgut B, Guler M, Akpolat N, Demır T, Celıker doi.org/10.1097/ICO.0b013e3181606086.
U. The impact of tacrolimus on vascular endothelial 106. Magalhaes OA, Marinho DR, Kwitko S. Topical
growth factor in experimental corneal neovascular- 0.03% tacrolimus preventing rejection in high-
ization. Curr Eye Res. 2011;36(1):34–40. https://doi. risk corneal transplantation: a cohort study. Br
org/10.3109/02713683.2010.516620. J Ophthalmol. 2013;97(11):1395–8. https://doi.
95. Shoughy SS, Aljassar FM, Tabbara KF. Aqueous org/10.1136/bjophthalmol-2013-303639.
penetration of topical tacrolimus. Am J Ophthalmol 107. Zhai L-Y, Zhang X-R, Liu H, Ma Y, Xu
Case Rep. 2020;17:100582. https://doi. H-C. Observation of topical tacrolimus on high-
org/10.1016/j.ajoc.2019.100582. risk penetrating keratoplasty patients: a randomized
clinical trial study. Eye (Lond). 2020;34(9):1600–7. tive, randomised, multicentre study. Eye (Lond).
https://doi.org/10.1038/s41433-019-0717-3. 2009;23(11):2063–70. https://doi.org/10.1038/
108. Faramarzi A, Abbasi H, Feizi S, et al. Topical 0.03% eye.2008.402.
tacrolimus versus systemic mycophenolate mofetil 120. Mayer K, Reinhard T, Reis A, Voiculescu A,
as adjuncts to systemic corticosteroids for prevent- Sundmacher R. Synergistic antiherpetic effect of
ing graft rejection after repeat keratoplasty: one-year acyclovir and mycophenolate mofetil following
results of a randomized clinical trial. Eye (Lond). keratoplasty in patients with herpetic eye disease:
2021;35(10):2879–88. https://doi.org/10.1038/ first results of a randomised pilot study. Graefes
s41433-020-01375-z. Arch Clin Exp Ophthalmol. 2003;241(12):1051–4.
109. Hill JC. The use of cyclosporine in high-risk kera- https://doi.org/10.1007/s00417-003-0724-7.
toplasty. Am J Ophthalmol. 1989;107(5):506–10. 121. Reis A, Reinhard T, Voiculescu A, et al.
https://doi.org/10.1016/0002-9394(89)90494-7. Mycophenolate mofetil versus cyclosporin A
110. Hill JC. Systemic cyclosporine in high-risk in high risk keratoplasty patients: a prospec-
keratoplasty: short- versus long-term therapy. tively randomised clinical trial. Br J Ophthalmol.
Ophthalmology. 1994;101(1):128–33. https://doi. 1999;83(11):1268–71. https://doi.org/10.1136/
org/10.1016/S0161-6420(13)31253-6. bjo.83.11.1268.
111. Algros M-P, Angonin R, Delbosc B, Cahn J-Y, 122. Sloper CM, Powell RJ, Dua HS. Tacrolimus (FK506)
Kantelip B. Danger of systemic cyclosporine for in the management of high-risk corneal and lim-
corneal graft. Cornea. 2002;21(6):613. bal grafts. Ophthalmology. 2001;108(10):1838–44.
112. Abudou M, Wu T, Evans JR, Chen https://doi.org/10.1016/s0161-6420(01)00759-x.
X. Immunosuppressants for the prophylaxis of cor- 123. Holland EJ, Mogilishetty G, Skeens HM, et al.
neal graft rejection after penetrating keratoplasty. Systemic immunosuppression in ocular surface stem
Cochrane Database Syst Rev. 2015;(8):CD007603. cell transplantation: results of a 10-year experience.
https://doi.org/10.1002/14651858.CD007603.pub2. Cornea. 2012;31(6):655. https://doi.org/10.1097/
113. Bali S, Filek R, Si F, Hodge W. Systemic immuno- ICO.0b013e31823f8b0c.
suppression in high-risk penetrating keratoplasty: a 124. Baradaran-Rafii A, Eslani M, Djalillian
systematic review. J Clin Med Res. 2016;8(4):269– AR. Complications of keratolimbal allograft surgery.
76. https://doi.org/10.14740/jocmr2326w. Cornea. 2013;32(5):561. https://doi.org/10.1097/
114. Den S, Omoto M, Shimmura S, Tsubota K, ICO.0b013e31826215eb.
Shimazaki J. Prospective, randomized study on 125. Holland EJ. Living related conjunctival lim-
efficacy of systemic cyclosporine a in high–risk bal allograft for the treatment of stem cell defi-
corneal transplantation. Invest Ophthalmol Vis Sci. ciency - discussion. Ophthalmology. 2001;108(1):
2006;47(13):1286. 133–4.
115. Shimazaki J, Den S, Omoto M, Satake Y, Shimmura 126. Chan CC, Holland EJ. Keratolimbal allograft. In:
S, Tsubota K. Prospective, randomized study of the Mannis MJ, Holland EJ, editors. Cornea. 5th ed.
efficacy of systemic cyclosporine in high-risk corneal New York: Elsevier; 2022.
transplantation. Am J Ophthalmol. 2011;152(1):33– 127. Shanbhag SS, Patel CN, Goyal R, Donthineni PR,
39.e1. https://doi.org/10.1016/j.ajo.2011.01.019. Singh V, Basu S. Simple limbal epithelial trans-
116. Reinhard T, Mayweg S, Sokolovska Y, et al. Systemic plantation (SLET): review of indications, surgical
mycophenolate mofetil avoids immune reactions technique, mechanism, outcomes, limitations, and
in penetrating high-risk keratoplasty: preliminary impact. Indian J Ophthalmol. 2019;67(8):1265–77.
results of an ongoing prospectively randomized https://doi.org/10.4103/ijo.ijo_117_19.
multicentre study. Transpl Int. 2005;18(6):703–8. 128. Iyer G, Srinivasan B, Agarwal S, Tarigopula
https://doi.org/10.1111/j.1432-2277.2005.00126.x. A. Outcome of allo simple limbal epithelial transplan-
117. Reinhard T, Reis A, Böhringer D, et al. Systemic tation (alloSLET) in the early stage of ocular chemi-
mycophenolate mofetil in comparison with systemic cal injury. Br J Ophthalmol. 2017;101(6):828–33.
cyclosporin A in high-risk keratoplasty patients: https://doi.org/10.1136/bjophthalmol-2016-309045.
3 years’ results of a randomized prospective clini- 129. Kaur A, Jamil Z, Priyadarshini SR. Allogeneic
cal trial. Graefes Arch Clin Exp Ophthalmol. simple limbal epithelial transplantation: an appro-
2001;239(5):367–72. https://doi.org/10.1007/ priate treatment for bilateral stem cell deficiency.
s004170100285. BMJ Case Rep. 2021;14(2):e239998. https://doi.
118. Birnbaum F, Böhringer D, Sokolovska Y, org/10.1136/bcr-2020-239998.
Sundmacher R, Reinhard T. Immunosuppression 130. Férnandez-Buenaga R, Aiello F, Zaher SS, Grixti A,
with cyclosporine A and mycophenolate mofetil Ahmad S. Twenty years of limbal epithelial therapy:
after penetrating high-risk keratoplasty: a retrospec- an update on managing limbal stem cell deficiency.
tive study. Transplantation. 2005;79(8):964. https:// BMJ Open Ophthalmol. 2018;3(1):e000164. https://
doi.org/10.1097/01.TP.0000158022.62059.F2. doi.org/10.1136/bmjophth-2018-000164.
119. Birnbaum F, Mayweg S, Reis A, et al. Mycophenolate 131. Serna-Ojeda JC, Basu S, Vazirani J, Garfias Y,
mofetil (MMF) following penetrating high-risk Sangwan VS. Systemic immunosuppression for
keratoplasty: long-term results of a prospec- limbal allograft and allogenic limbal epithelial cell
transplantation. Med Hypothesis Discov Innov 134. Eslani M, Cheung AY, Holland EJ. 3-s2.0-
Ophthalmol J. 2019;9(1):23–32. B9780323672405001645?indexOverride=GLO
132. Graue-Hernandez EO. Simple limbal epithe- BAL. In: Mannis MJ, Holland EJ, editors. Cornea,
lial transplantation. In: Mannis MJ, Holland EJ, 5th ed. New York: Elsevier; 2022. p. 1607.e1–14.e1.
editors. Cornea. New York: Elsevier; 2022. p. 1634. 135. Basu S, Sureka SP, Shanbhag SS, Kethiri AR,
e1–e1638. Singh V, Sangwan VS. Simple limbal epithelial
133. Daya SM, Watson A, Sharpe JR, et al. Outcomes and transplantation: long-term clinical outcomes in 125
DNA analysis of ex vivo expanded stem cell allograft cases of unilateral chronic ocular surface burns.
for ocular surface reconstruction. Ophthalmology. Ophthalmology. 2016;123(5):1000–10. https://doi.
2005;112(3):470–7. https://doi.org/10.1016/j. org/10.1016/j.ophtha.2015.12.042.
ophtha.2004.09.023.
Management of the Vascularized
Cornea Before Corneal Graft
4
Surgery: Fine-Needle Diathermy
and Inhibition of VEGF
Nadim S. Azar, Matias Soifer, and Victor L. Perez
Key Points corneal transplantation, namely fine-needle

• Corneal neovascularization (CNV) is a patho- diathermy and laser therapy.
logical condition that appears as a result of • Medical therapies are also used for corneal
chronic persistence of different infectious, neovascularization prior to corneal transplan-
immune-related and mechanical factors. tation, notably anti-VEGF therapy via topical
• Preoperative CNV is a key risk factor for graft drops, subconjunctival injections and deliv-
rejection. It is necessary to treat corneal neo- ered through scleral lens.
vascularization prior to corneal transplant sur-
gery to prevent rejection.
• Many surgical techniques exist for the man- Introduction and Epidemiology
agement of corneal neovascularization prior to
Corneal neovascularization is a condition that
affects around ten million people in the world. In
the USA, it constitutes a feature of roughly 5% of
Supplementary Information The online version con- all ophthalmic presentations, with an estimated
tains supplementary material available at https://doi.
incidence of 1.4 million new cases yearly [1].
org/10.1007/978-3-031-32408-6_4.
The angiogenic drive emanates as a response to
the presence of an irritating ocular surface patho-
N. S. Azar · M. Soifer gen over time, such as allergies, microbial kerati-
Foster Center for Ocular Immunology, Duke Eye tis, immune-mediated processes, persistent
Institute, Durham, NC, USA
epithelial defects, limbal stem cell deficiency,
Department of Ophthalmology, Duke University chronic contact lens use and more. A chronic
Medical Center, Durham, NC, USA
e-mail: [email protected]; angiogenic response that persists as a response to
[email protected] the different etiologies can lead to a pathological
V. L. Perez (*) CNV that may affect visual acuity. [2]
Foster Center for Ocular Immunology, Duke Eye Corneal neovascularization represents a high-
Institute, Durham, NC, USA risk factor for rejection in patients that will
Department of Ophthalmology, Duke University undergo a corneal transplant. This was reported
Medical Center, Durham, NC, USA in a study by Cursiefen et al., where around 20%
Foster Center for Ocular Immunology at Duke Eye of human corneal buttons obtained by penetrat-
Center, Duke University School of Medicine, ing keratoplasty (PK) showed signs of neovascu-
Durham, NC, USA larization and angiogenesis [3, 4]. It is also

https://doi.org/10.1007/978-3-031-32408-6_4
60 N. S. Azar et al.
necessary to study corneal neovascularization linical Considerations for Treatment

C
prior to corneal transplant surgery, as CNV per se Administration and Management
is a well-known risk predictor of graft rejection
and subsequently failure [3, 5]. In fact, in this There are important clinical considerations to
same study, histopathologic evaluation of neo- be taken into account in treatment and manage-
vascularized corneal buttons that underwent PK ment of CNV prior to PK. A comprehensive
was done and showed that graft rejection was patient and slit lamp evaluation is essential
present in around 30% of the buttons [4]. CNV toward identifying, staging, and managing the
has shown to increase corneal edema, inflamma- corneal neovascularization and its associated
tion, scarring and lipid deposition and to worsen risk of rejection/failure of the patient’s trans-
the prognosis of following PKs, thereby corrobo- plant. The principal risk factor for postoperative
rating the necessity to treat it prior to PK [3]. In corneal graft rejection is the presence of host
fact, the collaborative corneal transplantation neovascularization prior to surgery. In fact, the
studies (CCTS) evaluated the survival of corneal risk is proportional to the area and depth of cor-
transplants in high-risk patients defined as having neal vascularity [8]. Therefore, a thorough and
two or more quadrants of corneal stromal vascu- complete clinical assessment of the corneal ves-
larization and/or a history of graft rejection previ- sels is mandatory. Another important consider-
ously, and observed that corneal rejection rate ation is noting whether the vessels are epithelial
ranged between 16% and 41% regardless of (superficial) or stromal (deep) since the depth
human leukocyte antigen (HLA) or ABO- may offer clues to their etiology and establish a
compatibility matching [6]. CNV prior to, but prognosis. In short, superficial neovasculariza-
also following PK can adversely affect graft lon- tion represent the “benign” spectrum of etiolo-
gevity [7]. Importantly, CNV presence can be gies that occurs as a result of ocular surface
recognized on slit lamp, and since it is a clinically disorders or contact lens wear. However, stro-
visible objective biomarker, it can be noted, mal vascularization points toward serious causes
graded and treated adequately in order to prevent such as interstitial keratitis, which can be micro-
rejection, increase graft survival and inform the bial or autoimmune. These warrant investiga-
patient about the transplant prognosis. As such, tion since they may present a recurrent course
management of the vascularized cornea before and flare again, which results in poor graft sur-
corneal graft surgery, and more specifically PK vival. It has also been shown that deep corneal
surgery, is vital and easy to detect and is a predic- angiogenesis is more associated with risk of
tor of graft stability and decreased rejection. graft failure than superficial vascularization,
which itself is more correlated with direct post-
operative visual acuity [9].
Surgical and Medical Treatments Significantly, it is crucial to assess whether
for Corneal Neovascularization there is active inflammation on the hosts ocular
Prior to Corneal Transplantation surface and dampen as much of it as possible.
Active inflammation conducts increased vessel
Many treatments are available for the manage- formation and predisposes to rejection and over-
ment of corneal neovascularization prior to cor- all failure. Every ophthalmic inflammatory dis-
neal transplants. These are commonly aimed at ease should be in remission before doing elective
ablating the vessels (laser photocoagulation, cau- corneal transplantation, at least for 6 months
terization or fine-needle diathermy) or regulating [10]. Although CNV is the biggest risk factor for
angiogenesis (anti-VEGF antibodies). The graft rejection, an ocular surface evaluation is
upcoming sections highlight modern manage- mandatory to correct potential additive features
ment techniques for corneal neovascularization such as inflammatory dry eye syndrome [11] or
with their associated efficacy and pertinent side lid abnormalities prone to increased inflamma-
effects (Video 4.1). tion and, consequently, a worse prognosis.
4 Management of the Vascularized Cornea Before Corneal Graft Surgery: Fine-Needle Diathermy… 61
Treating inflammatory conditions that can be Although there is limited literature assess-
associated with corneal neovascularization such ing the efficacy and safety of the technique, it
as the ones described previously in the risk fac- has been used for the management of CNV. Our
tors is necessary for the reduction of neovascular- group, among other surgical teams, has adopted
ization. Finally, the patient should be counseled the use of FND during corneal transplant sur-
regarding the high-risk category of their trans- gery. We target the diathermy treatment toward
plant [12]. A strict anti-inflammatory and anti- bleeding stromal vessels in the host bed. This
microbial protocol has to be explained with its usually happens after removing the recipient
rationale to help the patient understand the need tissue but before performing the transplant.
for frequent postoperative topical drops, the pos- This technique’s efficacy relies on its ability to
sibility of a graft rejection, and its management be used during surgery while taking advantage
given their prognosis. of clear visualization of blood vessels and neo-
vascularization, but also direct access to them.
Fine-needle diathermy, however, can poten-
Surgical and Laser Treatments tially damage the corneal endothelium beneath
for Management of Corneal the treated area, but also corneoscleral limbus
Neovascularization [14]. It has also been noted that FND releases
proangiogenic factors, which would lead to the
Fine-Needle Diathermy stimulation of vascularization, which goes
Fine-needle diathermy (FND) is a mainstay and against its intended effect and limits its use [15].
an important method of targeting corneal neovas- Some of the downsides include the necessity for
cularization (Table 4.1). It consists of applying retreatments for optimal results, but also serious
thermal energy to the corneal layers up to the side effects such as corneal perforations, notably
deep stroma in order to obliterate corneal neovas- in thin corneas, as it involves the use of a needle
cularization. It involves using a stainless-steel [1, 16]. Other side effects might include striae
cutting needle attached to a nylon suture. The creation, corneal scarring and intracorneal bleed-
needle is usually inserted around the limbus in a ing that usually resolves in weeks [17]. To coun-
parallel fashion to the blood vessels targeted. It teract intrastromal bleeding, FND guided by
can also be inserted directly into the blood vessel angiography can be used to selectively treat affer-
lumen in case the vessel is large. A diathermy ent feeder vessels [18, 19]. In some instances,
unit set at a low current setting (around 0.5–1 mA) whitening of the stromal cornea occurs near the
is put in contact with the needle for around 1 s, needle pricks, but it is usually resolvable over a
which blanches the selected blood vessels [1]. day or 2 [16, 17]. Overall, it appears as a reason-
This technique was studied and described by able treatment modality due to the fact that is
Pillai et al. [13]. affordable, simple and effective. It is usually
Table 4.1 Surgical therapies for managing corneal neovascularization prior to corneal transplant: clinical benefits,
drawbacks and complications
Surgical
therapies Clinical benefits Drawbacks Complications
Fine-needle Relatively cheap, simple and Afferent both afferent and efferent Corneal perforations and
diathermy effective, it can penetrate into corneal vessels, retreatment scarring, striae formation,
different depths of necessary, can release proangiogenic intracorneal and intrastromal
neovascularization invasion factors bleeding, corneal opacification
Laser Simple procedure, patient Afferent corneal vessel re-budding Damage to corneal
tolerability, very targeted: only does not show significant effects endothelium and lens, corneal
corneal efferent vessels are when neovascularization is sizeable thinning and bleeding, iris
affected or when corneas are very inflamed, deposits and atrophy, pupil
costly and several sessions needed peaking
done under topical anesthesia and, unlike laser CNV prior to PK was done [21]. Although practical
treatment, it destroys both afferent and efferent in an outpatient setting and tolerable by patients,
corneal vessels. FND has recently been altered laser treatment is not as effective in cases in cases of
by using an electrolysis needle that is thinner and sizeable corneal neovascularization [22]. Overall
more flexible than the previously described [20]. general side effects of laser therapy might include
It involves direct thermal energy to cauterize the damage to corneal endothelium or even the lens, but
vessels instead of electrical currents, as originally also corneal thinning and bleeding, iris deposits and
described in diathermy. It was reported in three atrophy, and pupil peaking [16, 23].
cases of lipid keratopathy on whom no postoper-
ative complications were observed [20], but the
sample size was small, and more extensive Medical Antiangiogenic Treatments
research should be done to assess this procedure. for Management of Corneal
Neovascularization
Laser Treatment
Another effective strategy for targeting corneal vas- Anti-VEGF/VEGF receptor agents have appeared
cularization is through the use of laser therapy as notable players in the anti-angiogenesis ther-
(Table 4.1). In general, laser photocoagulation works apy prior to PK. Many antiangiogenic therapeutic
by destruction of corneal efferent vessels, specifi- techniques interfere with the VEGF system, the
cally as these are wide and have a slow blood flow most important ones consisting of VEGF neutral-
relatively, in comparison to the afferent vessels that izing direct antibodies, VEGF receptor antibod-
are deeper and thinner and have a much faster blood ies and receptor tyrosine kinase inhibitors that act
flow [16]. Targeting of afferent vessels would thus downstream in the VEGF pathway. VEGF is a
lead to their reappearance and the necessity to resort member of a family of proteins consisting of sev-
to different modalities of treatments. Frequency- eral subtypes (VEGF-A, VEGF-B, VEGF-C and
doubled Nd:YAG (532 nm), an important type of VEGF-D), with the VEGF-A isoforms being the
laser photocoagulation, is an effective treatment that most studied and notable inducer of pathologic
can decrease the area of corneal vascularization CNV. The main categories we will discuss in the
without causing any significant side effects, as stro- chapter are VEGF neutralizing antibodies and
mal hemorrhage was a rare noted complication by also use of scleral devices incorporating anti-
Parsa et al., in a patient where an attempt to reduce VEGF on the ocular surface (Table 4.2).
Table 4.2 Antiangiogenic medical therapies for corneal neovascularization prior to corneal transplant: clinical bene-
fits, drawbacks and side effects
Medical
therapies Clinical benefits Drawbacks Side effects
Anti-VEGF Work on newly formed vessels, Costly, reduced antiangiogenic Subconjunctival injection: risk
agents extensive data regarding use, effects on mature already existing of vasospasm and vascular
different modes of administration, vessels, resistance to treatment, effects, local side effects from
safer and more effective than nonresponsiveness to treatment, injection
anti-inflammatory drugs, directed short half-lives Topical injections: adhesions
mechanism of action between corneal epithelium
and basement membrane,
stromal thinning, delayed
wound healing, epithelial
defects
Scleral lens Constantly bathes ocular surface Costly, scarcity of data regarding Adhesions, stromal thinning,
devices and increases the bioavailability scleral lens devices prior to PK, delayed wound healing in
with of drugs, mechanical protective similar drawbacks to anti-VEGF some cases
anti-VEGF effects on cornea, custom-fitted treatments that are administered
under the device
VEGF vascular endothelial growth factor, PK penetrating keratoplasty
Anti-VEGF Local Therapy bevacizumab have been shown to decrease neo-

VEGF is as a corneal angiogenic factor that stems vascularization to some extent, thereby improv-
from inflammation and is described as an essen- ing corneal clarity [52, 53]. A direct comparison
tial component of neovascularization in rat cor- between subconjunctival and topical administra-
neas [24]. VEGF is proven to be involved in a tions has shown that both are equally as effective
causal relationship in corneal neovascularization in reducing corneal angiogenesis [54].
through extensive research and evidence [25–30]. Subconjunctival injection has three main draw-
VEGF-A, a subtype from the VEGF family of backs: it has a limited duration of action in the
proteins, augments the replication of endothelial cornea [55], may result in the quicker entry of
cells through mitosis and facilitates their migra- the drug into the systemic circulation, thus
tion but is also involved in the creation of blood exposing the patient to the risk of vasospasm and
vessels [25]. vascular effects and finally, local side effects
Anti-VEGF antibodies, which are currently from the injection. However, topical drop admin-
used as intravitreal injections for retinal prolifer- istration is estimated to affect adhesion between
ative vascular disorders, offer promising applica- epithelium and basement membrane in compari-
tions for the management of corneal son to subconjunctival injection, resulting in
neovascularization [31]. These can be adminis- stromal thinning and delayed wound healing [40,
trated via topical drops, subconjunctival injec- 56–59]. This effect is duration and dose-depen-
tion, and with the use of scleral lenses. Anti-VEGF dent [58] and has been shown to reverse after
antibodies help in corneal graft survival, as initial treatment [60].
mouse models report improvement in murine In short, anti-VEGF therapy studies are still
graft survival [32, 33]. small and cannot provide extremely conclusive
Bevacizumab is the most studied anti-VEGF evidence. Anti-VEGF agents may be safer and
therapy: it consists of a humanized murine mono- more effective than anti-inflammatory drugs as
clonal IgG1 antibody that reduces corneal neo- they generally present with less side effects and
vascularization [34]. Many clinical studies, have a more directed mechanism of action ratio-
mainly prospective case series and case reports, nale. As a whole, they appear to be especially
were effectuated using bevacizumab therapy for useful for the treatment of newly created vessels
corneal neovascularization. These studies involve which signify ongoing disease, while they are
different etiologies from which the corneal neo- less effective in treating mature previously pres-
vascularization emanates, such as preceding a PK ent vessels in chronic neovascularization. As
[35, 36], after a PK [37], and on rejected/reject- such, they might be used in conjunction with ste-
ing PK [38–41]. Experimental bevacizumab roids and other anti-inflammatory drugs to target
studies have demonstrated statistically significant both immature and mature vessels [1]. Major
effects in decreasing neovascularization, by using challenges to anti-VEGF antibody use mainly
different parameters such as length, density and pertain to high costs of the modality, resistance to
area of CNV [27, 32, 39, 42–46]. It has shown treatment, short half-life and partial to nonre-
particularly good results on newly formed vessels sponsiveness to the treatment [61–63].
rather than already existing ones [35, 38, 39, 47,
48]. This discrepancy is theorized to occur as pre- cleral Lens Devices with the Use
S
existing vessels become less dependent on VEGF of Anti-VEGF Therapy
for growth and survival [49] and are covered by The use of scleral lens devices such as the pros-
pericytes [47, 50]. Moreover, bevacizumab thetic replacement of the ocular surface (PROSE)
effects were also greater when vessels were provide a reservoir for the antiangiogenic
smaller and less numerous and early in the CNV molecules so that it these are constantly bathing
course [38, 48, 51]. the ocular surface, thus increasing the bioavail-
In terms of route of administration, topical, ability of the drug. The PROSE lens is a large,
subconjunctival and intraocular formulations of rigid, gas-permeable, custom-designed lens that
vaults over the corneal surface. Despite its advan- mainly focused on basic science in animal mod-
tages, there is still scarcity of data observing out- els. It is vital to expand our knowledge regarding
comes on scleral lens or PROSE with corneal neovascularization in order to target more
Bevacizumab prior to corneal transplants. On an specific molecular angiogenic pathways and
interventional case series of 13 patients with cor- reduce the burden of a worldwide health
neal neovascularization, one drop of 1% bevaci- problem.
zumab was instilled each morning in the fluid
reservoir of the PROSE device. This was repeated Take Home Notes
after 6 h with a second drop of bevacizumab for a • Corneal neovascularization is a pathologic
median of 6 months. The response was highly phenomenon that can happen as a result of dif-
positive as 12/13 cases presented regression of ferent infectious or immune-related insults to
the neovascularization and 10/13 improved their the ocular surface.
visual acuity [64]. Another case series of five • Corneal neovascularization treatment prior to
patients with a similar protocol had similar results corneal transplantation has been shown to
and followed patients for up to 2 years without increase the longevity of the graft and prevent
noting epithelial side effects [65]. Although nei- rejection.
ther of these studies tackled corneal transplanta- • Fine-needle diathermy is a cheap and effective
tion outcomes in these populations, their results method that uses thermal energy to cauterize
suggest that the use of bevacizumab with scleral vessels.
lens devices may improve the success of grafts if • Laser therapy is a simple and tolerable proce-
surgery is undertaken. dure but is less effective on sizeable
neovascularization.
• Anti-VEGF local therapy, albeit costly, is a
Conclusion very effective method to treat corneal neovas-
cularization and offers several modes of
Corneal neovascularization can appear as a administration.
pathologic response to different factors affecting • Novel scleral lens devices bathe the corneal
the ocular surface, such as infectious, immune- surface with anti-VEGF therapy and increase
mediated and mechanical effects. It is important the bioavailability of the therapy, but they are
to treat CNV prior to corneal transplants, espe- expensive.
cially on penetrating keratoplasties, as it has been
shown to increase graft longevity and prevent
rejection. Prevention of CNV is very challenging,
but many different modalities have been used to References
treat it prior to PK. Many surgical and medical
therapies have offered significant therapeutic 1. Gupta D, Illingworth C. Treatments for cor-
neal neovascularization: a review. Cornea.
benefits for the management of CNV prior to PK, 2011;30(8):927–38.
mainly the use of fine-needle diathermy, lasers 2. Lee P, Wang CC, Adamis AP. Ocular neovascular-
and medical management with antiangiogenic ization: an epidemiologic review. Surv Ophthalmol.
treatments such as anti-VEGF antibodies and 1998;43(3):245–69.
3. Chang JH, Gabison EE, Kato T, Azar DT. Corneal
coupling it with scleral lens devices. It is neces- neovascularization. Curr Opin Ophthalmol.
sary to address the clinical benefits of each ther- 2001;12(4):242–9.
apy prior to initiating it, but also to target the 4. Cursiefen C, Küchle M, Naumann GO. Angiogenesis
specific side effects discussed in this chapter that in corneal diseases: histopathologic evaluation of
254 human corneal buttons with neovascularization.
are associated with the different modalities in Cornea. 1998;17(6):611–3.
question, in order to optimize patient response 5. Mayer DJ, Casey TA. Reducing the risk of corneal
and promote corneal graft survival. Overall, graft rejection. A comparison of different methods.
research done on corneal neovascularization is Cornea. 1987;6(4):261–8.
6. The collaborative corneal transplantation studies 21. Parsa CF, Temprano J, Wilson D, Green
(CCTS). Effectiveness of histocompatibility matching WR. Hemorrhage complicating YAG laser feeder
in high-risk corneal transplantation. The Collaborative vessel coagulation of cornea vascularization. Cornea.
Corneal Transplantation Studies Research Group. 1994;13(3):264–8.
Arch Ophthalmol. 1992;110(10):1392–403. 22. Baer JC, Foster CS. Corneal laser photocoagulation for
7. Dana MR, Schaumberg DA, Kowal VO, Goren MB, treatment of neovascularization. Efficacy of 577 nm
Rapuano CJ, Laibson PR, et al. Corneal neovascu- yellow dye laser. Ophthalmology. 1992;99(2):173–9.
larization after penetrating keratoplasty. Cornea. 23. Marsh RJ. Argon laser treatment of lipid keratopathy.
1995;14(6):604–9. Br J Ophthalmol. 1988;72(12):900–4.
8. Maguire MG, Stark WJ, Gottsch JD, Stulting 24. Amano S, Rohan R, Kuroki M, Tolentino M, Adamis
RD, Sugar A, Fink NE, et al. Risk factors for cor- AP. Requirement for vascular endothelial growth
neal graft failure and rejection in the collabora- factor in wound- and inflammation-related cor-
tive corneal transplantation studies. Collaborative neal neovascularization. Invest Ophthalmol Vis Sci.
Corneal Transplantation Studies Research Group. 1998;39(1):18–22.
Ophthalmology. 1994;101(9):1536–47. 25. Andreoli CM, Miller JW. Anti-vascular endothelial
9. Vail A, Gore SM, Bradley BA, Easty DL, Rogers growth factor therapy for ocular neovascular disease.
CA. Corneal graft survival and visual outcome. A mul- Curr Opin Ophthalmol. 2007;18(6):502–8.
ticenter study. Corneal Transplant Follow-up Study 26. Cursiefen C, Rummelt C, Küchle
Collaborators. Ophthalmology. 1994;101(1):120–7. M. Immunohistochemical localization of vascular
10. Jabbehdari S, Rafii AB, Yazdanpanah G, Hamrah P, endothelial growth factor, transforming growth fac-
Holland EJ, Djalilian AR. Update on the manage- tor alpha, and transforming growth factor beta1 in
ment of high-risk penetrating keratoplasty. Curr human corneas with neovascularization. Cornea.
Ophthalmol Rep. 2017;5(1):38–48. 2000;19(4):526–33.
11. Soifer M, Mousa HM, Stinnett SS, Galor A, Perez 27. Gan L, Fagerholm P, Palmblad J. Vascular endothelial
VL. Matrix metalloproteinase 9 positivity predicts growth factor (VEGF) and its receptor VEGFR-2 in the
long term decreased tear production. Ocul Surf. regulation of corneal neovascularization and wound
2021;19:270–4. healing. Acta Ophthalmol Scand. 2004;82(5):557–63.
12. Soifer M, Mousa HM, Levy RB, Perez 28. Hosseini H, Nejabat M, Mehryar M, Yazdchi T,
VL. Understanding immune responses to surgical Sedaghat A, Noori F. Bevacizumab inhibits corneal
transplant procedures in Stevens Johnsons Syndrome neovascularization in an alkali burn induced model
Patients. Front Med. 2021;8:656998. https://www. of corneal angiogenesis. Clin Exp Ophthalmol.
frontiersin.org/article/10.3389/fmed.2021.656998. 2007;35(8):745–8.
13. Pillai CT, Dua HS, Hossain P. Fine needle diathermy 29. Kim B, Tang Q, Biswas PS, Xu J, Schiffelers RM, Xie
occlusion of corneal vessels. Invest Ophthalmol Vis FY, et al. Inhibition of ocular angiogenesis by siRNA
Sci. 2000;41(8):2148–53. targeting vascular endothelial growth factor pathway
14. Kirwan RP, Zheng Y, Tey A, Anijeet D, Sueke H, Kaye genes: therapeutic strategy for herpetic stromal kera-
SB. Quantifying changes in corneal neovasculariza- titis. Am J Pathol. 2004;165(6):2177–85.
tion using fluorescein and indocyanine green angiog- 30. Philipp W, Speicher L, Humpel C. Expression of
raphy. Am J Ophthalmol. 2012;154(5):850–858.e2. vascular endothelial growth factor and its receptors
15. Junghans BM, Collin HB. The limbal vascular in inflamed and vascularized human corneas. Invest
response to corneal injury. An autoradiographic study. Ophthalmol Vis Sci. 2000;41(9):2514–22.
Cornea. 1989;8(2):141–9. 31. Heier JS, Antoszyk AN, Pavan PR, Leff SR, Rosenfeld
16. Feizi S, Azari AA, Safapour S. Therapeutic PJ, Ciulla TA, et al. Ranibizumab for treatment of
approaches for corneal neovascularization. Eye Vis neovascular age-related macular degeneration: a
Lond Engl. 2017;4:28. phase I/II multicenter, controlled, multidose study.
17. Faraj LA, Elalfy MS, Said DG, Dua HS. Fine nee- Ophthalmology. 2006;113(4):633.e1–4.
dle diathermy occlusion of corneal vessels. Br J 32. Cursiefen C, Cao J, Chen L, Liu Y, Maruyama K,
Ophthalmol. 2014;98(9):1287–90. Jackson D, et al. Inhibition of hemangiogenesis and
18. Romano V, Steger B, Kaye SB. Fine-needle dia- lymphangiogenesis after normal-risk corneal trans-
thermy guided by angiography. Cornea. 2015;34(9): plantation by neutralizing VEGF promotes graft sur-
e29–30. vival. Invest Ophthalmol Vis Sci. 2004;45(8):2666–73.
19. Spiteri N, Romano V, Zheng Y, Yadav S, Dwivedi 33. Yatoh S, Kawakami Y, Imai M, Kozawa T, Segawa
R, Chen J, et al. Corneal angiography for guid- T, Suzuki H, et al. Effect of a topically applied
ing and evaluating fine-needle diathermy treatment neutralizing antibody against vascular endothelial
of corneal neovascularization. Ophthalmology. growth factor on corneal allograft rejection of rat.
2015;122(6):1079–84. Transplantation. 1998;66(11):1519–24.
20. Wertheim MS, Cook SD, Knox-Cartwright NE, Van 34. Rodrigues EB, Farah ME, Maia M, Penha FM,
DL, Tole DM. Electrolysis-needle cauterization of Regatieri C, Melo GB, et al. Therapeutic monoclonal
corneal vessels in patients with lipid keratopathy. antibodies in ophthalmology. Prog Retin Eye Res.
Cornea. 2007;26(2):230–1. 2009;28(2):117–44.
35. Mackenzie SE, Tucker WR, Poole multiple models of ocular neovascularization. Am J
TRG. Bevacizumab (avastin) for corneal neovascu- Pathol. 2006;168(6):2036–53.
larization--corneal light shield soaked application. 50. Ferrari G, Dastjerdi MH, Okanobo A, Cheng SF,
Cornea. 2009;28(2):246–7. Amparo F, Nallasamy N, et al. Topical ranibizumab
36. Gerten G. Bevacizumab (avastin) and argon laser to as a treatment of corneal neovascularization. Cornea.
treat neovascularization in corneal transplant surgery. 2013;32(7):992–7.
Cornea. 2008;27(10):1195–9. 51. Lin CT, Hu FR, Kuo KT, Chen YM, Chu HS, Lin
37. DeStafeno JJ, Kim T. Topical bevacizumab therapy YH, et al. The different effects of early and late beva-
for corneal neovascularization. Arch Ophthalmol. cizumab (Avastin) injection on inhibiting corneal
2007;125(6):834–6. neovascularization and conjunctivalization in rab-
38. Awadein A. Subconjunctival bevacizumab for vas- bit limbal insufficiency. Invest Ophthalmol Vis Sci.
cularized rejected corneal grafts. J Cataract Refract 2010;51(12):6277–85.
Surg. 2007;33(11):1991–3. 52. Avisar I, Weinberger D, Kremer I. Effect of subcon-
39. Bock F, König Y, Kruse F, Baier M, Cursiefen junctival and intraocular bevacizumab injections on
C. Bevacizumab (Avastin) eye drops inhibit cor- corneal neovascularization in a mouse model. Curr
neal neovascularization. Graefes Arch Clin Exp Eye Res. 2010;35(2):108–15.
Ophthalmol. 2008;246(2):281–4. 53. Lee SH, Leem HS, Jeong SM, Lee K. Bevacizumab
40. Dastjerdi MH, Al-Arfaj KM, Nallasamy N, Hamrah accelerates corneal wound healing by inhibiting TGF-
P, Jurkunas UV, Pineda R, et al. Topical bevacizumab beta2 expression in alkali-burned mouse cornea.
in the treatment of corneal neovascularization: results BMB Rep. 2009;42(12):800–5.
of a prospective, open-label, noncomparative study. 54. Ozdemir O, Altintas O, Altintas L, Ozkan B, Akdag C,
Arch Ophthalmol. 2009;127(4):381–9. Yüksel N. Comparison of the effects of subconjuncti-
41. Doctor PP, Bhat PV, Foster CS. Subconjunctival val and topical anti-VEGF therapy (bevacizumab) on
bevacizumab for corneal neovascularization. Cornea. experimental corneal neovascularization. Arq Bras
2008;27(9):992–5. Oftalmol. 2014;77(4):209–13.
42. Manzano RPA, Peyman GA, Khan P, Carvounis PE, 55. Mohammadpour M. Deep intrastromal injec-
Kivilcim M, Ren M, et al. Inhibition of experimental tion of bevacizumab for the management of cor-
corneal neovascularisation by bevacizumab (Avastin). neal neovascularization. Cornea. 2013;32(1):
Br J Ophthalmol. 2007;91(6):804–7. 109–10.
43. Bachmann BO, Bock F, Wiegand SJ, Maruyama K, 56. Kim T-i, Chung JL, Hong JP, Min K, Seo KY, Kim
Dana MR, Kruse FE, et al. Promotion of graft sur- EK. Bevacizumab application delays epithelial heal-
vival by vascular endothelial growth factor a neutral- ing in rabbit cornea. Invest Ophthalmol Vis Sci.
ization after high-risk corneal transplantation. Arch 2009;50(10):4653–9.
Ophthalmol. 2008;126(1):71–7. 57. Koenig Y, Bock F, Horn F, Kruse F, Straub K,
44. Habot-Wilner Z, Barequet IS, Ivanir Y, Moisseiev Cursiefen C. Short- and long-term safety profile and
J, Rosner M. The inhibitory effect of different con- efficacy of topical bevacizumab (Avastin) eye drops
centrations of topical bevacizumab on corneal neo- against corneal neovascularization. Graefes Arch Clin
vascularization. Acta Ophthalmol. 2010;88(8): Exp Ophthalmol. 2009;247(10):1375–82.
862–7. 58. Kim SW, Ha BJ, Kim EK, Tchah H, Kim T-i. The
45. Han YS, Lee JE, Jung JW, Lee JS. Inhibitory effects effect of topical bevacizumab on corneal neo-
of bevacizumab on angiogenesis and corneal neovascularization. Ophthalmology. 2008;115(6):
vascularization. Graefes Arch Clin Exp Ophthalmol. e33–8.
2009;247(4):541–8. 59. Murata M, Takanami T, Shimizu S, Kubota Y, Horiuchi
46. Kim TI, Kim SW, Kim S, Kim T, Kim EK. Inhibition S, Habano W, et al. Inhibition of ocular angiogenesis
of experimental corneal neovascularization by using by diced small interfering RNAs (siRNAs) specific to
subconjunctival injection of bevacizumab (Avastin). vascular endothelial growth factor (VEGF). Curr Eye
Cornea. 2008;27(3):349–52. Res. 2006;31(2):171–80.
47. Cursiefen C, Hofmann-Rummelt C, Küchle M, 60. Carrasco MA. Subconjunctival bevacizumab for cor-
Schlötzer-Schrehardt U. Pericyte recruitment in neal neovascularization in herpetic stromal keratitis.
human corneal angiogenesis: an ultrastructural study Cornea. 2008;27(6):743–5.
with clinicopathological correlation. Br J Ophthalmol. 61. Chames P, Van Regenmortel M, Weiss E, Baty
2003;87(1):101–6. D. Therapeutic antibodies: successes, limita-
48. You IC, Kang IS, Lee SH, Yoon KC. Therapeutic tions and hopes for the future. Br J Pharmacol.
effect of subconjunctival injection of bevacizumab 2009;157(2):220–33.
in the treatment of corneal neovascularization. Acta 62. Bergers G, Hanahan D. Modes of resistance
Ophthalmol. 2009;87(6):653–8. to anti-angiogenic therapy. Nat Rev Cancer.
49. Jo N, Mailhos C, Ju M, Cheung E, Bradley J, 2008;8(8):592–603.
Nishijima K, et al. Inhibition of platelet-derived 63. Tranos P, Vacalis A, Asteriadis S, Koukoula S,
growth factor B signaling enhances the efficacy of Vachtsevanos A, Perganta G, et al. Resistance to
anti-vascular endothelial growth factor therapy in antivascular endothelial growth factor treatment in
age-related macular degeneration. Drug Des Dev delivery system for bevacizumab. Semin Ophthalmol.
Ther. 2013;7:485–90. 2009;24(3):149–55.
64. Yin J, Jacobs DS. Long-term outcome of using
Prosthetic Replacement of Ocular Surface Ecosystem
(PROSE) as a drug delivery system for bevacizumab
in the treatment of corneal neovascularization. Ocul Suggested Reading
Surf. 2019;17(1):134–41.
65. Lim M, Jacobs DS, Rosenthal P, Carrasquillo KG. The Perez VL. Corneal vessel cauterization. Durham: Duke
Boston Ocular Surface Prosthesis as a novel drug Eye Center; 2022.
Part II
Penetrating Keratoplasty
Main Issues to Overcome
in Modern Penetrating
5
Keratoplasty
Farideh Doroodgar, Sana Niazi, Hassan Hashemi,

and Mohammad Ali Javadi
Key Points raising eye pressure and hastening infectious

• Immunologic rejection and endothelial crystalline keratopathy (ICK) development,
decompensation are two important causes of removing any loose or vascularized sutures as
graft failure. soon as possible are measures that can be taken
• Special attention should be paid to the condi- to increase the chances of survival in regraft.
tion of the Ocular Surface when deciding on • In this chapter, we discuss the pitfalls of mod-
Penetrating keratoplasty as it may negatively ern penetrating keratoplasty (PK) along with
influence its outcomes significantly. other issues and techniques in order to achieve
• The large enough second graft with no remnants the best possible results by taking adequate
of the first transplant, systemic steroid, and action on the variables that negatively influ-
immunosuppressive medications before surgery ence the outcomes of PKP, avoiding or con-
and continue for a long time afterward, avoiding trolling the associated risks.
Supplementary Information The online version con-

org/10.1007/978-3-031-32408-6_5.
F. Doroodgar (*) · S. Niazi Eye Research Center, Farabi Eye Hospital, Tehran
Translational Ophthalmology Research Center, University of Medical Sciences, Tehran, Iran
Tehran University Medical Science, Tehran, Iran e-mail: [email protected]
Negah Aref Ophthalmic Research Center, Shahid M. A. Javadi
Beheshti University of Medical Science, Tehran, Iran Department of Ophthalmology, Labbafinezhad
e-mail: [email protected]; Hospital, Shahid Beheshti University of Medical
[email protected] Sciences, Tehran, Iran
H. Hashemi
Noor Ophthalmology Research Center, Noor Eye
Hospital, Tehran, Iran

https://doi.org/10.1007/978-3-031-32408-6_5
72 F. Doroodgar et al.
Introduction improved outcomes in PK and other surgical

procedures [5]. However, like any other surgery,
Corneal blindness is a significant cause of revers- this type of surgery has complications, including
ible blindness, making corneal transplantation minor (astigmatism, slow rehabilitation, and
the most prevalent type of human organ trans- prolonged use of topical steroids) and major
plantation. Corneal transplantation, performed complications (graft failure). Considering the
for therapeutic, tectonic, and optical purposes, is diversity of operative incisions in different kera-
thus a procedure that successfully restores vision toplasty modifications, clarification is required
in thousands of patients each year. Different sur- to evaluate their postoperative complications;
gical techniques have been suggested for corneal therefore, in the present chapter, we discuss the
transplantation; penetrating keratoplasty (PK) pitfalls of modern PK alongside other issues and
ranks among the oldest and most common surgi- techniques.
cal techniques, which offers good long-term First, we discuss the reasons for early and late
visual rehabilitation, most suitable in the pres- graft rejection after penetrating keratoplasty,
ence of endothelial dysfunction or severe deep which can help a better comprehension of the
corneal scarring, affecting the visual axis up to mechanisms of this complication. Regarding the
the Descemet membrane (DM) level. Other sur- direct association of visual acuity after PK with
gical techniques have also been proposed for patients’ quality of life [6], we pay attention to
reducing complications in special indications [1]. the short-term and long-term visual outcomes of
In this chapter, we focus on the pitfalls of modern PK in section “The Short-Term and Long-Term
PK and technical issues of the trephination, sutur- Functional and Refractive Outcomes of
ing technique, and immunological aspects in a Penetrating Keratoplasty”. In section “Clinical
review of randomized controlled trials of pene- Aspects of Graft Failure After Penetrating
trating keratoplasty techniques [2]. There was no Keratoplasty”, we focus on graft rejection and
indication that any one technique was superior in failure, which is considered the measure of surgi-
increased quality of life or cost-effectiveness. cal success, and explain the risk factors of graft
In modern keratoplasty, selectivity is the main failure after PK. Knowing about the risk factors
principle, i.e., whether to replace only the can provide a better understanding of the indica-
affected corneal layer through layer-by-layer or tions of PK and the implementation of effective
interlayer transplantation. One of the important prevention and treatments in high-risk patients
modern advances that aim to improve clinical [7]. sections “Management of Postpenetrating
outcomes is replacing manual surgical incision Keratoplasty Astigmatism” and “Postoperative
of the cornea with new techniques, like a laser. Care” include postoperative care and
Production of surgical graft profiles such as management.
“top-hat,” “mushroom,” and “zig-zag” shapes
(which increases the wound surface area,
decreases surgically associated astigmatism and he Reasons for Early and Late
T
nonastigmatic aberrations, and thus improves Graft Rejection After Penetrating
wound healing and biomechanical stability), Keratoplasty
higher precision, the inherent automatability and
repeatability (which eases the learning curve and Graft rejection is the most common cause of cor-
decreases complications), earlier suture removal, neal graft failure, defined as the irreversible loss
and improved visual outcomes (short- and long- of refractive quality. Immunologic rejection
term) are the main advantages of using an infra- (allograft) is the leading cause of graft failure.
red femtosecond laser (FSL) [3, 4]. The use of More than half the cases of graft decompensation
other lasers, such as nonmechanical excimer- are caused by allograft rejection [8]. Another
assisted lasers and pulsed ultraviolet light from important cause of graft failure is endothelial
excimer lasers, has also been announced with decompensation.
5 Main Issues to Overcome in Modern Penetrating Keratoplasty 73
Immunological Rejection are the presence of stromal blood vessels in one or

more quadrants of the recipient cornea (high risk),
Allograft rejection after PK is the main cause of preoperative glaucoma, young age, prior anterior
graft failure. Despite immunosuppressive agents segment surgery, active ocular inflammation, ocu-
prescribed to high-risk patients, immunologic lar surface disease, herpes simplex keratitis, neu-
rejection occurs by destroying the donor tissue by rotrophic keratopathy, large and eccentric grafts,
the host’s immune system. This causes reversible and anterior synechiae. Of note, all studies report
or irreversible damage to the grafted cornea [9]. a high rate for other or undetected causes, which
A corneal graft failure is result of an activated shows that the cause of graft failure is still not
sequence of complex immune responses. The well-established [10]. Others have reported the
host immune system has an efferent immune main risk factors of immunologic rejection as new
response against these foreign antigens. This vascularization of the recipient cornea over two or
response culminates in rejection and graft decom- more quadrants, corneal opacity due to an infec-
pensation, resulting in irreversible damage to the tious origin, posttraumatic corneal opacity or con-
cornea [9]. The details of the cellular mecha- genital glaucoma, graft diameter >8 mm, and a
nisms of immunologic rejection are described therapeutic indication of PK [17]. In another
elsewhere [9, 10]. Preoperative major histocom- study, the risk factors of allograft rejection after
pability complex (MHC) and non-MHC antigens PK included corneal vascularization, long opera-
matching, and new treatment modalities are still tion time, and younger donor age [18]. The pre-
controversial and under investigation. sented risk factors for graft failure and allograft
rejection should be considered by surgeons when
contemplating PK.
Endothelial Cell Failure Graft survival should be interpreted with cau-
tion when comparing different studies to assess
The most common form of graft rejection is the success of corneal transplantation. There is
endothelial rejection, while epithelial and subepi- no consensus on graft failure; some define it as
thelial rejections are infrequent. Epithelial rejec- the loss of optical clarity, whereas others have a
tion occurs in roughly 2% of graft rejections, variable effect on vision. Late endothelial fail-
beginning days to weeks earlier as a line (which ure’s gradual loss of clarity could give an uncer-
consists of lymphocytes, plasma cells, and neu- tain endpoint.
trophils) located near engorged limbal vessels. Measures that can be taken to increase the
The commonest consists of a line of Keratic chances of survival of corneal transplantation
Precipitates (KP) beginning inferiorly at the in those who have a second operation are:
graft-host junction and marching superiorly [11, If the immunological system is to blame for
12]. Molecular evidence shows a steadily the first operation’s success:
decreasing trend of endothelial cell density 1. The second operation should be performed as
(ECD) after PK, predictive of graft failure due to soon as feasible after the first, and no evidence
endothelial decompensation [13, 14]. of inflammation should be seen throughout
Other reasons have also been described for the procedure.
corneal graft failure, which serves as risk factors 2. The second graft should be large enough to
rather than causes; factors such as glaucoma, non- incorporate all of the previously grafted tis-
viral infections, endothelial cell failure, viral her- sue, with no remnants of the first transplant
petic infections [15], intraocular pressure (IOP) unless the previous transplantation was, par-
elevation/glaucoma, diseases of the ocular sur- ticularly huge and off-center.
face, recurrence of the primary disease, wound 3. Systemic steroid and immunosuppressive
dehiscence/hypotonia and trauma [16], uncorrect- medications should be started 2–7 days before
able refractive error, and primary donor failure surgery and continued for a long time
[17]. Risk factors for corneal allograft rejection afterward.
4. After surgery, immunosuppressive drops such Following transplantation, treatments such

as tacrolimus, in addition to topical steroids, as gene therapy may be used. Because of its
should be given indefinitely as supplementary inexpensive cost, minor histocompatibility
therapy. At the same time, caution should be complex tissue matching can be performed.
exercised to avoid raising eye pressure and ABO antigen testing, on the other hand, is not
hastening ICK development. as specific as significant histocompatibility
5. After surgery, any loose or vascularized tests. The cornea remains the only organ with
sutures should be removed as soon as possi- the unique immune privilege to assess major
ble, and the patient should have unrestricted histocompatibility complex tissue matching
access to the treating physician. [15].
Specifically, most major histocompatibility
Topical administration of Adhesion Molecules testing only works for class II and has not been
AB enhanced graft survival in animal experi- demonstrated successful for class I. Unfortunately,
ments, but this has yet to be investigated in because significant human histocompatibility
humans. Several adhesion molecules are impli- complex genes are highly polymorphic, any ran-
cated in corneal transplant rejection, two of dom allocation of human leukocyte antigens
which are VLA-4 (Very Late Antigen) and LFA-1 (HLA) will take a very long time to obtain the
(Leukocyte Function Adhesion) [19]. required matching level, which is highly unethi-
cal for our patients [15]. Suppose a highly spe-
cific, low-cost, and time-efficient HLA matching
cular Surface Problems and Dry Eye
O is developed in the future. In that case, HLA test-
Disease ing may be used as a routine evaluation to
increase the number of grafts that survive in
One crucial reason for graft rejection is ocular transplant recipients.
surface problems and dry eye disease (DED), As noted in the literature, high-risk situations
affecting millions globally. Immunopathological such as corneal vascularization, DED, and pro-
graft failure is one of the most prevalent ocular longed use of antiglaucoma medicine can lower
surface problems. Hyperosmolarity is also cru- the corneal graft survival rate [24]. In light of
cial in dry eye disease’s inflammatory cycle [20]. this, the injection of anti–vascular endothelial
Maintaining a healthy and moist ocular sur- growth factor therapy (anti-VEGF) [25] into the
face depends on healthy epithelia, tear film, and subconjunctival, the use of Lifitegrast [26], an
eyelid, increasing corneal graft survival. Several antagonist of LFA-1 and an inhibitor of T cell
previous studies revealed that dry eye disease, as formation, in dry eye management as well as
a form of the ocular surface disorder (OSD), can subconjunctival, an adjunct nonpreservative top-
lead to graft rejection and has two known causes, ical lubricant in glaucoma treatment [27], will
lipid or aqueous tear deficiency [21, 22]. Mild likely be beneficial and produce a promising
inflammatory processes can progress to serious result related to a higher graft survival rate in the
chronic diseases (i.e., cicatrizing conjunctivitis) future.
that are contraindications for whole penetrating The aforementioned items conclude that a
keratoplasty. prospective clinical trial examining the impact of
The goal of managing dry eye or OSD is to preexisting DED on corneal transplant survival is
manage the tear film’s hyperosmolarity state to warranted. An immunological understanding of
decrease the immune system’s expression of a HLA’s function in corneal graft rejection and a
response to any foreign substances and inflam- thorough understanding of the need for HLA tis-
matory antigens. Although the cornea has immu- sue matching will open up new avenues for pre-
nological privilege because of its avascular and venting graft rejection. Due to continuing
lymph node-free nature, effective immune rejec- research into potential pharmacotherapies with
tion avoidance is preferable to immune suppres- novel targets, several viable therapy options for
sion with immune modifying agents [23]. DED are envisaged.
he Short-Term and Long-Term

T recommended protocols have been discussed in a
Functional and Refractive section on postoperative management of pene-
Outcomes of Penetrating trating keratoplasty [36–38].
Keratoplasty The new FSL technique, which cuts both
recipient and donor within a liquid interface,
While anatomic success is relatively frequent, could solve this problem. Excimer-assisted treph-
functional failure is relatively common. A sys- ination has shown better alignment in all sutures-
tematic review of 13 studies (530 eyes) on out keratoplasty patients. It leaves the cornea’s
visual outcomes of PK reported mean best-cor- curvature undisturbed, reducing shear and com-
rected visual acuity (BCVA) in the logarithm of pression artifacts in the tissue [5, 36, 39].
the minimum angle of resolution (Log-MAR) FS technology has enabled perfect limbal-
within the range of 0.05–0.40 Log-MAR; nota- oriented centration through optical coherence
bly, studies with longer follow-ups (>2 years) tomography. This solves the problem of centra-
have reported a mean BCVA of 0.05 or 0.1 Log- tion with trephination in the recipient’s eye. FSL
MAR [28]. Fukuoka reported mean BCVA of can use side-cut profiles that improve the fitting
1.54 ± 0.68, 0.06 ± 0.22, 0.03 ± 0.17, and between donor disc and recipient bed. They
0.14 ± 0.42 Log-MAR after 10 years, 20 years, combine the optical benefits of PK and increased
and 25 years, respectively [21, 29]. A mean visual rehabilitation [4, 40]. Different side-cut
follow-up of 14 years also revealed 73.2% of profiles can be chosen in FSL PK; mushroom-
patients having a BSCVA of 20/40 or better and shaped incisions can be performed manually
an open-angle glaucoma rate of 5.4% [30, 31]. with an FSL or a microkeratome, but a zig-zag
The application of new microkeratome-assisted incision can only be performed with FSL
with transplantation of a two-piece mushroom- (Fig. 5.2).
shaped graft showed favorable visual outcomes The mushroom PK aims to minimize the
comparable to conventional PK; BCVA of replacement of the recipient healthy endothe-
20/40 and 20/20 were observed in 100% and lium (typically keratoconus eyes) and maintain a
>50% of eyes, respectively [32]. Studies com- large diameter of the superficial, refractive part
paring the quality of vision after PK with other of the graft for optimizing the postoperative
surgical techniques, such as deep anterior refractive results. The top-hat design was
lamellar keratoplasty (DALK), have also intended to also address the advantage of supply-
reported better BCVA after PK in the long run ing a larger amount of donor cells to eyes with
[28, 33]. The difference in techniques used dur- decompensated endothelial cells [41–44]. The
ing modern PK can also impact the visual potential advantages of these profiles include
outcome. improved graft adaptation, better and more sta-
At least 1 year is usually required for the PK ble wound healing leading to earlier suture
wound to heal sufficiently. Up to 30% of eyes removal, and eventually prolonged graft sur-
still have astigmatism, uncorrectable with spec- vival; however, none of these potential advan-
tacles, and considered an important cause of poor tages have been approved by reliable studies
visual outcomes of keratoplasty [34, 35]. The until now, except for more favorable visual out-
incidence of astigmatism can be reduced by bet- comes [45].
ter separation of the corneal button from the cor- Apart from the visual acuity of the patients
nea of donor and recipient, termed trephination. after PK, most of which can be corrected by spec-
Trephination systems include handheld, motor tacles, an important outcome after PK is graft
trephine, excimer laser, or FSL-based (Fig. 5.1; failure/rejection, the clinical aspects of which are
Video 5.1). The modifications of the different explained in the following.
76
Types of Trephines
Mechanical Trephines Nonmeehanieal Trephines
Freestanding Blade/Hand- Motor Trephines (Mikro- Suction Trephines Guided Trephines (GTs,
Types Advantages
Held Trephines Keratron,Asmotom) (Hessburg-Barron) Hanna)
The 193-nm Excimer

1. No trauma to intraocular tissues
Laser
2. Avoid deformation and
compression of tissue during
trephination
3. Reduction of horizontal torsion
(”Erlangen orientation teeth/notches”)
The 2.94-µm 4. Reduction of vertical tilt (congruent
Erbium-Y AG Laser cut edges)
5. Reduction of host and donor
decentration
6. Feasibility of “harmonization” of
donor and host topography
The Femtosecond Laser 7. Reduction of anterior chamber
inflammation early after PKP
8. Reduction of astigmatism after
suture removal
9. Higher regularity of corneal
topography
10. Significantly better visual acuity
with spectacle correction
I l. Feasibility of trephination with
unstable cornea (e.g., “open eye”,
descemetocele, after radial
keratotomy, iatrogenic keratectasia
after LASIK)
12. Arbitrary shape (e,g., elliptical)
F. Doroodgar et al.
Fig. 5.1 Different types of trephines

a d
b e
Fig. 5.2 The shapes of sidecuts used in laser-assisted keratoplasty. Standard cut (a), top hat (b), mushroom (c), zig-zag
(d), and Christmas tree (e)
linical Aspects of Graft Failure

C Graft survival of PK has been compared with
After Penetrating Keratoplasty other keratoplasty procedures, such as DALK,
and some have claimed worse graft survival of
In PK, all five layers of the cornea are removed cornea after PK compared with DALK. The use
and replaced with donor tissue. The rejection of new microkeratome-assisted transplantation of
rates reported for PK range from 5.8% to 41%, a two-piece mushroom-shaped graft showed a
depending on the duration of follow-up. graft immunologic rejection rate of <5% and
Rejection can occur a few weeks after a cornea graft survival >95% for 5 years. Any independent
transplant, but it is more common after several postoperative risk factors, including infection,
months [46, 47], while others have reported rejection, disease recurrence, eyelid or glaucoma
favorable short-term (91% survival at 1 year) surgery, or a repeat graft, decreased graft survival
[48] and long-term outcomes (25-year graft to 34% at 10 years [32, 48, 51, 52].
survival rate of 85.4% [31] and 98.8%, 97.0%,
and 93.2% at 10 years, 20 years, and 25 years
after surgery, respectively) [29]. One of the linical Aspects for Prevention
C
main causes of graft failure after PK is the and Treatment of Corneal Transplant
progressive donor endothelial cell loss (ECL) Rejection
after PK, migration from areas with high to
low-density ECD. The higher probability of No standard or uniform guideline for preventing
ECL after PK for bullous keratopathy than transplant rejection in high-risk cases of corneal
after PK for keratoconus suggested the posi- transplantation exists. A history of HSK, a his-
tive role of a healthy recipient endothelium tory of rejection, and transplantation in actively
(with high ECD) on the long-term survival of infected eyes are all factors to consider before
donor tissue [49, 50]. surgery. Despite advances in anti-inflammatory
drug regimens, glaucoma surgery, and other more severe pain when there are more of
treatments, the high-risk group still faces a 60% them in the subbasal layer of the cornea.
chance of cornea transplantation. The main goal Low-graft inflammation can sometimes lead
of follow-up is to prevent immunological rejec- to rejection without the start of KP. KP can
tion reactions [53–55]. occasionally appear after ocular edema has
subsided. The appearance of subepithelial
infiltrates (SEIs) or stromal edema should be
reoperative Precautions to Reduce
P treated as fully as other full-thickness graft
the Risk of Transplant Rejection rejections. The incidence of immune reac-
tions was reported to be 10% in layered trans-
The use of tacrolimus as a topical or systemic plantation and 23% in complete
treatment for vernal keratoconjunctivitis is bene- transplantation in a 3-year study.
ficial. Stevens-Johnson, pemphigoid and
Mooren’s ulcer should be postponed for at least
1 year after the inflammation have been con- Angiogenesis Suppression
trolled. Blepharitis, eyelid inflammation, infec-
tion, and eyelid disorders like ectropion and Inflammation, corneal edema, limbus stem cell
entropion can all be treated. A stem cell trans- defects, and hypoxia are all factors that stimulate the
plant should be performed at least 3 months growth of blood vessels in the cornea. Corneal
before corneal transplant surgery if the patient growth appears to be influenced by angiogenesis.
has a large corneal stem cell defect (due to factors There is growth from both bone marrow precursors
such as chemical burns or atopic keratoconjuncti- and lymphatic vessels when it comes to lymphatic
vitis) [54]. vessel growth. Although VEGF is the most impor-
tant molecule in vascular endothelial cell prolifera-
tion, other factors such as nitric oxide and
he Ways of Diagnosing Corneal
T proinflammatory cytokines are also involved.
Transplant Rejection Although VEGF is the most important molecule in
vascular endothelial cell proliferation, other factors
1. Look into the patient’s complaints, such as such as nitric oxide, matrix metalloproteinases, pro-
eye pain or photosensitivity, as well as slit inflammatory cytokines, and some growth factors
lamp findings, like the presence of KP or cells (platelet-derived growth factor (PDGF) and fibro-
in the anterior chamber. The disadvantage of blast growth factor b (bFGF)) are also involved [60].
this method is that it does not allow for early Inatomi et al. introduced a clinical classifica-
detection of rejection before clinical rejection tion for corneal vascular growth that includes
occurs. four levels.
2. The number of KPs, which can only show an The first level is vascular growth in the corneal
increase in the thickness of the laryngeal and environment, then vascular growth in the
endothelial layers and cannot detect inflam- mid-periphery. In the third level, moderate vascu-
matory cells, is also being tracked [56–58]. lar growth is seen throughout the cornea, and at
3. In-Vivo Confocal Microscopy is able to show the end, severe vascular growth throughout the
microstructural changes in corneal cells and cornea [60].
shows an increase in ICs (Immune Cells, The location, depth, length, and diameter of
Activated Keratocytes) in the early stages of vessels, their branched appearance, and the state
transplant rejection. They can be found in all of blood flow in the arteries should all be consid-
layers of the cornea, particularly in the sub- ered when assessing corneal arteries. Long-term
basal region and endothelium [56–59]. There contact lens use after deep anterior lamellar kera-
is a relationship between clinical symptoms toplasty (DALK) is an example of lipid keratopa-
and IC (AK) density. The patient experiences thy [60, 61].
Treatment of Corneal Angiogenesis Various studies have suggested several meth-

ods for blocking blood vessels in addition to
The removal of the angiogenic stimulus is the pri- pharmacological treatments, including cryother-
mary treatment for corneal angiogenesis. Despite apy, laser thermal cauterization, fine needle dia-
the risk of side effects, topical corticosteroids are thermy (FND), and photodynamic therapy
still the first line of defense. Immunosuppressive (PDT). Although these methods are most effec-
drugs such as 0.05% cyclosporine A and tacroli- tive in treating adult arteries, they can also be
mus are other anti-inflammatory drugs [60]. used to treat young and immature arteries when
Cyclosporine A systemically inhibits endo- combined with monoclonal antibodies.
thelial cell migration and inhibits angiogenesis. Clinical studies have used vascular cauteriza-
There are conflicting studies on the effect of topi- tion with argon lasers and Nd:YAG lasers.
cal cyclosporine A [60]. Fine needle diathermy is effective in the treat-
Tacrolimus inhibits the production of cyto- ment of corneal arteries in 80% of cases and is an
kines by T lymphocytes. It also reduces the pro- effective, easy, safe, and inexpensive method. In
duction of immunoglobulins [60]. a study using the Nd: YAG laser, 53% of the
Anti-VEGF monoclonal antibodies, such as arteries were completely blocked after 3 months,
bevacizumab and ranibizumab, are another class while 37% reopened. Because fine needle dia-
of drugs used to treat corneal arteries. These thermy can increase VEGF production, it is best
drugs work by slowing the growth of young, combined with the topical application of anti-
active vessels while having no effect on mature VEGF monoclonal antibodies to reduce the risk
or older vessels. Bevacizumab drops (5 mg/mL of corneal recurrence [60, 69].
five times daily) have been shown in animal Verteporfin photodynamic therapy is a
studies to be effective in reducing corneal ves- method of selectively blocking corneal arteries.
sels that do not respond to anti-inflammatory However, this method is costly, and there is a risk
therapies [62]. of laser-related side effects as well as the genera-
Various studies have shown that subconjuncti- tion of oxygen-free radicals [70].
val injections of pegaptanib and aflibercept have Amniotic membrane transplantation
antiangiogenic effects [63, 64]. (AMT) is a surgical technique used to repair cor-
FD006 is a new monoclonal antibody with neal epithelial defects and reduce inflammation
potent antiangiogenic properties that have and angiogenesis. Less invasive surgical proce-
recently been introduced [65]. dures, such as sequential sector conjunctival epi-
The safety of using topical monoclonal anti- theliectomy (SSCE), have been used in cases of
bodies on the eye’s surface has been questioned. damage to part of the corneal stem cells. Corneal
Although there have been reports of delays in vascular growth was associated with this damage.
epithelial defect repair and increased expression Amniotic Membrane Transplantation has anti-
of matrix metalloproteinases after topical appli- inflammatory and antiangiogenic properties [71].
cation, there do not appear to be any serious side A conjunctival and limbus autograft
effects [60, 66]. (CLAU) is removed from one eye when only one
Drugs in the tetracycline family, such as mino- eye is affected, and the other eye is perfectly
cycline and doxycycline, have antiangiogenic healthy. It is the most effective treatment of cor-
properties in addition to anti-collagenase proper- neal stem cells associated with damage to more
ties. Recently, a gel and solution containing 2% than two-thirds of limbus stem cells. Another
doxycycline were developed, and antiangiogenic option is cultured limbal epithelial transplants,
effects were observed in animal and human stud- which are used for complete defects in the stem
ies [60, 67]. cells of one or both eyes. Living-related conjunc-
Fasudil is a Rho Kinase Inhibitor (ROCK tival limbal allograft (lr-CLAL), as well as ker-
Inhibitor) drug that has been shown to have anti- atolimbal allograft (KLAL), are recommended in
angiogenic properties in laboratory studies [68]. cases where there is a significant corneal stem
cell defect [60]. Other surgical procedures rec- such as age ranges compared or considering age
ommended for stem cell defects include simple as a continuous variable; also, the results can dif-
limbal epithelial transplant (SLET) and culti- fer based on the preoperative estimation of the
vated oral mucosal epithelial transplantation likelihood of graft survival [73, 77].
(COMET). The advantage of autograft methods There was no significant difference in results
over allogeneic methods is that there is no risk of between PK and DALK in pediatric keratoconus.
allogeneic transplant rejection, and there is no Low-quality donor tissues increased the inci-
need for suppressive drugs in the postoperative dence of graft epithelium abnormalities. With
treatment regimen [72]. adequate therapy, graft clarity and visual results
The blood vessels in the cornea should be may be good. In chronic or delayed-onset mus-
investigated to determine the source of their pres- tard gas keratitis, PKP should be considered a
ence. Anti-herpes medication is required in addi- high-risk transplant1 [80, 81] (Fig. 5.3).
tion to transplant rejection treatment. Obviously,
the blood vessels will regress as well, but if the Donor’s Sex
blood vessels remain active despite immunologi- The higher risk of failure in female patients (haz-
cal treatment, cautery or a topical injection of ard ratio = 1.42) has also been reported as the
2.5 mg bevacizumab 0.1% should be employed spurious result of analysis with multiple compar-
as an additional therapy. If the cause is a specific isons [74, 75]. Female grafts from male donors
underlying condition (e.g., HSK), anti-herpes may be subject to alloimmune reactivity in
medication is required in addition to transplant female recipients, as Y gene antigens are only not
rejection treatment. expressed in females. Lass and colleagues have
also observed that the higher ECD in females at
baseline did not influence the failure of PK at
The Risk Factors of Graft Failure 5 years [79, 82, 83].
The risk factors of graft failure after PK include Graft Size

the following. A study by Barraquer and colleagues shows graft
diameters of 7.0–7.4 and 8.0 mm had the best
raft Characteristics: Donor’s Age
G 10-year survival estimate (70% and 69%) in PKs
and Sex, Graft Size, and Eye Banking larger than 8.75 mm. The mechanism of graft fail-
Practices ure and rejection is supposed to be the proximity
of lumbar vasculature to conjunctival vessels and,
Donor’s Age thus, antigenic materials, which makes immuno-
The effect of donor’s sex and age on the graft sur- logic rejection more probable. In the first half of
vival rate is controversial. Gal and colleagues the twentieth century, a diameter of 8.00–8.25 mm
suggested no effect on donor’s age. In a study by was suggested as optimal for full-thickness grafts.
Barraquer and colleagues, grafts from donors In PKs for bullous keratopathy or Fuchs’ dystro-
aged 80 were at a slightly higher risk of failure phy (with insufficient recipient ECD), an inverse
than patients under 80 years [73–75]. The Corneal association was observed between ECL and tre-
Donor Study (CDS) has reported no or weak phine diameter [50, 75, 82–88].
association between donor age and failure up to In keratoconus, the cone is frequently inferi-
97 years of age, with follow-up ranging from orly displaced and should be completely removed
<1 year to 22 years. A comparison of grafts that to prevent residual or recurrent disease. In
survived for 5 years postoperatively also showed advanced scarred conus, corneal topography is
a higher ECL in the donors aged between 70 and unreliable and should not be used for surgical
76 years [46, 76–79]. This difference in the
results of studies can be related to the different 1
If there are no risk factors, transplant rejection reduces as
methodologies used for the statistical analysis, the recipient becomes older.
Fig. 5.3 PKP in chronic or delayed-onset mustard gas keratitis should be considered as a high-risk graft; however, with
appropriate management, graft clarity and visual outcomes may be favorable
planning. Donor size will be adjusted in relation Eye Banking Practices

to the host limbal white-to-white measurement DALK has shown that poor-quality corneas are
and conus extension, so grafts larger than 8.5 mm not associated with an increased risk of graft fail-
may be required [1]. ure in penetrating keratoplasty, following trans-
In a phakic eye undergoing penetrating kera- plantation. Other conditions of the graft, such as
toplasty, a larger donor button is used to ensure preservation status and time between donor’s
the greatest fit. The donor cornea is predicted to death and transplant (enucleation within 12 h),
be 0.26 mm smaller than the donor corneas cut have not been identified as significant risk fac-
from the endothelial side. Penetrating grafts used tors. By reducing the requirement for a perfectly
to treat keratoconus had a superior visual out- healthy endothelium and a high ECD, DALK
come and graft survival rate2 [89] (Video 5.2). increases the number of donor cornea available
for transplantation—but more follow-up is
• A corneal transplant size of about 7.5 mm is required immediately after DALK [75, 76,
associated with a higher rate of rejection, 93–97].
researchers have found. Because of the graft’s
small size, a portion of the cone may remain, ecipient-Related Factors: Primary
R
causing significant astigmatism and, in some Diagnosis, Number of Surgeries,
cases, graft failure. Neovascularization, and Comorbidities
• In cases of keratoconus, the graft should be
large enough to cover the entire cone in order Primary Diagnosis
to avoid damaging eye orbits. The Asian Cohort (Singapore) and Cornea donor
• However, no clinical investigation has been study (CDS) showed a higher failure rate in pseu-
conducted to prove that a smaller transplant dophakic bullous keratopathy than Fuch’s dystro-
size is related to a lower rejective reaction phy grafts. The presence of trauma and chemical
than a corneal transplant rejection [90]. Some burns at the time of PK also deserves special men-
factors that increase the chances of corneal tion, considering the worst prognosis among all
transplant rejection include closeness to most diagnoses. It has been postulated that the lower
receptor and receptor Langer cells and prox- ECL in these patients may be the main cause of
imity to more blood vessels in the limbus area lower graft survival rates. Based on the results of
[91, 92]. 895 PK from 778 patients, patients with keratoco-
nus (the most frequent diagnosis) showed the best
2
Considering size to prevent rejection. 10-year survival estimate (95%), followed by
endothelial and stromal dystrophies (both 55%), 10-year survival estimate. Corneal vasculariza-
infectious leukomas (corneal ulcers with different tion quadrants are another characteristic associ-
infectious etiologies; 49%), trauma (33%), and ated with graft failure and rejection after
chemical burns (14%); other diagnoses resulted in keratoplasty. Preoperative administration of anti-
a 10-year survival rate of 37%, which showed a angiogenic pharmacologic agents may be able to
4.6-fold higher odds (hazard ratio) of failure in improve survival [75, 105–112].
patients with stromal dystrophies, sixfold for
endothelial dystrophies, 7.4-fold for infectious Infections
leukomas, tenfold for trauma, and 11.9-fold for Corneal infectious disease is a major cause of
the chemical burn; 9.5 for other diagnoses, com- blindness worldwide. Different types of keratitis
pared with keratoconus3 [48, 75, 98]. may influence the long-term prognosis of grafts.
Active microbial infection at the time of PK
Primary or Repeat PK increases the graft failure rate (hazard
Another recipient-related factor impacting the ratio = 5.10). The use of oral antivirals and immu-
outcome is the primary or repeat PK perfor- nosuppression with cyclosporin A or
mance. The prognosis of repeat surgeries has Mycophenolate Mofetil has reduced the recur-
been suggested by several studies. This is proba- rence rate of herpes simplex [54, 113–116].
bly due to deteriorated condition of the corneal In summary, keratitis induced by herpes sim-
bed, increased IOP during the previous kerato- plex has a lower likelihood of successful corneal
plasty, and violation of anterior chamber immu- transplantation if:
nological privilege. In an earlier study by
Thompson and colleagues, similar results were • During surgery, there has been inflammation.
obtained, considering the better survival of pri- • In the cornea, blood vessels have formed.
mary grafts and keratoconus. The patient’s age • Due to decreased tear production and
had less impact on the second or more grafts decreased corneal sensitivity, there is a higher
when compared to first grafts4 [9, 75, 93, 100– risk of persistent epithelial defects (PED) after
104] (Video 5.3). surgery. Suture loosening, corneal invasion,
and graft failure are all symptoms of PED [91,
Corneal Neovascularization 92, 117].
The loss of corneal angiogenic privilege after PK
increases the risk of graft failure. Barraquer and A prophylactic dose of antiviral drug should
colleagues also showed that primary PK grafts be started 1–2 weeks before corneal transplanta-
with avascular recipient corneas had the best tion and continue for at least 6 months after. In
endothelial insufficiency caused by herpes sim-
plex that is treated with DSEAK/DMEK antiviral
3
The benefits of steroid treatment should be weighed drugs can last for up to a year [118].
against the risks of long-term steroid therapy, especially in
those with an underlying condition that causes damage to
the eyes. Glaucoma
4
Systemic sensitization to donor antigens is more likely This effect has not been approved in subgroup
than local alterations caused by the main graft to hasten analysis, separated based on the indication of
the rejective reaction. This rejection is unrelated to MHC PK. More studies are needed to establish the
congruence and could be caused by MHC components
shared by both the first and second donors. The survival of
exact mechanism of this effect. CDS also showed
the second link will not be improved by matching the that cases with a history of glaucoma (before PK
donor-recipient HLA17. Therefore, the chance of rejec- and those who used IOP-lowering medications at
tion increases after the second corneal transplant opera- PK) had a higher graft failure rate than those with
tion [99]. No increases in the chances of transplant
rejection have been observed in the other eye of someone
no history of glaucoma (58% vs. 22%, respec-
whose first eye has had a rejection corneal transplant in tively; hazard ratio = 7.2). Considering the higher
any of the authors’ clinical studies. rate of endothelial decompensation in patients
with glaucoma, the higher failure is related to induced by synechyma on the corneal endothe-
ECL decline. Stewart and colleagues showed a lium, may lead to graft failure5 [54].
3-year transplant survival rate of 86% in eyes
without glaucoma, 72% in eyes with glaucoma Systemic Chronic Diseases
(73% in eyes with medically managed glaucoma Children with Peters’ anomaly and those with
and 63% in surgically managed glaucoma) [98, associated cataracts and glaucoma are at risk for
119–121]. graft failure. Smoking and poor outcomes in the
Antiglaucoma drugs increase the risk of first decade of life may also play a role. African-
corneal transplant rejection for three reasons Americans and nonwhite race have also been
as follows: noted as risk factors but not confirmed as inde-
pendent risk factors [48, 98, 116, 122, 127–130].
• Graft rejection is higher as a result of increased During endothelial transplantation, roughly
intraocular inflammation. 10% of immune reactions and transplant rejec-
• Epithelial cell surface abnormalities cause tion are delicate and dispersed KPs on the endo-
persistent inflammation. thelium’s surface, less as Khodadoust lines, and
• Decomposition of endothelial cells without respond well to topical steroids. The reasons for
immunological responses [122]. this response are:
• There have been reports of pilocarpine drops
when a rejection occurs, which subside after 1. The limbus area’s distance from the endothe-
the medicine is stopped and topical steroids lium layer.
are started [123]. 2. A less amount of stroma is transferred with
• Dorzolamide is a reversible carbonic anhy- the tissue.
drase II inhibitor that does not build up inside 3. In connective tissue, there is no corneal
the cornea with repeated use. It acts as a selec- epithelium.
tive inhibitor of type II and has a minor effect 4. Receptor dendritic cells are unable to access
on type I. All of the patients in the study had the grafted layer. Thereby, selective e ndothelial
complicated ocular histories, which included keratoplasty should be considered in cases
several surgeries and compromised corneas where only the endothelium is affected.
[124].
Surgery-Related Factors
Ocular Surface Diseases
The ocular surface is formed by three-component The exact effect of the surgical details on the out-
tissues: the cornea, conjunctiva, and limbus. come has been only evaluated scantly. Some fac-
Corneal infection in any form of bacterial, fungal tors related to surgery, such as loose sutures and
(Video 5.4), or viral infection can increase the iris synechiae to the graft margin, have been
possibility of corneal graft failure. Any disease demonstrated as risk factors for graft failure. The
that endangers the healthy and moist surface possibility of combined surgeries as a risk factor
accelerates corneal transplant rejection [22]. has also been raised. Still, combined surgery lost
its effect in multivariate analysis [12, 75]. Others
Inflammation have also shown that combined surgeries did not
Anterior synechiae on the iris can impair ocular significantly affect graft survival [95, 131]. This
immune privilege and increase the risk of graft
rejection. Auto-immune diseases include uveitis,
ocular mucous membrane pemphigoid, eye-
5
Immune reactions can occur after any inflammation in
the eyes that have had corneal transplantation, and YAG
involved collagen vascular disorders, Steven- laser capsulotomy is no exception. To decrease these
Johnson syndrome, and atopic immunological reactions, steroid drops should be used
keratoconjunctivitis. Both glaucoma and traction, following capsulotomy [125, 126].
is while Fasolo et al. showed a 2.8-fold greater tism, spherical equivalent, graft rejection, graft
risk of graft failure after PK with pars plana vit- failure, and complications. However, it resulted
rectomy [93]. The low number of combined sur- in higher ECD and better BCVA [40].
geries in study populations is the main limitation Further studies are required to compare the
in the available literature [132]. surgical outcome of modern vs. conventional
Other surgical interventions at the time of sur- PK. The large studies considered a strict inclu-
gery, including cataract extraction, removal, and sion criterion that comprises only a specific group
insertion of IOL, did not alter the odds of rejec- of patients (with moderate risk) and is performed
tion. The study also found that none of the surgi- on specific races. It has to be noted that the fre-
cal procedures combined with PK, including quency of these risk factors varied among stud-
posterior-lens implantation or pars plana vitrec- ies, based on the study population.
tomy, influenced graft survival [130, 131, 133].
Others have also confirmed higher graft failure
rates in children who received PK combined with Penetrating Keratoplasty
other intraocular surgeries than those without. in Keratoconus
Others have also claimed that concurrent vitrec-
tomy results in a higher rate of immunologic An elective PK is now reserved in extreme cases
rejection. As stated in the literature, the surgical when the DM and endothelium appear to be sepa-
procedures, considered as “combined,” are vari- rated due to previous corneal hydrops. In such
able. Cox regression analysis of 37 variables cases, a lamellar approach can still be used, espe-
showed concurrent cataract extraction or IOL cially if the scars do not disrupt the visual axis.
removal without lens implantation and intraocu- The procedure will need to be modified to a PK
lar silicone oil after the PK as significant predic- intraoperatively if a large tear is discovered (lon-
tors of graft failure after PK; two-thirds of these ger than 2–3 clock hours). Although the PK strat-
patients experienced graft failure eventually egy for keratoconus is not dissimilar to that used
[116, 134–136]. for other etiologies, there are a few things to keep
The main focus of this section was on the risk in mind:
factors of graft rejection/failure after PK; how-
ever, the rates reported among different studies Decentered Grafts
are different for several reasons. First, various Decentered grafts can also create significant
predictive treatments have been suggested for irregular astigmatism in the visual axis, requir-
reducing the graft rejection rate after PK, espe- ing the patient to wear stiff glasses for visual
cially in high-risk patients, such as immunosup- rehabilitation and, in certain situations, a
pressive agents [15], and variation in their use for second-centered graft. Keratoconus patients
the study population of different studies can be a may benefit from employing same-diameter tre-
source of different rates reported. Another issue phines for both donor and host tissues, which
that must be considered is that the surgical details shrinks the donor button and lowers postopera-
of PK are continually evolving, and several mod- tive myopia. Reducing donor size when the
ifications have been introduced since its introduc- anterior lens-to-retina length is less than
tion (1950) [137]. Therefore, the difference in the 20.19 mm could result in significant postopera-
surgical techniques used during each procedure tive hyperopia [138–142].
can be an important source of diverse results of
studies. A review of seven comparative studies Suturing Technique
showed that FSL PK was not superior to conven- The surgeon can use one of the following suture
tional PK in postoperative topographic astigma- techniques after placing the four cardinal 10-0
nylon sutures: single continuous suture (SCS), Management of Postpenetrating

double continuous suture (DCS), combination Keratoplasty Astigmatism
of continuous and interrupted suture (CCIS), or
interrupted suture (IS). IS should always be the When astigmatism is too severe to be treated with
closure method in cases where partial or com- the excimer laser alone, it can be reduced to a
plete suture removal is required. These include level where PRK or LASIK can be performed
pediatric keratoplasty (sutures becoming loose with or without relaxing incisions. Similarly,
too quickly), vascularization in the host cornea, emmetropia-like refractive outcomes could be
multiple previous rejections, or inflammatory achieved by combining relaxing incisions with
concomitant conditions [143]. IOL implantation or IOL implantation with an
Furthermore, research comparing astigma- excimer laser [147].
tism in keratoconus patients treated with a sin-
gle continuous suture against a DCS found that
astigmatism was identical in both groups after Postoperative Care
removing the sutures (DCS 4.6 D, SCS 5.2 D)
[144]. As a result, it is apparent that any suture I ntraocular Pressure Measurement
closure technique can work. The final choice is after Penetrating Keratoplasty
made by the surgeon. The needle is placed 90%
deep into the donor cornea and then through The Goldmann applanation tonometry (GAT)
the host cornea to provide a rudimentary idea remains the gold standard of IOP assessment,
of normal graft suturing. The ideal bite is as despite changes in corneal anatomy that occur
close to DM as possible, and a similar amount after PK. The iCare, Tono-Pen XL, ocular
of tissue should be acquired in both the donor response analyzer, and PDCT contour tonometry
and host corneas to approximate Bowman’s are gaining attention. All save the iCare tend to
layer. overstate IOP when compared to GAT. Future
They hypothesized that the time between developments may improve the method’s depend-
transplanting and recurrence with DALK would ability even more. Although Goldmann tonome-
be shorter than with PK, albeit this has yet to be try is still the most appropriate modality for IOP
confirmed or supported by other investigations measurement in post-PK eyes, its limitations in
[145]. More research on long-term results is these situations underscore the need for new
needed to assess the influence of DALK on kera- approaches that are more accessible and conve-
toconus [1]. nient to employ other tonometers have yielded
Penetrating keratoplasty is a safe and success- inconsistent results, with some promising and
ful treatment for patients with keratoconus who others discouraging [148].
are contact lens intolerant or have poor corrected
visual acuity. The severity of the problem and the
trephination and suturing techniques performed anagement of Patients after
M
does not affect the ultimate visual results [30]. Surgery to Reduce the Risk of Corneal
DALK is the first surgical option for patients Transplant Rejection
with keratoconus and possibly other corneal
stromal pathologies with normal endothelium. After a high-risk corneal transplant, the main
Because of its benefits, such as better globe goals are prevention, early diagnosis, and
integrity preservation, reduced intraoperative appropriate treatment. The patient should be edu-
complications, and elimination of endothelial cated on the symptoms of transplantation and
graft rejection, DALK can be considered an suture loosening in order to make an early diag-
alternative to PK for this condition [146] nosis. Topical steroids and antibiotics should be
(Fig. 5.4). prescribed to prevent transplant rejection and
infection after suturing. Regular follow-up can
86
Fig. 5.4 Comprehensive schematic illustration of key penetrating keratoplasty considerations

F. Doroodgar et al.
Table 5.1 Comparison of different topical corticosteroids in terms of available types, potency, penetration, and side
effects
Increasing Penetrating Anti-inflammatory Available
intraocular pressure eyes effect Dosages types Topical corticosteroid
Low Low Medium 0.10% Drops Fluorometholone (acetate)
0.10% Ointment
Low Low High 0.2–0.5% Drops Loteprednol (etabonate)
High Medium High 0.10% Drops Dexamethasone (sodium
Complex Ointment phosphate)
High High High 0.125–1% Drops Prednisolone (acetate,
phosphate)
Very high High Very high 0.05% Drops Difluprednate
High High Very high 0.10% Drops Betamethasone (acetate and
0.10% Ointment sodium phosphate)
Low Low High 0.5–1% Drops Hydrocortisone
1–3% Ointment
Low Low Medium 1% Drops Medrysone
Low Low High 1% Drops Rimexolone
also aid in the early detection of cornea transplant for preventing corneal transplant rejection
rejection. according to a 2011 survey [53]) reaches a high
level in the aqueous humor in 120 min and per-
Corticosteroids sists for a day [151].
Corticosteroids are still the most commonly pre- Topical corticosteroid drops (betamethasone
scribed drugs for corneal transplant prevention 0.1% or prednisolone acetate 1%) in two doses,
and treatment [53, 54, 72]. In cases of high-risk four times a day for 3 months, are used to prevent
corneal transplant rejection, specific immunosup- transplant rejection in penetrating keratoplasty
pressive drugs are used. transplants with no risk factors and a low risk of
The most commonly prescribed drugs to pre- rejection (ACAID time), and at a daily dose of
vent and treat rejection in all types of corneal 4–6 months and gradually reducing the dose over
transplants are topical corticosteroids (betameth- 6–12 months, they are used in penetrating kerato-
asone and prednisolone acetate 1%). How these plasty transplants with a high risk of corneal
drugs are administered in postcorneal transplant transplant rejection. Patients with herpes simplex
regimens varies greatly [53]. Table 5.1 Introduce keratopathy complications should be treated with
widely used steroids in ophthalmology [149, an oral prophylactic dose of acyclovir or valaci-
150]. clovir for a long time. Although anterior layer
grafting eliminate graft rejection, graft rejection
How Important Topical Steroids Work in the stromal layer and epithelium remains a pos-
and What Are They Used For sibility. Prednisolone acetate drops 6–4 times a
The ideal steroid will pass through the stromal day are recommended for these patients in the
fluid layer, depending on how much-unchanged first days after surgery, with the dose gradually
drug enters the systemic circulation. The penetra- decreasing over 6–12 months because the major-
tion of acetate derivatives is lower in the case of ity of transplant rejection after endothelium trans-
an epithelial defect, but regardless of that, acetate plantation has been observed when topical
and alcohol derivatives are more effective than corticosteroid drops are completely stopped6
phosphate derivatives in inhibiting corneal [152, 153].
inflammation [149].
Prednisolone acetate 1% (the preferred and 6
Topical steroid after Low-Risk surgery should be used
most commonly prescribed topical corticosteroid for at least 2 months until ACAID is created, with eye
If the pseudo facial and intraocular pressures activated and cause ACAID, but the probability
are normal, topical corticosteroid drops can be of graft rejection is very high until the ACAID
safely continued for longer, but the dose must be phenomenon occurs in 6–8 weeks after surgery
limited or less potent drugs such as loteprednol [155].
etabonate or fluorometholone be used to prevent The first corneal transplant rejection can occur
cataracts in patients with clear crystalline lenses. after 3–4 weeks and a week in a low-risk or high-
Shimazaki discovered that long-term use of risk operation, respectively. Start a topical steroid
fluorometholone 0.1% was effective in prevent- and, depending on the response, taper and dis-
ing corneal transplant rejection in low-risk pene- continue within 3–4 weeks. If KP occurs during
trating keratoplasty (PKP) patients with no follow-up, we will continue with the rejection
significant side effects, but considering the con- treatment as planned. If the patient’s corneal
dition of the lens in terms of the onset of posterior transplantation was due to keratoconus or corneal
subcapsular cataracts (PSC), eye pressure control dystrophies, start topical steroids immediately,
is required at each visit [151]. It is recommended and if the patient’s corneal transplantation was
that the daily dose of topical corticosteroids be due to herpes, start an oral antiviral drug or pro-
continued for a long time until the surgeon sees phylactic dose.
significant side effects. For 1 year after transplan-
tation, some surgeons use Fluorometholone Long-Term Topical Steroid Use
drops or Prednisolone lotion drops on a daily In high-risk cases, such as a history of glaucoma
basis. If the transplant is not in the high-risk [157] or vitreous and retinal surgery, multiple
group, the immune privilege (IP) mode will be cases of rejection attacks, as well as pseudopha-
restored after some time, making the patient no kic patients who have one eye [158], recommend
different from other patients [154, 155]. taking topical steroids for a long time.
Difluprednate 0.05% is not currently avail- There is no general rule for all patients, and
able in Iran and many other countries. the risk of developing Infectious Crystalline
Injecting 20–40 mg of Triamcinolone sub- Keratopathy is the worst part. In a patient with a
conjunctival capsule during surgery is helpful in history of corneal transplantation, during a rou-
patients who are unlikely to cooperate postopera- tine examination, start topical steroids every
tively or have a high risk of corneal transplant 2–4 h and reduce them within 2–3 months.
rejection. Some surgeons perform a conjunctival
injection of 40 mg methylprednisolone acetate Immunosuppressive Therapies
1 mL or 0.1% dexamethasone disodium phos- Immunosuppressive therapies, either topical or
phate. In terms of improved transplant rejection systemic, are frequently required to prevent graft
and relapse, there was no statistically significant rejection. According to Holland et al.’s research,
difference between the two groups at the end of the use of corticosteroid-sparing diets is recom-
the study [156]. mended in high-risk transplants to prevent cor-
Corticosteroid injections into the conjunctiva neal transplant rejection due to the many side
have not been shown to prolong epithelial defect effects of oral corticosteroids [159]. Calcineurin
repair. There is no evidence that using oral corti- Inhibitors, Anti-proliferative drugs, and blocking
costeroids to prevent corneal transplant rejection the activation and action of T cells are all included
is beneficial. The probability of rejective reac- in these diets. Calcineurin inhibitors include
tions is very low when dynamic mechanisms are Cyclosporine A and Tacrolimus. In low-risk
transplants, systemic immunosuppressive ther-
pressure and lens position monitored regularly. In patients apy usually has no place [160]. However, in high-
with a low risk of rejection, steroid treatment should be risk patients, the use of these drugs is
started at 2- to 3-h intervals and continued until the KP has
entirely disappeared. Even daily use of a steroid drop can
recommended (Fig. 5.5).
cause eye strain, cataracts in phakic patients, and corneal Cyclosporine is a cyclic undecapeptide, while
infectious crystalline keratopathy (ICK). tacrolimus is a macrocyclic lactone; However,
5
Pharmacological views
Oral Immunization with

donor-specific Allo-antigen
Induction of the Allo-specific Corticosteroids

Lymphadenectomy Immunosuppressive therapies See Table 1
Tolerance
Gene Therapy
Mycophenolate Mofetil
(MMF)
Cyclosporine A
Blocking the Activation &
Immunosuppressive drugs Anti-proliferative drugs Rapamycin
Action of T Cells
Tacrolimus
Main Issues to Overcome in Modern Penetrating Keratoplasty
Azathioprine
Monoclonal antibodies against T Regulating the immune reaction Inhibiting immune access to the
lymphocyte antigens Blocking Co-stimulatory Signals using Cytokines Inhibition of Antigen-presenting cells
grafts
Fig. 5.5 Management of patients after surgery to reduce the risk of corneal transplant rejection
89
both inhibit calcineurin and are classified as fusion protein to B7 molecules. The CD28 recep-
immunosuppressive drugs [161]. Tacrolimus tor T lymphocyte is responsible for CLTA4-Ig’s.
blocks calcineurin to limit T lymphocyte IL-2 Both topical CLTA2g drops and monoclonal anti-
production. Cyclosporine A inhibits T lympho- bodies against CD28 have improved connective
cyte migration, and vascular development tissue survival in various studies [169, 172–174].
increases lymphocyte infiltration in the conjunc- Belatacept (more fusion to B7) has not yet been
tiva and lacrimal gland and tears production. used to prevent corneal transplantation rejection
Despite differences in potency, tacrolimus, and [169, 173]. Also, a study showed that Tocilizumab
cyclosporine both have high graft survival rates may promote corneal allograft survival, possibly
[162, 163]. However, cyclosporine A has low by modulating the Treg-Th17 balance in rat [175].
efficiency in avoiding transplant rejection and is Regulating the immune reaction using cyto-
not frequently used nowadays [53, 54, 150, 159, kines also could be a beneficial approach to man-
164, 165]. agement transplantation. Transfection of
The use of anti-proliferative drugs (such as interleukin-4 (IL-4) and interleukin-10 (IL-10)
mycophenolate mofetil, rapamycin, and azathio- genes into transplanted corneal epithelium via
prine) along with the mentioned immunosuppres- viral carriers increases its survival in animal stud-
sive drugs can be more effective in preventing ies [169, 170]. In rat eyes, topical administration
transplant rejection. Mycophenolate mofetil (MMF- of α-melanocytic stimulating hormone [176],
known as CellCept) inhibits T and B lymphocyte α-Tumor necrosis factor, or α-TNF, in a mouse
proliferation. It had fewer side effects and a more study, less transplant rejection [177].
significant effect in preventing graft rejection than Despite the effect of inhibition of antigen-
cyclosporin A [53, 54, 150, 159, 164]. Rapamycin presenting cells to reduce inflammation, it did not
inhibits B and T cell activity and lowers Ig produc- affect reducing the rate of corneal transplant
tion, with an effect that is 80–100 times stronger rejection [54, 169, 178].
than cyclosporine A. It is a good substitute for Inhibiting immune access to the grafts has sig-
mycophenolate mofetil [53, 54, 150, 159, 164, 166]. nificant effects in preventing graft rejection via
In complicated patients, azathioprine is currently obstructing the movement of immune cells and
used as a supplement to oral cyclosporine A or oral corneal vascularization attenuation. Among the
tacrolimus [53, 54, 150, 159, 164]. methods of obstructing the movement of immune
Some experimental approaches to blocking cells, we can mention: Leukocyte function anti-
the activation and action of T cells are monoclo- gen 1 or LFA-on T lymphocytes attach to intra-
nal antibodies against T lymphocyte antigens, cellular adhesion molecule 1 (ICAM-1). As a
blocking co-stimulatory signals, regulating the result, cytokine production increases [54, 167,
immune response with cytokines and peptides, 169, 178]. Antigen function of white blood cells
inhibiting antigen-presenting cells, and inhibiting and very late antigens 1 and 4 are among these
immune access to grafts [53, 54, 167–170]. antibodies (VLA-1 and VLA-4). Liftegrast 5%
Monoclonal antibodies against CD3, CD4, (Xiidra) is one of the topical treatments for dry
CD8, IL-12, and αβ T-cell receptors have benefi- eye that has been shown to help transplanted cor-
cial effects after parenteral administration in neas survive longer. Its structure is similar to that
experimental animal models [171]. Xanax (dacli- of ICAM-1 [178–180].
zumab) and Simulect (basiliximab) bind to the Several genes are involved in increasing trans-
alpha subunit (Tac/CD25) of interleukin 2 to pre- plant survival. Induction of the allospecific
vent T lymphocyte activation [54, 168, 170]. Tolerance could be effective in combination with
Alemtuzumab (CD52 receptor antibody), in a anti-inflammatory and other immune-regulating
small number of studies, has been shown to properties to prevent transplant rejection. For
improve transplant survival [169]. example, in a study, oral immunization with
Abatacept (cytotoxic T-lymphocyte-associated donor-specific alloantigen tolerance was induced
protein 4) is a high affinity-bonded recombinant in animals [181]. Also, lymphadenectomy,
another offered treatment, prevents antigen- immune cells in different layers of the cornea
presenting by blocking the function of VEGF-3 was studied in a study of 38 patients who under-
[182, 183]. The transfer of the interleukin-10 went corneal transplantation (15 with rejection
gene in animal studies and programmed cell and 23 without rejection) using in vivo confocal
death ligand (PDL-1) gene transfer (decreasing microscopy (IVCM). The authors concluded that
CD8 lymphocytes, natural killer proinflammatory the density of these cells increases in patients
cytokines) [54, 184–186] showed that gene ther- with corneal transplant rejection, particularly in
apy could be a different and impressive treatment the subbasal and endothelial layers, and that this
method in the future. increase is linked to pain and light sensitivity in
these patients [56]. In a separate study, 45 patients
who had corneal transplants were followed up
he Stages and Approach
T prospectively, with AK cells counted with IVCM
to the Corneal Transplant Rejection on the first and seventh days after surgery, as well
as monthly. In this study, patients who had rejec-
1. When a corneal transplant that is already clear tive transplant reactions 2 months ago had an
and free of problems develops edema without increase in AK cells. The number of cells in the
any inflammatory or clinical signs, it is con- other group remained constant for 4 months after
sidered “possible.” transplantation [187].
2. When corneal edema is accompanied by Rapid diagnosis and initiation of treatment
inflammation, but there is no endothelial with high and frequent doses of the drug accord-
rejection line, it is considered “probable.” ing to the stage, KPs, and clinical findings is the
3. When corneal edema is accompanied by an second most effective strategy in the manage-
endothelial rejection line, this is referred to as ment of corneal transplant rejection.
“definite rejection.”
Corneal transplant rejection usually starts with Better Management

photophobia, then mild to moderate AC reaction, of Transplantation Considering KPs,
then the emergence of subtle pigmented KPs, then Vascularization, and Iris Adhesions
the pigmentation of KPs, and finally the forma-
tion of Khodadoust line and corneal edema. A subconjunctival injection of betamethasone
Basically, hyperacute rejection does not occur 2–4 mg is helpful if the rejection is more severe. Oral
during corneal transplant rejection. The term steroids and intravenous injections may be added to
“hyperacute rejection” has been used in some the regimen in addition to the above. Steroid drops
articles to rule out corneal transplantation, but are started at 1-h intervals, and betamethasone oint-
this is a phenomenon unique to organ transplan- ment can be used at night while sleeping.
tation. The rejection of the graft in the stromal KP is not pigmented at first and disappears
layer is mentioned immediately after the endo- with proper treatment, but due to the severity of
thelial layer is expelled in the articles that have the reaction, large, and pigmented KP can occur,
used this term, which starts with the sign of cor- which may disappear or remain for a long time
neal opacity and involves the center of the cornea (old KPs) after treatment except for inflamma-
for a short time [9]. tion. No special treatment is required in such
Up to one-third of transplant rejection cases cases where the number, size, and color of KPs
result in transplant failure. Two to 3 months have not changed over time.
before the onset of clinical symptoms, a confocal When severe corneal edema prevents KP
microscopic examination of the stroma reveals from being seen, the edema can be treated and
the presence of activated keratocytes (AK). It is a reduced to allow KP to be seen. KP may be
way to diagnose corneal transplant rejection gone, but edema persists in the final stages of
before the clinical signs begin. The density of complete transplant rejection, which cannot be
improved. If there are predisposing factors for should be observed more than in other people
rejective reactions, they should be removed at [55].
the same time as starting treatment because stro-
mal edema may prevent KPs from being seen
after KP edema has resolved. The presence of harmaceuticals and Clinical Aspects
P
active blood v essels, especially the cornea, is of Different Drugs in Keratoplasty
one of the active factors in the rejection of cor-
neal transplantation, which should be elimi- The clinical factors that should be taken into
nated by subconjunctival injection of anti-VEGF account when using various corticosteroids as the
or fine needle courtesy. cornerstone of treatment, immunosuppressants,
As a supplement, it has been suggested that and antibodies are summarized in Tables 5.2, 5.3,
iris adhesions to the cornea be isolated. It is and 5.4, respectively [156, 169, 177, 188–196].
emphasized that treatment for corneal transplant Figure 5.6 also includes a summary of safeguards
rejections should be continued until the KPs have that should be taken before undergoing corneal
vanished completely. During this time, the eye transplantation.
pressure should be recorded at each visit, and the In conclusion, the most effective strategy for
clarity or onset of opacity in keratoconus patients treating corneal transplantation is prevention and
Table 5.2 Clinical points in using corticosteroid drugs

Drugs Using
Betamethasone 0.1%, As previously stated, the most commonly used drug for corneal transplantation is
Prednisolone Acetate prednisolone acetate drops 1%.
1%, or Dexamethasone Corticosteroids are usually prescribed at a dose of one drop every 3 h at the start of
0.1% treatment, with the dose gradually decreasing depending on the severity of the transplant
rejection and the clinical response to treatment. Corticosteroid drops should be used until
the KPs (keratic precipitate) deposits on the cornea are completely gone.
In a patient with corneal transplant rejection, we usually stop treatment 2–3 weeks after
no KPs or cells are seen in the anterior chamber, and the KPs resolve after starting the
steroid after 4–5 weeks, but a re-examination is needed a few weeks later to ensure there
is no recurrence of KPs.
Patients with a history of recurrent corneal transplant rejection should continue to use
low-dose corticosteroid drops on a long-term basis (once daily or every other day).
Betamethasone eye Betamethasone eye ointment can also be used once a night before bed as a
ointment complementary treatment in patients with a history of recurrence. However, keep in mind
that the ointment’s bioavailability is lower than that of frequent drops.
Patients should be followed up on a regular basis. In patients with a history of high blood
pressure or herpes simplex keratopathy, corticosteroids should be used with caution.
Injection of 20 mg of In case of severe rejection or patients who cannot use the drug frequently, an injection of
Triamcinolone acetonide 20 mg of triamcinolone acetonide or 2 mg of dexamethasone can be used.
or 2 mg of Increased intraocular pressure is the most serious side effect of triamcinolone acetonide
Dexamethasone injection or dexamethasone submucosal injection, and patients should be closely
monitored for it.
Oral corticosteroids There is no agreement on the use of oral corticosteroids (prednisolone) at a daily dose of
(Prednisolone) 1 mg/kg body weight in the treatment of corneal transplant rejection, and a small clinical
study has been done.
(continued)
Table 5.2 (continued)

Drugs Using
Intravenous Several studies have been performed on the use of intravenous corticosteroids (single-
corticosteroids dose methyl prednisolone, 125–500 mg) in the acute phase of corneal transplant rejection.
(single-dose Patients with acute endothelial layer transplant rejection were randomly divided into two
methylprednisolone, groups in one clinical study: Topical corticosteroids were combined with a single pulse of
125–500 mg) methylprednisolone (500 mg) in one group, while topical corticosteroids were used alone
in the other. In both groups, topical corticosteroid treatment consisted of a single
subconjunctival injection of 2 mg of betamethasone followed by treatment with 0.01%
dexamethasone drops for the first 24 h. In terms of transplant rejection, relapse,
recurrence, and posttransplant rejection failure, there was no statistically significant
difference between the two groups at the end of the study.
Oral Prednisolone The use of oral prednisolone in the treatment of corneal transplant rejection seems to be
useful only in cases where 6 months or less have elapsed since corneal transplant surgery,
and on examination, in addition to the rejection symptoms, there is corneal edema.
Subconjunctival Another study also showed that endothelial transplant rejection after penetrating
injection of 20 mg keratoplasty (PKP) with subconjunctival injection of 20 mg triamcinolone acetonide with
Triamcinolone 1% prednisolone acetate compared with the intravenous pulse of 500 mg
methylprednisolone with prednisolone acetate 1% was preferred.
Intracameral injection of In several studies, the effect of intracameral injection of triamcinolone and
Triamcinolone and dexamethasone in endothelial layer transplantation and penetrating keratoplasty (PKP)
Dexamethasone was investigated.
Because of the risk of infection and cataracts from injecting into the anterior chamber,
this method is not recommended.
Table 5.3 Clinical points in using immunosuppressive drugs

Drugs Using
Topical and systemic tacrolimus Some studies have recommended the use of immunoregulator drugs such as
topical and systemic tacrolimus as well as CellCept.
The patient’s condition, whether he or she has only one eye or the other eye is
healthy, and whether or not there is a systemic disease such as diabetes or
kidney failure, will determine how to begin and how long to use them. The
anterior segment specialist is consequently compelled to choose. Topical
steroids and tacrolimus drops are used for a long time, and in some cases
forever, after the inflammation has been controlled.
Cyclosporine 0.05% drops Other clinical studies have shown that cyclosporine 0.05% drops have no effect
in the treatment of acute corneal transplant rejection.
Tacrolimus drops 0.05% In another study, the therapeutic effect of tacrolimus drops 0.05% on
endothelial layer transplantation after PKP was evaluated, and patients were
randomly divided into two groups:
Both groups received corticosteroid therapy, with tacrolimus drops 0.05%
added as an adjunct drug in one group. A single subconjunctival injection of
1 mg betamethasone, 1% prednisolone acetate drops, and 1 mg oral
prednisolone tablets per weight (1 mg/kg) was given to both groups as
corticosteroid treatment. Treatment with 0.05% tacrolimus drops was found to
be ineffective in treating the acute phase of corneal endothelial rejection at the
conclusion of the study, though during recurrences, it might reduce rejection.
Table 5.4 Clinical points in using antibodies

Drugs Using
CD3 receptor antibody The first monoclonal antibody approved by the US Food and Drug
(muromonab or OKT3) Administration (FDA) to prevent kidney transplant rejection is the CD3 receptor
antibody (muromonab or OKT3).
CD3 is one of the receptors on T lymphocytes, and its systemic consumption
causes a significant reduction in the number of T lymphocytes and immune
system suppression. Diets containing this monoclonal antibody have been used
to prevent and treat acute renal, liver, and heart transplants. Acute corneal
transplant rejection has also been treated with this antibody. The use of this
antibody systemically has a number of negative consequences, including an
increased risk of infection and cytokine release syndrome.
Muromonab In one case, muromonab was injected into the anterior chamber to treat acute
corneal transplant rejection, which resulted in a severe inflammatory intraocular
response.
α-Tumor necrosis factor or α-TNF α-Tumor necrosis factor or α-TNF is associated with apoptosis, and therefore its
inhibition can greatly inhibit immune function.
In a study in mice, after corneal allograft transplantation for one of the groups,
α-TNF factor receptor drops were started and decreased corneal transplant
rejection in that group.
Antibodies have no place in the treatment of corneal transplant rejection.
quickly diagnose to start treatment with high, fre- tions, rapamycin, and azithromycin may be
quent doses of corticosteroids. used as substitutes.
–– Combination of oral tacrolimus or oral
Take Home Notes cyclosporine and oral Mycophenolate
• In high-risk cases, the main goals of corneal Mofetil.
follow-up are transplantation rejection pre- –– With more research being done on mono-
vention, rapid diagnosis, and appropriate clonal antibodies, these drugs will play a
treatment. bigger role in transplant rejection preven-
• The following steps are currently recom- tion regimens in the future.
mended as a treatment regimen for preventing • The most effective treatment for acute corneal
transplant rejection: transplant rejection is still corticosteroid drops.
–– Use high-dose topical corticosteroids that Because of its high penetration of healthy cor-
are gradually reduced and kept at a low neal epithelium and strong immunosuppressive
dose for a long time. effect, 1% prednisolone acetate is the most
–– Topical tacrolimus or topical cyclosporine widely used corticosteroid drop in the treat-
2% in combination with topical corticoste- ment of acute corneal transplantation, but in
roids for a long time. Iran, betamethasone 0.1% drops are the drug of
–– Preoperative oral corticosteroids and sys- choice.
temic mycophenolate mofetil prescrip-
5
Stevens-Johnson Diagnosis of Early active treatment

Herpes
syndrome/Toxic epidermal the causative using antiviral agents
Corneal Trauma simiplex
necrosis, Chemical burns, disease and steroids
neovascular- keratitis
Chronic inflammtion,’
Micro-scarring ization
Herpes Endothellal
simplex Active treatment with
cell density Stevens-Johnson
Limbal stem keratitis immunosuppressants,
Repeated corneal syndrome and
cell Anti-vascular
transplantation, Pre-operative Preventive chemical burns
deficiency endothelial growth
Chronic inflammation,
consideration cares factors, Various
Limbal stem cell
angiogenesis inhibitors
deficiency
Poor visual
Repeat acuity Corneal
penetrating neovascularization
keratoplasty and corneal opacity
Emergency
penetrating Penetrating Postoperative
keratoplasty glaucoma
Keratoplasty medication
Glaucoma
Retinal
surgery
Main Issues to Overcome in Modern Penetrating Keratoplasty
Post-operative Graft
Complications Failure
Repeat
Rejection
penetrating
keratoplasty
Ocular surface Suture problems
Persistent epithelial
defect, Infectious
Late Emergency keratitis, Graft
Epithelial Endothelial rejection
endothelial penetrating
problems cell density
failure keratoplasty
Fig. 5.6 An overview of preventative, preoperative, and postoperative measures

95
References 17. Sellami D, et al. Epidemiology and risk fac-

tors for corneal graft rejection. Transplant Proc.
2007;39:2609–11. (Elsevier).
1. Arnalich-Montiel F, Alió del Barrio JL, Alió
18. Inoue K, Amano S, Oshika T, Tsuru T. Risk fac-
JL. Corneal surgery in keratoconus: which type,
tors for corneal graft failure and rejection in pen-
which technique, which outcomes? Eye Vis.
etrating keratoplasty. Acta Ophthalmol Scand.
2016;3:2.
2001;79:251–5.
2. Frost NA, Wu J, Lai TF, Coster DJ. A review
19. Hori J, Isobe M, Yamagami S, Tsuru T. Acceptance
of randomized controlled trials of penetrat-
of second corneal allograft by combination of anti-
ing keratoplasty techniques. Ophthalmology.
VLA-4 and anti-LFA-1 monoclonal antibodies in
2006;113:942–9.
mice. Transplant Proc. 1998;1:200–1.
3. Trufanov SV, Budnikova EA, Rozinova VN
20. Baratta RO, Schlumpf E, Buono BJD, DeLorey S,
[Modern modifications of penetrating keratoplasty
Calkins DJ. Corneal collagen as a potential thera-
with complex operative incision]. Vestn oftalmol.
peutic target in dry eye disease. Surv Ophthalmol.
2019;135:260–6.
2022;67:60–7.
4. Adeyoju J, Konstantopoulos A, Mehta JS, Hossain
21. Alio JL, et al. Corneal graft failure: an update. Br J
P. Femtolaser-assisted keratoplasty: surgi-
Ophthalmol. 2021;105:1049.
cal outcomes and benefits. J EuCornea. 2020;8:
22. Siregar S. How ocular surface disorder affected cor-
1–13.
neal graft survival. In: Dry eye syndrome-modern
5. Tóth G, et al. Comparison of excimer laser versus
diagnostic techniques and advanced treatments.
femtosecond laser assisted trephination in penetrat-
Rijeka: IntechOpen; 2021.
ing keratoplasty: a retrospective study. Adv Ther.
23. Beining MW, et al. In-office thermal systems for
2019;36:3471–82.
the treatment of dry eye disease. Surv Ophthalmol.
6. Castellanos-González JA, Orozco-Vega R, González
2022;67:1405–18.
Ojeda A, Martínez Ruiz AM, Fuentes-Orozco
24. Alió del Barrio JL, et al. Corneal transplantation
C. Evaluation of the quality of life related to vision
after failed grafts: options and outcomes. Surv
after penetrating keratoplasty. Archiv Soc Espan
Ophthalmol. 2021;66:20–40.
Oftalmol. 2021;96:69–73.
25. Ji YW, et al. Corneal lymph angiogenesis facilitates
7. Selver OB, Karaca I, Palamar M, Egrilmez S, Yagci
ocular surface inflammation and cell trafficking in
A. Graft failure and repeat penetrating keratoplasty.
dry eye disease. Ocul Surf. 2018;16:306–13.
Exp Clin Transplant. 2021;19:72–6.
26. Perez VL, Pflugfelder SC, Zhang S, Shojaei A, Haque
8. Chotikavanich S, Prabhasawat P, Satjapakasit
R. Lifitegrast, a novel integrin antagonist for treat-
O. Ten-year survival of optical penetrating kera-
ment of dry eye disease. Ocul Surf. 2016;14:207–15.
toplasty and risk factors for graft failure in Thai
27. Zhang X, Vadoothker S, Munir WM, Saeedi
patients. J Med Assoc Thail. 2020;103:883–90.
O. Ocular surface disease and glaucoma medi-
9. Panda A, Vanathi M, Kumar A, Dash Y, Priya
cations: a clinical approach. Eye Contact Lens.
S. Corneal graft rejection. Surv Ophthalmol.
2019;45:11–8.
2007;52:375–96.
28. Song Y, Zhang J, Pan Z. Systematic review and meta-
10. Qazi Y, Hamrah P. Corneal allograft rejection:
analysis of clinical outcomes of penetrating kerato-
immunopathogenesis to therapeutics. J Clin Cell
plasty versus deep anterior lamellar keratoplasty for
Immunol. 2013;2013:006.
keratoconus. Exp Clin Transplant. 2019;18:417–28.
11. Guilbert E, et al. Long-term rejection incidence and
29. Fukuoka S, et al. Extended long-term results of
reversibility after penetrating and lamellar kerato-
penetrating keratoplasty for keratoconus. Cornea.
plasty. Am J Ophthalmol. 2013;155:560–569.e562.
2010;29:528–30.
12. Anderson E, Chang V. Corneal allograft rejection
30. Javadi MA, et al. Outcomes of penetrating kerato-
and failure. Am Acad Ophthalmol. 2015;
plasty in keratoconus. Cornea. 2005;24:941–6.
13. Benetz BA, et al. Endothelial morphometric mea-
31. Pramanik S, Musch DC, Sutphin JE, Farjo
sures to predict endothelial graft failure after
AA. Extended long-term outcomes of penetrat-
penetrating keratoplasty. JAMA Ophthalmol.
ing keratoplasty for keratoconus. Ophthalmology.
2013;131:601–8.
2006;113:1633–8.
14. Lass JH, et al. Endothelial cell density to predict
32. Busin M, Madi S, Scorcia V, Santorum P, Nahum
endothelial graft failure after penetrating kerato-
Y. A two-piece microkeratome-assisted mush-
plasty. Arch Ophthalmol. 2010;128:63–9.
room keratoplasty improves the outcomes and
15. Di Zazzo A, Kheirkhah A, Abud TB, Goyal S, Dana
survival of grafts performed in eyes with diseased
R. Management of high-risk corneal transplantation.
stroma and healthy endothelium (An American
Surv Ophthalmol. 2017;62:816–27.
Ophthalmological Society Thesis). Trans Am
16. Gómez‐Benlloch A, et al. Causes of corneal trans-
Ophthalmol Soc. 2015;113:T1.
plant failure: a multicentric study. Acta Ophthalmol.
33. Garrido C, Cardona G, Güell JL, Pujol J. Visual
2021;99:e922–8.
outcome of penetrating keratoplasty, deep anterior
lamellar keratoplasty and Descemet membrane 50. Chung S-H, Kim HK, Kim MS. Corneal endothe-
endothelial keratoplasty. J Opt. 2018;11:174–81. lial cell loss after penetrating keratoplasty in relation
34. Lam F, Rahman M, Ramaesh K. Traumatic wound to preoperative recipient endothelial cell density.
dehiscence after penetrating keratoplasty—a cause Ophthalmologica. 2010;224:194–8.
for concern. Eye. 2007;21:1146–50. 51. Arundhati A, et al. Comparative study of long-
35. Tzelikis PF, et al. Traumatic wound dehiscence term graft survival between penetrating kerato-
after corneal keratoplasty. Arq Bras Oftalmol. plasty and deep anterior lamellar keratoplasty. Am J
2015;78:310–2. Ophthalmol. 2021;224:207–16.
36. Latz C, Asshauer T, Rathjen C, Mirshahi 52. Borderie VM, Guilbert E, Touzeau O, Laroche
A. Femtosecond-laser assisted surgery of the eye: L. Graft rejection and graft failure after anterior
overview and impact of the low-energy concept. lamellar versus penetrating keratoplasty. Am J
Micromachines. 2021;12:122. Ophthalmol. 2011;151:1024–1029.e1021.
37. Seitz B, Hager T, Langenbucher A, Naumann 53. Azevedo Magalhaes O, Shalaby Bardan A, Zarei-
GO. Reconsidering sequential double running suture Ghanavati M, Liu C. Literature review and sug-
removal after penetrating keratoplasty: a prospec- gested protocol for prevention and treatment of
tive randomized study comparing excimer laser and corneal graft rejection. Eye. 2020;34:442–50.
motor trephination. Cornea. 2018;37:301–6. 54. Jabbehdari S, et al. Update on the management of
38. Seitz B, et al. Penetrating keratoplasty for kerato- high-risk penetrating keratoplasty. Curr Ophthalmol
conus – excimer versus femtosecond laser trephina- Rep. 2017;5:38–48.
tion. Open Ophthalmol J. 2017;11:225–40. 55. Javadi M. Corneal graft rejection: mechanism, clini-
39. Boden KT, et al. Novel liquid interface for femtosec- cal manifestations, diagnosis, and treatment. Bina J
ond laser-assisted penetrating keratoplasty. Curr Eye Ophthalmol. 2004;10:90–105.
Res. 2020;45:1051–7. 56. Chirapapaisan C, et al. In vivo confocal microscopy
40. Liu Y, Li X, Li W, Jiu X, Tian M. Systematic review demonstrates increased immune cell densities in cor-
and meta-analysis of femtosecond laser–enabled neal graft rejection correlating with signs and symp-
keratoplasty versus conventional penetrating kerato- toms. Am J Ophthalmol. 2019;203:26–36.
plasty. Eur J Ophthalmol. 2021;31:976–87. 57. Abou Shousha M, et al. In vivo characteristics of
41. Maier P, Böhringer D, Birnbaum F, Reinhard corneal endothelium/descemet membrane complex
T. Improved wound stability of top-hat profiled for the diagnosis of corneal graft rejection. Am J
femtosecond laser-assisted penetrating keratoplasty Ophthalmol. 2017;178:27–37.
in vitro. Cornea. 2012;31:963–6. 58. Van Den Berg R, et al. Descemet’s membrane thick-
42. Seitz B, et al. Inverse mushroom-shaped nonme- ening as a sign for the diagnosis of corneal graft
chanical penetrating keratoplasty using a femtosec- rejection: an ex vivo study. Cornea. 2017;36:1535.
ond laser. Am J Ophthalmol. 2005;139:941–4. 59. Eleiwa TK, et al. Diagnostic performance of three-
43. Saelens IE, Bartels MC, Van Rij G. Manual trephina- dimensional endothelium/descemet membrane com-
tion of mushroom keratoplasty in advanced kerato- plex thickness maps in active corneal graft rejection.
conus. Cornea. 2008;27:650–5. Am J Ophthalmol. 2020;210:48–58.
44. McAllum P, Kaiserman I, Bahar I, Rootman 60. Roshandel D, et al. Current and emerging thera-
D. Femtosecond laser top hat penetrating kerato- pies for corneal neovascularization. Ocul Surf.
plasty: wound burst pressures of incomplete cuts. 2018;16:398–414.
Arch Ophthalmol. 2008;126:822–5. 61. Javadi MA, Feizi S. Mustard gas eye injuries in
45. Maier P, Birnbaum F, Reinhard T. Therapeutic chemical warfare victims; 2015. (Farhang Farda).
applications of the femtosecond laser in corneal sur- 62. Koenig Y, et al. Short-and long-term safety profile
gery. Klinische Monatsblatter fur Augenheilkunde. and efficacy of topical bevacizumab (Avastin®)
2010;227:453–9. eye drops against corneal neovascularization.
46. Williams KA, et al. How effective is penetrating cor- Graefes Arch Clin Exp Ophthalmol. 2009;247:
neal transplantation? Factors influencing long-term 1375–82.
outcome in multivariate analysis. Transplantation. 63. Dastjerdi MH, Sadrai Z, Saban DR, Zhang Q, Dana
2006;81:896–901. R. Corneal penetration of topical and subconjunc-
47. Borderie VM, et al. Predicted long-term out- tival bevacizumab. Invest Ophthalmol Vis Sci.
come of corneal transplantation. Ophthalmology. 2011;52:8718–23.
2009;116:2354–60. 64. Oliveira HB, et al. VEGF TrapR1R2 sup-
48. Anshu A, et al. Long-term review of penetrating presses experimental corneal angiogenesis. Eur J
keratoplasty: a 20-year review in Asian eyes. Am J Ophthalmol. 2010;20:48–54.
Ophthalmol. 2021;224:254–66. 65. Wang Q, et al. Pharmacological characteris-
49. Reinhard T, Böhringer D, Hüschen D, Sundmacher tics and efficacy of a novel anti-angiogenic anti-
R. Chronic endothelial cell loss of the graft after pen- body FD006 in corneal neovascularization. BMC
etrating keratoplasty: influence of endothelial cell Biotechnol. 2014;14:1–9.
migration from graft to host. Klinische Monatsblatter 66. Dekaris I, Gabrić N, Drača N, Pauk-Gulić M, Miličić
fur Augenheilkunde. 2002;219:410–6. N. Three-year corneal graft survival rate in high-
risk cases treated with subconjunctival and topical 83. Böhringer D, et al. Matching of the minor histocom-
bevacizumab. Graefes Arch Clin Exp Ophthalmol. patibility antigen HLA-A1/HY may improve prog-
2015;253:287–94. nosis in corneal transplantation. Transplantation.
67. Jovanovic V, Nikolic L. The effect of topical doxy- 2006;82:1037–41.
cycline on corneal neovascularization. Curr Eye Res. 84. Skeens HM, Holland EJ. Large-diameter penetrat-
2014;39:142–8. ing keratoplasty: indications and outcomes. Cornea.
68. Hata Y, et al. Antiangiogenic properties of fasudil, 2010;29:296–301.
a potent Rho-Kinase inhibitor. Jpn J Ophthalmol. 85. Lass JH, et al. Donor age and factors related to
2008;52:16–23. endothelial cell loss 10 years after penetrating ker-
69. Faraj LA, Elalfy MS, Said DG, Dua HS. Fine nee- atoplasty: specular Microscopy Ancillary Study.
dle diathermy occlusion of corneal vessels. Br J Ophthalmology. 2013;120:2428–35.
Ophthalmol. 2014;98:1287–90. 86. Epstein AJ, de Castro TN, Laibson PR, Cohen EJ,
70. Al-Torbak AA. Photodynamic therapy with verte- Rapuano CJ. Risk factors for the first episode of
porfin for corneal neovascularization. Mid E Afr J corneal graft rejection in keratoconus. Cornea.
Ophthalmol. 2012;19:185. 2006;25:1005–11.
71. Lee HS, Lee JH, Kim CE, Yang JW. Anti-neovascular 87. Williams KA, Roder D, Esterman A, Muehlberg
effect of chondrocyte-derived extracellular matrix SM, Coster DJ. Factors predictive of corneal graft
on corneal alkaline burns in rabbits. Graefes Arch survival: report from the Australian Corneal Graft
Clin Exp Ophthalmol. 2014;252:951–61. Registry. Ophthalmology. 1992;99:403–14.
72. Ferrari G, et al. Topical ranibizumab as a treatment 88. Bidaut-Garnier M, et al. Evolution of corneal graft
of corneal neovascularization. Cornea. 2013;32:992. survival over a 30-year period and comparison of sur-
73. Group, C.D.S.I. The effect of donor age on corneal gical techniques: a cohort study. Am J Ophthalmol.
transplantation outcome: results of the cornea donor 2016;163:59–69.
study. Ophthalmology. 2008;115:620–626.e626. 89. Goble RR, Lea SJH, Falcon MG. The use of
74. Stulting RD, et al. Effect of donor and recipi- the same size host and donor trephine in pen-
ent factors on corneal graft rejection. Cornea. etrating keratoplasty for keratoconus. Eye. 1994;8:
2012;31:1141. 311–4.
75. Barraquer RI, Pareja-Aricò L, Alba Gómez-Benlloch 90. Li C, et al. Effect of corneal graft diameter on thera-
RM. Risk factors for graft failure after penetrating peutic penetrating keratoplasty for fungal keratitis.
keratoplasty. Medicine. 2019;98:e15274. Int J Ophthalmol. 2012;5:698.
76. Williams KA, Lowe M, Bartlett C, Kelly T-L, 91. Beekhuis WH. Current clinician’s opinions on
Coster DJ. Risk factors for human corneal graft risk factors in corneal grafting. Results of a survey
failure within the Australian corneal graft registry. among surgeons in the eurotransplant area. Cornea.
Transplantation. 2008;86:1720–4. 1995;14:39–42.
77. Mannis MJ, et al. The effect of donor age on pen- 92. Wilson SE. Graft failure after penetrating kerato-
etrating keratoplasty for endothelial disease: graft plasty. Surv Ophthalmol. 1990;34:325–56.
survival after 10 years in the Cornea Donor Study. 93. Fasolo A, et al. Risk factors for graft failure after
Ophthalmology. 2013;120:2419–27. penetrating keratoplasty: 5-year follow-up from the
78. Group, C.D.S.I. Donor age and corneal endothelial corneal transplant epidemiological study. Cornea.
cell loss 5 years after successful corneal transplan- 2011;30:1328–35.
tation: specular microscopy ancillary study results. 94. Patel SV. Graft survival after penetrating kerato-
Ophthalmology. 2008;115:627–632.e628. plasty. Am J Ophthalmol. 2011;151:397–8.
79. Lass JH, et al. Baseline factors related to endothelial 95. Sugar J, et al. Donor risk factors for graft failure in
cell loss following penetrating keratoplasty. Arch the cornea donor study. Cornea. 2009;28:981.
Ophthalmol. 2011;129:1149–54. 96. Dunn SP, et al. The effect of ABO blood incompat-
80. Feizi S, et al. Penetrating keratoplasty versus deep ibility on corneal transplant failure in conditions
anterior lamellar keratoplasty in children and ado- with low-risk of graft rejection. Am J Ophthalmol.
lescents with keratoconus. Am J Ophthalmol. 2009;147:432–438.e433.
2021;226:13–21. 97. Feizi S, Javadi MA, Kanavi MR, Javadi F. Effect
81. Velásquez-Monzón K, Navarro-Peña MC, Klunder- of donor graft quality on clinical outcomes after
Klunder M, Tsatsos M, Ramírez-Ortiz MA. Pediatric deep anterior lamellar keratoplasty. Cornea.
penetrating keratoplasty and graft rejection: experi- 2014;33:795–800.
ence at the Hospital Infantil de México Federico 98. The Writing Committee for the Cornea Donor Study
Gómez. Boletín médico del Hospital Infantil de Research Group, et al. Factors predictive of corneal
México. 2020;77:23–7. graft survival in the cornea donor study. JAMA
82. Anshu A, Lim LS, Htoon HM, Tan DT. Postoperative Ophthalmol. 2015;133:246.
risk factors influencing corneal graft survival in 99. Banerjee S, Dick AD, Nicholls SM. Factors affect-
the Singapore Corneal Transplant Study. Am J ing rejection of second corneal transplants in rats1.
Ophthalmol. 2011;151:442–448.e441. Transplantation. 2004;77:492–6.
100. Kitazawa K, et al. Moderately long-term safety and 114. Serna-Ojeda JC, et al. Long-term outcomes of pedi-
efficacy of repeat penetrating keratoplasty. Cornea. atric penetrating keratoplasty for herpes simplex
2018;37:1255–9. virus keratitis. Am J Ophthalmol. 2017;173:139–44.
101. Al-Mezaine H, Wagoner M. Repeat penetrat- 115. Kuffova L, et al. High-risk corneal graft rejection
ing keratoplasty: indications, graft survival, in the setting of previous corneal herpes simplex
and visual outcome. Br J Ophthalmol. 2006;90: virus (HSV)-1 infection. Invest Ophthalmol Vis Sci.
324–7. 2016;57:1578–87.
102. Khairallah AS. Outcome of repeat penetrating kera- 116. Shiuey EJ, et al. Development of a nomogram to pre-
toplasty in eyes with failed penetrating keratoplasty. dict graft survival after penetrating keratoplasty. Am
Saudi Med J. 2016;37:1029–32. J Ophthalmol. 2021;226:32–41.
103. Yalniz-Akkaya Z, et al. Repeat penetrating kerato- 117. Whitcup SM, Nussenblatt RB, Price FW Jr, Chan
plasty: indications and prognosis, 1995–2005. Eur J C-C. Expression of cell adhesion molecules in cor-
Ophthalmol. 2009;19:362–8. neal graft failure. Cornea. 1993;12:475–80.
104. Thompson RW Jr, Price MO, Bowers PJ, Price FW 118. Tan DT, et al. Penetrating keratoplasty in Asian
Jr. Long-term graft survival after penetrating kerato- eyes: the Singapore corneal transplant study.
plasty. Ophthalmology. 2003;110:1396–402. Ophthalmology. 2008;115:975–982.e971.
105. Bachmann B, Taylor RS, Cursiefen C. Corneal neo- 119. Banitt M, Lee R. Management of patients with com-
vascularization as a risk factor for graft failure and bined glaucoma and corneal transplant surgery. Eye.
rejection after keratoplasty: an evidence-based meta- 2009;23:1972–9.
analysis. Ophthalmology. 2010;117:1300–1305. 120. Hollander DA, et al. Graft failure after penetrat-
e1307. ing keratoplasty in eyes with Ahmed valves. Am J
106. Merz PR, Röckel N, Ballikaya S, Auffarth GU, Ophthalmol. 2010;150:169–78.
Schmack I. Effects of ranibizumab (Lucentis®) and 121. Stewart RM, et al. Effect of glaucoma on corneal
bevacizumab (Avastin®) on human corneal endothe- graft survival according to indication for penetrating
lial cells. BMC Ophthalmol. 2018;18:1–8. keratoplasty. Am J Ophthalmol. 2011;151:257–262.
107. Al-Debasi T, Al-Bekairy A, Al-Katheri A, Al Harbi e251.
S, Mansour M. Topical versus subconjunctival 122. Price MO, Thompson RW, Price FW. Risk factors
anti-vascular endothelial growth factor therapy for various causes of failure in initial corneal grafts.
(Bevacizumab, Ranibizumab and Aflibercept) for Arch Ophthalmol. 2003;121:1087–92.
treatment of corneal neovascularization. Saudi J 123. Massry GG, Assil KK. Pilocarpine-associated
Ophthalmol. 2017;31:99–105. allograft rejection in postkeratoplasty patients.
108. Fasciani R, Mosca L, Giannico MI, Ambrogio SA, Cornea. 1995;14:202–5.
Balestrazzi E. Subconjunctival and/or intrastromal 124. Konowal A, et al. Irreversible corneal decompensa-
bevacizumab injections as preconditioning therapy tion in patients treated with topical dorzolamide. Am
to promote corneal graft survival. Int Ophthalmol. J Ophthalmol. 1999;127:403–6.
2015;35:221–7. 125. Cahane M, Ashkenazi I, Urinowski E, Avni
109. Hos D, et al. Risk of corneal graft rejection after I. Corneal graft rejection after neodymium-yttrium-
high-risk keratoplasty following fine-needle vessel aluminum-garnet laser posterior capsulotomy.
coagulation of corneal neovascularization combined Cornea. 1992;11:534–7.
with bevacizumab: a pilot study. Transplant Dir. 126. DeBacker CM, El-Naggar S, Sugar J, Lai WW. Effect
2019;5:e452. of neodymium: YAG laser posterior capsulotomy on
110. Koenig Y, Bock F, Kruse FE, Stock K, Cursiefen corneal grafts. Cornea. 1996;15:15–7.
C. Angioregressive pretreatment of mature corneal 127. Lass JH, et al. The effect of donor diabetes his-
blood vessels before keratoplasty: fine-needle ves- tory on graft failure and endothelial cell density 10
sel coagulation combined with anti-VEGFs. Cornea. years after penetrating keratoplasty. Ophthalmology.
2012;31:887–92. 2015;122:448–56.
111. Cursiefen C, et al. Antisense oligonucleotide eye 128. Zhang X, et al. Association of smoking and other risk
drops against IRS-1 inhibit corneal neovasculariza- factors with Fuchs’ endothelial corneal dystrophy
tion: interim results of a randomized phase II clinical severity and corneal thickness. Invest Ophthalmol
trial. Invest Ophthalmol Vis Sci. 2009;50:4953. Vis Sci. 2013;54:5829–35.
112. Cursiefen C, et al. Aganirsen antisense oligo- 129. Jurkunas UV, Bitar MS, Funaki T, Azizi B. Evidence
nucleotide eye drops inhibit keratitis-induced of oxidative stress in the pathogenesis of fuchs
corneal neovascularization and reduce need for endothelial corneal dystrophy. Am J Pathol.
transplantation: the I-CAN study. Ophthalmology. 2010;177:2278–89.
2014;121:1683–92. 130. Karadag R, et al. Survival of primary penetrat-
113. Altenburger AE, Bachmann B, Seitz B, Cursiefen ing keratoplasty in children. Am J Ophthalmol.
C. Morphometric analysis of postoperative corneal 2016;171:95–100.
neovascularization after high-risk keratoplasty: her- 131. Alice LY, et al. Perioperative and postoperative risk
petic versus non-herpetic disease. Graefes Arch Clin factors for corneal graft failure. Clin Ophthalmol.
Exp Ophthalmol. 2012;250:1663–71. 2014;8:1641.
132. Hayashi T, et al. Pars plana vitrectomy combined risk corneal transplantation. J Ophthal Vis Res.
with penetrating keratoplasty and transscleral- 2017;12:81.
sutured intraocular lens implantation in complex 151. Shimazaki J, Iseda A, Satake Y, Shimazaki-Den
eyes: a case series. BMC Ophthalmol. 2020;20:369. S. Efficacy and safety of long-term corticosteroid
133. Rahman I, Carley F, Hillarby C, Brahma A, Tullo eye drops after penetrating keratoplasty: a prospec-
A. Penetrating keratoplasty: indications, outcomes, tive, randomized, clinical trial. Ophthalmology.
and complications. Eye. 2009;23:1288–94. 2012;119:668–73.
134. Zhang Y, et al. Indications and outcomes of penetrat- 152. Ang M, Soh Y, Htoon HM, Mehta JS, Tan
ing keratoplasty in infants and children of Beijing, D. Five-year graft survival comparing Descemet
China. Cornea. 2018;37:1243–8. stripping automated endothelial keratoplasty
135. Majander A, Kivelä TT, Krootila K. Indications and and penetrating keratoplasty. Ophthalmology.
outcomes of keratoplasties in children during a 40‐ 2016;123:1646–52.
year period. Acta Ophthalmol. 2016;94:618–24. 153. Jordan CS, Price MO, Trespalacios R, Price
136. Sugar A, et al. Factors associated with corneal FW. Graft rejection episodes after Descemet strip-
graft survival in the cornea donor study. JAMA ping with endothelial keratoplasty: part one:
Ophthalmol. 2015;133:246–54. clinical signs and symptoms. Br J Ophthalmol.
137. Young AL, Kam K, Jhanji V, Cheng LL, Rao SK. A 2009;93:387–90.
new era in corneal transplantation: paradigm shift 154. Niederkorn JY. High risk corneal allografts and why
and evolution of techniques. Hong Kong Med J. they lose their immune privilege. Curr Opin Allergy
2012;18:509–16. Clin Immunol. 2010;10:493.
138. Olson RJ. Variation in corneal graft size related to tre- 155. Niederkorn JY. Corneal transplantation and immune
phine technique. Arch Ophthalmol. 1979;97:1323–5. privilege. Int Rev Immunol. 2013;32:57–67.
139. Wilson SE, Bourne WM. Effect of recipient-donor 156. Hudde T, Minassian D, Larkin D. Randomised
trephine size disparity on refractive error in kerato- controlled trial of corticosteroid regimens in endo-
conus. Ophthalmology. 1989;96:299–305. thelial corneal allograft rejection. Br J Ophthalmol.
140. Javadi MA, Mohammadi MJ, Mirdehghan SA, 1999;83:1348–52.
Sajjadi SH. A comparison between donor- 157. Dunn SP, et al. Corneal graft rejection ten years after
recipient corneal size and its effect on the ultimate penetrating keratoplasty in the cornea donor study.
refractive error induced in keratoconus. Cornea. Cornea. 2014;33:1003.
1993;12:401–5. 158. Armitage WJ, et al. High-risk corneal transplanta-
141. Kuryan J, Channa P. Refractive surgery after corneal tion: recent developments and future possibilities.
transplant. Curr Opin Ophthalmol. 2010;21:259–64. Transplantation. 2019;103:2468.
142. Lanier JD, Bullington RH, Prager TC. Axial length 159. Holland EJ, et al. Systemic immunosuppression in
in keratoconus. Cornea. 1992;11:250–4. ocular surface stem cell transplantation: results of a
143. Javadi MA, et al. Comparison of the effect of three 10-year experience. Cornea. 2012;31:655–61.
suturing techniques on postkeratoplasty astigmatism 160. Streilein JW. Immunobiology and immunopathol-
in keratoconus. Cornea. 2006;25:1029–33. ogy of corneal transplantation. Chem Immunol.
144. Solano JM, Hodge DO, Bourne WM. Keratometric 1999;73:186–206.
astigmatism after suture removal in penetrating 161. Barbarino JM, Staatz CE, Venkataramanan R, Klein
keratoplasty: double running versus single running TE, Altman RB. PharmGKB summary: cyclosporine
suture techniques. Cornea. 2003;22:716–20. and tacrolimus pathways. Pharmacogenet Genomics.
145. Romano V, Iovieno A, Parente G, Soldani AM, 2013;23:563–85.
Fontana L. Long-term clinical outcomes of deep 162. Fahr A. Cyclosporin clinical pharmacokinetics. Clin
anterior lamellar keratoplasty in patients with kera- Pharmacokinet. 1993;24:472–95.
toconus. Am J Ophthalmol. 2015;159:505–11. 163. Thomson A, Bonham C, Zeevi A. Mode of action of
146. Javadi MA, Feizi S, Yazdani S, Mirbabaee F. Deep tacrolimus (FK506): molecular and cellular mecha-
anterior lamellar keratoplasty versus penetrating nisms. Ther Drug Monit. 1995;17:584–91.
keratoplasty for keratoconus: a clinical trial. Cornea. 164. Kharod-Dholakia B, Randleman JB, Bromley JG,
2010;29:365–71. Stulting RD. Prevention and treatment of corneal
147. Feizi S, Zare M. Current approaches for manage- graft rejection: current practice patterns of the
ment of postpenetrating keratoplasty astigmatism. J Cornea Society (2011). Cornea. 2015;34:609–14.
Ophthalmol. 2011;2011:708736. 165. Xie L, Shi W, Wang Z, Bei J, Wang S. Prolongation
148. Dumitrescu O-M, Istrate S, Macovei M-L, Gheorghe of corneal allograft survival using cyclosporine
AG. Intraocular pressure measurement after pen- in a polylactide-co-glycolide polymer. Cornea.
etrating keratoplasty. Diagnostics. 2022;12:234. 2001;20:748–52.
149. McGhee C. Pharmacokinetics of ophthalmic corti- 166. Thompson P, Xu D, Brunette I, Chen H. Combined
costeroids. Br J Ophthalmol. 1992;76:681. effect of rapamycin and cyclosporine in the preven-
150. Abud TB, Di Zazzo A, Kheirkhah A, Dana tion of rat corneal allograft rejection. Transplant
R. Systemic immunomodulatory strategies in high- Proc. 1998;4:1033–5.
167. Fu H, Larkin DF, George AJ. Immune modula- mice at high risk of rejection. BMC Ophthalmol.
tion in corneal transplantation. Transplant Rev. 2004;4:1–7.
2008;22:105–15. 183. Chen L, et al. Vascular endothelial growth factor
168. Williams KA, Coster DJ. Use of monoclo- receptor-3 mediates induction of corneal alloimmu-
nal antibodies in corneal transplantation. Clin nity. Nat Med. 2004;10:813–5.
Immunotherapeut. 1994;2:32–41. 184. Gong N, Pleyer U, Volk H, Ritter T. Effects of
169. Thiel M, Kaufmann C, Coster D, Williams local and systemic viral interleukin-10 gene trans-
K. Antibody-based immunosuppressive agents for fer on corneal allograft survival. Gene Ther.
corneal transplantation. Eye. 2009;23:1962–5. 2007;14:484–90.
170. Singh S, et al. Monoclonal antibodies: a review. Curr 185. Nosov M, et al. Role of lentivirus‐mediated
Clin Pharmacol. 2018;13:85–99. overexpression of programmed death‐ligand 1
171. Thiel MA, Coster DJ, Williams KA. The poten- on corneal allograft survival. Am J Transplant.
tial of antibody‐based immunosuppressive agents 2012;12:1313–22.
for corneal transplantation. Immunol Cell Biol. 186. Pleyer U, et al. Survival of corneal allografts fol-
2003;81:93–105. lowing adenovirus-mediated gene transfer of
172. Merediz S, Zhang E-P, Wittig B, Hoffmann interleukin-4. Graefes Arch Clin Exp Ophthalmol.
F. Ballistic transfer of minimalistic immunologically 2000;238:531–6.
defined expression constructs for IL4 and CTLA4 187. Kocaba V, Colica C, Rabilloud M, Burillon
into the corneal epithelium in mice after orthotopic C. Predicting corneal graft rejection by confocal
corneal allograft transplantation. Graefes Arch Clin microscopy. Cornea. 2015;34:S61–4.
Exp Ophthalmol. 2000;238:701–7. 188. Webster AC, Pankhurst T, Rinaldi F, Chapman JR,
173. Bluestone JA, St Clair EW, Turka LA. CTLA4Ig: Craig JC. Monoclonal and polyclonal antibody ther-
bridging the basic immunology with clinical appli- apy for treating acute rejection in kidney transplant
cation. Immunity. 2006;24:233–8. recipients: a systematic review of randomized trial
174. Thiel MA, et al. Local or short‐term systemic data. Transplantation. 2006;81:953–65.
costimulatory molecule blockade prolongs rat 189. Hashemian MN, et al. Topical tacrolimus as adju-
corneal allograft survival. Clin Exp Ophthalmol. vant therapy to corticosteroids in acute endothe-
2005;33:176–80. lial graft rejection after penetrating keratoplasty:
175. Wu X-S, et al. Tocilizumab promotes cor- a randomized controlled trial. Cornea. 2018;37:
neal allograft survival in rats by modulating 307–12.
Treg-Th17 balance. Int J Ophthalmol. 2019; 190. Reinhard T, Sundmacher R. Adjunctive intracameral
12:1823. application of corticosteroids in patients with endo-
176. Hamrah P, et al. Local treatment with alpha- thelial immune reactions after penetrating kerato-
melanocyte stimulating hormone reduces cor- plasty: a pilot study. Transpl Int. 2002;15:81–8.
neal allorejection. Transplantation. 2009; 191. Maris PJGJ, Correnti AJ, Donnenfeld
88:180. ED. Intracameral triamcinolone acetonide as treat-
177. Qian Y, Dekaris I, Yamagami S, Dana MR. Topical ment for endothelial allograft rejection after pen-
soluble tumor necrosis factor receptor type I sup- etrating keratoplasty. Cornea. 2008;27:847–50.
presses ocular chemokine gene expression and 192. Kim BZ, et al. New Zealand trends in corneal
rejection of allogeneic corneal transplants. Arch transplantation over the 25 years 1991-2015. Br J
Ophthalmol. 2000;118:1666–71. Ophthalmol. 2017;101:834–8.
178. Amouzegar A, Chauhan SK. Effector and regula- 193. Alldredge OC, Krachmer JH. Clinical types of cor-
tory T cell trafficking in corneal allograft rejection. neal transplant rejection: their manifestations, fre-
Mediat Inflamm. 2017;2017:8670280. quency, preoperative correlates, and treatment. Arch
179. Donnenfeld ED, et al. Safety of lifitegrast ophthal- Ophthalmol. 1981;99:599–604.
mic solution 5.0% in patients with dry eye disease: a 194. Hill JC, Maske R, Watson P. Corticosteroids in cor-
1-year, multicenter, randomized, placebo-controlled neal graft rejection: oral versus single pulse therapy.
study. Cornea. 2016;35:741. Ophthalmology. 1991;98:329–33.
180. Semba CP, Gadek TR. Development of lifitegrast: 195. Randleman JB, Stulting RD. Prevention and treat-
a novel T-cell inhibitor for the treatment of dry eye ment of corneal graft rejection: current practice pat-
disease. Clin Ophthalmol. 2016;10:1083. terns (2004). Cornea. 2006;25:286–90.
181. Ma D, Mellon J, Niederkorn JY. Oral immunisation 196. Poon A, Constantinou M, Lamoureux E, Taylor
as a strategy for enhancing corneal allograft survival. HR. Topical Cyclosporin A in the treatment of acute
Br J Ophthalmol. 1997;81:778–84. graft rejection: a randomized controlled trial. Clin
182. Plšková J, Holáň V, Filipec M, Forrester JV. Lymph Exp Ophthalmol. 2008;36:415–21.
node removal enhances corneal graft survival in
Improvements in Surgical
Techniques and Suturing
6
in Penetrating Keratoplasty
Abdo Karim Tourkmani and Colm McAlinden
Key Points benefit from corneal transplantation, what

• Full-thickness corneal transplantation has impact surgery may have on other conditions,
stood the test of time and is still a boon to cor- and what impact other conditions may have on
neal surgeons in the twenty-first century. From graft success and survival.
the first description by Reisinger in 1824 and • PKP is indicated for corneal diseases affecting
the landmark case report by Zirm in 1905, all layers of the cornea, except the epithelium,
many modifications ensued including the including Combined endothelial and stromal
switching of square grafts to round grafts by diseases (e.g. Fuchs endothelial dystrophy
Castroviejo. The development of the surgical with keratoconus), endothelial disease (e.g.
microscope, suture materials, and topical ste- Fuchs endothelial dystrophy) with significant
roids has had a profound impact on the suc- secondary stromal scarring and keratoconus
cess of penetrating keratoplasty (PKP). The with previous hydrops.
femtosecond laser is one of the latest develop-
ments which adds precision to depth and
diameter of tissue penetration as well as the istory of Penetrating Keratoplasty
H
ability to create various edge profiles for (PKP)
grafts, such as mushroom, top hat and zigzag
configurations. In 1905, Austrian Ophthalmologist Eduard Zirm
• The most important aspect of pre-operative was the first surgeon to successfully perform a
planning is an assessment of the contribution full-thickness corneal transplant which remained
to visual impairment by corneal disease. One clear [1]. Interestingly, the indication was for a
needs to consider how likely the patient will bilateral chemical injury, which even today
would be deemed a poor candidate for transplan-
Supplementary Information The online version con- tation. There were many attempts prior to the
tains supplementary material available at https://doi. report from Zirm, dating back to the early 1800s,
org/10.1007/978-3-031-32408-6_6.
which all failed, with graft transparency not last-
ing more than 2–3 weeks [2]. Franz Reisinger
A. K. Tourkmani (*) first introduced the term keratoplasty in 1824 [3],
Department of Ophthalmology, Royal Gwent and Michael Marcus described the first set of
Hospital, Newport, UK basic steps of full-thickness keratoplasty in 1840
C. McAlinden [4]. Towards the end of the 1800s, there was great
Corneo Plastic Unit & Eye Bank, Queen Victoria interest and work on lamellar corneal grafting;
Hospital, East Grinstead, UK
https://doi.org/10.1007/978-3-031-32408-6_6
104 A. K. Tourkmani and C. McAlinden
however, with the turn of the century, following stem cell deficiency may require stem cell
the landmark case report by Zirm, there was a transplantation prior to PKP [9].
resurgence in interest in full-thickness • Inflammation/infection:
keratoplasty. –– Active inflammation/infection is a risk fac-
In the 50 years following the publication by tor for graft failure. Inflammation should be
Zirm, many advances in keratoplasty ensued. well controlled prior to surgery. In eyes
One of the most notable was the work of Ramon with infections such as fungal, acantham-
Castroviejo, progressing from square-shaped oeba and herpetic keratitis, they should be
grafts to circular grafts as well as progressing quiescent for many months prior to PKP.
from overlay sutures to appositional sutures [5]. The exception to this would be tectonically
Richard Townley Paton developed the first eye grafting in co-existing perforation [10].
bank in 1944 in New York, USA, which paved • Lid abnormalities:
the way for greater access to corneal tissue [6]. –– Lid conditions such as ectropion and entro-
Other advances such as the development of the pion should be surgically corrected prior to
surgical microscope, suture materials, and topical PKP to reduce the risk of failure
steroids have had a profound impact on the suc- post-operatively.
cess of penetrating keratoplasty (PKP). In more • Glaucoma:
recent times, the femtosecond laser has added to –– Glaucoma is an independent risk factor for
the evolution of PKP [7]. The laser adds preci- graft failure. Further, poorly controlled intra-
sion to depth and diameter of tissue penetration ocular pressure (IOP) is associated with
as well as the ability to create various edge pro- endothelial cell loss [11]. Hence good IOP
files for grafts, such as mushroom, top hat and control is essential pre-operatively. Glaucoma
zigzag configurations [8]. drainage devices (GDD) are highly associ-
ated with graft failure occurring in approxi-
mately 45% of eyes at 3 years [12]. The
Pre-operative Considerations optimal glaucoma procedure and timing in
PKP eyes remain to be established [13].
In the past two decades, lamellar corneal grafting
has superseded PKP, with the selective replacement Patients need to understand the lengthy post-
of only the diseased layers of the cornea. However, operative period of visual rehabilitation as well
PKP still has a very important role. Perhaps, the as post-operative topical therapy as well as the
most important aspect of pre-operative planning is life-long risk of rejection and graft failure. A full
an assessment of the contribution to visual impair- ophthalmic and systemic history is vital as well
ment by corneal disease. Ocular co-morbidities as anaesthetic considerations. With advancing
need to be carefully considered. One needs to con- patient age, there are lower rejection rates [14];
sider how likely the patient will benefit from cor- however, advanced age is a risk factor for supra-
neal transplantation, what impact surgery may have choroidal haemorrhage.
on other conditions, and what impact other condi- Generally, PKP is indicated for corneal dis-
tions may have on graft success and survival. eases affecting all layers of the cornea except the
The following co-morbidities require specific epithelium. These include:
consideration:
• Combined endothelial and stromal diseases
• Ocular surface disease (OSD): (e.g. Fuchs endothelial dystrophy with
–– This is the leading cause of graft failure, keratoconus)
and hence it is imperative that pre-operative • Endothelial disease (e.g. Fuchs endothelial
OSD is aggressively treated. These mea- dystrophy) with significant secondary stromal
sures need to be maintained in the post- scarring
operative period also. Those with limbal • Keratoconus with previous hydrops
6 Improvements in Surgical Techniques and Suturing in Penetrating Keratoplasty 105
Relative contraindications to PKP include: sia, 5 patients (4.3%) developed suprachoroidal

haemorrhages [17]. Sudden drops in intraocular
• Epithelial disease (e.g. limbal stem cell failure pressure intraoperatively should be avoided;
and severe dry eye) when entering the anterior chamber, this should
• Significant stromal vascularisation be performed slowly, followed by an injection of
• Multiple previous failed grafts viscoelastic. The duration of the eye left open
should be kept to a minimum.
However, even in cases of full-thickness scars or There are many variations to the surgical tech-
perforations, some surgeons may consider deep nique of PKP; however, the main goals are to
anterior lamellar keratoplasty (DALK) or safely perform surgery without complications
Descemet Stripping Endothelial Keratoplasty and achieve good wound apposition with mini-
(DSEK) [15], particularly in higher-risk eyes, to mal astigmatism induction. The main general
avoid the risk of endophthalmitis or expulsive steps of the procedure are summarised below:
haemorrhage with open-sky PKP. It can be sur-
prising how well some patients see despite small • Scleral fixation ring: the purpose of this ring is
residual scars. to provide scleral structural support when the
eye is open to reduce the risk of suprachoroi-
dal haemorrhage. The ring is sized just within
Surgical Techniques the interpalpebral opening and sutured to the
sclera, typically with 7-0 silk sutures.
A clear surgical plan is a prerequisite to safe sur- • The diameter of the host cornea is measured,
gery, with efficient use of time to maintain as and the centre is marked with a marking pen.
short as possible operating time (Video 6.1). Sizing depends on numerous factors including
Modifications need to be planned such as if cata- pathology, the extent of pathology, host cor-
ract surgery is planned with PKP. neal diameter and risk of rejection. A trephine
Povidone iodine pre-operatively is important to make a circular cut on both the donor and
in reducing post-operative endophthalmitis. This recipient cornea. These can be hand-held,
can be administered as 5% eye drops many min- motor (e.g. Asmotom), suction (e.g. Hessburg-
utes prior to draping and further with 10% painted Barron) or guided (e.g. Hanna). However, the
to the lashes and peri-ocular region. In patients most modern method of performing the cuts is
undergoing combined PKP with cataract surgery, with the use of a femtosecond laser. A
the pupil is dilated pre-operatively, whereas, in 7.5–8 mm trephination to the host cornea is a
those who are phakic, 2% pilocarpine drops are typical size. The trephination is ideally cen-
administered. tred on the corneal marking; however, it may
Suprachoroidal haemorrhage is a dreaded need to be decentred to account for the pathol-
complication with the main risk factors being ogy, e.g., peripheral thinning. It is a trade-off
age, glaucoma (and in particular, high pre- with increasing astigmatism induction with
operative IOP), previous surgery, hypertension, larger decentrations.
anticoagulant therapy, tachycardia, arteriosclero- • Donor cornea trephination: the punch trephi-
sis and previous suprachoroidal haemorrhage nation is performed with the corneal
[16]. Steps should be taken to reduce these fac- endothelial side up and is sized 0.25 mm

tors pre-operatively, e.g. ensure good blood pres- larger than the planned host cornea trephina-
sure control and safely stop anticoagulants tion. Alternatively, the femtosecond laser may
pre-operatively. Ingraham and colleagues be used (Fig. 6.1).
reported, in a group of 714 patients undergoing • Host cornea trephination: the size of the host
PKP under general anaesthesia, 4 patients corneal trephination is 0.25–0.5 mm less than
(0.56%) developed suprachoroidal haemor- the donor cornea. The trephine is placed on
rhages, and in 116 patients under local anaesthe- the host cornea, and most commonly, suction
a b
c d
Fig. 6.1 The CORONET® donor corneal trephine punch. (a) Base to place donor cornea epithelium side down (i.e.
endothelium side up), (b) trephine guide added, (c) independent trephine, (d) trephine added to the system
Fig. 6.2 30° blade used to penetrate the anterior chamber after almost full-thickness trephination
is applied. Alternatively, a free-hand trephine Fig. 6.2) can be used to gently penetrate the
without suction may be used. The cutting anterior chamber in a controlled manner.
blade is lowered by rotating the dial. The tre- Viscoelastic can then be injected into the ante-
phine can be made to almost full-thickness or rior chamber.
full-thickness, whereby a small aqueous leak • Right and left-bevelled corneal scissors are
is usually observed. In almost full-thickness used to excise the host cornea in a perpendicu-
trephinations, a blade (e.g. MVR or 30° blade, lar fashion; however, slight inward angling
creates a posterior wound ledge that may

assist with a watertight wound.
• With a Paton spatula, the donor cornea is care-
fully positioned and sutured into the host, and
suturing may begin. This starts with four car-
dinal 10-0 nylon interrupted sutures, typically
starting at the 12 o’clock position, followed by
6 o’clock, then 3 and 9 o’clock. The second
suture is the most important in terms of astig-
matism and graft alignment. Fine double-
toothed forceps are used to grasp the donor
Fig. 6.4 Instrument set with commonly used PKP instru-
cornea and the suture passed, aiming for 90% ments (not exhaustive). From left to right: straight tying
depth. The suture is tied in a 3-1-1 fashion or forceps, toothed forceps, rhexis forceps (for cataract
a slipknot. A further 8–12 interrupted sutures extraction), Vannas scissors, keratoplasty suture marker,
are placed (16 in total is typical) (Fig. 6.3). spring scissors and needle holder
Alternatively, a running suture may be placed
after eight interrupted sutures. Suture knots
should be buried in the donor graft as this will Suture Materials and Sizes
be further from the limbus. One should avoid
the temptation to over-tighten sutures, as this Suture material can be classified into non-
can lead to cheese wiring, flatter corneas, high absorbable or absorbable, and structurally, mono-
astigmatism, hyperopia and surface healing filament or multi-filament. It can also be classified
issues. A Bowman-to-Bowman layer align- as natural or synthetic. Absorbable sutures are
ment and apposition should be the aim. used for temporary wound closure and undergo
Further details on suturing techniques and enzymatic degradation or hydrolysis with time.
types will be discussed later in this chapter. A Non-absorbable sutures are used for longer-term
typical PKP instrument set is shown in wound closure, and although they are non-
Fig. 6.4. absorbable, they may degrade with time and, in
• Suture modifications can be made to optimise the process, loose tensile strength. Monofilament
astigmatism, aided with the use of an intraop- sutures contain a single strand of the material,
erative keratoscope or aberrometry. which is more resistant to harbouring microor-
• Surgery is completed with an injection of sub- ganisms. Multi-filament sutures have multiple
conjunctival steroids and antibiotics. strands that are braided together, permitting good
handling properties but may attract
microorganisms.
Absorbable
• Synthetic
–– Polygalactin 910 (Vicryl)
–– Polydioxanone (PDS)
• Natural:
–– Fast/plain/chromic gut
Fig. 6.3 16 interrupted 10-0 nylon corneal sutures

Non-absorbable important to emphasise some principles that are

common to whichever type of suturing is
• Synthetic chosen.
–– Nylon (e.g. Ethilon)
–– Polypropylene (e.g. Prolene) • The material chosen for suturing would be a
–– Polyester (e.g. Dacron, Ethibond, nylon suture. Only in the case of a double con-
Mersilene) tinuous suture, the surgeon may choose a 10-0
• Natural: nylon for both sutures, or 10-0 for one and
–– Silk 11-0 for the second suture [20].
• The depth of the needle passage in both graft
Nylon is the most effective suture material in and host should be around 90%. This reduces
keratoplasty surgery. Its relatively elastic, bio- the chance of having steps at the level of the
compatible and strength properties are superior graft-host junction [21].
to other materials for this purpose. Prolene or • The length of the bite should be tailored to
Mersilene can cause stromal fibrosis. Also, the each individual case, however, the longer the
lack of elasticity can cause issues when post- bite, the greater the compression area (hence
operative corneal oedema resolves. Silk is com- less chance of post-operative leaks).
monly used to suture the scleral fixation ring • Whether using interrupted sutures or continu-
intraoperatively. It is very easy to handle and ous sutures, there is a need for implanting first
knot but is associated with significant tissue reac- the initial four cardinal sutures. The surgeon
tion; hence its use is limited to intraoperative in usually starts with the suture at the 12 o’clock
PKP surgery. position, then the one opposite to it (6 o’clock),
Suture sizes are based on the United States aiming for an equal distribution of the graft
Pharmacopeia (USP) system. The initial sizes over the trephination area. Then, two sutures
introduced were #1 to #6, with higher numbers would be placed along the 3–9 o’clock merid-
indicating larger suture diameter. Later, size 0 ian, choosing first the clock hour in which the
was introduced as a thinner suture. Further gap between graft and host is larger.
thinner suture diameters were introduced, and • The knot has to be locked, so that it does not
the system was modified to 1-0 to 12-0, with loosen; then buried into the cornea, so that it is
1-0 being the thickest and 12-0 being the thin- not exposed (to avoid the risk of suture-related
nest. The diameter of 10-0 is approximately infection and corneal neovascularisation,
20 μm. amongst others) [22].
We will now proceed with a more detailed

Suturing in Penetrating explanation of each one of the suturing tech-
Keratoplasty niques referenced above.
Correct suturing is of paramount importance for

an appropriate wound apposition, ensuring the Interrupted Suturing
chances of wound leaks, loose stitches or any
other graft-host junction problems are reduced to This is the technique usually employed at the
the minimum possible [18]. Several suturing beginning of the learning curve for a PKP surgeon
techniques have been described in penetrating [23]. It is usually easier than the others, and the
keratoplasty (PKP). These include interrupted surgeon does not rely on the same suture for the
suturing, continuous single suturing and continu- whole process. It is more forgiving with a needle
ous double suturing. For continuous suturing, becoming bent or blunt from numerous (both suc-
whether single or double, they can be classified cessful and unsuccessful) passes, or a suture break-
as torque or anti-torque suturing [19]. ing halfway through the surgery (common events
We will describe each one of these types in for any surgeon irrespective of the experience, but
greater detail later in this text, however, it is far more common in the early learning curve).
The surgeon would start the suturing by plac-

ing the four cardinal sutures, as explained earlier
in the text. Then, the surgeon would continue by
placing interrupted sutures halfway between the
sutures that have been already placed (45–225°
meridian and 135–315° meridian). The principles
to follow would be:
• Start in the quadrant where the gap between

graft and host is greatest.
• Continue with suturing along the same axis
but in the opposite quadrant. Fig. 6.6 This image is of the same patient as in Fig. 6.5,
• Repeat the action on the meridian 90° from of following cataract surgery and removal of all interrupted
the meridian that has just been sutured. graft sutures. Note, a suture to the main corneal wound is
shown in this image
At this stage, the donor button has been fixed

to the recipient's eye with eight interrupted
sutures in the 90–270° meridian, 0–180° merid- Continuous Suturing (Single)
ian, 45–225° meridian and 135–315° meridian.
Ideally, each one of the sutures should be equidis- The operating surgeon may choose to use a single
tant from one another. continuous suture. The basic principles would
The surgeon then should continue by placing still apply. Preferably, a 10-0 nylon suture would
the next interrupted suture in the area where the be used [23]. The first four interrupted cardinal
gap between the graft and host is greatest, equi- sutures still need to be placed for an appropriate
distantly from the sutures to either side; then con- fixation of the donor button to the host recipient
tinue with the suture along the same meridian but eye, which would then allow for an appropriate
in the opposite quadrant; then repeat the action in continuous suture to be placed.
the meridian 90° away. The running suture would start in the outer lip
As a rule of thumb, the above steps should be of the graft-host junction (i.e. within the “host”
repeated until 16 interrupted 10-0 nylon sutures lip), so that, at the completion of the suture, it
have been placed. However, for non-standard would exit from the inner lip (i.e. within the
PKPs (very large or very small, irregular trephi- “donor” lip), allowing for the knot to be buried in
nations, etc.), more or less sutures may be the graft-host junction. Approximately three
required (Figs. 6.5 and 6.6). equidistant bites would be placed per quadrant,
plus a bite overriding the cardinal interrupted
sutures. Thus, usually, 16 bites are applied. The
four cardinal interrupted sutures can be removed,
the running suture is tightened and adjusted, and
the knot is tied and buried.
The suture can be torque (the bites through the
cornea are radial; the thread joining consecutive
bites is oblique over the cornea) or anti-torque
(the bites through the cornea are oblique; the
thread joining consecutive bites are radial over
the cornea). Typically, if a single continuous
suture is chosen, a torque orientation is utilised
Fig. 6.5 PKP for herpetic corneal perforation with 15 (Fig. 6.7).
interrupted 10-0 nylon corneal sutures. The superior car-
dinal suture became loose and had been removed. Note
this patient later developed a white dense cataract with
posterior synechiae
a b
Fig. 6.7 Schematic diagram showing (a) Torque (the joining consecutive bites are radial over the cornea) run-
bites through the cornea are radial; the thread joining con- ning suture. The dashed lines depict the suture deep to the
secutive bites is oblique over the cornea) and (b) anti- corneal surface. The solid lines depict the suture superfi-
torque (the bites through the cornea are oblique; the thread cial to the corneal surface
Continuous Suturing (Double) Spadea et al. compared the outcomes of a

single-running suture to a double torque–anti-
Double Torque torque suture in PKP [24]. The first group of 35
eyes received a 16 bite 10-0 nylon running suture,
This would consist of a double torque type suture, and the second group of 33 eyes received an 8 + 8
in which the bites of the second suture would run bite 10-0 nylon double running suture, with one
in between the bites of the first suture [19]. The running clockwise and one running anti-
first suture would usually be 10-0 nylon, with the clockwise (i.e. double torque–anti-torque). In the
second suture being either again 10-0 nylon or single-running suture group, five patients had a
11-0 nylon. suture adjustment in the post-operative period,
and no adjustment was made in any patients with
the double-running suture. Sutures were removed
Double Torque–Anti-torque between 12 and 20 months post-operatively.
Corneal topography was performed in all patients
In this technique, the first suture would be a 10-0 after all sutures were removed. The topographic
nylon torque suture, with the second one being astigmatism ranged from 1.07 to 8.09 D (mean
either a 10-0 or 11-0 anti-torque nylon suture. 3.51 ± 1.93 SD) in the single-running suture
When placed correctly, these sutures create the group and from 1.13 to 9.36 D (3.42 ± 1.94) in
appearance of equilateral triangles with geomet- the double-running torque–anti-torque group.
rical perfection. Also, the theoretical advantage The refractive cylinder ranged from 1.00 to
of using an anti-torque suture in combination 4.50 D (2.62 ± 1.20) in the single-running suture
with a torque suture would be “locking” the group and from 1.00 to 4.50 D (2.73 ±1.17) in the
donor graft button in the current “post-torque double-running torque–anti-torque group. The
suture” position, rather than inducing further mean topographic keratometric values ranged
rotation (hence increased cylinder) within the from 37.45 to 52.24 D (45.47 ± 3.82) in the
corneal donor button with a further second torque single-running suture group and from 40.90 to
suture (Fig. 6.8). 51.84 D (46.25 ± 2.99) in the double-running
a b
Fig. 6.8 DALK case with corneal leukomas secondary to torque suture is used in this case with an excellent astig-
infectious keratitis associated to contact lens misuse matic refractive outcome. (Courtesy of Jorge L. Alio)
before (a) and after (b) surgery. Double torque–anti-
torque–anti-torque group. No significant change In a group of 40 patients, Sharma et al. studied

in mean central corneal power and astigmatism the post-operative topographic patterns following
occurred in either group before and after the torque, anti-torque and no torque continuous
suture removal. Comparing final astigmatism, the suturing in PKP [26]. Patients were randomly
two groups’ corneal curvature and refractive cyl- assigned to a group, and 10-0 nylon was used in
inder showed no statistically significant differ- all cases. Patients with 3 D or more astigmatism
ences (P > 0.05). No patient developed 1 month post-operatively underwent suture
post-operative glaucoma, wound dehiscence and adjustment at the slit lamp. With a jeweller and
suture-related complications during the 36 tying forceps, the suture was pulled from the flat-
months of the follow-up period [24]. ter meridians to the steeper meridians. The anti-
Vajpayee et al. compared three continuous torque group showed a significantly higher
running suture techniques, namely, torque, anti- number of prolate pattern maps post-operatively
torque and no torque in PKP [25]. Fifty-three compared with the other suture groups, and fur-
patients were randomly assigned to one of the ther, a significantly higher proportion of patients
three suture types. In total, 17 eyes had torque, 18 in the prolate group required suture adjustment.
eyes had anti-torque and 18 eyes had no torque. Following suture adjustment, the frequency of
All patients with astigmatism of more than 3 D at prolate patterns decreased, but most reverted to
4 weeks underwent suture adjustment at the slit the prolate pattern at 3 months. Among the eyes
lamp. Astigmatism at 4 weeks (before any suture with a prolate corneal pattern, the torque group
adjustment) was higher in the torque group on had the highest astigmatism (11.21 ± 6.21 D),
keratometry, refraction and topography, but this whereas the astigmatism in the anti-torque group
was not statistically significant. Thirty eyes had was 6.1 ± 2.52 D, and in the no-torque group, it
more than 3 D of astigmatism and underwent was 8.85 ± 2.6 D [26].
suture adjustment (12 torque group, 11 anti- Nuzzi et al. recently compared five suturing
torque group, 7 no-torque group). Four of these techniques in a group of 100 eyes from 100
30 eyes required a second suture adjustment (2 patients undergoing PKP [27]. Each patient was
torque group, 2 no-torque group). One eye in the randomly assigned a suture technique group,
torque group required a third suture adjustment. which consisted of: interrupted (10-0 nylon, 16
There were no significant differences in astigma- bites), single running (10-0 nylon, 16–24 bites),
tism following suture adjustments between the double running (two 10-0 sutures, 12–16 bites),
three groups, and this remained the case at 3 and double running anti-torque (two 10-0 sutures with
6 months follow-up [25]. cross-stitch sutures in the opposite direction), and
double running with 10-0 and 11-0 sutures. In the may choose a 1-1-1 technique. The author of this
three double-running suture groups, the first suture chapter favours using a 3-1-1 technique.
was placed at 90% depth and the second at 50–90% The 10-0 nylon suture would be passed through
depth. Patients with greater than 3 D of astigma- the donor cornea, then through the host corneal
tism post-operatively had suture adjustment. In the rim, leaving a short end of around 1–1.5 corneal
interrupted suture group, selective suture removal diameters over the donor button, and a longer end
was performed along the corresponding meridian peripherally. Holding the long end with a needle
at 2–3 months post-operatively. In the single-run- holder (or any other adequate holding instrument
ning group, the suture was adjusted immediately of the surgeon’s preference), the longer end would
post-operatively or at later follow-up visits. In the be looped around a tying forceps three times; then
three double running groups, complete removal of the short end would be picked up with the tip of
a single 10-0 nylon suture was performed at 2–3 the tying forceps, and the knot tightened to an
months post-operatively. At 12 months post-oper- adequate tension—enough to appose the graft and
atively, the mean keratometric astigmatism was host, but not too tight to induce high astigmatism
5.86 ± 1.87 D in the interrupted suture group, or loosen up neighbouring sutures. When the
5.34 ± 1.99 D in the single running suture group, desired tension is achieved, the long end can be
3.00 ± 1.38 D in the double 10-0 running suture pulled centripetally and held flat on the donor cor-
group, 2.89 ± 1.11 D in the double 10-0 running nea. Then, the shorter end is pulled towards the
anti-torque suture group and 3.06 ± 1.22 D in the long end (centripetally as well), locking the knot
double running 10-0 and 11-0 sutures. In terms of in position. Then, a one loop knot is performed,
statistical significance of these differences, there with care not to loosen up the initial knot, aiming
was no difference between the interrupted suture for this second knot to be in line with the first one.
group and single-running suture group; however, The process is repeated with an extra loop. The
the difference between each of these two groups 1-1 knots need to be tied with “opposite” orienta-
and the three double-running suture groups was tion (clockwise/counterclockwise), otherwise if
statistically significant. There were no statistically all the 3-1-1 knots are tied with the same orienta-
significant differences between the three double- tion of the loop, it will end up being a sliding knot
running suture groups. In terms of complications, (i.e. loosening up with time).
no endophthalmitis or graft ulceration occurred in The technique is largely similar for a 2-1-1 or
any group, however in terms of wound leak, this a 2-1-1-1 technique. However it is different for a
occurred in 10% in the interrupted suture group, 1-1-1 technique. With this technique, a 1 loop
5.3% in the single-running suture group and 0% in knot is not possible to lock after the first knot,
all double-running suture groups [27]. but rather after the second knot. Care needs to be
taken that the second knot is aligned with the
first knot, so as it is tightened, it is not locked
ying, Locking and Burying
T before reaching the desired tension. This is usu-
the Knot ally less controlled than any of the other tech-
niques and the only theoretical advantage is
Whichever technique is chosen by the surgeon, easier burial of the knot, which, if the technique
the suture knot would need to be locked (to is good, is easy to do with any of the other meth-
avoid loosening of tension) then buried within ods as well.
the corneal tissue (ideally avoiding burying the
knot in the graft-host junction). Once again, dif-
ferent techniques can be chosen for this Wound Apposition
purpose.
The two most commonly used techniques Appropriate wound apposition is of paramount
would be either a 3-1-1 knot or a 2-1-1-1 knot. importance to avoid early postoperative leaks,
Eventually, a 2-1-1 can be used. Some surgeons loose sutures, or steps between graft and host
junction that may lead to delayed graft-host Postoperative Suture Adjustment

wound dehiscence (sometimes even at the time of
removal of graft sutures, which would require re- For interrupted suturing, this can be rather sim-
suturing and a delayed healing process) [28]. ple, as the only way to do an adjustment would be
A suture passage at approx. 90% depth in both via selective or complete suture removal [29]. A
graft and host would usually ensure that no steps corneal suture can eventually stay in place for an
occur on the corneal surface. This however may unlimited period of time, as long as the suture is
be more difficult to achieve when uneven thick- not loose or broken (which would represent a risk
nesses between graft and host are present (such of infection and/or corneal neovascularisation).
as that typically seen in patients with keratoconus Sutures that become loose or broken in the early
or corneal melting/thinning) [21]. As a rule of postoperative period may (or likely) need to be
thumb, in a standard case, a 90% suture depth replaced. Sutures that break or become loose
passage and a relatively longer bite would be rec- after several months can simply be removed.
ommended, to ensure a greater compression area, From a refractive perspective, selective suture
and reduce the chances of gaps in the interface at removal can be performed after 6–12 months.
any level. Time varies with the age of the patient and the
The knot would be tightened to appose both indication for graft. As an example, younger ker-
wound lips and not to have a loose stitch. atoconic patients would have a quicker recovery
Excessive tightening is to be avoided as very tight time compared to elder patients with neuro-
sutures induce high astigmatism, and in some trophic corneas. As such, a young keratoconic
cases, induce suture-related cheese wiring. patient could be considered for suture removal 6
months after surgery, whereas an elderly patient
with neurotrophic cornea may need the sutures in
Intraoperative Suture Adjustment place for 18 months or longer. The interrupted
suture can be removed corresponding to the steep
Intraoperative suture adjustment differs depend- axis as measured by corneal topography. Hence,
ing on the suturing technique used (interrupted by removing sutures along the steep axis, this
sutures vs. running sutures). For interrupted would become flatter and by a coupling effect,
sutures, it has already been highlighted how these the flatter meridian would become steeper.
should be placed (four cardinal sutures, then four In a running suture, the options are either
oblique sutures etc.). As the initial four sutures removal or adjustment. In an adjustment, the
are placed, one should see a symmetrical distri- suture tension can be locally modified, ideally
bution of the donor graft over the trephination under keratoscopic control, until the surgeon
area. One should also see tension lines between obtains a rounder keratoscopic reflex [30]. With
consecutive bites, conforming a “diamond” suture removal, whether under the microscope or
square configuration. As we continue with the at the slit lamp, the recommendation would be to
next four sutures, then a rather homogenous cut the suture at many different points after the
“octagon” configuration should be seen. At this application of topical anaesthesia and topical
point, a keratoscope can be used every so often to povidone-iodine. This way, the passes of the
check for the roundness of the corneal kerato- suture running intrastromally at the time of
scopic mires, and adjust the sutures accord- removal would be significantly less than pulling a
ingly—even repeat those that may be too tight or full suture from the same point, hence reducing
too loose. the chances of promoting a corneal abscess.
For running sutures, intraoperative adjustment
should be performed before the knot is tied, with Take Home Notes
the help of an intraoperative keratoscope. When • Ocular surface disease (OSD) is the leading
the keratoscopic mires look round and regular, cause of graft failure and hence it is impera-
the knot can then be tied and buried. tive that pre-operative OSD is aggressively
treated. However, this needs to be maintained donor’s material. Report of new instruments. Am J
Ophthalmol. 1941;24:1–20.
post-operatively. 6. Farge EJ. Eye banking: 1944 to the present. Surv
• There are many variations to the surgical tech- Ophthalmol. 1989;33:260–3.
nique of penetrating keratoplasty; however, 7. Seitz B, Brunner H, Viestenz A, et al. Inverse
the main goals are to safely perform surgery mushroom-shaped nonmechanical penetrating kerato-
plasty using a femtosecond laser. Am J Ophthalmol.
without complications and achieve good 2005;139:941–4.
wound apposition with minimal astigmatism 8. Daniel MC, Bohringer D, Maier P, et al. Comparison
induction. of long-term outcomes of femtosecond laser-assisted
• For a DALK procedure, the same principles keratoplasty with conventional keratoplasty. Cornea.
2016;35:293–8.
would apply, bearing in mind that the suture 9. Tsubota K, Satake Y, Kaido M, et al. Treatment of
should go full-thickness through the DALK severe ocular-surface disorders with corneal epi-
donor button, and not 90% depth as for a thelial stem-cell transplantation. N Engl J Med.
PKP. 1999;340:1697–703.
10. Hossain P, Tourkmani AK, Kazakos D, et al.
• Nylon is the most effective suture material in Emergency corneal grafting in the UK: a 6-year anal-
keratoplasty surgery. Its relatively elastic, bio- ysis of the UK Transplant Registry. Br J Ophthalmol.
compatible and strength properties are supe- 2018;102:26–30.
rior to other materials for this purpose. The 11. Reinhard T, Bohringer D, Sundmacher R. Accelerated
chronic endothelial cell loss after penetrating kera-
10-0 size is commonly used which has a diam- toplasty in glaucoma eyes. J Glaucoma. 2001;10:
eter of approximately 20 μm. 446–51.
• Correct suturing is of paramount importance 12. Kwon YH, Taylor JM, Hong S, et al. Long-term
for an appropriate wound apposition, ensuring results of eyes with penetrating keratoplasty and
glaucoma drainage tube implant. Ophthalmology.
the chances of wound leaks, loose stitches or 2001;108:272–8.
any other graft-host junction problems are 13. Tourkmani AK, Sanchez-Huerta V, De Wit G, et al.
reduced to the minimum possible. Weighing of risk factors for penetrating keratoplasty
• The suture knot would need to be locked (to graft failure: application of Risk Score System. Int J
Ophthalmol. 2017;10:372–7.
avoid loosening of tension) and then buried 14. Bradley BA. Rejection and recipient age. Transpl
within the corneal tissue (ideally avoiding Immunol. 2002;10:125–32.
burying the knot in the graft-host junction). 15. Tourkmani AK, Ansari AS, Hossain PN, et al.
The two most commonly used techniques Tectonic Descemet stripping endothelial keratoplasty
for the management of corneal perforation: a case
would be either a 3-1-1 knot or a 2-1-1-1 series. Cornea. 2020;39:1571–5.
knot. 16. Bandivadekar P, Gupta S, Sharma N. Intraoperative
suprachoroidal hemorrhage after penetrating kera-
toplasty: case series and review of literature. Eye
Contact Lens. 2016;42:206–10.
17. Ingraham HJ, Donnenfeld ED, Perry HD. Massive
References suprachoroidal hemorrhage in penetrating kerato-
plasty. Am J Ophthalmol. 1989;108:670–5.
1. Zirm E. Eine erfolgreiche totale Keratoplastik. 18. Melles GR, Binder PS. A comparison of wound heal-
Graefes Arch Ophthalmol. 1906;64:580–93. ing in sutured and unsutured corneal wounds. Arch
2. Armitage WJ, Tullo AB, Larkin DF. The first suc- Ophthalmol. 1990;108:1460–9.
cessful full-thickness corneal transplant: a commen- 19. McNeill JI, Wessels IF. Adjustment of single con-
tary on Eduard Zirm’s landmark paper of 1906. Br J tinuous suture to control astigmatism after pen-
Ophthalmol. 2006;90:1222–3. etrating keratoplasty. Refract Corneal Surg. 1989;5:
3. Reisinger F. Die Keratoplastik, ein Versuch zur 216–23.
Erweiterung der Augenheilkunst. Bayerische 20. Holland E. A comparison of suture types in the stimu-
Annalen. 1824;1:207–15. lation of corneal inflammation. Invest Ophthalmol Vis
4. Marcus M. Angabe eines Operationsverfahren zur Sci Suppl. 1990;31(4):270.
Ausführung der Transplantatio Corneae. In: Schmidt 21. Busin M, Arffa RC. Deep suturing technique for pen-
CC, editor. Jahrbücher der in-und ausländischen gesa- etrating keratoplasty. Cornea. 2002;21:680–4.
mten Medicin. Leipzig: Otto Weigand; 1841. p. 89. 22. Shahinian L Jr, Brown SI. Postoperative complica-
5. Keratoplasty CR. Comments on the technique of tions with protruding monofilament nylon sutures.
corneal transplantation. Source and preservation of Am J Ophthalmol. 1977;83:546–8.
23. Javadi MA, Naderi M, Zare M, et al. Comparison 27. Nuzzi R, Burato C, Tridico F, et al. Advantages
of the effect of three suturing techniques on post- of double running sutures in astigmatism after
keratoplasty astigmatism in keratoconus. Cornea. penetrating keratoplasty. Clin Ophthalmol.
2006;25:1029–33. 2022;16:797–802.
24. Spadea L, Cifariello F, Bianco G, et al. Long-term 28. Abou-Jaoude ES, Brooks M, Katz DG, et al.
results of penetrating keratoplasty using a single or Spontaneous wound dehiscence after removal of
double running suture technique. Graefes Arch Clin single continuous penetrating keratoplasty suture.
Exp Ophthalmol. 2002;240:415–9. Ophthalmology. 2002;109:1291–6; discussion 1297.
25. Vajpayee RB, Sharma V, Sharma N, et al. Evaluation 29. Van Meter WS, Gussler JR, Soloman KD, et al.
of techniques of single continuous suturing in pen- Postkeratoplasty astigmatism control. Single con-
etrating keratoplasty. Br J Ophthalmol. 2001;85: tinuous suture adjustment versus selective interrupted
134–8. suture removal. Ophthalmology. 1991;98:177–83.
26. Sharma V, Sharma N, Vajpayee RB, et al. Study of 30. Solano JM, Hodge DO, Bourne WM. Keratometric
corneal topographic patterns with single continu- astigmatism after suture removal in penetrating kera-
ous suturing techniques in penetrating keratoplasty. toplasty: double running versus single running suture
Cornea. 2003;22:5–9. techniques. Cornea. 2003;22:716–20.
Closing the Wound: Can Sutures
Be Avoided?
7
Luca Menabuoni, Alessandra Balestrazzi,
Luca Buzzonetti, Romina Fasciani,
Claudio Macaluso, Luigi Mosca, Roberto Pini,
Giada Magni, Paolo Matteini, Fulvio Ratto,
Michele Rossi, and Francesca Rossi

• Description of the laser welding approach and
its recent history Suturing is a critical step in the surgical practice:
• Photothermal effects in laser welding of cor- it allows closuring an open wound and supports
neal tissue the recovery of the injured tissue. Nowadays, a
• Mechanical load resistance of laser welded variety of suturing techniques are available,
corneas in PK while the fundamental goals remain the same: an
• Laser welding in Deep Anterior Lamellar excellent adhesion of the wound walls, providing
Keratoplasty (DALK): a clinical case a strong mechanical response of the treated tis-
• The development of biocompatible patches to sue, prevention of inflammation and bacterial
be welded as adhesives in ocular surgery contamination, and a satisfactory functional and
cosmetic outcome.
In ocular surgery, suturing is usually performed
in corneal transplantation. It can be performed
Supplementary Information The online version con- using interrupted sutures, continuous running sin-
gle suture, double sutures or a combination of
org/10.1007/978-3-031-32408-6_7.
interrupted and continuous sutures [1, 2].
L. Menabuoni C. Macaluso
Casa di Cura Villa Donatello, Florence, Italy University of Parma, Parma, Italy
A. Balestrazzi R. Pini · G. Magni · P. Matteini · F. Ratto · M. Rossi
Cornea and Ocular Surface Unit, ASL ROMA 2, F. Rossi (*)
Rome, Italy Institute of Applied Physics, Italian National
Research Council, Florence, Italy
L. Buzzonetti
e-mail: [email protected]; [email protected];
Ophthalmology Department, Bambino Gesù IRCCS
[email protected]; [email protected];
Children’s Hospital, Rome, Italy
[email protected]
R. Fasciani · L. Mosca
Cornea and Refractive Surgery Unit, Department of
Opthalmology, Agostino Gemelli University
Polyclinic Foundation-IRCCS, Rome, Italy
e-mail: [email protected];
[email protected]
https://doi.org/10.1007/978-3-031-32408-6_7
118 L. Menabuoni et al.
Needles and stitches are also used in cataract Laser Closuring of Corneal Wounds
surgery and more in general in corneal wounds
management. Laser welding of biological tissues is a tech-
However, the suturing procedure is time-con- nique used to join tissues by inducing a photo-
suming, and the post-treatment results can be thermal effect between the two edges of the
affected by several side effects that might impair wound. It has been proposed in several surgical
the surgery outcome. The use of needle and of the specialities over the last 30 years: neurosurgery,
suturing thread induces a mechanical trauma in eye surgery, dermatology, aesthetic medicine,
the tissue. The presence of the foreign body at the dentistry, etc., are just some of the fields where
wound site can exacerbate or prolong the inflam- laser welding is applied [7]. Laser welding hold
matory phase. The emergence of infection, suture the promise to provide instantaneous, watertight
erosion, and granuloma are associated with fur- seals without the introduction of foreign materi-
ther problems [3–5]. The poor apposition of the als. For this reason, laser welding has progres-
donor and recipient flaps due to sutures can result sively gained relevance in the clinical setting,
in post-operative astigmatism [1]. making surgery less invasive, faster and less
These possible adverse effects related to cor- risky, and appearing as a valid alternative to tra-
neal transplantation may adversely affect the ditional surgical methods.
quality of the patient’s post-operative course. In In the laser welding approach, optimised by
addition, in susceptible subjects, clinical condi- our research team for corneal tissue, we proposed
tions could worsen to the point of requiring medi- the application of the photo-enhancing dye
cal care, representing an additional cost to the Indocyanine Green (ICG) in the tissue and the
national health service. use of a diode laser emitting at 810 nm. Most of
On this basis, for about 20 years we have been biological tissues, and corneal tissue in particu-
working on corneal laser welding, implementing lar, are transparent to the light in the near-infrared
this consolidated technique with the use of new region, while the stained tissue presents the opti-
materials characterised by biocompatibility and cal absorbance peak at this specific wavelength.
low production costs. Laser welding is a tech- In other words: only the presence of the biologi-
nique that exploits the selective interaction cal tissue stained with the ICG absorbs the radia-
between a specific wavelength and an absorber, tion emitted by the laser. This interaction triggers
resulting in a photothermal effect. This controlled a confined photothermal effect resulting in the
and localised temperature increase determines selective fusion of wound walls at low irradiation
the welding of the edges of biological tissue, power per target area, thus reducing the risk of
especially if it consists of thermosensitive mole- thermal damage to surrounding tissues. The
cules such as collagen. welding effect may be modulated in the depth of
Different solutions were proposed to over- the transparent tissue, thus resulting in a more
come all these problems: the use of the femtosec- effective closure of the wound.
ond laser for corneal tissue cutting can support Laser closuring of ocular tissues can be per-
the introduction of self-closuring wounds or the formed following different original approach,
reduction of suturing need (e.g. in cataract sur- that has been studied and well-described over the
gery or intrastromal corneal surgery) [2]. The use past 20 years [7–23].
of biocompatible adhesives or patches can sup- In the literature, the main absorption is due to
port the realisation of a suture-less transplanta- the water content of the cornea: this means that
tion or the conjunctiva treatment [6]. the photothermal effect is not spatially confined
In this chapter, we describe the proposed tech- and mainly located in the external corneal sur-
nique for a full laser-assisted corneal surgery and face, immediately impacting the laser light.
the biocompatible patches that can be used for Moreover, to the best of our knowledge, only one
the closuring or the local medication of ocular of the designed protocols reached the test phase
tissues. in animal models in vivo [19], and our original
7 Closing the Wound: Can Sutures Be Avoided? 119
approach has been tested in the clinical practice results and lowering the occurrence of
in selected patients [12, 15, 18]: both are based astigmatism.
on the combined use of the exogenous dye (typi- Another important aspect that the combina-
cally ICG) and the 810 nm laser. tion of the femtosecond laser with laser welding
improves, is the possibility to design patient-
tailored profiles, with enhanced mechanical
Photothermal Laser Welding resistance [26, 27]. Several donor/host-profiled
in Keratoplasty configurations have been described in the litera-
ture, such as zigzag, Christmas tree, mushroom,
During the last 20 years of research activity, at top hat [28, 29] or anvil ones [17, 18]. Each pro-
the Institute of Applied Physics, we developed an file choice is due to patient’s anatomy and
original photothermal approach to the laser weld- pathology and to the surgeon’s experience.
ing of corneal tissue, thus supporting the closur- During clinical practice, it has been shown that
ing of surgical wounds in corneal transplantation the anvil cut shape leads to improved sealing: it
[7, 8, 12, 13, 15, 16, 24, 25]. has a larger surface of the donor/recipient
It is based on the use of a near-infrared diode wound edge, and an observed mechanical stabil-
laser emitting at 810 nm and the topical appli- ity [30].
cation of the chromophore ICG, which shows Theoretical studies [30, 31] and clinical
high optical absorption at the laser wavelength observations [18, 24] pointed out that the femto-
emission. The protocol is as follows: (1) ICG is second laser cut profiles support a stronger
prepared in the form of a saturated aqueous mechanical load resistance to IOP. The qualita-
solution of commercially available Indocyanine tively theoretical studies showed that the over-
Green for biomedical applications (e.g. lap in the zigzag, Christmas tree, top hat,
IC-GREEN Akorn, Buffalo Grove, IL or ICG- mushroom and anvil patterns led to higher pres-
Pulsion Medical Systems AG, Germany). (2) sure resistance as compared with non-profiled
The solution is accurately positioned within the femtosecond laser-assisted trephination. The
surgical wound walls, staining the stromal tis- laser profiled flaps are stronger to mechanical
sue in depth. (3) The wound edges are approxi- stress and strain due to increasing internal pres-
mated, and laser welding (8 W/cm2 power sure in respect to the traditional straight cut.
density) is performed under a surgical micro- Moreover, the femtosecond laser-assisted anvil
scope [18]. wound configuration was found to be the most
The laser used in pre-clinical tests and in clini- stable graft configuration. Thanks to the periph-
cal applications is typically an AlGaAs diode eral lamellar ring, in the anvil profiled PK, the
laser (e.g. Mod. WELD 800 by El.En. SpA, Italy) donor cornea easily fit into the recipient bed: the
emitting at 810 nm and equipped with a fiberop- angled edge created a key-hole effect between
tic delivery system. the donor and the host tissues and increased the
This approach is of particular interest when surface area of junction, as the clinical practice
combined with femtosecond laser approach in a evidenced (Fig. 7.1).
fully laser-based transplantation: laser cutting Thanks to the laser welding procedure, the
enables to realise a clear-cut plane, with the donor flap presents very good mechanical stabil-
same geometrical and morphological profile ity not only in the immediate post-surgery times
both in the donor and in the recipient tissues. but even 1 year after the treatment [14]. Good
By doing this, a perfect matching of the wound mechanical stability after the treatment was clini-
walls is obtained, thus reducing the occurrence cally observed during a re-surgery after corneal
of postoperative astigmatism or of reduced rejection. It is well known that the photothermal
visual acuity [2]. The use of the laser welding effect underlying the laser welding of the wound
approach supports the reduction of stitches, walls is based on the controlled thermal denatur-
thus further improving the post-operative ation of interfibrillar links [32–38]. During irra-
Fig. 7.1 Deformation (in mm) and stress evaluation (von maintains intact the connection at the donor/host inter-
Mises values in N/m2, in false colours) in the transplanted face, while at the same IOP values, the zigzag profile is
cornea in anvil (left) mushroom (centre) zigzag profile losing contact (see arrow)
(right). At high internal pressure load, the anvil profile
a b
Fig. 7.2 TEM micrograph of a stromal region 50 μm far from the weld site (a). The appearance of a stromal region
close to the welded area (b). Scale bar 500 nm
diation, the temperature in the stained tissue is at lagen from the two sides of the walls is
around 55 °C, thus preserving the morphology of interwoven, and thus, the wound is sealed.
the single collagen fibre. Its regular spatial distri- The clinical observation in re-surgery was a
bution however is lost, as the interfibrillar bridges clear resistance of the transplanted flap at the weld
are denaturised at these temperatures. New site (Fig. 7.2), very close to the mechanical resis-
bridges are realised immediately after irradiation, tance of a natural and closed cornea, while re-cut-
during the cooling phase. The result is an imme- ting a transplanted flap is usually an easy-to-perform
diate closure of the wound walls, where the col- procedure in sutured donor flaps (Fig. 7.3).
a b c
Fig. 7.3 Porcine cornea in laser welding tests, modelling matoxylin and eosin) of intact cornea (a), 10× image of
the sealing of a surgical wound. Sections (10 μm thick- the laser welded cut (b); 40× image of the surgical cut (c)
ness) coloured with standard histological staining (hae-
a b
Fig. 7.4 Pre-operative slit-lamp image (a) and corneal map (b) of a patient’s cornea (right eye) selected for a suture-
less, laser-welded, DALK. The cornea presents a post-herpetic stromal scar in the central part
All-Laser Transplantation cornea. The pre-operative BCVA was 1/20, and

the intraocular pressure was 14 mmHg (Fig. 7.4).
The combined use of femtosecond laser and pho- The pachymetry revealed a thinnest point of
tothermal laser welding of the surgical wound 430 μm. Donor and recipient corneal flaps were
supports the all-laser corneal transplantation, trephined with a femtosecond laser. After laser
such as Deep Anterior Lamellar Keratoplasty trephination, the donor lamella was applied in the
(DALK) [25, 39]. recipient bed: no sutures were used to stabilise
In this particular approach, the first attempt to the donor cornea, while the laser welding proce-
perform an all-laser DALK procedure with the dure was selected and performed, as described in
support of laser welding was performed in “Photothermal Laser Welding in Keratoplasty”
Tuscany in 2006 [36]. In brief, this suture-less section (see Video 7.1). No adverse events or
surgery has been performed on a 55 years-old complications occurred during surgery or in the
patient with a dense scar in the central area of the post-operative period. A satisfactory intraopera-
tive wound apposition was observed under the onitoring the Thermal Effects
M
operating microscope at 360° before and after of Laser Welding
laser welding (Fig. 7.5). The central corneal was
clear with a slight interface, and the wound was The laser welding protocol is based on the induc-
well-apposed on the 360° in each examination tion of a controlled photothermal effect within
(Fig. 7.6). the corneal tissue. A specific thermal range has to
In this laser welding procedure, it was possi- be reached: below 50 °C, the procedure is inef-
ble to observe that, thanks to controlled heating fective, while over 70 °C undesired thermal dam-
of the graft–host junction, the structure of the ages can occur [40]. The induced temperature
collagen proteins denature and create a quick sta- values are related to the energy density, i.e. the
bilisation and healing of the wound at the graft– fluence, delivered to the tissue, and this is related
host junction. The good apposition was to the distance of the fibre tip from the target. The
maintained for all the follow-up, with an increas- welding effect is thus due to the surgeon ability to
ing CDVA during the 6 months period. This laser maintain the correct distance during the irradia-
welding technique can support the more common tion. When the correct temperature is induced,
suturing approach in DALK and PK or eventually the main clinical evidence is a whitening of the
totally substitute it, at least in the DALK cornea at the welded site [9, 41]. It is thus obvi-
procedure. ous the whole procedure, and its efficacy is
strongly surgeon-depended.
Over the last 10 years, the research activities
were thus devoted to design a new approach to
control and monitor the temperature effects dur-
ing the welding procedure. Several attempts can
be found in the literature [20, 23], almost based
on the NIR monitoring of the temperature at the
irradiation site. In our research group, we devel-
oped a robotic platform enabling to control and
standardise the irradiation conditions: the
movement of the fibre tip and its distance from
the target is controlled by a robotic arm [42–46].
This solution enables to overcome the problems
related to the manual procedure. The proposed
Fig. 7.5 Laser welding procedure in DALK. In the platform is based on the teleoperation of a robot
image, it is possible to see the stained wound walls (green controlled with visual feedback, where machine
arrow), the NIR laser tip (red arrow) and the red pointer vision and image processing are part of the con-
(inside the green circle), indicating the welding area
trol system. A collaborative robot arm was cho-
a b c
Fig. 7.6 Post-operative slit-lamp images of a laser welded DALK at 1 (a), 7 (b) and 30 (c) p.o. days
sen, with a control system that autonomously In the past, few attempts to close scleral
decides to stop the procedure in case of adverse wounds were performed [51], by the use of chito-
events: two cameras, one dedicated to the general san patches [52–54]. This was the starting point
vision and the other one devoted to the control of to push the research activities towards the design
the temperature, together with the general control and optimisation of new biocompatible materials
system based on a visual servoing continuously to support the healing of ocular tissues [55, 56].
adjust the trajectory. The system was tested on The working principle is as follows: the bio-
porcine corneas, with good welding results. material is casted or electro-spinned with the
staining dye, typically ICG or cyanines. The bio-
materials are designed in order to be biocompat-
aser Closuring with Biocompatible
L ible, biodegradable, to present a good handling,
Materials good chemical stability and good mechanical
load resistance when necessary.
Recent research activities are focused on the The 810 nm diode laser is used to seal the
design and optimisation of biocompatible patch onto the tissue: as said, the tissue is trans-
patches. These new materials can be used for the parent to this wavelength, while the stained patch
closuring of wounds [47, 48] and for local medi- directly absorbs the light. As a result, a photo-
cation by the controlled delivery of drugs in situ. thermal effect is induced: as the realised patches
The first successful ex vivo tests were per- are thin (thickness around 20–40 μm), the ther-
formed in our lab for the closuring of capsular mal effect is transferred towards the ocular tissue.
tissue [49, 50]. The approach is similar to the In the area of the spot, we thus observe a confined
photothermal welding described in the previous denaturation of the biological tissue together
paragraphs. However, here the fibre tip is posi- with the denaturation of the patch, thus leading to
tioned onto the flap to be secured, which is previ- the immediate sealing of the tissue and the patch.
ously stained with ICG (Fig. 7.7). The controlled photothermal effect can also be
used to enable the release of drugs encapsulated
within the patch [57].
Conclusions
The research activities offered a large variety of

different solutions to the problems that can occur
in the suturing procedure, in particular when ocu-
lar tissues are involved. The use of a NIR diode
laser, in combination with an absorbing dye used
to stain the biological tissue, is proposed to
realise a suture-less surgery, while providing an
immediate sealing of the surgical wounds and
improving the healing process. The combined
use of biocompatible materials is proposed for
the local treatment of ocular tissues or to close
extremely thin tissues.
Moreover, the laser welding procedure was
tested in combination with the laser cutting
Fig. 7.7 ICG-stained capsular patch laser welded by
diode laser radiation, to close a capsulorhexis. Laser spots
approach in corneal transplantation, demonstrat-
are clearly evident at the periphery of the patch. Tests are ing to be a flexible tool for performing patient-
performed ex vivo in porcine eyes tailored surgery.
Take Home Notes 8. Pini R, Rossi F, Menabuoni L, et al. Diode laser weld-
ing for cornea suturing: an experimental study of the
• Suturing is a challenging procedure in PK: it repair process. In: Manns F, Soderberg PG, Ho A, edi-
can induce several side effects in the post- tors. Ophthalmic technologies XIV; 2004. p. 245–52.
operative period, such as foreign body reac- 9. Rossi F, Pini R, Menabuoni L, et al. Experimental
tion, partial closuring of the wound walls, study on the healing process following laser welding
of the cornea. J Biomed Opt. 2005;10(2). https://doi.
uneven healing in the corneal perimeter lead- org/10.1117/1.1900703.
ing to astigmatism. 10. Pini R, Rossi F, Menabuoni L, et al. Preliminary study
• Laser welding of the surgical wound walls can on the closure of the lens capsule by laser welding -
be performed with a NIR diode laser (810 nm) art. no. 61381C. In: Manns F, Soderberg PG, Ho
A, editors. Ophthalmic technologies XVI; 2006. p.
and the preliminary staining of the tissues. C1381.
• A spatially confined and controlled thermal 11. Pini R, Rossi F, Menabuoni L. Laser welding of
effect is induced, thus resulting in an immedi- biological tissue: experimental studies in ophthal-
ate sealing and in a stable flap over the healing mology - art. no. 619103. In: Grzymala R, Haeberle
O, editors. Biophotonics and new therapy frontiers;
time. 2006. p. 19103.
• New materials can be used as adhesives: once 12. Menabuoni L, Pini R, Rossi F, et al. Laser-assisted
stained, they can be welded to the ocular tis- corneal welding in cataract surgery: retrospective
sues, thus improving local treatment of study. J Cataract Refract Surg. 2007;33(9):1608–12.
https://doi.org/10.1016/j.jcrs.2007.04.013.
wounds and pathological tissues. 13. Rossi F, Matteini P, Ratto F, et al. Laser tissue welding
in ophthalmic surgery. J Biophotonics. 2008;1(4):331–
42. https://doi.org/10.1002/jbio.200810028.
14. Rossi F, Matteini P, Pini R, et al. Long term obser-
References vation of low power diode laser welding after pen-
etrating keratoplasty in human patients. In: Manns F,
1. Krootila K, Wetterstrand O, Holopainen J. Post- Soderberg PG, Arthur HO, editors. Ophthalmic tech-
keratoplasty astigmatism. In: Hjortdal J, editor. nologies XX; 2010.
Corneal transplantation. Cham: Springer; 2016. 15. Buzzonetti L, Capozzi P, Petrocelli G, et al. Laser
https://doi.org/10.1007/978-3-319-24052-7_12. welding in penetrating keratoplasty and cataract sur-
2. Williams GP, Mehta JS. Technology: femtosecond gery in pediatric patients: early results. J Cataract
laser in keratoplasty. In: Hjortdal J, editor. Corneal Refract Surg. 2013;39(12):1829–34. https://doi.
transplantation. Cham: Springer; 2016. https://doi. org/10.1016/j.jcrs.2013.05.046.
org/10.1007/978-3-319-24052-7_15. 16. Rossi F, Menabuoni L, Malandrini A, et al. “All-
3. Christo CG, van Rooij J, Geerards AJ, Laser” endothelial corneal transplant in human
et al. Suture- related complications follow- patients. In: Manns F, Soderberg PG, Ho A, editors.
ing keratoplasty: a 5-year retrospective Ophthalmic technologies XXII; 2012.
study. Cornea. 2001;20(8):816–9. https://doi. 17. Menabuoni L, Canovetti A, Rossi F, et al. The ‘anvil’
org/10.1097/00003226-200111000-00008. profile in femtosecond laser-assisted penetrating
4. Bar-Sela SM, Spierer O, Spierer A. Suture-related keratoplasty. Acta Ophthalmol. 2013;91(6):e494–5.
complications after congenital cataract surgery: https://doi.org/10.1111/aos.12144.
Vicryl versus Mersilene sutures. J Cataract Refract 18. Canovetti A, Malandrini A, Lenzetti I, et al. Laser-
Surg. 2007;33(2):301–4. https://doi.org/10.1016/j. assisted penetrating keratoplasty: 1-year results in
jcrs.2006.10.039. patients using a laser-welded anvil-profiled graft.
5. Jeganathan SV, Ghosh S, Jhanji V, et al. Resuturing Am J Ophthalmol. 2014;158(4):664–70. https://doi.
following penetrating keratoplasty: a retrospective org/10.1016/j.ajo.2014.07.010.
analysis. Br J Ophthalmol. 2008;92(7):893–5. https:// 19. Rodriguez Galarza RM, McMullen RJ Jr. Descemet’s
doi.org/10.1136/bjo.2007.133421. membrane detachments, ruptures, and separations in
6. Liu J, Huang Y, Yang W, Sun X, Yingni X, Peng Y, ten adult horses: clinical signs, diagnostics, treatment
Song W, Yuan J, Li R. Sutureless transplantation using options, and preliminary results. Vet Ophthalmol.
a semi-interpenetrating polymer network bioadhesive 2020;23(4):611–23. https://doi.org/10.1111/
for ocular surface reconstruction. Acta Biomater. vop.12793.
2022;153:273–86. https://doi.org/10.1016/j. 20. Strassmann E, Livny E, Loya N, et al. CO2 laser
actbio.2022.09.049. welding of corneal cuts with albumin solder using
7. Pini R, Rossi F, Matteini P, et al. Laser tissue welding radiometric temperature control. Ophthalmic Res.
in minimally invasive surgery and microsurgery. In: 2013;50(3):174–9.
Pavesi L, Fauchet PM, editors. Biophotonics. Berlin: 21. Rasier R, Ozeren M, Artunay O, et al. Corneal tissue
Springer; 2008. p. 275–99. welding with infrared laser irradiation after clear cor-
neal incision. Cornea. 2010;29(9):985–90. https://doi. by second-harmonic generation imaging. In: Manns F,
org/10.1097/ICO.0b013e3181cc7a3e. Soderberg PG, Arthur HO, editors. Ophthalmic tech-
22. Savage HE, Halder RK, Kartazayeu U, et al. NIR nologies XX; 2010.
laser tissue welding of in vitro porcine cornea and 36. Matteini P, Cicchi R, Ratto F, et al. Thermal transitions
sclera tissue. Lasers Surg Med. 2004;35(4):293–303. of fibrillar collagen unveiled by second-harmonic
https://doi.org/10.1002/lsm.20094. generation microscopy of corneal stroma. Biophys
23. Barak A, Eyal O, Rosner M, et al. Temperature- J. 2012;103(6):1179–87. https://doi.org/10.1016/j.
controlled CO2 laser tissue welding of ocular tissues. bpj.2012.07.055.
Surv Ophthalmol. 1997;42 Suppl 1:S77–81. https:// 37. Rossi F, Pini R, Menabuoni L. Experimental
doi.org/10.1016/s0039-6257(97)80029-x. and model analysis on the temperature dynam-
24. Menabuoni L, Malandrini A, Canovetti A, Lenzetti ics during diode laser welding of the cornea. J
I, Pini R, Rossi F. The use of femtosecond laser Biomed Opt. 2007;12(1):014031. https://doi.
and corneal welding in the surgery of keratoco- org/10.1117/1.2437156.
nus. In: Alió J, editor. Keratoconus. Essentials in 38. Rossi F, Matteini P, Pini R, et al. Temperature control
ophthalmology. Cham: Springer; 2017. https://doi. during diode laser welding in a human cornea - art. no.
org/10.1007/978-3-319-43881-8_24. 663215. In: Vogel A, editor. Therapeutic laser appli-
25. Rossi F, Canovetti A, Malandrini A, et al. An “all-laser” cations and laser-tissue interaction III. Spie-Int Soc
endothelial transplant. J Vis Exp. 2015;(101):e52939. Optical Engineering: Bellingham; 2007. p. 63215.
https://doi.org/10.3791/52939. 39. Menabuoni L, Pini R, Fantozzi M, et al. “All–
26. Nuzzo V, Aptel F, Savoldelli M, et al. Histologic and laser” sutureless lamellar keratoplasty (ALSL–
ultrastructural characterization of corneal femtosec- LK): a first case report. Invest Ophthalmol Vis Sci.
ond laser trephination. Cornea. 2009;28(8):908–13. 2006;47(13):2356.
https://doi.org/10.1097/ICO.0b013e318197ebeb. 40. Niemz MH. Laser-tissue interactions. Berlin:
27. Chamberlain WD, Rush SW, Mathers WD, et al. Springer; 2007.
Comparison of femtosecond laser-assisted kerato- 41. Rossi F, Pini R. Modeling the temperature rise dur-
plasty versus conventional penetrating keratoplasty. ing diode laser welding of the cornea. In: Ophthalmic
Ophthalmology. 2011;118(3):486–91. https://doi. technologies XV, vol. 5688; 2005. p. 185–193. https://
org/10.1016/j.ophtha.2010.08.002. doi.org/10.1117/12.610814.
28. Busin M, Robert CA. Microkeratome-assisted mush- 42. Rossi F, Pini R, Menabuoni L, et al. Robotic con-
room keratoplasty with minimal endothelial replace- solle for ocular surgery: a preliminary study. In:
ment. Am J Ophthalmol. 2005;140(1):138–40. Ophthalmic technologies XXIV; 2014.
29. Farid M, Kim M, Steinert RF. Results of pen- 43. Menabuoni L, Malandrini A, Canovetti A, et al. Laser
etrating keratoplasty performed with a femtosecond assisted robotic surgery in keratoplasty. Investig
laser zigzag incision initial report. Ophthalmology. Ophthalmol Vis Sci. 2017;58(8):3.
2007;114(12):2208–12. https://doi.org/10.1016/j. 44. Rossi F, Micheletti F, Magni G, et al. A robotic
ophtha.2007.08.048. platform for laser welding of corneal tissue. In:
30. Canovetti A, Rossi F, Rossi M, et al. Anvil-profiled Novel biophotonics techniques and applications IV;
penetrating keratoplasty: load resistance evaluation. 2017.
Biomech Model Mechanobiol. 2019;18(2):319–25. 45. Rossi F, Micheletti F, Magni G, et al. Laser assisted
https://doi.org/10.1007/s10237-018-1083-y. robotic surgery in cornea transplantation. In: Design
31. Lee HP, Zhuang H. Biomechanical study on the edge and quality for biomedical technologies X; 2017.
shapes for penetrating keratoplasty. Comput Methods 46. Russo S, Petroni G, Quaglia C, et al. ESPRESSO: a
Biomech Biomed Engin. 2012;15(10):1071–9. https:// novel device for laser-assisted surgery of the anterior
doi.org/10.1080/10255842.2011.571677. eye segment. Minim Invasive Ther Allied Technol.
32. Matteini P, Rossi F, Menabuoni L, et al. Microscopic 2016;25(2):70–8. https://doi.org/10.3109/13645706.
characterization of collagen modifications induced by 2015.1092450. PMID: 26429150.
low-temperature diode-laser welding of corneal tis- 47. Bal-Ozturk A, Cecen B, Avci-Adali M, et al. Tissue
sue. Lasers Surg Med. 2007;39(7):597–604. https:// adhesives: from research to clinical translation. Nano
doi.org/10.1002/lsm.20532. Today. 2021;36:101049. https://doi.org/10.1016/j.
33. Matteini P, Ratto F, Rossi F, et al. Photothermally- nantod.2020.101049.
induced disordered patterns of corneal col- 48. Smeets R, Tauer N, Vollkommer T, et al. Tissue
lagen revealed by SHG imaging. Opt Exp. adhesives in reconstructive and aesthetic surgery-
2009;17(6):4868–78. application of silk fibroin-based biomaterials. Int J
34. Matteini P, Ratto F, Rossi F, et al. Investigation on Mol Sci. 2022;23(14):7687. https://doi.org/10.3390/
fibrous collagen modifications during corneal laser ijms23147687.
welding by second harmonic generation micros- 49. Pini R, Menabuoni L, Lenzetti I, et al.
copy. In: Manns F, Soderberg PG, Ho A, editors. 50. Pini R, Rossi F, Menabuoni L, et al. A new technique
Ophthalmic technologies XIX; 2009. for the closure of the lens capsule by laser welding.
35. Matteini P, Rossi F, Ratto F, et al. Quantitative analy- Ophthalmic Surg Lasers Imaging. 2008;39(3):260–1.
sis of thermally-induced alterations of corneal stroma https://doi.org/10.3928/15428877-20080501-12.
51. Rossi F, Matteini P, Menabuoni L, et al. Sutureless in suine dura mater and vocal folds. In: Biophotonics:
closure of scleral wounds in animal models by the use photonic solutions for better health care IV, vol. 9129;
of laser welded biocompatible patches. In: Manns F, 2014. https://doi.org/10.1117/12.2051487.
Soderberg PG, Ho A, editors. Ophthalmic technolo- 55. Milanesi A, Magni G, Centi S, et al. Optically acti-
gies XXI; 2011. vated and interrogated plasmonic hydrogels for
52. Rossi F, Matteini P, Esposito G, et al. In vivo experi- applications in wound healing. J Biophotonics.
mental study on laser welded ICG-loaded chitosan 2020;13(9):e202000135. https://doi.org/10.1002/
patches for vessel repair. In: Kollias N, Choi B, Zeng jbio.202000135.
H, editors. Photonic therapeutics and diagnostics VII; 56. Ratto F, Magni G, Aluigi A, et al. Cyanine-doped
2011. nanofiber mats for laser tissue bonding. Nanomaterials
53. Rossi F, Matteini P, Esposito G, et al. In vivo laser (Basel). 2022;12(9):1613. https://doi.org/10.3390/
assisted end-to-end anastomosis with ICG-infused nano12091613.
chitosan patches. In: Sroka R, Lilge LD, editors. 57. Matteini P, Ratto F, Rossi F, et al. Chitosan films doped
Medical laser applications and laser-tissue interac- with gold nanorods as laser-activatable hybrid bioad-
tions V; 2011. hesives. Adv Mater. 2010;22(38):4313–6. https://doi.
54. Rossi F, Matteini P, Ratto F, et al. Laser bonding with org/10.1002/adma.201002228.
ICG-infused chitosan patches: preliminary experiences
Femtosecond-Assisted Penetrating
Keratoplasty and Deep Anterior
8
Lamellar Keratoplasty
Alfredo Vega-Estrada and Jorge L. Alió
Key Points emtosecond Laser Action

F
• Femtosecond laser-assisted keratoplasty pro- on Corneal Tissue
vides highly reproducible corneal incisions.
• Incisions performed using the femtosecond Femtosecond laser dissection is based on the cre-
laser heal faster and stronger. This may have ation of extremely precise incisions in the corneal
the advantage of earlier suture removal. tissue by separating lamellar fibers within the
• A significant investment, together with a stroma. This effect is explained under the princi-
learning curve period, is necessary when ples of optical breakdown and laser photodisrup-
adopting femtosecond laser technology for tion. Achieving optical breakdown will produce
assisting keratoplasty procedures. two different effects that have in common tissue
• Femtosecond laser-assisted keratoplasty pro- ablation. The first one, called plasma-induced
cedure induces healing patterns in the periph- ablation, determines a physics process, and the
eral cornea that can be grade according to the second one is called photodisruption and
degree of induced scaring. describes the mechanical behavior of matter.
• The potential of femtosecond-assisted surgery Laser-induced optical breakdown (LIOB)
to improve wound healing and vision recovery plays a significant role in the mechanism of
has been observed in numerous animal and plasma-mediated photodisruption, the most
clinical studies. important interaction between femtosecond
pulses and tissues. LIOB requires a small focal
Supplementary Information The online version con- spot size to achieve the threshold Fluence
tains supplementary material available at https://doi. (energy/area) for plasma formation [1].
org/10.1007/978-3-031-32408-6_8.
This spatial resolution is determined by the
width of the free-electron distribution, which is
A. Vega-Estrada the function of focalized size and peak intensity
Vissum, Grupo Miranza, Alicante, Spain which at the same time is determined in the fem-
Universidad Miguel Hernández de Elche, tosecond laser systems by the numerical aperture
Alicante, Spain (NA). NA is defined as the sine of the angle of the
Hospital Virgen de los Lirios de Alcoy, marginal ray coming from the focal point, multi-
Alicante, Spain plied by the refractive index of the medium from
e-mail: [email protected] which the input beam comes (Fig. 8.1).
J. L. Alió (*) In femtosecond laser systems, NA represents
Vissum Miranza, Miguel Hernández University, one of the most important factors that determine
Alicante, Spain
e-mail: [email protected] laser precision. To achieve the larger NA possi-
https://doi.org/10.1007/978-3-031-32408-6_8
128 A. Vega-Estrada and J. L. Alió
Fig. 8.1 Numerical

aperture (NA). (Courtesy
of Jaypee Brothers
published in W
Femtosecond Laser W
Assisted Corneal Graft
Surgery. Editor. Jorge
L. Alio)
W
NA =
f
Fig. 8.2 Laser

interaction with the
corneal tissue. (Courtesy Ultrashort laser pulse
of Jaypee Brothers
Optical absorption
published in
Femtosecond Laser
Assisted Corneal Graft Scattering
Wavefront distortion
Surgery. Editor. Jorge
L. Alio)
Plasma generation
Nonlineal
selffocussing
Photodisruption
Plasma desfoccusing
ble, the energy delivered will condense in a Cavitation bubble size can vary depending on
decreased spot size [2]. laser energy and occupies a greater space volume
There are two ways that the systems use in than previous solid material, which induces a
order to increase the NA: one is to increase the lamellar tissue separation. The femtosecond laser
lens diameter, and the other, is to decrease the can then be programmed to align several millions
focal distance (Figs. 8.1 and 8.2). of cavitation bubbles to create larger tissue sepa-
Thus, the optical breakdown process consists ration aiming to compose several patterns of cor-
in focusing the photon energy beam at a specific neal tissue dissection. Femtosecond laser
depth into the corneal stroma, concentrating incisions are composed of a great number of
enough energy over the threshold level, which overlapped bubbles with a specific per-pulse
frees the chemical bonds, thus transforming the energy. This bubble overlap will create micron-
solid area of tissue covered by the photon beam sized tissue separation that, when placed all
into a vapor state, creating a cavitation gas bubble together, will be responsible for the tissue
(Fig. 8.3). dissection.
8 Femtosecond-Assisted Penetrating Keratoplasty and Deep Anterior Lamellar Keratoplasty 129
In the next section, we will describe the inci-

sion module used in the Intralase femtosecond
laser platform, which is the one in which the
authors have more experience.
Incision Design Software
IntraLase-enabled keratoplasty (IEK) module is

the incision design software in the Intralase fem-
tosecond laser system [4]. Corneal graft surgery
includes several techniques, such as penetrating
Fig. 8.3 Schematic view of cavitation bubble creation
keratoplasty (PK), deep anterior lamellar kerato-
within the corneal stroma. (Courtesy of Jaypee Brothers plasty (DALK), descemet’s stripping automated
published in Femtosecond Laser Assisted Corneal Graft keratopalsty (DSAEK), and others that can be
Surgery. Editor. Jorge L. Alio) used depending on the surgical technique. Tissue
dissection design within the IEK module is based
on the combination of different cutting elements
This way, tissue dissection will depend on cor- that, when overlapping each other, creates com-
neal tissue transparency, laser per-pulse energy, plex graft incision in order to give more precise
and spot separation. Selecting the appropriate dissection when performing keraoplasty proce-
balance when programming the femtosecond dures (Fig. 8.4).
laser to obtain high-quality incisions is impor- In the next section, we describe those cutting
tant. To optimize incision quality, femtosecond elements that are used by the IEK module:
energy pulses must be precisely positioned within
the corneal tissue to permit bubbles to interact osterior Side Cut
P
[3]. That is why femtosecond lasers use a certain Posterior side cut is a vertical cylindrical-shaped
type of contact patient interface along with a suc- incision with a certain diameter that starts from
tion ring that holds the eye in a fixed position dur- the base of the cornea at its posterior depth and
ing the entire treatment. extends up to meet the lamellar incision at the
In summary, femtosecond laser dissection is anterior depth point.
based on a complex physical process, including This posterior cylinder can be warped into a
the creation of millions of cavitation bubbles that, conic section by adjusting the side cut angle
when placed together, can separate the tissue and parameter, as shown in the figure typically used
therefore create a highly precise tissue in zigzag profiles.
dissection.
Lamellar Cut
Lamellar cut is a planar incision parallel to the
Femtosecond-Assisted surface of the cornea that extends from the outer
Keratoplasty: Profiles, Outcomes, diameter to the inner diameter, where it is
Advantages, and Disadvantages expected to meet the anterior side cut.
This forms a ring incision whenever the inner
Nowadays, there are many femtosecond laser diameter is present and becomes a full lamellar
platforms for dissecting corneal tissue. Each of incision with an inner diameter equal to zero.
these platforms uses independent algorithms and
software to create its own incision architecture nterior Side Cut
A
and dissection patterns in order to be used for Anterior side cut is the analog section to the pos-
performing keratoplasty procedures. terior side cut, a cylindrical-shaped incision that
Anterior side cut
Anterior side cut

Ring lamellar cut Ring lamellar cut
Posterior side cut
Posterior side cut

Full lamellar cut
Fig. 8.4 Different incision uses in the Intralase-enabled keratoplasty module. (Courtesy of Jaypee Brothers published
in Femtosecond Laser Assisted Corneal Graft Surgery. Editor. Jorge L. Alio)
extends from the posterior depth to the surface of

the cornea. Anterior side cut is also used for
astigmatic keratotomy to correct residual astig-
matism following corneal graft surgery.
Keratoplasty Incision Profiles
Combining these previously described incisions

produces a huge tool when designing the corneal Fig. 8.5 Tophat profile
graft, allowing the surgeon to perform tailored solu-
tions to optimize clinical results in each case.
Posterior side cut, anterior side cut, and lamellar cut and (3) a narrower anterior side cut. This type of
can be combined in a limitless number of complex architecture maximizes the posterior tissue to the
edge profile graft combinations [5] with a very high transplanted and may be better suited for posterior
level of precision in dimensions and centration, corneal disease such as endothelial defect, Fuchs
which is not possible with other techniques [6]. dystrophy, or bullous keratopathy (Fig. 8.5).
Several profiles have been extensively tested, Stepped incisions such as tophat were recog-
and it is up to the surgeon needs to adapt them to nized long time ago to provide better wound
the specific case scenario in the field of corneal healing responses [7, 8]. Thus, self-healing pro-
graft surgery. file grafts such as tophat provide a superior tissue
In the following lines we will describe the tis- apposition, prevent wound leakage, and can
sue dissection profiles that the authors usually potentially accelerate wound healing.
uses when performing corneal graft surgery. Original tophat incision uses 90° angulation
as standard settings, and can evolve into different
Tophat-Based Profile shapes by reducing the posterior side cut angle
The most instructional and well-known shape might such as square-zag which reduces the deepest
be the tophat profile, which is composed of three diameter which the objective of facilitating deep
single steps imitating that typical hat morphology: stromal dissection and direct visualization during
(1) a wider posterior side cut; (2) ring lamellar cut; deep lamellar surgery [9].
Mushroom-Based Profile Zigzag slanted incisions increase the wound

The inverted version of a tophat is known as a apposition area, avoiding vertical cuts and offer-
mushroom and is often used for anterior surface ing superior biomechanical properties, as well as
surgery, being composed of a narrower posterior stronger wound healing and earlier postoperative
side cut and a wider anterior side cut, both inter- period [5].
sected by the ring lamellar cut (Fig. 8.6).
The broader anterior section maximizes the
superficial stromal area of the transplanted, Femtosecond-Assisted Penetrating
which makes it more suitable for anterior lamel- Keratoplasty Outcomes
lar corneal graft surgeries such as those caused
by corneal burns or scarring, keratoconus or mar- For the conventional penetrating keratoplasty pro-
ginal pellucid degeneration [10]. cedure (PKP) indications like advanced Fuchs’
dystrophy, severe pseudophakic bullous keratopa-
Zigzag-Based Profile thy, extensive trauma, corneal amyloidosis, and
One of the more complex edge profile incisions is severe keratoconus, the femtosecond laser-assisted
the zigzag graft, which compared to the squared PKP (FSPKP) has been done with high efficacy
angled profiles (tophat, mushroom), presents and safety outcomes [11–13]. There are some
superior biomechanical, sealing, and coaptation authors that have proposed that FSPKP could be
properties [5]. Zigzag can be challenging in terms superior to PKP in case of repeated PKP and global
of graft configuration, which is the reason why it is trauma with corneal laceration [14, 15]. Over the
considered a second step in the starting user learn- past decade, several researchers have demonstrated
ing curve. Zigzag profile is composed of a slanted that FSPKP is highly safe and enables faster suture
anterior and posterior side cuts connected by the removal with better refractive and visual outcomes
ring lamellar cut. It can be adapted to maximize than conventional PKP [13, 16]. Farid et al. pub-
either anterior or posterior surface, which makes it lished a study comparing results between FSPKP
suitable to a wide variety of situations (Fig. 8.7). (zigzag pattern) and PKP in relation with refractive
and visual outcomes [17]. Significant differences
in the average postoperative astigmatism between
both groups after the first month (P = 0.013) and
the third one (P = 0.018) were reported. By the
third month, the average astigmatism was 3D in the
zigzag FSPKP group and 4.46 D in the conven-
tional PKP group. Additionally, significant differ-
ences in best spectacle-corrected visual acuity
(BSCVA) were achieved after the first (P = 0.0003)
and the third month (P = 0.006), with 81% of the
FSPKP zigzag group versus 45% of the conven-
Fig. 8.6 Mushroom profile tional group achieving BSCVA of more than or
equal to 20/40 by the third month (P = 0.03). In the
same line, Chamberlain et al. [13] and Gaster et al.
[18] report similar results in terms of postoperative
topographic astigmatism.
On the other hand, many authors have pub-
lished the higher stability of the FS tophat dissec-
tion pattern with higher wound leakage pressure
[12, 16, 18–21] together with other dissection
patterns that could provide less risk of endothe-
lial rejection and, therefore, less complications
Fig. 8.7 Zigzag profile [22, 23].
Optical coherence tomography analysis reveals was no significant difference in terms of visual
no wound dehiscence and excellent appearance of acuity and rejection rate [29].
graft host tissue apposition and confirms that The few limitations of the femtosecond laser
FSLAPKP offers benefits of rapid wound healing in PKP include difficulty to cut in peripheral cor-
and predictable wound geometry [24]. neal opacity and difficulty to achieve the desired
Contrast sensitivity function, quality of vision, planar cuts in eyes with significant anterior and
and higher orders aberration analysis were posterior surface irregularities like descemeto-
reported with favorable results using FSPKP [25, cele [30], which can be overcome by optimiza-
26]. In terms of the ultrastructure and microbio- tion and innovations of the femtosecond
logical studies, it has been reported that femto- technology. Additionally, in order to perform FS
second laser preserves the ultrastructure of the laser dissection, you need to use the laser plat-
disrupted corneal collagen fibers [27], prompts form in both the patient and tissue donor that also
an initial increase in keratocyte activation and needs to be dissected on an artificial anterior
dendritic cells that decrease over time [28], favors chamber which, in the end, increases the overall
corneal reinnervation as early as 1 month postop- surgical time. Finally, femtosecond laser plat-
eratively [28]. All these advantages support the forms are expensive to purchase and maintain,
safety and efficacy of femtosecond laser when representing a high cost and not suitable for all
performing penetrating keratoplasty procedures. ophthalmological departments.
Additionally, from the standpoint of mechani-
cal strength, femtosecond laser wound configura-
tions are biomechanically more stable and create Femtosecond-Assisted Deep
more surface area for healing than the conven- Anterior Lamellar Keratoplasty
tional corneal trephination penetrating kerato- Outcomes
plasty [29]. Additionally in theory less suture
tension is required to achieve a water-tight wound There are several factors that make it difficult to
in FSPK than in conventional PKP. In FSPK, carry out a proper analysis of the outcomes
laser cutting is precisely controlled, and the edge obtained after the deep anterior lamellar kerato-
profile is predictable and controllable. Further, in plasty (DALK) procedure. The different methods
FSPK, laser-produced radial alignment markers of baring the Descemet membrane (DM), the
guarantee the accurate placement of cardinal technologies used to perform the dissection of
sutures. With regard to manual trephination, tilt- the recipient and the donor tissue, and the differ-
ing trephination and subsequent cuts with scis- ent suturing techniques, among others, must be
sors could have an uneven or irregular edge considered as factors that will impact the results
profile [29]. obtained by different authors. Video 8.1 describes
Some authors believed that the application of the different steps in both the donor's and the
the femtosecond laser might stimulate the prolif- recipient's cornea during a FSDALK procedure.
eration of keratocytes, thereby enhancing the Nevertheless, we can get an insight into the
graft–bed connection and wound stability and outcomes after the DALK procedure by analyz-
enabling earlier suture removal [29]. ing the different scientific data that have been
In despite of the aforementioned results, in a published over the last few years.
recent review article that analyzed the published One of the main complications of keratoplasty
data since 2007 until 2019 comparing the out- procedures reported by several investigators is
comes of both FSPKP and conventional PKP, the significant amount of refractive error mainly
authors did not find significant results between related to the induction of irregular astigmatism
both techniques [29]. Specifically, the authors [31]. A theoretical advantage of FSDALK in
reported that 12 months after the procedure, there terms of postoperative corneal astigmatism is that
the same configuration of the incisions should that the anterior aspect of the side cut from mul-
lead to a more regular corneal surface in, there- tiplanar incisions might be open in those areas
fore, less astigmatism. Nevertheless, a recent where topographical astigmatism is higher and in
review of femtosecond DALK (FSDALK) found this way reduce the corneal cylinder [32].
that different authors report no significant differ- Regarding wound integrity, there are some
ence in postoperative astigmatism when compar- reports which claim that FSDALK will perform
ing FSDALK with conventional manual better in terms of stronger incisions. Laboratory
trephination procedure [32]. In the same line, our research demonstrates that FSDALK incision
group report a few years ago showed no statisti- will support as high as 800 mmHg pressure
cally significant difference in the corneal cylinder which is significantly above physiological pres-
when comparing FSDALK with the conventional sure [32]. The possibility of removing sutures
technique [33]. early in the postoperative period is another factor
In terms of visual acuity, some authors that that supports the fact that wound integrity will
have analyzed the effect of the residual stromal perform better than conventional PK [32].
bed after the DALK procedure have found that In terms of graft rejection and endothelial
there is a correlation between the tissue remain- cell loss, endothelial-mediated rejection is
ing after baring the DM and the vision that the unlikely to occur in DALK procedures as this
patient can achieve after the surgery [31]. Those tissue is not being replaced during the surgery.
authors have shown that the less the stromal In this line, a study comparing FSDALK and the
residual tissue remains over the DM, the better manual technique found that FSDALK has a
the outcomes in terms of visual acuity after the lower rejection rate when compared with the
procedure. Li et al. published, in 2016, a com- manual trepanation technique [32]. Another
parative study analyzing the outcomes of issue of interest is that the energy released by
FSDALK and vacuum trephine and reported that the FS laser when passing close to the Descemet
FSDALK performed better in terms of visual membrane may reduce the endothelial cell pop-
acuity at 6 and 12 months postoperative [32]. It ulation. Nevertheless, the few reports published
must be remarked that the FSDALK procedure in the scientific literature addressing this matter
did not involve Descemet membrane stromal bar- demonstrate no significant differences in the
ing in the abovementioned series [32]. On the endothelial cell count when comparing the
other hand, other authors have not found signifi- FSDALK with the conventional manual trepa-
cant differences when comparing FSDALK and nation [32].
conventional DALK in terms of visual outcomes. The few limitations of the femtosecond laser
That is the case of Shehadeh–Mashor and col. in DALK procedures include difficulty cutting in
that found better vision in the early postoperative peripheral corneal opacity and achieving the
in the FSDLAK group but not at 6 and 12 months desired planar cuts in eyes with significant ante-
after the surgery [32]. In the same way, Bleriot rior and posterior surface irregularities [11].
et al. did not observe significant differences in Additionally, in order to perform FS laser dissec-
terms of vision when comparing FSDALK and tion, you need to use the laser platform in both
manual technique [32]. Moreover, similar find- the patient and tissue donor that also needs to be
ings were reported by our research group when dissected on an artificial anterior chamber which,
comparing the visual acuity of patients that in the end, increases the overall surgical time.
underwent FSDALK and manual DALK [32]. Finally, femtosecond laser platforms are expen-
Other advantages that FSDALK will theoreti- sive to purchase and maintain, representing a
cally have in terms of refraction, specifically in high cost and not suitable for all ophthalmologi-
terms of managing postoperative astigmatism, is cal departments.
ealing Patterns in Femtosecond

H Earlier suture removal seems to support increased
Laser-Assisted Keratoplasty wound integrity [35].
The wound configurations in femtosecond-
The cutting process in high pulse energy assisted DALK may combine the mechanical and
femtolaser is driven by mechanical forces wound healing advantages found for stepped cor-
applied by the expanding bubbles, which dis- neal wounds in PKP with the advantages of
rupt the tissue at a larger radius than the plasma lamellar surgery [35].
created at the laser focus. While in low pulse It also offers the advantages of better donor-
energy femto-laser, spot separations smaller recipient fit with increased surface area contact,
than the spot sizes are used for overlapping which may accelerate wound healing and earlier
plasmas, which directly evaporate the tissue suture removal [36].
inside the plasma volume, effectively separat- Our group established a grading for the side
ing tissue without a need for secondary cut corneal healing as observed and registered
mechanical tearing [34]. photographically by slit lamp photography with
Shtein et al. conducted in vivo confocal micro- illumination at 45° light angle of incidence con-
scopic evaluation of corneal wound healing after cerning the slit lamp observation optics placed
femtosecond laser-assisted keratoplasty [28]. orthogonal to the corneal vertex as observed by
Their findings confirmed that keratocyte activa- the first Purkinje reflex [33]. The grading of the
tion increased significantly following surgery in scar was performed as follows: Grade 0 = trans-
both the central (p < 0.001) and peripheral parent scar, Grade 1 = faint healing opacity,
(p = 0.002) cornea. The level of activation then Grade 2 = evident healing opacity, Grade 3 = sig-
decreases over time, but it is unknown whether nificant opacity with some cosmetic imbalance,
activated keratocytes can completely return to and Grade 4 = highly significant opacity with
their original state or remain permanently bio- very significant cosmetic imbalance (Table 8.1).
chemically altered. They also observed an In a study made by our research group [33] for
increase in dendritic cells postoperatively which femtosecond laser-assisted mushroom configura-
decreases after 6 months as wound healing pro- tion DALK, 52% of cases show healing pattern
gresses, and they found a significant association grade 3 or 4.
between the degree of keratocyte activation and The reasons for the femtosecond-assisted
dendritic cells. DALK to show a more active wound healing
They also observed accelerated nerve regen- leading to leucomatous wound could be either
eration which may be an indicator of improved due to the larger area of contact between the
corneal healing after FLAK, as femtosecond donor and recipient tissues and/or to femtosec-
laser allow shaped trephination with better wound ond laser-related biological activation of the cor-
apposition and less collateral damage when com- neal tissues, which should be related to the level
pared to potential crush injury from the mechani- of energy used for the creation of the side cut.
cal trephine [28]. Another study compared Femtosecond-
Early laboratory studies demonstrated that assisted DALK to mechanical DALK in 20 pedi-
femtosecond laser-enabled keratoplasty using the atric patients and concluded that femtosecond
tophat configuration provided better wound cuts could improve the success rate of big-bubble
integrity compared with manual PKP [12, 20]. technique. By improving donor/recipient fit
Laboratory experiments have shown an increased through femtosecond-created side cuts, the post-
burst pressure indicating increased wound stabil- operative spherical equivalent is reduced, and
ity. This is difficult to test in human patients. healing is accelerated [10].
Table 8.1 Analysis of side cut corneal wound healing pattern

Grade 0
Transparent scar
Grade 1
Faint healing opacity
Grade 2
Evident healing opacity
Grade 3
Significant opacity with some cosmetic imbalance
Grade 4
Highly significant opacity with very significant cosmetic imbalance
Future Perspective and Conclusions • Excellent wound apposition between the cor-
neal donor and receptor can be achieved, pro-
In summary, we can say that femtosecond laser- viding more biomechanical and safer
assisted keratoplasty offers the surgeon the ability dissection patterns.
to create dissection patterns in the corneal tissue • Nowadays, there are no clear advantages in
that are otherwise not possible with manual trephi- terms of outcomes of final corneal astigma-
nation. Many of the scientific work that has been tism or BCVA when comparing femtosecond-
published in the scientific literature demonstrates assisted keratoplasty and conventional
that femtosecond laser-assisted keratoplasty can keratoplasty procedures.
provide multiplanar self-healing profiles, which • Nowadays, several ongoing trials are on the
provides better tissue coaptation and biomechani- way to develop more desirable cutting shapes
cal stability helping to optimize clinical results, in order to enhance wound alignment and
reduce the number of sutures and earlier suture attachment for performing suture-less
removal compared to the mechanical procedure. keratoplasty in the future and to use femtosec-
Nevertheless, some of the review articles that have ond technology also for DSAEK.
been recently published comparing FSPK and con-
ventional PKP do not report a clear advantage of
laser-assisted techniques. However, we must keep
in mind that most of these studies have a limited References
methodology, so it would be necessary to carry out
1. Roach WP, Thompson ET. Selected papers on ultra-
well-designed trials with a higher quality of scien- short laser pulse bioeffects. SPIE milestone series,
tific evidence in order to reach conclusive results. vol. 174; 2002.
The potential of femtosecond-assisted surgery 2. Slade S. Laser refractive cataract surgery, sci-
to improve wound healing and vision recovery has ence, medicine and industry. Wayne: Bryn Mawr
Communications LLC; 2012.
been observed in numerous animal and clinical 3. Binder PS, Sarayba M, Ignacio T, et al.
studies. As a result, more trials are underway to Characterization of submicrojoule femtosecond laser
develop more desirable cutting shapes to enhance corneal tissue dissection. J Cataract Refract Surg.
wound alignment and attachment for performing 2008;34:146–52.
4. Ortiz JG, Alió JL. Femtosecond laser corneal graft
suture-less keratoplasty in the future. In addition, applications with IntraLase technology. In: Alió JL,
in the way that our understanding of the physical editor. Femtosecond laser assisted keratoplasty, vol.
and mechanical forces involved in the process 3. p. 18; 30.
improves, we will be able to create custom shape 5. Farid M, Kim M, Steinert RF. Results of pen-
etrating keratoplasty performed with a femtosecond
dissections and increase biomechanical wound laser zigzag incision initial report. Ophthalmology.
stability that, in the future, will allow us to perform 2007;114:2208–12.
early suture removal and decrease the amount of 6. Sarayba MA, Maguen E, Salz J, et al. Femtosecond
astigmatism related to keratoplasty procedures. laser keratome creation of partial thickness donor cor-
neal buttons for lamellar keratoplasty. J Refract Surg.
2007;23:58–65.
Take Home Notes 7. Barraquer JI. Technique of penetrating keratoplasty.
• Femtosecond laser keratoplasty procedure is Am J Ophtalmol. 1950;33:6–17.
highly accepted among corneal surgeons 8. Busin M. A new lamellar wound configuration for
penetrating keratoplasty surgery. Arch Ophthalmol.
because it is highly reproducible and intuitive 2003;121:260–5.
technology that provides and guarantees an 9. Neuhann TF, Lege BB, Joergensen JS, et al.
excellent wound apposition. Today it is mostly Femtosecond laser profiled penetrating Keratoplasty
used for PKP and DALK procedures, while with the IntraLase FS60 IEK. Chicago: ASCRS;
2008.
the technique has limitations for its use in pos- 10. Price FW, Price MO. Femtosecond laser shaped pen-
terior lamellar techniques. etrating keratoplasty: one-year results utilizing a top-
• Using femtosecond laser during keratoplasty hat configuration. Am J Ophthalmol. 2008;145:210–4.
procedures might have the potential advan- 11. Alió JL, Soria F, Vega A, Abdou A. Femtosecond laser-
assisted penetrating keratoplasty with IntraLase technol-
tages of earlier suture removal. ogy (use of the different technologies, surgical practical
pearls, outcomes). In: Alió JL, editor. Femtosecond vision after femtosecond laser-assisted descemet
laser assisted keratoplasty, vol. 8. p. 58; 74. stripping endothelial keratoplasty and penetrating
12. Steinert RF, Ignacio TS, Sarayba MA. “Top-hat”- keratoplasty: a randomized, multicenter clinical trial.
shaped penetrating keratoplasty using the femtosec- Am J Ophthalmol. 2011;152(4):556–66.
ond laser. Am J Ophthalmol. 2007;143(4):689–91. 26. Chamberlain W, Omid N, Lin A, et al. Comparison
13. Chamberlain WD, Rush SW, Mathers WD, et al. of corneal surface higher-order aberrations after endo-
Comparison of femtosecond laser-assisted kerato- thelial keratoplasty, femtosecond laser-assisted kera-
plasty versus conventional penetrating keratoplasty. toplasty, and conventional penetrating keratoplasty.
Ophthalmology. 2011;118(3):486–91. Cornea. 2012;31(1):6–13.
14. Rush SW, Fraunfelder FW, Mathers WD, et al. 27. Nuzzo V, Aptel F, Savoldelli M, et al. Histologic and
Femtosecond laser-assisted keratoplasty in failed ultrastructural characterization of corneal femtosec-
penetrating keratoplasty and globe trauma. Cornea. ond laser trephination. Cornea. 2009;28(8):908–13.
2011;30(12):1358–62. 28. Shtein RM, Kelley KH, Musch DC, et al. In vivo
15. Graef S, Maier P, Boehringer D, et al. Femtosecond confocal microscopic evaluation of corneal wound
laserassisted repeat keratoplasty: a case series. healing after femtosecond laser-assisted keratoplasty.
Cornea. 2011;30(6):687–91. Ophthalmic Surg Lasers Imaging. 2012;43(3):205–13.
16. Bahar I, Kaiserman I, McAllum P, et al. Femtosecond 29. Wen-yan P, Zhi-ming T, Xiu-fen L, Shi-you
laserassisted penetrating keratoplasty: stability evalu- Z. Comparing the efficacy and safety of f emtosecond
ation of different wound configurations. Cornea. laser-assisted vs conventional penetrating kerato-
2008;27(2):209–11. plasty: a metaanalysis of comparative studies. Int
17. Farid M, Steinert RF, Gaster RN, et al. Comparison of Ophthalmol. 2021;41:2913–23.
penetrating keratoplasty performed with a femtosec- 30. Yoo SH, Hurmeric V. Femtosecond laser-assisted
ond laser zig-zag incision versus conventional blade keratoplasty. Am J Ophthalmol. 2011;151(2):189–91.
trephination. Ophthalmology. 2009;116(9):1638–43. 31. Alió JL, Vega A, Soria F, Abdou A. Femtosecond
18. Gaster RN, Dumitrascu O, Rabinowitz YS. Penetrating laser-assisted anterior lamellar keratoplasty with
keratoplasty using femtosecond laser-enabled kerato- IntraLase technology (use of the different technolo-
plasty with zig-zag incisions versus a mechanical tre- gies, surgical practical pearls, outcomes). In: Alió JL,
phine in patients with keratoconus. Br J Ophthalmol. editor. Femtosecond laser assisted keratoplasty, vol.
2012;96(9):1195–9. 9. p. 78; 86.
19. Ignacio TS, Nguyen TB, Chuck RS, et al. Top hat 32. Chamberlain WD. Femtosecond laser-assisted deep
wound configuration for penetrating keratoplasty anterior lamellar keratoplasty. Curr Opin Ophthalmol.
using the femtosecond laser: a laboratory model. 2019;30:256–63.
Cornea. 2006;25(3):336–40. 33. Alio JL, Abdelghany AA, Barraquer R, et al.
20. Maier P, Böhringer D, Birnbaum F, et al. Improved Femtosecond laser assisted deep anterior lamel-
wound stability of top-hat profiled femtosecond laser- lar keratoplasty outcomes and healing patterns
assisted penetrating keratoplasty in vitro. Cornea. compared to manual technique. Biomed Res Int.
2012;31(8):963–6. 2015;2015:397891.
21. Birnbaum F, Wiggermann A, Maier PC, et al. Clinical 34. Latz C, Asshauer T, Rathjen C, Mirshahi
results of 123 femtosecond laser-assisted penetrating A. Femtosecond-laser assisted surgery of the eye:
keratoplasties. Graefes Arch Clin Exp Ophthalmol. overview and impact of the low-energy concept.
2013;251(1):95–103. Micromachines (Basel). 2021;12(2):122. https://
22. Fung SS, Iovieno A, Shanmuganathan VA, et al. doi.org/10.3390/mi12020122. PMID: 33498878;
Femtosecond laser-assisted lock-and-key shaped PMCID: PMC7912418.
penetrating keratoplasty. Br J Ophthalmol. 35. Price FW Jr, Price MO, Grandin JC, Kwon R. Deep
2012;96(1):136–7. anterior lamellar keratoplasty with femtosecond-
23. Gilmer W, McLeod SD, Jeng BH. Laboratory laser zigzag incisions. J Cataract Refract Surg.
model of bursting pressures of femtosecond laser- 2009;35(5):804–8. https://doi.org/10.1016/j.
assisted penetrating keratoplasty wounds using jcrs.2009.01.011. Erratum in: J Cataract Refract Surg.
novel pattern designs. Br J Ophthalmol. 2012;96(9): 2009 Jul;35(7):1325. PMID: 19393877.
1273–4. 36. Jonas JB, Vossmerbaeumer U. Femtosecond laser
24. Heur M, Tang M, Yiu S, et al. Investigation of fem- penetrating keratoplasty with conical incisions and
tosecond laser-enabled keratoplasty wound geom- positional spikes. J Refract Surg. 2004;20(4):397.
etry using optical coherence tomography. Cornea. https://doi.org/10.3928/1081-5 97X-2 0040701-1 5.
2011;30(8):889–94. PMID: 15307404.
25. Cheng YY, van den Berg TJ, Schouten JS, Pels E,
Wijdh RJ, van Cleynenbreugel H, et al. Quality of
Femtosecond Laser-Assisted
Tuck-In Penetrating Keratoplasty
9
Jorge L. Alió del Barrio, Olena Al-Shymali,
and Jorge L. Alió

• Tuck-in lamellar keratoplasty provides struc-
tural support to those extremely weakened Keratoglobus (KTG) is a rare noninflammatory
corneas with diffuse and severe thinning up to corneoscleral ectasia [1]. It is a bilateral disorder
the periphery, such as keratoglobus. characterized by generalized corneal thinning
• However, in patients where hydrops or endo- especially peripherally [2], that results in the
thelial damage is present, such tuck-in lamel- globular protrusion of the cornea. Previously, it
lar keratoplasty technique is not possible. was thought-out equivalent to congenital glau-
• Femtosecond laser-assisted tuck-in PKP is a coma and megalocornea until it was described
1-step technique that allows us to approach by Verrey as an individual clinical entity in 1947
cases where extreme peripheral thinning and [3]. Afterward, this was backed by Cavara in
corneal endothelial damage are present. 1950 [4].
Although the exact etiology of KTG remains
unknown, it has been described in both acquired
and congenital forms [1]. The former has been
associated with vernal keratoconjunctivitis and
chronic marginal blepharitis due to frequent
eye rubbing, dysthyroid eye disease, and idio-
pathic orbital inflammation [5, 6]. Congenital
KTG has been linked to Leber’s congenital
amaurosis, blue sclera syndrome, osteogenesis
Supplementary Information The online version con- imperfecta, and connective tissue disorders like
tains supplementary material available at https://doi. Ehlers–Danlons syndrome, type IV in particu-
org/10.1007/978-3-031-32408-6_9. lar, Marfan syndrome and Rubinstein–Tayabi
syndrome [1, 7–9].
The extreme corneal thinning and protrusion
J. L. A. del Barrio · J. L. Alió (*)
Vissum Miranza, Miguel Hernández University, in patients with KTG lead to visual deterioration
Alicante, Spain due to extreme myopia, irregular astigmatism,
e-mail: [email protected] and scaring [10]. The condition can be compli-
O. Al-Shymali cated by acute hydrops causing pain [11] or by
Cornea, Cataract and Refractive Surgery Unit, corneal perforations occurring spontaneously or
Vissum (Miranza Group), Alicante, Spain following minimal trauma [1, 12].
https://doi.org/10.1007/978-3-031-32408-6_9
140 J. L. A. del Barrio et al.
Since KTG is a rare ectasia, diverse algo- Surgical Technique

rithms were proposed. Still, no agreement on a
standard method was reached. Using conserva- Preoperative Considerations
tive methods such as spectacles or contact lenses
to correct myopia, and astigmatism is not always An explicit examination of the cornea in a KTG
possible, especially in children, and sometimes patient should be done under the slit lamp.
are ineffective with a corneal perforation risk Moreover, a detailed pachymetric map must be
owing to trauma. The surgical approach consists made, especially of the corneal periphery. For
of various lamellar keratoplasty and epikerato- this technique, we advise a minimum thickness
plasty techniques that provide structural support of 300 microns peripherally in order to have a
to these extremely weakened corneas. However, higher chance of succeeding during this techni-
in patients where hydrops or endothelial damage cally demanding surgery. If no Descemet mem-
are present, a two-step procedure with a second brane tears and endothelial damage are present,
central corneal penetrating graft will be neces- the surgeon may opt for conventional lamellar
sary, which means two corneoscleral buttons per keratoplasty or a tuck-in lamellar keratoplasy
eye are required [2, 10]. Although conventional technique [2, 15, 16]. However, when in addition
penetrating keratoplasty (PKP) is the gold- to the peripheral thinning, we have a damaged
standard procedure for many corneal diseases, it Descemet membrane or endothelium, we have
is impracticable in KTG because of peripheral two options: the 2-step procedure [10] of a lamel-
thinning. Therefore, large limbus to limbus lar epikeratoplasty followed by a central corneal
grafts is used, which may cause limbal stem cells penetrating graft which takes a longer time and
and angle structure disruptions as well as needs two corneoscleral buttons per eye, or we go
increase the risk of neovascularization and graft for our described femtosecond laser-assisted
rejection [1, 13]. tuck-in PKP [14], a 1-step surgery that requires
Recently our team proposed a 1-step tech- only one corneoscleral button per eye.
nique for KTG cases with endothelial damage,
femtosecond laser-assisted tuck-in PKP, where
both extreme peripheral thinning and corneal Technique
endothelial damage exists [14]. On account of the
thinning disorders in KTG, the aim of our tech- onor Cornea Preparation
D
nique is to insure tectonic support and optimal In an artificial anterior chamber and assisted by
thickness to the extremely subtle peripheral cor- the iFS femtosecond laser (AMO Inc, Irvine,
nea, in addition to replacing the damaged central CA), an anterior side cut of 8.5 mm in diameter
part of the cornea. and 310 microns deep is performed, continued by
a 300 μm deep ring lamellar cut up to 9.5 mm of
diameter (8.4–9.5 mm). Then, following the ring
Indications lamellar dissection plane through the anterior
side cut, a crescent blade is used to complete the
Femtosecond laser-assisted tuck-in PKP is indi- lamellar dissection until the limbal area. Once
cated in patients with KTG that have both periph- this maneuver is finished, a final full-thickness
eral corneal thinning and endothelial damage due trephination with an 11 mm trephine from the
to hydrops or Descemet membrane tears. In this endothelial side is done. Eventually, a donor but-
way, we enhance the thickness of the peripheral ton that is 11 mm in total diameter with an
recipient stroma and simultaneously exchange 8.5 mm central full-thickness area and a 1.25 mm
the central damaged part of the KG cornea by a partial-thickness circumferential peripheral edge
full-thickness penetrating corneal graft. is obtained (Fig. 9.1a).
9 Femtosecond Laser-Assisted Tuck-In Penetrating Keratoplasty 141
a b The retained margin of the recipient

Donor button
cornea consisted of:
8.5 mm full thickness center

anterior portions
300 µm
lamellar pocket
1.25 mm partial thickness circumferential peripheral edge

posterior portions
c d
Suture passage Final result
Donor cornea
Tucked-In peripehral circumferential edge
Recipient cornea Recipient cornea: retained edge

Tucked-In peripheral edge Donor cornea
Fig. 9.1 Femtosecond laser-assisted Tuck-in PKP surgi- the anterior portion of the recipient’s stromal pocket; (d)
cal diagram: (a) final appearance of the donor cornea; (b) the peripheral circumferential donor rim remains tucked
final appearance of the peripheral recipient stromal into the recipient’s stromal pocket, enhancing the periph-
pocket; (c) the thread passes through the central 8.5 mm eral corneal thickness
portion of the donor cornea with partial-thickness bite and
ecipient Cornea Preparation

R recipient cornea is excised with the assistance of
Once the preparation of the donor cornea is fin- Vannas scissors. Subsequently, a recipient periph-
ished, the recipient cornea is prepared using the eral stromal pocket made of an anterior and poste-
same pattern as the femtosecond laser, performing rior portion is obtained (Fig. 9.1b).
an anterior side cut of 8.5 mm in diameter at 160
microns depth and a ring lamellar cut depth of 150 Donor-Recipient Apposition
μm with 9.5 mm in diameter (recipient side cut The donor cornea is then sutured using 10/0 nylon
depth should be estimated as half of the mean pre- interrupted sutures. The thread is passed through
operative peripheral recipient corneal thickness). the donor 8.5 mm full-thickness button in a par-
Similarly to the donor cornea, a crescent blade is tial-thickness manner (up to the level where the
used to complete the peripheral lamellar dissec- peripheral rim starts) and the anterior portion of
tion following the previously dissected ring lamel- the recipient stromal pocket as close as it could be
lar plane, creating a lamellar pocket. The to the limbus (Fig. 9.1c). Similarly, all 16 sutures
dissection is aimed at the most corneal periphery are placed on the whole cornea. Then, in order to
by reaching the limbus. After the injection of vis- make sure that the peripheral donor edge remains
coelastic into the anterior chamber (AC) through within the peripheral recipient stromal pocket,
a paracentesis, the AC is entered using a 15-degree this peripheral donor edge is unfolded and tucked
sharp blade following the contour of the 8.5 mm in the stromal recipient pocket with the assistance
diameter anterior side cut. Further, the central of a blunt spatula (Fig. 9.1d) (Video 9.1).
Postoperative Management tion. Probably, the most challenging step is the

manual expansion of the lamellar plane with the
Postoperative care does not differ from the one crescent blade on the recipient cornea, along with
used for a standard PKP surgery: topical antibi- the creation of the lamellar recipient pocket. In
otic four times a day for 1 week and topical ste- this step, it is critical to keep the correct depth,
roids every 2 h for 1 week and four times a day but as we work inside a thin peripheral cornea,
thereafter. inadvertent tears of the anterior lip of the cornea
may occur, requiring extra suturing to reconstruct
the tear and ensure the correct positioning of the
Complications donor cornea.
The experience with this technique is still lim-

ited, and the only postoperative complications Outcomes
that we have observed were an Urrets-Zavalia
syndrome due to retained viscoelastic and suture- As already discussed, published evidence with
related infectious keratitis. However, intra and this technique is still short, and it is limited to a
postoperative potential complications with this publication of our group describing the outcome
Femto Tuck-In PKP technique should be equiva- obtained from both eyes of the same KG patient
lent to those described for standard (Fig. 9.2) [14]. This patient experienced a great
PKP. Especially important for this technique is bilateral visual improvement (corrected distance
the fact of performing an adequate preoperative vision of 20/50 in both eyes), with a residual
plan for the femtosecond laser dissection planes refractive cylinder of five diopters. Corneal
(diameters and depths) in relation to the periph- topography showed high but regular astigmatism,
eral preoperative pachymetry in order to avoid and the anterior segment optical coherence
too superficial or too deep surgical planes with tomography (OCT) adequately restored the
subsequent risk of anterior or posterior perfora- peripheral corneal thickness (Fig. 9.3).
a b c
Fig. 9.2 Preoperative external anterior segment photo- Slit lamp postoperative anterior segment picture 6 months
graph after intracameral air injection of a patient with after the transplant (c). Observe the complete transpar-
keratoglobus and chronic hydrops in relation to a severe ency recovery and the peripheral partial-thickness donor
tear of the corneal endothelium (a); Intraoperative photo- rim tucked into the host cornea
graph immediately after completing the Tuck-in PKP (b);
9 Femtosecond Laser-Assisted Tuck-In Penetrating Keratoplasty 143
a
b
Fig. 9.3 Anterior segment optical coherence tomography white arrows). In those areas not covered by the peripheral
(OCT) pictures (a, b). The peripheral donor partial- donor rim, the thickness remains low (a)
thickness rim is tucked into the host stromal pocket ((b),
Conclusions and it is more accurate than a free-hand manual

technique with surgical blades. Also, the short
When severe keratoglobus associates with previous ring lamellar cut assists the depth of the peripheral
corneal hydrops and severe endothelial damage, dissection that is performed up to the limbal area.
routine PKP with an almost limbus to limbus diam-
eter graft associates with a poor visual outcome due Take Home Notes
to a high risk of rejection and secondary glaucoma • Femtosecond laser-assisted tuck-in PKP
[13]. Such cases were previously managed with a ensures tectonic support and optimal thick-
tuck-in or epikeratoplasty lamellar technique to ness to the thinned peripheral cornea, in addi-
first reinforce the peripheral corneal thickness, fol- tion to replace the damaged endothelium.
lowed by a PKP several months later for visual • The addition of the femtosecond laser to assist
rehabilitation [10]. However, this approach requires the host and donor cornea partial-thickness
two surgical procedures and the use of two donor trephinations is not essential, but it reduces the
corneas. On the other hand, femtosecond laser- risk of a full-thickness cut that may happen
assisted tuck-in PKP offers an alternative option to with manual suction trephines in these
rehabilitate these patients but with a single surgical extremely thin corneas, and it is more accurate
procedure and only one donor cornea. than a free-hand manual technique with surgi-
Tuck-in PKP in keratoglobus eyes with previ- cal blades.
ous corneal hydrops combines the advantages of
PKP (the central full-thickness graft rehabilitates
the stroma and the endothelium) and lamellar Conflict of Interest None of the authors have any con-
tuck-in techniques (the peripheral partial- flict of interest to disclose.
thickness rim tucked into the intrastromal host
pocket integrates perfectly and provides tectonic
support to the recipient tissue). Simultaneously, References
damage to the recipient’s limbal stem cells is
avoided as the limbal region is not manipulated. 1. Wallang B, Das S. Keratoglobus. Eye.
The addition of the femtosecond laser to assist 2013;27:1004–12.
the host and donor cornea partial-thickness treph- 2. Javadi MA, Kanavi MR, Ahmadi M, Yazdani
S. Outcomes of epikeratoplasty for advanced kerato-
inations is not essential, but it reduces the risk of globus. Cornea. 2007;26:154–7.
a full-thickness cut that may happen with manual 3. Verrey F. Keratoglobe aigu. Ophthalmologica.
suction trephines in these extremely thin corneas, 1947;114:284–8.
4. Cavara V. Keratoglobus and keratoconus. A contributors and outcomes. Trans Am Ophthalmol Soc.
tion to the nosological interpretation of keratoglobus. 1999;97:187–98.
Br J Ophthalmol. 1950;34(10):621–6. 12. Baillif S, Garweg J, Grange J, Burillon C, Kodjikian
5. Cameron J, Al-Rajhi A, Badr I. Corneal ectasia L. Keratoglobus: review of the literature. J Fr
in vernal keratoconjunctivitis. Ophthalmology. Ophthalmol. 2005;28:1145–9.
1989;96:1615–23. 13. Cowden J, Copeland R, Schneider M. Large diam-
6. Jacobs D, Green W, Maumenee A. Acquired kerato- eter therapeutic penetrating keratoplasties. Refract
globus. Am J Ophthalmol. 1974;77:393–9. Corneal Surg. 1989;5:244–8.
7. Koenekoop R. An overview of Leber congenital 14. Alió Del Barrio J, Al-Shymali O, Alió J. Femtosecond
amaurosis: a model to understand human retinal laser-assisted tuck-in penetrating keratoplasty for
development. Surv Ophthalmol. 2004;49:379–98. advanced keratoglobus with endothelial damage.
8. Cameron J. Corneal abnormalities in Ehlers-Danlos Cornea. 2017;36(9):1145–9.
syndrome type VI. Cornea. 1993;12:54–9. 15. Kaushal S, Jhanji V, Sharma N, Tandon R, Titiyal
9. Nelson M, Talbot J. Keratoglobus in Rubinstein Taybi JS, Vajpayee RB. “ Tuck In” lamellar keratoplasty
syndrome. Br J Ophthalmol. 1989;73:385–7. (TILK) for corneal ectasias involving corneal periph-
10. Jones DH, Kirkness CM. A new surgical technique ery. Br J Ophthalmol. 2008;92:286–90.
for keratoglobus-tectonic lamellar keratoplasty fol- 16. Vajpayee RB, Bhartiya P, Sharma N. Central lamel-
lowed by secondary penetrating keratoplasty. Cornea. lar keratoplasty with peripheral intralamellar tuck:
2001;20:885–7. a new surgical technique for keratoglobus. Cornea.
11. Grewal S, Laibson P, Cohen E, Rapuano C. Acute 2002;21:657–60.
hydrops in the corneal ectasias: associated fac-
Rescuing Failed Penetrating
Keratoplasty Grafts
10
Jorge L. Alió del Barrio, Scott Robbie, Marcus Ang,
Andrea Montesel, and Jorge L. Alió

• Corneal transplantation is the world's most
frequent type of human tissue graft but still Corneal transplantation is the world's most fre-
remains a critical shortage of corneal graft quent type of human tissue graft, with more than
tissue. 180,000 corneal grafts performed every year
• The global shortage of corneal tissue is likely worldwide [1]. Corneal transplantation aims to
to grow as the cumulative number of trans- restore visual function when it is severely and
planted patients increases over time, making irreversibly impaired by end-stage corneal dis-
corneal graft failure one of the most common ease after conservative surgical or medical treat-
indications for corneal transplantation today. ment options have failed. Corneal transplantation
• PK is one of the commonest forms of corneal techniques have evolved considerably over the
transplantation and remains the procedure of last two decades, specifically with regard to
choice for selected indications. lamellar techniques and the optimization and
• Options for managing a failed PK depend on standardization of endothelial keratoplasty tech-
the clinical context, the failure mechanism, niques in particular; however, there remains a
and any comorbidity. critical shortage of corneal graft tissue, and it has
been estimated that there is only 1 corneal donor
available for every 70 needed [1]. The global
J. L. A. del Barrio (*) · J. L. Alió
shortage of corneal tissue is likely to grow as the
Vissum Miranza, Miguel Hernández University, cumulative number of transplanted patients
Alicante, Spain increases over time, making corneal graft failure
S. Robbie
Cornea Unit, Guy’s and St Thomas’ Hospital,
London, UK
M. Ang
Singapore Eye Research Institute, Singapore National
Eye Centre, Singapore, Singapore A. Montesel
Cornea, Cataract and Refractive Surgery Unit,
Department of Ophthalmology and Visual Science,
Vissum (Miranza Group), Alicante, Spain
Duke-NUS Graduate Medical School, Singapore,
Singapore Jules Gonin Eye Hospital, University of Lausanne,
e-mail: [email protected] Lausanne, Switzerland
https://doi.org/10.1007/978-3-031-32408-6_10
one of the most common indications for corneal The first question one should pose when con-
transplantation today [2]. sidering a repeat PK is, as always, whether sur-
Corneal graft failure is defined as the failure of gery is indicated at all. Even for an eye that is
the corneal transplant to adequately restore the blind—either from corneal or noncorneal pathol-
visual function of the patient in the absence of ogy—corneal transplantation may still provide
concomitant pathology and is typically classified significant benefits in terms of making the eye
into [2]: more comfortable and reducing the risk of infec-
tion (although a conjunctival flap may be a more
• Primary graft failure (PGF) is defined as the suitable alternative in this situation while not
presence of a diffusely edematous corneal necessarily precluding subsequent corneal trans-
graft on the first postoperative day that fails to plantation) [4]. Immunological graft failure may
clear at any time postoperatively, with a lack usually be diagnosed with confidence 2–3 months
of an identifiable cause of corneal graft failure after the onset of treatment for endothelial rejec-
within three months from the transplant [2]. It tion. Management for any acute (or suspected)
is the result of inadequate donor endothelial endothelial rejection should, therefore, have
cell function, inadequate tissue preservation, already proceeded according to local practice
or surgical trauma from harvesting or the patterns prior to considering further options, with
procedure. topical corticosteroids the mainstay and the use
• Secondary graft failure (SGF) involves the of subconjunctival steroids, systemic steroids (in
deterioration of vision as a consequence of the the form of either oral prednisone or pulsed intra-
loss of corneal transparency in a previously venous methylprednisolone) and other immuno-
functional corneal graft, usually due to the suppressive therapy instituted as appropriate.
presence of diffuse and irreversible corneal Rates of reversibility of severe endothelial rejec-
edema in a previously functional transplant as tion are between 51% and 64% [5–8].
a consequence of endothelial failure from The second question to consider is whether an
immunological (rejection) or nonimmunolog- endothelial keratoplasty might be performed.
ical (chronic late endothelial loss) events. It This confers a number of advantages discussed
may also refer to cases of transparency loss in below and should not be discounted, even in the
the absence of endothelial failure, such as cor- context of a chronically decompensated PK,
neal stromal scarring (herpetic disease, because of the capacity for remodeling of any
trauma) or surface opacification (limbal stem stromal opacification that might take a year or
cell deficiency). The latter causes will not be more [9]. Third, consideration should be given to
discussed in the current review. the surgical aim and whether the existing graft, at
• Morphological graft failure refers to clear its best, ever provided the patient with useful
grafts without transparency loss but confer- vision. For example, if the graft profile was excel-
ring poor visual function as a result of severe lent (low/moderate and regular corneal astigma-
irregular astigmatism (such as severe recur- tism), affording the patient a high level of
rent ectasia). It may be considered a subtype unaided, spectacle-corrected, or contact lens-
of SGF. corrected visual acuity prior to failure, then every
effort should be made to maintain this with endo-
Despite the rise of lamellar keratoplasty tech- thelial keratoplasty. At the other extreme, if the
niques in the last two decades, PK remains one of profile of the graft was highly irregular even
the commonest forms of corneal transplantation before failure, and the patient had been unable to
and remains the procedure of choice for selected tolerate a contact lens to correct for this, then it
indications [3]. Options for the management of a may be more advantageous to attempt a redo PK
failed PK depend on the clinical context, the and improve on the existing levels of
mechanism of failure, and any comorbidity. astigmatism.
10 Rescuing Failed Penetrating Keratoplasty Grafts 147
K Graft Failure: Incidence and Risk

P (e.g., where there is severe stromal opacification
Factors in addition to endothelial dysfunction). While
there has been a progressive reduction in the vol-
The survival rate for PK has been reported umes of primary PKs undertaken, the number of
recently to be between 52% and 98.8% at 10-year PKs constituting regrafts has remained relatively
follow-ups [3, 10–15]. This extended range stable in recent years, accounting for 11.3–19.0%
derives from highly variable methods of data col- of the total PKs performed in the US from 2005
lection and reporting by investigators from differ- to 2019 [26, 27]. Mattahei and coworkers, in a
ent parts of the world [3, 16]. Furthermore, systematic review of the global indications for
datasets published before the advent of endothe- PK surgery, found that between 1980 and 2014,
lial keratoplasty likely provide a more compre- PK regrafts represented 9.3–18.8% of the total
hensive overview of the risks and benefits of number of PK performed in developed countries
repeat PK. PK is now less commonly performed [16].
for low-risk indications such as keratoconus or Preoperative medical management and plan-
Fuchs dystrophy, where lamellar techniques are ning are fundamental to success. Subsequent PKs
now generally regarded as the gold standard. PK usually carry all the risks of the first transplant
is increasingly being reverted to in more complex plus the increased risk of allograft rejection, as
cases where the technical difficulty inherent to well as cumulative comorbidities such as glau-
lamellar techniques preclude their success [17]. coma [22, 28]. Consequently, every effort should
Outcomes with PK are strongly related to the pri- be made to optimize the condition of the ocular
mary diagnosis—the indication for transplanta- surface, control intraocular pressure, address
tion being the strongest predictor of survival [18]. inflammation (including systemic inflammatory
Transplantation performed for keratoconus and diseases), and manage other pathologies that
stromal dystrophies have a better prognosis than might impact the risk of rejection such as corneal
grafts performed for endothelial diseases, bullous neovascularization and herpetic eye disease [22,
keratopathy, and postinfective corneal scars, for 29]. Preoperative imaging, including anterior
example [15, 17, 19]. Moreover, when a PK is segment optical coherence tomography, may be
performed for therapeutic and tectonic indica- helpful in anticipating and planning for addi-
tions, as opposed to optical rehabilitation, the tional procedures that may be indicated more fre-
prognosis is even poorer [13, 20]. Other well- quently with repeat transplantation, including
established risk factors for PK failure include a synechiolysis, iridectomy, and pupilloplasty [30].
history of previous failed PK, the presence of For grafts at high risk of rejection—usually
active ocular inflammation (including surface defined as one where there are two or more quad-
inflammation), corneal neovascularization, other rants of neovascularization or where previous
ocular comorbidities (such as herpetic eye dis- rejection has occurred—there is some, albeit lim-
eases or glaucoma), aphakia, larger graft diame- ited, evidence to support the use of both pre- and
ter, and the undertaking of combined surgical postoperative systemic immunosuppression [31].
procedures [15, 18, 21–25]. Management varies widely, but presurgical sys-
temic corticosteroid treatment improves corneal
transplant survival in murine models and courses
Repeat PK of pre- and postoperative oral prednisone (1 mg/
kg) or intravenous methylprednisolone are often
As a consequence of the recent introduction of prescribed in this context, usually in combination
posterior lamellar grafts, repeat PK remains the with topical steroid treatment and, less com-
only surgical option for PK grafts that have failed monly, steroid-sparing agents such as mycophe-
and where the functional potential of the existing nolate mofetil, cyclosporin A, tacrolimus,
transplant has been fundamentally compromised sirolimus, and topical cyclosporin [6, 31–33]. A
significant proportion of surgeons in one study comorbidities such as glaucoma and cataract are
prescribe topical steroids for use indefinitely in more common in such cases [41, 42]. With regard
high-risk cases and oral acyclovir for a median of to host trephination, consideration should be given
6 months in those cases involving Herpes simplex to the indications for the transplant and the reasons
keratopathy [33]. There is moderate evidence for for failure. The capacity to heal varies consider-
the benefit of mycophenolate mofetil on corneal ably between patients and eyes, and this is further
transplant survival, but evidence for the effective- affected by the duration of any steroid treatment
ness of other noncorticosteroid immunosuppres- and even whether the wound interface is that of a
sive agents is less conclusive [31, 34, 35]. The primary PK or a regraft, with evidence for a stron-
case for human leukocyte antigen (HLA) match- ger wound-healing response in the latter [43]. It
ing of tissue to the recipient has yet to be demon- might be assumed, at the point where one might
strated in studies determining whether this results consider suture removal (usually ≥1 year after sur-
in a reduction in the risk of allograft rejection gery), that the graft-host interface is sufficiently
compared with random tissue allocation; equally, healed as to make re-trephination a more obvious
there some evidence to suggest a role for the indi- choice but, in fact, the strength of the graft-host
rect presentation of minor histocompatibility interface is never much more than that afforded by
antigens (such as male H-Y antigen) in modulat- the endothelium and epithelium—this is what
ing the risk of corneal transplant rejection and makes PKs particularly vulnerable to traumatic
graft survival, but this has yet to inform practice injury and, while most traumatic wound dehis-
widely, and the effect may not be significant in cence occurs within 18–24 months of transplanta-
high-risk cases [36, 37]. tion, this can occur even decades after the surgery
Given the risks of allograft rejection in eyes [44–46]. Consequently, most PKs may be replaced
that have undergone multiple PKs, these may by simply accessing the graft-host junction, peel-
ultimately be better candidates for a keratopros- ing out the original graft, and suturing in a trans-
thesis, such as the Boston type 2 keratoprosthesis plant of the same diameter [46, 47]. The appearance
(BKPro), if the ocular surface is not compromised on slit lamp examination preoperatively of vessels
by dryness or a lack of blinking [28]. Refinements and stromal opacification at the graft-host inter-
to the device itself and the management of eyes face may give some indication as to the degree of
implanted with it have led to an expansion of the healing that has already taken place and, if signifi-
indications for the BKPro in recent years. There cant, a little more resistance might be expected
is some evidence to support the use of the BKPro with this approach, but remains the favored option
earlier in the pathway of eyes at risk of further in most cases.
failed PKs, with one study demonstrating that the As a general rule, the transplant should be as
likelihood of maintaining 20/200 or better vision “large as possible and as small as necessary”—
at 2 years and a clear graft at 5 years was signifi- that is to say, a balance should be struck, in terms
cantly higher with the BKPro than PK in eyes of the optical advantages conferred by a larger
with a previously failed PK (without a higher risk graft and the reduced risks of allograft rejection
of postoperative glaucoma) [38–40]. associated with smaller diameter transplants [17,
18]. While peeling out the original donor is rela-
tively easily performed, repeat PK does afford the
Surgical Considerations opportunity to excise the old transplant together
with a rim of host cornea by performing a slightly
Basic surgical principles apply when undertaking larger trephination—if this can be accomplished
a repeat PK. Particular care in surgical planning while maintaining sufficient clearance from the
should be undertaken when the original transplant limbus, ideally at least 1.5 mm. This may be pre-
was a therapeutic keratoplasty for microbial kera- ferred if decentration, high levels of astigmatism,
titis, because disruption of the anterior segment or irregularity have been a problem with the origi-
anatomy, eccentrically-sited transplants, and nal transplant. Otherwise, greater weight should
be afforded to the risks of allograft rejection in the up with particular attention to optimizing the
context of a repeat PK [28]. A “peephole” trans- ocular surface and managing postoperative
plant, sited within a larger diameter, failed trans- inflammation.
plant, may offer considerable functional benefit
while minimizing the risks of further failure,
albeit with a flatter profile and higher levels of Outcomes
irregular astigmatism [48]. Conversely, a graft
that has failed due to infection or a corneal melt- Repeat PKs have a higher risk of failure com-
ing disorder may require excision of diseased or pared with primary PK, with a worse prognosis
thinned tissue far outside of the existing trans- and poorer visual outcomes. The 5-year graft
plant, sometimes up to the limbus—with the failure rate for repeat PK is between 34% and
caveat that it is almost always worth considering a 70% [41, 53–56], with numerous factors playing
tectonic anterior lamellar keratoplasty in an acute a role in the survival of the transplant. Repeated
setting rather than a penetrating keratoplasty [20, PKs share the same risk factors as primary PK,
49]. Where a large diameter keratoplasty, such as but there is a higher risk of immunological
a total corneal transplant, is indicated, every effort allograft rejection associated with violation of
should be made to ensure that the underlying the immune-privileged status of the cornea. This
cause has been managed appropriately, that risk increases with the number of prior corneal
inflammation has settled, and that the host limbal grafts performed, especially if they have failed as
stem cell populations are spared by undermining a result of allograft rejection [25, 53, 57]. In addi-
and retracting the corneal limbus prior to trephi- tion, the host corneal bed is usually suboptimal,
nation [20, 50]. Some vacuum trephines may not with high rates of altered iridocorneal angle anat-
be available in size required or even suitable for omy, peripheral anterior synechiae (PAS), and
application in these circumstances—in which corneal surface disease. Moreover, when com-
case marking of the surface, or partial trephina- pared with patients receiving primary PK, PK
tion, with a hand-held trephine, followed by a regraft patients are slightly older, have higher use
freehand dissection may permit more controlled of IOP-lowering drugs, and are more prone to
excision of the host button [49]. develop postoperative corneal neovasculariza-
Given the increased risk of allograft rejection tion, all of which are associated with poor graft
attached to repeat PK, any advantages of a con- survival [56]. Finally, as for primary PK, the risk
tinuous suture are usually outweighed by those of a failed graft is closely associated with the
conferred by the use of interrupted sutures. original graft indication, with the best rates of
Interrupted sutures may permit the early removal survival reported for eyes with a primary diagno-
of a combined continuous suture, a loose suture, sis of keratoconus and stromal dystrophies [54].
a suture driving a foreign body response, or cor- With regard to visual outcomes, repeat PK is
neal neovascularization but also facilitate the generally effective in improving the preoperative
management of astigmatism postoperatively best-corrected visual acuity (BCVA), but the
[51]. The application of a Flieringa ring for results achieved are usually worse than those for
scleral support is advisable when performing PK primary PK. In fact, a BCVA of 20/40 or better
in aphakic eyes [51]. Transplants at high risk of has been reported as being only 4.8–43.1% of
rejection and any transplant performed in the clear PK regrafts at the last follow-up visit [54,
context of ocular surface disease will likely ben- 58–65]. In addition, the standard of final visual
efit from a central temporary tarsorrhaphy at the acuity achieved is inversely proportional to the
end of the procedure [52]. This may afford a suf- number of corneal regrafts performed [59].
ficient view of the transplant during the immedi- Despite this, a recent meta-analysis performed by
ate postoperative period while facilitating corneal Wang and coworkers found that visual outcomes
protection and epithelialization [52]. Repeat PK in PK regrafts were not inferior to the ones for
merits frequent and careful postoperative follow- secondary endothelial grafts [66].
escemet Stripping Endothelial

D tion and ensure that it will fit inside the area of
Keratoplasty (DSEK) after Failed PK the prior PK [74]. Finally, Descemet membrane
(DM) stripping in these DSEK under PK cases
Endothelial keratoplasty (EK) has essentially can be more challenging and may also lead to
replaced PK for the treatment of end-stage cor- inadvertent disruption of the graft-host junction
neal endothelial disease [67]. The advantages of [75]. Moreover, DM stripping has not been
Descemet stripping EK (DSEK) over that of a reported to reduce the graft dislocation rate in
primary PK include a reduced need for sutures, such cases [75], so surgeons may therefore
resulting in a tectonically stronger eye, mitigated choose not to perform DM stripping when per-
effects of ocular surface disease, and more rapid forming a DSEK under a failed PK graft.
postoperative visual rehabilitation [68–71]. In
cases of failed PK, however, positioning of a
DSEK under the failed transplant may still be Outcomes
performed (Fig. 10.1), with the main theoretical
advantage being a reduced risk of allograft rejec- Noncomparative studies have reported 3-year
tion compared with a repeat PK [72, 73]. As such, graft survival rates of 69–81% for DSEK under
patients with a failed PK who undergo a DSEK PK, and 21–70% for repeat PK [14, 47, 56, 59,
instead of a repeat PK are less likely to require 62, 63, 76, 77]. Furthermore, a direct compara-
systemic immunosuppression. tive long-term study found graft survival was
prolonged significantly in eyes that underwent
DSEK under a failed PK, compared with eyes
Surgical Considerations undergoing repeat PK at up to 5 years follow-up
(cumulative survival probability 86.4% versus
Technically, performing a DSEK under a failed 51.3%, respectively; log-rank P value = 0.013)
PK does not differ much from a standard DSEK [78]. Graft rejection episodes, specifically, were
case. Special attention should be paid to the pos- also reported to be more common in the repeat
terior graft–host junction morphology (ideally PK group compared with the DSEK under PK
imaged by anterior segment OCT) as significant group, while the incidence of postoperative com-
steps between the donor and the host posterior plications was no greater in the DSEK under PK
surfaces may preclude correct graft adherence if group [78].
the donor tissue is decentered or oversized com- Another potential advantage of performing
pared to the PK posterior diameter. Subsequently, DSEK instead of PK in eyes with corneal decom-
the diameter of the DSEK donor lenticule should pensation is the promise of faster postoperative
ideally match that of the posterior diameter of the visual rehabilitation and possibly better long-
PK, and careful attention should be taken to cen- term visual outcomes [79–81]. Most studies on
ter the DSEK graft within the PK graft-host junc- DSEK after failed PK have reported significant
Fig. 10.1 Slit lamp picture and anterior segment OCT image (Visante, Zeiss, Germany) of a patient that received
Descemet’s stripping DSAEK under a previously failed PK
improvements in postoperative visual acuity [82– of transplant rejection [73]. A recent comparative
85]. Best-corrected visual acuity (BCVA) has study demonstrated that DMEK delivers better
been reported to increase from 1.00 logMAR pre- visual outcomes than DSAEK for managing
operatively to 0.40 logMAR at 1 year postopera- failed PK grafts [90]. This study found that,
tively in 60 eyes (P < 0.0001) in one case series despite an overall failure rate in the medium-term
[86] and from 1.43 logMAR preoperatively to that was similar between the two techniques
0.55 logMAR at 1 year postoperatively in 22 eyes (DMEK-43% vs. DSAEK-50%), the primary fail-
(P = 0.001) in another [84]; however, compara- ure rate was higher in the DMEK group (due to
tive case series of DSAEK under a failed PK and persistent graft detachment), while the secondary
repeat PK have not found statistically significant failure rate was higher in the DSAEK group. The
differences in improvement in postoperative rejection rate was reported to be similar (DMEK-
BCVA between the two techniques [61, 78]. This 21% vs. DSAEK-29%) [90]. Thus, while grafting
may be explained by the presence of stromal a DMEK under a failed PK is more technically
scarring or preexisting astigmatism of the failed challenging than DSAEK, with a higher risk of
PK in eyes that undergo DSEK under the PK early failure, if successful, it confers better long-
[78]. In such cases, one may perform a DALK in term visual outcomes and survival rates.
the previous PK over the DSAEK graft [87].
Graft dislocation, which has a reported rate of
5.9–6.3%, remains the main concern when per- Descemet Stripping DMEK Under PK
forming DSEK under failed PK [74, 85].
Suggested risk factors for this include the Table 10.1 summarizes the reported outcomes of
presence of a glaucoma drainage device, donor DMEK for the management of secondary failure
grafts that are smaller than the previous PK, and of PK [91–100]. Performing DMEK in this sce-
a mismatch of the posterior graft–host junction. nario carries the additional intraoperative risk of
dehiscence of the PK graft-host junction when
performing descemetorhexis maneuvers.
escemet Membrane Endothelial
D Consequently, it is recommended that the desce-
Keratoplasty (DMEK) After Failed PK metorhexis is made 0.25–0.5 mm smaller than
the PK diameter, thereby avoiding the graft-host
EK in the form of Descemet Stripping Endothelial junction. It is also recommended that the diame-
Keratoplasty (DSEK) has proven efficacy in ter of the DMEK is either matched or undersized
restoring clarity to full-thickness grafts with (by 0.25 mm) compared with the PK diameter,
endothelial decompensation. In the absence of noting that oversized DMEK grafts are associ-
significant scarring or surface irregularity, this has ated with higher re-bubbling rates [93].
been the preferred surgical approach for the man- DMEK under a failed PK is also associated
agement of a decompensated PK over repeat PK with a higher rate of re-bubbling—much higher
for more than a decade because it affords a lower than after primary DMEK cases—up to 60% in
risk of rejection, prolongs graft survival, has some series [91–96, 99]. This higher detachment
greater safety in the presence of ocular surface rate and need to re-bubble is thought to be related
disease, affords better intraocular pressure control to the DM tags and remnants associated with a
(and therefore fewer cases of secondary glau- more challenging descemetorhexis and the more
coma), and typically allows for faster visual irregular posterior corneal surfaces encountered
recovery and a more acceptable refractive out- (with pronounced gaps at the graft-host junction).
come [78, 82, 88, 89]; however, in recent years, Under these circumstances, gas tamponade with
Descemet membrane endothelial keratoplasty SF6 20% at the end of the surgery may be par-
(DMEK) has been demonstrated to confer advan- ticularly advantageous, given that this has been
tages over DSEK in terms of the speed of visual demonstrated to reduce the need for re-bubbling
recovery, better visual outcomes, and a lower risk compared with air tamponade [95].
152
Table 10.1 Reported outcomes of DMEK for failed PK management

PK Postop ECC loss Re-bubbling Rejection
Author Year N Time Descemetorhexis Tamponade Preop BCVA BCVA (time point) rate rate Graft failure
Anshu et al. [91] 2013 6 6m Yes Air 20/70 20/30 33% (6 m) 50% 0% 17% PGF
Snellen Snellen 0% SGF
Heinzelmann et al. 2017 19 21.5 m Yes Air 20/400 20/200 n/r 26% 10.5% 5% PGF
[92] (1–42 m) Snellen Snellen 11% SGF
Lavy et al. [86] 2017 11 16 m Yes Air Range 5 eyes 59% (16 m) 36% 0% 9% PGF
HM-20/30 ≥20/25 18% SGF
Snellen
Pasari et al. [93] 2019 93 21 m (1 Yes Air 20/100 (6 m) 20/30 (6 m) 31% (6 m) 53% (DMEK > 2% 5% PGF
m–7 y) Snellen Snellen 44% (12 m) PK) 11% SGF
47% (24 m) 27% (DMEK =
PK)
33% (DMEK <
PK)
Pierné et al. [94] 2019 28 6 m Yes Air 1.34 0.54 24% (6 m) 50% 0% 0% PGF
LogMAR LogMAR 0% SGF
Schrittenlocher et al. 2020 52 12 m Yes Air/SF6 1.07 0.56 (6 m) 36% (6 m) 59.6% 13.5% 34.6%
[95] 20% LogMAR 0.38 (12 m) 37% (12 m)
LogMAR 40% (24 m)
Sorkin et al. [96] 2019 29 6 m Yes Air 0.97 0.51 38% (6 m) 58.6% 6.9% 27.6% PGF
LogMAR LogMAR 6.9% SGF
Sorkin et al. [96] 2019 10 6 m Yes (Femto) Air 1.32 0.35 43.8% (6 m) 10% 10% 0% PGF
LogMAR LogMAR 0% SGF
Alió del Barrio et al. 2019 8 6 m No SF6 20% 20/800 20/40 50% (6 m) 25% 0% 0% PGF
[97] Snellen Snellen 0% SGF
Güell et al. [98] 2019 26 23.08m No SF6 20% 20/80 (6 m) 20/32 (6 m) 27.7% (6 m) 11.53% 0% 0% PGF
(6–48 m) Snellen Snellen 34.4% (12 m) 7.7% SGF
N study sample, BCVA best-corrected visual acuity, HM hand movements, ECC endothelial cell count, m months, y years, Femto femtosecond laser-assisted descemetorhexis,
SF6 20% sulfur hexafluoride, DMEK>PK DMEK graft diameter larger than PK graft diameter, DMEK=PK DMEK graft diameter equal than PK graft diameter, DMEK<PK
DMEK graft diameter smaller than PK graft diameter, PGF primary graft failure, SGF secondary graft failure, LogMAR logarithm of the minimum angle of resolution, Snellen
Snellen visual acuity.
J. L. A. del Barrio et al.
A further concern is the relatively high rate of topathologic analysis that the excised DM pre-
endothelial cell loss and graft failure observed fol- sented a continuous extracellular matrix
lowing DMEK under PK. This may be accounted connecting the host and donor DM, indicating
for by the higher rate of re-bubbling and additional that a primary intention wound healing after PK
manipulation of the DMEK donor required in such may occur in some cases at this tissue level, and
cases [93]. In contrast, known risk factors in repeat so it could enable in such cases a DMrhexis out-
PK, such as the number of previous PKs, preexist- side the failed graft and the transplantation of a
ing anterior synechiae, and corneal neovascular- DMEK graft larger than the previous
ization, appear not to have a significant role in PK. Nevertheless, this finding is extracted from
modulating the risk of failure for DMEK under PK an isolated case report [101].
[93]; however, previous glaucoma surgery in the
operated eye has been shown to be an important
risk factor for DMEK failure under PK. This rela- on-Descemet Stripping DMEK
N
tively common situation may carry with it comor- Under PK
bidities that ultimately limit any visual gains [93].
In comparison with routine DMEK, visual recov- As it has been suggested previously with
ery in eyes with preexisting PK is generally slower, DSEK [85], another approach to minimizing
with improvements in visual acuity observed up to posterior surface damage to the PK during
6 months postoperatively [94, 95]. descemetorhexis, and therefore reducing the
To minimize any damage to the posterior PK risk of re-bubbling and DMEK failure, is to
surface during DMEK under PK surgery, femto- avoid this surgical step altogether—thereby
second laser-assisted descemetorhexis has been making the surgery simpler, faster and less
attempted. This technique, which involves using risky [97, 98]. Contrary to evidence support-
the femtosecond laser to perform a side cut in the ing a role for descemetorhexis in routine
posterior surface of the PK and so facilitate the DMEK [93], recent studies have demonstrated
DMrhexis maneuvers, has shown encouraging that nonstripping DMEK under a failed PK
results in reducing the rate of graft re-bubbling may be advantageous as long as the host DM
(10%) and failure (0%) [96, 100]; however, the is anatomically intact (no scarring or signifi-
study sample was limited (n = 10), and no differ- cant irregularities), with lower incidences of
ences in endothelial cell loss rates were observed graft re-bubbling and failure reported com-
in the short- or long-term compared to manual pared with manual stripping techniques
DMrhexis DMEK under PK (Table 10.1). (Fig. 10.2 and Table 10.1) [97, 98].
Recently, Steindor et al. published a case
report where they stripped the DM starting from
outside the failed PK and observed through a his-
Fig. 10.2 Slit lamp picture and anterior segment OCT PK. Images are taken 2 years after surgery. The inferior
image (MS-39, CSO, Italy) of a patient that received non- peripheral scar appeared as a consequence of microbial
Descemet’s stripping DMEK under a previously failed keratitis 1 year after surgery
Conclusion 4. Alino AM, Perry HD, Kanellopoulos AJ, Donnenfeld

ED, Rahn EK. Conjunctival flaps. Ophthalmology.
1998;105(6):1120–3.
Corneal graft failure is a growing medical 5. Qazi Y, Hamrah P. Corneal allograft rejection:
challenge because of the constantly increasing immunopathogenesis to therapeutics. J Clin Cell
volume of transplanted patients and the Immunol. 2013;2013(9):6.
6. Kharod-Dholakia B, Randleman JB, Bromley JG,
increase in life expectancy worldwide. The Stulting RD. Prevention and treatment of corneal
introduction of stromal and endothelial lamel- graft rejection: current practice patterns of the
lar transplant techniques has brought specific Cornea Society (2011). Cornea. 2015;34(6):609–14.
complications and new forms of graft failure 7. Hudde T, Minassian DC, Larkin DF. Randomised
controlled trial of corticosteroid regimens in endo-
that will demand different approaches to those thelial corneal allograft rejection. Br J Ophthalmol.
encountered with classical repeat PK. On the 1999;83(12):1348–52.
other hand, these novel and exciting lamellar 8. Yamazoe K, Yamazoe K, Shimazaki-Den S,
techniques open new doors to rescuing failed Shimazaki J. Prognostic factors for corneal graft
recovery after severe corneal graft rejection fol-
PKs, helping overcome previous limitations lowing penetrating keratoplasty. BMC Ophthalmol.
linked to successive penetrating grafts and 2013;13:5.
offering better and faster visual results with 9. Pasari A, Price MO, Feng MT, Price FW Jr.
much-reduced intra- and postoperative risks. Descemet membrane endothelial keratoplasty for
failed penetrating keratoplasty: visual outcomes and
Corneal specialists should be aware of the graft survival. Cornea. 2018;38(2):151–6.
modern surgical alternatives available to them 10. Borderie VM, Boëlle PY, Touzeau O, Allouch C,
for the rehabilitation of a patient’s visual func- Boutboul S, Laroche L. Predicted long-term out-
tion when faced with a failed corneal trans- come of corneal transplantation. Ophthalmology.
2009;116(12):2354–60.
plant. A comprehensive understanding of the 11. Fukuoka S, Honda N, Ono K, Mimura T, Usui
outcomes and limitations of these options is T, Amano S. Extended long-term results of pen-
also critical. etrating keratoplasty for keratoconus. Cornea.
2010;29(5):528–30.
12. Mannis MJ, Holland EJ, Gal RL, Dontchev M,
Take Home Notes Kollman C, Raghinaru D, et al. The effect of donor
• Corneal graft failure is a growing medical age on penetrating keratoplasty for endothelial dis-
challenge because of the constantly increasing ease: graft survival after 10 years in the cornea donor
volume of transplanted patients and the study. Ophthalmology. 2013;120(12):2419–27.
13. Tan DTH, Janardhanan P, Zhou H, Chan YH, Htoon
increase in life expectancy worldwide. HM, Ang LPK, et al. Penetrating keratoplasty in
• Endothelial keratoplasty techniques open new asian eyes. The Singapore Corneal Transplant Study.
doors to rescuing failed PKs, helping over- Ophthalmology. 2008;115(6):975–83.
come previous limitations linked to successive 14. Thompson RW, Price MO, Bowers PJ, Price
FW. Long-term graft survival after penetrating kera-
penetrating grafts and offering better and toplasty. Ophthalmology. 2003;110(7):1396–402.
faster visual results with much reduced intra- 15. Williams KA, Lowe M, Bartlett C, Kelly T-L,
and postoperative risks. Coster DJ. Risk factors for human corneal graft fail-
ure within the Australian Corneal Graft Registry.
Transplantation. 2008;86(12):1720–4.
16. Matthaei M, Sandhaeger H, Hermel M, Adler W,
References Jun AS, Cursiefen C, et al. Changing indications
in penetrating keratoplasty: a systematic review
1. Gain P, Jullienne R, He Z, Aldossary M, Acquart S, of 34 years of global reporting. Transplantation.
Cognasse F, et al. Global survey of corneal trans- 2017;101(6):1387–99.
plantation and eye banking. JAMA Ophthalmol. 17. Williams KA, Coffey NE, et al. The Australian cor-
2016;134(2):167–73. neal graft registry 2018 report. Adelaide: Flinders
2. Alió JMA, El Sayyad F, Barraquer R, Arnalich- University; 2018.
Montiel F, Alió del Barrio JL. Corneal 18. Barraquer RI, Pareja-Aricò L, Gómez-Benlloch A,
graft failure: an update. Br J Ophthalmol. Michael R. Risk factors for graft failure after pene-
2020;105(8):1049–58. trating keratoplasty. Medicine. 2019;98(17):e15274.
3. Patel SV. Graft survival after penetrating kerato- 19. Williams KA, Esterman AJ, Bartlett C, Holland
plasty. Am J Ophthalmol. 2011;151(3):397–8. H, Hornsby NB, Coster DJ. How effective is pen-
etrating corneal transplantation? Factors influenc- 34. Birnbaum F, Mayweg S, Reis A, Bohringer D, Seitz
ing long-term outcome in multivariate analysis. B, Engelmann K, et al. Mycophenolate mofetil
Transplantation. 2006;81(6):896–901. (MMF) following penetrating high-risk keratoplasty:
20. Ang M, Mehta JS, Sng CC, Htoon HM, Tan long-term results of a prospective, randomised, mul-
DT. Indications, outcomes, and risk factors for ticentre study. Eye. 2009;23(11):2063–70.
failure in tectonic keratoplasty. Ophthalmology. 35. Sinha R, Jhanji V, Verma K, Sharma N, Biswas NR,
2012;119(7):1311–9. Vajpayee RB. Efficacy of topical cyclosporine A 2%
21. Perera C, Jhanji V, Lamoureux E, Pollock G, Favilla in prevention of graft rejection in high-risk kerato-
I, Vajpayee RB. Clinical presentation, risk factors plasty: a randomized controlled trial. Graefes Arch
and treatment outcomes of first allograft rejection Clin Exp Ophthalmol. 2010;248(8):1167–72.
after penetrating keratoplasty in early and late post- 36. Böhringer D, Grotejohann B, Ihorst G, Reinshagen
operative period. Eye. 2012;26(5):711–7. H, Spierings E, Reinhard T. Rejection prophy-
22. Price MO, Thompson RW Jr, Price FW Jr. Risk fac- laxis in corneal transplant. Dtsch Arzteblatt Int.
tors for various causes of failure in initial corneal 2018;115(15):259–65.
grafts. Arch Ophthalmol. 2003;121(8):1087–92. 37. Hopkinson CL, Romano V, Kaye RA, Steger B,
23. Sugar A, Gal RL, Kollman C, Raghinaru D, Stewart RM, Tsagkataki M, et al. The influence
Dontchev M, Croasdale CR, et al. Factors associated of donor and recipient gender incompatibility
with corneal graft survival in the cornea donor study. on corneal transplant rejection and failure. Am J
JAMA Ophthalmol. 2015;133(3):246–54. Transplant. 2017;17(1):210–7.
24. Sugar A, Tanner JP, Dontchev M, Tennant B, 38. Hager JL, Phillips DL, Goins KM, Kitzmann AS,
Schultze RL, Dunn SP, et al. Recipient risk fac- Greiner MA, Cohen AW, et al. Boston type 1 kera-
tors for graft failure in the cornea donor study. toprosthesis for failed keratoplasty. Int Ophthalmol.
Ophthalmology. 2009;116(6):1023–8. 2016;36(1):73–8.
25. Wilson SE, Kaufman HE. Graft failure after 39. Ahmad S, Mathews PM, Lindsley K, Alkharashi M,
penetrating keratoplasty. Surv Ophthalmol. Hwang FS, Ng SM, et al. Boston type 1 keratopros-
1990;34(5):325–56. thesis versus repeat donor keratoplasty for corneal
26. Park CY, Lee JK, Gore PK, Lim CY, Chuck graft failure: a systematic review and meta-analysis.
RS. Keratoplasty in the United States: a 10-year Ophthalmology. 2016;123(1):165–77.
review from 2005 through 2014. Ophthalmology. 40. Fadous R, Levallois-Gignac S, Vaillancourt L,
2015;122(12):2432–42. Robert MC, Harissi-Dagher M. The Boston kera-
27. America EBAo. The 2019 EBAA statistical report: toprosthesis type 1 as primary penetrating corneal
highlights, trends in eye banking and corneal trans- procedure. Br J Ophthalmol. 2015;99(12):1664–8.
plantation. 2019. 41. Ramamurthy S, Reddy JC, Vaddavalli PK, Ali
28. Maguire MG, Stark WJ, Gottsch JD, Stulting MH, Garg P. Outcomes of repeat keratoplasty for
RD, Sugar A, Fink NE, et al. Risk factors for cor- failed therapeutic keratoplasty. Am J Ophthalmol.
neal graft failure and rejection in the collabora- 2016;162(83-8):e2.
tive corneal transplantation studies. Collaborative 42. Srujana D, Kaur M, Urkude J, Rathi A, Sharma
Corneal Transplantation Studies Research Group. N, Titiyal JS. Long-term functional and ana-
Ophthalmology. 1994;101(9):1536–47. tomic outcomes of repeat graft after optically
29. Hos D, Le VNH, Hellmich M, Siebelmann S, failed therapeutic keratoplasty. Am J Ophthalmol.
Roters S, Bachmann BO, et al. Risk of corneal 2018;189:166–75.
graft rejection after high-risk keratoplasty following 43. Abdelkader A. Changes in corneal wound healing
fine-needle vessel coagulation of corneal neovascu- and graft biomechanics after primary penetrating
larization combined with bevacizumab: a pilot study. keratoplasty versus repeat penetrating keratoplasty.
Transplant Direct. 2019;5(5):e452. Cornea. 2019;38(8):1006–10.
30. Ang M, Baskaran M, Werkmeister RM, Chua J, 44. Raber IM, Arentsen JJ, Laibson PR. Traumatic
Schmidl D, Aranha Dos Santos V, et al. Anterior seg- wound dehiscence after penetrating keratoplasty.
ment optical coherence tomography. Prog Retin Eye Arch Ophthalmol. 1980;98(8):1407–9.
Res. 2018;66:132–56. 45. Binder PS, Abel R Jr, Polack FM, Kaufman
31. Abudou M, Wu T, Evans JR, Chen HE. Keratoplasty wound separations. Am J
X. Immunosuppressants for the prophylaxis of cor- Ophthalmol. 1975;80(1):109–15.
neal graft rejection after penetrating keratoplasty. 46. Pettinelli DJ, Starr CE, Stark WJ. Late traumatic
Cochrane Database Syst Rev. 2015;8:CD007603. corneal wound dehiscence after penetrating kerato-
32. Kim HK, Choi JA, Uehara H, Zhang X, Ambati plasty. Arch Ophthalmol. 2005;123(6):853–6.
BK, Cho YK. Presurgical corticosteroid treat- 47. Kirkness CM, Ezra E, Rice NS, Steele AD. The
ment improves corneal transplant survival in mice. success and survival of repeat corneal grafts. Eye.
Cornea. 2013;32(12):1591–8. 1990;4(1):58–64.
33. Koay PY, Lee WH, Figueiredo FC. Opinions on risk 48. Seitz B, Langenbucher A, Kuchle M, Naumann
factors and management of corneal graft rejection in GO. Impact of graft diameter on corneal power
the United kingdom. Cornea. 2005;24(3):292–6. and the regularity of postkeratoplasty astigmatism
before and after suture removal. Ophthalmology. 64. Yalniz-Akkaya Z, Nurozler AB, Yildiz EH, Onat
2003;110(11):2162–7. M, Budak K, Duman S. Repeat penetrating kerato-
49. Bessant DA, Dart JK. Lamellar keratoplasty in the plasty: indications and prognosis, 1995-2005. Eur J
management of inflammatory corneal ulceration and Ophthalmol. 2009;19(3):362–8.
perforation. Eye. 1994;8(1):22–8. 65. Barut Selver O, Karaca I, Palamar M, Egrilmez S,
50. Jones DH, Kirkness CM. A new surgical technique Yagci A. Graft failure and repeat penetrating kerato-
for keratoglobus-tectonic lamellar keratoplasty plasty. Exp Clin Transplant. 2018;19(1):72–6.
followed by secondary penetrating keratoplasty. 66. Wang F, Zhang T, Kang YW, He JL, Li SM, Li
Cornea. 2001;20(8):885–7. SW. Endothelial keratoplasty versus repeat penetrat-
51. Lee RM, Lam FC, Georgiou T, Paul B, Then KY, ing keratoplasty after failed penetrating keratoplasty:
Mavrikakis I, et al. Suturing techniques and post- a systematic review and meta-analysis. PLoS ONE.
operative management in penetrating kerato- 2017;12(7):1–11.
plasty in the United Kingdom. Clin Ophthalmol. 67. Lee WB, Jacobs DS, Musch DC, Kaufman SC,
2012;6:1335–40. Reinhart WJ, Shtein RM. Descemet’s stripping
52. Koenig SB, Harris GJ. Temporary suture tarsor- endothelial keratoplasty: safety and outcomes: a
rhaphy after penetrating keratoplasty. Cornea. report by the American Academy of Ophthalmology.
1991;10(2):121–2. Ophthalmology. 2009;116(9):1818–30.
53. Aboshiha J, Jones MNA, Hopkinson CL, Larkin 68. Ang M, Htoon HM, Cajucom-Uy HY, Tan D, Mehta
DFP. Differential survival of penetrating and lamel- JS. Donor and surgical risk factors for primary
lar transplants in management of failed corneal graft failure following Descemet’s stripping auto-
grafts. JAMA Ophthalmol. 2018;136(8):859–65. mated endothelial keratoplasty in Asian eyes. Clin
54. Al-Mezaine H, Wagoner MD. Repeat penetrating Ophthalmol. 2011;5:1503–8.
keratoplasty: indications, graft survival, and visual 69. Ang M, Mehta JS, Anshu A, Wong HK, Htoon HM,
outcome. Br J Ophthalmol. 2006;90(3):324–7. Tan D. Endothelial cell counts after Descemet’s
55. Kitazawa K, Wakimasu K, Kayukawa K, Yokota stripping automated endothelial keratoplasty ver-
I, Inatomi T, Hieda O, et al. Moderately long-term sus penetrating keratoplasty in Asian eyes. Clin
safety and efficacy of repeat penetrating kerato- Ophthalmol. 2012;6:537–44.
plasty. Cornea. 2018;37(10):255–9. 70. Ang M, Mehta JS, Lim F, Bose S, Htoon HM, Tan
56. Weisbrod DJ, Sit M, Naor J, Slomovic AR. Outcomes D. Endothelial cell loss and graft survival after
of repeat penetrating keratoplasty and risk factors for Descemet’s stripping automated endothelial kerato-
graft failure. Cornea. 2003;22(5):429–34. plasty and penetrating keratoplasty. Ophthalmology.
57. Jabbehdari S, Rafii AB, Yazdanpanah G, Hamrah 2012;119(11):2239–44.
P, Holland EJ, Djalilian AR. Update on the man- 71. Ang M, Soh Y, Htoon HM, Mehta JS, Tan D. Five-
agement of high-risk penetrating keratoplasty. Curr year graft survival comparing descemet stripping
Ophthalmol Rep. 2017;5(1):38–48. automated endothelial keratoplasty and penetrating
58. Beckingsale P, Mavrikakis I, Al-Yousuf N, keratoplasty. Ophthalmology. 2016;123(8):1646–52.
Mavrikakis E, Daya SM. Penetrating keratoplasty: 72. Ezon I, Shih CY, Rosen LM, Suthar T, Udell
outcomes from a corneal unit compared to national IJ. Immunologic graft rejection in descemet’s strip-
data. Br J Ophthalmol. 2006;90(6):728–31. ping endothelial keratoplasty and penetrating kera-
59. Bersudsky V, Blum-Hareuveni T, Rehany U, Rumelt toplasty for endothelial disease. Ophthalmology.
S. The profile of repeated corneal transplantation. 2013;120(7):1360–5.
Ophthalmology. 2001;108(3):461–9. 73. Anshu A, Price MO, Price FW Jr. Risk of corneal
60. Keane MC, Galettis RA, Mills RAD, Coster DJ, transplant rejection significantly reduced with
Williams KA. A comparison of endothelial and Descemet’s membrane endothelial keratoplasty.
penetrating keratoplasty outcomes following failed Ophthalmology. 2012;119(3):536–40.
penetrating keratoplasty: a registry study. Br J 74. Straiko MD, Terry MA, Shamie N. Descemet strip-
Ophthalmol. 2016;100(11):1569–75. ping automated endothelial keratoplasty under
61. Kitzmann AS, Wandling GR, Sutphin JE, Goins failed penetrating keratoplasty: a surgical strat-
KM, Wagoner MD. Comparison of outcomes of egy to minimize complications. Am J Ophthalmol.
penetrating keratoplasty versus Descemet’s stripping 2011;151(2):233–7 e2.
automated endothelial keratoplasty for penetrating 75. Clements JL, Bouchard CS, Lee WB, Dunn SP,
keratoplasty graft failure due to corneal edema. Int Mannis MJ, Reidy JJ, et al. Retrospective review
Ophthalmol. 2012;32(1):15–23. of graft dislocation rate associated with descemet
62. Patel NP, Kim T, Rapuano CJ, Cohen EJ, Laibson stripping automated endothelial keratoplasty after
PR. Indications for and outcomes of repeat pen- primary failed penetrating keratoplasty. Cornea.
etrating keratoplasty, 1989-1995. Ophthalmology. 2011;30(4):414–8.
2000;107(4):719–24. 76. Al-Mezaine H, Wagoner MD, King Khaled Eye
63. Rapuano CJ, Cohen EJ, Brady SE, Arentsen JJ, Specialist Hospital Cornea Transplant Study
Laibson PR. Indications for and outcomes of G. Repeat penetrating keratoplasty: indications,
repeat penetrating keratoplasty. Am J Ophthalmol. graft survival, and visual outcome. Br J Ophthalmol.
1990;109(6):689–95. 2006;90(3):324–7.
77. Williams KA, Roder D, Esterman A, Muehlberg descemet membrane endothelial keratoplasty in the
SM, Coster DJ. Factors predictive of corneal graft treatment of failed penetrating keratoplasty. Cornea.
survival. Report from the Australian Corneal Graft 2019;38(9):1077–82.
Registry. Ophthalmology. 1992;99(3):403–14. 91. Anshu A, Price MO, Price FW Jr. Descemet mem-
78. Ang M, Ho H, Wong C, Htoon HM, Mehta JS, Tan brane endothelial keratoplasty and hybrid techniques
D. Endothelial keratoplasty after failed penetrating for managing failed penetrating grafts. Cornea.
keratoplasty: an alternative to repeat penetrating ker- 2013;32(1):1–4.
atoplasty. Am J Ophthalmol. 2014;158(6):1221–7. 92. Heinzelmann S, Bohringer D, Eberwein P, Lapp T,
79. Ang M, Lim F, Htoon HM, Tan D, Mehta JS. Visual Reinhard T, Maier P. Descemet membrane endothe-
acuity and contrast sensitivity following Descemet lial keratoplasty for graft failure following penetrat-
stripping automated endothelial keratoplasty. Br J ing keratoplasty. Graefes Arch Clin Exp Ophthalmol.
Ophthalmol. 2016;100(3):307–11. 2017;255(5):979–85.
80. Fuest M, Ang M, Htoon HM, Tan D, Mehta JS. Long- 93. Pasari A, Price MO, Feng MT, Price FW Jr.
term visual outcomes comparing descemet stripping Descemet membrane endothelial keratoplasty for
automated endothelial keratoplasty and penetrating failed penetrating keratoplasty: visual outcomes and
keratoplasty. Am J Ophthalmol. 2017;182:62–71. graft survival. Cornea. 2019;38(2):151–6.
81. Bose S, Ang M, Mehta JS, Tan DT, Finkelstein 94. Pierne K, Panthier C, Courtin R, Mazharian A,
E. Cost-effectiveness of Descemet’s stripping Souedan V, Gatinel D, et al. Descemet membrane
endothelial keratoplasty versus penetrating kerato- endothelial keratoplasty after failed penetrating ker-
plasty. Ophthalmology. 2013;120(3):464–70. atoplasty. Cornea. 2019;38(3):280–4.
82. Anshu A, Price MO, Price FW Jr. Descemet’s strip- 95. Schrittenlocher S, Schlereth SL, Siebelmann
ping endothelial keratoplasty under failed penetrating S, Hayashi T, Matthaei M, Bachmann B, et al.
keratoplasty: visual rehabilitation and graft survival Long-term outcome of descemet membrane endo-
rate. Ophthalmology. 2011;118(11):2155–60. thelial keratoplasty (DMEK) following failed
83. Covert DJ, Koenig SB. Descemet stripping and penetrating keratoplasty (PK). Acta Ophthalmol.
automated endothelial keratoplasty (DSAEK) in 2020;98(7):e901–6.
eyes with failed penetrating keratoplasty. Cornea. 96. Sorkin N, Mimouni M, Santaella G, Trinh T,
2007;26(6):692–6. Cohen E, Einan-Lifshitz A, et al. Comparison
84. Heitor de Paula F, Kamyar R, Shtein RM, Sugar A, of manual and femtosecond laser-assisted des-
Mian SI. Endothelial keratoplasty without Descemet cemet membrane endothelial keratoplasty for
stripping after failed penetrating keratoplasty. failed penetrating keratoplasty. Am J Ophthalmol.
Cornea. 2012;31(6):645–8. 2020;214:1–8.
85. Nottage JM, Nirankari VS. Endothelial keratoplasty 97. Alio Del Barrio JL, Montesel A, Ho V, Bhogal
without Descemet’s stripping in eyes with previous M. Descemet membrane endothelial keratoplasty
penetrating corneal transplants. Br J Ophthalmol. under failed penetrating keratoplasty without host
2012;96(1):24–7. descemetorhexis for the management of secondary
86. Anshu A, Lim LS, Htoon HM, Tan DT. Postoperative graft failure. Cornea. 2020;39(1):13–7.
risk factors influencing corneal graft survival in 98. Güell JL, Morral M, Barbany M, Gris O, Elies D,
the Singapore Corneal Transplant Study. Am J Manero F. Descemet membrane endothelial kera-
Ophthalmol. 2011;151(3):442–8. toplasty after penetrating keratoplasty. J EuCornea.
87. Fuest M, Siregar SR, Farrag A, Htoon HM, Tan D, 2019;2:10–3.
Mehta JS. Long-term follow-up of deep anterior 99. Lavy I, Liarakos VS, Verdijk RM, Parker J, Muller
lamellar keratoplasty after Descemet stripping auto- TM, Bruinsma M, et al. Outcome and histopathol-
mated endothelial keratoplasty. Graefes Arch Clin ogy of secondary penetrating keratoplasty graft fail-
Exp Ophthalmol. 2018;256(9):1669–77. ure managed by descemet membrane endothelial
88. Price FW Jr, Price MO. Endothelial keratoplasty to keratoplasty. Cornea. 2017;36(7):777–84.
restore clarity to a failed penetrating graft. Cornea. 100. Sorkin N, Trinh T, Einan-Lifshitz A, Mednick Z,
2006;25(8):895–9. Santaella G, Telli A, et al. Outcomes of femtosec-
89. Mitry D, Bhogal M, Patel AK, Lee BS, Chai SM, ond laser-assisted Descemet membrane endothelial
Price MO, et al. Descemet stripping automated keratoplasty for failed penetrating keratoplasty. Can
endothelial keratoplasty after failed penetrating J Ophthalmol. 2019;54(6):741–5.
keratoplasty: survival, rejection risk, and visual out- 101. Steindor FA, Clemens AC, Herwig-Carl MC, et al.
come. JAMA Ophthalmol. 2014;132(6):742–9. Wound healing of descemet membrane after pen-
90. Einan-Lifshitz A, Mednick Z, Belkin A, Sorkin N, etrating keratoplasty and its relevance for des-
Alshaker S, Boutin T, et al. Comparison of descemet cemet membrane endothelial keratoplasty surgeons.
stripping automated endothelial keratoplasty and Cornea. 2021;40(7):910–3.
Excimer Laser-Assisted
Keratoplasty: Penetrating
11
Keratoplasty “Excimer-PKP”
and Deep Anterior Lamellar
Keratoplasty “Excimer-DALK”
Loay Daas, Loïc Hamon, Elias Flockerzi,

Shady Suffo, and Berthold Seitz
Key Points xcimer Laser-Assisted Penetrating

E
• A comparison between excimer laser-assisted, Keratoplasty “Excimer-PKP”
Femtosecond laser-assisted, and mechanical (Video 11.1)
trephination regarding donor and recipient
centration, the occurrence of “vertical tilt” and Corneal transplantation is the oldest, most com-
“horizontal torsion” of the graft in the recipi- mon, and most successful transplantation in
ent bed, all-sutures-out—keratometric astig- humans. The first penetrating keratoplasty (PKP)
matism, and visual acuity. was performed in 1905 by Eduard Zirm in
• The Homburg/Erlangen technique of nonme- Olmütz, now in the Czech Republic [1]. With the
chanical excimer laser-assisted trephination. increasing expertise of the microsurgeon, the
• Indication, technique, and contraindications technique of PKP exceeds by far the simple
for excimer laser-assisted deep anterior lamel- replacement of two collagen discs and is crucial
lar keratoplasty “excimer-DALK”. for achieving good postoperative visual acuity.
Visual acuity after successful PKP depends
largely on the clarity of the graft and postoperative
astigmatism after suture removal [2]. One of the
most common causes of patient dissatisfaction
after PKP is severe irregular astigmatism [3]. A
clear transplant after PKP with high irregular astig-
matism cannot be considered successful anymore.
The following factors are the most important cause
tains supplementary material available at https://doi.org/ of unsatisfactory postoperative astigmatism [4, 5]:
10.1007/978-3-031-32408-6_11.
1. Decentration of the recipient bed [6]
2. Vertical tilt [7]
L. Daas (*) · L. Hamon · E. Flockerzi · S. Suffo
B. Seitz
3. Horizontal torsion [7]
Department of Ophthalmology, Saarland University 4. Mechanical deformation of the incision edges
Medical Center, Homburg, Germany
e-mail: [email protected]; [email protected]; When performing PKP, every step, from donor
[email protected]; [email protected];
selection to intraoperative trephination and sutur-
[email protected]
https://doi.org/10.1007/978-3-031-32408-6_11
160 L. Daas et al.
ing technique to careful postoperative follow-up, which is the prerequisite for congruent cutting
can affect the final refractive outcome [3, 8]. surfaces and angles in both the donor and recipi-
This includes horizontal positioning of the ent. For this purpose, an artificial anterior cham-
patient’s head and the limbal plane, type of treph- ber should be used for donor trephination. To
ination, the correct placement of the cardinal avoid the increase of astigmatism after suture
sutures, and suture technique [9–13]. removal, decentration, “vertical tilt” and “hori-
Horizontal positioning of the patient’s head zontal torsion”, a trephination for PKP that cre-
and the limbal plane is an indispensable require- ates a tension-free systemic fit of a circular donor
ment to avoid decentration, “vertical tilt” and disc into a circular recipient bed with congruent
“horizontal torsion” (Fig. 11.1). In cases of unproblematic watertight fitting incision edges
doubt, centering should be based on the corneal system should be used for PKP. Furthermore, the
limbus and not on the pupil. An “optimal trephi- noncontact technique avoids inappropriate treph-
nation” requires a full visual inspection, no con- ination in case of instable corneas, such as perfo-
tact, optimal centering of donor, and recipient, rated ulcers, status postcorneal hydrops or
the identical shape of donor and recipient (usu- iatrogenic keratectasia after LASIK (Fig. 11.2).
ally circular), congruent cutting angles, symmet- Currently, these requirements for optimal
rical 360° alignment of donor and recipient, no trephination are best met by nonmechanical
completion of trephination with scissors, and no excimer laser-assisted trephination, which pres-
injury to intraocular structures (iris, lens). Donor ents many advantages in terms of topographic
and recipient trephination should be performed astigmatism and visual acuity after suture
from the epithelial side with the same system, removal [14, 15]. Seitz et al. [15] prospectively
Fig. 11.1 Intraoperative

determinants of a
astigmatism after
penetrating keratoplasty.
(a) “Decentration” =
donor and/or recipient
Decentration
trephination, (b)
“vertical tilt” =
incongruent cutting
angle between donor
and recipient, (c)
“horizontal torsion” =
horizontal discrepancy
between the donor and
recipient form and/or
asymmetrical graft
fitting—“The second
b Vertical tilt
cardinal suture is
crucial!”
Horizontal torsion
11 Excimer Laser-Assisted Keratoplasty: Penetrating Keratoplasty “Excimer-PKP” and Deep Anterior… 161
compared the nonmechanical excimer laser- The type of trephination also has a major impact
assisted trephination and mechanical trephina- on the proper placement of the primary four or
tion using a hand-held motor trephine eight cardinal sutures [11, 16]. For this aim, the
(Micorkeraton, Geuder). They showed that the Homburg/Erlangen technique of nonmechanical
former resulted in a significantly better postop- excimer laser-assisted trephination has a great
erative visual acuity after suture removal (20/28 advantage when compared to other trephination
vs. 20/39, p < 0.01) as well as a significantly systems due to the presence of “orientation teeth
lower surface regularity index (SRI) (0.91 ± 0.45 and notches” along the metal masks. Using a
vs. 1.05 ± 0.46, P = 0.04) after suture removal. curved donor mask on the corneoscleral disc in
an artificial anterior chamber results in eight “ori-
entation teeth” (Fig. 11.3). During the recipient
trephination, eight corresponding “orientation
notches” are lasered on the patient’s cornea with
the help of a recipient mask. The “orientation
teeth” on the edge of the graft [14] and corre-
sponding notches on the edge of the recipient
support the symmetrical positioning of the first
eight cardinal sutures, thus reducing the “hori-
zontal torsion” (Fig. 11.4) and improve the opti-
cal quality after transplantation [5]. Furthermore,
this procedure ensures donor and recipient cen-
tration [17]. These beneficial influences on the
main intraoperative determinants of astigmatism
after keratoplasty result in lower postoperative
keratometric astigmatism (3.1 ± 2.1 vs. 6.2 ± 2.9,
P < 0.001), higher topographic regularity, and
improved spectacle-corrected visual acuity after
suture removal [15, 16, 18]. Additional functions
and benefits of these cardinal sutures include
symmetrical horizontal distribution of the donor
Fig. 11.2 Inappropriate mechanical trephination results tissue in the recipient bed, stabilization of the
in case of an unstable cornea due to tissue irregularities anterior chamber to make sure that additional
Rotating laser beam guided by Helium-neon laser
Donor mask Recipient mask
Donor Recipient
a b
Fig. 11.3 Principle of excimer laser trephination using a 193 nm excimer laser along a metal mask, in the donor (a)
and recipient (b) (schematic sketch, sagittal view)
162 L. Daas et al.
± 3.0 D/4.4 ± 3.1 D/4.0 ± 2.9 D vs. 6.2 ± 2.9

D/7.1 ± 3.2 D/7.4 ± 3.3 D, p ≤ 0.005) [21]. The
technical data of the excimer laser used in our
center are summarized in Table 11.1. The masks
used for the recipient/donor corneal incisions are
Table 11.1 Technical data of the SCHWIND-AMARIS

excimer laser
Working laser ArF-Excimer laser
Type ArF-Excimer laser
Laser class 4
Wavelength 193 nm
Mode Pulsed
Pulse energy (beam 14 mJ max.
Fig. 11.4 The first cardinal suture: “orientation teeth” on output)
the edge of the graft and corresponding notches on the Pulse frequency 500 Hz (AMARIS/AMARIS
edge of the recipient (arrow) 500E)
750 Hz (AMARIS 750S)
suturing is uniform and a good adaptation of the 1050 Hz (AMARIS 1050RS)
Pulse duration 3–15 ns
donor and recipient wound margins to the extent
Pulse-to-pulse <3%
of Bowman’s layer. Plus and minus steps (over- stability
riding or underriding of the donor, respectively) Beam diameter 6 × 3 mm
of the anterior corneal surface of the donor should (output)
be avoided. In contrast, the wound adaptation of Beam divergence 1 × 2 mrad
the posterior corneal surface plays a subordinate Treatment parameter
Energy 0.67–1.0 mJ (nominal)
role. Concerning the correct placement of the
Treatment area App. 193 mm under beam output;
second cardinal suture, unintentional deviations Reference: lower edge of
from circular recipient openings can represent a objective
challenge, even for an experienced keratoplasty Beam diameter 0.54 mm FWHM (full width half
surgeon. (treatment area) maximum)
Fluence (nominal) Low fluence: 250 mJ/cm2
Conventional mechanical trephination of the
High fluence: 500 mJ/cm2
recipient is often associated with a deformation
Aiming laser Diode laser
of corneal tissue layers, distortion of cutting Laser class 1
edges, and irregular cutting surfaces due to the Wavelength 650 nm
axial and radial forces that arise when using Power (middle, <0.3 mW
mechanical trephines [7, 16]. beam output)
Prospective clinical studies have shown that Mode cw (continuous wave)
Laser arm 90° swivelling
the noncontact excimer laser-assisted PKP tech-
Cross laser Diode laser
nique (1) improves donor and recipient trephina- (option)
tion centering, (2) reduces vertical tilt, and (3) (For AMARIS 750S/1050RS only)
horizontal torsion of the graft in the recipient Laser class 1
bed, resulting in significantly lower corneal astig- Wavelength 635 nm
matism after removal of the corneal sutures, Power <0.3 mW
Mode cw
higher regularity of topography, and conse-
OCP laser (option for all AMARIS types)
quently, better vision when glasses are used for Laser class 1
correction [15, 18–20]. Excimer laser-assisted Wavelength 1280–1360 nm
trephination is also superior to femtosecond Energy/power <1 mW
laser-assisted trephination in terms of postopera- Mode cw
tive refractive/pentacam/anterior segment-OCT Source: SCHWIND eye-tech-solutions GmbH Website, user
astigmatism after removal of corneal sutures (4.3 manual for SCHWIND-AMARIS Excimer Laser (QR Code)
determined according to the indication for PKP Surgical Technique

and are available in the following sizes: 6.5/6.6,
7.0/7.1, 7.5/7.6, 8.0/8.1, and 8.5/8.6 mm. The steps of a standardized excimer laser-assisted
In summary, the Homburg/Erlangen technique DALK procedure [27] are as follows:
of nonmechanical excimer laser-assisted trephi-
nation can significantly improve donor and recip- 1. Donor: In equivalence to excimer laser-
ient centration, reduce “vertical tilt” and assisted penetrating keratoplasty (excimer-
“horizontal torsion” of the graft in the recipient PKP): Using excimer laser with mask, a
bed, resulting in significantly less—all-sutures- complete trephination of the donor cornea
out—keratometric astigmatism, more regular with "orientation teeth" is performed [3].
topography, and better visual acuity compared to 2. Recipient cornea: An excimer laser trephina-
conventional trephination [7, 15, 18–20]. tion of the recipient with congruent “notches”
to 80% of midperipheral corneal thickness at
the trephination site is performed. The corneal
xcimer Laser-Assisted Deep
E thickness is measured by anterior segment
Anterior Lamellar Keratoplasty optical coherence tomography (AS-OCT).
“Excimer-DALK” Successful manual excision of this superficial
stromal lamella is followed by intrastromal air
In recent years, an increase in lamellar proce- injection using a 30-gauge needle according
dures in corneal surgery has been observed, espe- to the “big bubble” technique described by
cially in posterior lamellar keratoplasty Anwar and Teichmann [28]. After performing
(“Descemet membrane endothelial keratoplasty” the “big bubble”, a paracentesis is performed,
(DMEK)). According to the German Keratoplasty and air is injected into the anterior chamber.
Registry [22], 3.0% of all 9042 corneal trans- The formation of a kidney-shaped intracam-
plants in 2020 in Germany were performed as eral air bubble indicates the successful separa-
deep anterior lamellar keratoplasty (DALK). tion of the deep stromal tissue and Descemet’s
This low percentage may be explained by a cur- membrane, without perforation. After punc-
rently missing standardized surgical technique to ture and deflation of the “big bubble”, the
perform DALK. intracameral air bubble moves toward the cen-
Especially young patients with advanced ter. If complete exposure of Descemet’s mem-
keratoconus who also suffer from atopic eczema brane (“naked Descemet’s”) is successful, the
and cannot be treated with a rigid, gas-perme- surgical procedure can be continued as
able contact lens, riboflavin-ultraviolet A cross- DALK. Otherwise, a “conversion” to excimer-
linking (CXL), or implantation of intracorneal PKP with full graft is mandatory.
ring segments (ICRS) profit from this method. 3. Donor lamella: The donor cornea is fixed
Prerequisites for performing DALK include an without Descemet’s membrane (“anterior
intact corneal endothelium with sufficient donor lamella”) using two continuous sutures,
endothelial cell density, and an intact according to Hoppenreijs et al. [9] (Fig. 11.5).
Descemet’s membrane (DALK is not an option 4. At the end of surgery, an air/gas bubble
after corneal hydrops) [23–25]. DALK is con- (approximately 80%) is routinely injected
traindicated in endothelial corneal dystrophies into the anterior chamber to ensure attach-
and any form of herpetic keratitis—due to the ment of the patient's own Descemet’s mem-
possible reactivation of herpes in the patient's brane to the donor lamella. This avoids the
own endothelium [26]. development of a so-called “double anterior
164 L. Daas et al.
We believe excimer-DALK contributes to a

certain standardization of deep anterior lamellar
keratoplasty. Furthermore, the “big bubble” rep-
resents a safe and more controlled technique to
finalize tissue dissection after the excision of
about 80% of the superficial stromal lamella,
especially in the absence of an intraoperative
AS-OCT. Complete exposure of Descemet’s
membrane, so-called “naked Descemet’s” with
excision of the predescemetal stromal lamella
[35], is an important prognostic factor with
Fig. 11.5 Slit lamp biomicroscopy photography: 1 day regard to postoperative visual acuity outcome
after excimer laser-assisted DALK (Figs. 11.6 and 11.7). If the residual thickness of
the recipient bed is less than 80 μm and homoge-
chamber”. To prevent a postoperative rise of
intraocular pressure (“air block”), an Nd:YAG
laser iridotomy at the 6 o’clock position
should be routinely created preoperatively in
analogy to the procedure prior to performing
DMEK [29, 30].
In a review paper, the American Academy of

Ophthalmology (AAO) evaluated DALK to be
equivalent to PKP in terms of postoperative
visual acuity outcomes [31]. Endothelial cell loss
after successful DALK (without intraoperative
perforation of Descemet's membrane) (12 months
Fig. 11.6 Intraoperative imaging: complete exposure of
postoperatively: −12.9% ± 17.6%) is lower than
Descemet’s membrane in excimer laser-assisted DALK
after PKP (−27.7% ± 11.1%) [32]. Endothelial after successful “big bubble” dissection followed by man-
immune reaction of the graft is rarely observed ual excision of the superficial stromal lamina, but prior to
after DALK [33]. Another significant advantage excision of the predescemetal stromal lamina (arrow)
over PKP is that the globe remains intact and is
not unroofed as in PKP.
The intraoperative conversion rate to PKP is at
least 16.2%. Therefore, detailed informed con-
sent for both surgeries is absolutely necessary
[34]. However, the conversion rate decreases
with the expertise of the surgeon.
The excimer laser-assisted technique allows
for combining the advantages of DALK (fast
visual recovery and less immune reaction) and
excimer laser trephination (optimal visual recov-
ery with low postoperative astigmatism) with a
low intraoperative perforation rate. Moreover, Fig. 11.7 Intraoperative imaging: complete exposure of
Descemet’s membrane (so-called “naked Descemet’s”) in
excimer-DALK does not result in any disadvan- excimer laser-assisted DALK with successful excision of
tages for the patient in case of conversion to the predescemetal stromal lamella and crystal clear view
excimer-PKP. into the anterior chamber
neous in its thickness, the visual results are 5. Seitz B, Langenbucher A, Naumann GOH. The
shown to be satisfactory. However, postoperative penetrating keratoplasty. A 100-year success story.
Ophthamology. 2005;102:1128–36.
visual acuity development takes longer and may 6. Riedel T, Seitz B, Langenbucher A, Naumann
last between 2 and 5 years according to other GOH. Morphological results after eccentric penetrat-
studies [36]. ing keratoplasty. Ophthalmologe. 2001;98:639–46.
7. Naumann GOH. Part II: corneal transplantation in
anterior segment diseases. The Bowman Lecture
Take Home Notes (Number 56) 1994. Eye. 1995;9:395–421.
• Compared to motor trephination, the 8. Seitz B, Szentmáry N, El-Husseiny M, Viestenz A,
Homburg/Erlangen technique of nonmechani- Langenbucher A, Naumann GOH. The penetrat-
cal excimer laser-assisted trephination can ing keratoplasty (PKP) – a century of success. In:
Hjortdal J, editor. Corneal transplantation. Berlin:
significantly improve donor and recipient cen- Springer; 2016. p. 67–92.
tration, reduce “vertical tilt” and “horizontal 9. Hoppenreijs VPT, Van Rij G, Beekhuis WH, Rijneveld
torsion” of the graft in the recipient bed, and WJ, Rinkel-Van DE. Causes of high astigmatism
results in significantly less all-sutures-out after penetrating keratoplasty. Doc Ophthalmol.
1993;85:21–34.
keratometric astigmatism, more regular topog- 10. Naumann GOH, Sautter H. Surgical procedures on
raphy, and better visual acuity. the cornea. In: Blodi FC, Mackensen G, Neubauer
• The recipient endothelium must be intact for H, editors. Surgical ophthalmology. Berlin: Springer;
DALK to be an option. 1991. p. 433–97.
11. Olson RJ. Modulation of postkeratoplasty astig-
• Contraindications for DALK are acute corneal matism by surgical and suturing techniques. Int
hydrops, any form of herpetic keratitis or Ophthalmol Clin. 1983;23:137–51.
endothelial corneal dystrophy. 12. Van Rij G, Cornell FM, Waring GO III, Wilson LA,
• The surgeon should aim for a complete expo- Beekhuis H. Postoperative astigmatism after cen-
tral vs eccentric penetrating keratoplasties. Am J
sure of Descemet’s membrane (so-called Ophthalmol. 1985;99:317–20.
“naked Descemet’s”) to achieve the best post- 13. Van Rij G, Waring GO. Configuration of corneal tre-
operative visual acuity results. phine opening using five different trephines in human
donor eyes. Arch Ophthalmol. 1988;106:1228–33.
14. Behrens A, Seitz B, Küchle M, Langenbucher A,
Conflict of Interest All authors certify that they have no Kus MM, Rummelt C, Naumann GOH. “Orientation
affiliation or interest in any organization or entity that has teeth” in non-mechanical laser corneal trephina-
a financial or nonfinancial interest in the topics or materi- tion for penetrating keratoplasty: 2.94 μm Er:YAGv
als covered in this chapter. 193 nm ArF excimer laser. Br J Ophthalmol.
1999;83:1008–12.
Funding No funding or support was received for this 15. Seitz B, Langenbucher A, Kus MM, Küchle M,
contribution. Naumann GOH. Nonmechanical corneal trephi-
nation with the excimer laser improves outcome
after penetrating keratoplasty. Ophthalmology.
1999;106:1156–65.
References 16. Seitz B, Langenbucher A, Naumann
GOH. Trephination in penetrating keratoplasty. In:
1. Zirm E. A successful total keratoplasty. Graefes Arch Reinhard T, Larkin F, editors. Essentials in ophthal-
Clin Exp Ophthalmol. 1906;64:580–93. mology - corneal and external eye disease. Berlin:
2. Naydis I, Klemm M, Hassenstein A, Richard Springer; 2006. p. 123–52.
G, Katz T, Linke SJ. Postkeratoplasty astigma- 17. Lang GK, Naumann GOH, Koch JW. A new elliptical
tism. Comparison of three suturing techniques. excision for corneal transplantation using an excimer
Ophthamologe. 2011;108:252–9. laser. Arch Ophthalmol. 1990;108:914–5.
3. Seitz B, Daas L, Milioti G, Szentmáry N, 18. Seitz B, Hager T, Langenbucher A, Naumann
Langenbucher A, Suffo S. Excimer laser-assisted pen- GOH. Reconsidering sequential double running
etrating keratoplasty: on 1 July 2019 excimer laser suture removal after penetrating keratoplasty – a pro-
penetrating keratoplasty celebrates its 30th anniver- spective randomized study comparing excimer laser
sary. Ophthalmologe. 2019;116:1221–30. and motor trephination. Cornea. 2018;37:301–6.
4. Artana LG. Computerized corneal topography in the 19. Seitz B, Langenbucher A, Nguyen NX, Kus MM,
treatment of high astigmatism after penetrating kerato- Küchle M, Naumann GOH. Results of the first 1000
plasty. Klin Monatsbl Augenheilkd. 1995;206:312–6. consecutive elective nonmechanical keratoplasties
166 L. Daas et al.
with the excimer laser - a prospective study over more 29. Cursiefen C, Heindl LM. Perspectives of deep
than 12 years. Ophthalmologe. 2004;101:478–88. anterior lamellar keratoplasty. Ophthalmologe.
20. Szentmáry N, Langenbucher A, Naumann GOH, Seitz 2011;108:833–9.
B. Intra-individual variability of penetrating kerato- 30. Seitz B, Daas L, Flockerzi E, Suffo S. Descemet
plasty outcome after excimer laser versus motorized membrane endothelial keratoplasty DMEK -
corneal trephination. J Refract Surg. 2006;22:804–10. donor and recipient step by step. Ophthalmologe.
21. Tóth G, Szentmáry N, Langenbucher A, Akhmedova 2020;117:811–28.
E, El-Husseiny M, Seitz B. Comparison of excimer 31. Reinhart WJ, Musch DC, Jacobs DS, Lee WB,
laser versus femtosecond laser assisted trephination Kaufman SC, Shtein RM. Deep anterior lamellar ker-
in penetrating keratoplasty: a retrospective study. Adv atoplasty as an alternative to penetrating keratoplasty:
Ther. 2019;36:3471–4382. a report by the American Academy of Ophthalmology.
22. Flockerzi E, Maier P, Böhringer D, Reinshagen H, Ophthalmology. 2011;118:209–18.
Kruse F, Cursiefen C, Reinhard T, Geerling G, Torun 32. Cheng YYY, Visser N, Schouten JS, Wijdh RJ, Pels E,
N, Seitz B. Trends in corneal transplantation from Cleynenbreugel H, Eggink CA, Zaal MJW, Rijneveld
2001 to 2016 in Germany: a report of the DOG- WJ, Nuijts RMMA. Endothelial cell loss and visual
Section cornea and its keratoplasty registry. Am J outcome of deep anterior lamellar keratoplasty versus
Ophthalmol. 2018;188:91–8. penetrating keratoplasty: a randomized multicenter
23. Cursiefen C, Schaub F, Bachmann B. Update: deep clinical trial. Ophthalmology. 2011;118:302–9.
anterior lamellar keratoplasty (DALK) for keratoco- 33. Hos D, Matthaei M, Bock F, Maruyama K, Notara M,
nus. Ophthalmologe. 2016;113:204–12. Clahsen T, Hou Y, Le VNH, Salabarria AC, Horstmann
24. Schaub F, Heindl LM, Enders P, Roters S, Bachmann J, Bachmann BO, Cursiefen C. Immune reactions
BO, Cursiefen C. Deep anterior lamellar keratoplasty. after modern lamellar (DALK, DSAEK, DMEK) ver-
Experiences and results of the first 100 consecutive sus conventional penetrating corneal transplantation.
DALK from the university eye hospital of cologne. Prog Retin Eye Res. 2019;73:100768.
Opthalmology. 2017;114:1019–26. 34. Myerscough J, Friehmann A, Bovone C, Mimouni M,
25. Seitz B, Cursiefen C, El Husseiny M, Viestenz A, Busin M. Evaluation of the risk factors associated with
Langenbucher A, Szentmary N. DALK and pen- conversion of intended deep anterior lamellar kerato-
etrating laser keratoplasty for advanced keratoconus. plasty to penetrating keratoplasty. Br J Ophthalmol.
Ophthalmologe. 2013;110:839–48. 2020;104:764–7.
26. Holbach LM, Asano N, Naumann GO. Infection of 35. Dua HS, Faraj LA, Said DG, Gray T, Lowe
the corneal endothelium in herpes simplex keratitis. J. Human corneal anatomy redefined: a novel pre-
Am J Ophthalmol. 1998;26:592–4. Descemet’s layer (Dua’s layer). Ophthalmology.
27. Daas L, Hamon L, Ardjomand N, Safi T, Seitz 2013;120:1778–85.
B. Excimer laser-assisted DALK: a case report 36. Sarnicola E, Sarnicola C, Cheung AY, Holland EJ,
from the Homburg Keratoconus Center (HKC). Sarnicola V. Surgical corneal anatomy in deep ante-
Ophthalmologe. 2021;118:1245–8. rior lamellar keratoplasty: suggestion of new acro-
28. Anwar M, Teichmann KD. Big-bubble technique to nyms. Cornea. 2019;38:515–22.
bare Descemet’s membrane in anterior lamellar kera-
toplasty. J Cataract Refract Surg. 2002;28:398–403.
Part III
Epithelial Lamellar Keratoplasty
Keratolimbal Grafts: Outcomes,
Innovations and Alternatives
12
Rafael I. Barraquer and Juan Alvarez de Toledo
Key Points Introduction and Historical

• Limbal stem cell deficiency is one of the lead- Background
ing causes of corneal graft failure, with very
difficult clinical and surgical management The tissue-specific, layer-by-layer approach to
• Limbal graft is the only viable way today the treatment of corneal disease characterizes the
available for the treatment of corneal stem cell most relevant trends in keratoplasty during the
deficiency first quarter of our century. Increased awareness
• Different types of corneal limbal graft trans- of the different origins and varying involvement
plants have been proposed with variable of the corneal layers in a particular condition led
success to question the rationale for a penetrating kerato-
• In this chapter, the reader will find a descrip- plasty (PK)—no matter how successful its
tion of the different alternatives today avail- record- and naturally called for the development
able for limbal stem cell transplantation and and application of the progressively dominant
how to choose each, with a description of rel- lamellar techniques.
evant technical surgical details and the postop- However, in the case of the ocular surface, the
erative clinical management essential for the importance of the regeneration dynamics of the
success of the technique epithelium and the role of the limbus had been
recognized at least since the mid-twentieth cen-
tury, long before the concept of “limbal stem cell
Supplementary Information The online version con- deficiency” (LSCD) was formulated. In 1964,
tains supplementary material available at https://doi.org/
10.1007/978-3-031-32408-6_12.
during a discussion at the first World Cornea
Congress [1], José I. Barraquer described the use
“epithelial conjunctivo-corneal limbus taken
R. I. Barraquer (*) from the other eye” for the treatment of s uperficial
Centro de Oftalmologia Barraquer, Barcelona, Spain
burns of a single affected eye. In 1966, Strampelli
International University of Catalonia, et al. published a case of an opaque and highly
Barcelona, Spain
vascularized cornea treated by the transplantation
Institut Universitari Barraquer, Univestitat Autònoma of a complete limbal ring from the fellow eye [2].
de Barcelona, Barcelona, Spain
The following year, Strampelli presented a sec-
J. A. de Toledo ond case and described his technique in more
Institut Universitari Barraquer, Univestitat Autònoma
de Barcelona, Barcelona, Spain
detail at the 2nd International Corneo-Plastic
Conference in London [3]. During the 1970 and
Centro de Oftalmologia Barraquer, Barcelona, Spain
https://doi.org/10.1007/978-3-031-32408-6_12
170 R. I. Barraquer and J. A. de Toledo
1980s, Joaquin Barraquer performed in Barcelona during the 1970s and 1980s [6, 7]. Richard Thoft
similar cases of 360° ring limbal transplants was most likely unaware of this when he tried
(Fig. 12.1) and later published his technique “conjunctival transplantation” for unilateral
(Figs. 12.2 and 12.3) [4]. chemical burns (published in 1977) [8], which
While relevant research on corneal epithelial failed to obtain a functional corneal surface – as
regeneration and homeostasis can be traced back the donor tissue probably did not include
to the 1940s [5], the crucial role of the “limbal CESC. He later described, in 1984, a “kerato-
stem cells”—or, more precisely, corneal epithe- epithelioplasty” procedure [9]—subsequently
lial stem cells (CESC)—was understood only modified by Turgeon et al. in 1990 [10], which
a b
c d
Fig. 12.1 (a) Right eye (RE) of a patient who had a patient affected by a chemical burn, before the autologous
chemical burn in her left eye. This RE was amblyopic due limbal ring transplantation. (c) The immediate postopera-
to childhood unilateral aphakia. The image shows the sta- tive result, 15 days after surgery. (d) One year after a PK
tus of the ocular surface two months after the removal of a rehabilitate the visual function. (Courtesy of Prof. Joaquin
360° ring of limbal conjunctiva. (b) Left eye (LE) of the Barraquer, performed in 1981)
12 Keratolimbal Grafts: Outcomes, Innovations and Alternatives 171
a b
c d
Fig. 12.2 (a) Schematic design of a ring-shaped autolo- viously been subjected to a complete peritomy (b), a
gous limbal transplant according to J. Barraquer. A ring of peripheral keratectomy marked with a superficial 9 mm.
limbal tissue including 5 mm. of limbal conjunctiva and partial trephination (c), and dissection of the anomalous
peripheral corneal epithelium is obtained from the donor tissue at the limbus with a crescent blade (d). (Art by
eye and then anchored in the recipient eye, which has pre- Emilio Iglesias MD, PhD, 1981, from [4])
involved the application of several thin disks or expansion techniques are not available. In 1994,
lenticles from cadaveric peripheral cornea. Tsai and Tseng first described a proper limbal
Although these grafts probably included only a allograft technique using cadaveric donor tissue
scarce quantity of donor CESC, they represent [12]. One year later, Kwitko et al. reported the
the first attempt at using allografts for ocular sur- use of conjunctiva from a living-related donor—
face reconstruction, which would allow treating siblings or other living relatives—for treating
bilateral diseases. bilateral LSCD [13]. Their technique was modi-
In 1989, Kenyon and Tseng were the first to fied by Kenyon and Rapoza to include limbus and
publish a series of limbal transplantations conjunctiva from the living relative, being the
acknowledging the CESC theory [11]. They first description of a living-related conjunctivo-
employed two arcuate segments of conjunctival limbal autograft (lr-CLAL) procedure [14]. This
and peripheral corneal tissue from the fellow eye chapter will review the classical (nontissue engi-
to treat unilateral LSCD. Their technique remains neered) techniques of limbal transplantation,
a standard treatment for most unilateral severe especially the keratolimbal and their outcomes,
ocular surface disease, especially where ex vivo innovations and alternatives.
a b
c d
Fig. 12.3 (a) Complete removal of the fibro-vascular tis- and corneal circumferences. (d) Limbal area starts repop-
sue covers recipient cornea and limbus. (b) Recipient’s ulating the epithelial layer from the stem cells present in
eye is prepared with a bare ocular surface. (c) The ring this area. (Art by Emilio Iglesias MD, PhD, 1981, from
autograft is then anchored with sutures in the conjunctival [4])
lassification of Ocular Surface

C 2. The type of donor: either autologous (fellow
Transplantation Procedures eye, etc.) or homologous (cadaveric, living-
related or living nonrelated).
The transplantation procedures for ocular surface 3. Whether the tissue was transplanted directly
reconstruction were named as they emerged along or previously cultivated (or otherwise tissue
several decades, increasingly since the 1980s. As engineered) ex vivo.
this generated a varied terminology using differ-
ent criteria, which was prone to cause confusion, This classification established a framework
Holland and Schwartz proposed in 1996 to clas- that could eventually incorporate novel tech-
sify these techniques according to the anatomical niques and introduced a series of standard abbre-
location of the donor tissue and whether the viations or acronyms. It favored a better
source was the same individual or not [15]. This understanding of each technique and clarified the
was further developed by a committee from the communication between surgeons, allowing a
Cornea Society into a classification published in more accurate comparison of the outcomes. In
2011 (Table 12.1) [16], which defined the possi- the case of tissue-engineered procedures, despite
ble procedures according to three features: the specific terms and acronyms proposed by this
classification, the general term “cultivated limbal
1. The histological type of the transplanted tis- epithelial transplantation” (CLET) still domi-
sue: conjunctiva, limbus/conjunctiva, limbus/ nates in the literature.
cornea and other mucosal tissues.
Table 12.1 Classification of ocular surface transplantation procedures, according to the Cornea Society 2011 [16]
Procedure Abbreviation Donor Transplanted tissue
Conjunctival transplantation
Conjunctival autograft CAU Fellow eye Conjunctiva
Cadaveric conjunctival allograft c-CAL Cadaveric Conjunctiva
Living related conjunctival allograft Ir-CAL Living relative Conjunctiva
Living nonrelated conjunctival allograft Inr-CAL Living Conjunctiva
non-relative
Limbal transplantation
Conjunctival limbal autograft CLAU Fellow eye Limbus/
conjunctiva
Cadaveric conjunctival limbal allograft c-CLAL Cadaveric Limbus/
conjunctiva
Living related conjunctival limbal allograft Ir-CLAL Living relative Limbus/
conjunctiva
Living nonrelated conjunctival limbal allograft Inr-CIAL Living Limbus/
non-relative conjunctiva
Keratolimbal autograft KLAU Fellow eye Limbus/cornea
Keratolimbal allograft KLAL Cadaveric Limbus/cornea
Other mucosal transplantation
Oral mucosa autograft DMAU Recipient Oral mucosa
Nasal mucosa autograft NMAU Recipient Nasal mucosa
Intestine mucosa autograft IMAU Recipient Intestinal mucosa
Peritoneal mucosa autograft PMAU Recipient Peritoneum
Ex vivo cultivated conjunctival transplantation
Ex vivo cultivated conjunctival autograft EVCAU Recipients eye(s) Conjunctiva
Ex vivo cultivated cadaveric conjunctival allograft EVc-CAL Cadaveric Conjunctiva
Ex vivo cultivated living-related conjunctival allograft EVIr-CAL Living relative Conjunctiva
Ex vivo cultivated living nonrelated conjunctival EVInr-CAL Living Conjunctiva
allograft non-relative
Ex vivo limbal transplantation
Ex vivo cultivated limbal autograft EVLAU Recipients eye(s) Limbus/cornea
Ex vivo cultivated cadaveric limbal allograft EVc-LAL Cadaveric Limbus/cornea
Ex vivo cultivated living-related limbal allograft EVIr-LAL Living relative Limbus/cornea
Ex vivo cultivated living nonrelated limbal allograft EVInr-LAL Living Limbus/cornea
non-relative
Other ex vivo cultivated mucosal transplantation
Ex vivo cultivated oral mucosa autograft EVOMAU Recipient Oral mucosa
Tissue Options important contributor to ocular surface homeo-

stasis. This includes its crucial role in the
Regarding the procedures for ocular surface reha- production of mucins—from the goblet cells—
bilitation, the first feature to be considered is the and of cytokines, among other (patho-)physio-
possible histological components of the graft. logical roles.
The main options include conjunctiva alone or Conjunctival tissue for transplantation can be
reaching to the limbus, peripheral superficial cor- obtained from the same eye, from the patient’s
nea including the limbus, and other mucous fellow eye, or from a donor. It can be harvested
membranes. from either the bulbar or fornix conjunctiva. The
While conjunctiva is commonly framed as latter has been reported as a greater source of
“the invading tissue” in cases of LSCD, healthy conjunctival stem cells [17]. However, bulbar
conjunctiva is nonetheless necessary as it is an conjunctiva—especially the superior quadrant—
is more often used due to easier access and faster graft also contributes to the repopulation of the
healing. In any case, a conjunctival graft will not corneal surface supplying new epithelial cor-
provide CESC if it does not extend to include the neal cells. Although most of these will be late
limbus. transient and postmitotic cells, it has been
The CESC is known to be present around the argued that oligopotent stem cells capable of
limbus at the epithelial basal cell layer. As the generating either conjunctival or corneal phe-
corneal epithelium is continuous with that of the notypes depending on the environment can be
conjunctiva, the precise location of the CESC is found dispersed throughout the entire ocular
not anatomically obvious. Furthermore, their dis- surface, including the cornea [19]. A downside
tribution is not uniform around the cornea, espe- of a keratolimbal graft is the greater impact on
cially in several discrete crypts or “niches” the (living) donor eye as more CESC are
related to Vogt’s palisades [18]. This creates vari- harvested.
ability regarding the actual amount of CESC Among other tissues, oral mucosa is probably
transplanted by a particular technique. the most frequently used for ocular surface recon-
Under the “limbal transplantation” heading in struction. It is commonly applied as a substitute
Table 12.1, we actually find two different modali- for conjunctiva, especially when the latter is not
ties: “conjunctival limbal” and “keratolimbal” available from the same individual due to bilat-
grafts. The former will include a variable amount eral disease, multiple surgeries or when a large
of CESC depending on how close-cropped to the surface graft is required. This includes treatment
cornea it has been harvested, which is surgeon of symblepharon and reconstruction of the con-
dependent. If the tissue is mostly conjunctival, a junctival fornixes, in recalcitrant pterygia and in
conjunctival limbal graft may include a few association with biological keratoprosthesis.
CESC. Conversely, a keratolimbal graft typically Alternatively, other mucous membrane tissues
comprises a superficial layer of the peripheral such as nasal, peritoneal and intestinal (rectal)
cornea up to the limbus, including some conjunc- have been used in some cases. While oral and
tiva and superficial sclera. It will, therefore, con- other mucosa lack CESC and normally show a
tain most of the CESC present in the collected different phenotype, the advantage of being
sector. autologous can be an attractive feature in some
Conjunctival limbal grafts are mainly used for situations, especially as a source for cultivated
the reconstruction of a conjunctival defect adja- grafts [20].
cent to the cornea, as in pterygium or limbal
tumor surgery, or in cases with localized LSCD
in which there is healthy and functional limbus in Donor Options
the wider remaining sectors. On the other hand,
including few CESC in the graft also means A key parameter for the success and long-term
lesser aggression to the donor site. prognosis in ocular surface transplantation is the
The keratolimbal tissue is preferred in cases relationship between donor and recipient, as can
with severe or complete LSCD, as it not only be represented by the degree of histocompatibil-
provides the lacking CESC but the normal sup- ity. The preferred tissue source is—in principle—
porting limbal stroma from the donor site as the autologous, as no immune homograft reaction
well. This probably favors the recreation of the can occur. Unfortunately, this is not an option in
limbal niches, promoting the long-term main- cases with bilateral disease, at least for the same
tenance of CESC. Additionally, this kind of tissue from the fellow eye.
The second-best source is a compatible living corneal—stroma and epithelium—including

donor: a parent or sibling with at least half of the Vogt’s palisades (Fig. 12.4c, d). As discussed
major histocompatibility antigens identical. In above, this has an impact in the amount of CESC
this scenario, HLA types I and II should be deter- harvested.
mined for all the relatives and potential family In CLAU, donor conjunctiva is dissected care-
donors. fully, separating Tenon’s capsule up to its limbal
In cases of nonrelated living or cadaveric insertion, and a superficial sheet of limbal
donors, some degree of tissue matching should (peripheral corneal) epithelium is cut with scis-
be attempted to improve the results. Tissue engi- sors, without including Bowman’s membrane.
neering techniques might offer the possibility of This technique is mainly used for pterygium sur-
modifying the immunogenicity of cultivated gery (Fig. 12.5). However, there is little hard evi-
cells, to decrease the immune rejection to homol- dence on whether the limbal component of these
ogous ex vivo amplified grafted cells. grafts—which may be variable depending on a
particular surgeon’s technique—represents any
significant benefit compared to a standard con-
Limbal Autograft Techniques junctival graft.
In KLAU (Video 12.1), a portion of the periph-
Limbal autografts, including conjunctival limbal eral superficial corneal stroma and the epithelium
and keratolimbal (respectively CLAU and is included in the donor tissue. After dissecting
KLAU) remain the corneal surface rehabilitation the conjunctiva centripetally, a 0.15–0.20 mm
procedures with the largest record of success and groove is performed with a blade at the limbal
best prognosis [21, 22]. Apart from avoiding the sclera; lamellar dissection of the peripheral
risk of immune rejection, autologous tissue gen- superficial corneal stroma is then performed, dis-
erally offers better viability than the cadaveric, as secting approximately 1.0–1.5 mm of the periph-
the latter involves some degree of postmortem eral cornea. This donor tissue will include most
decay, somewhat more traumatic harvesting, and of the CESC niches present at the local Vogt’s
preservation methods that are only partially phys- palisades. In cases with severe unilateral LSCD,
iologic. The living-related sources do not have the KLAU grafts are usually obtained from the
these last limitations but are relatively rare. fellow eye-one or two limbal segments of up to
The donor tissue in CLAU comprises con- 90° width each (Fig. 12.6). For treating a local-
junctiva with some peripheral limbal epithelial ized LSCD, a “translocation of the limbus” tech-
cells (Fig. 12.4a, b), while in KLAU, it includes nique can be used, which corresponds to a KLAU
the limbal conjunctiva and peripheral superficial from the same eye (Fig. 12.7).
a b
c d
Fig. 12.4 (a) Delimitation of the area of a CLAU. Only included. (c, d) KLAU dissection includes peripheral
an epithelial part of the limbal area is removed. (b) superficial corneal stroma, epithelium and superficial lim-
Dissection of the conjunctiva is done by splitting Tenon’s bal sclera to ensure the inclusion of all the niches of the
capsule, which would induce retraction of the graft if CESC
a b
c d
Fig. 12.5 (a) Recurrent pterygium after four surgeries tectomy and CLAU. (d) Slit lamp appearance of the result.
with multiple corneal adherences and inferior-nasal sym- The clear and re-epithelialized cornea shows some thin-
blepharon (detail in b). (c) Result after careful removal of ning of the inferior due to the keratectomy
the recurrent pterygium, symblepharon, superficial kera-
a b
c d
Fig. 12.6 (a) Keratolimbal autograft (KLAU) in a patient staining). (c) Complete epithelial layer with no staining, 6
with a unilateral chemical burn in his RE. Two autografts days after the surgery. (d) Stable ocular surface and
were positioned in the superior and inferior limbus one improvement in corneal transparency 9 years after
day after surgery. (b) Centripetal re-epithelialization from KLAU. Patient’s BCVA reached 0.9 with RGP CL
the limbal autografts at day 2 postoperative (fluorescein
a b
c d
Fig. 12.7 (a) Limbal stem cell deficiency (LSCD) in the (limbal translocation) and grafted in the superior nasal
only eye of a patient after more than 20 intravitreal anti- zone. (c) Two months later, a complete and healthy epithe-
VEGF injections through the upper nasal sclera. LSCD lium covers the cornea. (d) Corneal superficial stroma still
may be due to toxicity from the antiseptic or anaesthetic shows a tenuous opacity, but BCVA and subjective symp-
drugs applied during repeated procedures. (b) Single tomatology improved
sector-KLAU from the inferior quadrant of this only eye
Limbal Allograft Techniques obtained from older donors showed >3% of p63+
cells, considered as the minimum value to predict
In cases with total bilateral LSCD, allogeneic a favorable outcome [23]. Routine hypothermic
limbal transplantation is the best option—short storage in liquid media at 4 °C is generally used
of a keratoprosthesis—to repopulate the to preserve the donor limbal tissue, but novel
affected ocular surface with CESC. Initial methods of preservation like hypothermic air-
series of successful limbal allografts were pub- lifted conditions have demonstrated better main-
lished in the 1990s. Turgeon et al. reported on tenance of the epithelial structure, cell phenotype
13 patients in which Thoft’s technique (kerato- and higher viability of the stem cell pool [24].
epithelioplasty) was performed to stabilize the Limbal allotransplants can be performed in
ocular surface affected with persistent epithe- different modalities. Isolated allografts are usu-
lial defects [10]. Tsai and Tseng reported a ally of the keratolimbal type (KLAL).
series of 16 eyes with several causes of LSCD Conjunctival limbal allografts (CLAL) combine
(chemical burns, Stevens-Johnson syndrome, the disadvantages of a high risk of rejection due
congenital sclerocornea, Terrien’s degeneration to the vascularized tissue and low yield of CESC
and chronic conjunctivitis) in which a limbal and are rarely performed. KLAL can be sectorial
ring-shaped allograft was grafted as a source of or ring-shaped (especially from cadaveric donors)
CESC [12]. as the latter can supply more CESC. All of these
Donor age does not seem to be a critical factor can also be combined with keratoplasty, as dis-
for a successful clinical result. CESC cultures cussed below.
Ring-shaped KLAL is the most widely used

technique to treat bilateral LSCD (Fig. 12.8). A
360° ring of keratolimbal tissue is obtained from
a fresh cadaver eye (Fig. 12.9), with the inner and
outer diameters appropriately marked with dif-
ferent trephines. Ring width and thickness should
be enough to include the whole corneoscleral
limbal zone including the CESC niches.
In the recipient’s eye, a complete limbal perit-
omy and removal of the superficial corneal vascular
pannus and scarring, including any perilimbal fibro-
sis, is mandatory. When the recipient Bowman’s
membrane is absent, the human amniotic mem-
brane (hAM) can be fixated under the limbal
allograft, covering the corneal stroma to promote Fig. 12.8 Schematic representation of a ring-shaped
epithelial repopulation, decrease the stromal inflam- KLAL. After removal of the pannus and fibrovascular tis-
sue in the recipient’s cornea and limbal area, a ring-shaped
mation and inhibit the neovascularization. A stable limbal allograft including limbal conjunctiva, superficial
tear film and good eyelid function are paramount to sclera and cornea is placed in the limbal area and secured
achieving long-term success (Fig. 12.10). with sutures
a b
c d
Fig. 12.9 (a) Harvesting of a ring-shaped KLAL from a stroma. (c) The outer scleral diameter is also marked and
fresh cadaveric eye. Conjunctiva is cut 3–5 mm. from the trephined with a 13–14 mm. trephine. (d) A careful lamel-
limbus and reflected over the corneal surface. (b) The lar dissection is performed to obtain a ring of limbal tissue
inner diameter of the ring is marked with an 8–9 mm. tre- containing all the pool of the donor’s CESC
phine, which penetrates 0.15–0.20 mm in the corneal
a b
c d
Fig. 12.10 (a) Bilateral chemical burn with superficial epithelialization, still with closure lines that are visible
pannus and central corneal leukoma. (b) Ring-shaped with fluorescein in the central cornea. (d) Final result one
KLAL four days after surgery. Only the outer border of year after surgery with a stable epithelium and clear cor-
the graft was sutured with 8-0 vicryl sutures, and no suture nea. The patient is maintained under oral cyclosporine A
was placed at the inner circle. (c) Complete re-
Sectorial KLAL (Fig. 12.11) consists in fix-

ating two arcuate segments of limbal tissue from
a donor (living-related or cadaveric) over the
recipient limbus—typically at the vertical
meridians, after removing the abnormal superfi-
cial corneal and limbal tissue. It is indicated in
cases with total bilateral LSCD with less exten-
sive or without vascularization and scarring of
the ocular surface (Fig. 12.12). hAM can also be
applied under the sectorial grafts to promote
re-epithelialization.
Clear information of the advantages and pos-
sible complications of the different options
should be given to the patients and to the poten-
tial living-related donors, highlighting the bene- Fig. 12.11 Schematic representation of a sectorial
fits of being the source of tissue for their affected KLAL. After the removal of the pannus and fibrovascular
relatives. tissue in the recipient’s cornea and limbal area, two wide
limbal allografts of 90° width are placed at the superior
and inferior limbus
a b
c d
Fig. 12.12 (a) Bilateral LSCD in a patient due to chronic New epithelial centripetal growth from the two limbal
contact lens abuse. Persistent epithelial defects and super- allografts. (d) Complete reepithelialization and recovery
ficial stromal scarring. (b) Sectorial KLAL with two lim- of the corneal transparency observed one year after the
bal grafts placed in the vertical meridian of the limbus. (c) surgery
Combined Techniques DALK or PK is performed following the standard

techniques after all the superficial tissue invading
A combined conjunctival autograft and kerato- the cornea has been removed. Once the corneal
limbal allograft (CLAU + KLAL), sometimes suture is completed, one or two autografts of up to
referred as “the modified Cincinnati procedure” 3 clock hours (90°) each are obtained from the fel-
[25], uses two fragments of recipient’s conjunc- low eye and anchored at the recipient limbus with
tiva obtained from the fellow eye and two sectors monofilament sutures (10-0 nylon or 11-0 polyes-
of a cadaveric donor keratolimbal ring. The con- ter), usually at the vertical sectors – most com-
junctival grafts are placed superior and inferiorly, monly the upper when a single graft is placed.
while the keratolimbal allograft sectors are In bilateral LSCD with central corneal stromal
placed nasally and temporally. opacity and/or substance loss, a ring-shaped
In cases of LSCD associated with opacity and/ KLAL can also be combined with simultaneous
or substance loss affecting the central corneal DALK or PK plus superficial keratectomy
stroma, limbal transplantation (KLAU, CLAU or (Fig. 12.15). This procedure may also be indi-
KLAL) can be combined with keratoplasty—either cated in cases where the depth and removability
DALK (Fig. 12.13) or PK (Fig. 12.14) (Video of the corneal opaque tissue are in doubt, and an
12.2). The addition of CESC will improve ocular isolated KLAL may not provide the rapid visual
surface stability and reduce the risk of neovascular- recovery required (or demanded) by the patient
ization of the corneal graft. In unilateral cases (Fig. 12.16). The inner diameter of the KLAL
where the fellow eye can be the donor (KLAU), a ring is the same as that of the central corneal
a b
c d
Fig. 12.13 (a) Corneal opacity and neovascularization limbal conjunctiva with a CLAU from the fellow eye, a
after previous failed pterygium surgery. (b) Lamellar ker- lamellar corneal graft was sutured in the corneal bed. (d)
atectomy was performed manually until reaching a trans- Result 6 months after the procedure with a stable ocular
parent corneal stromal plane. (c) After reconstructing the surface and a transparent corneal graft
a b
c d
Fig. 12.14 (a) Total superficial corneal neovasculariza- in position with good revascularization. (d) Transparent
tion after a previous conjunctival flap performed to treat corneal graft and stable ocular surface 1 year after the
bacterial keratitis with risk of perforation. (b) Combined procedure
PK with KLAU in the superior limbus. (c) KLAU is fixed
graft. After this has been fixated with eight tem-

porary 10-0 nylon or 11-0 polyester interrupted
sutures, the ring-shaped KLAL is placed on top
of the peripheral recipient cornea, and its outer
perimeter fixated to the sclera with 10-0 nylon or
9-0 vicryl sutures. The inner circle of the KLAL
is secured to the corneal graft with eight addi-
tional monofilament sutures of the same material
used for the temporary fixation, which is then
replaced by other sutures including the three tis-
sues in their bites. Finally, 8-0 or 9-0 vicryl is
used to secure the conjunctiva included in with
the donor tissue and that of the recipient.
Fig. 12.15 Schematic representation of a ring-shaped
KLAL combined with central lamellar or penetrating ker-
A single large diameter DALK or PK, tre-
atoplasty. After the removal of the pannus and fibrovascu- phined eccentrically in the donor (Fig. 12.17),
lar tissue in the recipient’s cornea and limbal area, a will include a sector of the donor’s limbal area
central keratoplasty is performed and fixed with eight with its CESC. This procedure, named “limbo-
temporary monofilament sutures. Then, a ring-shaped
limbal allograft (a) including limbal conjunctiva, superfi-
keratoplasty” by Sundmacher et al. in 1997 [26]
cial sclera, and cornea is placed in the limbal area and is technically less demanding that the previously
fixed to the sclera with eight vicryl sutures at the outer described but has a higher risk of epithelial fail-
edge. The central circle of the ring is sutured to the kera- ure due to the smaller proportion (up to 40%) of
toplasty with eight additional monofilament sutures (b),
while the initial temporary sutures are replaced with eight
the limbal zone included in the graft. With the
additional that include the three tissues recent advances in immunosuppression, better
Fig. 12.16 Combined ring-shaped KLAL with PK in a erative intra-tissular hemorrhage that occurs before re-
case of Stevens-Johnson syndrome. The limits of the connection of the blood microcirculation has been
donor conjunctiva is highlighted (arrows). Typical postop- completed
a b
c d
Fig. 12.17 (a) Limbokeratoplasty of 9 mm of diameter pattern with fluorescein in the superior and central cornea;
performed in the single eye of a congenital aniridia epithelial defects and conjunctival cell staining pattern in
affected patient. (b) Around 40% of the donor limbus is the inferior cornea, demonstrating that LSCD is present in
included in the graft when trephined eccentrically. (c) the inferior limbus. Graft failed 3 years after surgery due
Epithelial irregularities in the inferior cornea are seen to recurrence of LSCD
under retroillumination. (d) Normal corneal epithelium
long-term results have been reported with this and fibrosis of the donor area, even granuloma
procedure in less severe special indications like and symblepharon in the upper fornix can occur
gelatinous drop-like corneal dystrophy [27]. when the donor area is not properly repaired,
especially in individuals predisposed to scarring
or poorly compliant of the postoperative steroid
Postoperative Management, treatment.
Outcomes and Complications After One possible concern in cases of autografts is
Limbal Autografts to induce LSCD in the donor eye, which has been
rarely reported, even after CLAU [28]. Although
In order to achieve an optimal outcome, ocular LSCD has never been described in a healthy
surface reconstruction procedures must be ratio- donor eye when at least half of the limbal circum-
nally planned and staged. The main issues after a ference had been respected, any subclinical
limbal autograft relates to the healing process in LSCD must be ruled out in the potential donor
both the donor and recipient eyes. eye. One case of Mooren’s ulcer has been
For a CLAU, the preferred area to obtain the reported after a CLAU procedure for recurrent
donor tissue is the superior or superior-temporal pterygium [29].
conjunctival quadrant—of either the fellow or the KLAU are as a rule very successful, provided
same eye—due to the protection offered by the some guidelines are strictly followed. Rare and
upper lid in the postoperative period. Scarring relatively minor complications have been
reported after KLAU [30], including infections of been confirmed as beneficial in more recent
the donor site, filamentary keratitis, negative flu- reports [34].
orescein staining and subconjunctival hemor- While oral immunosuppression is a key factor
rhage. Astigmatism can be induced if the corneal in the long-term success of limbal allografts,
stroma is removed too deep or too centrally. It most ophthalmologists not specialized in this
has been documented that the CESC completely field are not familiar with the required protocols
repopulate the donor limbal area within 1 year and their general side effects. Even though these
after the procedure [31]. are not serious in most cases [35], specific knowl-
However, autologous limbal transplantation edge and experience is required for proper man-
(CLAU or KLAU) will not work if the ocular sur- agement. According to the postoperative
face and cornea of the recipient’s eye (including evolution of each patient, these treatments must
the stroma) are severely inflamed, and there is a be frequently adjusted, which requires the par-
tear film deficiency or eyelid malposition. All ticipation of a specialist in immunosuppres-
these factors should be medically or surgically sion—usually an internist—in their monitoring.
corrected in advance. Postoperative treatment to Holland et al. published in 2012 their 10-year
ensure the survival of the CESC and a promote results with a protocol involving two oral immu-
the correct regrowth of a stable corneal epithelial nosuppressants (mycophenolate mofetil and
layer usually includes from a judicious use of ste- tacrolimus) combined with 1 mg/kg oral predni-
roids and unpreserved lubricants to advanced sone, all of which should be started 1 week before
topical treatments such as carboxyl-methyl- surgery. Tacrolimus levels are adjusted to 8–10
glucose polysulfate, autologous serum or growth ng/ml the first month postoperative and to 5–8
factors-enriched plasma drops and amniotic ng/ml at 6 months postoperative. Oral prednisone
membrane extracts. is slowly tapered and discontinued after 3 months
[36]. One year after the procedure, monotherapy
can be considered, and 3 years after surgery, oral
Postoperative Management, medication can be stopped if the ocular surface is
Outcomes and Complications after stable. Other protocols consist in combinations
Limbal Allografts of azathioprine and cyclosporine A with
prednisone.
Limbal allografts are at high risk of immune In Holland’s protocol, absolute contraindica-
rejection due to the vascularity of the limbus, tions for oral immunosuppression include
which negates the immune privilege of the cen- patients with a history of previous malignancy 5
tral cornea, and because their antigen load is years before, nonadherence to a strict clinical or
much larger than that of a standard PK, due to the laboratory follow up or medications, and signifi-
presence of different cell types including cant health issues like diabetes, uncontrolled
Langerhans’. hypertension, renal insufficiency, severe heart
Initial reports on lr-CLAL showed increased diseases or other organ failures. Age over 70 is
graft survival when performed with high HLA also a contraindication, and patients between 60
matching (0-1 mismatches) [32]. An early deep and 70 years are selected for immunosuppression
review of the evidence-based published results of depending on their general health.
limbal transplantations [33] found significantly A further report from the Holland group in
better results with autologous tissue but no differ- 2017 found KLAL to achieve a true ocular sur-
ences between KLAL and lr-CLAL (p = 0.328). face stability in 72.7% of cases with a mean fol-
Patients with Stevens-Johnson syndrome (SJS) low up of 9.1 years, provided the appropriate
and those with concurrent hAM transplantation selection criteria and proper immunosuppression
had poorer prognosis and long-term surface were applied “and the procedure repeated as
improvement. The use of living-related tissue has needed” [37]. Obviously, including this last
option conditions the meaning “long-term suc- 0.3 mm2 of a live limbal fragment including a
cess”, as the actual survival of a particular trans- CESC niche [45]. Depending of the donor source
plant is frequently shorter. (cadaver or live tissue) the growth potential is dif-
The topical measures previously commented ferent, being necessary a larger amount of cadav-
regarding the postoperative management of eric tissue (0.5 mm2) to obtain a similar
autologous limbal grafts also apply to allografts, proliferative rate as with the live tissue.
plus the possible role of topical immunosuppres- A large series of 125 eyes treated with autolo-
sives [38]. gous SLET found an overall success of 76% after
Discontinuation of oral immunosuppression 1.5 years of follow-up, with progressive conjunc-
remains a controversial issue. Acute rejection has tivalization in 18.4% of treated eyes [46]. The
been described [39] in a series of six patients main factors of failure were acid injury, severe
more than 3 years after a KLAL procedure, sug- symblepharon, SLET combined with kerato-
gesting that donor cells are still present, thus at plasty and postoperative loss of the transplants,
risk of acute rejection. Therefore, long-term or which highlights the importance of performing
indefinite immunosuppression should be consid- the procedure in quiet eyes without inflammation
ered despite a good mid-term result. Rare com- and with previously repaired eyelid or conjuncti-
plications of a limbal allograft include the val malposition. Success with allogeneic SLET
possible transmission of a donor infection [40], a has also been reported by the same group [47].
conjunctival neoplasia [41], and even a systemic Since its description, SLET has been applied
malignancy [42]. to a variety of conditions—from chemical inju-
ries to ocular surface tumors, among many oth-
ers, with a success comparable to both CLAU
imple Epithelial Limbal
S and CLET, with the restoration of the corneal
Transplantation (SLET): A Real epithelium in 83% of operations and improve-
Innovation? ment in visual acuity in 69% of reported cases
[48]. Successful SLET has been reported after
In 2012, Sangwan et al. presented a novel surgi- failed CLET for unilateral chronic ocular burns
cal technique for transplanting CESC that they [49]. SLET has been combined with CLAU for
called “Single Epithelial Limbal Transplantation” severe chemical burn [50], with pre-descemetic
(SLET) [43]. This was initially described as an DALK in a case of a massive corneal epibulbar
autograft from the fellow eye limbus and later as dermoid [51], and with PK for keratolysis after
an allograft from a fresh cadaveric coneo-scleral chemical burn [52], or in severe congenital cor-
rim [44]. It involves first fixating an hAM graft neal opacities [53]. Several modifications have
with fibrin glue over the ocular surface – previ- been proposed, including a “mini-SLET” for pte-
ously bared by a superficial keratectomy to rygium surgery [54] or in pediatric cases [55], a
remove the abnormal tissue. A small piece from glueless technique [56], and a SLET variant
the donor limbus is harvested, cut into tiny using autologous fornix conjunctiva for the
fragments and placed over the hAM in a circular explants instead of limbus [57].
fashion – avoiding the visual axis. A layer of SLET has been described as “an ingenious,
fibrin glue is applied to fixate these small tissue low cost and effective technique for limbal stem
“explants”, and finally, a bandage soft contact cell transplantation” [58], and “a paradigm shift
lens is fitted over the cornea. These multiple frag- in limbal transplantation” [59]. It represents an
ments of CESC-containing tissue will originate a in vivo or in situ CESC expansion, which employs
new epithelial multilayer, favored by the known a small piece of the donor limbus, thus protecting
beneficial effects of hAM on cell growth. the donor eye in autologous cases. This would
The minimal amount of tissue needed for extend the indications to those with partial bilat-
assure a correct growth of epithelial cells has eral involvement, as an eye with partially dam-
been established by in vitro studies in about aged limbus—thus not eligible as donor CLAU/
KLAU—could still donate the small biopsy for Conclusion

SLET. Moreover, SLET bypasses the need for
sophisticated and expensive ex vivo cell expan- The management of LSCD is a complex and
sion technology. This may be particularly rele- challenging field, where the many causes and
vant in countries where these are not available. factors involved, together with the heterogeneity
Prior to the description of SLET, Kim et al. of specific situations and the multiple options
presented in 2008 another very ingenious and available, make it difficult to choose the best
somewhat related technique in which the small treatment. As a result, the decision in a particular
limbal biopsy was subjected to “in vivo” expan- case will be highly influenced by the preferences
sion over an hAM placed for two weeks on the of the surgeon and the patient. Nevertheless, a
cornea of a patient’s relative. After this time, the few fundamental factors can be selected in an
hAM with the expanded donor CESC was grafted attempt at establishing a useful and simple deci-
on the patient’s cornea [60]. sion tree, as shown in Fig. 12.18.
Non Inflammatory and inflammatory ecology
Sequential conjuntival
epitheliotomy (SCE)
Partial LSCD
Sector-keratolimbal autografts
(S-KLAU)
Sector-keratolimbal allografts
Limbal Stem Cell Deficiency (LSCD)
(S-KLAL)
Keratolimbal autografts
(KLAU)
Unilateral LSCD
Non Inflammatory etiology (p.e. alca I burn)
Cultivated limbal epithelial
transplantation (CLET)
Keratolimbal allografts (KLAL)

cadaveric or living-related
Bilateral LSCD
Non Inflammatory etiology (p.e. alkali burn)
Cultivated oral mucosal epithelial
transplantation (COMET)
Schirmer Test
>10 mm
Bilateral LSCD Boston

LSCD
Cultivated oral mucosal

Total
Inflammatory etiology (p.e. OCP, SJS) or

keratoprosthesis: BKP epith. transp. (COMET)
conenital (p.e. Aniridia)
SchirmerTest < 5 mm
or keratinization
Osteo-odonto Tibial Osteo

keratoprothesis (OOKP) keratoprosthesis (T-OKP)
Fig. 12.18 Author’s proposed decision tree for the selection of the different surgical procedures for LSCD
In summary, Keratolimbal transplants are a. In partial LSCD, the milder cases might be
established effective and safe techniques for ocu- manageable with medical measures alone
lar surface reconstruction in many conditions or combined with nontransplant interven-
involving moderate to severe LSCD. In unilateral tions such as the sequential conjunctival
cases, KLAU offers the best combination of effi- epitheliectomy of Dua.
cacy, safety, availability and low cost. This is b. Moderate partial LSCD may require a lim-
being challenged by autologous SLET, especially ited CESC transplant, which could be a
in cases with partial bilateral involvement. In sector-KLAU or a mini-SLET (rarely
bilateral noninflammatory total LSCD, the differ- justifying the cost of a CLET or the risks
ent variants of KLAL (preferably living-related if associated with KLAL).
available and with good histocompatibility) and c. In cases with partial LSCD associated with
its combinations with other techniques are pref- a conjunctival defect, the techniques in the
erable unless CLET (or COMET) is available and previous item may be substituted or com-
its cost not an issue. Allogeneic SLET may be bined with sector-CLAU.
another option, as well as a KP. These become 2. In cases of total LSCD, whether the condition
the preferred alternative in bilateral total LSCD is unilateral or bilateral.
of inflammatory origin. a. In unilateral total LSCD (typically after a
An adequate knowledge of the pathology and chemical burn), the options include KLAU,
pathophysiology of the condition, together with CLET and now also autologous SLET.
familiarity with all surgical options and arma- 3. In bilateral total LSCD the main decision fac-
mentarium, are requisites for the selection of the tor might be whether the cause is or not
best procedure. Even techniques like oral mucosa inflammatory.
transplantation, which may have been considered a. Noninflammatory bilateral total LSCD can
old-fashioned at a certain point in time, can be be treated with KLAL, allogeneic CLET,
very useful to solve some complex situations. and possibly with autologous cultivated
Paramount for success is attention to the preop- oral mucosa epithelial transplantation
erative preparation of the ocular environment and (COMET) or allogeneic
adnexa, as well as the postoperative management SLET. Keratoprosthesis (KP, including the
to promote surface stability and avoid complica- simpler, allogeneic tissue haptic as in the
tions. When allografts are required, establishing Boston-KP) becomes an option to
a close collaboration with the specialist in sys- consider.
temic immunosuppression remains crucial, as the 4. In bilateral total LSCD with inflammatory
application and monitoring of the adequate treat- cause (as in SJS and mucous membrane pem-
ment protocol appear to substantially improve the phigoid), the preferred treatment are the dif-
long-term results. ferent types of KP. A fourth decision factor is
As long as the technological complexity, lim- whether there is or not a severe dry eye.
ited availability and high cost of ex vivo cell a. In a relatively preserved wet ocular surface
expansion and regeneration methods hinder their (Schirmer test >10 mm), the best option for
practical application, the classical limbal trans- a total bilateral LSCD with (chronic)
plant techniques will remain a valid option offer- inflammatory cause is a KP (i.e.,
ing excellent results when properly indicated and Boston-KP, although there may be a role
performed. for COMET.
b. In cases with severe dry eye (Schirmer
Take Home Notes <5 mm and/or keratinization 9), the options
1. The initial approach to the selection of the become progressively limited to the autol-
adequate surgical technique is based on the ogous tissue-supported KPs, such as the
observation whether the LSCD is partial or osteo-odonto-KP, the tibia-KP, or a trans-
total. mucosal Boston-KP [61].
References structure and a putative limbal stem cell niche. Br J

Ophthalmol. 2005;89:529–32.
19. Majo F, Rochat A, Nicolas M, et al. Oligopotent stem
1. Barraquer JI. Panel three discussion. In: King JH,
cells are distributed throughout the mammalian ocular
McTigue JW, editors. The world cornea congress
surface. Nature. 2008;456:250–4.
I. Washington: Butterworths; 1965. p. 354.
20. Prabhasawat P, Ekpo P, Uiprasertkul M, et al. Long-
2. Strampelli B, Restivo Manfredi ML. Total kera-
term result of autologous cultivated oral mucosal
tectomy in leukomatous eye associated with auto-
epithelial transplantation for severe ocular surface
graft of a keratoconjunctival ring removed from
disease. Cell Tissue Bank. 2016;17:491–503.
the contralateral eye. Ann Ottalmol Clin Ocul.
21. Daya SM. Conjunctival-limbal autograft. Curr Opin
1966;92:778–86.
Ophthalmol. 2017;28:370–6.
3. Strampelli B. Ring autokeratoplasty. In: Rycroft PV,
22. Kate A, Basu S. A review of the diagnosis and treat-
editor. Corneoplastic surgery. Oxford: Pergamon
ment of limbal stem cell deficiency. Front Med.
Press; 1969. p. 253–75.
2022;25(9):836009.
4. Barraquer J, Rutllán J. Microsurgery of the cornea:
23. Nieto-Nicolau N, Martínez-Conesa EM, Casaroli-
an atlas and textbook. Barcelona: Ediciones Scriba;
Marano RP. Limbal stem cells from aged donors are a
1984. p. 160–2.
suitable source for clinical application. Stem Cells Int.
5. Buschke W, Friedenwald JS, Fleischmann W. Studies
2016;2016:3032128.
on the mitotic activity of the corneal epithelium;
24. Li C, Dong N, Wu H, et al. A novel method for
methods; the effects of colchicine, ether, cocaine and
preservation of human corneal limbal tissue. Invest
ephedrin. Bull Johns Hopkins Hosp. 1943;73:143–67.
Ophthalmol Vis Sci. 2013;54:4041–7.
6. Davanger M, Evensen A. Role of the pericorneal
25. Chan CC, Biber JM, Holland EJ. The modi-
papillary structure in renewal of corneal epithelium.
fied Cincinnati procedure: combined conjunctival
Nature. 1971;229:560–1.
limbal autografts and keratolimbal allografts for
7. Schermer S, Galvin S, Sun TT. Differentiation-related
severe unilateral ocular surface failure. Cornea.
expression of a major 64K corneal keratin in vivo and
2012;31:1264–72.
in culture suggests limbal location of corneal epithe-
26. Sundmacher R, Reinhard T, Althaus C. Homologous
lial stem cells. J Cell Biol. 1986;103:49–62.
central limbo-keratoplasty in limbus stem cell dam-
8. Thoft RA. Conjunctival transplantation. Arch
age. Retrospective study of 3 years’ experience.
Ophthalmol. 1977;95:1425–7.
Ophthalmologe. 1997;94:897–901.
9. Thoft RA. Keratoepithelioplasty. Am J Ophthalmol.
27. Lang SJ, Böhringer D, Reinhard T. Penetrating lim-
1984;97:1–6.
bokeratoplasty for gelatinous dorneal dystrophy. Klin
10. Turgeon PW, Nauhein RC, Roat MI, et al. Indications
Monatsbl Augenheilkd. 2019;236:169–72.
for keratoepitelioplasty. Arch Ophthalmol.
28. Cheung AY, Sarnicola E, Holland EJ. Long-term ocu-
1990;108:33–6.
lar surface stability in conjunctival limbal autograft
11. Kenyon KR, Tseng SCG. Limbal autograft transplan-
donor eyes. Cornea. 2017;36:1031–5.
tation for ocular surface disorders. Ophthalmology.
29. Kim EC, Jun AS, Kim MS, et al. Mooren ulcer occur-
1989;96:709–23.
ring at donor site after contralateral conjunctivol-
12. Tsai RJF, Tseng SCG. Human allograft limbal trans-
imbal autograft for recurrent pterygium. Cornea.
plantation for corneal surface reconstruction. Cornea.
2012;31:1357–8.
1994;13:389–400.
30. Miri A, Said DG, Dua HS. Donor site complications
13. Kwitko S, Raminho D, Barcaro S, et al. Allograft con-
in autolimbal and living-related allolimbal transplan-
junctival transplantation for bilateral ocular surface
tation. Ophthalmology. 2011;118:1265–71.
disorders. Ophthalmology. 1995;102:1020–5.
31. Busin M, Breda C, Bertolin M, et al. Corneal epi-
14. Kenyon KR, Rapoza PA. Limbal allograft transplan-
thelial stem cells repopulate the donor area within
tation for ocular surface disorders. Ophthalmology.
1 year from limbus removal for limbal autograft.
1995;102(suppl):101–2.
Ophthalmology. 2016;123:2481–8.
15. Holland EJ, Schwartz OS. The evolution of epithe-
32. Daya SM, Ilari FA. Living related conjunctiva lim-
lial transplantation for severe ocular surface dis-
bal allograft for the treatment of stem cell deficiency.
ease and a proposed classification system. Cornea.
Ophthalmology. 2001;108:126–33; discussion 133–4.
1996;15:549–56.
33. Health Quality Ontario. Limbal stem cell transplanta-
16. Daya SM, Chan CC, Holland EJ. Cornea Society
tion: an evidence-based analysis. Ont Health Technol
nomenclature for ocular surface rehabilitative proce-
Assess Ser. 2008;8(7):1–58.
dures. Cornea. 2011;30:1115–9.
34. Titiyal JS, Sharma N, Agarwal AK, et al. Live related
17. Wei ZG, Cotsarelis G, Sun TT, et al. Label-retaining
versus cadaveric limbal allograft in limbal stem cell
cells are preferentially located in fornical epithelium:
deficiency. Ocul Immunol Inflamm. 2015;23:232–9.
implications on conjunctival epithelial homeostasis.
35. Krakauer M, Welder JD, Pandya HK, et al. Adverse
Invest Ophthalmol Vis Sci. 1995;36:236–46.
effects of systemic immunosuppression in keratolim-
18. Dua HS, Shanmuganathan VA, Powell-Richards AO,
bal allograft. J Ophthalmol. 2012;2012:576712.
et al. Limbal epithelial crypts: a novel anatomical
36. Holland EJ, Mogilishetty G, Skeens HM, et al. vated limbal epithelial transplantation (CLET) for
Systemic immunosuppression in ocular surface stem unilateral chronic ocular burns. Br J Ophthalmol.
cell transplantation: results of a 10-year experience. 2018;102:1640–5.
Cornea. 2012;31:655–61. 50. Panthier C, Bouvet M, Debellemaniere G, et al.
37. Movahedan A, Cheung AY, Eslani M, et al. Long-term Conjunctival limbal autografting (CLAU) combined
outcomes of ocular surface stem cell allograft trans- with customised simple limbal epithelial transplanta-
plantation. Am J Ophthalmol. 2017;184:97–107. tion (SLET) in a severe corneal chemical burn: Case
38. Pfau B, Kruse FE, Rohrschneider K, et al. Comparison report. Am J Ophthalmol Case Rep. 2020;20:100906.
between local and systemic administration of cyclo- 51. Choudhary DS, Agrawal N, Hada M, et al. Massive
sporin A on the effective level in conjunctiva, aqueous corneal-epibulbar dermoid managed with pre-
humor and serum. Ophthalmologe. 1995;92:833–9. descemetic DALK and SLET. GMS Ophthalmol
39. Eslani M, Haq Z, Movahedan A, et al. Late acute Cases. 2021;11:Doc05. https://doi.org/10.3205/
rejection after allograft limbal stem cell transplanta- oc000178.
tion: evidence for long-term donor survival. Cornea. 52. Kunapuli A, Fernandes M. Successful outcome
2017;36:26–31. of simultaneous allogeneic simple limbal epithe-
40. Cheung AY, Govil A, Friedstrom SR, et al. Probable lial transplantation with therapeutic penetrating
donor-derived cytomegalovirus disease after keratoplasty (PKP) for limbal stem cell deficiency
keratolimbal allograft transplantation. Cornea. and sterile keratolysis after chemical injury. Cornea.
2017;36:1006–8. 2021;40:780–2.
41. Sepsakos L, Cheung AY, Nerad JA, et al. Donor- 53. Showail M, Mireskandari K, Ali A. Simple limbal
derived conjunctival-limbal melanoma after a kerato- epithelial transplantation (SLET) in conjunction with
limbal allograft. Cornea. 2017;36:1415–8. keratoplasty for severe congenital corneal opacities.
42. Miller AK, Young JW, Wilson DJ, et al. Transmission Can J Ophthalmol. 2021;56(3):e78–82.
of donor-derived breast carcinoma as a recurrent mass 54. Hernández-Bogantes E, Amescua G, Navas A, et al.
in a keratolimbal allograft. Cornea. 2017;36:736–9. Minor ipsilateral simple limbal epithelial transplan-
43. Sangwan VS, Basu S, MacNeil S, et al. Simple lim- tation (mini-SLET) for pterygium treatment. Br J
bal epithelial transplantation (SLET): a novel surgical Ophthalmol. 2015;99:1598–600.
technique for the treatment of unilateral limbal stem 55. Pannu A, Sati A, Mishra SK, et al. Innovative tech-
cell deficiency. Br J Ophthalmol. 2012;96:931–4. nique of mini-simple limbal epithelial transplan-
44. Bhalekar S, Basu S, Sangwan VS. Successful man- tation in pediatric patients. Indian J Ophthalmol.
agement of immunological rejection following 2021;69:2222–4.
allogeneic simple limbal epithelial transplantation 56. Malyugin BE, Gerasimov MY, Borzenok SA. Glueless
(SLET) for bilateral ocular burns. BMJ Case Rep. simple limbal epithelial transplantation: the report of
2013;2013:bcr2013009051. the first 2 cases. Cornea. 2020;39:1588–91.
45. Kethiri AR, Basu S, Shukla S, et al. Optimizing the 57. Sakimoto T, Sakimoto A, Yamagami S. Autologous
role of limbal explant size and source in determining transplantation of conjunctiva by modifying simple
the outcomes of limbal transplantation: an in vitro limbal epithelial transplantation for limbal stem cell
study. PLoS One. 2017;12(9):e0185623. deficiency. Jpn J Ophthalmol. 2020;64:54–61.
46. Basu S, Sureka SP, Shanbhag SS, et al. Simple lim- 58. Singh A, Virender, Sangwan VS. Mini-review: regen-
bal epithelial transplantation: long-term clinical erating the corneal epithelium with simple limbal
outcomes in 125 cases of unilateral chronic ocular epithelial transplantation. Front Med. 2021;8:673330.
surface burns. Ophthalmology. 2016;123:1000–10. 59. Vazirani J. Commentary: SLET - A paradigm shift
47. Vasquez-Perez A, Nanavaty MA. Modified allogenic in limbal transplantation. Indian J Ophthalmol.
simple limbal epithelial transplantation followed by 2019;67:1277–8.
keratoplasty as treatment for total timbal stem cell defi- 60. Kim JT, Chun YS, Song KY, et al. The effect of
ciency. Ocul Immunol Inflamm. 2018;26:1189–91. in vivo grown corneal epithelium transplantation on
48. Shanbhag SS, Patel CN, Goyal R, et al. Simple lim- persistent epithelial defects with limbal stem cell defi-
bal epithelial transplantation (SLET): review of ciency. J Korean Med Sci. 2008;23:502–8.
indications, surgical technique, mechanism, out- 61. Camacho L, Soldevila A, de la Paz MF. Transmucosal
comes, limitations, and impact. Indian J Ophthalmol. Boston keratoprosthesis type I in a patient with
2019;67:1265–77. advanced ocular cicatricial pemphigoid. Cornea.
49. Basu S, Mohan S, Bhalekhar S, et al. Simple lim- 2020;39:1563–5.
bal epithelial transplantation (SLET) in failed culti-
Simple Limbal Epithelial
Transplantation
13
Anahita Kate and Sayan Basu

• The chapter will cover the indications and the
process of case selection for simple limbal The functional integrity of the corneal epithelium
epithelial transplantation (SLET). is maintained by the migration and turnover of
• A detailed description of the surgical tech- epithelial cells from the limbal palisades [1, 2].
nique of SLET has been presented. Deficiency of these stem cells or damage to the
• The normal postoperative course with the surrounding microenvironment can result in epi-
modalities for monitoring the outcomes has thelial instability, defects, corneal vasculariza-
been described. tion, and eventual scarring [1, 2]. Management of
• The common complications that can occur this entity usually requires surgical intervention,
after SLET along with the management of the and over the years, several different procedures
same, have been discussed. have emerged which aim to reestablish the cor-
neal epithelium. These include conjunctival lim-
bal autograft /allograft (CLAu/CLAL),
keratolimbal allograft (KLAL), cultivated limbal
epithelial transplantation (CLET), etc. [3–5]. In
conjunctival limbal grafts, a large area of the lim-
Supplementary Information The online version con- bus is harvested, and thus there exists a risk of
tains supplementary material available at https://doi. inducing iatrogenic limbal stem cell deficiency
org/10.1007/978-3-031-32408-6_13.
(LSCD) in the healthy eye [6, 7]. This risk is cir-
cumvented in CLET as only 3–4 mm of limbal
A. Kate biopsy is obtained [8]. However, the surgery is a
Shantilal Shanghvi Cornea Institute, LV Prasad Eye two-stepped procedure and requires extensive
Institute, Vijayawada, India laboratory support and regulatory approval. Thus,
S. Basu (*) to overcome these limitations, Sangwan et al pro-
Shantilal Shanghvi Cornea Institute, LV Prasad Eye posed a single-stage surgery involving in vivo
Institute, Hyderabad, Telangana, India
expansion of corneal epithelial cells from a small
Center for Ocular Regeneration (CORE), LV Prasad harvest of limbal stem cells [9]. This novel proce-
Eye Institute, Hyderabad, Telangana, India
dure, simple limbal epithelial transplantation
Prof. Brien Holden Eye Research Centre, (SLET), has been gaining popularity because of
Champalimaud Translational Centre for Eye
Research, LV Prasad Eye Institute, Hyderabad, India its relatively simple technique and its efficacy in
e-mail: [email protected] restoring the normal corneal epithelial pheno-
https://doi.org/10.1007/978-3-031-32408-6_13
192 A. Kate and S. Basu
type. This chapter will focus on the indications, same eye has not been associated with adverse
surgical steps, complications, and outcomes of outcomes to the donor eye [12].
SLET.
Preoperative Workup
Indications
Case Selection
SLET can be autologous (auSLET) or allogeneic
(alloSLET), depending on the source of the lim- A stepwise approach with meticulous ocular
bal epithelial stem cells (LESCs). In the latter examination to ensure proper case selection is of
case, the LECS can be harvested from a living or utmost importance in order to achieve ideal post-
a cadaveric donor. AuSLET is performed in eyes operative outcomes (Fig. 13.1). Several disorders
with unilateral LSCD while alloSLET is reserved with corneal scarring and vascularization may
for bilateral LSCD. Since ocular burns are the mimic LSCD and identifying the true cases is
most common causes of unilateral and bilateral essential to avoid unwarranted stem cell trans-
LSCD, SLET is most commonly performed in plantation. Ancillary tests such as confocal
eyes with this pathology. Table 13.1 details the microscopy and impression cytology can confirm
indications for SLET. Although SLET is indi- the presence of LSCD by identifying the con-
cated in eyes with established LSCD, the surgery junctival epithelial cells within the cornea [13–
has also been carried out in eyes with acute ocu- 15]. An anterior segment optical coherence
lar burns to promote epithelialization and tomography (AS-OCT) is now a commonly
decrease the inflammation associated with a per- available device and can also be used to differen-
sistent defect [10]. In this scenario, an alloSLET tiate true LSCD from its masquerades. The nor-
is carried out while reserving the autologous tis- mal corneal epithelium is hyporeflective with a
sue for addressing the ensuing LSCD. The con- uniform thickness, and a reversal of this pattern is
current presence of LSCD in eyes with congenital seen in LSCD. Varma et al have described a ratio
corneal opacities has prompted the combination of the epithelial to stromal reflectivity on the
of SLET with corneal transplantation in these AS-OCT line scans and have reported good sen-
eyes, and stable outcomes have been reported sitivity and specificity of this parameter in the
with the same [11]. SLET can also be performed diagnosis of LSCD [16]. The next step in the
in patients with prior failed limbal stem cell evaluation of these cases is the assessment of
transplantation (LSCT) and this includes eyes visual potential, and typically no intervention is
wherein a limbal biopsy has been previously har- carried out in cases where visual recovery is not
vested [12]. Obtaining multiple biopsies from the expected. The last prerequisite for performing
SLET is the presence of a wet ocular surface, and
thus SLET is contraindicated in eyes with aque-
ous deficiency dry eye or a dermalised surface.
Table 13.1 Indications for simple limbal epithelial Evaluation of the stromal thickness on the
transplantation
AS-OCT line scan is also important, as eyes with
Unilateral
thinned-out corneas are at risk for perforation
Ocular chemical burns
Postsurgical (ocular surface neoplasia) during the intraoperative dissection of the pan-
Pterygia nus. Shanbhag et al. have detailed a grading sys-
Bilateral tem based on the preoperative clinical features
Ocular chemical burns that assess the likely prognosis and outcomes fol-
Vernal keratoconjunctivitis
Stevens–Johnson syndrome lowing SLET [17]. Table 13.2 enlists the param-
Mucous membrane pemphigoid eters to help prognosticate the outcomes of SLET
Sjogren’s syndrome (primary/secondary) based on the presenting characteristic.
Congenital corneal opacities
13 Simple Limbal Epithelial Transplantation 193
Fig. 13.1 Algorithmic approach to case selection for confocal microscopy, IC impression cytology, CAG con-
cases suitable for simple limbal epithelial transplantation junctival autograft, CLAu conjunctival limbal autograft,
(SLET). LSCD limbal stem cell deficiency, AS-OCT ante- KPro keratoprosthesis
rior segment optical coherence tomography, IVCM in vivo
Sequence of Surgeries itself. Assessment of the grade of scarring can be

difficult as the fibrovascular pannus obscures the
Most disorders that result in LSCD usually have underlying corneal stroma. The enface infrared
significant ocular comorbidities that must be image of the cornea that is given with the AS-OCT
addressed to reestablish a stable ocular surface line scan can provide insight into corneal clarity
and for visual rehabilitation. Adnexal involve- [18]. This is based on the extent to which the
ment in the form of malposed lids or lashes is structures of the anterior chamber can be visual-
common, and correction of these entities prior to ized through the corneal scarring, and SLET can
SLET is vital. Conjunctival cicatrization, when be performed in isolation if iris details are dis-
present, can not only affect the outcome of the cernible. Cases with significant scarring will
surgery but also hamper the placement of contact require simultaneous or sequential keratoplasty
lenses [17]. The symblephara can be addressed to restore a clear visual axis. However, restraint
before SLET or in conjunction with the same. In must be exercised before deciding to surgically
unilateral cases, a conjunctival graft can be har- intervene, as stromal remodeling with a reduction
vested from the ipsilateral or fellow eye, whereas in the scar density may continue to occur several
in bilateral cases, a mucous membrane graft can years after SLET. With rigid contact lenses, a sig-
be used to cover the bare area. nificant improvement in visual acuity is often
Corneal stromal scarring can occur due to the noted, and keratoplasty can be deferred in these
underlying disease or secondary to the LSCD cases.
Table 13.2 Features prognosticating the outcomes after simple limbal epithelial transplantation
Parameter Excellent Good Fair Poor
Prior Corneal perforation No No Yes Yes
history AMG Yes Yes No No
LK/PK No No Yes Yes
SLET/LSCT No Yes Yes Yes
Multiple surgeries No No Yes Yes
Glaucoma No No No Yes
Clinical Eyelids No No No Yes
features Entropion/ No No No Yes
ectropion No No Yes (good Bell’s) Yes (poor Bell’s)
Irregular margin/ Complete Complete Incomplete Poor blink rate
keratinization
Lagophthalmos
Blink
Conjunctiva Minimal Mild Moderate Severe
Inflammation Grade 0 Grade 1 Grade 2 Grade 3
Symblephara
Ocular wettability Good Good Good Dry ocular
surface
Cornea >400 μ 300–400 μ 200–300 μ <200 or >600 μ
Stromal thickness Anterior chamber Anterior chamber Hazy view of No view of
Clarity details clearly details discerned anterior chamber anterior chamber
visible on infrared on infrared image details on infrared structures
image image
Anterior segment Organized Organized Disorganized Disorganized
AMG amniotic membrane grafting, LK lamellar keratoplasty, PK penetrating keratoplasty, SLET simple limbal epithe-
lial transplant, LSCT limbal stem cell transplant. Adapted from Shanbhag et al. [17]
Presurgical Care Technique
Management of the underlying systemic and ocu- Anesthesia

lar pathology prior to SLET is important to
achieve ideal postoperative outcomes. This General anesthesia is required when SLET is per-
includes control of the ocular allergy in eyes with formed in children. In adults, harvesting the lim-
vernal keratoconjunctivitis, decreasing the sur- bal biopsy can be done under topical anesthesia
face inflammation in eyes with ocular burns, cic- while SLET is performed under a peribulbar
atrizing conjunctivitis, etc. Additionally, systemic block.
immunosuppression may be required periopera-
tively in patients with underlying autoimmune
disorders such as Stevens–Johnson Syndrome, Donor Eye
mucous membrane pemphigoid, etc. The use of
topical brimonidine tartrate 0.15% is recom- In autologous SLET, the limbal graft is usually
mended as it induces localized vasoconstriction obtained from the superior limbus. Caution must
and decreases bleeding intraoperatively [17]. The be exercised while choosing this site, and relying
medication is instilled in both the donor and the solely on limbal pigmentation is not recom-
recipient eyes 10–15 min before commencing the mended as they do not confirm the location of the
surgery. LESC. A subconjunctival bleb is created with
preservative-free lignocaine following which a its basement side up and secured with fibrin glue
limbus-based conjunctival flap is fashioned by (Tisseel Kit, Baxter AG, Vienna, Austria). The
dissecting between the conjunctiva and the edges of the membrane are then tucked under-
Tenon’s layer. This flap extends across one clock neath the free conjunctiva, and the excess tissue
hour (3–4 mm) of the limbus. Further dissection is excised. Care is taken to ensure that there are
of the limbal tissue is carried out using a no redundant or loose folds within the
15-number blade, and care is taken to proceed in membrane.
a horizontal manner to remain in the superficial The limbal biopsy is retrieved with nontoothed
plane of the limbus. The onset of bleeding marks forceps and cut into 6–10 pieces. Triangular sec-
the posterior border of the limbus. The dissection tions are made with the middle part of the blades
is continued until the clear gray cornea is visible. of the scissors to avoid placing tentative cuts. The
The conjunctival tissue is then excised carefully transplants are placed in the mid-periphery and
so as to not leave any remanent tissue abutting concentrically with their epithelial side up, which
the limbal biopsy, which is then harvested flush is identified by their smooth, shiny, and pig-
to the cornea. A nontoothed forceps is used to mented surface. Fibrin glue is used to affix the
handle the limbal tissue to avoid traumatizing the biopsied bits. After the glue has polymerized,
LESC. The biopsied tissue is then placed in a which typically takes around a minute, a bandage
bowl of balanced salt solution until it is utilized. contact lens (BCL) is placed. Any excess glue
In eyes requiring alloSLET, the biopsy is that is present must be carefully removed by
obtained from a donor aged 60 or less, with visi- sharp dissection to avoid displacement of the
ble palisades of Vogt and an intact epithelium. transplants. A suture tarsorrhaphy is carried out
The donor tissue should be utilized within 48 h of in children to protect the transplants from inad-
procurement. The limbal tissue is harvested using vertent trauma (Video 13.1).
a pinch biopsy technique wherein the tissue is
grasped with Lims’s forceps and then excised.
The size of the biopsy is similar to that of auS- Postoperative Care
LET, as a longer biopsy may result in a greater
antigenic load [17]. Both the donor and the recipient eye receive top-
ical antibiotics (moxifloxacin 0.5%) until the
corneal and conjunctival epithelial defects heal.
Recipient Eye The status of healing is monitored at each visit
with fluorescein stain and a BCL is maintained
A 360-degree peritomy is carried out 2–3 mm in the recipient eye until the surface is com-
away from the limbus in the subtenon space. The pletely epithelialized. A tapering dose of topical
dissection is advanced using both blunt and sharp corticosteroids (prednisolone acetate 1%) is
dissection in a similar circumferential pattern administered in conjunction with the antibiotics.
until the entire pannus has been removed. This The steroids are started at a six times/day dose
allows the pannus to be removed in toto and and tapered over 6 weeks in auSLET while a
reduces the intraoperative risk of perforation. maintenance dose of one to two times a day is
This is followed by a blunt tenotomy and removal continued in eyes with alloSLET. The latter
of a frill of Tenon’s tissue from beneath the con- group of patients also requires systemic immu-
junctiva. This allows the surrounding conjunctiva nosuppression to stave off rejection episodes.
to recess and creates space for laying down the This is given in the form of a staggered regimen
human amniotic membrane (hAM). Additionally, of pulse doses of intravenous methylpredniso-
this step aids in creating an area of the bare sclera, lone in isolation or in combination with oral
which will prevent early degradation of the hAM cyclosporine and prednisolone. A standardized
and rapid postoperative conjunctivalization. The protocol for the same has been described
hAM is then placed over the corneal surface with Shanbhag et al. [17].
SLET Modifications Mechanism of Action
Sandwich Technique The corneal epithelial cells arise from within the
limbal harvests and spread circumferentially
Amescua et al. have described a technique around each transplant. This multidirectional
where two layers of the cryopreserved amniotic growth of the epithelial sheet is a significant dif-
membrane are used, and the limbal biopsies are ference between SLET and other in vivo options
sandwiched between these layers [19]. The dou- of stem cell transplantation such as CLAu,
ble layer offers additional protection to the CLAL, and KLAL where the epithelial cells
transplants and is a viable alternative if fresh migrate in a unidirectional and centripetal pat-
AM is not available. Also, in pediatric cases tern. As a result, the center of the cornea is the
where retaining the BCL is a concern, the sand- last area to epithelialize, rendering this area sus-
wich technique can be used to prevent displace- ceptible to healing issues. This is in contrast to
ment of the transplants. However, these SLET where the rate of epithelialization is simi-
membranes are associated with a higher risk of lar in both the center and the periphery of the cor-
detachment and subsequent loss of the limbal nea. Factors such as the age of the donor, number
transplants [17]. of transplants from a single biopsy, and size of
the transplants may also affect the speed of epi-
thelial sheet formation, especially in eyes with
Minor Ipsilateral SLET (Mini-SLET) cadaveric alloSLET [25, 26]. Slower growth rates
from transplants from the same source are attrib-
This technique has been adopted to address pte- uted to intraoperative tissue handling and the use
rygia in eyes where harvesting a conjunctival of excessive fibrin glue [25].
autograft may not be feasible or desirable such The AM acts as a substrate for epithelial cell
as eyes with multiple prior surgeries or those proliferation and helps keep the conjunctival
with glaucoma [20, 21]. In such cases, a mini- cells at bay until complete corneal epithelializa-
SLET is performed where the biopsy is har- tion has occurred. This process typically takes up
vested from the same eye, and the transplants are to 2 weeks though stratification and epithelial
placed over the affected area alone. The rate of thickening may continue to occur in the 3–4
recurrence of mini-SLET is comparable to that weeks after SLET [25, 27]. Long-term retention
of conjunctival autograft [21, 22]. The procedure of the AM with its eventual thinning has been
has also been described in eyes with partial noted in different studies [19, 27, 28]. Confocal
LSCD [23]. microscopy studies following SLET have dem-
onstrated the resorption of the limbal biopsy
fragment after 6 months of the surgery [29].
Glueless SLET However, despite the lack of visible biopsied tis-
sues, the stemness is preserved. This has been
Here the SLET transplants are inserted within demonstrated by immunohistochemistry for stem
stomal pockets created in the donor cornea [24]. cell markers in eyes that underwent keratoplasty
The procedure can be considered in low-resource after SLET, thus underscoring the potential of
settings where the availability of fibrin glue avail- SLET in creating a self-sustaining milieu for epi-
ability is a concern. However, embedding the thelial cell turnover [28, 30]. This aspect has also
LESCs in the intrastromal area may be associated been highlighted by the ability of a post-SLET
with a risk of epithelial ingrowth due to the mis- eye to re-epithelize a large area of defect within a
directed proliferation of the stem cells. time frame similar to that of a normal eye [31].
Outcomes [32]. Histopathological differences have also

been reported in the pattern of wound healing
Monitoring Outcomes when compared to the normal corneal epithelium
[33]. These factors probably account for the
Epithelialization after SLET begins within 48 h changes in the stromal scar density, and the pro-
of the surgery and is completed by the second gressive nature of this change must be considered
postoperative week [25]. However, the healing before planning keratoplasty for visual rehabili-
of the corneal surface following SLET is a com- tation. The use of impression cytology and con-
plex process that extends beyond mere epitheli- focal microscopy has also been described for the
alization of the cornea. Several tools can be used assessment of outcomes after SLET [30, 33]. The
to monitor this course postoperatively, which in presence of pure corneal epithelial phenotype is
turn may help in the early identification and ideal; however, a mixed phenotype with both
prompt management of recurrences of LSCD. conjunctival and corneal epithelial cells can also
The replacement of the hyperreflective thick occur. In such cases, the degree to which con-
conjunctival epithelium by the hyporeflective junctival cells predominate the phenotype often
corneal epithelium can be assessed with the determines the outcomes [33]. Serial monitoring
AS-OCT line scan (Figs. 13.2 and 13.3). The of post-SLET eyes is feasible with confocal
thickness of the corneal epithelium reverts to microscopy, which can reveal the development of
normal within 3 months of the surgery postop- the multilayered corneal epithelium along with a
erative period [32]. transition zone between the corneal and conjunc-
Although the reflectivity patterns of the epi- tival epithelial cells [29]. This device can also be
thelium and stroma approach near normal levels used to identify eyes with partial success as they
in the first postoperative year, remodeling within present with activated nuclei and dysmorphic
these layers continues to occur beyond this period epithelial patterns [29].
a b
c d
Fig. 13.2 (a, c) Preoperative image of a case of unilateral autologous simple limbal epithelial transplantation
total limbal stem cell deficiency with a superior sym- (SLET) showing a stable ocular surface and a clear visual
blepharon. Optical coherence tomography (OCT) line axis. The SLET transplants are also visible (yellow arrow-
scan of the same eye depicting a hyperreflective thickened heads). Postoperative OCT line scan showing a normal
epithelium with a relatively spared underlying corneal hyporeflective corneal epithelium with a compact under-
stroma. (b, d) Clinical photograph of the same eye after lying stroma
a b
c d
Fig. 13.3 (a, c) Preoperative image of a case of total lim- simple limbal epithelial transplantation (SLET) depicting
bal stem cell deficiency with a lateral permanent tarsor- a well-epithelized corneal surface with visible SLET
rhaphy. A hyperreflective pannus is seen on the optical transplants (yellow arrowheads) Reversal of the epithelial
coherence tomography (OCT) line scan of the same eye. reflectivity with a normal hyporeflective pattern of the
No thinning of the underlying corneal stroma is noted. corneal epithelium is seen on the postoperative OCT line
(b, d) Clinical photograph of the same eye after allogeneic scan
Clinical Efficacy modalities of allogeneic LSCT such as CLET,

KLAL, and CLAL [44–46].
Anatomical success following SLET is typically Although the success rate of SLET drops to
defined as the restoration of a well-epithelialized around 71% when performed in pediatric cases
avascular corneal surface. Several studies have of LSCD, it is still higher than the success rate of
determined the anatomic success rate to be CLET in pediatric eyes, which is around 47%
around 80% in cases of both partial and total [28, 47]. The repeatability of this outcome by
LSCD [17, 28, 33–39]. This rate is higher than surgeons of varying experience has also been
the success rate of CLET, which is around 70% demonstrated, highlighting the relative ease of
[40, 41]. On comparing the outcomes of SLET the surgical learning curve [28]. The ability of
with CLAU, the outcomes were found to be simi- SLET to maintain a stable ocular surface ranges
lar, and since the quantity of harvested limbal tis- from 75–80% at the end of the first year after sur-
sue is significantly less in SLET versus CLAU, gery in both adults and children [28, 35]. Good
the former procedure is considered superior to functional outcomes have also been reported
latter [42]. The outcomes of alloSLET are similar after SLET, with a two-line improvement in
to that of auSLET, with a success rate ranging visual in nearly 70% of both adult and pediatric
from 71–83% (Figs. 13.2 and 13.3) [17, 33, 43]. cases [28, 34, 35]. Causes of suboptimal visual
These results were not affected by the source of recovery include the presence of stromal scar-
the donor LESC, i.e., if they were obtained from ring, amblyopia, etc., and hence, careful case
a living-related or a cadaveric donor [17]. selection for SLET is crucial to obtain good
Furthermore, they are comparable to other outcomes.
Approach to Keratoplasty Early Postoperative Complications
Keratoplasty in cases of LSCD is required in eyes Loss of the limbal transplants may occur in the
with significant stromal opacification in order to immediate postoperative period because of
visually rehabilitate them. Depending upon the excess glue, the reverse orientation of the
depth of the stromal scarring, a lamellar (LK) or hAM, loss of BCL, or inadvertent trauma. Free
penetrating keratoplasty (PK) can be planned. edges of the hAM may also cause its displace-
However, every attempt should be made to per- ment along with the transplants. Although most
form an LK as it is associated with lower rejection hematomas that collect beneath the hAM are
rates. Although SLET and PK/LK can be per- self-limiting, they may become large enough to
formed sequentially or in combination, the latter displace the hAM. In such cases, the bleed can
has been consistently associated with poorer out- be released from beneath the hAM with a
comes and a higher risk of failure of SLET [28, 26-gauge needle.
34, 48]. Hence performing the keratoplasty fol-
lowing the SLET is recommended. An additional
benefit of deferring the keratoplasty is that it pro- Late Complications
vides time for stromal remodeling to occur. This
process can decrease scar density, and a signifi- Rejection
cant proportion of these cases have good visual Acute episodes of rejection can occur following
outcomes with rigid contact lenses [49]. Retention alloSLET, especially if the immunosuppression
of clarity of the graft for greater than a year, along is not administered adequately. These cases pres-
with good visual outcomes, has been reported in ent with congestion, epithelial haze, and stippled
keratoplasties performed after SLET [50, 51]. staining of the cornea and often have foci of cel-
lularity around the limbal transplants. A rejection
line can also be seen adjacent to the limbus,
Complications which takes up fluorescein stain [52]. They are
managed by increasing the dose of topical ste-
Intraoperative Complications roids and by giving pulse doses of methylpred-
nisolone [52, 53]. With timely intervention and
1. Perforation of the cornea can occur during appropriate immunosuppression, these episodes
dissection of the pannus. Judicious use of the can typically be reversed.
AS-OCT will help identify the cases which
have a higher risk of the same. Careful Focal Recurrence
removal of the pannus has to be performed in Partial failure of SLET with focal recurrence of
such eyes while reserving the dissection over LSCD can ensue in a small subset of cases who
the thinned-out cornea for the last. undergo the procedure. Several such cases have
2. The amniotic membrane and the transplants symblephara that abut the cornea or extend over
may get displaced while removing the specu- it and are not addressed either prior to or in con-
lum due to the presence of tags of fibrin glue junction with SLET. These cases often require a
between the two. Isolating such attachments conjunctival autograft after the excision of the
and separating them by sharp dissection can fibrotic tissue to ensure optimal outcomes fol-
help prevent this complication. lowing SLET [17].
Primary Failure long-term topical and systemic immunosuppres-

Primary failure of SLET occurs when the surface sion to ensure the viability of the transplants.
fails to stabilize, and recurrence of the LSCD is The surgical technique of SLET has a quick
noted. It is usually secondary to intraoperative learning curve and is not dependent on sophisti-
technique-related issues such as superficial har- cated equipment or laboratory support. This in
vest of LESC or inadvertent trauma to the same. combination with the single-staged nature of the
Other risk factors that can predispose these eyes surgery, has eased the logistics associated with a
to failure of SLET include simultaneous kerato- stem cell transplant and has also reduced the cost
plasty with SLET, causative etiologies such as incurred for the same. These factors have facili-
acid injury, etc. [28, 34, 35]. These eyes typically tated the global adoption of the procedure for the
require a repeat stem cell transplant for their management of stem cell deficiency across dif-
management. ferent etiologies.
Miscellaneous Take Home Notes

Iatrogenic LSCD and pyogenic granulomas can • Simple limbal epithelial transplantation
rarely be observed in the donor eye [28, 34, 38]. (SLET) is a novel and effective technique of
Infective or sterile keratitis, persistent epithelial restoring a stable ocular surface in eyes with
defects leading to thinning, and eventual perfora- limbal stem cell deficiency (LSCD).
tion can occur in the recipient eye [17]. Migration • As the procedure is not dependent on special-
of the limbal transplants into the area of the visual ized infrastructure or surgical instruments, it
axis has also been reported [43]. Excessive pro- can be easily adopted by trained corneal
liferation of the epithelial cells can cause a build- surgeons.
up of the cells over the bandage contact lenses • Ideal cases for SLET include wet eyes with
[54]. And so, long-term retention of BCL must be unilateral LSCD, minimal adnexal stromal
avoided, especially in young patients. involvement, and a fairly clear corneal stroma.
• Systemic immunosuppression is required in
cases of allogeneic SLET to ensure the viabil-
Conclusion ity of the stem cells.
• Failure of SLET due to loss of transplants and
SLET is a simple and novel approach for the focal recurrences are the most common com-
management of LSCD, especially in eyes with plications and can be prevented by meticulous
unilateral disease. Proper case selection is vital to preoperative and intraoperative surgical
ensure optimal outcomes. Ideal candidates for planning.
SLET include unilateral cases of LSCD with wet
eyes, minimal adnexal, and corneal stromal
involvement. By following the established set of References
intraoperative steps, a successful outcome with a
1. Tseng SCG. Concept and application of limbal stem
well-epithelialized avascular corneal surface is cells. Eye. 1989;3:141–57.
ensured in a majority of cases. Preexisting sym- 2. Ordonez P, Di Girolamo N. Limbal epithelial stem
blephara, which are not adequately addressed, cells: role of the niche microenvironment. Stem Cells.
concurrent keratoplasties, and underlying causes 2012;30:100–7.
3. Ganger A, Singh A, Kalaivani M, Gupta N, Vanathi
such as acid injuries are associated with a higher M, Mohanty S, Tandon R. Outcomes of surgical inter-
risk of failure of the surgery. ventions for the treatment of limbal stem cell defi-
Although allogeneic SLET also has stable ciency. Indian J Med Res. 2021;154:51–61.
long-term outcomes, the subgroup of patients 4. Vazirani J, Mariappan I, Ramamurthy S, Fatima
S, Basu S, Sangwan VS. Surgical management of
who are suitable for the procedure is small, bilateral limbal stem cell deficiency. Ocul Surf.
which restricts its widespread utility. It requires 2016;14:350–64.
5. Deng SX, Kruse F, Gomes JAP, et al. Global consen- 19. Amescua G, Atallah M, Nikpoor N, Galor A, Perez
sus on the management of limbal stem cell deficiency. VL. Modified simple limbal epithelial transplantation
Cornea. 2020;39:1291–302. using cryopreserved amniotic membrane for unilat-
6. Miri A, Said DG, Dua HS. Donor site complications eral limbal stem cell deficiency. Am J Ophthalmol.
in autolimbal and living-related allolimbal transplan- 2014;158:469–75.
tation. Ophthalmology. 2011;118:1265–71. 20. Hernández-Bogantes E, Amescua G, Navas A,
7. Baradaran-Rafii A, Eslani M, Jamali H, Karimian F, Garfias Y, Ramirez-Miranda A, Lichtinger A, Graue-
Tailor UA, Djalilian AR. Postoperative c omplications Hernández EO. Minor ipsilateral simple limbal epi-
of conjunctival limbal autograft surgery. Cornea. thelial transplantation (mini-SLET) for pterygium
2012;31:893–9. treatment. Br J Ophthalmol. 2015;99:1598–600.
8. Pellegrini G, Traverso CE, Franzi AT, Zingirian M, 21. Sati A, Banerjee S, Kumar P, Kaushik J, Khera
Cancedda R, De Luca M. Long-term restoration of A. Mini-simple limbal epithelial transplantation ver-
damaged corneal surfaces with autologous cultivated sus conjunctival autograft fixation with fibrin glue
corneal epithelium. Lancet. 1997;349:990–3. after pterygium excision: a randomized controlled
9. Sangwan VS, Basu S, MacNeil S, Balasubramanian trial. Cornea. 2019;38:1345–50.
D. Simple limbal epithelial transplantation (SLET): 22. Jha A, Simba A. Conjunctival autograft versus com-
a novel surgical technique for the treatment of uni- bined amniotic membrane and mini-simple limbal
lateral limbal stem cell deficiency. Br J Ophthalmol. epithelial transplant for primary pterygium excision.
2012;96:931–4. J Ophthalmic Vis Res. 2022;17:4–11.
10. Iyer G, Srinivasan B, Agarwal S, Tarigopula 23. Pannu A, Sati A, Mishra SK, Kumar S, Dhar
A. Outcome of allo simple limbal epithelial transplan- S. Innovative technique of mini-simple limbal epi-
tation (alloSLET) in the early stage of ocular chemi- thelial transplantation in pediatric patients. Indian J
cal injury. Br J Ophthalmol. 2017;101:828–33. Ophthalmol. 2021;69:2222–4.
11. Showail M, Mireskandari K, Ali A. Simple limbal 24. Malyugin BE, Gerasimov MY, Borzenok SA. Glueless
epithelial transplantation (SLET) in conjunction with simple limbal epithelial transplantation: the report of
keratoplasty for severe congenital corneal opacities. the first 2 cases. Cornea. 2020;39:1588–91.
Can J Ophthalmol. 2021;56:e78–82. 25. Mittal V, Jain R, Mittal R. Ocular surface epitheli-
12. Basu S, Mohan S, Bhalekar S, Singh V, Sangwan alization pattern after simple limbal epithelial trans-
V. Simple limbal epithelial transplantation (SLET) plantation: an in vivo observational study. Cornea.
in failed cultivated limbal epithelial transplantation 2015;34:1227–32.
(CLET) for unilateral chronic ocular burns. Br J 26. Kethiri AR, Basu S, Shukla S, Sangwan VS, Singh
Ophthalmol. 2018;102:1640–5. V. Optimizing the role of limbal explant size and source
13. de Araújo AL, da Ricardo S, Sakai VN, de Barros JN, in determining the outcomes of limbal transplanta-
Gomes JAP. Impression cytology and in vivo confo- tion: an in vitro study. PLoS One. 2017;12:e0185623.
cal microscopy in corneas with total limbal stem cell 27. Kate A, Shanbhag SS, Goyal R, Basu S. Serial ante-
deficiency. Arq Bras Oftalmol. 2013;76:305–8. rior segment optical coherence tomography post
14. Banayan N, Georgeon C, Grieve K, Ghoubay D, autologous simple limbal epithelial transplantation.
Baudouin F, Borderie V. In vivo confocal microscopy BMJ Case Rep. 2020;13:e236692.
and optical coherence tomography as innovative tools 28. Basu S, Sureka SP, Shanbhag SS, Kethiri AR, Singh
for the diagnosis of limbal stem cell deficiency. J Fr V, Sangwan VS. Simple limbal epithelial transplanta-
Ophtalmol. 2018;41:e395–406. tion: long-term clinical outcomes in 125 cases of uni-
15. Chidambaranathan GP, Mathews S, Panigrahi AK, lateral chronic ocular surface burns. Ophthalmology.
Mascarenhas J, Prajna NV, Muthukkaruppan V. In 2016;123:1000–10.
vivo confocal microscopic analysis of limbal stroma 29. Pedrotti E, Chierego C, Cozzini T, Merz T, Lagali
in patients with limbal stem cell deficiency. Cornea. N, De Gregorio A, Fasolo A, Bonacci E, Bonetto
2015;34:1478–86. J, Marchini G. In vivo confocal microscopy of
16. Varma S, Shanbhag SS, Donthineni PR, Mishra DK, the corneal- conjunctival transition in the evalua-
Singh V, Basu S. High-resolution optical coherence tion of epithelial renewal after SLET. J Clin Med.
tomography angiography characteristics of limbal 2020;9:3574.
stem cell deficiency. Diagnostics. 2021;11:1130. 30. Prabhasawat P, Chirapapaisan C, Ngowyutagon P,
17. Shanbhag SS, Patel CN, Goyal R, Donthineni PR, et al. Efficacy and outcome of simple limbal epithelial
Singh V, Basu S. Simple limbal epithelial transplan- transplantation for limbal stem cell deficiency veri-
tation (SLET): review of indications, surgical tech- fied by epithelial phenotypes integrated with clinical
nique, mechanism, outcomes, limitations, and impact. evaluation. Ocul Surf. 2021;22:27–37.
Indian J Ophthalmol. 2019;67:1265–77. 31. Kate A, Basu S. Amniotic membrane granuloma in a
18. Kate A, Basu S. Mini-conjunctival autograft com- case of ocular chemical injury: clinical features, his-
bined with deep anterior lamellar keratoplasty for topathology, and outcomes. Cureus. 2021;13:e19171.
chronic sequelae of severe unilateral chemical burn: 32. Kate A, Mudgil T, Basu S. Longitudinal changes in
a case report. Int J Surg Case Rep. 2021;88:106508. corneal epithelial thickness and reflectivity following
simple limbal epithelial transplantation: an optical transplantation (alloSLET) from cadaveric donor eyes
coherence tomography-based study. Curr Eye Res. in patients with persistent corneal epithelial defects.
2021;2021:1–7. Br J Ophthalmol. 2021;105:180–5.
33. Prabhasawat P, Luangaram A, Ekpo P, Lekhanont 44. Shanbhag SS, Saeed HN, Paschalis EI, Chodosh
K, Tangpagasit W, Boonwong C, Inthasin N, J. Keratolimbal allograft for limbal stem cell defi-
Chirapapaisan C. Epithelial analysis of simple lim- ciency after severe corneal chemical injury: a system-
bal epithelial transplantation in limbal stem cell defi- atic review. Br J Ophthalmol. 2018;102:1114–21.
ciency by in vivo confocal microscopy and impression 45. Cheung AY, Eslani M, Kurji KH, Wright E, Sarnicola
cytology. Cell Tissue Bank. 2019;20:95–108. E, Govil A, Holland EJ. Long-term outcomes of
34. Vazirani J, Ali MH, Sharma N, et al. Autologous living-related conjunctival limbal allograft compared
simple limbal epithelial transplantation for unilateral with keratolimbal allograft in patients with limbal
limbal stem cell deficiency: multicentre results. Br J stem cell deficiency. Cornea. 2020;39:980–5.
Ophthalmol. 2016;100:1416–20. 46. Basu S, Fernandez MM, Das S, Gaddipati S,
35. Gupta N, Joshi J, Farooqui JH, Mathur U. Results Vemuganti GK, Sangwan VS. Clinical outcomes of
of simple limbal epithelial transplantation in uni- xeno-free allogeneic cultivated limbal epithelial trans-
lateral ocular surface burn. Indian J Ophthalmol. plantation for bilateral limbal stem cell deficiency. Br
2018;66:45–52. J Ophthalmol. 2012;96:1504–9.
36. Wang Y, Hu X, Yang K, Zhang Y, Deng S, Wang Z, Li 47. Sejpal K, Ali MH, Maddileti S, Basu S, Ramappa
S, Tian L, Jie Y. Clinical outcomes of modified simple M, Kekunnaya R, Vemuganti GK, Sangwan
limbal epithelial transplantation for limbal stem cell VS. Cultivated limbal epithelial transplanta-
deficiency in Chinese population: a retrospective case tion in children with ocular surface burns. JAMA
series. Stem Cell Res Ther. 2021;12:259. Ophthalmol. 2013;131(6):731–6. https://doi.
37. Shanbhag SS, Nikpoor N, Rao Donthineni P, Singh V, org/10.1001/jamaophthalmol.2013.2308.
Chodosh J, Basu S. Autologous limbal stem cell trans- 48. Kunapuli A, Fernandes M. Successful outcome of
plantation: a systematic review of clinical outcomes simultaneous allogeneic simple limbal epithelial
with different surgical techniques. Br J Ophthalmol. transplantation with therapeutic penetrating kera-
2020;104:247–53. toplasty (PKP) for limbal stem cell deficiency and
38. Mittal V, Jain R, Mittal R, Vashist U, Narang sterile keratolysis after chemical injury. Cornea.
P. Successful management of severe unilateral 2021;40:780–2.
chemical burns in children using simple limbal epi- 49. Basu S, Hertsenberg AJ, Funderburgh ML, et al.
thelial transplantation (SLET). Br J Ophthalmol. Human limbal biopsy–derived stromal stem
2016;100:1102–8. cells prevent corneal scarring. Sci Transl Med.
39. Queiroz AG, Barbosa MMO, Santos MS, Barreiro 2014;6:266ra172.
TP, Gomes JÁP. Assessment of surgical outcomes 50. Singh D, Vanathi M, Gupta C, Gupta N, Tandon
of limbal transplantation using simple limbal epi- R. Outcomes of deep anterior lamellar keratoplasty
thelial transplantation technique in patients with following autologous simple limbal epithelial trans-
total unilateral limbal deficiency. Arq Bras Oftalmol. plant in pediatric unilateral severe chemical injury.
2016;79:116–8. Indian J Ophthalmol. 2017;65:217–22.
40. Zhao Y, Ma L. Systematic review and meta-analysis 51. Gupta N, Farooqui JH, Patel N, Mathur U. Early
on transplantation of ex vivo cultivated limbal epithe- results of penetrating keratoplasty in patients with
lial stem cell on amniotic membrane in limbal stem unilateral chemical injury after simple limbal epithe-
cell deficiency. Cornea. 2015;34:592–600. lial transplantation. Cornea. 2018;37(10):1249–54.
41. Mishan MA, Yaseri M, Baradaran-Rafii A, Kanavi https://doi.org/10.1097/ICO.0000000000001681.
MR. Systematic review and meta-analysis investigat- 52. Vazirani J, Basu S, Sangwan VS. Allograft rejection
ing autograft versus allograft cultivated limbal epithe- after living-related simple limbal epithelial transplan-
lial transplantation in limbal stem cell deficiency. Int tation. Indian J Ophthalmol. 2021;69:433–5.
Ophthalmol. 2019;39:2685–96. 53. Bhalekar S, Basu S, Sangwan VS. Successful man-
42. Arora R, Dokania P, Manudhane A, Goyal agement of immunological rejection following
JL. Preliminary results from the comparison of allogeneic simple limbal epithelial transplantation
simple limbal epithelial transplantation with con- (SLET) for bilateral ocular burns. BMJ Case Rep.
junctival limbal autologous transplantation in severe 2013;2013:bcr2013009051.
unilateral chronic ocular burns. Indian J Ophthalmol. 54. Bhalekar S, Sangwan VS, Basu S. Growth of corneal
2017;65:35–40. epithelial cells over in situ therapeutic contact lens
43. Riedl JC, Musayeva A, Wasielica-Poslednik J, Pfeiffer after simple limbal epithelial transplantation (SLET).
N, Gericke A. Allogenic simple limbal epithelial BMJ Case Rep. 2013;2013:bcr2013009113.
Cultivated Limbal Epithelial
Transplantation (CLET)
14
Paolo Rama
Key Points the corneal epithelial turnover, resulting in dam-

• Limbal stem-cell transplantation (LSCT) is age to the corneal epithelial layer, which will ulti-
the surgical procedure indicated for the treat- mately repair due to conjunctival migration onto
ment of limbal stem-cell deficiency (LSCD). the cornea [6] (Fig. 14.1).
• Cultivated limbal epithelial transplantation Limbal stem cells guarantee regular turnover
(CLET) is the latest advanced cell therapy and response to injuries of the corneal epithelium
applied to the treatment of limbal stem-cell through differentiation and migration of cells
deficiency. from the limbal niches to the corneal surface.
• From a small limbal biopsy (1–2 mm2) stem Conjunctival migration, or “conjuctivalization”,
cells of the corneal epithelium can be expanded is a compensatory repair mechanism that protects
in vitro and cultivated on fibrin. the cornea from infection, stromal ulceration,
• CLET is a GMP-validated procedure that has melting and perforation. While it provides a sta-
recently been approved in Europe by the EMA ble and protective superficial layer to the cornea,
Competent Authority and is now available for it is often accompanied by persistent inflamma-
clinical use for the treatment of corneal burns. tion and severe visual impairment. Lamellar and/
or penetrating keratoplasty cannot be used suc-
Introduction
Limbal stem-cell transplantation is the surgical

procedure indicated when the limbus has been
irreversibly damaged [1]. The stem cells of the
corneal epithelium (LSCs) reside in the basal
layer of the limbus [2–5]. In limbal stem-cell
deficiency (LSCD), impairment of the limbal
stem-cell compartment causes a breakdown of
P. Rama (*)
Vita-Salute University, San Raffaele, Milan, Italy
Cornea and Ocular Surface Unit, San Raffaele Fig. 14.1 Severe limbal stem-cell deficiency after chemi-
Hospital, Milano, Italy cal burn with graft failure after penetrating keratoplasty
e-mail: [email protected] with recurrence of conjunctivalization
https://doi.org/10.1007/978-3-031-32408-6_14
204 P. Rama
cessfully in these cases as donor corneal epithe- ultivated Limbal Epithelial

C
lium is replaced by that of the recipient within Transplantation (CLET)
months. In the presence of corneal epithelial
stem-cell compartment deficiency, donor graft Autologous CLET
reepithelialization will not take place, with sub-
sequent epithelial defects and the ultimate recur- To overcome risks for the donor eye, Pellegrini
rence of conjunctivalization, and the risk of et al. [9] proposed to expand limbal stem cells in
rejection and failure (Fig. 14.1). culture to treat LSCD secondary to burns. The pio-
LSCD includes a group of heterogeneous dis- neering work of Rheinwald and Green showed that
eases including congenital abnormalities, it was possible to culture a layer of stratified squa-
acquired diseases such as chemical and thermal mous epithelium with stem cells taken from a
injuries, immunological diseases, toxicity and small skin biopsy to prepare cultivated skin grafts
infections [6]. Such diseases may damage not for the treatment of severe-burn patients [10, 11].
only the limbus but also the eyelids, conjunctiva, The same procedure was used to prepare autolo-
corneal nerves, stroma and lacrimal system. gous grafts of corneal epithelium with stem cells
Ocular surface disease is the most appropriate from a 1–2 mm2 limbal biopsy. Various protocols
term for such a complex disorder. Scrupulous for the cultivation of limbal stem cells for trans-
step-by-step reconstruction should be planned, plantation have been proposed and recently
treating the structures involved separately to pre- reviewed by Shortt et al. and Joe and Yeung,
pare the best recipient bed for limbal stem-cell including methods to extract cells from the biopsy
transplantation. Limbal stem-cell transplantation (mechanical disruption or enzymatic dissociation),
(LSCT) is, therefore, a step in the reconstruction substrates and carriers (fibrin sheet, amniotic
of the ocular surface, while lamellar or penetrat- membrane, polymers, contact lenses, collagen),
ing corneal grafts will ultimately restore corneal mediums with animal-derived components or
transparency, leading to the recovery of visual xeno-free [12, 13]. Although good clinical out-
capacity. comes have been reported with all of these differ-
Eyelid malposition and malocclusion should ent culture procedures, few studies have evaluated
first be treated. Conjunctival symblepharon the clonal characteristics of the cultivated cells and
should be then addressed using the appropriate their proliferative potential. When dealing with
procedures. Once the eyelids and conjunctiva stem-cell-based therapies for diseases involving
have been treated, tear film and inflammation cell-renewing tissue, it should be mandatory to
should be carefully evaluated. The minimum demonstrate the presence, survival, and concentra-
required tear film and the maximum amount of tion of stem cells in culture and in the graft and
inflammation that allows the successful long- validate the procedure under GMP conditions [14,
term survival of the grafted stem cells are not 15]. In February 2015, this therapy was approved
clear. In our previous clinical trials [7, 8], we by the European Medicine Agency (EMA) for the
excluded patients with the Schirmer test below 5 treatment of corneal burns (Holoclar®). Two recent
mm/5 min, but this was arbitrarily chosen, and publications summarize the history of CLET, from
one might suggest that the quality of tears might discovery to clinical approval, including the regu-
be even more important than the quantity. latory aspects [16, 17].
Unfortunately, at present, there is still no valid A pre-requisite for CLET is the presence of a
method for its assessment. In our clinical proto- small area of preserved limbus (2–3 mm), which
col for limbal transplantation, we exclude patients is biopsied, expanded in culture and transplanted
showing severe active inflammation. As for tear onto the LSCD-affected eye.
film, we are still far from having reproducible Ex vivo stem-cell expansion is a complex,
clinical assessment and inflammation grading, time-consuming and expensive procedure, but
with the exception of redness scoring. with several advantages compared with tradi-
14 Cultivated Limbal Epithelial Transplantation (CLET) 205
tional limbal grafting: fewer risks for the donor tem-Cell Expansion in Culture
S
eye, the possibility to treat partial bilateral LSCD, Cells are enzymatically dissociated, character-
and the possibility to re-graft following eventual ized and expanded in vitro on a feeding layer of
failure. lethally irradiated 3T3-J2 cells to a size of
approximately 2.2 cm2 [4, 7, 8]. Limbal biopsies
are processed within 24 h of withdrawal.
Surgical Procedure of CLET Following dissociation with a solution of trypsin
and EDTA, one aliquot of the cell suspension
Biopsy (10%) is plated on a lethally irradiated layer of
A 1–2 mm2 wide, approximately 150–200 μm 3T3-J2 cells for colony-forming efficiency analy-
deep, limbal biopsy is taken from the contralat- sis, while the remaining volume of the cell sus-
eral eye, or from an unaffected portion of the lim- pension (90%) is plated at high density on lethally
bus in partial bilateral cases (Fig. 14.2). irradiated layer of 3T3-J2 cells. When the culture
The procedure can be carried out under topical reaches sub-confluence, cells are again dissoci-
anaesthesia with oxybuprocaine, or para/retro- ated using trypsin, divided into two aliquots and
bulbar anaesthesia with carbocaine or Marcaine cryopreserved. Once the surgery is planned, one
without adrenaline depending on patient collabo- aliquot of cells is thawed and plated on a layer of
ration. The use of topical lidocaine should be lethally irradiated 3T3-J2 cells on a supportive
avoided due to its toxicity. Limbal tissue is nor- fibrin layer. The fibrin disk carrying cultivated
mally harvested in the superior quadrant, cells, 2.2 cm2 in dimension, is packed in sterile
although harvesting can be carried out from any stainless-steel containers with 4 ml of transport
quadrant if necessary. We previously showed that medium, placed in a sterile Petri dish, and
there are no differences in the efficacy of stem- inserted into a polystyrene box for transport.
cell isolation and growth comparing different Once packaged, the graft has a shelf-life of 36 h.
areas of the limbus [4]. The biopsy specimen is The second aliquot of frozen limbal cells culti-
then inserted into a sterile tube containing the vated from the original biopsy, when available
transport medium, and immediately sent to the after having prepared the graft, is kept cryopre-
laboratory where it will be processed within 24 served to be used for a second application, if
hours. Sutures are not required, but we usually required.
use two 10/0 nylon stitches to bring the conjunc-
tiva over the area of the corneal biopsy to reduce Grafting
risks and symptoms. Bandaging is generally not The anaesthesia can be para/retrobulbar, using a
required. long-lasting drug such as naropine to prolong the
a b
Fig. 14.2 2 mm2 limbal biopsy, 160 μm deep, in the healthy fellow eye. Biopsy and grafting were performed twice after
CLET failure for persistent severe inflammation
206 P. Rama
blocking of eye movement after surgery. When cal preservative-free dexamethasone 0.1% three
general anaesthesia is used, an associated para/ times per day for 2 weeks, then reduced to one
retrobulbar injection will help prevent eye move- drop twice daily for 1 week and one drop once
ment after surgery. Lidocaine and adrenaline daily for a further week, and then stopped. The
must not be used due to their toxic effects on the topical corticosteroid can be continued in the
cultivated cells. presence of persistent ocular inflammation.
The surgical procedure is as follows: Topical preservative-free antibiotics are used
only in the presence of epithelial defects.
1. Limbal peritomy a few millimetres outside Eye drops containing benzalkonium chloride
the limbus, with proper coagulation. A should be avoided. Benzalkonium chloride (as
4–5 mm pocket in the bulbar conjunctiva is well as other quaternary ammonium compounds)
created into which the fibrin-cultured epithe- is cytotoxic, and eye drops containing this preser-
lial sheet is inserted. vative might damage the newly regenerated cor-
2. Pannectomy: removal of fibrovascular corneal neal epithelium.
layer of conjunctival origin; try to find the
cleavage level between the pannus and the
cornea to avoid, when possible, keratectomy. Allogeneic CLET
3. Lavage with BSS, whilst checking for an
absence of consistent blood loss that could In total LSCD when the limbus is completely
form blood collections (‘sacks’) under the destroyed in both eyes, limbal tissue taken from a
epithelial graft. deceased donor or from a living relative can be
4. Transfer of the stem-cell graft on fibrin from used. In the literature, contrasting results have
the transport container to a suitable dish. It is been reported on the use of allogeneic keratolim-
best to use the protective film of the adhesive bal grafts, with an overall success rate of 73%
tab from surgical gowns, which is to be kept [18]. Both clinical successes and failures have
sterile; under the microscope, it is possible to been observed in the presence of systemic immu-
recognize the fibrin “nude” side (smooth and nosuppressive therapy [19–21] while positive
translucent) from the cell-seeded side (rough). clinical results have been reported in the absence
It is absolutely crucial to place the fibrin sheet of immunosuppression [22, 23] and/or in the
with the cultivated cells outside and not upside absence of allogeneic cell survival [24, 25]. In
down. The fibrin sheet is allowed to slide onto most cases, however, the interpretation of results
the recipient’s prepared graft area, using BSS has been hampered, either by the lack of a proper
and slight traction with forceps at the edge of genetic evaluation of the presumptive long-term
the graft as required. engraftment of allogeneic limbal grafts or by the
5. The excess of the fibrin sheet is trimmed, and inadequate length of follow-up. In the absence of
the edge is covered with the conjunctiva demonstrated surviving donor cells, a possible
applying 2 or 3 stitches of vicryl or silk 8/0. explanation for clinical success is that patients
6. Close the eyelids with Steri-Strips. with non-total limbal stem-cell deficiency have
been included, and the grafted allogeneic limbal
Post-operative Management cells might have induced modification of the
We prefer systemic treatment for the first 2 weeks microenvironment, and promoted proliferation of
to avoid inadvertent trauma and local toxicity. the patient’s own dormant stem cells, whose prog-
Oral doxycycline 100 mg (or if allergic, amoxi- eny gradually replaces donor cells. While remain-
cillin 500 mg) twice a day for 2 weeks, oral pred- ing in situ in the injured eye, these limbal cells are
nisone 0.5 mg/kg/day for 2 weeks, tapering the evidently unable to generate corneal epithelium,
dose after that to 0.25 mg/kg/day for 1 week and either because of the lack of a suitable microenvi-
0.125 mg/kg/day for 1 week, and then stopped. ronment for multiplication or because of fibrotic
After 2 weeks, topical treatment is started: topi- obstruction to their migration over the cornea.
This would explain the mixed population of donor involved separately, to prepare the best recipient
and recipient corneal cells observed at short-term bed for the cultivated cells.
follow-up. These findings are consistent with The procedure of ex vivo stem-cell expan-
reports showing that clinical improvement sion is crucial and mandatory to demonstrate
observed following allogeneic keratolimbal grafts the presence, survival, and concentration of
does not necessarily correlate with the long-term stem cells in culture and in the graft, and vali-
survival of donor cells [24, 25]. Similarly, cul- date the procedure under GMP conditions. We
tured allogeneic epidermal keratinocytes do not are still dependent on the presence of animal-
engraft permanently but provoke epidermal derived products, such as 3T3 feeder layer and
regeneration in partial-thickness skin burns, pre- fetal calf serum. Even though all these ingredi-
sumably by stimulating residual hair follicle stem ents have been proven to be safe and have been
cells [26]. approved for human use by regulatory agen-
cies, we hope to find a way to be free of them in
the future.
Conclusions We still lack a valid solution for total limbal
stem-cell deficiency cases. Contrasting results
Autologous cultivated limbal epithelial trans- have been reported on the use of allogeneic kera-
plantation is an effective and safe procedure to tolimbal grafts, and in the absence of allogeneic
treat limbal stem-cell deficiency when there is an cell survival, we cannot rely on this treatment for
undamaged, even small, portion (1–2 mm2 are long-term success in total bilateral diseases.
sufficient) of the limbus that will provide donor Future perspectives include: (1) finding other
cells to be expanded in vitro. Unilateral and par- sources of autologous stem cells able to function
tial bilateral limbal deficiency can thus be suc- like the corneal epithelium to treat bilateral lim-
cessfully treated with long-term survival and bal stem-cell deficiency, (2) preparation of a
without the need for systemic immunosuppres- “composite” graft with stem cells seeded with
sion (Figs. 14.3 and 14.4). other cells, such as keratocytes, fibroblasts, mela-
Limbal stem-cell deficiency is part of the nocytes, and/or other cells, on a 3D scaffold that
complex disorder known as ocular surface dis- might reproduce the “niche” where stem cells
ease, and scrupulous step-by-step reconstruction normally reside, (3) improve tear substitutes and/
should be planned, treating the structures or tissue engineering of the lacrimal gland to treat
a b
Fig. 14.3 (a) chemical burn with total corneal “conjunc- successful epithelial regeneration. Lamellar keratoplasty
tivalization” due to severe LSCD; (b) one year after has been planned for stromal scarring
CLET, stable and avascular epithelium demonstrating
208 P. Rama
a b
c d
Fig. 14.4 (a) severe burn with total LSCD; (b, c) one teen years after large (10 mm) lamellar keratoplasty. Very
year after CLET with a regular and avascular epithelial stable epithelium with excellent visual recovery (0.8)
layer with hazy, thinned, and irregular stroma; (d, e) nine-
severe dry eye, (4) more accurate modulation of ated progeny in the human ocular surface. J Cell
Biol. 1999;145(4):769–82. https://doi.org/10.1083/
the inflammatory response before and after jcb.145.4.769. PMID: 10330405.
grafting. 5. Tseng SC. Concept and application of limbal stem
cells. Eye. 1989;3:141–57. https://doi.org/10.1038/
Take Home Notes eye.1989.22. PMID: 2695347.
6. Deng SX, Borderie V, Chan CC, Dana R, Figueiredo
• Limbal stem-cell deficiency is an ocular sur- FC, Gomes JAP, et al. Global consensus on definition,
face disease, and scrupulous step-by-step classification, diagnosis, and staging of limbal stem
reconstruction, treating the structures involved cell deficiency. Cornea. 2019;38(3):364–75. https://
separately, should be planned to prepare the doi.org/10.1097/ICO.0000000000001820. PMID:
30614902.
best recipient bed for limbal stem-cell 7. Rama P, Bonini S, Lambiase A, Golisano O, Paterna P,
transplantation. De Luca M, Pellegrini G. Autologous fibrin-cultured
• Autologous cultivated limbal epithelial trans- limbal stem cells permanently restore the corneal sur-
plantation is an effective and safe procedure face of patients with total limbal stem cell deficiency.
Transplantation. 2001;72(9):1478–85. https://doi.
for the treatment of severe limbal stem-cell org/10.1097/00007890-2 00111150-0 0002.PMID:
deficiency with a long follow-up (over 20 11707733.
years). 8. Rama P, Matuska S, Paganoni G, Spinelli A, De
• It has several advantages over direct limbal Luca M, Pellegrini G. Limbal stem-cell therapy
and long-term corneal regeneration. N Engl J Med.
transplantation: (1) minimum risk for the 2010;363(2):147–55. https://doi.org/10.1056/
donor’s eye, (2) the procedure can be repeated NEJMoa0905955. PMID: 20573916.
due to the reduced size of the biopsy, (3) par- 9. Pellegrini G, Traverso CE, Franzi AT, Zingirian M,
tial bilateral LSCDs can be treated. Cancedda R, De Luca M. Long-term restoration of
damaged corneal surfaces with autologous cultivated
• Conflicting long-term results have been corneal epithelium. Lancet. 1997;349(9057):990–3.
reported on the use of allogeneic cells, and we https://doi.org/10.1016/S0140-6 736(96)11188-0 .
need to find other sources of autologous stem PMID: 9100626.
cells capable of functioning as the corneal epi- 10. Rheinwald JG, Green H. Serial cultivation of strains
of human epidermal keratinocytes: the formation
thelium to treat bilateral totala limbal stem- of keratinizing colonies from single cells. Cell.
cell deficiency. 1975;6(3):331–43. https://doi.org/10.1016/s0092-
8674(75)80001-8. PMID: 1052771.
11. Gallico GG, O’Connor NE, Compton CC, Kehinde
O, Green H. Permanent coverage of large burn
wounds with autologous cultured human epithe-
References lium. N Engl J Med. 1984;311(7):448–51. https://
doi.org/10.1056/NEJM198408163110706. PMID:
1. Deng SX, Kruse F, Gomes JAP, Chan CC, Daya 6379456.
S, Dana R, et al. Global consensus on the man- 12. Shortt AJ, Secker GA, Notara MD, Limb GA, Khaw
agement of limbal stem cell deficiency. Cornea. PT, Tuft SJ, Daniels JT. Transplantation of ex vivo
2020;39(10):1291–302. https://doi.org/10.1097/ cultured limbal epithelial stem cells: a review of
ICO.0000000000002358. PMID: 32639314. techniques and clinical results. Surv Ophthalmol.
2. Schermer A, Galvin S, Sun TT. Differentiation-related 2007;52(5):483–502. https://doi.org/10.1016/j.survo-
expression of a major 64K corneal keratin in vivo and phthal.2007.06.013. PMID: 17719371.
in culture suggests limbal location of corneal epithe- 13. Joe AW, Yeung SN. Concise review: Identification of
lial stem cells. J Cell Biol. 1986;103:49–62. https:// limbals stem cells: classical concepts and new chal-
doi.org/10.1083/jcb.103.1.49. PMID: 2424919. lenges. Stem Cells Transl Med. 2014;3(3):318–22.
3. Cotsarelis G, Cheng SZ, Dong G, Sun TT, Lavker https://doi.org/10.5966/sctm.2013-0137. PMID:
RM. Existence of slow-cycling limbal epithelial 24327757.
basal cells that can be preferentially stimulated to 14. De Luca M, Pellegrini G, Green H. Regeneration
proliferate: implications on epithelial stem cells. of squamous epithelia from stem cells of cultured
Cell. 1989;57:201–9. https://doi.org/10.1016/0092- grafts. Regen Med. 2006;1(1):45–57. https://doi.
8674(89)90958-6. PMID: 2702690. org/10.2217/17460751.1.1.45. PMID: 17465819
4. Pellegrini G, Golisano O, Paterna P, Lambiase Review.
A, Bonini S, Rama P, De Luca M. Location and 15. Pellegrini G, Rama P, De Luca M. Vision from the
clonal analysis of stem cells and their differenti- right stem. Trends Mol Med. 2011;17(1):1–7. https://
210 P. Rama
doi.org/10.1016/j.molmed.2010.10.003. PMID: ex vivo expanded stem cell allograft for ocular surface
21075055. reconstruction. Ophthalmology. 2005;112(3):470–7.
16. Pellegrini G, Rama P, Di Rocco A, Panaras A, De https://doi.org/10.1016/j.ophtha.2004.09.023. PMID:
Luca M. Concise review: Hurdles in a success- 15745776.
ful example of limbal stem cell-based regenerative 26. De Luca M, Albanese E, Bondanza S, Megna M,
medicine. Stem Cells. 2014;32(1):26–34. https://doi. Ugozzoli L, Molina F, et al. Multicentre experi-
org/10.1002/stem.1517. PMID: 24038592. ence in the treatment of burns with autologous and
17. Pellegrini G, Lambiase A, Macaluso C, Pocobelli allogenic cultured epithelium, fresh or preserved
A, Deng S, Cavallini GM, et al. From discovery to in a frozen state. Burns. 1989;15(5):303–9. https://
approval of an advanced therapy medicinal product- doi.org/10.1016/0305-4179(89)90007-7. PMID:
containing stem cells, in the EU. Regen Med. 2686683.
2016;11(4):407–20. https://doi.org/10.2217/rme-
2015-0051. PMID: 27091398.
18. Baylis O, Figueiredo F, Henein C, Lako M, Ahmad
S. 13 years of cultured limbal epithelial cell ther- Suggested Reading
apy: a review of the outcomes. J Cell Biochem.
2011;112(4):993–1002. https://doi.org/10.1002/ Kenyon KR, Tseng SC. Limbal autograft transplantation
jcb.23028. PMID: 21308743 Review. for ocular surface disorders. Ophthalmology.
19. Djalilian AR, Mahesh SP, Koch CA, Nussenblatt RB, 1989;96(5):709–22. https://doi.org/10.1016/s0161-
Shen D, Zhuang Z, Holland EJ, Chan CC. Survival of 6420(89)32833-8. PMID: 2748125.
donor epithelial cells after limbal stem cell transplan- Holland EJ. Epithelial transplantation for the management
tation. Invest Ophthalmol Vis Sci. 2005;46(3):803– of severe ocular surface disease. Trans Am Ophthalmol
7. https://doi.org/10.1167/iovs.04-0575. PMID: Soc. 1996;94:677–743. PMID: 8981714.
15728534. Frucht-Pery J, Siganos CS, Solomon A, Scheman L,
20. Mills RA, Coster DJ, Williams KA. Effect of immu- Brautbar C, Zauberman H. Limbal cell autograft trans-
nosuppression on outcome measures in a model of plantation for severe ocular surface disorders. Graefes
rat limbal transplantation. Invest Ophthalmol Vis Sci. Arch Clin Exp Ophthalmol. 1998;236(8):582–7.
2002;43(3):647–55. PMID: 11867579. https://doi.org/10.1007/s004170050125. PMID:
21. Ilary L, Daya SM. Long-term oucomes of keratolim- 9717653.
bal allografts for the treatment of severe ocular surface Jenkins C, Tuft S, Lui C, Buckley R. Limbal transplanta-
disorders. Ophthalmology. 2002;109(7):1278–84. tion in the management of chronic contact lens-associ-
https://doi.org/10.1016/s0161-6 420(02)01081-3 . ated epitheliopathy. Eye. 1993;7:629–33. https://doi.
PMID: 12093650. org/10.1038/eye.1993.145. PMID: 8287983.
22. Kwitko S, Marinho D, Barcaro S, Bocaccio F, Rymer Sangwan VS, Basu S, Macneil S, et al. Simple limbal epi-
S, Fernandes S, Neumann J. Allograft conjunctival thelial transplantation (SLET): a novel surgical tech-
transplantation for bilateral ocular surface disor- nique for the treatment of unilateral limbal stem cell
ders. Ophthalmology. 1995;102(7):1020–5. https:// deficiency. Br J Ophthalmol. 2012;96(7):931–4.
doi.org/10.1016/s0161-6420(95)30918-9. PMID: https://doi.org/10.1136/bjophthalmol-2011-301164.
9121746. PMID: 22328817.
23. Rao SK, Rajagopal R, Sitalakshmi G, Padmanabhan Sangwan VS, Sharp JAH. Simple limbal epithelial trans-
P. Limbal allografting from related live donors for plantation. Curr Opin Ophthalmol. 2017;28(4):382–6.
corneal surface reconstruction. Ophthalmology. https://doi.org/10.1097/ICU.0000000000000377.
1999;106(4):822–8. https://doi.org/10.1016/S0161- PMID: 28406800 Review.
6420(99)90173-2. PMID: 10201609. Dietrich-Ntoukas T, Hofmann-Rummelt C, Kruse FE,
24. Henderson TR, Coster DJ, Williams KA. The long Schlötzer-Schrehardt U. Comparative analysis of the
term outcome of limbal allografts: the search for sur- basement membrane composition of the human lim-
viving cells. Br J Ophthalmol. 2001;85:604–9. https:// bus epithelium and amniotic membrane epithelium.
doi.org/10.1136/bjo.85.5.604. PMID: 11316725. Cornea. 2012;31(5):564–9. https://doi.org/10.1097/
25. Daya SM, Watson A, Sharpe JR, Giledi O, Rowe A, ICO.0b013e3182254b78. PMID: 22382594.
Martin R, James SE. Outcomes and DNA analysis of
Mesenchymal Stem Cells
for Regeneration of the Ocular
15
Surface
Marina López-Paniagua, Sara Galindo,

Margarita Calonge, Inmaculada Pérez,
José M. Herreras, Ana de la Mata,
and Teresa Nieto-Miguel
Marina López-Paniagua, Sara Galindo, Ana de la Mata

and Teresa Nieto-Miguel contributed equally.
M. López-Paniagua · S. Galindo · M. Calonge CIBER-BBN (Biomedical Research Networking

A. de la Mata Center in Bioengineering, Biomaterials and
IOBA (Institute of Applied Ophthalmobiology), Nanomedicine), Carlos III National Institute of
University of Valladolid, Valladolid, Spain Health, Madrid, Spain
CIBER-BBN (Biomedical Research Networking Castile and Leon Networking Center for Regenerative
Center in Bioengineering, Biomaterials and Medicine and Cell Therapy, Valladolid, Spain
Nanomedicine), Carlos III National Institute of
University Clinic Hospital, Valladolid, Spain
Health, Madrid, Spain
Castile and Leon Networking Center for Regenerative
T. Nieto-Miguel (*)
Medicine and Cell Therapy, Valladolid, Spain
IOBA (Institute of Applied Ophthalmobiology),
e-mail: [email protected]; sgalindor@ioba.
University of Valladolid, Valladolid, Spain
med.uva.es; [email protected];
[email protected] CIBER-BBN (Biomedical Research Networking
Center in Bioengineering, Biomaterials and
I. Pérez
Nanomedicine), Carlos III National Institute of
Health, Madrid, Spain
Department of Cell Biology, Genetics, Histology and
Department of Nursing. Faculty of Nursing,
Pharmacology, University of Valladolid, Valladolid,
Spain
J. M. Herreras
https://doi.org/10.1007/978-3-031-32408-6_15
212 M. López-Paniagua et al.
Abbreviations transplant rejections, or congenital disorders can

disrupt the integrity of the corneal epithelium.
CD Cluster of differentiation This type of loss is an important cause of visual
CK Cytokeratin impairment and blindness that affects millions of
CLET Cultivated limbal epithelial people worldwide [1]. The corneal epithelium
transplantation has an extremely high turnover rate (4–7 days)
DED Dry eye disease that is mediated by the limbal epithelial stem
EVs Extracellular vesicles cells (LESCs) located in the palisades of Vogt
GVHD Graft-versus-host disease within the corneo-scleral limbal niche [2–4].
HLA-DR Human leukocyte antigen-DR LESC deficiency or dysfunction and/or the
IL Interleukin destruction of the niche microenvironment pro-
LESCs Limbal epithelial stem cells duces a condition known as limbal stem cell defi-
LSCD Limbal stem cell deficiency ciency (LSCD). LSCD reduces the regeneration
MSCs Mesenchymal stem cells and repair of the corneal epithelium, and the cor-
oGVHD Ocular graft-versus-host disease neal surface is gradually replaced by conjunctival
TNF-α Tumour necrosis factor alpha epithelium. This process is accompanied by
Treg Regulatory T cells chronic inflammation of the ocular surface,
TSG-6 Tumour necrosis factor-stimulated chronic pain, ulceration, and neovascularization,
gene/protein-6 all of which result in corneal blindness due to the
loss of corneal transparency [5].
At present, among the stem cell-based thera-
pies, cultivated limbal epithelial cell transplanta-
Key Points tion (CLET) is the treatment of choice for LSCD.
• Mesenchymal stem cells (MSCs) have signifi- In unilateral cases of LSCD, treatment by autolo-
cant therapeutic potential to regenerate the gous CLET is possible following acquisition of
ocular surface. limbal tissue from the contralateral healthy eye
• Preclinical evidence demonstrates that MSCs [6–11]. However, bilateral cases of LSCD are
can be used for the treatment of ocular surface more frequent; therefore, it is necessary to use
diseases. allogeneic limbal tissue. Consequently, this
• MSCs have been successfully applied in clini- requires one year of immunosuppression to avoid
cal settings for the treatment of some ocular immune rejection, resulting in an increased risk
surface diseases. of patient morbidity and associated medical costs
• Work must continue to overcome the technical [11]. To avoid this immunosuppression, it is nec-
and scientific challenges that remain unsolved essary to seek either an extraocular autologous
to establish the use of MSCs as a widely source of stem cells or a non-immunogenic allo-
accepted treatment for ocular surface geneic source.
diseases. In recent years, the use of mesenchymal stem
cells (MSCs) has remarkably increased in the
fields of cell therapy and regenerative medicine.
egeneration of the Ocular Surface
R Collectively, these stromal-derived cells retain
by Mesenchymal Stem Cells some intrinsic developmental and differentiation
features after they are derived from a variety of
The integrity of the corneal epithelium is crucial animal and human tissues, including bone mar-
for maintaining corneal transparency and visual row, adipose tissue, dental pulp, umbilical cord,
function. Corneal damage due to different cir- and ocular limbal stroma, among others [12].
cumstances such as chemical or thermal burns, They are defined by their adherence to plastic
eye surgeries, cicatrizing-autoimmune patholo- substrates when cultured in standard conditions
gies, severe dry eye disease (DED), infections, and their multipotent differentiation capacity to
15 Mesenchymal Stem Cells for Regeneration of the Ocular Surface 213
form bone, cartilage, and adipose tissue in vitro. and effective for the regeneration of the ocular
Importantly, the MSCs exhibit expression of a surface.
characteristic set of specific surface antigens,
including positive expression for the cluster of
differentiation (CD) 73, CD90, and CD105 [13]. reclinical Evidence of MSC Efficacy
P
However, they do not express antigens CD34, in Ocular Surface Regeneration
CD45, CD11b or CD14, CD19 or CD79α, or
human leukocyte antigen-DR (HLA-DR) mark- Currently, there are many published preclinical
ers [13]. studies showing the potential restorative effects
Moreover, MSCs present four potential advan- of MSCs for ocular surface pathologies in experi-
tages over LESCs with regard to their utility in mental models [26, 27]. These studies were con-
cell therapy and tissue regeneration. First, acqui- ducted with MSCs obtained from different
sition of MSCs is not restricted to deceased sources such as bone marrow, adipose tissue, lim-
donors or healthy eyes of living donors as they bal stroma, umbilical cord, and others, and they
can be easily obtained from several different liv- were administered by different routes. The most
ing tissues [12]. Second, they can be cultured relevant therapeutic preclinical studies that sup-
in vitro to clinical scales in a short period of time, port the use of MSCs for the treatment of ocular
thus overcoming the limitations of LESCs, which surface diseases are described below.
are difficult to isolate and culture [14, 15]. Third,
the stem cell phenotype is maintained even dur-
ing cryopreservation [16]. Fourth, they are not SCs for the Treatment of LSCD
M
immunogenic; therefore, immunosuppression is and Corneal Epithelial Damage
not necessary after allogeneic transplantation
[17, 18]. CLET is the current treatment of choice among
MSCs have additional advantages over stem cell-based interventions for LSCD. This
LESCs, especially for ocular surface repair. For surgical procedure aims to replace the destroyed
instance, the capacity of MSCs for differentiation limbal stem cell population by an autologous or
following transplantation enables them to allogeneic cell population with full functionality
undergo integration, proliferation, and differenti- [6, 7]. However, this treatment has some limita-
ation in the damaged tissues, and in many cases, tions such as the low availability of donor tissues,
facilitate tissue regeneration [19–21]. MSCs may or the difficulty in culturing the limbal epithelial
also reduce inflammation, apoptosis, and fibrosis cells [11]. Nevertheless, in recent years MSCs
and improve tissue regeneration by activating have been shown to be safe and effective and,
endogenous progenitor cells [22]. MSCs also therefore, good candidates for the treatment of
have immunomodulatory properties that enable LSCD [8, 11].
the regulation of T cells, B-cells, and natural In experimental models of corneal epithelial
killer cells, thus mitigating the secretion of damage and LSCD, transplantation of both bone
inflammatory cytokines [23, 24]. marrow- and adipose tissue-derived MSCs
Considering all, MSCs have emerged as very reduces the clinical signs of LSCD such as neo-
attractive candidates for cell-based therapies in vascularization, corneal opacity, and epithelial
numerous and highly varied clinical applications defects (Fig. 15.1). The cells can be administered
including the treatment of some ocular surface using routes such as sub-conjunctival injection
diseases such as LSCD, DED, or even as a poten- [29–37], topical administration [38, 39], applica-
tial treatment to improve corneal allograft sur- tion of MSC-bearing amniotic membrane [28,
vival [11, 25]. This chapter summarizes the main 40–43] or MSC-bearing biopolymers [44–47], or
existing preclinical and clinical evidence that by intravenous [48–53] and intraperitoneal injec-
currently supports MSC-based therapies as safe tion [51]. MSCs obtained from other cell sources
Fig. 15.1 Histological evaluation of ocular surface tis- Compared to healthy control eyes, untreated LSCD eyes
sues from a rabbit model of total limbal stem cell defi- had fewer epithelial layers, a disorganized corneal epithe-
ciency (LSCD) treated with human adipose tissue-derived lium and stroma, and the presence of inflammatory cells
mesenchymal stem cells (AT-MSCs). Representative (in dark purple) in the stroma of the central cornea.
images of periodic acid-Schiff staining of ocular surface However, LSCD eyes transplanted with AT-MSCs showed
tissues obtained from healthy control eyes, untreated fewer inflammatory cells and less disorganization in the
LSCD eyes, and LSCD eyes 8 weeks after being trans- epithelium and stroma of the central cornea than the
planted with AT-MSCs on amniotic membranes. untreated eyes. (Results from [28])
such as limbal stroma [35, 54, 55] or dental pulp Furthermore, other authors have also shown that
[56] are also able to decrease these clinical signs MSCs have an antioxidant effect on the ocular
in experimental models of LSCD. The preclinical surface of experimental models of corneal burns
data have also demonstrated that transplantation or LSCD [45–47, 49, 51]. Some authors have
of MSCs to treat LSCD does not induce adverse demonstrated migration and engraftment of the
events or toxicological effects, even with xenoge- cells on the ocular surface after topical adminis-
neic transplantation [28, 32, 38, 40, 41, 49, 51, tration [28, 38–40, 42, 56], sub-conjunctival
53, 54, 56, 57]. injection [29, 34, 35, 54], and intravenous injec-
The molecular mechanism(s) of MSC-based tion [48, 50, 52, 58]. However, others did not
tissue restoration is not yet fully understood. observe the presence of MSCs at the area of dam-
However, we do know that the transplanted cells age after topical administration on amniotic
reduce inflammation in the ocular surface of membranes [55], or sub-conjunctival [30, 33,
experimental models of corneal epithelial dam- 37], intravenous, or intraperitoneal injections
age or LSCD, both by decreasing inflammatory [51]. Therefore, the evidence suggests that MSCs
infiltrates [28, 33, 38–40, 43, 57–59] and reduc- can promote therapeutic effects at a distance
ing proinflammatory cytokines such as tumour from the target tissues by releasing trophic
necrosis factor-alpha (TNF-α), IL-6, and IL-1β, factors.
among others [29–31, 34, 37, 53]. In addition, Additionally, some preclinical data showed
some authors have described the tumour necrosis recovery of the differentiated corneal epithelial
factor-stimulated gene/protein-6 (TSG-6) as one cell markers cytokeratin (CK) 3 and CK12 [28,
of the molecules involved in the anti-inflammatory 41, 43, 47, 50, 56, 60] and the limbal epithelial
effect of MSCs in the cornea [29, 37, 51, 53]. stem cell markers p63, CK15, and ATP-binding
cassette sub-family G member 2 [28, 29, 41, 50, 72], modulation of macrophage infiltration [77]
56, 58, 61] in the ocular surface of the MSC- or macrophage maturation [76], decreased proin-
transplanted experimental LSCD models. flammatory factors such as TNF-α [72, 76], IL-1
Although transdifferentiation of MSCs into cor- [72], or IL-6 [76], and/or increased anti-
neal and limbal epithelial cells has not been dem- inflammatory factors such as IL-10 [72, 76] or
onstrated in vivo, MSCs seem to contribute to the epidermal growth factor [72].
recovery of the corneal and limbal phenotype by One of the most severe forms of DED occurs
secreting factors and helping resident stem cells. in the context of chronic graft-versus-host dis-
ease (GVHD) that can develop after allogeneic
haematopoietic stem cell transplantation, appear-
MSCs for the Treatment of DED ing in 60% of patients [78]. GVHD with ocular
damage (oGVHD) is caused by the immune
DED is a multifactorial and inflammatory-based response produced by the immunocompetent
pathology [62] that affects between 5.5% and cells from the donor graft that “attack” the recipi-
35% of the world population [63]. It presents ent ocular surface (conjunctiva, cornea, limbus,
with varying severity of symptoms such as pain and tear film) and all of the glands that produce
and blurred vision, and the most severe cases can tear components. This attack produces chronic
lead to corneal ulcers, infections, and even perfo- ocular inflammation and ocular tissue destruction
rations [64, 65]. DED is also characterized by an [79–83].
increase of inflammatory molecules and reactive Because of the high immunoregulatory and
oxygen species and by a decrease of anti- immunosuppressive capacity and the ocular anti-
inflammatory and growth factors in the ocular inflammatory and ocular tissue regenerative
surface [66, 67]. potential of MSCs, they have been successfully
In this context, MSCs have been proposed as a tested as therapy in vivo models of DED associ-
possible treatment for patients affected by the ated with oGVHD [83–86]. Sub-conjunctival
most severe forms of DED. MSCs isolated from injection of bone marrow-derived MSCs in a
bone marrow [68–72], adipose tissue [73–75], or mouse model of GVHD decreased both the pres-
umbilical cord [76] have been therapeutically ence of CD3+ T cells in corneal tissues and cor-
administered in experimental in vivo DED mod- neal keratinization [84, 85]. In addition, other
els using different routes of delivery such as topi- authors showed that for mice with GHVD, MSCs
cal application through eye drops [69], intraorbital can engraft into lacrimal gland tissues and secrete
injection around or directly into lacrimal glands collagen type I that reduces the pathogenic fibro-
[70, 73–75], and intraperitoneal [71] or intrave- sis of the gland [86]. All of these preclinical
nous injections [68, 72, 76, 77]. These studies results suggest that MSCs are a promising cell
have shown that MSC therapy to treat DED therapy to treat DED, although more studies are
improves tear volume and tear film stability [69– needed to optimize it [87–89].
72, 74–76], maintains corneal epithelial integrity
[72, 74], increases the number of conjunctival
goblet cells [69, 70], and reduces ocular surface MSCs Promote Corneal Allograft
hyperemia [74–76]. Some studies also reported Survival
lacrimal gland regeneration [72, 77]. Moreover,
several authors found decreased ocular surface Corneal transplantation or keratoplasty is the
inflammation following MSC treatment. The most frequent type of human tissue transplant
reduced inflammation was associated with [90]. In low-risk patients, the survival rate of full-
decreased lymphocytic foci [71, 73] or CD4+ T thickness corneal grafts at 1 year is around 90%
cell infiltration [70], maintained or increased reg- (even without donor-recipient major histocom-
ulatory T cell (Treg) and Th2 presence [68, 71, patibility complex matching). However, in high-
risk patients with corneal neovascularization and there are several other potential stem cell niches
inflammation, the long-term prognosis is lower in the ocular surface that could help maintain cel-
than 50% [91, 92]. Topical corticosteroids are lular homeostasis of the corneal stroma, conjunc-
currently the most common immunosuppressive tiva, and meibomian glands [102]. And although
drugs used in corneal transplantation. However, the main stem cell deficiency at the ocular sur-
their effectiveness is lower in high-risk patients, face is the LSCD, causing corneal opacity, other
and prolonged application can provoke numerous pathologies are starting to be thought of as ame-
side effects [93, 94]. Therefore, alternative thera- nable to therapy with stem cells, as reviewed in a
peutic strategies are required to improve the previous section on preclinical studies. The fol-
prognosis of long-term corneal transplantation lowing are the most relevant ocular surface
and to diminish the adverse side effects of the pathologies for which stem cell treatment, most
current pharmacological treatments. specifically with MSCs, have already been trans-
Preclinical studies have shown that systemic lated into clinical practice and published.
and sub-conjunctival administration of MSCs
can prolong corneal allograft survival. Therefore,
their administration in combination with or in the MSCs for the Treatment of LSCD
absence of immunosuppressive drugs could help
prevent immune rejection of the corneal graft The destruction or dysfunction of the stem cells
[95–97]. The mechanism by which MSCs modu- residing in the limbal niche, leading to LSCD,
late corneal allograft survival has not been fully can have several aetiologies: chemical injuries,
elucidated yet; however, it has been associated immune-mediated cicatrizing diseases of the
with inhibition of antigen-presenting cell activa- ocular surface (e.g., Stevens-Johnson syndrome
tion, change in Th1/Th2 balance, reduction of and its spectrum, mucous membrane pemphi-
CD4+ T cell infiltration, and induction of Treg goid, atopic keratoconjunctivitis, ocular rosacea),
proliferation [95, 96, 98, 99]. These immuno- sequelae of infectious keratitis, or primary causes
modulatory and immunosuppressive actions are such as congenital aniridia or ectodermal dyspla-
related to the MSC-dependent secretion of solu- sia. All of these conditions lead to neovascular
ble factors such as TSG-6, hepatocyte growth pannus, an unstable corneal surface, and eventu-
factor, nitric oxide, and prostaglandin E2 [100, ally, visual deficit and chronic nociceptive pain
101]. Despite the encouraging preclinical results [11]. Diseases leading to LSCD are difficult to
obtained so far, there are still many issues and manage, requiring complex medical and surgical
challenges that need to be overcome before the approaches. Upon the development of LSCD, the
clinical application of this therapeutic approach problem becomes unsolvable unless new stem
in humans is attempted. These include determi- cells can be provided in the correct location
nation (1) if one or a few sources of MSCs pro- [103]. Since the first transplantations of autolo-
duce better clinical results than others, (2) the gous limbal tissue in 1989 [104] and the culti-
best dose and route of administration, and also vated autologous limbal cells in 1997 [105] to the
(3) the most effective frequency and timing of more recent techniques of delivering limbal tis-
cell administration [95, 96]. sue (simple limbal epithelial transplantation) in
2012 [106] or the cultivation of autologous and
allogeneic stem cells (reviewed in [11]), many
linical Evidence of MSC Efficacy
C cases have been successfully treated.
in Ocular Surface Pathology There is still a big need for the development of
safer, more accessible techniques that avoid the
Most studies of ocular surface stem cell func- necessity of immunosuppression when the source
tional failure have focused on the LESCs that of tissue or cells must be allogeneic, as it is often
reside in the corneoscleral limbal niche. However, the case in bilateral diseases. This can be achieved
with MSCs due to their many beneficial proper- with a size range of ∼30 to ∼200 nm in diameter.
ties, especially the absence of immunogenicity. EVs are lipid-encapsulated membranous vesicles
The use of allogeneic bone marrow-derived that are released by cells into the extracellular
MSCs has already been applied in the clinic. A spaces and contain components (protein, DNA,
randomized controlled clinical trial demonstrated and RNA) from the cells that release them. While
the benefits of this stem cell type, which was that trial was run for only 14 days, the signs and
assessed to be comparable and slightly superior symptoms of the GVHD-dependent DED were
to CLET in the management of LSCD [8]. This significantly mitigated. Thus, this cell-free
methodology avoids the use of immunosuppres- approach for delivering MSC components to treat
sion but can only be applied in places where a DED in general and specifically DED associated
Cell Processing Unit that complies with the with chronic GVHD is promising. The long-term
accepted standards of good manufacturing proce- effects and safety remain to be demonstrated, and
dures [107] is available. Therefore, work must MSC exosome-based therapy still faces chal-
progress to find solutions that are more accessi- lenges such as determining the stability during
ble and that perhaps can do more to replace the storage and transport, and determination of the
damage limbal niche instead of just providing heterogeneity of the exosome composition.
stem cells.
Conclusions and Future

MSCs for the Treatment of DED Perspectives
The most severe forms of DED are still difficult MSC-based therapies for ocular surface pathol-
to manage with current therapies. Undoubtedly, ogy, from corneal blindness due to LSCD, to
DED associated with chronic GVHD is one of immune-based inflammatory diseases such as
the most, if not the most, severe form of DED. It DED, or to corneal transplantation, show great
can be devastating with unbearable pain, photo- potential to reduce the onset of vision loss.
phobia, and reduced quality of life [108]. The Current preclinical evidence has already been
therapeutic efficacy of MSCs in the treatment of partially translated into clinical applications.
DED was first reported in a 2012 clinical study of These studies, of course, still need to be con-
22 chronic GVHD patients with refractory firmed with larger controlled clinical trials, and
DED. The patients were intravenously transfused some questions and technical problems remain to
with allogeneic MSCs, and 55% achieved clini- be solved. Among them, it should be elucidated if
cal improvement that was attributed to the gen- some MSC sources are better than others, and
eration of CD8+CD28-Tcells [109]. what are the safest and most clinically effective
In 2020, 7 patients with severe Sjögren’s MSC doses and routes of administration. In addi-
syndrome-associated DED were treated with adi- tion, it is essential to develop standardized proto-
pose tissue-derived MSCs that were delivered by cols for the culture and characterization of MSCs
a single transconjunctival injection into the main so that the results obtained in different preclinical
lacrimal gland. The treatment was well tolerated, and clinical centres can be properly compared.
and patients showed great improvement that Despite all the challenges and unknowns that
lasted up to 16 weeks [110]. remain, the future of MSCs in the ocular surface
In 2022, a clinical trial demonstrated the ben- is certainly promising (Fig. 15.2).
eficial effects of exosomes from human umbilical Over the last few years, EVs derived from
cord MSCs that were administered as eye drops MSCs have strongly emerged as a potential
to treat DED associated with chronic GVHD in alternative to MSC treatment. EVs appear to
14 patients [111]. Exosomes are a sub-type of replicate many of the therapeutic effects of
extracellular vesicles (EVs) of endosomal origin MSCs but without most of the safety risks and
Fig. 15.2 Mesenchymal stem cells for the regeneration of the ocular surface: from preclinical to clinical evidence
regulatory issues related to live cell therapies Take Home Notes

[112, 113]. As a consequence, MSC-derived • MSC-based treatments for ocular surface
EVs could represent a safer and more cost- pathology have shown potential therapeutic
effective alternative than cell therapies with value.
live MSCs. Currently, a lot of preclinical evi- • Preclinical studies have revealed that MSCs
dence supports the idea that MSC-derived EV can prolong corneal allograft survival.
application in corneal disease models induces • Preclinical evidence supporting the use of
anti-fibrotic, anti-apoptotic, and anti-inflamma- MSCs for treating LSCD and DED has already
tory effects, and that it promotes corneal epi- been translated into clinical practice.
thelial cell proliferation. These observations are • Although the results obtained so far on the use
consistent with the induction by EVs of accel- of MSCs for ocular surface pathology are very
erated corneal epithelial wound healing and encouraging, more preclinical and clinical
reduced corneal epithelial defects [114, 115]. studies are needed to confirm them.
The therapeutic development of EVs is still at • The clinical future of MSC-based therapy, and
an early stage, and the EV mechanism of action potentially MSC-derived EV therapy, in the
in ocular surface diseases remains to be fully ocular surface, is undoubtedly very
elucidated. Nevertheless, the solid evidence promising.
obtained from preclinical studies strongly sug-
gests that, in the near future, isolated MSC-
derived EVs could become a new therapeutic Acknowledgements We thank B. Bromberg (Certified
strategy for patients suffering from ocular sur- Editor in Life Sciences, Xenofile Editing) for his assis-
tance in the final editing and preparation of this manu-
face diseases. script. This work was funded by the Ministry of Science
and Innovation (grant PID2019- 105525RB- 100/ Med. 2010;363:147–55. https://doi.org/10.1056/

AEI/10.13039/501100011033 MICINN/FEDER, EU), NEJMoa0905955.
Spain; the Carlos III National Institute of Health, CIBER- 7. Ramírez BE, Sánchez A, Herreras JM, Fernández I,
BBN (CB06/01/003 MICINN/FEDER, EU), Spain; and García-Sancho J, Nieto-Miguel T, Calonge M. Stem
the Regional Center for Regenerative Medicine and Cell cell therapy for corneal epithelium regeneration
Therapy of Castilla y León, Spain. following good manufacturing and clinical proce-
dures. Biomed Res Int. 2015;2015:1–19. https://doi.
org/10.1155/2015/408495.
8. Calonge M, Pérez I, Galindo S, Nieto-Miguel
References T, López-Paniagua M, Fernández I, Alberca M,
García-Sancho J, Sánchez A, Herreras JM. A proof-
1. Flaxman SR, Bourne RRA, Resnikoff S, Ackland of-concept clinical trial using mesenchymal stem
P, Braithwaite T, Cicinelli MV, Das A, Jonas cells for the treatment of corneal epithelial stem cell
JB, Keeffe J, Kempen J, Leasher J, Limburg H, deficiency. Transl Res. 2019;206:18–40. https://doi.
Naidoo K, Pesudovs K, Silvester A, Stevens GA, org/10.1016/J.TRSL.2018.11.003.
Tahhan N, Wong T, Taylor H, Arditi A, Barkana 9. Behaegel J, Zakaria N, Tassignon M-J, Leysen I,
Y, Bozkurt B, Bron A, Budenz D, Cai F, Casson Bock F, Koppen C, Ní Dhubhghaill S. Short- and
R, Chakravarthy U, Choi J, Congdon N, Dana R, long-term results of xenogeneic-free cultivated
Dandona R, Dandona L, Dekaris I, Del Monte M, autologous and allogeneic limbal epithelial stem cell
Deva J, Dreer L, Ellwein L, Frazier M, Frick K, transplantations. Cornea. 2019;38:1543–9. https://
Friedman D, Furtado J, Gao H, Gazzard G, George doi.org/10.1097/ICO.0000000000002153.
R, Gichuhi S, Gonzalez V, Hammond B, Hartnett 10. Shimazaki J, Satake Y, Higa K, Yamaguchi T, Noma
ME, He M, Hejtmancik J, Hirai F, Huang J, Ingram H, Tsubota K. Long-term outcomes of cultivated cell
A, Javitt J, Joslin C, Khairallah M, Khanna R, sheet transplantation for treating total limbal stem
Kim J, Lambrou G, Lansingh VC, Lanzetta P, cell deficiency. Ocul Surf. 2020;18:663–71. https://
Lim J, Mansouri K, Mathew A, Morse A, Munoz doi.org/10.1016/j.jtos.2020.06.005.
B, Musch D, Nangia V, Palaiou M, Parodi MB, 11. Calonge M, Nieto-Miguel T, de la Mata A, Galindo
Pena FY, Peto T, Quigley H, Raju M, Ramulu P, S, Herreras JM, López-Paniagua M. Goals and
Reza D, Robin A, Rossetti L, Saaddine J, Sandar challenges of stem cell-based therapy for corneal
M, Serle J, Shen T, Shetty R, Sieving P, Silva blindness due to limbal deficiency. Pharmaceutics.
JC, Sitorus RS, Stambolian D, Tejedor J, Tielsch 2021;13:1483. https://doi.org/10.3390/
J, Tsilimbaris M, van Meurs J, Varma R, Virgili pharmaceutics13091483.
G, Wang YX, Wang NL, West S, Wiedemann P, 12. Rohban R, Pieber TR. Mesenchymal stem and pro-
Wormald R, Zheng Y. Global causes of blindness genitor cells in regeneration: tissue specificity and
and distance vision impairment 1990–2020: a sys- regenerative potential. Stem Cells Int. 2017;2017:1–
tematic review and meta-analysis. Lancet Glob 16. https://doi.org/10.1155/2017/5173732.
Health. 2017;5:e1221–34. https://doi.org/10.1016/ 13. Dominici M, Le Blanc K, Mueller I, Slaper-
S2214-109X(17)30393-5. Cortenbach I, Marini F, Krause DS, Deans RJ,
2. Cotsarelis G, Cheng SZ, Dong G, Sun TT, Lavker Keating A, Prockop DJ, Horwitz EM. Minimal cri-
RM. Existence of slow-cycling limbal epithe- teria for defining multipotent mesenchymal stromal
lial basal cells that can be preferentially stimu- cells. The International Society for Cellular Therapy
lated to proliferate: implications on epithelial position statement. Cytotherapy. 2006;8:315–7.
stem cells. Cell. 1989;57:201–9. https://doi. https://doi.org/10.1080/14653240600855905.
org/10.1016/0092-8674(89)90958-6. 14. Zhang L, Coulson-Thomas VJ, Ferreira TG, Kao
3. Li W, Hayashida Y, Chen Y-T, Tseng SC. Niche reg- WWY. Mesenchymal stem cells for treating ocular
ulation of corneal epithelial stem cells at the limbus. surface diseases. BMC Ophthalmol. 2015;15:155.
Cell Res. 2007;17:26–36. https://doi.org/10.1038/ https://doi.org/10.1186/s12886-015-0138-4.
sj.cr.7310137. 15. O’Callaghan AR, Daniels JT. Concise Review:
4. Schermer A, Galvin S, Sun TT. Differentiation- Limbal epithelial stem cell therapy: controversies
related expression of a major 64K corneal keratin and challenges. Stem Cells. 2011;29:1923–32.
in vivo and in culture suggests limbal location of cor- https://doi.org/10.1002/stem.756.
neal epithelial stem cells. J Cell Biol. 1986;103:49– 16. Luetzkendorf J, Nerger K, Hering J, Moegel A,
62. https://doi.org/10.1083/jcb.103.1.49. Hoffmann K, Hoefers C, Mueller-Tidow C, Mueller
5. Dua HS, Saini JS, Azuara-Blanco A, Gupta LP. Cryopreservation does not alter main character-
P. Limbal stem cell deficiency: concept, aetiology, istics of good manufacturing process–grade human
clinical presentation, diagnosis and management. multipotent mesenchymal stromal cells including
Indian J Ophthalmol. 2000;48:83–92. immunomodulating potential and lack of malig-
6. Rama P, Matuska S, Paganoni G, Spinelli A, De nant transformation. Cytotherapy. 2015;17:186–98.
Luca M, Pellegrini G. Limbal stem-cell therapy https://doi.org/10.1016/j.jcyt.2014.10.018.
and long-term corneal regeneration. N Engl J
17. Ho MSH, Mei SHJ, Stewart DJ. The immunomodu- tissue-derived mesenchymal stem cells in experi-
latory and therapeutic effects of mesenchymal stro- mental corneal failure due to limbal stem cell niche
mal cells for acute lung injury and sepsis. J Cell damage. Stem Cells. 2017;35:2160–74. https://doi.
Physiol. 2015;230:2606–17. https://doi.org/10.1002/ org/10.1002/stem.2672.
jcp.25028. 29. Di G, Du X, Qi X, Zhao X, Duan H, Li S, Xie L,
18. Griffin MD, Ritter T, Mahon BP. Immunological Zhou Q. Mesenchymal stem cells promote diabetic
aspects of allogeneic mesenchymal stem cell thera- corneal epithelial wound healing through TSG-6–
pies. Hum Gene Ther. 2010;21:1641–55. https://doi. dependent stem cell activation and macrophage
org/10.1089/hum.2010.156. switch. Invest Opthalmol Vis Sci. 2017;58:4344.
19. Zuk PA, Zhu M, Mizuno H, Huang J, Futrell https://doi.org/10.1167/iovs.17-21506.
JW, Katz AJ, Benhaim P, Lorenz HP, Hedrick 30. Yao L, Li Z, Su W, Li Y, Lin M, Zhang W, Liu Y,
MH. Multilineage cells from human adi- Wan Q, Liang D. Role of mesenchymal stem cells on
pose tissue: implications for cell-based thera- cornea wound healing induced by acute alkali burn.
pies. Tissue Eng. 2001;7:211–28. https://doi. PLoS ONE. 2012;7:e30842. https://doi.org/10.1371/
org/10.1089/107632701300062859. journal.pone.0030842.
20. Phinney DG, Prockop DJ. concise review: mesen- 31. Ke Y, Wu Y, Cui X, Liu X, Yu M, Yang C, Li
chymal stem/multipotent stromal cells: the state of X. Polysaccharide hydrogel combined with mesen-
transdifferentiation and modes of tissue repair—cur- chymal stem cells promotes the healing of corneal
rent views. Stem Cells. 2007;25:2896–902. https:// alkali burn in rats. PLoS ONE. 2015;10:e0119725.
doi.org/10.1634/stemcells.2007-0637. https://doi.org/10.1371/journal.pone.0119725.
21. Kuo TK, Ho JH, Lee OK. Mesenchymal stem cell 32. Lin H-F, Lai Y-C, Tai C-F, Tsai J-L, Hsu H-C, Hsu
therapy for nonmusculoskeletal diseases: emerging R-F, Lu S-N, Feng N-H, Chai C-Y, Lee C-H. Effects
applications. Cell Transplant. 2009;18:1013–28. of cultured human adipose-derived stem cells trans-
https://doi.org/10.3727/096368909X471206. plantation on rabbit cornea regeneration after alkaline
22. Joe AW, Gregory-Evans K. Mesenchymal stem cells chemical burn. Kaohsiung J Med Sci. 2013;29:14–8.
and potential applications in treating ocular disease. https://doi.org/10.1016/j.kjms.2012.08.002.
Curr Eye Res. 2010;35:941–52. https://doi.org/10.3 33. Ghazaryan E, Zhang Y, He Y, Liu X, Li Y, Xie J,
109/02713683.2010.516466. Su G. Mesenchymal stem cells in corneal neovas-
23. Chamberlain G, Fox J, Ashton B, Middleton cularization: comparison of different application
J. Concise Review: Mesenchymal stem cells: their routes. Mol Med Rep. 2016;14:3104–12. https://doi.
phenotype, differentiation capacity, immuno- org/10.3892/mmr.2016.5621.
logical features, and potential for homing. Stem 34. Shukla S, Mittal SK, Foulsham W, Elbasiony E,
Cells. 2007;25:2739–49. https://doi.org/10.1634/ Singhania D, Sahu SK, Chauhan SK. Therapeutic
stemcells.2007-0197. efficacy of different routes of mesenchymal
24. Ren G, Chen X, Dong F, Li W, Ren X, Zhang Y, Shi stem cell administration in corneal injury. Ocul
Y. Concise Review: Mesenchymal stem cells and Surf. 2019;17:729–36. https://doi.org/10.1016/j.
translational medicine: emerging issues. Stem Cells jtos.2019.07.005.
Transl Med. 2012;1:51–8. https://doi.org/10.5966/ 35. Li G, Zhang Y, Cai S, Sun M, Wang J, Li S, Li
sctm.2011-0019. X, Tighe S, Chen S, Xie H, Zhu Y. Human limbal
25. Nieto-Miguel T, Galindo S, López-Paniagua M, niche cells are a powerful regenerative source for
Pérez I, Herreras JM, Calonge M. Cell therapy using the prevention of limbal stem cell deficiency in a
extraocular mesenchymal stem cells. In: Alió J, Alió rabbit model. Sci Rep. 2018;8:6566. https://doi.
del Barrio J, Arnalich-Montiel F, editors. Corneal org/10.1038/s41598-018-24862-6.
regeneration. Essentials in ophthalmology. Cham: 36. Pan J, Wang X, Li D, Li J, Jiang Z. MSCs inhib-
Springer; 2019. p. 231–62. its the angiogenesis of HUVECs through the miR-
26. Beeken LJ, Ting DSJ, Sidney LE. Potential of mes- 211/Prox1 pathway. J Biochem. 2019;166:107–13.
enchymal stem cells as topical immunomodula- https://doi.org/10.1093/jb/mvz038.
tory cell therapies for ocular surface inflammatory 37. Zhang N, Luo X, Zhang S, Liu R, Liang L, Su W,
disorders. Stem Cells Transl Med. 2021;10:39–49. Liang D. Subconjunctival injection of tumor necro-
https://doi.org/10.1002/sctm.20-0118. sis factor-α pre-stimulated bone marrow-derived
27. Galindo S, de la Mata A, López-Paniagua mesenchymal stem cells enhances anti-inflammation
M, Herreras JM, Pérez I, Calonge M, Nieto- and anti-fibrosis in ocular alkali burns. Graefes Arch
Miguel T. Subconjunctival injection of mes- Clin Exp Ophthalmol. 2021;259:929–40. https://doi.
enchymal stem cells for corneal failure due to org/10.1007/s00417-020-05017-8.
limbal stem cell deficiency: state of the art. Stem 38. Zeppieri M, Salvetat ML, Beltrami AP, Cesselli
Cell Res Ther. 2021;12:60. https://doi.org/10.1186/ D, Bergamin N, Russo R, Cavaliere F, Varano GP,
s13287-020-02129-0. Alcalde I, Merayo J, Brusini P, Beltrami CA, Parodi
28. Galindo S, Herreras JM, López-Paniagua M, Rey E, PC. Human adipose-derived stem cells for the treat-
de la Mata A, Plata-Cordero M, Calonge M, Nieto- ment of chemically burned rat cornea: preliminary
Miguel T. Therapeutic effect of human adipose
results. Curr Eye Res. 2013;38:451–63. https://doi. SK, Chauhan SK. Restoration of corneal trans-
org/10.3109/02713683.2012.763100. parency by mesenchymal stem cells. Stem Cell
39. Oh JY, Kim MK, Shin MS, Lee HJ, Ko JH, Wee Rep. 2016;7:583–90. https://doi.org/10.1016/j.
WR, Lee JH. The anti-inflammatory and anti- stemcr.2016.09.001.
angiogenic role of mesenchymal stem cells in cor- 49. Lee RH, Yu JM, Foskett AM, Peltier G, Reneau
neal wound healing following chemical injury. Stem JC, Bazhanov N, Oh JY, Prockop DJ. TSG-6 as a
Cells. 2008;26:1047–55. https://doi.org/10.1634/ biomarker to predict efficacy of human mesenchy-
stemcells.2007-0737. mal stem/progenitor cells (hMSCs) in modulat-
40. Ma Y, Xu Y, Xiao Z, Yang W, Zhang C, Song E, Du ing sterile inflammation in vivo. Proc Natl Acad
Y, Li L. Reconstruction of chemically burned rat cor- Sci. 2014;111:16766–71. https://doi.org/10.1073/
neal surface by bone marrow-derived human mes- pnas.1416121111.
enchymal stem cells. Stem Cells. 2006;24:315–21. 50. Lan Y, Kodati S, Lee HS, Omoto M, Jin Y, Chauhan
https://doi.org/10.1634/stemcells.2005-0046. SK. Kinetics and function of mesenchymal stem
41. Rohaina CM, Then KY, Ng AMH, Wan Abdul cells in corneal injury. Invest Opthalmol Vis Sci.
Halim WH, Zahidin AZM, Saim A, Idrus 2012;53:3638. https://doi.org/10.1167/iovs.11-9311.
RBH. Reconstruction of limbal stem cell deficient 51. Roddy GW, Oh JY, Lee RH, Bartosh TJ, Ylostalo J,
corneal surface with induced human bone mar- Coble K, Rosa RH, Prockop DJ. Action at a distance:
row mesenchymal stem cells on amniotic mem- systemically administered adult stem/progenitor
brane. Transl Res. 2014;163:200–10. https://doi. cells (mscs) reduce inflammatory damage to the cor-
org/10.1016/j.trsl.2013.11.004. nea without engraftment and primarily by s ecretion
42. Pınarlı FA, Okten G, Beden U, Fışgın T, Kefeli M, of TNF-α stimulated gene/protein 6. Stem Cells.
Kara N, Duru F, Tomak L. Keratinocyte growth factor- 2011;29:1572–9. https://doi.org/10.1002/stem.708.
2 and autologous serum potentiate the regenerative 52. Ye J, Yao K, Kim JC. Mesenchymal stem cell trans-
effect of mesenchymal stem cells in cornea damage plantation in a rabbit corneal alkali burn model:
in rats. Int J Ophthalmol. 2014;7:211–9. https://doi. engraftment and involvement in wound healing.
org/10.3980/j.issn.2222-3959.2014.02.05. Eye. 2006;20:482–90. https://doi.org/10.1038/
43. Jiang T-S, Cai L, Ji W-Y, Hui Y-N, Wang Y-S, Hu D, sj.eye.6701913.
Zhu J. Reconstruction of the corneal epithelium with 53. Yun YI, Park SY, Lee HJ, Ko JH, Kim MK, Wee
induced marrow mesenchymal stem cells in rats. WR, Reger RL, Gregory CA, Choi H, Fulcher
Mol Vis. 2010;16:1304–16. SF, Prockop DJ, Oh JY. Comparison of the anti-
44. Cejkova J, Trosan P, Cejka C, Lencova A, Zajicova inflammatory effects of induced pluripotent stem
A, Javorkova E, Kubinova S, Sykova E, Holan cell–derived and bone marrow–derived mesen-
V. Suppression of alkali-induced oxidative injury in chymal stromal cells in a murine model of corneal
the cornea by mesenchymal stem cells growing on injury. Cytotherapy. 2017;19:28–35. https://doi.
nanofiber scaffolds and transferred onto the dam- org/10.1016/j.jcyt.2016.10.007.
aged corneal surface. Exp Eye Res. 2013;116:312– 54. Acar U, Pinarli FA, Acar DE, Beyazyildiz E, Sobaci
23. https://doi.org/10.1016/j.exer.2013.10.002. G, Ozgermen BB, Sonmez AA, Delibasi T. Effect
45. Cejka C, Cejkova J, Trosan P, Zajicova A, Sykova of allogeneic limbal mesenchymal stem cell ther-
E, Holan V. Transfer of mesenchymal stem cells and apy in corneal healing: role of administration
cyclosporine A on alkali-injured rabbit cornea using route. Ophthalmic Res. 2015;53:82–9. https://doi.
nanofiber scaffolds strongly reduces corneal neovas- org/10.1159/000368659.
cularization and scar formation. Histol Histopathol. 55. Nili E, Li FJ, Dawson RA, Lau C, McEwan B,
2016;31:969–80. https://doi.org/10.14670/ Barnett NL, Weier S, Walshe J, Richardson NA,
HH-11-724. Harkin DG. The impact of limbal mesenchymal
46. Cejka C, Holan V, Trosan P, Zajicova A, Javorkova stromal cells on healing of acute ocular surface
E, Cejkova J. The favorable effect of mesenchymal wounds is improved by pre-cultivation and implan-
stem cell treatment on the antioxidant protective tation in the presence of limbal epithelial cells.
mechanism in the corneal epithelium and renewal Cell Transplant. 2019;28:1257–70. https://doi.
of corneal optical properties changed after alkali org/10.1177/0963689719858577.
burns. Oxidative Med Cell Longev. 2016;2016:1–12. 56. Gomes JÁP, Geraldes Monteiro B, Melo GB, Smith
https://doi.org/10.1155/2016/5843809. RL, Pereira C, da Silva M, Lizier NF, Kerkis A,
47. Holan V, Trosan P, Cejka C, Javorkova E, Zajicova A, Cerruti H, Kerkis I. Corneal reconstruction with
Hermankova B, Chudickova M, Cejkova J. A com- tissue-engineered cell sheets composed of human
parative study of the therapeutic potential of mesen- immature dental pulp stem cells. Invest Opthalmol
chymal stem cells and limbal epithelial stem cells Vis Sci. 2010;51:1408. https://doi.org/10.1167/
for ocular surface reconstruction. Stem Cells Transl iovs.09-4029.
Med. 2015;4:1052–63. https://doi.org/10.5966/ 57. Espandar L, Caldwell D, Watson R, Blanco-Mezquita
sctm.2015-0039. T, Zhang S, Bunnell B. Application of adipose-
48. Mittal SK, Omoto M, Amouzegar A, Sahu A, derived stem cells on scleral contact lens carrier in
Rezazadeh A, Katikireddy KR, Shah DI, Sahu an animal model of severe acute alkaline burn. Eye
Contact Lens Sci Clin Pract. 2014;40:243–7. https:// topical mesenchymal stem cell therapy in the
doi.org/10.1097/ICL.0000000000000045. treatment of experimental dry eye syndrome
58. Ahmed SK, Soliman AA, Omar SMM, Mohammed model. Stem Cells Int. 2014;2014:1–9. https://doi.
WR. Bone marrow mesenchymal stem cell trans- org/10.1155/2014/250230.
plantation in a rabbit corneal alkali burn model (a 70. Lee MJ, Ko AY, Ko JH, Lee HJ, Kim MK, Wee
histological and immune histo-chemical study). Int J WR, Khwarg SI, Oh JY. Mesenchymal stem/stro-
Stem Cells. 2015;8:69–78. https://doi.org/10.15283/ mal cells protect the ocular surface by suppress-
ijsc.2015.8.1.69. ing inflammation in an experimental dry eye. Mol
59. Almaliotis D, Koliakos G, Papakonstantinou E, Ther. 2015;23:139–46. https://doi.org/10.1038/
Komnenou A, Thomas A, Petrakis S, Nakos I, mt.2014.159.
Gounari E, Karampatakis V. Mesenchymal stem 71. Aluri HS, Samizadeh M, Edman MC, Hawley
cells improve healing of the cornea after alkali injury. DR, Armaos HL, Janga SR, Meng Z, Sendra VG,
Graefes Arch Clin Exp Ophthalmol. 2015;253:1121– Hamrah P, Kublin CL, Hamm-Alvarez SF, Zoukhri
35. https://doi.org/10.1007/s00417-015-3042-y. D. Delivery of bone marrow-derived mesenchymal
60. Gu S, Xing C, Han J, Tso MOM, Hong stem cells improves tear production in a mouse model
J. Differentiation of rabbit bone marrow mesenchy- of Sjögren’s syndrome. Stem Cells Int. 2017;2017:1–
mal stem cells into corneal epithelial cells in vivo 10. https://doi.org/10.1155/2017/3134543.
and ex vivo. Mol Vis. 2009;15:99–107. 72. Abughanam G, Elkashty OA, Liu Y, Bakkar MO,
61. Reinshagen H, Auw-Haedrich C, Sorg RV, Tran SD. Mesenchymal stem cells extract (MSCsE)-
Boehringer D, Eberwein P, Schwartzkopff J, based therapy alleviates xerostomia and keratocon-
Sundmacher R, Reinhard T. Corneal surface recon- junctivitis sicca in Sjogren’s syndrome-like disease.
struction using adult mesenchymal stem cells in Int J Mol Sci. 2019;20:4750. https://doi.org/10.3390/
experimental limbal stem cell deficiency in rab- ijms20194750.
bits. Acta Ophthalmol. 2011;89:741–8. https://doi. 73. Park SA, Reilly CM, Wood JA, Chung DJ,
org/10.1111/j.1755-3768.2009.01812.x. Carrade DD, Deremer SL, Seraphin RL, Clark
62. Craig JP, Nichols KK, Akpek EK, Caffery B, Dua KC, Zwingenberger AL, Borjesson DL, Hayashi
HS, Joo C-K, Liu Z, Nelson JD, Nichols JJ, Tsubota K, Russell P, Murphy CJ. Safety and immuno-
K, Stapleton F. TFOS DEWS II definition and clas- modulatory effects of allogeneic canine adipose-
sification report. Ocul Surf. 2017;15:276–83. https:// derived mesenchymal stromal cells transplanted
doi.org/10.1016/j.jtos.2017.05.008. into the region of the lacrimal gland, the gland of
63. Gayton J. Etiology, prevalence, and treatment of dry the third eyelid and the knee joint. Cytotherapy.
eye disease. Clin Ophthalmol. 2009;3:405. https:// 2013;15:1498–510. https://doi.org/10.1016/j.
doi.org/10.2147/OPTH.S5555. jcyt.2013.06.009.
64. Calonge M, Enríquez-de-Salamanca A, Diebold 74. Villatoro AJ, Fernández V, Claros S, Rico-Llanos
Y, González-García MJ, Reinoso R, Herreras JM, GA, Becerra J, Andrades JA. Use of adipose-derived
Corell A. Dry eye disease as an inflammatory dis- mesenchymal stem cells in keratoconjunctivitis sicca
order. Ocul Immunol Inflamm. 2010;18:244–53. in a canine model. Biomed Res Int. 2015;2015:1–10.
https://doi.org/10.3109/09273941003721926. https://doi.org/10.1155/2015/527926.
65. Wei Y, Asbell PA. The core mechanism of dry eye 75. Bittencourt MKW, Barros MA, Martins JFP,
disease is inflammation. Eye Contact Lens Sci Clin Vasconcellos JPC, Morais BP, Pompeia C,
Pract. 2014;40:248–56. https://doi.org/10.1097/ Bittencourt MD, Evangelho KDS, Kerkis I,
ICL.0000000000000042. Wenceslau CV. Allogeneic mesenchymal stem cell
66. Hagan S, Martin E, Enríquez-de-Salamanca A. Tear transplantation in dogs with keratoconjunctivitis
fluid biomarkers in ocular and systemic disease: sicca. Cell Med. 2016;8:63–77. https://doi.org/10.3
potential use for predictive, preventive and person- 727/215517916X693366.
alised medicine. EPMA J. 2016;7:15. https://doi. 76. Lu X, Li N, Zhao L, Guo D, Yi H, Yang L, Liu X,
org/10.1186/s13167-016-0065-3. Sun D, Nian H, Wei R. Human umbilical cord mes-
67. Choi SW, Cha BG, Kim J. Therapeutic contact lens enchymal stem cells alleviate ongoing autoimmune
for scavenging excessive reactive oxygen species on dacryoadenitis in rabbits via polarizing macrophages
the ocular surface. ACS Nano. 2020;14:2483–96. into an anti-inflammatory phenotype. Exp Eye
https://doi.org/10.1021/acsnano.9b10145. Res. 2020;191:107905. https://doi.org/10.1016/j.
68. Xu J, Wang D, Liu D, Fan Z, Zhang H, Liu O, Ding exer.2019.107905.
G, Gao R, Zhang C, Ding Y, Bromberg JS, Chen 77. Dietrich J, Ott L, Roth M, Witt J, Geerling G, Mertsch
W, Sun L, Wang S. Allogeneic mesenchymal stem S, Schrader S. MSC transplantation improves lacri-
cell treatment alleviates experimental and clini- mal gland regeneration after surgically induced dry
cal Sjögren syndrome. Blood. 2012;120:3142–51. eye disease in mice. Sci Rep. 2019;9:18299. https://
https://doi.org/10.1182/blood-2011-11-391144. doi.org/10.1038/s41598-019-54840-5.
69. Beyazyıldız E, Pınarlı FA, Beyazyıldız Ö, 78. Li F, Zhao S. Control of cross talk between angio-
Hekimoğlu ER, Acar U, Demir MN, Albayrak genesis and inflammation by mesenchymal stem
A, Kaymaz F, Sobacı G, Delibaşı T. Efficacy of cells for the treatment of ocular surface dis-
eases. Stem Cells Int. 2016;2016:1–8. https://doi. Exp Pharmacol. 2021;13:345–58. https://doi.
org/10.1155/2016/7961816. org/10.2147/JEP.S237487.
79. Ogawa Y, Shimmura S, Dogru M, Tsubota 90. Gain P, Jullienne R, He Z, Aldossary M, Acquart
K. Immune processes and pathogenic fibrosis in S, Cognasse F, Thuret G. Global survey of corneal
ocular chronic graft-versus-host disease and clini- transplantation and eye banking. JAMA Ophthalmol.
cal manifestations after allogeneic hematopoietic 2016;134:167–73. https://doi.org/10.1001/
stem cell transplantation. Cornea. 2010;29:S68–77. jamaophthalmol.2015.4776.
https://doi.org/10.1097/ICO.0b013e3181ea9a6b. 91. Williams KA, Esterman AJ, Bartlett C, Holland
80. Ogawa Y, Okamoto S, Wakui M, Watanabe R, H, Hornsby NB, Coster DJ. How effective is
Yamada M, Yoshino M, Ono M, Yang H-Y, penetrating corneal transplantation? Factors
Mashima Y, Oguchi Y, Ikeda Y, Tsubota K. Dry influencing long-term outcome in multivariate anal-
eye after haematopoietic stem cell transplantation. ysis. Transplantation. 2006;81:896–901. https://doi.
Br J Ophthalmol. 1999;83:1125–30. https://doi. org/10.1097/01.tp.0000185197.37824.35.
org/10.1136/bjo.83.10.1125. 92. Alio JL, Montesel A, El Sayyad F, Barraquer RI,
81. Shikari H, Antin JH, Dana R. Ocular graft- Arnalich-Montiel F, Alio Del Barrio JL. Corneal
versus-host disease: a review. Surv Ophthalmol. graft failure: an update. Br J Ophthalmol.
2013;58:233–51. https://doi.org/10.1016/j. 2021;105:1049–58. https://doi.org/10.1136/
survophthal.2012.08.004. bjophthalmol-2020-316705.
82. Ogawa Y, Kawakami Y, Tsubota K. Cascade of 93. Tahvildari M, Amouzegar A, Foulsham W, Dana
inflammatory, fibrotic processes, and stress-induced R. Therapeutic approaches for induction of tolerance
senescence in chronic GVHD-related dry eye dis- and immune quiescence in corneal allotransplanta-
ease. Int J Mol Sci. 2021;22:6114. https://doi. tion. Cell Mol Life Sci. 2018;75:1509–20. https://
org/10.3390/ijms22116114. doi.org/10.1007/s00018-017-2739-y.
83. Shimizu S, Sato S, Taniguchi H, Shimizu E, He 94. Renfro L, Snow JS. Ocular effects of topical and
J, Hayashi S, Negishi K, Ogawa Y, Shimmura systemic steroids. Dermatol Clin. 1992;10:505–12.
S. Observation of chronic graft-versus-host dis- 95. Murphy N, Lynch K, Lohan P, Treacy O, Ritter
ease mouse model cornea with in vivo confocal T. Mesenchymal stem cell therapy to promote cor-
microscopy. Diagnostics. 2021;11:1515. https://doi. neal allograft survival. Curr Opin Organ Transplant.
org/10.3390/diagnostics11081515. 2016;21:559–67. https://doi.org/10.1097/
84. Sanchez-Abarca LI, Hernandez-Galilea E, Lorenzo MOT.0000000000000360.
R, Herrero C, Velasco A, Carrancio S, Caballero- 96. Oh JY, Kim E, Yun YI, Lee RH. Mesenchymal
Velazquez T, Rodriguez-Barbosa JI, Parrilla M, stromal cells for corneal transplantation: litera-
Del Canizo C, San Miguel J, Aijon J, Perez-Simon ture review and suggestions for successful clini-
JA. Human bone marrow stromal cells differentiate cal trials. Ocul Surf. 2021;20:185–94. https://doi.
into corneal tissue and prevent ocular graft-versus- org/10.1016/j.jtos.2021.02.002.
host disease in mice. Cell Transplant. 2015;24:2423– 97. Treacy O, Lynch K, Murphy N, Chen X, Donohoe
33. https://doi.org/10.3727/096368915x687480. E, Canning A, Lohan P, Shaw G, Fahy G, Ryan AE,
85. Martínez-Carrasco R, Sánchez-Abarca LI, Nieto- Ritter T. Subconjunctival administration of low-dose
Gómez C, Martín García E, Sánchez-Guijo F, murine allogeneic mesenchymal stromal cells pro-
Argüeso P, Aijón J, Hernández-Galilea E, Velasco motes corneal allograft survival in mice. Stem Cell
A. Subconjunctival injection of mesenchymal Res Ther. 2021;12:227. https://doi.org/10.1186/
stromal cells protects the cornea in an experimen- s13287-021-02293-x.
tal model of GVHD. Ocul Surf. 2019;17:285–94. 98. Lynch K, Treacy O, Chen X, Murphy N, Lohan
https://doi.org/10.1016/j.jtos.2019.01.001. P, Islam MN, Donohoe E, Griffin MD, Watson
86. Rusch RM, Ogawa Y, Sato S, Morikawa S, Inagaki E, L, McLoughlin S, O’Malley G, Ryan AE, Ritter
Shimizu E, Tsubota K, Shimmura S. MSCs become T. TGF-β1-licensed murine MSCs show superior
collagen-type I producing cells with different pheno- therapeutic efficacy in modulating corneal allograft
type in allogeneic and syngeneic bone marrow trans- immune rejection in vivo. Mol Ther. 2020;28:2023–
plantation. Int J Mol Sci. 2021;22:4895. https://doi. 43. https://doi.org/10.1016/j.ymthe.2020.05.023.
org/10.3390/ijms22094895. 99. Murphy N, Treacy O, Lynch K, Morcos M, Lohan
87. Al-Jaibaji O, Swioklo S, Connon CJ. Mesenchymal P, Howard L, Fahy G, Griffin MD, Ryan AE, Ritter
stromal cells for ocular surface repair. Expert Opin T. TNF-α/IL-1β—licensed mesenchymal stromal
Biol Ther. 2019;19:643–53. https://doi.org/10.1080/ cells promote corneal allograft survival via myeloid
14712598.2019.1607836. cell-mediated induction of Foxp3 + regulatory
88. Dietrich J, Schrader S. Towards lacrimal gland T cells in the lung. FASEB J. 2019;33:9404–21.
regeneration: current concepts and experimental https://doi.org/10.1096/fj.201900047R.
approaches. Curr Eye Res. 2020;45:230–40. https:// 100. Oh JY, Lee RH, Yu JM, Ko JH, Lee HJ, Ko AY,
doi.org/10.1080/02713683.2019.1637438. Roddy GW, Prockop DJ. Intravenous mesenchymal
89. Baiula M, Spampinato S. Experimental phar- stem cells prevented rejection of allogeneic cor-
macotherapy for dry eye disease: a review. J neal transplants by aborting the early inflammatory
response. Mol Ther. 2012;20:2143–52. https://doi. GVHD defined by NIH consensus criteria. Bone
org/10.1038/mt.2012.165. Marrow Transplant. 2012;47:1470–3. https://doi.
101. Mittal SK, Foulsham W, Shukla S, Elbasiony E, org/10.1038/bmt.2012.56.
Omoto M, Chauhan SK. Mesenchymal stromal 109. Weng J, He C, Lai P, Luo C, Guo R, Wu S, Geng
cells modulate corneal alloimmunity via secre- S, Xiangpeng A, Liu X, Du X. Mesenchymal stro-
tion of hepatocyte growth factor. Stem Cells Transl mal cells treatment attenuates dry eye in patients
Med. 2019;8:1030–40. https://doi.org/10.1002/ with chronic graft-versus-host disease. Mol
sctm.19-0004. Ther. 2012;20:2347–54. https://doi.org/10.1038/
102. Ramos T, Scott D, Ahmad S. An update on ocular mt.2012.208.
surface epithelial stem cells: cornea and conjunc- 110. Møller-Hansen M, Larsen A-C, Toft PB, Lynggaard
tiva. Stem Cells Int. 2015;2015:1–7. https://doi. CD, Schwartz C, Bruunsgaard H, Haack-Sørensen
org/10.1155/2015/601731. M, Ekblond A, Kastrup J, Heegaard S. Safety and
103. Elhusseiny AM, Soleimani M, Eleiwa TK, ElSheikh feasibility of mesenchymal stem cell therapy in
RH, Frank CR, Naderan M, Yazdanpanah G, patients with aqueous deficient dry eye disease.
Rosenblatt MI, Djalilian AR. Current and emerg- Ocul Surf. 2021;19:43–52. https://doi.org/10.1016/j.
ing therapies for limbal stem cell deficiency. Stem jtos.2020.11.013.
Cells Transl Med. 2022;11:259–68. https://doi. 111. Zhou T, He C, Lai P, Yang Z, Liu Y, Xu H, Lin X,
org/10.1093/stcltm/szab028. Ni B, Ju R, Yi W, Liang L, Pei D, Egwuagu CE,
104. Keivyon KR, Tseng SCG. Limbal autograft Liu X. miR-204–containing exosomes amelio-
transplantation for ocular surface disorders. rate GVHD-associated dry eye disease. Sci Adv.
Ophthalmology. 1989;96:709–23. https://doi. 2022;8:eabj9617. https://doi.org/10.1126/sciadv.
org/10.1016/S0161-6420(89)32833-8. abj9617.
105. Pellegrini G, Traverso CE, Franzi AT, Zingirian M, 112. Rani S, Ryan AE, Griffin MD, Ritter
Cancedda R, De Luca M. Long-term restoration of T. Mesenchymal stem cell-derived extracellular
damaged corneal surfaces with autologous cultivated vesicles: toward cell-free therapeutic applications.
corneal epithelium. Lancet. 1997;349:990–3. https:// Mol Ther. 2015;23:812–23. https://doi.org/10.1038/
doi.org/10.1016/S0140-6736(96)11188-0. mt.2015.44.
106. Sangwan VS, Basu S, MacNeil S, Balasubramanian 113. Fuloria S, Subramaniyan V, Dahiya R, Dahiya S,
D. Simple limbal epithelial transplantation (SLET): Sudhakar K, Kumari U, Sathasivam K, Meenakshi
a novel surgical technique for the treatment of uni- DU, Wu YS, Sekar M, Malviya R, Singh A, Fuloria
lateral limbal stem cell deficiency. Br J Ophthalmol. NK. Mesenchymal stem cell-derived extracellu-
2012;96:931–4. https://doi.org/10.1136/ lar vesicles: regenerative potential and challenges.
bjophthalmol-2011-301164. Biology. 2021;10:172. https://doi.org/10.3390/
107. López-Paniagua M, De La Mata A, Galindo S, biology10030172.
Blázquez F, Calonge M, Nieto-Miguel T. Advanced 114. Deng SX, Dos Santos A, Gee S. Therapeutic poten-
therapy medicinal products for the eye: defini- tial of extracellular vesicles for the treatment of
tions and regulatory framework. Pharmaceutics. corneal injuries and scars. Transl Vis Sci Technol.
2021;13:347. https://doi.org/10.3390/ 2020;9:1. https://doi.org/10.1167/tvst.9.12.1.
pharmaceutics13030347. 115. McKay TB, Yeung V, Hutcheon AEK, Guo
108. Jacobs R, Tran U, Chen H, Kassim A, Engelhardt X, Zieske JD, Ciolino JB. Extracellular vesi-
BG, Greer JP, Goodman SG, Clifton C, Lucid C, cles in the cornea: insights from other tissues.
Vaughan LA, Savani BN, Jagasia M. Prevalence and Anal Cell Pathol. 2021;2021:1–12. https://doi.
risk factors associated with development of ocular org/10.1155/2021/9983900.
Induced Pluripotent Stem Cells
in Epithelial Lamellar Keratoplasty
16
Sanja Bojic, Francisco Figueiredo,
and Majlinda Lako
Abbreviations MZOC Multizone ocular cells

PMC Post-mitotic cells
ATMP Advanced therapeutic medicinal RA Trans-retinoic acid
products RPE Retinal pigmented epithelium
CLET Cultivated limbal epithelial SDIA Stromal cell-derived inducing activity
transplantation SEAM Self-formed ectodermal autonomous
ECM Extracellular matrix multizone
EGF Epithelial growth factor SLET Simple limbal epithelial
ESC Embryonic stem cells transplantation
FACS Fluorescence-activated cell sorting SMILE Small incision lenticule extraction
FBS Fetal bovine serum TAC Transient amplifying ell
FGF Fibroblast growth factor TDC Terminally differentiated cells
GMP Good manufacturing practice TGF Transforming growth factor
HAM Human amniotic membrane
iPSC Induced pluripotent stem cells
KGF Keratinocyte growth factor
KRT Cytokeratin Key Points
LSC Limbal stem cells
LSCD Limbal stem cell deficiency • Limbal stem cell biology and limbal stem cell
MSC Mesenchymal stem cells deficiency;
• Limbal stem cell transplantation—advantages
and limitations;
S. Bojic (*) · M. Lako • History and biology of induced pluripotent
Faculty of Medical Sciences, Biosciences Institute, stem cells;
Newcastle University, Newcastle upon Tyne, UK
e-mail: [email protected]; • Differentiation of induced pluripotent stem
[email protected] cells to corneal epithelial lineages—advan-
F. Figueiredo tages and limitations;
Faculty of Medical Sciences, Biosciences Institute, • Main strategies for inducing corneal epithelial
Newcastle University, Newcastle upon Tyne, UK differentiation in induced pluripotent stem
Department of Ophthalmology, Royal Victoria cells.
Infirmary, Newcastle upon Tyne Hospitals NHS
Foundation Trust, Newcastle upon Tyne, UK
https://doi.org/10.1007/978-3-031-32408-6_16
226 S. Bojic et al.
Introduction metric division, stem cells give rise to one daugh-

ter cell that remains in the niche as a stem cell,
As the outermost part and a major protective while the other daughter cell enters differentiation
shield of the eye, the cornea is directly exposed to and becomes a transient amplifying cell (TAC).
the environment. Corneal scarring and opacity Unlike stem cells that divide sparsely, TACs show
are the fourth most common cause of blindness very high mitotic activity. By going through mul-
globally, according to the World Health tiple divisions to increase the number of cells
Organization [1]. Corneal diseases represent a resulting from each stem cell, they protect stem
major cause of blindness with the affected popu- cells from going through the cell cycle too often,
lation estimated to be more than 10 million peo- which could result in DNA damage accumulation
ple worldwide [2, 3]. Physical injuries such as in the stem cell pool over time. Eventually, after a
abrasions, thermal or chemical burns, infections, certain number of divisions, TACs differentiate
refractive surgeries, contact lens wearing, or into post-mitotic cells (PMCs), incapable of fur-
insufficient tear production are common reasons ther division. PMCs are fully committed to differ-
for corneal damage [4]. Besides its barrier func- entiation and mature into terminally differentiated
tion, the cornea is also responsible for three- cells (TDCs). Similar to the other squamous epi-
fourths of the total refractive power of the human thelia, the corneal TDCs are continuously shed
eye, and preserving its transparency is vital for from the corneal surface, while new TACs are con-
normal sight. tinuously produced from LSCs residing in the
The cornea is composed of three cellular lay- basal layers of the limbus. TACs migrate from the
ers of different developmental origins: the outer corneal periphery toward the central region of the
epithelial layer (of ectodermal origin), the stroma, cornea, and simultaneously ascend from basal to
and the inner endothelial layer (both of neural superficial layers of the cornea, to differentiate and
crest origin) [5]. The epithelium as the outermost replace cells continuously lost from the corneal
layer serves as a principal barrier to foreign mate- surface [8]. In homeostasis, the rate of cell prolif-
rials, including pathogens, due to the presence of eration is equal to the rate of cell desquamation,
tight junctions between the epithelial cells and maintaining the corneal epithelial mass constant.
continuous cell turnover. The integrity of the cor- The lifespan of human corneal epithelial cells is
neal surface is vital for its transparency. The epi- approximately 7–14 days [5].
thelial layer of the cornea renews continuously
throughout life. A pool of epithelial stem cells
located in the limbal region of the cornea serves Limbal Stem Cells (LSCs)
as a lifetime reservoir of undifferentiated cells
which enables constant regeneration of the cor- Stem cells are undifferentiated cells, with unlim-
nea. Due to their location, corneal epithelial stem ited potential for self-renewal and the potential to
cells are known as limbal stem cells (LSCs). differentiate into different cell types [9].
LSCs were reported for the first time in 1989 by Tissue-specific stem cells, also known as adult
Cotsarelis and colleagues, as slow cycling, label stem cells, are present in almost every adult tis-
retaining cells [6]. sue, serving as a constant reservoir of new cells
The constant corneal renewal and homeostasis for tissue regeneration. LSCs as adult stem cells
are achieved by three different processes that hap- are responsible for lifetime-long corneal epithe-
pen simultaneously in the corneal epithelium: pro- lial maintenance and regeneration. Most of the
liferation, migration, and differentiation. Crucial time stem cells are quiescent (in a growth-arrested
for maintaining corneal integrity is the balance state), but they can enter the cell cycle on demand
between proliferation and differentiation [7]. (e.g., when tissue is injured) and give rise to their
Preserving the constant pool of stem cells in the highly proliferative progenies—TACs. Although
limbus throughout the lifetime is achieved by their dividing less frequently, LSCs show a high clo-
sparse asymmetric cell divisions. During asym- nogenic potential [10].
16 Induced Pluripotent Stem Cells in Epithelial Lamellar Keratoplasty 227
A critical role in preventing differentiation Table 16.1 Commonly used molecular markers of iPSC,
limbal stem cells, and corneal epithelial cells
and maintaining a balance between quiescence,
proliferation, and regeneration is played by the Corneal
Limbal stem epithelial
stem cell microenvironment, known as the iPSC cells cells
stem cell niche [11]. The concept of stem cell Intracellular Nanog p63, ΔNp63 KRT3
niche as a unique microenvironment that sup- markers (p40),
ports an undifferentiated state (“stemness”) ΔNp63α
and self-renewal of stem cells was proposed by Oct4 KRT14 KRT12
Sox2 KRT15
Schofield in the late 70s [12]. Some of the key
c-Myc KRT19
LSC niche factors are limbal extracellular
Lin28 Sox17
matrix (ECM), the basement membrane, in Surface SSEA4 ABCG2 Connexin 43
particular, as well as soluble growth factors markers TRA- ABCB5 Integrin α2
and survival molecules released by different 1-60 and α6
niche cells, cell–matrix interactions, and cell– TRA- Integrin β1
cell contacts of limbal stem cells with sur- 1-81 and α9
GPHA2
rounding niche cells [13–15]. Anatomically,
TSPAN7
the LSC niche is located deep in the Palisades
of Vogt, radially oriented fibrovascular ridges
of the limbal stroma [16]. Interpalisade ridges scriptomics including GPHA2, CCL20, SOX17,
are occupied by the epithelial pegs that contain and TSPAN7 [19, 20, 30, 31].
limbal stem cells [17]. The full list of positive and negative LSC
Mature corneal epithelial cells express differ- markers is provided in Table 16.1.
entiation markers such as cytokeratin 3 (KRT3)
and 12 (KRT12). Limbal epithelial cells, on the
contrary, express putative stem cell markers and Limbal Stem Cell Deficiency (LSCD)
lack the expression of differentiation markers
such as KRT3 and KRT12 [18]. The unique, LSCs could be depleted or destroyed by numer-
exclusive marker of LSCs, has not yet been dis- ous factors, such as burns, infections, and auto-
covered. There is, however, an extensive set of immune diseases, causing disturbance of the
putative LSC markers proposed so far [4, 19, 20]. corneal regeneration process, which results in an
The correct identification of LSCs relies on a ocular surface disorder called limbal stem cell
combination of the expression of putative LSC deficiency (LSCD). LSCD is a chronic, painful,
markers together with a lack of expression of cor- progressive disorder that leads to conjunctival
neal differentiation markers. The most widely ingrowth (conjunctivalization) and neovascular-
used putative LSC marker is the epithelial tran- ization of the corneal surface, inflammation, cor-
scription factor p63 and its isoform p63α (partic- neal scarring, consequential loss of transparency,
ularly ΔNp63α, which is an N-terminally and eventually loss of vision.
truncated transcript of the alpha isoform of p63) The common causes of LSCD are chemical
[21, 22]. Another form of p63 commonly used is and thermal burns, chronic inflammation, micro-
p40 (ΔNp63) [23]. Other transcription factors bial infections, and extended contact lens wear,
characteristic of LSCs are C/EBPδ and Bmi1 among others. Inherited conditions such as con-
[24]. Different cytoplasmatic proteins (certain genital aniridia, ectodermal dysplasia, and xero-
cytokeratins such as KRT14, KRT15, and derma pigmentosum, but also autoimmune
KRT19) and cell membrane or transmembrane diseases such as Stevens-Johnson syndrome and
proteins (e.g., ABCG2, ABCB5, Notch-1, and ocular cicatricial pemphigoid, could also lead to
CD200) are also proposed as potential positive LSCD [32].
LSC markers [25–29]. Recently, novel LSC LSCD can affect only one eye (unilateral) or
markers were discovered using single-cell tran- both eyes (bilateral) while depending on the
228 S. Bojic et al.
extent of the damage could be classified as either plantation of limbal epithelial tissue pieces, or ex
partial or total [33]. The ablation of the LSC pool vivo cultivated limbal epithelial transplantation
for continuous regeneration of the corneal sur- (CLET), which requires expansion of LSCs in
face results in recurrent corneal erosions or per- vitro before transplantation [4].
sistent epithelial defects accompanied by The choice of LSC source for transplantation
inflammation, and consequential, pain, irritation, depends on whether the disease is unilateral or
tearing, redness, photophobia, epiphora, and bilateral. Acquired forms of LSCD could be uni-
blepharospasm [34, 35]. This may further lead to lateral or bilateral, while autoimmune diseases
corneal scarring, thinning, and even perforation. such as Steven Johnson syndrome usually have
In LSCD, not only does the regenerative stem bilateral manifestations. In the case of unilateral
cell function of the limbus fail but also its barrier LSCD, a limbal biopsy is taken from the healthy
function. As a result, conjunctival epithelium contralateral eye (autologous transplantation). In
along with blood vessels invades the surface of bilateral cases, donor tissue is obtained from a
the cornea leading to conjunctivalization of the living related or unrelated donor, or a cadaver
cornea, a hallmark of LSCD. Other signs of (allogeneic transplantation). In the case of allo-
LSCD are corneal epithelial haze, loss of limbal geneic transplantation, long-term systemic
architecture, persistent epithelial defects, corneal immunosuppression is necessary to try to pre-
neovascularization, corneal scarring, and possi- vent the risk of allograft rejection [4]. Although
bly keratinization [4, 36]. LSC transplantation can improve vision in
Early diagnosis is very important as it can LSCD, evaluation of the 3-year outcomes of
prevent further damage to the ocular surface. allogeneic CLET showed a decrease in LSCD
Although the diagnosis is largely made on clini- severity and an increase in visual acuity up to 12
cal grounds based on slit-lamp examination, months posttreatment, but thereafter LSCD
further diagnostic methods, such as corneal severity score and visual acuity progressively
impression cytology and in vivo confocal deteriorated [38]. Moreover, the 5-year graft sur-
microscopy, are very useful to confirm LSCD vival defined by the absence of recurrence of the
[4, 37]. clinical signs of LSCD was 71% for autologous
The main aims of LSCD treatment are to sup- CLET and 0% for allogeneic LSC transplanta-
port the disturbed or absent corneal regeneration tion [39].
by restoring the number of LSCs by transplanta- Long-term immunosuppressive therapy,
tion and to re-establish the normal LSC niche potential risk of allograft rejection and disease
microenvironment. However, an optimization of transfection, shortage of donors, a limited num-
the corneal surface is a very important initial step ber of LSC passages, and risk of cell/gene con-
as part of the LSCD management. The choice of tamination by 3T3 feeder cells are some of the
treatment depends on the severity and the extent drawbacks of LSC transplantation [40, 41]. To
of the disease. For patients with severe LSCD, overcome the limiting factors of LSC transplan-
ocular surface reconstruction is required. In those tation, alternative sources of cells for cell replace-
cases, cell replacement therapy using autologous ment therapy have been investigated. Other
or allogeneic limbal grafts is the treatment of autologous epithelial stem cells such as conjunc-
choice [4, 36]. tival, epidermal, oral mucosal, and hair follicle
Transplantation of LSCs can be achieved by cells, or allogeneic amniotic epithelial cells have
taking a large biopsy for direct transplantation of been proposed, but none of these cell types repre-
whole limbal grafts from the contralateral healthy sents a better solution than transplantation of
eye in unilateral LSCD or a living-related donor LSCs so far.
in bilateral LSCD (which might result in iatro- Most of the cells were used only in preclinical
genic LSCD). A safer option is taking a smaller models, with exemptions of oral mucosal epithe-
biopsy for either simple limbal epithelial trans- lial stem cells and mesenchymal stem cells that
plantation (SLET), which includes direct trans- have been used in clinical trials [42, 43].
Mesenchymal stem cells (MSCs) are easily Induced Pluripotent Stem Cells
accessible and nonimmunogenic, thereby their (iPSCs)
clinical use would not require immunosuppres-
sion. However, the potential for trans-In 2006, Yamanaka and his team created mouse
differentiation of MSCs of neural crest and iPSCs using retroviral transduction into mouse
mesoderm origin into corneal epithelial cells of somatic cells, by triggering the ectopic expres-
ectodermal origin is still questionable, particu- sion of four specific genes—POU domain class 5
larly in vivo [44, 45]. transcription factor 1 (OCT3/4), the sex-
Other possible sources for the generation of determining region Y-box2 (SOX2), Kruppel-like
LSCs and corneal epithelial cells are pluripotent factor 4 (KLF4), and myelocytomatosis onco-
stem cells such as embryonic stem cells (ESCs) gene (c-MYC), under ESC culture conditions
and induced pluripotent stem cells (iPSCs). [50]. Just a year later, they created the human
Pluripotent stem cells represent a renewable iPSCs [51]. Human iPSCs were generated by ret-
source of autologous cells and are easily expand- roviral transduction of the same four transcrip-
able and bankable [46, 47]. Besides representing tion factors OCT3/4, SOX2, KLF4, and c-MYC,
an unlimited source of autologous cells, the other also known as “OSKM factors” or “Yamanaka’s
advantage of pluripotent stem cells compared to cocktail” [51]. OSKM factors were selected after
tissue-specific, adult stem cells, is their wider dif- testing many genes potentially involved in the
ferentiation potential—they can differentiate into first stages of ESC development [50]. Since then,
virtually any cell type. To date, multiple proto- iPSCs have been produced from a wide variety of
cols have been developed for directed differentia- cell types across different species, suggesting a
tion of both ESCs and iPSCs toward LSCs and universal molecular mechanism behind the
corneal epithelium but were only tested in pre- somatic cell reprogramming [52]. This revolu-
clinical studies. tionary discovery led Yamanaka to be awarded a
The discovery of iPSCs unlocked an abun- Nobel Prize in Physiology and Medicine in 2012.
dance of possibilities in multiple areas of regen- Human iPSCs are remarkably similar to
erative medicine including corneal regeneration. human ESCs in terms of morphology, gene
Although with a similar capacity to iPSCs, ESCs expression, surface antigens, proliferative poten-
are the subject of serious ethical debates and tial, pluripotency, and telomerase activity. They
therefore less interesting for wider application can differentiate into cell types of all three pri-
than iPSCs. The possibility to use cells derived mary germ layers (ectoderm, endoderm, and
from an individual patient’s own iPSCs could mesoderm), both in vitro and in vivo (teratoma
represent a cornerstone of personalized and pre- formation). The formation of teratomas after
cision medicine. The important advantage of iPSC transplantation into immunosuppressed
using a patient’s own iPSCs to derive cells for mice is one of their hallmark characteristics [53].
cell replacement therapy is the lack of immune Their pluripotent potential and capability to dif-
response (rejection) toward the body’s own cells. ferentiate into cells of all three germ layers make
Consequently, autologous cell transplantation iPSCs very valuable for regenerative medicine,
with a patient’s own cells would not require a but at the same time, limit their potential clinical
long-term immunosuppression [48]. However, use, as undifferentiated cells can give rise to tera-
there is still a long way to go to achieve a safe toma upon transplantation [53].
clinical application of cells derived from iPSCs. The use of proto-oncogenes c-MYC and KLF4
The negative side of iPSC clinical use including should be avoided in reprogramming events before
potential mutagenesis and tumorigenesis, high the clinical use of iPSC-derived cells. Reactivation
costs of both iPSCs and iPSC-derived cell proof transgene(s) could lead to tumor formation after
duction, as well as the reproducibility of differen- transplantation [53]. Replacement of c-MYC and
tiation in different clones are yet to be resolved KLF4 by NANOG and LIN28 was also used for the
[49]. successful reprogramming of human somatic cells
230 S. Bojic et al.
[54, 55]. Converting cells from one cell type to logical compatibility, as possible complications
another without prior dedifferentiation into iPSCs related to immune rejection would be eliminated
could eliminate the risk of teratoma formation by the generation and application of autologous
[53], in a process known as direct trans-differenti- iPSCs from individual patients.
ation which will be explained later. Directed differentiation protocols were devel-
Since iPSCs’ discovery, different cellular repro- oped for the generation of various cell types.
gramming strategies have been introduced, which Sequential addition of small molecules, growth
could be classified into two major groups—inte- factors, vitamins, survival factors, and exposure
grating and non-integrating. The integrating strate- to different substrates that are part of the natural
gies rely on viral transduction of retrovirus or stem cell niche induce and direct iPSC differen-
lentivirus, or transposons [56, 57]. The nonintegrat- tiation toward desired cell types. In the absence
ing approach includes the use of adenovirus, Sendai of inductive cues, spontaneous differentiation of
virus, episomal vectors, synthetic mRNAs, recom- iPSCs occurs, and they spontaneously shift
binant proteins, or small molecules [58–60]. toward the neuroectodermal fate [65, 66].
Initially, human skin fibroblasts were used as Takahashi and Yamanaka’s breakthrough dis-
a starting material for iPSC generation. However, covery of the underlying molecular mechanism
skin fibroblasts are not suitable for large-scale of somatic cell reprogramming opened a virtually
production, which creates a need for repeated unlimited number of possibilities for their use in
invasive skin biopsies [61]. Peripheral blood rep- human models of disease and development,
resents a particularly desirable source, as it can regenerative medicine, drug screening and dis-
be routinely obtained without the need for inva- covery, and toxicological research [18, 67, 68].
sive procedures. So far, it seems that any somatic The possibility of deriving human iPSCs from
cell can be reprogrammed to become iPSCs but almost any somatic cell has created an invaluable
with variable efficiencies [62]. The origin of cells opportunity for studying specific hereditary dis-
used for the generation of iPSCs might play a eases. Patient-specific disease models can help in
very important role in the iPSC phenotype and identifying new disease biomarkers, which can
transplantation outcome [63]. Residual epigene- support earlier diagnosis, or serve as a ground for
tic memory might be responsible for the propen- developing novel screening procedures. iPSC
sity of iPSC cell lines to preferentially disease models can also help to identify new
differentiate toward cell types of their origin, compounds capable of alleviating disease pathol-
especially in the initial passages. ogy in vitro, thereby supporting drug discovery
Another important point to consider is the and screening. Directed differentiation of iPSCs
rather common use of feeder cells and animal- toward different tissues creates a ground for the
derived products in the iPSC generation process. generation of future advanced therapeutic medic-
Murine-derived feeder cells are widely used for inal products (ATMPs) for tissue and organ
iPSC production. Conventional iPSC culture sys- repair, offering ground-breaking new opportuni-
tems also use serum-containing media, while ties in the field of regenerative medicine. Normal
fetal bovine serum (FBS)-containing medium is developmental processes, such as ocular morpho-
routinely used for the culture of feeder cells. genesis, can also be studied using iPSC differen-
However, the reduction or complete removal of tiation models [2, 69, 70].
serum and animal-derived products and imple-
mentation of feeder-free culture systems is a
mandatory step for the generation of clinical- i PSC Differentiation to Corneal
grade iPSCs [64]. Epithelial Cells
Importantly, since iPSCs are not derived from
human embryos, their use circumvents moral and Since 2012 when the first two studies on methods
ethical considerations related to the use of ESCs. for directed differentiation of iPSCs toward cor-
Another great advantage of iPSCs is immuno- neal epithelial cells were reported, multiple
research groups published protocols for the gen- replicating the corneal stem cell niche by expos-
eration of corneal and limbal epithelial cells from ing human ESCs seeded on collagen IV to a
iPSCs. We will summarize those studies with a medium conditioned by limbal fibroblast to
focus on the directed differentiation of human induce corneal epithelial differentiation led to a
iPSCs toward corneal epithelium. The relevance successful generation of corneal epithelial-like
of studies related to murine iPSC-derived corneal cells. The method published by Ahmad et al.
epithelium from the clinical point of view is inspired multiple future studies on iPSCs and
much lower due to the differences between mouse served as a baseline for the development of many
and human physiology. iPSC differentiation protocols toward corneal
The microenvironment of a stem cell niche, epithelial-like cells.
including various ECM proteins and secreted In vitro mimicking of LSC niche cues could
growth factors, as well as physical characteristics be achieved using a conditioned medium (condi-
of the surrounding tissue, plays a vital role in the tioned by either limbal or corneal fibroblasts),
determination of stem cell fate [71]. Culture sub- feeder cells (such as 3T3-J2 mouse embryonic
strates (surfaces on which cells live and grow) fibroblasts or PA6 mouse stromal cell line), or
such as feeder cells, human amniotic membrane, natural scaffolds (such as denuded human amni-
or various ECM proteins, together with soluble otic membrane—HAM, decellularized human
components in the culture medium (e.g., vita- organ-cultured corneas, or corneal stromal
mins, growth factors, survival molecules, among lenticules).
others), provide cues for iPSC fate decision and
specification. Collagen IV and laminin in the lim- i PSCs Differentiation Toward Corneal
bal basement membrane, as well as limbal fibro- Epithelial Cells Using Conditioned
blasts in the stroma, are some of the important Medium
components of the LSC niche [14]. Therefore, In 2012, using an improved protocol for ESC
collagen IV, a key component of limbal stroma, differentiation toward corneal epithelium pub-
and laminin, a basement membrane protein, are lished by Ahmad et al. [73], Shalom-Feuerstein
widely used as substrates for cell cultivation [72]. and colleagues successfully induced corneal dif-
Many of the studies tried to mimic the LSC ferentiation in iPSCs and developed a cellular
niche environment in order to induce and direct model of iPSC-derived corneal epithelial cells in
differentiation toward corneal epithelium by a follow-up study. They reported using BMP4
using collagen IV or laminin coating as a sub- coupled with collagen IV and corneal fibro-
strate for cell cultivation, in combination with blasts-derived conditioned medium, in a proto-
conditioned medium produced by corneal or lim- col that recapitulates corneal epithelial linage
bal fibroblasts. Other studies, on the other hand, development, for induction of corneal epithelial
tried to mimic natural ocular surface develop- differentiation of iPSCs [68]. They introduced
ment by inducing a cascade of signals that are two major modifications to the method designed
known to direct the development of the corneal for ESC differentiation toward corneal epithe-
epithelium. lium by Ahmad et al. [73]. Instead of limbal
fibroblasts, they used corneal fibroblasts, iso-
lated from the entire cornea. Moreover, they
i PSCs Differentiation Toward Corneal added BMP4, a major regulator of embryonic
Epithelial Cells Using Limbal epithelial commitment and found that this
Niche Cues enhanced the efficiency of corneal epithelial dif-
ferentiation. Most of the cells generated using
The first study reporting the successful differen- this protocol expressed mature corneal epithelial
tiation of human pluripotent stem cells into cor- cell marker KRT3 (>90%), while 20–25% of
neal epithelial cells was published in 2007, by cells were limbal epithelial progenitor cells as
Ahmad and colleagues [73]. Their concept of evidenced by the expression of KRT14 [68].
232 S. Bojic et al.
i PSCs Differentiation Toward Corneal bal commitment of iPSCs. Moreover, generated

Epithelial Cells Using Feeder Cells cells were capable of further maturation toward
In 2012, Hayashi et al. used dermal fibroblasts corneal epithelial cells [18].
and, for the first time, limbal epithelial cells, to
derive iPSCs which were then directed toward i PSCs Differentiation Toward Corneal
corneal epithelial differentiation using the so- Epithelial Cells Using Natural Scaffolds
called “stromal cell-derived inducing activity (Denuded HAM, Decellularized Human
(SDIA) method” [74]. SDIA method was initially Organ-Cultured Corneas, or Corneal
used for the induction of neuroectodermal differ- Stromal Lenticules)
entiation of iPSCs toward dopaminergic neurons Sareen et al. published a study that maintained
and retinal pigmented epithelium (RPE) [75]. In the iPSCs on a natural niche replacement repre-
this study, Hayashi and colleagues showed that sented by de-epithelialized human organ-cultured
the long-term SDIA differentiation method could corneas and denuded HAM, both closely resem-
induce the generation of corneal epithelial cells, bling limbal basement membrane in composi-
but only after the induction of differentiation of tion. In this study again limbal-derived iPSCs
neural, RPE, and lens cells. This finding is con- cultured on de-epithelialized human corneas,
sistent with the ocular development timeline, as showed more advanced differentiation, evidenced
corneal epithelial cells develop after RPE and by the expression of the mature corneal epithelial
lens, supporting the authors’ theory that the markers, and compared to fibroblast-derived
“SDIA differentiation method in vitro mimics the iPSCs cultured on HAM. A certain level of reten-
process of ocular development in vivo” [74]. tion of methylation-related epigenetic signatures
Importantly, limbal epithelial cell-derived iPSCs in limbal-derived iPSCs could support the limbal
exhibited higher corneal epithelial marker expres- epithelial differentiation [46].
sion and larger corneal epithelial cell colony Qin et al. proposed the utilization of decellu-
numbers than dermal fibroblast-derived iPSCs. larized corneal stromal lenticules, by-products of
This difference may be caused by the retention of small incision lenticule extraction (SMILE), as
the epigenomic signature of their original parent potential scaffolds that could support the survival
cells after reprogramming. and proliferation of corneal epithelial-like cells
In 2017, Aberdam et al. published an improved derived from iPSCs [76]. Lenticules were decel-
protocol based on their previous work [68]. In lularized to remove any cellular or nuclear mate-
their previous study, most of the cells produced rial that could potentially cause allograft
were mature corneal epithelial cells [68]. In this rejection. The growth of corneal epithelial stem
study, they thoroughly modified their previous cells on lenticules was previously investigated,
culture conditions in an attempt to produce a pure and it was demonstrated that they were capable
population of iPSC-derived limbal epithelial of proliferating for at least three passages in cul-
cells, able to further differentiate [18]. In the first ture, suggesting that lenticules might not inter-
phase, they introduced feeders made of irradiated fere with the stemness and proliferative potential
fibroblasts isolated from the peripheral cornea of corneal epithelial stem cells [77].
which were seeded on collagen IV-coated dishes.
In phase two, cells were seeded on collagen IV
coated dishes containing irradiated mouse 3T3- i PSCs Differentiation Toward Corneal
J2 feeder layer. Furthermore, the addition of Epithelial Cells Using Defined Factors
TGFβ inhibitor, the cell survival rock inhibitor
(Y-27632), EGF (epidermal growth factor) or While previous studies on pluripotent stem cells
KGF (keratinocyte growth factor) was added at induction of corneal differentiation relied on var-
different time points of the newly developed pro- ious undefined or animal-derived components
tocol. This modified protocol improved the lim- (such as feeder cells, amniotic membrane or use
of the conditioned medium, alone or in combina- i PSCs Differentiation Toward Corneal

tions), in 2014, Mikhailova et al. proposed for Epithelial Cells Using a Combination
the first time the use of defined factors for directed of Limbal Niche Cues with Defined
differentiation of iPSCs toward corneal epithelial Factors Containing Animal-Derived
progenitors. They used two small molecular Components
inhibitors (i.e., SB-505124 and IWP-2) in combi- In their attempt to better reflect the complexity of
nation with basic fibroblast growth factor (bFGF) the whole eye development, Hayashi et al. gener-
to mimic “signalling cues active during ocular ated a so-called “self-formed ectodermal autono-
surface ectoderm development” [78]. mous multi-zone (SEAM)” of ocular cells derived
They showed that inhibition of the TGFβ from iPSCs [69]. A proportion of iPSCs exposed
pathway using SB-505124 and the Wnt signal- to this protocol spontaneously formed circular
ling pathway using IWP-2 together with acti- 2D colonies composed of four concentric zones:
vation of the FGF signaling pathway directed the innermost zone 1 containing neuronal cells,
iPSC differentiation toward a relatively pure zone 2 containing neural crest cell-like cells and
population of corneal epithelial-like progeni- retina-like cells (neuro-retina and RPE), zone 3
tor cells, capable of terminal differentiation containing ocular surface epithelial-like cells
into mature corneal epithelial-like cells [78]. (corneal, limbal, and conjunctival-like cells), and
In their next study, the same group of authors the outermost zone 4 with nonocular surface
evaluated the protein expression of both ESC- epithelial-like cells. Lens-like cells could be
and iPSC-derived limbal epithelial stem cells found at the margin of zones 2 and 3. Each zone
and compared it to native human corneal epi- had distinctive cell morphology, with the visible
thelial cells [79]. They identified 860 unique delineation with adjacent zones. The innermost
proteins present in all three samples and zone was formed first, followed by the emergence
showed that protein expression profiles were of three more radial concentric cell zones. To
nearly identical in limbal epithelial cells some extent, the concentric SEAM seems to
derived from ESCs and iPSCs, proving that mimic the whole eye development, which might
their differentiation protocol is reproducible be useful for studies of ocular morphogenesis
and leads to the production of homogeneous [69]. In this study, a laminin 511 E8 fragment
cell populations [79]. Mikhailova et al. further (LN511E8) was used as a substrate in combina-
investigated the use of the cells generated by tion with defined factors for the full length of the
their protocol on bioengineered collagen protocol. Corneal epithelium-like cells generated
matrices in serum-free conditions for potential using this protocol can be purified and sorted by
clinical application showing that the prolifera- fluorescence-activated cell sorting (FACS),
tion of cells on bioengineered matrices was expanded, and differentiated to form a transplant-
significantly higher than on collagen-coated able corneal epithelial sheet that was able to
control wells [80]. recover corneal function in an animal model of
LSCD [69, 70].
Shibata et al. continued work previously done
i PSCs Differentiation Toward Corneal by Hayashi et al. (the same research group) and
Epithelial Cells Using a Combination reported that laminin isoforms differentially
of Limbal Niche Cues with Defined regulate the ocular cell differentiation from

Factors iPSCs, and that SEAM contains four concentric
zones only when iPSCs are cultivated on
Many studies use a combined approach, a combi- LN511E8 form of laminin [71]. To investigate
nation of limbal niche cues (different ECM com- the capability of various laminin isoforms to
ponents such as collagen IV, laminin, or Matrigel) facilitate the generation of iPSC-derived corneal
with defined chemical factors. epithelial cells, iPSCs were differentiated on five
234 S. Bojic et al.
different laminin isoforms (i.e., LN111E8, which led to low expression of BMP target genes.
LN211E8, LN332E8, LN411E8, and LN511E8). The addition of SB431542, influenced the bal-
iPSCs differentiated on LN332E8 isoform gener- ance of co-SMADS in favor of BMP signaling,
ated the highest proportion of corneal epithelial- leading to successful differentiation of the nonre-
like cells. Their results suggest that iPSCs sponsive iPSC line toward corneal epithelial
differentiation on LN332E8 enhances the yield progenitor-like cells [8].
of iPSC-derived corneal epithelial cells, and it In 2019, Li et al. published a stepwise, chemi-
shortens the culture period needed for corneal cally defined method to induce the differentiation
epithelial-like cell sheet preparation compared to of “multizone ocular cells (MZOC)” from iPSCs,
iPSC differentiation on LN511E8 [71]. Although which contained differentiated cell types includ-
successful in corneal epithelial-like cell sheet ing the neural retina, retinal pigment epithelium,
preparation, the SEAM method is very time- surface ectoderm, neural crest, and lens cells.
consuming, complex, and requires a certain level Resembling the SEAM method, the surface ecto-
of expertise, therefore, the outcome is not easily derm zone of MZOC could be mechanically iso-
reproducible. Better yield, shorter differentiation lated and further induced into corneal
time, and simpler methodology make this proto- epithelial-like cells [2]. The differentiating cells
col easier to follow and reproduce compared to “spontaneously and progressively formed five
the one on LN511E8. Moreover, Shibata et al. identifiable zones” (zones 1 and 2 formed first,
further simplified the protocol excluding the need followed by zones 3 and 4, with zone 5 formed
for FACS sorting, by introducing magnetic- last). Zone 1 expressed the neural retina markers,
activated cell sorting as an alternative to further zone 2 RPE markers, and zone 3 the surface ecto-
facilitate the efficient production of transplant- dermal markers together with the LSC marker, in
able corneal epithelial-like cell sheets [71]. the peripheral region. Zone 4 cells primarily
In 2018, Kamarudin et al. published their expressed neural crest markers and zone 5 lens
defined feeder-free monolayer differentiation markers. The ocular surface ectoderm (zone 3)
method [8]. Induction of corneal differentiation was mechanically isolated from MZOCs and
was achieved in pluripotent stem cells (two iPSC directed into corneal epithelial cells. Comparable
lines and one ESC line) by supplementation of with LSC reaching replicative senescence
growth factors and small molecules during the in vitro after a certain number of passages, the
two-stage differentiation protocol. For the first iPSC-derived corneal epithelial-like cells also
seven days (the first stage of differentiation) exhibited limited proliferative capacity in a con-
ESCs and iPSCs seeded on Matrigel-coated tinuous passage. Therefore, further optimization
plates were exposed to BMP4, all-trans-retinoic of this method is necessary for the long-term
acid (RA) and EGF, alone or in combination. expansion of these cells [2].
BMP4 and RA were selected according to their
capability to promote non-neural ectodermal eno-Free iPSCs Differentiation Toward
X
commitment. EGF, on the other hand, stimulates Corneal Epithelial Cells Using
the proliferation of corneal epithelial progenitors. a Combination of Limbal Niche Cues
Importantly, this study revealed certain intra-line with Defined Factors
differences in the capability of iPSCs to differen- To fulfil the need for more standardized, xeno-
tiate into corneal epithelial-like cells which were free methods for the generation of iPSC-derived
dependent on the level of endogenous BMP sig- ocular cells for transplantation purposes,
naling and could be restored via the activation of Hongisto et al. developed a protocol for the gen-
this pathway by a specific TGFβ inhibitor eration of a relatively pure population of limbal
SB431542. The nonresponsive iPSC line had a epithelial-like cells and RPE-like cells from plu-
lower level of BMP signaling activity due to ripotent stem cells (both ESCs and iPSCs), to
lower expression of receptors and effectors, help with efficient and large-scale production of
autologous cells for cell therapies. They pro- Direct Trans-differentiation

posed a relatively simple, short, and easily
reproducible protocol that relies on the utiliza- The risk of teratoma formation hinders the clini-
tion of defined molecular cues and circumvents cal use of iPSC-derived corneal epithelial cells. If
the use of undefined animal components, such rigorous purification of corneal epithelial-like
as serum and feeders. iPSCs differentiation cells is not performed adequately, and some
toward limbal epithelial cells included two undifferentiated iPSCs remained, their remaining
phases: the first one, corneal induction, con- pluripotent capacity has the potential to form
ducted in suspension culture (without any plate teratomas after transplantation [53].
coating) and the second one, corneal differentia- An alternative to the directed differentiation of
tion, in adherent culture (using a combination of iPSCs into corneal epithelial-like cells is a direct
collagen IV and laminin-521 used for coating). reprogramming method (also known as direct
iPSCs-derived limbal epithelial- like cells trans-differentiation), which is a process wherein
derived by this protocol were capable of further mature, fully differentiated somatic cells are
maturation and differentiation after exposure to induced to become the other cell type without
appropriate conditions [65]. Besides proposing going through the intermediate state of pluripo-
the xeno-free differentiation protocol, they also tency (iPSC stage) [82]. Following that approach,
implemented a monolayer culture of undifferen- rat hair follicle stem cells were transdifferenti-
tiated iPSCs on recombinant laminin-521 in a ated into corneal epithelial-like cells by induced
defined, commercially available xeno-free overexpression of PAX6 together with the use of
medium before the start of induction, ensuring soluble factors [83], as well as murine vibrissa
the xeno-free generation of starting cell material hair follicle stem cells [84]. Moreover, human
for further differentiation. Moreover, they also skin-derived precursor cells were directly trans-
developed a xeno-free cryopreservation method differentiated into corneal epithelial-like cells by
to provide a readily available stock of iPSC- culturing them with a set of three growth factors
derived limbal epithelial stem cells for potential (EGF, KGF, and HGF) [85]. Trans-differentiation
cell therapy, as well as for quality and safety of stem cells from human-exfoliated deciduous
testing. Importantly, the purity of derived cells teeth toward a phenotype of corneal epithelium
improved after cryopreservation, and even fur- was also reported [86]. More recently, Kitazawa
ther after passaging and prolonged culture, et al. reported direct reprogramming of human
which was documented by the increased expres- fibroblasts into corneal epithelial cells by overex-
sion of the putative stem cell marker ΔNp63 pression of PAX6, OVOL2, and KLF4 [87]. Some
[65]. In 2018, the same protocol was published of the advantages of direct reprogramming are
as a detailed video article, showing step-by-step the circumvention of the pluripotent state, which
instructions for robust, xeno-and feeder-free is potentially tumorigenic, and carcinogenic, but
production of limbal epithelial stem cells from also avoiding expensive and lengthy production
pluripotent stem cells [81]. and characterization of iPSC lines [88].
Although the feeder-free culture might be To examine the trans-differentiation approach,
more scalable and less laborious than feeder- Cieslar-Pobuda and colleagues tried different
based systems, the major drawback of this pro- combinations of transcription factors—a combi-
tocol, and most likely feeder-free iPSC culture nation of three transcription factors important for
in general, is that prolonged feeder-free culture limbal epithelial development ΔNp63, TCF4, and
using single cell passaging led to an accumula- C/EBPδ, plus either OCT4 or KLF4 or both. Cells
tion of karyotype changes. Therefore, frequent infected with any combination of transcription
karyotyping is necessary, and the use of low factors showed the same strong expression of
passage iPSCs for differentiation is required mature corneal epithelial cell markers KRT3 and
[65]. KRT12, but cells infected with ΔNp63, TCF4,
236 S. Bojic et al.
and C/EBPδ proliferated and differentiated much commercially available, but it was documented
faster than cells transduced with four or five fac- that long-term feeder-free culture of iPSCs leads
tors. Although KRT3 and KRT12 expression sug- to accumulation of karyotypic abnormalities
gests a change toward the corneal epithelial state, which could influence growth and differentiation
no further markers were examined, nor clonoge- but also the safety profile of the final cell product
nicity of generated cells [53]. [65, 81]. Therefore, further optimization of pro-
tocols for iPSC generation and differentiation
toward corneal epithelial cells is necessary.
Conclusions Multiple groups published methods for prefer-
ential differentiation of iPSCs into corneal and
Recently, in the first clinical trial based on iPSC- limbal epithelial cells. It has been documented
derived cells, iPSC-derived RPE cells were trans- that different iPSC clones show a different pro-
planted into a patient with neovascular age-related pensity for corneal differentiation, likely caused
macular degeneration [89]. iPSC-derived corneal by differences in levels of endogenous signaling
epithelial cells might be the next candidate for pathways involved in corneal epithelial develop-
transplantation as LSCD is a promising target for ment (such as BMP4) or their epigenetic status.
future transplantation trials [65, 90]. This issue might be resolved by further optimiza-
Only a few thousand p63-bright limbal epithe- tion of the protocol specifically for each iPSC
lial cells are required for cell transplantation onto line, which unfortunately is a time-consuming,
the human eye. A minimum of approximately expensive, and nonsustainable process.
p63-bright 3000 cells is required for successful Most of the protocols include two steps: the
transplantation as reported by Rama et al. [22]. induction phase and the differentiation phase.
One of the limiting factors in LSC transplanta- In the induction phase direction of iPSC dif-
tion is the fact that primary ex vivo expanded ferentiation toward surface ectoderm is essential.
LSCs undergo rapid replicative senescence in TGFβ and Wnt antagonists as well as the FGF
vitro. iPSCs as a potentially unlimited source of pathway activators are used to induce ectodermal
corneal epithelial cells opened the possibility to differentiation, while BMP4 is used to propel dif-
overcome replicative senescence of LSCs and ferentiation toward surface ectoderm.
enable the production of sufficient cell numbers Components of basement membrane such as col-
for transplantation and cell banking. Moreover, lagen IV and different isoforms of laminin are
iPSCs can be obtained from the patient’s own used to mimic the natural niche microenviron-
cells, and then directed toward differentiation ment. Monolayer culture might be a more easily
into the desired cell type, making them suitable reproducible, quicker method compared to 3D
for autologous cell transplantation. However, corneal organoids, or 2D SEAM which are lon-
undefined culture conditions and the use of ger, more complex and difficult to standardize
animal-derived components impede the clinical across different laboratories and different iPSC
translation of iPSC-derived cells into therapeu- lines [81].
tics. GMP-compliant generation, culture, and dif- A rigorous purification to get a highly homog-
ferentiation protocols are mandatory for enous cell population, safe for transplantation
cell-based medicinal products. There is certainly purposes, is necessary before clinical utilization
a need for more standardized, xenogeneic-free as sufficient purity is mandatory for cell thera-
protocols for iPSC generation, culture, and dif- pies. The final product must be devoid of any
ferentiation together with reasonable financial undifferentiated cells, noncorneal iPSC-derived
costs. Strictly defined conditions would prevent cells or cellular impurities that might originate
batch-to-batch variations, which is an important from feeder cells. For that reason, improved
advantage of serum-free, xeno-free, and feeder- protocols for highly homogenous production of
free protocols. Several defined xeno-free proto- corneal epithelial cells from iPSCs are necessary
cols for a prolonged culture of iPSCs are already to consider their potential future use in cell ther-
apy. Hayashi et al. included the FACS purifica- combination with a conditioned medium pro-
tion step to gain a homogenous population of duced by corneal or limbal fibroblasts;
corneal epithelial cells, including corneal stem • Mimicking the ocular surface development is
cells, before transplantation of ex vivo expanded achieved by inducing a cascade of defined sig-
corneal epithelial cell sheets in an experimentally nals known to direct the development of cor-
induced rabbit model of LSCD. Shibata et al. fur- neal epithelium;
ther simplified the purification process, by intro- • Improved protocols for highly homogenous
ducing magnetic-activated cell sorting as an production of corneal epithelial cells from
alternative to FACS, intending to facilitate effi- induced pluripotent stem cells as well as
cient production of the transplantable corneal GMP-compliant generation, culture, and dif-
epithelial-like cell sheets. So far, iPSC-derived ferentiation protocols are necessary to con-
corneal epithelial cell sheets are transplanted suc- sider their use in cell therapy.
cessfully in an experimentally induced animal
model of LSCD. For the next step, in human
transplantation of iPSC-derived corneal and lim-
bal epithelial cells, the development of safe, References
robust, and GMP-compliant differentiation pro-
1. Resnikoff S, Pascolini D, Mariotti SP, Pokharel
tocols is necessary. Although Hongisto et al. GP. Global magnitude of visual impairment caused
developed a xeno-free protocol, no GMP- by uncorrected refractive errors in 2004. Bull
compliant protocols for iPSC differentiation World Health Organ. 2008;86(1):63–70. https://doi.
toward corneal epithelium have been published org/10.2471/blt.07.041210.
2. Li Z, Duan H, Li W, Hu X, Jia Y, Zhao C, et al. Rapid
yet. Adaptation of research protocols relying on differentiation of multi-zone ocular cells from human
research-grade ingredients to clinical-grade pro- induced pluripotent stem cells and generation of
tocols using GMP-grade reagents is a crucial step corneal epithelial and endothelial cells. Stem Cells
in the clinical translation of iPSC-derived corneal Dev. 2019;28(7):454–63. https://doi.org/10.1089/
scd.2018.0176.
and limbal epithelial cells. 3. Whitcher JP, Srinivasan M, Upadhyay MP. Corneal
Besides their therapeutical use in regenerative blindness: a global perspective. Bull World Health
medicine, iPSC-derived corneal and limbal epi- Organ. 2001;79(3):214–21.
thelial cells can be potentially used for research 4. Bojic S. Optimisation of protocols for ex vivo expan-
sion of limbal stem cells and their enrichment.
purposes, including investigation of ocular mor- Doctoral dissertation, Institute of Genetic Medicine:
phogenesis, disease modeling, drug screening, Newcastle University, Newcastle upon Tyne, United
toxicity testing, or genetic engineering. Kingdom; 2020.
5. Eghrari AO, Riazuddin SA, Gottsch JD. Overview
of the cornea: structure, function, and development.
Take Home Notes Prog Mol Biol Transl Sci. 2015;134:7–23. https://doi.
org/10.1016/bs.pmbts.2015.04.001.
• Induced pluripotent stem cells represent an 6. Cotsarelis G, Cheng S-Z, Dong G, Sun T-T, Lavker
unlimited source of autologous cells and a RM. Existence of slow-cycling limbal epithe-
lial basal cells that can be preferentially stimu-
powerful resource for cell-based replacement lated to proliferate: implications on epithelial
corneal therapies; stem cells. Cell. 1989;57(2):201–9. https://doi.
• Two main strategies are used for the induction org/10.1016/0092-8674(89)90958-6.
of corneal epithelial differentiation of induced 7. Thoft RA, Friend J. The X, Y, Z hypothesis of corneal
epithelial maintenance. Invest Ophthalmol Vis Sci.
pluripotent stem cells (alone or in combina- 1983;24(10):1442–3.
tion)—mimicking the LSC niche environment 8. Kamarudin TA, Bojic S, Collin J, Yu M, Alharthi S,
and mimicking natural ocular surface Buck H, et al. Differences in the activity of endog-
development; enous bone morphogenetic protein signaling impact
on the ability of induced pluripotent stem cells to
• Mimicking the limbal stem cell niche environ- differentiate to corneal epithelial-like cells. Stem
ment is achieved by using feeder cells or vari- Cells. 2018;36(3):337–48. https://doi.org/10.1002/
ous extracellular protein matrix coatings in stem.2750.
238 S. Bojic et al.
9. Weissman IL. Stem cells: units of development, 2010;363(2):147–55. https://doi.org/10.1056/

units of regeneration, and units in evolution. Cell. NEJMoa0905955.
2000;100(1):157–68. https://doi.org/10.1016/ 23. Yu M, Bojic S, Figueiredo GS, Rooney P, de
s0092-8674(00)81692-x. Havilland J, Dickinson A, et al. An important role
10. Takács L, Tóth E, Berta A, Vereb G. Stem cells of the for adenine, cholera toxin, hydrocortisone and tri-
adult cornea: from cytometric markers to therapeutic iodothyronine in the proliferation, self-renewal and
applications. Cytometry A. 2009;75(1):54–66. https:// differentiation of limbal stem cells in vitro. Exp Eye
doi.org/10.1002/cyto.a.20671. Res. 2016;152:113–22. https://doi.org/10.1016/j.
11. Mushtaq M, Kovalevska L, Darekar S, Abramsson exer.2016.09.008.
A, Zetterberg H, Kashuba V, et al. Cell stemness is 24. Barbaro V, Testa A, Di Iorio E, Mavilio F, Pellegrini
maintained upon concurrent expression of RB and G, De Luca M. C/EBPdelta regulates cell cycle
the mitochondrial ribosomal protein S18-2. Proc and self-renewal of human limbal stem cells. J Cell
Natl Acad Sci. 2020;117(27):15673–83. https://doi. Biol. 2007;177(6):1037–49. https://doi.org/10.1083/
org/10.1073/pnas.1922535117. jcb.200703003.
12. Schofield R. The relationship between the spleen 25. Bojic S, Hallam D, Alcada N, Ghareeb A, Queen
colony-forming cell and the haemopoietic stem cell. R, Pervinder S, et al. CD200 expression marks
Blood Cells. 1978;4(1-2):7–25. a population of quiescent limbal epithelial stem
13. Dziasko MA, Daniels JT. Anatomical features and cells with holoclone forming ability. Stem Cells.
cell-cell interactions in the human limbal epithe- 2018;36(11):1723–35. https://doi.org/10.1002/
lial stem cell niche. Ocul Surf. 2016;14(3):322–30. stem.2903.
https://doi.org/10.1016/j.jtos.2016.04.002. 26. Umemoto T, Yamato M, Nishida K, Yang J, Tano
14. Mei H, Gonzalez S, Deng SX. Extracellular matrix Y, Okano T. Limbal epithelial side-population cells
is an important component of limbal stem cell niche. have stem cell-like properties, including quiescent
J Funct Biomater. 2012;3(4):879–94. https://doi. state. Stem Cells. 2006;24(1):86–94. https://doi.
org/10.3390/jfb3040879. org/10.1634/stemcells.2005-0064.
15. Redondo PA, Pavlou M, Loizidou M, Cheema 27. Ksander BR, Kolovou PE, Wilson BJ, Saab KR, Guo
U. Elements of the niche for adult stem cell expan- Q, Ma J, et al. ABCB5 is a limbal stem cell gene
sion. J Tissue Eng. 2017;8:2041731417725464. required for corneal development and repair. Nature.
https://doi.org/10.1177/2041731417725464. 2014;511(7509):353–7. https://doi.org/10.1038/
16. Davanger M, Evensen A. Role of the pericorneal nature13426.
papillary structure in renewal of corneal epithe- 28. Thomas PB, Liu YH, Zhuang FF, Selvam S, Song
lium. Nature. 1971;229(5286):560–1. https://doi. SW, Smith RE, et al. Identification of notch-1
org/10.1038/229560a0. expression in the limbal basal epithelium. Mol Vis.
17. Townsend WM. The limbal palisades of Vogt. Trans 2007;13:337–44.
Am Ophthalmol Soc. 1991;89:721–56. 29. Merjava S, Neuwirth A, Tanzerova M, Jirsova K. The
18. Aberdam E, Petit I, Sangari L, Aberdam D. Induced spectrum of cytokeratins expressed in the adult
pluripotent stem cell-derived limbal epithelial cells human cornea, limbus and perilimbal conjunctiva.
(LiPSC) as a cellular alternative for in vitro ocular Histol Histopathol. 2011;26(3):323–31. https://doi.
toxicity testing. PLoS One. 2017;12(6):e0179913. org/10.14670/hh-26.323.
https://doi.org/10.1371/journal.pone.0179913. 30. Li DQ, Kim S, Li JM, Gao Q, Choi J, Bian F, et al.
19. Collin J, Queen R, Zerti D, Bojic S, Dorgau B, Single-cell transcriptomics identifies limbal stem cell
Moyse N, et al. A single cell atlas of human cornea population and cell types mapping its differentiation
that defines its development, limbal progenitor cells trajectory in limbal basal epithelium of human cornea.
and their interactions with the immune cells. Ocul Ocul Surf. 2021;20:20–32. https://doi.org/10.1016/j.
Surf. 2021;21:279–98. https://doi.org/10.1016/j. jtos.2020.12.004.
jtos.2021.03.010. 31. Dou S, Wang Q, Qi X, Zhang B, Jiang H, Chen S,
20. Altshuler A, Amitai-Lange A, Tarazi N, Dey S, et al. Molecular identity of human limbal heteroge-
Strinkovsky L, Hadad-Porat S, et al. Discrete lim- neity involved in corneal homeostasis and privilege.
bal epithelial stem cell populations mediate corneal Ocul Surf. 2021;21:206–20. https://doi.org/10.1016/j.
homeostasis and wound healing. Cell Stem Cell. jtos.2021.04.010.
2021;28(7):1248–61. https://doi.org/10.1016/j. 32. Ahmad S. Concise review: limbal stem cell defi-
stem.2021.04.003. ciency, dysfunction, and distress. Stem Cells Transl
21. Di Iorio E, Barbaro V, Ruzza A, Ponzin D, Pellegrini Med. 2012;1(2):110–5. https://doi.org/10.5966/
G, De Luca M. Isoforms of DeltaNp63 and the migra- sctm.2011-0037.
tion of ocular limbal cells in human corneal regenera- 33. Kolli S, Ahmad S, Lako M, Figueiredo F. Successful
tion. Proc Natl Acad Sci U S A. 2005;102(27):9523–8. clinical implementation of corneal epithelial stem cell
https://doi.org/10.1073/pnas.0503437102. therapy for treatment of unilateral limbal stem cell
22. Rama P, Matuska S, Paganoni G, Spinelli A, De deficiency. Stem Cells. 2010;28(3):597–610. https://
Luca M, Pellegrini G. Limbal stem-cell therapy doi.org/10.1002/stem.276.
and long-term corneal regeneration. N Engl J Med.
34. Espana EM, Grueterich M, Romano AC, Touhami 45. Nieto-Nicolau N, Martín-Antonio B, Müller-Sánchez
A, Tseng SC. Idiopathic limbal stem cell deficiency. C, Casaroli-Marano RP. In vitro potential of human
Ophthalmology. 2002;109(11):2004–10. https://doi. mesenchymal stem cells for corneal epithelial regen-
org/10.1016/s0161-6420(02)01250-2. eration. Regen Med. 2020;15(3):1409–26. https://doi.
35. Utheim TP, Aass Utheim Ø, Salvanos P, Jackson org/10.2217/rme-2019-0067.
CJ, Schrader S, Geerling G, et al. Concise Review: 46. Sareen D, Saghizadeh M, Ornelas L, Winkler MA,
Altered versus unaltered amniotic membrane as a Narwani K, Sahabian A, et al. Differentiation of
substrate for limbal epithelial cells. Stem Cells Transl human limbal-derived induced pluripotent stem
Med. 2018;7(5):415–27. https://doi.org/10.1002/ cells into limbal-like epithelium. Stem Cells Transl
sctm.17-0257. Med. 2014;3(9):1002–12. https://doi.org/10.5966/
36. Ghareeb AE, Lako M, Figueiredo FC. Recent sctm.2014-0076.
advances in stem cell therapy for limbal stem 47. Huang CY, Liu CL, Ting CY, Chiu YT, Cheng YC,
cell deficiency: a narrative review. Ophthalmol Nicholson MW, et al. Human iPSC banking: barri-
Therapy. 2020;9(4):809–31. https://doi.org/10.1007/ ers and opportunities. J Biomed Sci. 2019;26(1):87.
s40123-020-00305-2. https://doi.org/10.1186/s12929-019-0578-x.
37. Osei-Bempong C, Figueiredo FC, Lako M. The 48. Shi Y, Inoue H, Wu JC, Yamanaka S. Induced pluripo-
limbal epithelium of the eye–a review of limbal tent stem cell technology: a decade of progress. Nat
stem cell biology, disease and treatment. Bioessays. Rev Drug Discov. 2017;16(2):115–30. https://doi.
2013;35(3):211–9. https://doi.org/10.1002/ org/10.1038/nrd.2016.245.
bies.201200086. 49. Erbani J, Aberdam D, Larghero J, Vanneaux
38. Shortt AJ, Bunce C, Levis HJ, Blows P, Doré CJ, V. Pluripotent stem cells and other innovative strat-
Vernon A, et al. Three-year outcomes of cultured egies for the treatment of ocular surface diseases.
limbal epithelial allografts in aniridia and Stevens- Stem Cell Rev Rep. 2016;12(2):171–8. https://doi.
Johnson syndrome evaluated using the clinical out- org/10.1007/s12015-016-9643-y.
come assessment in surgical trials assessment tool. 50. Takahashi K, Yamanaka S. Induction of pluripotent
Stem Cells Transl Med. 2014;3(2):265–75. https:// stem cells from mouse embryonic and adult fibroblast
doi.org/10.5966/sctm.2013-0025. cultures by defined factors. Cell. 2006;126(4):663–
39. Borderie VM, Ghoubay D, Georgeon C, Borderie 76. https://doi.org/10.1016/j.cell.2006.07.024.
M, de Sousa C, Legendre A, et al. Long-term results 51. Takahashi K, Tanabe K, Ohnuki M, Narita M, Ichisaka
of cultured limbal stem cell versus limbal tissue T, Tomoda K, et al. Induction of pluripotent stem
transplantation in stage III limbal deficiency. Stem cells from adult human fibroblasts by defined factors.
Cells Transl Med. 2019;8(12):1230–41. https://doi. Cell. 2007;131(5):861–72. https://doi.org/10.1016/j.
org/10.1002/sctm.19-0021. cell.2007.11.019.
40. Tseng SC, Chen SY, Shen YC, Chen WL, 52. Karagiannis P, Takahashi K, Saito M, Yoshida Y, Okita
Hu FR. Critical appraisal of ex vivo expan- K, Watanabe A, et al. Induced pluripotent stem cells
sion of human limbal epithelial stem cells. and their use in human models of disease and devel-
Curr Mol Med. 2010;10(9):841–50. https://doi. opment. Physiol Rev. 2019;99(1):79–114. https://doi.
org/10.2174/156652410793937796. org/10.1152/physrev.00039.2017.
41. Baylis O, Figueiredo F, Henein C, Lako M, Ahmad 53. Cieślar-Pobuda A, Rafat M, Knoflach V, Skonieczna
S. 13 years of cultured limbal epithelial cell ther- M, Hudecki A, Małecki A, et al. Human induced plu-
apy: a review of the outcomes. J Cell Biochem. ripotent stem cell differentiation and direct transdiffer-
2011;112(4):993–1002. https://doi.org/10.1002/ entiation into corneal epithelial-like cells. Oncotarget.
jcb.23028. 2016;7(27):42314–29. https://doi.org/10.18632/
42. Kolli S, Ahmad S, Mudhar HS, Meeny A, Lako M, oncotarget.9791.
Figueiredo FC. Successful application of ex vivo 54. Yu J, Vodyanik MA, Smuga-Otto K, Antosiewicz-
expanded human autologous oral mucosal epithelium Bourget J, Frane JL, Tian S, et al. Induced pluripo-
for the treatment of total bilateral limbal stem cell tent stem cell lines derived from human somatic
deficiency. Stem Cells. 2014;32(8):2135–46. https:// cells. Science. 2007;318(5858):1917–20. https://doi.
doi.org/10.1002/stem.1694. org/10.1126/science.1151526.
43. Calonge M, Pérez I, Galindo S, Nieto-Miguel T, 55. Okita K, Ichisaka T, Yamanaka S. Generation of
López-Paniagua M, Fernández I, et al. A proof-of- germline-competent induced pluripotent stem
concept clinical trial using mesenchymal stem cells cells. Nature. 2007;448(7151):313–7. https://doi.
for the treatment of corneal epithelial stem cell defi- org/10.1038/nature05934.
ciency. Transl Res. 2019;206:18–40. https://doi. 56. Kang X, Yu Q, Huang Y, Song B, Chen Y, Gao X,
org/10.1016/j.trsl.2018.11.003. et al. Effects of integrating and non-integrating repro-
44. Zhang L, Coulson-Thomas VJ, Ferreira TG, Kao gramming methods on copy number variation and
WW. Mesenchymal stem cells for treating ocular genomic stability of human induced pluripotent stem
surface diseases. BMC Ophthalmol. 2015;15(Suppl cells. PLoS One. 2015;10(7):e0131128. https://doi.
1):155. https://doi.org/10.1186/s12886-015-0138-4. org/10.1371/journal.pone.0131128.
240 S. Bojic et al.
57. Kumar D, Talluri TR, Anand T, Kues WA. Transposon- tion. Nature. 2016;531(7594):376–80. https://doi.
based reprogramming to induced pluripotency. Histol org/10.1038/nature17000.
Histopathol. 2015;30(12):1397–409. https://doi. 70. Hayashi R, Ishikawa Y, Katori R, Sasamoto Y,
org/10.14670/hh-11-656. Taniwaki Y, Takayanagi H, et al. Coordinated genera-
58. Ma X, Kong L, Zhu S. Reprogramming cell fates by tion of multiple ocular-like cell lineages and fabrica-
small molecules. Protein Cell. 2017;8(5):328–48. tion of functional corneal epithelial cell sheets from
https://doi.org/10.1007/s13238-016-0362-6. human iPS cells. Nat Protoc. 2017;12(4):683–96.
59. Gurumoorthy N, Nordin F, Tye GJ, Wan Kamarul https://doi.org/10.1038/nprot.2017.007.
Zaman WS, Ng MH. Non-integrating lentivi- 71. Shibata S, Hayashi R, Okubo T, Kudo Y, Katayama T,
ral vectors in clinical applications: a glance Ishikawa Y, et al. Selective laminin-directed differen-
through. Biomedicine. 2022;10(1):107. https://doi. tiation of human induced pluripotent stem cells into
org/10.3390/biomedicines10010107. distinct ocular lineages. Cell Rep. 2018;25(6):1668–
60. Haridhasapavalan KK, Borgohain MP, Dey C, 79.e5. https://doi.org/10.1016/j.celrep.2018.10.032.
Saha B, Narayan G, Kumar S, et al. An insight into 72. Schlötzer-Schrehardt U, Dietrich T, Saito K, Sorokin
non-integrative gene delivery approaches to gener- L, Sasaki T, Paulsson M, et al. Characterization
ate transgene-free induced pluripotent stem cells. of extracellular matrix components in the lim-
Gene. 2019;686:146–59. https://doi.org/10.1016/j. bal epithelial stem cell compartment. Exp Eye
gene.2018.11.069. Res. 2007;85(6):845–60. https://doi.org/10.1016/j.
61. Raab S, Klingenstein M, Liebau S, Linta L. A com- exer.2007.08.020.
parative view on human somatic cell sources for 73. Ahmad S, Stewart R, Yung S, Kolli S, Armstrong L,
iPSC generation. Stem Cells Int. 2014;2014:768391. Stojkovic M, et al. Differentiation of human embry-
https://doi.org/10.1155/2014/768391. onic stem cells into corneal epithelial-like cells by
62. Yamanaka S. Induced pluripotent stem cells: past, in vitro replication of the corneal epithelial stem cell
present, and future. Cell Stem Cell. 2012;10(6):678– niche. Stem Cells. 2007;25(5):1145–55. https://doi.
84. https://doi.org/10.1016/j.stem.2012.05.005. org/10.1634/stemcells.2006-0516.
63. Scesa G, Adami R, Bottai D. iPSC preparation and 74. Hayashi R, Ishikawa Y, Ito M, Kageyama T, Takashiba
epigenetic memory: does the tissue origin mat- K, Fujioka T, et al. Generation of corneal epithelial
ter? Cell. 2021;10(6):1470. https://doi.org/10.3390/ cells from induced pluripotent stem cells derived
cells10061470. from human dermal fibroblast and corneal limbal
64. Nakagawa M, Taniguchi Y, Senda S, Takizawa N, epithelium. PLoS One. 2012;7(9):e45435. https://doi.
Ichisaka T, Asano K, et al. A novel efficient feeder- org/10.1371/journal.pone.0045435.
free culture system for the derivation of human 75. Kawasaki H, Suemori H, Mizuseki K, Watanabe K,
induced pluripotent stem cells. Sci Rep. 2014;4:3594. Urano F, Ichinose H, et al. Generation of dopaminer-
https://doi.org/10.1038/srep03594. gic neurons and pigmented epithelia from primate ES
65. Hongisto H, Ilmarinen T, Vattulainen M, Mikhailova cells by stromal cell-derived inducing activity. Proc
A, Skottman H. Xeno- and feeder-free differentiation Natl Acad Sci U S A. 2002;99(3):1580–5. https://doi.
of human pluripotent stem cells to two distinct ocular org/10.1073/pnas.032662199.
epithelial cell types using simple modifications of one 76. Qin S, Zheng S, Qi B, Guo R, Hou G. Decellularized
method. Stem Cell Res Ther. 2017;8(1):291. https:// human stromal lenticules combine with corneal
doi.org/10.1186/s13287-017-0738-4. epithelial-like cells: a new resource for corneal tis-
66. Klimanskaya I, Hipp J, Rezai KA, West M, Atala sue engineering. Stem Cells Int. 2019;2019:4252514.
A, Lanza R. Derivation and comparative assessment https://doi.org/10.1155/2019/4252514.
of retinal pigment epithelium from human embry- 77. Barbaro V, Ferrari S, Fasolo A, Ponzin D, Di Iorio
onic stem cells using transcriptomics. Cloning Stem E. Reconstruction of a human hemicornea through
Cells. 2004;6(3):217–45. https://doi.org/10.1089/ natural scaffolds compatible with the growth of cor-
clo.2004.6.217. neal epithelial stem cells and stromal keratocytes. Mol
67. Eintracht J, Toms M, Moosajee M. The use of induced Vis. 2009;15:2084–93.
pluripotent stem cells as a model for developmental 78. Mikhailova A, Ilmarinen T, Uusitalo H, Skottman
eye disorders. Front Cell Neurosci. 2020;14:265. H. Small-molecule induction promotes corneal epi-
https://doi.org/10.3389/fncel.2020.00265. thelial cell differentiation from human induced plu-
68. Shalom-Feuerstein R, Serror L, De La Forest DS, ripotent stem cells. Stem Cell Rep. 2014;2(2):219–31.
Petit I, Aberdam E, Camargo L, et al. Pluripotent stem https://doi.org/10.1016/j.stemcr.2013.12.014.
cell model reveals essential roles for miR-450b-5p 79. Mikhailova A, Jylhä A, Rieck J, Nättinen J, Ilmarinen
and miR-184 in embryonic corneal lineage specifi- T, Veréb Z, et al. Comparative proteomics reveals
cation. Stem Cells. 2012;30(5):898–909. https://doi. human pluripotent stem cell-derived limbal epithelial
org/10.1002/stem.1068. stem cells are similar to native ocular surface epi-
69. Hayashi R, Ishikawa Y, Sasamoto Y, Katori R, Nomura thelial cells. Sci Rep. 2015;5:14684. https://doi.
N, Ichikawa T, et al. Co-ordinated ocular development org/10.1038/srep14684.
from human iPS cells and recovery of corneal func- 80. Mikhailova A, Ilmarinen T, Ratnayake A, Petrovski
G, Uusitalo H, Skottman H, et al. Human pluripotent
stem cell-derived limbal epithelial stem cells on bio- Int J Ophthalmol. 2012;5(2):158–63. https://doi.
engineered matrices for corneal reconstruction. Exp org/10.3980/j.issn.2222-3959.2012.02.08.
Eye Res. 2016;146:26–34. https://doi.org/10.1016/j. 86. Tsai CL, Chuang PC, Kuo HK, Chen YH, Su WH, Wu
exer.2015.11.021. PC. Differentiation of stem cells from human exfoli-
81. Hongisto H, Vattulainen M, Ilmarinen T, Mikhailova ated deciduous teeth toward a phenotype of corneal
A, Skottman H. Efficient and scalable directed dif- epithelium in vitro. Cornea. 2015;34(11):1471–7.
ferentiation of clinically compatible corneal lim- https://doi.org/10.1097/ico.0000000000000532.
bal epithelial stem cells from human pluripotent 87. Kitazawa K, Hikichi T, Nakamura T, Nakamura M,
stem cells. JoVE. 2018;140:58279. https://doi. Sotozono C, Masui S, et al. Direct reprogramming
org/10.3791/58279. into corneal epithelial cells using a transcriptional net-
82. Kelaini S, Cochrane A, Margariti A. Direct repro- work comprising PAX6, OVOL2, and KLF4. Cornea.
gramming of adult cells: avoiding the pluripotent 2019;38(Suppl 1):S34–41. https://doi.org/10.1097/
state. Stem Cells Cloning. 2014;7:19–29. https://doi. ico.0000000000002074.
org/10.2147/sccaa.S38006. 88. Casaroli-Marano RP, Nieto-Nicolau N, Martínez-
83. Yang K, Jiang Z, Wang D, Lian X, Yang T. Corneal Conesa EM, Edel M. Potential role of induced plurip-
epithelial-like transdifferentiation of hair follicle stem otent stem cells (IPSCs) for cell-based therapy of the
cells is mediated by pax6 and beta-catenin/Lef-1. Cell ocular surface. J Clin Med. 2015;4(2):318–42. https://
Biol Int. 2009;33(8):861–6. https://doi.org/10.1016/j. doi.org/10.3390/jcm4020318.
cellbi.2009.04.009. 89. Mandai M, Watanabe A, Kurimoto Y, Hirami Y,
84. Blazejewska EA, Schlötzer-Schrehardt U, Zenkel M, Morinaga C, Daimon T, et al. Autologous induced
Bachmann B, Chankiewitz E, Jacobi C, et al. Corneal stem-cell-derived retinal cells for macular degenera-
limbal microenvironment can induce transdifferentiation. N Engl J Med. 2017;376(11):1038–46. https://
tion of hair follicle stem cells into corneal epithelial- doi.org/10.1056/NEJMoa1608368.
like cells. Stem Cells. 2009;27(3):642–52. https://doi. 90. Susaimanickam PJ, Maddileti S, Pulimamidi VK,
org/10.1634/stemcells.2008-0721. Boyinpally SR, Naik RR, Naik MN, et al. Generating
85. Saichanma S, Bunyaratvej A, Sila-Asna M. In minicorneal organoids from human induced pluripo-
vitro transdifferentiation of corneal epithelial-like tent stem cells. Development. 2017;144(13):2338–51.
cells from human skin-derived precursor cells. https://doi.org/10.1242/dev.143040.
Part IV
Stromal Lamellar Keratoplasty
Bowman Layer Transplantation
17
Achraf Laouani, Lydia van der Star, Silke Oellerich,
Korine van Dijk, Gerrit R. J. Melles,
and Viridiana Kocaba

• The standard treatment options for eyes with
advanced progressive keratoconus that are not The Bowman layer (BL) is an acellular and non-
eligible for UV-cross-linking or intracorneal regenerating layer located between the corneal
ring segments are DALK and PK. epithelial basement membrane and the anterior
• Eyes with advanced progressive keratoconus corneal stroma. The collagen fibrils are randomly
and subjectively good vision would benefit interwoven, forming a dense, felt-like sheet, in
from less invasive treatment options. contrast to the underlying stroma, where collagen
• In this chapter, we present Bowman layer fibers run in alignment across the diameter of the
transplantation as a new treatment option for cornea to form characteristic lamellae [1]. The
eyes with advanced progressive keratoconus. disruption of this anatomical barrier can be con-
genital, or more commonly secondary to frag-
mentation seen in keratoconus (KC) [2, 3]. It has
also been suggested that disruption of this layer
may result in an abnormal wound-healing
A. Laouani · L. van der Star · K. van Dijk V. Kocaba

Netherlands Institute for Innovative Ocular Surgery, Netherlands Institute for Innovative Ocular Surgery,
Rotterdam, The Netherlands Rotterdam, The Netherlands
Melles Cornea Clinic, Rotterdam, The Netherlands Melles Cornea Clinic, Rotterdam, The Netherlands
e-mail: [email protected]; [email protected]
Amnitrans Eye Bank, Rotterdam, The Netherlands
S. Oellerich
Tissue and Cell Therapy Group, Singapore Eye
Netherlands Institute for Innovative Ocular Surgery,
Research Institute, Singapore, Singapore
Rotterdam, The Netherlands
G. R. J. Melles (*)
Netherlands Institute for Innovative Ocular Surgery,
Rotterdam, The Netherlands
Melles Cornea Clinic, Rotterdam, The Netherlands
Amnitrans Eye Bank, Rotterdam, The Netherlands
https://doi.org/10.1007/978-3-031-32408-6_17
246 A. Laouani et al.
a b
Fig. 17.1 Schematic representation of Bowman layer implanted intrastromally in a manually dissected pocket.
(BL) inlay and onlay transplantation techniques. (a) (b) Bowman layer onlay. BL graft (red) is positioned onto
Bowman layer inlay. An isolated BL graft (red) is the patient’s anterior stroma
response, manifesting clinically as subepithelial more than 69 Diopter [14]. For those advanced,
stromal scarring [4, 5]. In view of these observa- progressive KC eyes with a subjectively accept-
tions, it has been suggested that restoring this able contact lens-corrected visual acuity before
anatomical barrier, by transplanting a healthy BL transplantation, even with poor contact lens
donor BL graft, may help to stabilize the cornea tolerance, BL transplantation can be indicated to
against further ectasia and maintain a better epi- postpone or even avoid the more invasive pene-
thelial–stromal interaction [6]. trating keratoplasty (PK) or DALK procedures,
BL transplantation was first developed as an which were previously the only available treat-
“inlay” graft, by inserting the BL graft in a manu- ment options for this group of patients [6–8,
ally created intrastromal pocket [6, 7]. In eyes 12–14].
with advanced, progressive KC (preoperative Advanced, progressive KC is also an indica-
maximum keratometry (Kmax) >69 Diopter), tion for the recently introduced BL onlay trans-
this technique resulted in a flattening of plantation [10]. In addition, the BL onlay
5–7 Diopter and stabilization of the corneas procedure can help in managing subepithelial
against further ectasia [6–8]. This technique, haze after excimer laser surface ablation and has
however, may be technically challenging, since the potential of reducing superficial corneal scar-
performing a midstromal dissection in eyes with ing and anterior corneal irregularities after her-
advanced KC and very thin corneas bears an petic infection [15, 16]. BL onlay grafting was
increased risk for Descemet membrane perfora- additionally applied to reduce fluctuation in
tion as also seen with deep anterior lamellar kera- visual acuity and refractive error after previous
toplasty (DALK) [9]. To avoid the stromal radial keratotomy (RK) surgery and to treat a
dissection and the associated risk of perforation, case with recurrent corneal erosions [17, 18].
BL transplantation was further developed into an
onlay procedure in which the isolated BL graft is
placed directly onto the anterior stroma resulting Surgical Technique
in a safer and technically easier procedure
(Fig. 17.1) [10, 11]. Bowman Layer Graft Preparation
Isolated BL grafts can either be prepared from

Indications whole donor globes that were obtained less than
24 h postmortem and whose corneas are consid-
Bowman layer inlay transplantation was devel- ered ineligible for PK or from an anterior corneal
oped to treat patients with advanced, progressive button after stripping a Descemet membrane endo-
keratoconus that were not considered to be eligi- thelial keratoplasty (DMEK) graft [15, 19–21].
ble anymore for ultraviolet corneal cross-linking For the preparation, donor globes or anterior
or intracorneal ring segments treatments [6–8, corneal buttons are mounted with the epithelial
12, 13]. Based on recent clinical outcomes, the side up on a globe holder or artificial anterior
BL inlay procedure is most effective in eyes with chamber, respectively, and the epithelium is
a preoperative maximum keratometry (Kmax) of removed using surgical spears. To incise the BL,
17 Bowman Layer Transplantation 247
a 30-gauge needle is used just within the limbal underlying stroma. The resulting BL graft usu-
area, 360° around. In a next step, the peripheral ally has a diameter of 9 to 11 mm (Fig. 17.2).
BL can be lifted and grasped with a McPherson After removal of remnant epithelial cells, the
forceps. With careful peeling to avoid tearing of grafts can be stored in organ-culture medium
the graft, the entire BL can be removed from the until the time of transplantation.
a b
c d
e f
Fig. 17.2 Bowman layer (BL) graft preparation. (a) The can be removed from the underlying stroma. (f) The
donor tissue is fixed with its epithelial side up. After epi- resulting BL graft usually has a diameter of 9 to 11 mm
thelial removal, a 30-gauge needle is used just within the and forms a single or a double roll. (Figure reprinted from
limbal area, 360° around, to incise the BL. (b) The periph- Dragnea DC, Birbal RS, Ham L, Dapena I, Oellerich S,
eral BL can be lifted and grasped with a McPherson for- van Dijk K, Melles GRJ. Bowman layer transplantation in
ceps. (c–e) By careful and circular peeling, the entire BL the treatment of keratoconus. Eye Vis (Lond). 2018;5:24)
As the manual preparation of BL grafts may immersed in 70% alcohol to remove any residual
require a high technical ability and is rather time epithelial cells, and rinsed with balanced salt
consuming, the use of a femtosecond laser to solution, is stained with trypan blue (VisionBlue;
facilitate the preparation has also been explored DORC International BV) and inserted along the
[22, 23]. Bowman layer grafts prepared with the surgical glide into the lamellar pocket. The BL
femtosecond laser seemed to be significantly graft is then unfolded and centered, using a can-
thicker, containing anterior stroma, but were rela- nula and jets of balanced salt solution (Fig. 17.3)
tively smoother than manually prepared tissue [6–8]. Once the BL graft is fully unfolded, the
[19, 22]. The potential optical impact of these dif- anterior chamber is inflated up to physiological
ferences between manually and femtosecond pressure with balanced salt solution, and the con-
laser-prepared graft has not been evaluated yet. junctiva is re-approximated to the superior lim-
bus. No sutures are required. Postoperative
medications include antibiotics for the first post-
Bowman Layer Inlay Grafting operative week and corticosteroids for the first
month after which the steroids are tapered.
Under retrobulbar anesthesia, a localized, supe- As manual dissection in very advanced kera-
rior conjunctival peritomy is performed. Then, toectatic, thin corneas may be challenging and
1–2 mm behind the limbus, a 5 mm partial thick- can result in a Descemet membrane perforation
ness scleral groove is created and dissected up to as also described for DALK [9, 12], the use of an
the cornea using a crescent knife. A paracentesis operating microscopy with intraoperative ante-
is then fashioned, through which the anterior rior segment optical coherence tomography
chamber is completely filled with air. Next, a (AS-OCT) has been described to facilitate visual-
dedicated set of curved spatulas (Melles spatula ization of the dissection plane especially when
set; DORC International, Zuidland, The blood, edema, or scarring obscures the surgeon’s
Netherlands) is used to dissect a mid-stromal view of the air–endothelial reflex during the man-
pocket. Utilizing the air–endothelial reflex, the ual dissection [24, 25].
dissection plane aims to be at 50% stromal depth, The use of a femtosecond laser to facilitate the
from limbus-to-limbus, 360° around, within the dissection of the stromal pocket has also been
recipient cornea. After the lamellar pocket has explored [20, 23]. As the presence of anterior cor-
been created and most air is removed from the neal scarring and/or an uneven thickness profile
anterior chamber, a surgical glide [BD Visitec (which are often seen in advanced keratoconus
(Fichman); Beaver-Visitec International, patients) could interfere with the optimal creation
Waltham, MA, USA] is inserted into the created of the femtosecond stromal pocket, this technique
pocket. The BL graft, which is once again may not be suited for all KC patients [11].
a b
c d
Fig. 17.3 Bowman layer (BL) inlay surgical technique. glide. (i, j) The graft is then unfolded and centered using a
(a) A scleral groove is created and dissected up to the cor- cannula. (Figure reprinted from Dragnea DC, Birbal RS,
nea and (b) paracentesis is then fashioned, (c) through Ham L, Dapena I, Oellerich S, van Dijk K, Melles
which air is injected filling the anterior chamber. (d–f) GRJ. Bowman layer transplantation in the treatment of
With curved spatulas, a mid-stromal pocked is made. keratoconus. Eye Vis (Lond). 2018;5:24)
After air removal, (g, h) the BL graft is inserted using a
e f
g h
i f
Fig. 17.3 (continued)

Bowman Layer Onlay Grafting ring, a manual keratotomy is carefully performed.

Epithelial remnants are removed by thoroughly
In very advanced KC eyes, the stromal dissection rinsing the stromal bed with BSS. The BL graft
can be challenging since the apex of the cone may (8.5–9.5 mm in diameter) is stained with 0.06% try-
be very thin and fragile, with an increasing risk of pan blue (VisionBlue; DORC International) and
perforation into the anterior chamber during sur- positioned onto the recipient cornea. The graft is
gery. Unlike the inlay technique, no corneal dis- then stretched at the periphery 360° by using thin
section is performed in the recently introduced forceps and flattened with a bent 30-gauge cannula
BL onlay grafting technique, reducing the intra- to squeeze out any interface fluid and to ensure the
operative risk of this procedure [10, 11]. adhesion of the BL graft onto the anterior stromal
During the procedure [10], first the recipient epi- bed without folds (Fig. 17.4). The BL transplant is
thelium is removed from the corneal surface with a subsequently allowed to dry-in for 45 min and a soft
hockey stick knife. In case of anterior stromal scar- bandage lens is placed on the eye [10].
a b
c d
Fig. 17.4 Bowman layer onlay surgical technique. (a) stroma. (c) The graft is stretched using thin forceps. (d)
The epithelium is first removed using a hockey stick knife. The graft is flattened with a 30-gauge cannula
(b) The stained graft is positioned onto the recipient
Clinical Outcomes grafts are usually re-epithelialized and appeared

to be well integrated in the corneal surface within
Bowman Layer Inlay Grafting 2–3 weeks. The average Kmax decreased in this
group from 75 Diopters to 70 Diopters up to
BL inlay transplantation resulted in a stabilization 1 year after surgery [10]. No changes were
of progressive KC in 88% of the eyes up to 8 years observed in the posterior corneal parameters.
postoperatively, which is comparable to the success Anterior- and posterior-order aberrations, espe-
rate of both UV cross-linking and intracorneal ring cially the corneal front lower order aberrations,
segments [14]. Furthermore, a flattening effect also seemed to improve throughout the post-
with an average Kmax reduction of 7 Diopter was operative follow-up period. The best contact lens
observed within the first postoperative month for corrected visual acuity remained stable [10, 11].
eyes with preoperative Kmax of more than 69 BL onlay transplantation was also performed
Diopters with no significant further changes up to to reduce fluctuations in visual acuity and refrac-
8 years postoperatively. For eyes with preoperative tive error in an eye after RK and the authors
Kmax of less than 69 Diopters, on the other hand, reported a reduction in the subjective complaints
no significant change in Kmax was observed after of visual fluctuation after the procedure [17]. In
the procedure [14]. For eyes with a preoperative two cases with superficial corneal scarring sec-
Kmax of more than 69 Diopters, van Dijk et al. also ondary to herpes simplex and varicella zoster
showed that the posterior corneal curvature flattens virus (HSV and VZV, respectively), BL onlay
postoperatively and stabilizes thereafter [7]. A sim- transplantation led to an improvement in corneal
ilar flattening effect was also described in other clarity, and no viral reactivation occurred
reports [21, 26], while no consistent conclusion can throughout the follow-up period [16]. Finally, a
yet be drawn on the effect of BL inlay transplanta- case report on a patient with a history of recurrent
tion on the posterior surface [7, 27]. painful corneal erosions, BL onlay grafting was
While most patients achieved a subjectively performed to restore the corneal surface. Until
acceptable vision with contact lenses and full 1.5 years postoperatively, the epithelium was
daily wear after BL inlay transplantation, average smooth over the graft, and the patients had no
best contact lens corrected visual acuity did not complaints and no recurrence of the epithelial
change significantly from pre- to postoperatively corneal erosion [18].
[13, 14, 28]. Possibly as a result of the corneal
flattening for some eyes an improvement in
spectacle-corrected visual acuity and a decrease Complications
in corneal higher order aberrations (especially
spherical aberration) in the first postoperative Since no sutures are required for both the BL
year were observed [8, 29]. The mid-stromal inlay and onlay procedure, suture-related compli-
positioning of the BL graft resulted, however, in cations can be avoided with these procedures.
an increase in corneal backscatter, which was Due to the acellularity of the BL graft, allograft
found to occur up to 5 years postoperatively, and rejection may be unlikely for both procedures
may be caused by interface irregularities [8, 13]. and topical steroids may be rapidly discontinued,
However, these changes did not correlate with a minimizing the risk of glaucoma development or
decrease of best-corrected visual acuity [8, 13]. cataract formation [8, 11].
Bowman Layer Onlay Graft Bowman Layer Inlay Graft
The clinical outcomes of BL onlay transplanta- In the case of the inlay graft technique, the main
tion have been reported for a pilot study includ- intraoperative complication (7.9–12.5%) [14, 21]
ing five patients with advanced KC [10]. The may be a Descemet membrane perforation while
a b
Fig. 17.5 Pre- and postoperative images of an eye that after Bowman layer onlay transplantation. Note that the
underwent Bowman layer onlay transplantation for scar (white arrow in a) is no longer present in b. Orange
advanced keratoconus. (a) Preoperative and (b) 2 years arrows point to notches in the BL-onlay graft
dissecting the mid-stromal pocket [7, 14, 21]. As on the first postoperative day due to inadvertent
with DALK, these perforations can be managed removal of the bandage lens [10]. The graft was
expectantly by aborting the operation, which then repositioned and covered with amniotic
allows healing, and reattempting again at a later membrane and a new bandage lens. It was also
date [9, 11, 12]. Alternatively, the surgeon may observed that most BL onlay grafts showed some
proceed with PK, depending on the size and posi- notches located at the peripheral edges progress-
tion of the perforation. ing over time (Fig. 17.5), but without any detect-
In a group of 35 eyes undergoing BL inlay able effect on the corneal curvature [10]. These
transplantation, 4 eyes (11.4%) showed contin- notches could be secondary to a difference in
ued corneal steepening and one patient under- thickness of the graft or a variable wound-healing
went successful BL inlay retransplantation for response.
unsatisfactory visual performance at 22 months For the BL onlay procedures performed for
after the initial BL inlay transplantation [14]. indications other than advanced, progressive KC,
Five eyes (four patients) developed an acute cor- no postoperative complications have been
neal hydrops within 8 years after surgery (at reported [16–18].
43–82 months postoperatively). These patients
had a history of severe eye rubbing and atopy and
had continued with eye rubbing after BL inlay Conclusion
transplantation. The eyes were treated with topi-
cal dexamethasone and NaCl, which resulted in BL graft transplantation is a promising new sur-
corneal clearing with some residual scarring and gical technique for the treatment of advanced,
visual restoration to pre-hydrops levels [14]. progressive keratoconus in order to postpone or
prevent a more invasive corneal surgery.
In eyes with advanced, progressive keratoco-
Bowman Layer Onlay Graft nus, both BL inlay and onlay grafting seem to
show comparable clinical outcomes in the first
For the five eyes included in the proof-of-concept postoperative months, resulting in corneal flat-
study of the BL onlay technique, it was reported tening, stabilization against further ectasia, and
that one eye with a partial dislocation of the graft enabling continued contact lens wear with high
tolerance. In addition, the BL onlay procedure 5. Melles GRJ, Binder PS, Anderson JA. Variation in
healing throughout the depth of long-term, unsutured,
was successfully applied to reduce superficial corneal wounds in human autopsy specimens and
corneal scarring and/or anterior corneal irregu- monkeys. Arch Ophthalmol. 1994;112(1):100–9.
larities and as a treatment of last resort in patients 6. Van Dijk K, Parker J, Tong CM, Ham L, Lie JT,
with recurrent corneal erosions. Groeneveld-van Beek EA, Melles GRJ. Midstromal
isolated Bowman layer graft for reduction of
Overall, the BL onlay approach is technically advanced keratoconus – a technique to postpone pen-
less challenging and shows promising results, etrating or deep anterior lamellar keratoplasty. JAMA
which need to be confirmed in a larger cohort of Ophthalmol. 2014;132(4):495–501.
patient. Less invasive and completely extraocu- 7. Van Dijk K, Liarakos VS, Parker J, Ham L, Lie JT,
Groeneveld-van Beek EA, Melles GRJ. Bowman
lar, the BL onlay procedure has the potential to layer transplantation to reduce and stabilize pro-
become the preferred surgical option for advanced gressive, advanced keratoconus. Ophthalmology.
KC. 2015;122(5):909–17.
8. Dragnea DC, Birbal RS, Ham L, Dapena I, Oellerich
S, van Dijk K, Melles GRJ. Bowman layer trans-
Take Home Notes plantation in the treatment of keratoconus. Eye Vis.
• Bowman layer inlay or onlay transplantation 2018;5:24.
may become an alternative treatment option 9. Karimian F, Feizi S. Deep anterior lamellar kerato-
for eyes with advanced progressive keratoco- plasty: indications, surgical techniques and complica-
tions. Mid E Afr J Ophthalmol. 2010;17(1):28–37.
nus in order to postpone or prevent a more 10. Dapena I, van der Star L, Groeneveld-van Beek EA,
invasive corneal surgery. Quilendrino R, van Dijk K, Parker JS, Oellerich S,
• Bowman layer transplantation can be per- Melles GRJ. Bowman layer onlay grafting: proof-of-
formed intrastromally or as an onlay graft and concept of a new technique to flatten corneal curva-
ture and reduce progression in keratoconus. Cornea.
induces a corneal flattening and stabilization 2021;40(12):1561–6.
of the keratoconus, maintaining the visual 11. Dapena I, Parker JS, Melles GRJ. Potential benefits
acuity and the comfort of the contact lens of modified corneal tissue grafts for keratoconus:
wear. Bowman layer ‘inlay’ and ‘onlay’ transplantation, and
allogenic tissue ring segments. Curr Opin Ophthalmol.
• Bowman layer onlay grafting is technically 2020;31(4):276–83.
less challenging, less invasive and completely 12. Parker JS, van Dijk K, Melles GR. Treatment options
extraocular. for advanced keratoconus: a review. Surv Ophthalmol.
• Bowman layer onlay grafts may also be 2015;60(5):459–80.
13. Van Dijk K, Parker JS, Baydoun L, Ilyas A, Dapena
applied to reduce superficial corneal scarring I, Groeneveld-van Beek EA, Melles GRJ. Bowman
and/or anterior corneal irregularities and as a layer transplantation: 5-year results. Graefes Arch
treatment of last resort in patients with recur- Clin Exp Ophthalmol. 2018;256(6):1151–8.
rent corneal erosions. 14. Van der Star L, van Dijk K, Vasiliauskaitė I, Dapena
I, Oellerich S, Melles GRJ. Long-term outcomes of
Bowman layer inlay transplantation for the treatment
of progressive keratoconus. Cornea. 2021;41:1150.
References 15. Lie J, Droutsas K, Ham L, Dapena I, Ververs B, Otten
H, van der Wees J, Melles GRJ. Isolated Bowman
1. Wilson SE, Hong JW. Bowman’s layer structure and layer transplantation to manage persistent subepi-
function: critical or dispensable to corneal function? thelial haze after excimer laser surface ablation. J
A hypothesis. Cornea. 2000;19(4):417–20. Cataract Refract Surg. 2010;36(6):1036–41.
2. Romero-Jiménez M, Santodomingo-Rubido J, 16. Dapena I, Musayeva A, Dragnea DC, Groeneveld-
Wolffsohn JS. Keratoconus: a review. Cont Lens van Beek EA, Ní Dhubhghaill S, Parker JS, van
Anter Eye. 2010;33(4):157–66. Dijk K, Melles GRJ. Bowman layer onlay trans-
3. Ambekar R, Toussaint KC Jr, Wagoner Johnson plantation to manage herpes corneal scar. Cornea.
A. The effect of keratoconus on the structural, 2020;39(9):1164–6.
mechanical, and optical properties of the cor- 17. Parker JS, Dockery PW, Parker JS, Dapena I, van Dijk
nea. J Mech Behav Biomed Mater. 2011;4(3): K, Melles GRJ. Bowman layer onlay graft for reduc-
223–36. ing fluctuation in visual acuity after previous radial
4. Melles GRJ, Binder PS. A comparison of wound heal- keratotomy. Cornea. 2020;39(10):1303–6.
ing in sutured and unsutured corneal wounds. Arch 18. Mulders-Al-Saady R, van der Star L, van Dijk K,
Ophthalmol. 1990;108(10):1460–9. Parker JS, Dapena I, Melles GRJ. Bowman layer
onlay graft for recurrent corneal erosions in map–dot– 24. Tong CM, van Dijk K, Melles GRJ. Update on
fingerprint dystrophy. Cornea. 2022;41:1062. Bowman layer transplantation. Curr Opin Ophthalmol.
19. Groeneveld-van Beek EA, Parker J, Lie JT, et al. 2019;30(4):249–55.
Donor tissue preparation for Bowman layer transplan- 25. Tong CM, Parker JS, Dockery PW, Birbal RS, Melles
tation. Cornea. 2016;35(12):1499–502. GRJ. Use of intraoperative anterior segment optical
20. García de Oteyza G, González Dibildox LA, Vázquez- coherence tomography for Bowman layer transplanta-
Romo KA, Tapia-Vazquez A, Davila-Alquisiras JH, tion. Acta Ophthalmol. 2019;97(7):e1031–2.
Martínez-Báez BE, Garcia-Albisua AM, Ramirez M, 26. Shah Z, Hussain I, Borroni D, Khan BS, Wahab
Hernández-Quintela E. Bowman layer transplantation S, Mahar PS. Bowman’s layer transplantation in
using a femtosecond laser. J Cataract Refract Surg. advanced keratoconus; 18-months outcome. Int
2019;45(3):261–6. Ophthalmol. 2021;42:1161.
21. Tourkmani AK, Mohammad T, McCance E, Potts 27. Tourkmani AK, Lyons C, Hossain PN,
J, Ford R, Anderson DF. One-year front versus cen- Konstantopoulos A, Anderson DF, Alio JL. 1 year
tral and paracentral corneal changes after Bowman posterior corneal changes after Bowman layer
layer transplantation for keratoconus. Cornea. transplantation for keratoconus. Eur J Ophthalmol.
2021;41(2):165–70. 2021;32:1370.
22. Parker JS, Huls F, Cooper E, Graves P, Groeneveld- 28. Zygoura V, Birbal RS, van Dijk K, Parker JS, Baydoun
van Beek EA, Lie J, Melles GRJ. Technical feasibil- L, Dapena I, Melles GRJ. Validity of Bowman layer
ity of isolated Bowman layer graft preparation by transplantation for keratoconus: visual performance at
femtosecond laser: a pilot study. Eur J Ophthalmol. 5–7 years. Acta Ophthalmol. 2018;96(7):e901–2.
2017;27(6):675–7. 29. Luceri S, Parker J, Dapena I, Baydoun L, Oellerich
23. Mittal V, Rathod D, Sehdev N. Bowman-stromal S, van Dijk K, et al. Corneal densitometry and higher
inlay using an intraocular lens injector for man- order aberrations after Bowman layer transplantation:
agement of keratoconus. J Cataract Refract Surg. 1-year results. Cornea. 2016;35(7):959–66.
2021;47(12):e49–55.
Anterior Lamellar Keratoplasty:
Current State of the Art
18
Enrica Sarnicola, Caterina Sarnicola,
and Vincenzo Sarnicola
Key Points • In order to achieve good visual outcome and

• Anterior lamellar keratoplasty (ALK) includes be classified as DALK, the stromal dissection
techniques whereby the diseased corneal has to create a residual bed that is thin
stroma is partially or totally replaced by donor (≤80 μm), smooth, and uniform in its
tissue, provided that the endothelium is still thickness.
functioning. • Surgical techniques used to be divided into
• Among ALK major advantages there is the two classes: predescemetic DALK (pdDALK)
avoidance of endothelial rejection, which is to indicate manual dissections techniques,
one the most common cause of graft failure which may be considered challenging and
with penetrating keratoplasty (PK). time consuming; and descemetic DALK
• Unsatisfactory visual outcomes related to the (dDALK) to refer to techniques that were
thick residual host bed and interface issue are thought to expose the Descemet’s membrane
a main downside in ALK. These techniques (DM) making the surgery faster and more reli-
never gain popularity and have progressively able, like with big bubble (BB) and
been dismissed. viscodissection.
• Only procedures that accomplish a very deep • It has been recently demonstrated that BB
stromal dissection, deep anterior lamellar ker- type 1 does not separate DM from stroma but
atoplasty (DALK), can achieve good visual is in fact an intrastromal bubble, whereas only
results, comparable with PK. BB type 2 truly exposes the DM. This newer
knowledge has made the term dDALK and
pdDALK confusing, creating the need for a
new classification.
org/10.1007/978-3-031-32408-6_18. • A new classification has been proposed: Deep
anterior lamellar keratoplasty (DALK) for all
the manual dissection techniques that are suf-
E. Sarnicola · V. Sarnicola (*) ficiently deep; subtotal anterior lamellar kera-
Ambulatorio di Chirurgia Oculare Santa Lucia,
Grosseto, Italy toplasty (STALK) for all the previous dDALK
techniques where a very thin layer of stroma is
Clinica degli Occhi Sarnicola, Grosseto, Italy
left behind together with the DM and the
C. Sarnicola endothelium indeed, and total anterior l amellar
Clinica degli Occhi Sarnicola, Grosseto, Italy
keratoplasty (TALK) for the cases were the
Ospedale San Donato, U.O.C. Oculistica, DM is truly exposed.
Arezzo, Italy
https://doi.org/10.1007/978-3-031-32408-6_18
258 E. Sarnicola et al.
Introduction outcomes similar to PK and shift the paradigm of

corneal transplantation [2, 10]. It was Archila, in
Corneal transplantation has been evolved rapidly 1984, who first wrote the term “ deep lamellar
in the past 20 years. Penetrating keratoplasty (PK) keratoplasty (DLK)” to differentiate his tech-
has been the dominant procedure for more than nique from all others ALKs that were too superfi-
half century, but it has now been substituted by cial, emphasizing the need of a deep dissection.
lamellar keratoplasties (LKs), less invasive proce- Due to the lack of a proper categorization, how-
dures that selectively replace only the diseased ever, misclassification between DALK and ALK
corneal layer [1–5]. These include deep anterior has been unfortunately common, and the two
lamellar keratoplasty (DALK) to address stromal terms have been often inappropriately used as
diseases; a variety of endothelial keratoplasty synonymous [9, 14, 20].
(EK) procedures for endothelial diseases, includ- A clear distinction between ALK and DALK
ing Descemet stripping endothelial keratoplasty is pivotal, especially when comparing visual out-
(DSEK), Descemet stripping automated endothe- comes with PK. Today, we know that the achieved
lial keratoplasty (DSAEK), and Descemet mem- residual host bed has to be equal or thinner than
brane endothelial keratoplasty (DMEK); as well 80 μm, smooth, and uniform in thickness to be
as ocular surface stem cell transplantation for epi- classified as a DALK, weather techniques that
thelial diseases due to limbal stem cell deficiency leave a significant thicker/uneven residual bed
(LSCD) [1, 2, 6–8]. This chapter will discuss LK should not be included in this category, and they
for stromal diseases with a healthy endothelium. should be indicated as ALK (that are not deep or
Initially, the expression LK was used to refer deep enough indeed) [9, 14, 15, 21, 22].
only to anterior lamellar keratoplasty and only to
manual dissection techniques, regardless of the
depth of the stromal removal [9]. Somewhat less nterior Lamellar Keratoplasty
A
than the entire thickness of cornea was removed, (ALK)
preserving the recipient endothelium, Descemet’s
membrane (DM), and a portion of deep stroma. ALK techniques have always had a poor diffu-
An intrastromal interface was thereby produced sion and are progressively being dismissed; today
where donor and host collagen lamellae meet in they are mostly the unintended result of a manual
apposition; hence the term “lamellar kerato- DALK that does not reach the appropriate depth.
plasty” [3]. With the development of endothelial There are some exceptions, however, that deserve
keratoplasty, LK necessarily became anterior at list a brief mention.
lamellar keratoplasty (ALK) [9].
Despite its undeniable advantages, including
the avoidance of endothelial rejection and longer Epikeratoplasty (EPK)
graft survival, ALK did not gain popularity,
mainly due to unsatisfactory visual results related Epikeratoplasty (EPK), also known as “epikera-
to the thick residual host bed and interface issues tophakia” or “onlay lamellar keratoplasty,” is an
[9–15]. Advances in technology and techniques, older surgical procedure initially conceived to
as well as a greater understanding of corneal correct aphakia, first described by Werblin and
physiology and optics, facilitated the resumption Kaufman in 1981 [23]. Other indications later
of lamellar surgery, showing that the key to obtain included high myopia, hyperopia, and
a good visual outcome, and to reduce the donor– keratoconus [24]. The technique consists in plac-
host interface, is to perform the deepest possible ing a lamellar donor graft on top of a de-
stromectomy, reaching a smooth and uniform in epithelialized host cornea and suturing it into a
thickness recipient surface [9, 16–19]. In fact, it prepared groove. The advantages of this proce-
was only with the introduction of newer deep dure are similar to those of other ALK procedure,
anterior lamellar keratoplasty (DALK) tech- plus its potential reversibility. The procedure,
niques that surgeons were able to achieve visual however, has been forsaken due to poorly pre-
18 Anterior Lamellar Keratoplasty: Current State of the Art 259
dictable visual outcomes, long visual recovery, nique, first described by Kaufman et al. in 2003,
risk of postoperative irregular astigmatism, pro- usually reserved for the treatment of corneal
gressive myopia, reduced contrast sensitivity, and opacities confined to the anterior third of the
interface opacity [25]. Today, this technique may corneal stroma (within the first 160 μm) [35,
still have some very limited indications like the 36].
management of extremely advanced keratoglo- The procedure includes both superficial kera-
bus, or thinning/perforation in brittle corneal syn- tectomy, as well as the preparation of an anterior
drome cases [26–29]. lamella from the donor tissue of the same thick-
ness to be placed onto the recipient bed, with or
without overlay sutures. Results obtained with
Lamellar Patch Grafts manual dissection are suboptimal due to the
irregular interface and thus poor visual outcome,
This group of ALKs includes different techniques while better results are reported with the use of
in which the graft is harvested in various shapes microkeratome or femtosecond laser. Compared
and depths, in accordance with the shape and the to the microkeratome, the femtosecond laser
depth of the affected cornea (i.e. lamellar cres- might be more accurate and precise; however, it
centic keratoplasty, “banana” graft, partial ring is more expensive and, in the presence of signifi-
lamellar keratoplasty, D-shaped lamellar kerato- cant corneal opacities, the quality of the stromal
plasty, C-shaped lamellar keratoplasty, “donut” bed could be inferior compared with that created
lamellar keratoplasty, annular lamellar kerato- using a microkeratome [35, 37–39].
plasty, ring lamellar keratoplasty, etc.). These The indications for SALK are very limited,
techniques have been used mainly for tectonic including only superficial irregular opacities,
purposes and are best suitable in cases with such as anterior corneal dystrophies and superfi-
peripheral corneal perforations/descemetoceles, cial scars, which are conditions where photo-
providing acceptables visual rehabilitation therapeutic keratectomy (PTK) is largely
because of the graft eccentric location. considered the treatment of choice. Recurrence
Although the technique of eccentric lamellar of corneal dystrophy after PTK or DALK is
patch grafting is technically challenging and common and often eventually requires repeated
lacks of standardized procedure, current indica- interventions. SALK has the theoretical advan-
tions include the management of a variety of cor- tage that it could be repeated with no increased
neal disorders characterized by peripheral risk of hyperopic shift or corneal thinning, as it
thinning and/or ectasia, with the aim of providing could instead occur with repeated PTK. Only
tectonic stability and/or improvement in corneal few small case series are published on this mat-
surface regularity: pellucid marginal degenera- ter [35, 38, 40, 41].
tion (PMD), peripheral ulcerative keratitis Contraindications of these techniques are
(PUK), Mooren’s ulcer, Fuchs marginal keratitis, superficial scars that are depressed because the
advanced Terrien’s marginal degeneration, and microkeratome cut follows the surface profile
also infectious disorders with peripheral melting and would lead to the same defect on the stromal
[15, 30–34]. bed; thin and/or irregular corneas; and deep-set
eyes with small palpebral aperture that can pose
difficulty in the fitting of the microkeratome [15]
uperficial Anterior Lamellar
S (Fig. 18.1).
Keratoplasty (SALK) Possible complications include dystrophy
recurrence, stromal melting, graft dislocation,
Superficial anterior lamellar keratoplasty epithelial ingrowth, infectious keratitis, and
(SALK) is an intended superficial ALK tech- astigmatism [37, 42].
amount of posterior stroma is left in place along

with the DM-endothelium, which is usually the
result of manual dissection techniques [17].
How much stroma can be left in place to pro-
vide good visual outcome? Josè Barraquer, who
outlined the requirements to achieve good
visual results with LK, addressed this matter in
1972 already. These requisites were later on
better refined with measured values, but they
essentially remain the same: obtain the deepest
possible interface to reduce scarring, attain a
posterior layer of uniform thickness, and create
a smooth surface of both the graft and the recip-
Fig. 18.1 Postoperative clinical photograph of a patient
who underwent microkeratome assisted SALK for postin- ient bed [9, 18].
fective scar. The slit lamp illumination shows the interface Although there is no unanimous agreement
between the donor (yellow arrow) and the recipient (white yet, the vast majority of articles in the litera-
arrow). The recipient bed has an uneven pachymetry,
ture show that visual recovery after pdDALK
being thinner in the center and thicker in the periphery,
and some residual scarring, which both affect the visual is slower (2–5 years of follow-up) but compa-
outcome rable with dDALK and PK, as long as the
residual recipient bed thickness does not mea-
sures more than 80 μm and is homogenous in
eep Anterior Lamellar
D its thickness. However, it is not always easy to
Keratoplasty judge the depth of the manual dissection intra-
operatively; therefore, dDALK techniques are
Classifications of Surgical Techniques usually preferred, as they make the surgeon
confident to have performed optimal stromal
Several DALK techniques have been described in removal with a good visual prognosis [9, 10,
the past 25 years, each aiming to expose the 19].
Descemet membrane (DM)-endothelium com- The more recent description of Dua’s layer
plex or to leave behind as little residual stroma as (DL), also called pre-descemetic Layer (PDL),
possible, aiming to create a good optical graft– has demonstrated that the recipient bed created
host interface and optimize visual recovery [16, in dDALK cases, which was thought to be a
22, 35]. DM–endothelium surgical exposure, in fact also
These techniques have been broadly classi- includes a very thin layer of stroma in most
fied in the literature as descemetic DALK cases [45]. Although the existence of the sixth
(dDALK) and predescemetic DALK new layer of cornea remains a contested subject,
(pdDALK) [17]. In dDALK cases, the dissec- the presence of some very thin stroma that
tion is achieved up to the DM (or at least what remains on top of was thought to be just DM–
it was thought to be up to the DM) thanks to a endothelium is unquestionable [46]. This
forceful deep stromal injection of air (big bub- knowledge generated confusion about what is
ble) or viscoelastic (viscodissection) which cre- intended with the terms “dDALK” and “
ates bubble formation that detaches the pdDALK,” supporting the need for a more
DM-endothelium from the posterior stromal. appropriate nomenclature.
This type of stromal dissection makes the sur- A new nomenclature has been proposed in
gery become faster and more reliable, allowing 2019, trying to respect both the previous classifi-
the surgeon to be confident to have performed cation and the new findings in microscopic
an optimal procedure with a good visual prog- anatomy. The new proposed classification goes
nosis [9, 17, 43, 44]. In pdDALK cases, a small as follow [9]:
• DALK—Deep Anterior Lamellar

Keratoplasty: it includes all the previously
called pdDALK techniques. This group basi-
cally includes manual techniques that leave
along with the DM a small amount of poste-
rior stroma, which is macroscopically evident
during surgery, but that does not measure
more than 80 μm of thickness (i.e. peeling off,
layer by layer manual dissection, hydrodissec-
tion, etc.).
• STALK—Sub-Total Anterior Lamellar
Keratoplasty: this group encompasses what Fig. 18.2 Big bubble (BB) type 1. This type of BB is
was called dDALK (except for the type 2 big well circumscribed, it has white margins (yellow arrow),
its diameter measures up to 8.5 mm, and it starts in the
bubble), in which the DM seems to be intraop-
center and enlarges circumferentially toward the
eratively exposed, but where a microscopic periphery
layer of stroma is in fact left in place too (i.e.
big bubble type 1, viscodissection,
air-viscobubble). Types of Big Bubble
• TALK—Total Anterior Lamellar There are three types of bubbles. BB Type 1 is
Keratoplasty: it includes the type 2 big bub- the most common type and exposes a residual
ble, the only technique previously classified as bed that intraoperatively looks like DM–endothe-
dDALK, which actually exposes the DM. lium, but that is in fact intrastromal. The recipient
bed obtained histologically also contains some
residual stroma, the alleged PDL. This type of
Surgical Techniques bubble has some distinctive features: it is well-
circumscribed, it has white margins, its diameter
Several techniques have been employed to measures up to 8.5 mm, it starts in the center and
achieve deep stromal dissection [16, 35, 47] enlarges circumferentially toward the periphery,
however, in the authors’ opinion, the most com- and it is quite resistant [45] (Fig. 18.2). BB type
mon are big bubble [43], viscodissection [44, 1 was erroneously considered a dDALK for many
48, 49], and some manual techniques [22, 50, years, but it has now been re-classified as STALK.
51]. BB type 2 is larger (up to 10.5 mm), typically
eccentric, it usually starts in the periphery and
ig Bubble (BB)
B enlarges centrally, and it has clear margins,
The big bubble (BB) technique was originally almost looking like a bubble of air in the anterior
described by Anwar and Teichmann in 2002, and chamber that does not move when the eye is
it is probably the most commonly used technique rotated (Fig. 18.3). This type of bubble is pretty
[9, 43, 52]. It was one of the first STALK/TALK rare and it is the only type of bubble that really
techniques (previously dDALK) described, and it cleaves off the DM–endothelium from the stroma
helped the world transitioning from PK to (TALK) and therefore it is very fragile [45, 53].
DALK. This technique involves a forceful injec- One should be extremely careful when opening
tion of air into the corneal stroma to produce a this type of bubble; given the high risk of DM
sudden separation of the stroma from the DM– rupture, some surgeons even suggest no to open
endothelium, resulting in rapid formation of a the bubble and to perform a manual dissection on
circular air pocket that is seen as a big bubble top of it instead [54]. The bubble usually resolves
[45] (Fig. 18.2). itself in the early postoperative time [55].
The deeper the air is injected, the higher are

the chances to create a big bubble [56]; therefore,
several modifications of the original technique
have been described trying to increase the bubble
success rate [22].
Cannula Big Bubble Technique

Aiming to inject the air as close as possible to
DM, the use of blunt instrument has been pro-
posed instead of a needle, by Sarnicola and Toro
in 2011, to let surgeons go as deeply as possible
into the corneal stroma, without being afraid of
perforating: the “cannula big bubble” tech-
Fig. 18.3 Big bubble (BB) type 2. BB type 2 (yellow nique. The surgical steps are similar to the BB
arrows) is typically eccentric, it usually starts in the technique described by Anwar, but with two
periphery and enlarges centrally, and it has clear margins,
almost looking like a bubble of air in the anterior chamber
important modifications. (1) After a partial cor-
that does not move when the eye is rotated. The black neal trephination, a smooth spatula is inserted at
arrow shows some central stromal emphysema (no the deepest point in the peripheral trephination
bubble) groove, and it is moved toward central/paracen-
tral cornea using a wiggling motion, creating a
very deep track. When a very deep plane is
BB type 3 consists in mixed type of bubbles: reached, two signs may be observed: reduced
BB type 1 and one or more smaller type 2 bub- resistance of the advancement of the spatula and
bles [45]. the appearance of DM folds. (2) The spatula can
then be withdrawn, leaving a corneal track where
Needle Big Bubble Technique to insert a 27-gauge, blunt tipped, bottom port,
With the original BB technique, a trephine is air injection cannula, attached to a 5 cc air filled
used to perform a partial thickness corneal syringe. After advancing the cannula, a little
trephination at about a 60–80% depth. A 27- or more forward to the center of the cornea, the air
30-gauge needle, attached to an air-filled is then injected using a firm continuous pressure
syringe, is then inserted deep into the paracen- till the formation of big bubble is noted [57]. A
tral stroma through the bottom of the trephina- comparative study over 507 eyes affected by ker-
tion groove, and advanced with the bevel parallel atoconus showed a significantly higher percent-
to the DM, facing downward. Air is injected, age of successful BB using a cannula (82%)
and a big bubble is formed between DM–endo- compared to using a needle (61%) (p < 0.01)
thelium and the corneal stroma (BB type 2) or, [58]. The advantages of using a blunt tipped bot-
more frequently, between the posterior stroma tom port cannula have been confirmed by several
and the PDL layer (BB type 1). Subsequently, studies [59–61].
anterior keratectomy is performed and a small
opening at the center of the anterior wall of the Pachy Bubble and Newer Devices
bubble is created. This opening should be per- Intraoperative corneal thickness measurement to
formed using the sharp tip of a pointed blade, create a pachymetry-guided intrastromal air injec-
held almost parallel to the surface. As the col- tion to increase the rate of BB formation has been
lapse of the air bubble occurs, the knife has to be proposed by Ghanem in 2012; the “pachy-
quickly withdrawn to avoid inadvertent perfora- bubble” technique. After an initial partial trephi-
tions. The remaining corneal stromal layers are nation (about 60–70%), intraoperative corneal
lifted with an iris spatula, severed with a blade, thickness measurements using ultrasound
and excised with scissors, before suturing the pachymetry are taken 0.8 mm internally from the
donor [43]. trephination groove in the 11–1 o’ clock position.
In this area, a 2-mm incision is made, parallel to to different biomechanics between the posterior
the groove, with a micrometer diamond knife, corneal lamellae and the anterior stroma [70].
calibrated to 90% depth of the thinnest measure- Newer femtosecond laser settings are the object
ment. The incision is then opened with toothed of current studies to create a cut that resembles the
forceps and widened superficially with a optimum cut achieved in the anterior cornea dur-
15-degree blade and used to start the stromal track ing refractive surgery. However, there are also
for the cannula BB technique [57, 62]. limitations in creating a residual host bed thinner
Intraoperative anterior segment optical coher- than 100 μm without damaging the endothelial
ence tomography (AS-OCT) and femtosecond cells and in producing a pachymetrically homoge-
lasers have been suggested to guide or create the neous host bed given the different curvature
deep stromal track to allow proper placement of between anterior and posterior surface, especially
needle/cannula; these tool seems to be helpful and in advanced keratoconus patients [71].
promising; however, they are very expensive, and
there are limited data in literature [63–67]. The Bubble Test
use of femtosecond lasers in DALK has also been While injecting air into the stroma, trying to
employed to substitute manual trephination and obtain a BB, corneal emphysema can occur and it
create precise shaped wound. The shaped wound may hinder visualization of the BB, making the
configuration may offer the advantage of better surgeon uncertain about how to proceed.
donor host apposition, with increased surface area Parthasarathy et al. described a technique in
contact, resulting in faster wound healing, greater which a small air bubble is injected into the ante-
tectonic stability, allowing earlier suture removal, rior chamber (AC) via a limbal paracentesis. If
and possibly reducing astigmatism as well. the small air bubble is seen at the periphery of the
However, there is no consensus on a standardized AC, it will confirm that separation of the DM
approach for wound design or postoperative man- induced by BB has been successfully accom-
agement, and most of the reports on the matter are plished. This is noted also while rotating the eye
laboratory studies or small case series with short- trying to make the air bubble move centrally; the
term outcomes [65]. The use of femtosecond laser air bubble moves circumferentially and remains
to prepare the residual stromal bed, on the other in the periphery of the AC, usually with a sausage
hand, has generated a lot of concern about its effi- configuration. If the small bubble is not seen at
cacy and safeness when used [35, 68, 69]. Studies the periphery of the AC and is instead located
have showed that deep femtosecond laser abla- centrally beneath the opaque corneal stroma, this
tion, using high energy, creates irregularities and would suggest that the big bubble has not been
bridges that give poor optical quality. This is due obtained [72] (Fig. 18.4).
a b
Fig. 18.4 Bubble test. (a) Picture shows positive bubble occupied by the BB. (b) When the BB is opened, the bub-
test in the setting of a successful big bubble (BB) type 1. ble test becomes negative, as the air bubble into the AC is
The small air bubble in the anterior chamber (AC) is seen free to move into the center of the AC (yellow arrow)
in the periphery of the AC (yellow arrow), as the center is
a b
c d
Fig. 18.5 New big bubble (BB) opening. (a) Once the a 2.2 keratome is used to incise the bubble, with a bottom
BB has been achieved, an anterior keratectomy should be upwards incision; (d) no collapse of the BB after the with-
performed and the stroma overlying the BB should be draw of the keratome
coated with a cohesive viscoelastic (yellow arrow); (b, c)
New Technique to Open the Big Bubble opening, a bottom upward incision cut using a
The technique to open the big bubble, as origi- 2.2 mm keratotome can be considered instead of
nally described by Anwar (see section “Needle the lifting motion cut and to perform a paracente-
Big Bubble Technique”), is associated with a sis to reduce the eye pressure before opening the
known risk of DM perforation; the maneuver is in bubble [16, 22] (Fig. 18.5).
fact called “the brave slash” by many corneal sur-
geons [43]. The BB opening technique has been Viscodissection
refined by Goshe et al. to avoid perforation. They Intrastromal injection of ophthalmic viscoelastic
suggested to coat the stroma overlying the BB can be used too to create a bubble and achieve a
with a cohesive viscoelastic prior to entering the quick and deep stromal dissection [44, 48, 74].
big bubble, and to incise the bubble using only This technique, also known as “visco-bubble,”
the tip of the blade in a “lifting” motion. These has been described by Melles et al. in 1999. This
measures limit the escape of air from the bubble, bubble mimics the behavior of the air BB type 1
preventing a sudden collapse of the BB while and it usually results in a STALK [9]. In this tech-
entering in it with the blade. Furthermore, they nique, a 30-gauge needle (or a blunt tipped bot-
suggested an air–viscoelastic exchange to main- tom port cannula) attached to a viscoelastic-filled
tain space in the bubble, to facilitate the removal syringe, is inserted into the corneal stroma as
of the anterior wall of the bubble [73]. To further close to the DM as possible [48]. To visualize the
limit the escape of air from the BB during its depth of the corneal track dissection during sur-
gery, Melles et al. proposed the creation of an air- trephination groove). Because of its high
to-
endothelium interface, which behaves as a viscosity, the viscoelastic device does not escape
convex mirror, exchanging anterior chamber as easily, creating a much higher intrastromal
aqueous fluid with air [49]. A dark, nonreflective pressure and increasing the chances of bubble
band can be seen between the tip of the needle/ formation, probably also increasing the pressure
cannula and the light reflex, representing the inside the small stromal air bubbles (from the
residual corneal tissue. Because the dark band failed BB attempt) that spontaneously merge to
becomes thinner when advancing the needle/can- form a large DM detachment [75].
nula into the deeper stromal layers, the corneal
depth of the needle/cannula can be judged from Manual DALK Techniques
the thickness of the dark band, helping to decide Despite technically more challenging, manual
when to inject the viscoelastic. A typical reflex, dissection techniques are still a valid option; they
that we like to call “golden ring”, outlines the for- are mainly adopted in cases where air- and/or
mation of a visco-bubble. Once the bubble has visco-dissection fail or when they are not indi-
formed, it can be opened, the stroma over the cated (i.e. keratoconus with history of previous
bubble is excised, and the recipient bed is thor- hydrops, deep dense stromal scars, opaque cor-
oughly irrigated to remove all residual viscoelas- nea with poor visibility, penetrating corneal
tic before suturing the donor [44, 48, 49]. wounds, etc.). These techniques used to be classi-
Although this procedure provides good fied as pdDALK, today just DALK (see section
results, it is not always easy to identify the reflex. “Classifications of Surgical Techniques”).
Some surgeons prefer to de-bulk the cornea
before attempting the viscobubble to enhance the Peeling Off (Video 18.1)
chances to inject the viscoelastic as close as pos- Malbran described this easy and rapid technique in
sible to the DM. In the authors’ opinion, the use 1966, which still represents a useful option, espe-
of cohesive viscoelastic should to be preferred cially in eyes with keratoconus. In the original
because it is easier to remove, reducing the risk technique, a partial corneal trephination is deep-
of postoperative double anterior chamber [22]. ened 360° with a blade and the edges are then
Intrastromal viscoelastic injection has also raised with a Paufique knife until there is enough
been suggested, by Sarnicola et al. in 2010, as a tissue to grasp; two forceps are then used to pull the
rescue bubble technique, thus as a second stroma away from the deeper layers [51, 76, 77].
approach after a failed air bubble: air-viscobubble This surgical technique basically separates the
(AVB) [17, 57, 75]. With this technique, when air anterior corneal stroma by pulling the deep stro-
BB fails, a superficial keratectomy is performed mal lamellae following the plane of their lowest
with a crescent blade and a new deeper corneal adhesion, which is usually very deep, allowing an
track is created into the stroma by using a blunt optimal visual recovery (formerly classified as
spatula. A visco-bubble is then attempted as a pdDALK). The pulling does not require great
second strategy to obtain a STALK, by using the force, and it is very easy especially in the area of
same blunt tipped bottom port cannula used for the cone of the keratoconus cases [9, 22, 78].
the cannula BB technique [17, 57, 75]. In a case Recently, the peeling off technique has been
series of 507 eyes affected by keratoconus, this re-proposed by some surgeons (Sarnicola, Fogla,
combined technique (AVB) incremented of 12% etc.) with some modifications, namely perform-
the percentage of bubble formation, bringing the ing a partial debulking before the pulling and the
total cases of successful bubble formation use of dedicated blunt instruments to deepen the
(STALK) to 94% [58]. When the BB fails, the trephination grove and find the appropriate very
cornea is generally pneumatized and offers many deep plane that allows for the stroma to be peeled
pathways of less resistance to air compared to the away (blunt tipped pocket stromal forceps or a
pre-Descemet space (i.e. leakage of air through blunt tipped 27 G DALK spatula) [22, 79]
the stroma, the trabecular meshwork, or the (Fig. 18.6).
a b c
d e f
g h i
Fig. 18.6 Peeling off DALK technique. In this specific plane is exposed 360°; (e) the appropriate depth of the
case, the peeling off technique has to be used after a failed plane can be identified by its shiny and smooth appear-
attempt of big bubble. (a) Anterior keratectomy using a ance, and by the absence of any area of air emphysema;
crescent blade. The white emphysema is the result of (f–h) once the stromal periphery is freed, the inner stro-
failed big bubble; (b–d) The trephination groove is deep- mal edges of the central stroma are grasped securely with
ened by creating a deep stromal track along the groove, forceps, which are used to pull the stromal tissue and to
using a 27 G DALK blunt spatula. Once the track has peel it from underlying deeper layers; (i) deep and smooth
reached the length of the spatula, it is opened using cor- residual recipient bed
neal scissors. This maneuver is repeated until a very deep
Layer-by-Layer Manual Dissection Although Tsubota et al. reported just one case
Dry manual dissection is probably the oldest of DM rupture over 17 eyes (repaired by injecting
described technique of LK, which reclaimed atten- air into the anterior chamber), this technique has
tion when Tsubota et al. applied the cataract surgery a high risk of intraoperative perforations [50, 80].
principle of “divide-and-conquer” to corneal trans- In order to reduce the risk of DM rupture, it might
plantation. After the initial trephination, the recipi- be helpful to reduce the intraocular pressure, by
ent cornea is divided into four quadrants in order to evacuating some aqueous through a peripheral
facilitate lamellar dissection at approximately 70% paracentesis, and to wet the stroma with some
of corneal depth. This division is then continued balanced salt solution (BSS) during the proce-
until a proper deep plane is exposed [50]. dure [22].
Injecting air into the stroma to create emphy- However, it is important to underline that a
sema can be helpful to understand the depth of smaller donor cannot be used in the presence of
the stromal dissection, as described by Archila in an intraoperative DM rupture because the dispar-
1984 [20, 81]. Some newest technological tools ity of curvature between the donor and the recipi-
such as intraoperative AS-OCT or a hand holder ent prevents the management of DM rupture [16,
pachymeter might be helpful as well [22, 66]. 22, 84, 85].
This technique is still a valid option, although
it is rarely the first technique of choice giving that
it is more challenging and time consuming than DALK Long-Term Outcomes
the others, and it requires a certain surgical expe-
rience to understand intraoperatively the depth of A major advantage of DALK is the avoidance of
the obtained dissection. endothelial rejection, one of the leading causes
of graft failure in PK [10–12]. When considering
Donor Preparation the endothelial cell count (ECC) as a proxy for
Donor cornea preparation can be performed graft survival, it is reasonable to think that DALK
using punch of the surgeon’s choice (with/with- surgery may provide a major advantage in
out suction). It is usually prepared from the endo- patients with long life expectancy [10].
thelial side and centration is crucial to avoid A study by Sarnicola et al. presented data on
creating an oval donor button with higher postop- 660 eyes, with various diagnoses, that underwent
erative astigmatism [82]. The DM–endothelium to DALK surgery with a mean follow-up of
complex has to be gently stripped off the donor 4.5 years (range 0.5–10 years). Graft survival
by using a dry weck surgical sponge, or with fine average was 99.3% (range 98.5–100%) [11].
non tooth forceps, taking care not to damage the Interestingly, patients were divided into five sub-
stroma. Trypan blue dye can be used to stain DM groups by different follow-ups (10–9 year,
and aid peeling; gently damaging the donor endo- 8–7 year, 6–5 year, 4–3 year, 2–1 year), and no
thelium with a swab before coloring with the significant difference was found in terms of graft
Trypan blue makes the staining stronger [16]. survival. These findings may indicate that, with
Most of the studies published in literature DALK, patients can expect having a clear graft
show that the diameter of the donor cornea gener- for a long period, as opposed to the progressive
ally oversized the recipient by 0.25 mm [52]. decline of graft survival over time that occurs
Nonetheless, we suggest the use of equivalent with PK. Similarly, the ECC with DALK shows
diameters in order to reduce postoperative myo- an average 11–12% loss only in the first 6 months
pic shift due to an oversized donor [82, 83]. The after surgery, and then it remains stable over
use of a smaller sized graft is still a matter of dis- time. The endothelial cell loss might be higher in
cussion; it could help to reduce postoperative cases that experienced an intraoperative DM rup-
myopia, but it has also been associated with a ture (19%); however, this does not seem to sig-
higher risk of DM wrinkles and glaucoma. We nificantly impact the graft survival [11, 80, 86,
suggest the use of a smaller sized graft in cases 87]. Graft failure rate was 0.6% and occurred
with anisometropia greater than 3 D myopic only within the first year postoperatively, due to
spherical equivalent is present in the eye to oper- infection or ocular surface complications. After
ate, or when both eyes require a DALK and have the initial postoperative period, thanks to the
a myopic spherical equivalent greater than 5 D. In recovery of the ocular surface defenses and to a
case of DM wrinkles appearance while suturing good stable ECC, DALK graft survival becomes
the donor, these should be managed by applying not “time dependent” and is likely to last life-
the suture, so that the wrinkles form in the periph- time, whereas with PK, the cumulative rate of
ery and therefore limit possible negative effect on graft failure increases significantly over time [2,
the VA. DM wrinkles become usually less visible 11]. These findings have been recently confirmed
and may completely disappear over time. by a 10-year graft survival comparative study
from Arundhati et al. of 362 primary DALK pro- Visual outcomes of DALK are comparable to
cedures and 306 primary PK procedures. The PK [9, 10, 17]. To be thorough, the reported results
survival rate for PK was 94.4%, 80.4%, and about DALK visual outcome actually vary among
72.0% at 1 year, 5 years, and 10 years, respec- the published studies. Although few studies have
tively, and 95.8%, 93.9%, and 93.9% at 1 year, found the outcomes of DALK to be inferior to PK,
5 years, and 10 years respectively for DALK many other studies have found comparable out-
(p = 0.001). Arundhati et al. also found that comes between PK and DALK. The reason for this
DALK resulted in fewer post-operative compli- incongruity may be imputed to a misclassification
cations and lower rates of graft rejection and fail- between DALK and ALK. In fact, no significant
ure; patients who underwent PK developed more difference in postoperative best spectacle-cor-
complications of glaucoma (29.3% vs. 11.6%, rected visual acuity (BSCVA) between STALK/
respectively; p < 0.001), allograft rejection TALK (previously dDALK) and PK has been
(16.6% vs. 1.7%, respectively; p < 0.001), epithe- found, whereas conflicting opinions have arisen
lial problems (10.4% vs. 5.5%, respectively; only regarding PK versus manual DALK (previ-
p = 0.018), and nonimmunological failure (7.8% ously pdDALK). Although there is not unanimous
vs. 1.9%, respectively; p < 0.001), compared to agreement yet, the vast majority of papers in the
DALK [88]. literature show that visual recovery after manual
DALK failures are not related to endothelial DALK (previously pdDALK) is slower but com-
rejection and DALK re-graft does not seems to parable with STALK/TALK (previously dDALK)
increase the risk of rejection [21, 89–92]. On the at a longer follow-up (usually 2–5 years), as long
contrary, a registry study over 4834 PKs by Kelly as the residual recipient bed thickness is equal or
et al. reports Kaplan–Meier survival rates of first less than 80 μm, regular in its thickness, and with
grafts for keratoconus to be 89%, 49%, and 17% a smooth surface [9, 10].
at 10 years, 20 years, and 23 years, respectively Among the main advantages of DALK over
[93]. Interestingly, a longitudinal review of 3992 PK, we also find that DALK is a “closed sky”
PKs for various diagnosis by Thompson et al. procedure, carrying a lower risk of endophthal-
showed that primary grafts had a twofold higher mitis and expulsive hemorrhage [10]. DALK also
10-year survival rate (82%) compared with initial offers a stronger wound integrity, lower risk of
re-grafts (41%) [86]. Second and third re-grafts glaucoma, and it allows for a safer staged cataract
have even a worse survival prognosis. Maguire surgery, with all the related advantages [10, 97].
et al. reported that the risk of failure 3 years after
PK increases from 17% with no previous grafts
to 53% with two or more previous PKs [94]. With urrent Indications and New
C
re-grafts, the recipient’s immune system might Prospectives
become sensitized to foreign corneal tissue,
developing an increased risk for immunologic DALK should be offered to patients suffering
rejection. from stromal diseases that have a presumably
Despite endothelial rejection is avoided with functioning endothelium, for optical, therapeutic,
DALK, epithelial, and stromal rejection can or tectonic purposes. Cases with history of preop-
still occur; however, these are usually easily erative DM rupture (i.e. penetrating corneal
managed with topical steroids [10]. The follow- wound, acute hydrops, etc.), or presumed previ-
ings have been identified as main risks factor for ous DM ruptures (i.e. radial keratotomy (RK),
rejection after DALK: shorter time of postoper- deep scars, etc.), can still be addressed with
ative local steroids (7-week median duration DALK; however, careful manual dissection
versus 1–4 years), younger age, African should be the preferred technique, as the DM can
American ethnicity, atopy, corneal neo-vascu- break at the site of previous scarring/rupture when
larization, and large limbus-to-limbus graft [95, air, fluid, or viscoelastic is injected [15]. Let us
96]. review some of the most common indications.
eratoconus and Other Ectatic

K outcomes are dependent on the indication for sur-
Disorders gery and are best for keratoconus and worse in
Keratoconus is probably the most common indi- cases with active vascularization or when a large
cation for DALK; the disease affects young graft has to be performed [108]. It is intuitive that
patients with a long life expectancy; therefore, it DALK would have a better prognosis in these
is easy to understand how they can definitely ben- patients. The preferred DALK surgical technique
efit from the long-term graft survival of DALK for keratoglobus should be manual dissection
[98–100]. Patients affected by keratoconus are [15, 85]. Vajpayee et al. have described a very
typically young, therefore with a healthy endo- interesting DALK modification called the “tuck
thelium; concomitant endothelial dystrophy is in” technique, with the aim of providing addi-
rare and usually does not compromise the endo- tional tectonic support to the peripheral cornea
thelium function at the age requiring stromal sur- and facilitate the donor suturing. The technique
gery [101]. In a series of 158 eyes affected by involves the creation of a peripheral, partial-
keratoconus, DALK showed a long-term graft thickness flange of about 2.5–3 mm of posterior
survival of 98% (6 years of average follow-up, stromal tissue of the donor lenticule. The flange
range 4–19 years) [102]. of the donor button is integrated into a 360° infe-
DALK can and should be performed even in rior stromal pocket in the host cornea, followed
keratoconus cases with a positive history for pre- by graft suturing. This procedure is indicated not
vious acute hydrops. Manual dissection tech- only for keratoglobus but also for any case with
niques should be the procedure of choice, given advanced peripheral corneal thinning, like PMD
the presumable presence of a DM break in the (PMD) [109, 110].
site of the previous hydrops. The key of perform- PMD is a rare ectatic disorder, which typi-
ing a successful surgery is to start the stromec- cally affects the inferior-peripheral cornea in a
tomy from the opposite site of the presumed DM crescentic fashion. Surgery is indicated when
break spot, dissecting the area of the hydrops as spectacles and contact lenses are unsuccessful in
last, and managing it as any other intraoperative providing satisfactory vision. Although a number
DM rupture [15, 80, 103–105]. of surgical techniques are available for patients
Recurrence of keratoconus after kerato- with PMD, there is currently no consensus on
plasty is largely related to placement of a small which method provides the most effective treat-
graft that did not remove the entire cone. A ment. Considering the peripheral location of the
repeated larger DALK, or even DALK in a previ- diseases, the advantages in term of rejections of
ous PK, can successfully manage this condition. lamellar techniques are evident [111, 112].
In the specific case of a DALK over a PK, manual Among these options, good outcomes have been
dissection techniques should be considered as reported with very large DALK for highly ectatic
preferred choice, to avoid the risk of a big bubble cases, even in eyes with previous perforations
burst when enlarging outside the full thickness [113, 114].
penetrating trephination grove [15, 106].
The surgical management of keratoglobus is Leukoma
particularly challenging due to diffuse limbus-to- Although superficial corneal opacities may be
limbus corneal thinning. There is not a unani- treated with PTK, deeper leukoma that are visu-
mous agreement about the best surgical treatment ally significant are candidates for DALK. Causes
for this disease; however, in order to avoid the of leukoma can vary and some of them deserve
placement of the graft–host junction at the few specifications.
thinned mid-periphery and to create better stabil- Among the post infective stromal scars, her-
ity, these cases require very large limbus-to- pes simplex virus (HSV) related scars are the
limbus grafts [107]. The graft proximity to the ones that may benefit the most from the low
limbus is well known as risk factor for graft immunological insult of DALK. In HSV infec-
rejection. Registry data have shown us that PK tions, especially for immune stromal keratitis, the
relationship between rejection and recurrence of scarring can be addressed with keratoplasty.
the infection is particularly significant. Rejection These eyes often present some degree of inflam-
can trigger an HSV recurrence and vice versa. In mation and corneal neovascularization, hence
a large cases series of 52 eyes, the combination of performing a DALK improves the graft survival
DALK, long-term oral antiviral therapy, and prognosis [2, 6, 108, 118–120].
long-term local steroids has shown good efficacy Stromal scarring in optical zone, resulting
in in both rehabilitating vision and preventing from penetrating or perforating corneal
recurrence of infection [115] (Fig. 18.7). Several wounds, is usually considered an indication to
other studies, despite confirming DALK as a PK. The frequent associated traumatic cataract is
good surgical treatment for post-herpetic stromal very often addressed together with the kerato-
scar, reported a certain percentage of HSV recur- plasty performing a triple procedure. Although
rence, probably due to a shorter postoperative providing satisfactory anatomical results, this
prophylactic antiviral treatment. On the contrary, approach does not allow an easy choice of the
PK outcomes are poor in cases with active vascu- refractive power of the IOL, with risk of poor
larization and should be avoided, as the survival visual results [97, 121, 122]. Recently, DALK
rate in these cases is only 60% with the first graft has been proposed to address this condition too,
[116]. Interestingly, a case report about the man- as long as there has been no significant loss of
agement of a stromal scar secondary to a HSV corneal tissue. In addition to the known advan-
recurrence on a PK graft shows that DALK is tages over PK, DALK allows the surgeon to per-
technically possible even in such cases, as long as form a staged procedure safely. Postponing
the endothelium is functional [117]. cataract surgery for a year after DALK provides
Severe thermal/chemical injuries often the surgeon with stable and more reliable param-
result in corneal opacity and LSCD. Eyes with eters to choose the appropriate IOL power and
significant LSCD are not candidates for conven- even to reduce the residual post keratoplasty
tional keratoplasty as the outcome of a corneal astigmatism. Performing DALK in eyes with a
transplant alone is poor in these patients because history of full-thickness perforating wound, and
of the LSCD will recur in the graft as soon as the therefore with a DM break, might be challenging;
donor epithelium fails; it is mandatory to perform however, it is not impossible. Given the presence
ocular surface stem cell grafting first. Once the of a break in the DM, the technique of choice is
ocular surface has been restored, residual corneal manual dissection. It is advisable to perform
a b
Fig. 18.7 DALK performed in herpes simplex scar. (a) Preoperative; (b) 1-year postoperative
DALK at list 1 year after the injury, to give progressive increase in the risk of rejection as
enough time to the cornea to cicatrize. In these with PK [130, 131].
cases, it is important to perform peripheral para- To be thorough, it has to be said that macular
centesis immediately after the trephination and corneal dystrophy may also have an associated
before starting the stromal dissection, as the cre- endothelial dysfunction. However, in advanced
ation of the paracentesis would be difficult in the cases, it might not be easy to properly evaluate
case of AC collapse. Stromal dissection should be the endothelial function, since corneal pachyme-
started from the opposite side of the corneal try is usually also abnormal. Given its remarkable
injury and continued in a manner that encircles advantages, DALK should nonetheless be per-
the corneal injury. Therefore, the perforated area formed where there is no clear evidence of endo-
is deliberately treated last. During the manual thelial dysfunction [126, 132].
dissection, once the deep plane is found, stromal
removal may be carried on with a crescent blade ctive Infections Unresponsive
A
to avoid any stromal traction that could open the to Medical Treatment
existing DM break, or potentially enlarge it. Misdiagnosis, lack of effective medical therapy,
When dissecting the area of the previous injury, and delay of treatment often compromise the suc-
aqueous leakage and collapse of the AC are usu- cess of treating fungal or amoebic infections.
ally seen, making the surgery more challenging. Conventional therapeutic PK (TPK) is the most
The leaking of the aqueous can be managed, commonly employed surgical procedure; how-
alternating fluid drainage with a sponge and care- ever, it is burdened with the risk of intraocular
ful stromal dissection maneuvers. A second spread of infectious organisms during the proce-
operator can keep gently drying the stroma to dure resulting in secondary endophthalmitis.
enhance the visibility of the stroma, while the Furthermore, given the frequent presence of
first surgeon completes the stromectomy. In cases severe inflammation, TPK is also considered at
without high posterior pressure, a gentle injection high risk of endothelial rejection [133]. Graft
of air into the AC may reform the AC and stop the clarity at 1 year postoperatively and the recur-
leakage of the fluid, making the completion of rence of infection have been found as high as
the stromectomy easier. At the end of the surgery, 51.3% and 30% respectively in TPK for fungal
once the donor is sutured, an air bubble should be keratitis [134]. The graft survival after TPK per-
left into the AC to promote the adherence between formed in cases of Acanthamoeba keratitis (AK)
the donor and recipient and reduce the risk of is also known to be poor and has been reported to
postoperative double AC [123]. be ranging from 55% to 78% at 1 year postopera-
tively, with a recurrence rate ranging from 38%
Stromal Dystrophies to 13% [134, 135]. For these reasons, TPK is usu-
PTK is usually the preferred initial therapeutic ally delayed and is performed in desperate cases
modality to treat stromal dystrophies that primar- in order to prevent impending corneal perforation
ily affect the anterior corneal stroma. However, or scleral extension [136, 137].
patients that show pan-stromal or posterior stro- Thanks to the sparing of the host endothelium,
mal involvement are best managed by DALK therapeutic DALK (TDALK) may represent a
[124]. Good visual outcomes have been reported better surgical option in these inflamed eyes,
for different types of stromal dystrophies with avoiding the risk of endothelial rejection.
both DALK and PK [125–129]. However, recur- Additionally, TDALK prevents the intraocular
rence of the dystrophy may occur in the donor spread of the infection, because the entry of the
cornea, regardless of the type of keratoplasty per- anterior chamber can be largely avoided.
formed, and clinically significant recurrence may However, TDALK could be less effective than
require multiple re-grafts. Performing DALK PK in eradicating the infection in very advanced
represents a better choice, since re-grafting in cases; therefore, an early surgical timing is cru-
DALK is relatively easy and is not burdened by cial to increase the chances of DALK to be radi-
a b c
Fig. 18.8 Early therapeutic DALK performed in a case of active Acanthamoeba keratitis, poorly responsive to medical
treatment. (a) Preoperative; (b) 1-week postoperative, (c) 18-months postoperative
cal [138–141]. Earlier surgical timing seems to active fungal keratitis. In both series, no episodes
be also the key to avoid complications due to proof rejection, recurrence, or graft failure were
longed inflammation and protracted toxic topical observed at 1 year of follow-up. Indications for
medical treatment [142–144]. an early TDALK were: poor response to targeted
In a comparative study between TDALK ver- medical therapy, a significant ulcer in optical
sus TPK for advanced microbial keratitis, Anshu zone that had not yet involved the entire stroma
et al. reported an infection recurrence rate of (deeper than 150 μm but less than 300 μm), sever-
15.3% (4 cases over 26 eyes) after TDALK and a ity and dangerousness of the infection
recurrence rate of 12% (12 cases over 100 eyes) (Acanthamoeba and fungal keratitis), and, in
after TPK [141]. However, all the recurrent cases some cases, patient compliance. Despite these
in the TDALK group were seen in manual DALK very good results, it is critical that only surgeons
cases. Three of them were successively treated with a low PK conversion rate should perform
with a second deeper/larger manual DALK, with- these procedures [138, 140] (Fig. 18.8; Video
out further recurrence. The remaining patient 18.1).
opted to be treated medically with resolution of
infection but developed graft failure. On the con- Descemetocele
trary, the recurrent cases belonging to the TPK Descemetocele is a severe complication of cor-
group were characterized by poor final outcome neal ulceration. A small descemetocele can be
and 9 out of 12 eyes ultimately required eviscera- successfully repaired with multilayer amniotic
tion. Considering that patients of the TPK group membrane grafting; however, surgical manage-
had actually more extensive and advanced ment of larger, or recurring, or infected desce-
lesions, and that the TDALK group had more metocele may require a keratoplasty. DALK can
favorable outcomes, Anshu et al. concluded that be technically performed even in these cases.
earlier intervention with lamellar surgery might Manual dissection should be the DALK tech-
have been a reasonable option [141]. nique of choice, and we recommend dissecting
Sarnicola et al. have more recently published the area of the descemetocele as last, using the
two studies demonstrating very good results of same precautions/approach suggested when per-
early TDALK in dangerous corneal infections, forming DALK for corneal penetrating injury,
precisely in 11 eyes affected by active because the descemetocele might break during
Acanthamoeba keratitis and 23 eyes affected by the surgery [15, 145, 146] (Fig. 18.9).
a b
c d
e f
Fig. 18.9 DALK performed in descemetocele, in the set- cemetocele intentionally as last; (d) recipient bed rupture
ting of HSV-related neurotrophic keratopathy. (a) in the area of the descemetocele (yellow arrow) with
Preoperative: the picture shows the recurrence of a desce- aqueous leakage; (e) 1-week postoperative; (f) 1-week
metocele (yellow arrow) after the reabsorbing of amniotic postoperative showing the presence of temporary tarsor-
membrane (white arrow); (b) Preoperative anterior seg- rhaphy to promote epithelial healing
ment OCT; (c) manual DALK dissecting the area of des-
ost Radial Keratotomy Corneal Shape

P long-term follow-up of patients has shown cor-
Disorders neal instability with frequent fluctuations of visual
RK was a popular refractive procedure used to acuity and hyperopic shifts, leading to poor visual
correct myopia before the era of excimer ablative acuity [147]. According to a recent study on
procedures. Despite initial satisfactory results, DALK performed in patients with RK, the indica-
tions for corneal transplant were significant irreg- Understanding the physiomechanical mecha-
ular astigmatism (50%), central scarring or haze nisms in DALK allows to correctly choose a
(40%), and progressive hyperopia with visual proper rescue strategy to successfully repair DM
fluctuation (10%) [148]. Although big bubble ruptures. Sarnicola et al. have described in detail
technique has been shown to be feasible, manual different approaches to adopt based on specific
dissection DALK techniques may be safer, con- scenarios of rupture. However, there are some
sidering the significant risk that the RK incisions general rules that can be valid in most cases [16,
may have reached the DM [15, 148]. 80, 85, 152, 153]. Once a recipient bed rupture is
encountered, the stromal removal should to be
continued and completed as deep and as smooth
omplications Unique to DALK
C as possible, trying to minimize any stromal
Surgery irregularities between the donor and recipient
layers that could keep the recipient bed rupture
Intraoperative unintentional rupture of recipi- patent. When completing the stromectomy in the
ent bed is a complication unique to DALK sur- area of DM rupture, it is not uncommon for the
gery, which can occur even in expert hands. The rupture to enlarge, especially in STALK (through
need of subsequent PK conversion cannot be BB type 1 or AVB) and TALK (through BB type
eliminated, but it can be largely avoided by learn- 2) cases, making subsequent stromectomy else-
ing how to manage recipient bed ruptures and where more difficult to perform. Therefore,
their consequences [80, 149, 150]. when completing the stromectomy, the area of
The rate of DM perforation varies in literature DM rupture should always intentionally dis-
from 4.5% to 45% of cases, leading to a PK con- sected last. Once the stromectomy is completed,
version in 0–86% of cases, depending of sur- the donor graft, denuded of its endothelium, can
geon’s skill [79]. In fact, the rate of DM rupture be sutured to the recipient. After suturing the
and PK conversion gradually decreases as sur- donor, an air bubble can be injected into the AC
geons become more experienced [149]. In a case (about 70% of the AC) to tamponade the rupture.
series of 1443 DALK procedures, Sarnicola et al. Same size diameter for donor and recipient
reported 119 (8%) intraoperative recipient bed should be used in cases of DM ruptures to avoid
ruptures, which were all successfully repaired curvature disparity between the graft donor and
with no need for PK conversion. One hundred recipient cornea, which could create a refractory
(84%) of the cases with ruptures resolved by the double AC. At the end of surgery, rotating the
first postoperative day, whereas 19 cases (16%) eye in different positions may facilitate the
developed double AC, which were all fixed by drainage of interface fluid, promoting adherence
using different strategies. Graft survival of the between the recipient bed and donor graft. The
eyes that experienced an intraoperative rupture postoperative head’s position of the patient plays
was 99% at last follow up, showing that it is a very important role; it has to be set so that the
worth trying to repair all DM ruptures in DALK, air bubble in the AC would tamponade the rup-
whereas immediate PK conversion should be ture (i.e. sitting position for superior ruptures,
avoided [80]. lying on the opposite site of a DM break for lat-
Only a few other studies in the literature have eral ruptures, and supine position with chin
reported on outcomes of intraoperative DM rup- hyperextension for inferior ruptures).
tures. Two large comparative studies about Pharmacological pupil dilation and close patient
DALK with and without DM ruptures (Senoo monitoring (for at list the first 6 h after surgery)
et al. [150] and Huang et al. [151]) also found are pivotal to prevent/manage pupillary block. At
that intraoperative recipient bed perforation did the patient’s discharge, the air in the AC should
not affect the graft survival or the visual acuity not be more than 50–60% and with good pupil
[150, 151]. dilation [80].
Double AC is the most frequent postoperative of other effective techniques is useful in cases
complication, and it usually occurs in cases that were BB fails or is not indicated.
experienced intraoperative recipient bed rupture. • When properly executed, DALK manual dis-
The reasons why such complication may occur section techniques can be as effective as sub-
are multiples. In some cases, the air bubble left in total anterior lamellar keratoplasty (STALK)
the AC at the end of surgery can shrink, becom- techniques.
ing insufficient to tamponade the rupture. These • Despite technically more challenging, DALK
cases can be fixed with simple re-bubbling. can be performed even in cases with a history
Another reason for double AC can be the patient’s of hydrops, RK, perforating wounds and des-
noncompliance with the head positioning; this is cemetocele. Manual dissection techniques
more common for cases with inferior rupture should be preferred in such cases.
because the required head position can be quite • With appropriate rescue techniques to repair
uncomfortable. Re-bubbling emphasizing the the recipient bed rupture and to manage post-
importance of head positioning to the patient can operative double AC, conversion to PK can be
be resolutive. In case of persistent double AC, the avoided in the majority of cases.
regularity of the sutures should carefully be
assessed. Sometimes, certain sutures can be con-
siderably tighter than others, pushing the donor
on the recipient bed not homogeneously and References
causing a persistent double chamber. In these
cases, the tight sutures should be replaced right 1. Colby K. Update on corneal transplant in 2021.
JAMA. 2021;325(18):1886–7. https://doi.
after the re-bubbling. The shape of the air bubble org/10.1001/jama.2020.17382.
in the AC during the re-bubbling can help evalu- 2. Tan DT, Dart JK, Holland EJ, Kinoshita S. Corneal
ating the tension of the sutures; a circular shape transplantation. Lancet. 2012;379(9827):1749–61.
of the bubble indicates a uniform distribution of https://doi.org/10.1016/S0140-6736(12)60437-1.
3. Rich LF. Expanding the scope of lamellar kerato-
the tension generated by the sutures, whereas an plasty. Trans Am Ophthalmol Soc. 1999;97:771–
oval or irregular shape of the air bubble suggests 814. PMCID: PMC1298281.
otherwise. Finally, significant donor–recipient 4. America EBAo. Eye banking statistical report.
disparity of curvature, in the presence of a recipi- Washington, DC: Eye Bank Association of America;
1997.
ent’s bed rupture during manual DALK, may 5. Sarnicola C, Sarnicola E, Perri P, Sarnicola
cause a persistent double chamber. In these cases, V. Recent developments in cornea and corneal
re-bubbling may not work, requiring surgical transplants. In: Grzybowski A, editor. Current
correction [80, 85, 153]. concepts in ophthalmology. Cham: Springer
International Publishing; 2020. p. 35–53. https://doi.
org/10.1007/978-3-030-25389-9.
Take Home Notes 6. Cheung AY, Sarnicola E, Denny MR, Sepsakos L,
• DALK has become the gold standard tech- Auteri NJ, Holland EJ. Limbal stem cell deficiency:
nique to treat stromal diseases, when the demographics and clinical characteristics of a large
retrospective series at a single tertiary referral cen-
endothelium is functioning. ter. Cornea. 2021;40(12):1525–31. https://doi.
• DALK avoidance of endothelial rejection org/10.1097/ICO.0000000000002770.
makes a very meaningful impact not only in 7. Cheung AY, Sarnicola E, Kurji KH, Govil A,
the prognosis of patients with keratoconus but Mogilishetty G, Eslani M, et al. Cincinnati proto-
col for preoperative screening and donor selection
also for high-risk corneal transplants (i.e. dis- for ocular surface stem cell transplantation. Cornea.
orders that need large grafts, presence of neo- 2018;37(9):1192–7. https://doi.org/10.1097/
vascularization, inflammation, active ICO.0000000000001662.
infections, etc.). 8. Cheung AY, Sarnicola E, Govil A, Holland
EJ. Combined conjunctival limbal autografts and
• The big bubble (BB) technique is the most living-related conjunctival limbal allografts for
common procedure; however, the mastership severe unilateral ocular surface failure. Cornea.
2017;36(12):1570–5. https://doi.org/10.1097/ Cornea. 2021;40(5):613–7. https://doi.org/10.1097/

ICO.0000000000001376. ICO.0000000000002525.
9. Sarnicola E, Sarnicola C, Cheung AY, Holland 22. Enrica S, Caterina S, Vincenzo S. Techniques of ante-
EJ, Sarnicola V. Surgical corneal anatomy in deep rior lamellar keratoplasty. In: Mannis MJ, Holland
anterior lamellar keratoplasty: suggestion of new EJ, editors. Cornea, vol. 2. 5th ed. Amsterdam:
acronyms. Cornea. 2019;38(4):515–22. https://doi. Elsevier; 2021.
org/10.1097/ICO.0000000000001845. 23. Werblin TP, Kaufman HE, Friedlander MH,
10. Reinhart WJ, Musch DC, Jacobs DS, Lee WB, Granet N. Epikeratophakia: the surgical cor-
Kaufman SC, Shtein RM. Deep anterior lamel- rection of aphakia. III. Preliminary results of
lar keratoplasty as an alternative to penetrating a prospective clinical trial. Arch Ophthalmol.
keratoplasty a report by the American Academy of 1981;99(11):1957–60. https://doi.org/10.1001/arch
Ophthalmology. Ophthalmology. 2011;118(1):209– opht.1981.03930020833002.
18. https://doi.org/10.1016/j.ophtha.2010.11.002. 24. Kang J, Cabot F, Yoo SH. Long-term follow-
11. Sarnicola V, Toro P, Sarnicola C, Sarnicola E, up of epikeratophakia. J Cataract Refract Surg.

Ruggiero A. Long-term graft survival in deep ante- 2015;41(3):670–3. https://doi.org/10.1016/j.
rior lamellar keratoplasty. Cornea. 2012;31(6):621– jcrs.2014.11.035.
6. https://doi.org/10.1097/ICO.0b013e31823d0412. 25. Uusitalo RJ, Uusitalo HM. Long-term fol-
12. Borderie VM, Sandali O, Bullet J, Gaujoux T, low-up of pediatric epikeratophakia. J
Touzeau O, Laroche L. Long-term results of deep Refract Surg. 1997;13(1):45–54. https://doi.
anterior lamellar versus penetrating keratoplasty. org/10.3928/1081-597X-19970101-12.
Ophthalmology. 2012;119(2):249–55. https://doi. 26. Cameron JA, Cotter JB, Risco JM, Alvarez
org/10.1016/j.ophtha.2011.07.057. H. Epikeratoplasty for keratoglobus associated with
13. Fontana L, Parente G, Sincich A, Tassinari blue sclera. Ophthalmology. 1991;98(4):446–52.
G. Influence of graft-host interface on the qual- https://doi.org/10.1016/s0161-6420(91)32271-1.
ity of vision after deep anterior lamellar kera- 27. Javadi MA, Kanavi MR, Ahmadi M, Yazdani
toplasty in patients with keratoconus. Cornea. S. Outcomes of epikeratoplasty for advanced kera-
2011;30(5):497–502. https://doi.org/10.1097/ toglobus. Cornea. 2007;26(2):154–7. https://doi.
ico.0b013e3181d25e4d. org/10.1097/01.ico.0000244878.38621.fc.
14. Sarnicola E, Sarnicola C, Cheung AY, Holland EJ, 28. Macsai MS, Lemley HL, Schwartz T. Management
Sarnicola V. Reply. Cornea. 2019;38(10):e45–e6. of oculus fragilis in Ehlers-Danlos type
https://doi.org/10.1097/ICO.0000000000002077. VI. Cornea. 2000;19(1):104–7. https://doi.
15. Sarnicola C, Sarnicola E, Sarnicola V, Anwar org/10.1097/00003226-200001000-00020.
M. Indications for anterior lamellar keratoplasty. In: 29. Walkden A, Burkitt-Wright E, Au L. Brittle cornea
Mannis MJ, Holland EJ, editors. Cornea, vol. 2. 5th syndrome: current perspectives. Clin Ophthalmol.
ed. Amsterdam: Elsevier; 2021. 2019;13:1511–6. https://doi.org/10.2147/OPTH.
16. Sarnicola E, Sarnicola C, Sarnicola V. Deep anterior S185287.
lamellar keratoplasty: surgical technique, indica- 30. Jhanji V, Young AL, Mehta JS, Sharma N, Agarwal
tions, clinical results and complications. In: Guell JL, T, Vajpayee RB. Management of corneal perforation.
editor. Cornea. ESASO Course series, vol. 6. Basel: Surv Ophthalmol. 2011;56(6):522–38. https://doi.
Karger; 2015. https://doi.org/10.1159/000381495. org/10.1016/j.survophthal.2011.06.003.
17. Sarnicola V, Toro P, Gentile D, Hannush 31. Huang D, Qiu WY, Zhang B, Wang BH, Yao
SB. Descemetic DALK and predescemetic YF. Peripheral deep anterior lamellar keratoplasty
DALK: outcomes in 236 cases of keratoconus. using a cryopreserved donor cornea for Terrien’s
Cornea. 2010;29(1):53–9. https://doi.org/10.1097/ marginal degeneration. J Zhejiang Univ Sci B.
ICO.0b013e3181a31aea. 2014;15(12):1055–63. https://doi.org/10.1631/jzus.
18. Barraquer JI. Lamellar keratoplasty. (Special tech- B1400083.
niques). Ann Ophthalmol. 1972;4(6):437–69. 32. Cheng CL, Theng JT, Tan DT. Compressive
19. Ardjomand N, Hau S, McAlister JC, Bunce C, C-shaped lamellar keratoplasty: a surgical alterna-
Galaretta D, Tuft SJ, et al. Quality of vision and tive for the management of severe astigmatism from
graft thickness in deep anterior lamellar and pen- peripheral corneal degeneration. Ophthalmology.
etrating corneal allografts. Am J Ophthalmol. 2005;112(3):425–30. https://doi.org/10.1016/j.
2007;143(2):228–35. https://doi.org/10.1016/j. ophtha.2004.10.033.
ajo.2006.10.043. 33. Cheung AY, Sarnicola E, Kurji KH, Genereux BM,
20. Archila EA. Deep lamellar keratoplasty dissection of Holland EJ. Three hundred sixty-degree fuchs
host tissue with intrastromal air injection. Cornea. superficial marginal keratitis managed with annular
1984;3(3):217–8. lamellar keratoplasty. Cornea. 2018;37(2):260–2.

21. Sarnicola C, Sarnicola E, Cheung AY, Panico https://doi.org/10.1097/ICO.0000000000001433.
E, Panico C, Sarnicola V. Deep anterior lamel- 34. Lohchab M, Prakash G, Arora T, Maharana P, Jhanji
lar keratoplasty after previous anterior lamel- V, Sharma N, et al. Surgical management of periph-
lar keratoplasty to improve the visual outcomes. eral corneal thinning disorders. Surv Ophthalmol.
2019;64(1):67–78. https://doi.org/10.1016/j. 2010;29(4):388–91. https://doi.org/10.1097/

survophthal.2018.06.002. ICO.0b013e3181ba7016.
35. Arenas E, Esquenazi S, Anwar M, Terry M. Lamellar 47. Luengo-Gimeno F, Tan DT, Mehta JS. Evolution of
corneal transplantation. Surv Ophthalmol. deep anterior lamellar keratoplasty (DALK). Ocul
2012;57(6):510–29. https://doi.org/10.1016/j. Surf. 2011;9(2):98–110. https://doi.org/10.1016/
survophthal.2012.01.009. s1542-0124(11)70017-9.
36. Kaufman HE, Insler MS, Ibrahim-Elzembely 48. Melles GR, Remeijer L, Geerards AJ, Beekhuis
HA, Kaufman SC. Human fibrin tissue adhesive WH. A quick surgical technique for deep, ante-
for sutureless lamellar keratoplasty and scleral rior lamellar keratoplasty using viscodissection.
patch adhesion: a pilot study. Ophthalmology. Cornea. 2000;19(4):427–32. https://doi.
2003;110(11):2168–72. https://doi.org/10.1016/ org/10.1097/00003226-200007000-00004.
S0161-6420(03)00832-7. 49. Melles GR, Rietveld FJ, Beekhuis WH, Binder
37. Patel AK, Scorcia V, Kadyan A, Lapenna L, Ponzin PS. A technique to visualize corneal incision and
D, Busin M. Microkeratome-assisted superficial lamellar dissection depth during surgery. Cornea.
anterior lamellar keratoplasty for anterior stromal 1999;18(1):80–6.
corneal opacities after penetrating keratoplasty. 50. Tsubota K, Kaido M, Monden Y, Satake Y, Bissen-
Cornea. 2012;31(1):101–5. https://doi.org/10.1097/ Miyajima H, Shimazaki J. A new surgical technique
ICO.0b013e31820c9fd1. for deep lamellar keratoplasty with single running
38. Fogla R, Knyazer B. Microkeratome-assisted suture adjustment. Am J Ophthalmol. 1998;126(1):1–
two-stage technique of superficial anterior lamel- 8. https://doi.org/10.1016/s0002-9394(98)00067-1.
lar keratoplasty for Reis-Bücklers corneal dystro- 51. Polack FM. Lamellar keratoplasty. Malbran’s
phy. Cornea. 2014;33(10):1118–22. https://doi. “peeling off” technique. Arch Ophthalmol.
org/10.1097/ICO.0000000000000189. 1971;86(3):293–5. https://doi.org/10.1001/archo
39. Shousha MA, Yoo SH, Kymionis GD, Ide T, Feuer pht.1971.01000010295010.
W, Karp CL, et al. Long-term results of femtosecond 52. Fogla R, Sahay P, Sharma N. Preferred practice pat-
laser-assisted sutureless anterior lamellar kerato- tern and observed outcome of deep anterior lamellar
plasty. Ophthalmology. 2011;118(2):315–23. https:// keratoplasty - a survey of Indian corneal surgeons.
doi.org/10.1016/j.ophtha.2010.06.037. Indian J Ophthalmol. 2021;69(6):1553–8. https://
40. Seitz B, Lisch W. Stage-related therapy of corneal doi.org/10.4103/ijo.IJO_3067_20.
dystrophies. Dev Ophthalmol. 2011;48:116–53. 53. McKee HD, Irion LC, Carley FM, Jhanji V,
https://doi.org/10.1159/000324081. Brahma AK. Dissection plane of the clear margin
41. Deshmukh R, Reddy JC, Rapuano CJ, Vaddavalli big-bubble in deep anterior lamellar keratoplasty.
PK. Phototherapeutic keratectomy: indications, Cornea. 2013;32(4):e51–2. https://doi.org/10.1097/
methods and decision making. Indian J Ophthalmol. ICO.0b013e318262e8ce.
2020;68(12):2856–66. https://doi.org/10.4103/ijo. 54. Goweida MB, Ragab AM, Liu C. Management of
IJO_1524_20. type 2 bubble formed during big bubble deep anterior
42. Beltz J, Madi S, Santorum P. Superficial ante- lamellar keratoplasty. Cornea. 2019;38(2):189–93.
rior lamellar keratoplasty: description of tech- https://doi.org/10.1097/ICO.0000000000001815.
nique and presentation of results. Min Oftalmol. 55. Gujar P. Type 2 big bubble deep anterior lamellar
2018;60:156–61. https://doi.org/10.23736/S0026- keratoplasty-serial anterior segment optical coher-
4903.18.01811-1. Madi S, editor. ence tomography documentation showing resolu-
43. Anwar M, Teichmann KD. Big-bubble tech- tion of bubble in the postoperative period. Indian
nique to bare Descemet’s membrane in anterior J Ophthalmol. 2017;65(10):1017–8. https://doi.
lamellar keratoplasty. J Cataract Refract Surg. org/10.4103/ijo.IJO_343_17.
2002;28(3):398–403. https://doi.org/10.1016/ 56. Pasricha ND, Shieh C, Carrasco-Zevallos OM,
s0886-3350(01)01181-6. Keller B, Cunefare D, Mehta JS, et al. Needle depth
44. Melles GR, Lander F, Rietveld FJ, Remeijer L, and big-bubble success in deep anterior lamellar
Beekhuis WH, Binder PS. A new surgical technique keratoplasty: an ex vivo microscope-integrated OCT
for deep stromal, anterior lamellar keratoplasty. study. Cornea. 2016;35(11):1471–7. https://doi.
Br J Ophthalmol. 1999;83(3):327–33. https://doi. org/10.1097/ICO.0000000000000948.
org/10.1136/bjo.83.3.327. 57. Sarnicola V, Toro P. Blunt cannula for desce-
45. Dua HS, Faraj LA, Said DG, Gray T, Lowe metic deep anterior lamellar keratoplasty. Cornea.
J. Human corneal anatomy redefined: a novel pre- 2011;30(8):895–8. https://doi.org/10.1097/
Descemet’s layer (Dua’s layer). Ophthalmology. ICO.0b013e3181e848c3.
2013;120(9):1778–85. https://doi.org/10.1016/j. 58. Sarnicola E, Sarnicola C, Sabatino F, Tosi
ophtha.2013.01.018. GM, Perri P, Sarnicola V. Cannula DALK ver-
46. Jafarinasab MR, Rahmati-Kamel M, Kanavi MR, sus needle DALK for keratoconus. Cornea.
Feizi S. Dissection plane in deep anterior lamellar 2016;35(12):1508–11. https://doi.org/10.1097/
keratoplasty using the big-bubble technique. Cornea. ICO.0000000000001032.
59. Fogla R, Padmanabhan P. Results of deep lamel- sia. Int J Ophthalmol. 2014;7(4):638–43. https://doi.
lar keratoplasty using the big-bubble technique org/10.3980/j.issn.2222-3959.2014.04.09.
in patients with keratoconus. Am J Ophthalmol. 72. Parthasarathy A, Por YM, Tan DT. Using a “small
2006;141(2):254–9. https://doi.org/10.1016/j. bubble technique” to aid in success in Anwar’s
ajo.2005.08.064. “big bubble technique” of deep lamellar kerato-
60. Fogla R. Deep anterior lamellar keratoplasty plasty with complete baring of Descemet’s mem-
in the management of keratoconus. Indian J brane. Br J Ophthalmol. 2008;92(3):422. https://doi.
Ophthalmol. 2013;61(8):465–8. https://doi. org/10.1136/bjo.2006.113357.
org/10.4103/0301-4738.116061. 73. Goshe J, Terry MA, Shamie N, Li J. Ophthalmic
61. Fournié P, Malecaze F, Coullet J, Arné JL. Variant of viscosurgical device-assisted incision modifica-
the big bubble technique in deep anterior lamellar ker- tion for the big-bubble technique in deep anterior
atoplasty. J Cataract Refract Surg. 2007;33(3):371– lamellar keratoplasty. J Cataract Refract Surg.
5. https://doi.org/10.1016/j.jcrs.2006.10.053. 2011;37(11):1923–7. https://doi.org/10.1016/j.
62. Ghanem RC, Ghanem MA. Pachymetry-guided jcrs.2011.09.014.
intrastromal air injection (“pachy-bubble”) for 74. Manche EE, Holland GN, Maloney RK. Deep lamel-
deep anterior lamellar keratoplasty. Cornea. lar keratoplasty using viscoelastic dissection. Arch
2012;31(9):1087–91. https://doi.org/10.1097/ Ophthalmol. 1999;117(11):1561–5. https://doi.
ICO.0b013e31823f8f2d. org/10.1001/archopht.117.11.1561.
63. Scorcia V, Busin M, Lucisano A, Beltz J, Carta 75. Muftuoglu O, Toro P, Hogan RN, Bowman RW,
A, Scorcia G. Anterior segment optical coher- Cavanagh HD, McCulley JP, et al. Sarnicola air-
ence tomography-guided big-bubble technique. visco bubble technique in deep anterior lamellar
Ophthalmology. 2013;120(3):471–6. https://doi. keratoplasty. Cornea. 2013;32(4):527–32. https://
org/10.1016/j.ophtha.2012.08.041. doi.org/10.1097/ICO.0b013e31826cbe99.
64. Buzzonetti L, Laborante A, Petrocelli G. Standardized 76. Malbran E, Stefani C. Lamellar keratoplasty in cor-
big-bubble technique in deep anterior lamellar kera- neal ectasias. Ophthalmologica. 1972;164(1):50–8.
toplasty assisted by the femtosecond laser. J Cataract https://doi.org/10.1159/000306704.
Refract Surg. 2010;36(10):1631–6. https://doi. 77. Malbran E. Lamellar keratoplasty in keratoconus.
org/10.1016/j.jcrs.2010.08.013. Int Ophthalmol Clin. 1966;6(1):99–109.
65. Chamberlain WD. Femtosecond laser-assisted 78. Smolek MK, McCarey BE. Interlamellar adhe-
deep anterior lamellar keratoplasty. Curr Opin sive strength in human eyebank corneas. Invest
Ophthalmol. 2019;30(4):256–63. https://doi. Ophthalmol Vis Sci. 1990;31(6):1087–95.
org/10.1097/ICU.0000000000000574. 79. Sarnicola C, Sarnicola V, Romani A, Sarnicola
66. Price FW. Intraoperative optical coherence tomog- E. Rediscovering a valuable manual Deep
raphy: game-changing technology. Cornea. Anterior Lamellar Keratoplasty (DALK) tech-
2021;40(6):675–8. https://doi.org/10.1097/ nique: outcomes of 42 “Peeling-off” DALK. Eur
ICO.0000000000002629. J Ophthalmol. 2022;33:900. https://doi.
67. De Benito-Llopis L, Mehta JS, Angunawela RI, org/10.1177/11206721221132622.
Ang M, Tan DT. Intraoperative anterior segment 80. Sarnicola C, Sarnicola E, Cheung AY, Sarnicola
optical coherence tomography: a novel assessment V. Deep anterior lamellar keratoplasty: can all rup-
tool during deep anterior lamellar keratoplasty. Am tures be fixed? Cornea. 2023;42:80–8. https://doi.
J Ophthalmol. 2014;157(2):334–41.e3. https://doi. org/10.1097/ICO.0000000000003054.
org/10.1016/j.ajo.2013.10.001. 81. Price FW. Air lamellar keratoplasty. Refract Corneal
68. de Macedo JP, de Oliveira LA, Hirai F, de Sousa Surg. 1989;5(4):240–3.
LB. Femtosecond laser-assisted deep anterior lamel- 82. Nanavaty MA, Vijjan KS, Yvon C. Deep anterior
lar keratoplasty in phototherapeutic keratectomy lamellar keratoplasty: a surgeon’s guide. J Curr
versus the big-bubble technique in keratoconus. Ophthalmol. 2018;30(4):297–310. https://doi.
Int J Ophthalmol. 2018;11(5):807–12. https://doi. org/10.1016/j.joco.2018.06.004.
org/10.18240/ijo.2018.05.15. 83. Huang T, Ouyang C, Hou C, Wu Q, Hu Y. Outcomes
69. Farid M, Steinert RF. Femtosecond laser- of same-size host and donor trephine in deep
assisted corneal surgery. Curr Opin Ophthalmol. anterior lamellar keratoplasty for keratoconus.
2010;21(4):288–92. https://doi.org/10.1097/ Am J Ophthalmol. 2016;166:8–13. https://doi.
ICU.0b013e32833a8dbc. org/10.1016/j.ajo.2016.03.018.
70. Ziebarth NM, Dias J, Hürmeriç V, Shousha MA, 84. Shi W, Li S, Gao H, Wang T, Xie L. Modified
Yau CB, Moy VT, et al. Quality of corneal lamel- deep lamellar keratoplasty for the treatment of
lar cuts quantified using atomic force microscopy. advanced-stage keratoconus with steep curvature.
J Cataract Refract Surg. 2013;39(1):110–7. https:// Ophthalmology. 2010;117(2):226–31. https://doi.
doi.org/10.1016/j.jcrs.2012.07.040. org/10.1016/j.ophtha.2009.07.005.
71. Lu Y, Shi YH, Yang LP, Ge YR, Chen XF, Wu Y, 85. Sarnicola E, Sarnicola C, Cheung AY, Panico E,
et al. Femtosecond laser-assisted deep anterior Panico C, Sarnicola V. Total or subtotal full thick-
lamellar keratoplasty for keratoconus and keratecta- ness recipient bed cut to repair donor-recipient cur-
vature disparity in cases of DM rupture with manual 98. Williams KA, Keane MC. Outcomes of corneal
DALK. Eur J Ophthalmol. 2020;30:1172. https:// transplantation in Australia, in an era of lamellar ker-
doi.org/10.1177/1120672120932833. atoplasty. Clin Exp Ophthalmol. 2022;50(4):374–85.
86. Thompson RW, Price MO, Bowers PJ, Price https://doi.org/10.1111/ceo.14089.
FW. Long-term graft survival after penetrating kera- 99. Eye Bank Association of America (EBAA). Eye
toplasty. Ophthalmology. 2003;110(7):1396–402. banking statistical report 2017. Washington, DC:
https://doi.org/10.1016/S0161-6420(03)00463-9. EBAA; 2018.
87. Zadok D, Schwarts S, Marcovich A, Barkana 100. Arnalich-Montiel F, Alió Del Barrio JL, Alió
Y, Morad Y, Eting E, et al. Penetrating kerato- JL. Corneal surgery in keratoconus: which type,
plasty for keratoconus: long-term results. Cornea. which technique, which outcomes? Eye Vis. 2016;3:2.
2005;24(8):959–61. https://doi.org/10.1097/01. https://doi.org/10.1186/s40662-016-0033-y.
ico.0000159729.51015.b4. 101. Cremona FA, Ghosheh FR, Rapuano CJ, Eagle RC,
88. Arundhati A, Chew MC, Lim L, Mehta JS, Lang Hammersmith KM, Laibson PR, et al. Keratoconus
SS, Htoon HM, et al. Comparative study of long- associated with other corneal dystrophies. Cornea.
term graft survival between penetrating kerato- 2009;28(2):127–35. https://doi.org/10.1097/
plasty and deep anterior lamellar keratoplasty. ICO.0b013e3181859935.
Am J Ophthalmol. 2021;224:207–16. https://doi. 102. Romano V, Iovieno A, Parente G, Soldani AM,
org/10.1016/j.ajo.2020.11.006. Fontana L. Long-term clinical outcomes of deep
89. Woo JH, Tan YL, Htoon HM, Tan DTH, Mehta anterior lamellar keratoplasty in patients with kera-
JS. Outcomes of repeat anterior lamellar keratoconus. Am J Ophthalmol. 2015;159(3):505–11.
toplasty. Cornea. 2020;39(2):200–6. https://doi. https://doi.org/10.1016/j.ajo.2014.11.033.
org/10.1097/ICO.0000000000002167. 103. Chew AC, Mehta JS, Tan DT. Deep lamellar kera-
90. Alió Del Barrio JL, Bhogal M, Ang M, Ziaei M, toplasty after resolution of hydrops in keratoconus.
Robbie S, Montesel A, et al. Corneal transplan- Cornea. 2011;30(4):454–9. https://doi.org/10.1097/
tation after failed grafts: options and outcomes. ICO.0b013e3181f0b1f3.
Surv Ophthalmol. 2021;66(1):20–40. https://doi. 104. Fuest M, Mehta JS. Strategies for deep anterior
org/10.1016/j.survophthal.2020.10.003. lamellar keratoplasty after hydrops in keratoconus.
91. Montesel A, Alió Del Barrio JL, Yébana Rubio P, Eye Contact Lens. 2018;44(2):69–76. https://doi.
Alió JL. Corneal graft surgery: a monocentric long- org/10.1097/ICL.0000000000000383.
term analysis. Eur J Ophthalmol. 2021;31(4):1700– 105. Jacob S, Narasimhan S, Agarwal A, Sambath J,
8. https://doi.org/10.1177/1120672120947592. Umamaheshwari G, Saijimol AI. Primary modi-
92. Gómez-Benlloch A, Montesel A, Pareja-Aricò L, fied predescemetic deep anterior lamellar kera-
Mingo-Botín D, Michael R, Barraquer RI, et al. toplasty in acute corneal hydrops. Cornea.
Causes of corneal transplant failure: a multicen- 2018;37(10):1328–33. https://doi.org/10.1097/
tric study. Acta Ophthalmol. 2021;99(6):e922–e8. ICO.0000000000001693.
https://doi.org/10.1111/aos.14708. 106. Lake D, Hamada S, Khan S, Daya SM. Deep ante-
93. Kelly TL, Williams KA, Coster DJ, Registry rior lamellar keratoplasty over penetrating kera-
ACG. Corneal transplantation for keratoconus: a toplasty for host rim thinning and ectasia. Cornea.
registry study. Arch Ophthalmol. 2011;129(6):691– 2009;28(5):489–92. https://doi.org/10.1097/
7. https://doi.org/10.1001/archophthalmol.2011.7. ICO.0b013e31818d3b3c.
94. Maguire MG, Stark WJ, Gottsch JD, Stulting 107. Wallang BS, Das S. Keratoglobus. Eye.
RD, Sugar A, Fink NE, et al. Risk factors for cor- 2013;27(9):1004–12. https://doi.org/10.1038/
neal graft failure and rejection in the collabora- eye.2013.130.
tive corneal transplantation studies. Collaborative 108. Williams KA, Lowe M, Bartlett C, Kelly TL, Coster
Corneal Transplantation Studies Research Group. DJ, Contributors A. Risk factors for human corneal
Ophthalmology. 1994;101(9):1536–47. https://doi. graft failure within the Australian corneal graft reg-
org/10.1016/s0161-6420(94)31138-9. istry. Transplantation. 2008;86(12):1720–4. https://
95. Gonzalez A, Price MO, Feng MT, Lee C, Arbelaez doi.org/10.1097/TP.0b013e3181903b0a.
JG, Price FW. Immunologic rejection episodes after 109. Vajpayee RB, Bhartiya P, Sharma N. Central
deep anterior lamellar keratoplasty: incidence and lamellar keratoplasty with peripheral intralamel-
risk factors. Cornea. 2017;36(9):1076–82. https:// lar tuck: a new surgical technique for kerato-
doi.org/10.1097/ICO.0000000000001223. globus. Cornea. 2002;21(7):657–60. https://doi.
96. Watson SL, Tuft SJ, Dart JK. Patterns of rejection org/10.1097/00003226-200210000-00005.
after deep lamellar keratoplasty. Ophthalmology. 110. Kaushal S, Jhanji V, Sharma N, Tandon R, Titiyal
2006;113(4):556–60. https://doi.org/10.1016/j. JS, Vajpayee RB. “Tuck In” Lamellar Keratoplasty
ophtha.2006.01.006. (TILK) for corneal ectasias involving corneal
97. Sarnicola C, Sarnicola E, Panico E, Panico C, periphery. Br J Ophthalmol. 2008;92(2):286–90.
Sarnicola V. Cataract surgery in corneal transplan- https://doi.org/10.1136/bjo.2007.12462.
tation. Curr Opin Ophthalmol. 2020;31(1):23–7. 111. Jinabhai A, Radhakrishnan H, O’Donnell C. Pellucid
https://doi.org/10.1097/ICU.0000000000000635. corneal marginal degeneration: a review. Cont
Lens Anter Eye. 2011;34(2):56–63. https://doi. 125. Mohamed A, Chaurasia S, Ramappa M, Murthy SI,
org/10.1016/j.clae.2010.11.007. Garg P. Outcomes of keratoplasty in lattice corneal
112. Moshirfar M, Edmonds JN, Behunin NL, dystrophy in a large cohort of Indian eyes. Indian
Christiansen SM. Current options in the man- J Ophthalmol. 2018;66(5):666–72. https://doi.
agement of pellucid marginal degeneration. J org/10.4103/ijo.IJO_1150_17.
Refract Surg. 2014;30(7):474–85. https://doi. 126. Sogutlu Sari E, Kubaloglu A, Unal M, Pinero D,
org/10.3928/1081597X-20140429-02. Bulut N, Erol MK, et al. Deep anterior lamellar
113. Millar MJ, Maloof A. Deep lamellar kerato- keratoplasty versus penetrating keratoplasty for
plasty for pellucid marginal degeneration: review macular corneal dystrophy: a randomized trial. Am
of management options for corneal perforation. J Ophthalmol. 2013;156(2):267–74.e1. https://doi.
Cornea. 2008;27(8):953–6. https://doi.org/10.1097/ org/10.1016/j.ajo.2013.03.007.
ICO.0b013e31816ed516. 127. Unal M, Arslan OS, Atalay E, Mangan MS, Bilgin
114. Al-Torbak AA. Deep anterior lamellar kerato- AB. Deep anterior lamellar keratoplasty for the
plasty for pellucid marginal degeneration. Saudi treatment of stromal corneal dystrophies. Cornea.
J Ophthalmol. 2013;27(1):11–4. https://doi. 2013;32(3):301–5. https://doi.org/10.1097/
org/10.1016/j.sjopt.2012.04.001. ICO.0b013e31825718ca.
115. Sarnicola V, Toro P. Deep anterior lamellar kera- 128. Vajpayee RB, Tyagi J, Sharma N, Kumar N, Jhanji
toplasty in herpes simplex corneal opacities. V, Titiyal JS. Deep anterior lamellar keratoplasty by
Cornea. 2010;29(1):60–4. https://doi.org/10.1097/ big-bubble technique for treatment corneal stromal
ICO.0b013e3181a317d3. opacities. Am J Ophthalmol. 2007;143(6):954–7.
116. Skarentzos K, Chatzimichael E, Panagiotopoulou https://doi.org/10.1016/j.ajo.2007.02.036.
EK, Taliantzis S, Konstantinidis A, Labiris 129. Shimazaki J, Shimmura S, Ishioka M, Tsubota
G. Corneal graft success rates in HSV kera- K. Randomized clinical trial of deep lamellar kerato-
titis: a systematic review. Acta Med (Hradec plasty vs penetrating keratoplasty. Am J Ophthalmol.
Kralove). 2020;63(4):150–8. https://doi. 2002;134(2):159–65. https://doi.org/10.1016/
org/10.14712/18059694.2020.57. s0002-9394(02)01523-4.
117. Ramamurthi S, Cornish KS, Steeples L, Ramaesh 130. Lewis DR, Price MO, Feng MT, Price
K. Deep anterior lamellar keratoplasty on a previously FW. Recurrence of granular corneal dystrophy type
failed full-thickness graft. Cornea. 2009;28(4):456– 1 after phototherapeutic keratectomy, lamellar kera-
7. https://doi.org/10.1097/ICO.0b013e31818c2af7. toplasty, and penetrating keratoplasty in a single
118. Sepsakos L, Cheung AY, Holland EJ. Outcomes of population. Cornea. 2017;36(10):1227–32. https://
keratoplasty after ocular surface stem cell trans- doi.org/10.1097/ICO.0000000000001303.
plantation. Cornea. 2017;36(9):1025–30. https://doi. 131. Marcon AS, Cohen EJ, Rapuano CJ,
org/10.1097/ICO.0000000000001267. Laibson PR. Recurrence of corneal stro-
119. Yao YF, Zhang B, Zhou P, Jiang JK. Autologous mal dystrophies after penetrating kerato-
limbal grafting combined with deep lamellar plasty. Cornea. 2003;22(1):19–21. https://doi.
keratoplasty in unilateral eye with severe chemi- org/10.1097/00003226-200301000-00005.
cal or thermal burn at late stage. Ophthalmology. 132. Reddy JC, Murthy SI, Vaddavalli PK, Garg P,
2002;109(11):2011–7. https://doi.org/10.1016/ Ramappa M, Chaurasia S, et al. Clinical outcomes
s0161-6420(02)01258-7. and risk factors for graft failure after deep ante-
120. Fogla R, Padmanabhan P. Deep anterior lamel- rior lamellar keratoplasty and penetrating kera-
lar keratoplasty combined with autologous limbal toplasty for macular corneal dystrophy. Cornea.
stem cell transplantation in unilateral severe chemi- 2015;34(2):171–6. https://doi.org/10.1097/
cal injury. Cornea. 2005;24(4):421–5. https://doi. ICO.0000000000000327.
org/10.1097/01.ico.0000151550.51556.2d. 133. Sharma N, Jain M, Sehra SV, Maharana P, Agarwal
121. Inoue Y. Corneal triple procedure. Semin T, Satpathy G, et al. Outcomes of therapeutic pen-
Ophthalmol. 2001;16(3):113–8. https://doi. etrating keratoplasty from a tertiary eye care centre
org/10.1076/soph.16.3.113.4202. in northern India. Cornea. 2014;33(2):114–8. https://
122. Oie Y, Nishida K. Triple procedure: cataract extrac- doi.org/10.1097/ICO.0000000000000025.
tion, intraocular lens implantation, and corneal graft. 134. Chen WL, Wu CY, Hu FR, Wang IJ. Therapeutic
Curr Opin Ophthalmol. 2017;28(1):63–6. https:// penetrating keratoplasty for microbial keratitis
doi.org/10.1097/ICU.0000000000000337. in Taiwan from 1987 to 2001. Am J Ophthalmol.
123. Sarnicola E, Sarnicola C, Cheung AY, Sarnicola 2004;137(4):736–43. https://doi.org/10.1016/j.
V. Deep anterior lamellar keratoplasty for corneal ajo.2003.11.010.
penetrating wounds. Eur J Ophthalmol. 2021;17:28. 135. Kashiwabuchi RT, de Freitas D, Alvarenga LS, Vieira
https://doi.org/10.1177/11206721211014385. L, Contarini P, Sato E, et al. Corneal graft survival
124. Moshirfar M, Bennett P, Ronquillo Y. Corneal dys- after therapeutic keratoplasty for Acanthamoeba ker-
trophy. Treasure Island, FL: StatPearls Publishing; atitis. Acta Ophthalmol. 2008;86(6):666–9. https://
2021. doi.org/10.1111/j.1600-0420.2007.01086.x.
136. Dart JK, Saw VP, Kilvington S. Acanthamoeba 2010;29(12):1448–51. https://doi.org/10.1097/

keratitis: diagnosis and treatment update 2009. Am ICO.0b013e3181e2ef9c.
J Ophthalmol. 2009;148(4):487–99.e2. https://doi. 146. Gabison EE, Doan S, Catanese M, Chastang P, Ben
org/10.1016/j.ajo.2009.06.009. M’hamed M, Cochereau I. Modified deep anterior
137. Kaufman AR, Tu EY. Advances in the manage- lamellar keratoplasty in the management of small
ment of Acanthamoeba keratitis: a review of the and large epithelialized descemetoceles. Cornea.
literature and synthesized algorithmic approach. 2011;30(10):1179–82. https://doi.org/10.1097/
Ocul Surf. 2022;25:26–36. https://doi.org/10.1016/j. ICO.0b013e3182031c81.
jtos.2022.04.003. 147. Waring GO, Lynn MJ, McDonnell PJ. Results of the
138. Sarnicola E, Sarnicola C, Sabatino F, Tosi GM, prospective evaluation of radial keratotomy (PERK)
Perri P, Sarnicola V. Early Deep Anterior Lamellar study 10 years after surgery. Arch Ophthalmol.
Keratoplasty (DALK) for acanthamoeba kera- 1994;112(10):1298–308. https://doi.org/10.1001/arc
titis poorly responsive to medical treatment. hopht.1994.01090220048022.
Cornea. 2016;35(1):1–5. https://doi.org/10.1097/ 148. Einan-Lifshitz A, Belkin A, Sorkin N, Mednick
ICO.0000000000000681. Z, Boutin T, Kreimei M, et al. Evaluation of
139. Sarnicola V, Sarnicola E, Sarnicola C, Sabatino F, big bubble technique for deep anterior lamellar
Tosi GM, Perri P. Reply. Cornea. 2016;35(6):e14–5. keratoplasty in patients with radial keratotomy.
https://doi.org/10.1097/ICO.0000000000000832. Cornea. 2019;38(2):194–7. https://doi.org/10.1097/
140. Sabatino F, Sarnicola E, Sarnicola C, Tosi GM, Perri ICO.0000000000001811.
P, Sarnicola V, et al. Early deep anterior lamellar 149. Smadja D, Colin J, Krueger RR, Mello GR, Gallois
keratoplasty for fungal keratitis poorly responsive A, Mortemousque B, et al. Outcomes of deep ante-
to medical treatment. Eye. 2017;31(12):1639–46. rior lamellar keratoplasty for keratoconus: learning
https://doi.org/10.1038/eye.2017.228. curve and advantages of the big bubble technique.
141. Anshu A, Parthasarathy A, Mehta JS, Htoon HM, Cornea. 2012;31(8):859–63. https://doi.org/10.1097/
Tan DT. Outcomes of therapeutic deep lamel- ICO.0b013e318242fdae.
lar keratoplasty and penetrating keratoplasty for 150. Senoo T, Chiba K, Terada O, Hasegawa K, Obara
advanced infectious keratitis: a comparative study. Y. Visual acuity prognosis after anterior chamber
Ophthalmology. 2009;116(4):615–23. https://doi. air replacement to prevent pseudo-anterior cham-
org/10.1016/j.ophtha.2008.12.043. ber formation after deep lamellar keratoplasty.
142. Bagga B, Garg P, Joseph J, Mohamed A, Kalra Jpn J Ophthalmol. 2007;51(3):181–4. https://doi.
P. Outcome of therapeutic deep anterior lamel- org/10.1007/s10384-006-0421-2.
lar keratoplasty in advanced. Indian J Ophthalmol. 151. Huang OS, Htoon HM, Chan AM, Tan D, Mehta
2020;68(3):442–6. https://doi.org/10.4103/ijo. JS. Incidence and outcomes of intraoperative des-
IJO_307_19. cemet membrane perforations during deep ante-
143. Laurik KL, Szentmáry N, Daas L, Langenbucher A, rior lamellar keratoplasty. Am J Ophthalmol.
Seitz B. Early penetrating keratoplasty à chaud may 2019;199:9–18. https://doi.org/10.1016/j.
improve outcome in therapy-resistant acanthamoeba ajo.2018.10.026.
keratitis. Adv Ther. 2019;36(9):2528–40. https://doi. 152. Sarnicola E, Sarnicola C, Cheung AY, Panico E,
org/10.1007/s12325-019-01031-3. Panico C, Sarnicola V. Manual deep anterior lamellar
144. Ehlers N, Hjortdal J. Are cataract and iris atro- keratoplasty after partial unintentional full-thickness
phy toxic complications of medical treatment trephination. Eur J Ophthalmol. 2020;31:774.
of acanthamoeba keratitis? Acta Ophthalmol https://doi.org/10.1177/1120672120932108.
Scand. 2004;82(2):228–31. https://doi. 153. Sarnicola V, Sarnicola E, Sarnicola C. Recovery
org/10.1111/j.1600-0420.2004.00237.x. techniques in DALK. In: Mannis MJ, Holland EJ,
145. Sharma N, Kumar C, Mannan R, Titiyal JS, editors. Cornea, vol. 2. 4th ed. Amsterdam: Elsevier;
Vajpayee RB. Surgical technique of deep anterior 2016.
lamellar keratoplasty in descemetoceles. Cornea.
Stromal Lenticule Addition
Keratoplasty (SLAK)
19
Leonardo Mastropasqua, Niccolò Salgari,
Manuela Lanzini, and Mario Nubile

• The concept of “remodeling” the keratoconic
cornea, by intrapocket implantation of human Keratoconus is a non-inflammatory degenerative
stromal lenticules, is presented in this chapter. ectatic disorder of the cornea in which progres-
The main difference resides in the fact that sive stromal thinning and apex protrusion pro-
diseased stroma is not replaced by the kerato- duce corneal optical function impairment [1]. Up
plasty technique but “augmented” in order to to the advanced stage, the corneal stromal is
reshape the transparent ectatic cornea. transparent, and visual function is mainly
• The use of femtosecond lasers system for the impeded by irregular astigmatism [2].
preparation of both the recipient intracorneal Visual rehabilitation in keratoconus relies on
dissection and the “shaped” lenticule from spectacle lenses when ectasia is in its early stages
donor corneas is clearly presented in the chap- but, when protrusion evolves, contact lens is the
ter. The main clinical advantages reside in the only option to achieve acceptable vision quality.
quick and simple surgical procedure that can be When gas-permeable lenses fail on multiple
performed suture-less under topical anesthesia. attempts, surgery has to be considered [3]. At
• The results as well as the pros and cons of the present, keratoconus is the second indication for
technique are presented, with updated litera- keratoplasty worldwide. The first indication for
ture review and description of the tomographic, keratoplasty became graft replacement in the
refractive, and microscopic tissue changes 2019 USA Eye-Bank report, which demonstrates
occurring after SLAK. Finally, future perspec- the increasing demand of tissues for keratoplasty
tives and field of improvement are described. repetition. Keratoconus usually develops in the
second decade of life [4] and this, combined with
limited graft survival and disease recurrence,
Supplementary Information The online version con- could be one of the reasons for the increasing
incidence of re-grafting in recent years [5].
org/10.1007/978-3-031-32408-6_19.
Deep anterior lamellar keratoplasty (DALK)
is considered the gold standard procedure for the
L. Mastropasqua · M. Lanzini · M. Nubile surgical treatment of keratoconus thanks to the
National High-Tech Eye Center, University of Chieti lower risk of rejection, increased postoperative
and Pescara, Chieti, Italy
e-mail: [email protected]; [email protected] biomechanical corneal strength and longer graft
survival compared to penetrating keratoplasty
N. Salgari (*)
Department of Translational Medicine, University of (PK) [6, 7]. DALK is a challenging procedure
Ferrara, Ferrara, Italy that may require conversion to PK if descemet-
https://doi.org/10.1007/978-3-031-32408-6_19
284 L. Mastropasqua et al.
endothelial rupture occurs [6, 8]. Unfortunately, procedure was complex at first and after lenticule
visual outcomes can be threatened by severe preparation with cryolathe, it involved lamellar
intraoperative (e.g. endophthalmitis, choroidal cap dissection of the recipient cornea, donor len-
hemorrhage) along with postoperative (e.g. rejec- ticule implantation on keratectomy bed, and then
tion, cataract, glaucoma, wound dehiscence, and repositioning and suturing of the recipient cap
stromal melting) complications that could also [21, 22]. Parallel to this work, a similar approach
require lifelong medical treatment or further sur- was proposed by Kaufman based on suturing the
geries [9]. Despite successful surgery, over a donor lenticule to the anterior corneal surface
third of patients should continue the use of con- after epithelium removal in a procedure called
tact lenses due to astigmatism or anisometropia “epikeratophakia” [23, 24].
after keratoplasty [10]. Dissection depth of early microkeratomes was
Attempts to modify corneal shape in order to quite variable and inadvertent perforation was
avoid keratoplasty have been made in recent even possible. Lenticule customization was based
years with variable results. Photorefractive kera- on mechanical carving and cutting. Wound dehis-
tectomy (PTK) and intracorneal ring segments cence of the lamellar cap was possible with risk
(ICRS) implantation combined with corneal col- of infections, epithelial ingrowth, and irregular
lagen cross-linking (CXL) is available proce- astigmatism development. Visual recovery was
dures but not widely adopted due to the difficult generally poor due to structural cryodamages to
prediction of visual results and application lim- the lenticules during grinding procedure. In a
ited to moderate cases [11, 12]. group of 32 patients, Troutman reported CDVA
>20/40 in 84% of patients and a target refraction
within 1.00 D in only 38.5% [19]. Similarly,
History of Stromal Keratophakia Swinger and Barraquer achieved CDVA >20/40 in
only 46.4% of their 46 patients [16]. Refractive
The optical function of the cornea relies on the results were far away from the desired target with
high regularity of its central area and small a mean postoperative astigmatism of
changes of this portion produce great modifica- 2.45 ± 0.37 D [25]. Given the increasing accu-
tion of its dioptric power [13, 14]. Attempts to racy and reproducibility of the laser subtractive
modify corneal shape by incision have been made techniques in the following years, keratophakia
since the 1800s but it was in the 1950s when a and epikeratophakia were soon abandoned.
novel concept introduced by José Ignacio
Barraquer led to the beginning of the refractive
surgery era [15–17]. Amid the first experiments ecent Advancements in Corneal
R
in this exciting surgery field, Barraquer proposed Surgery
two approaches to reshape corneal curvature by
means of tissue addition or subtraction [16, 18]. A renewed interest in stromal keratophakia began
The latter is now the cornerstone of all refractive in 2011 when, following the introduction of fem-
surgery procedures, while the first has been rap- tosecond laser technology in the refractive sur-
idly abandoned due to poor visual results mainly gery field, a novel group of techniques capable of
due to technological limitations [19, 20]. producing precise and optically functioning
The original idea of keratophakia (from Greek stromal lenticules was presented: the Refractive
words “kerato” for cornea and “phakia” for Lenticule Extraction (ReLEx) techniques [26].
“lens”) was to produce corneal curvature changes The small incision refractive lenticule extraction
by the addition of stromal tissue into the recipient (SMILE) became a valid alternative to the
stromal bed. The Barraquer’s “thickness law” excimer laser-based techniques available since
states that corneal flattening can be obtained by the 1980s [27].
tissue subtraction from the center or tissue addi- In 2011, Mohamed-Noriega et al. proved that
tion in the periphery and vice-versa [16]. The fresh extracted tissues after ReLEx are vital and
19 Stromal Lenticule Addition Keratoplasty (SLAK) 285
their vitality and optical efficiency were main- decided to rely on a spherical equivalent as
tained after cryopreservation [28]. Angunawela matching parameter. Cryopreserved lenticules
et al. showed that lenticules could be safely reim- were thawed and washed, soaked with riboflavin,
planted to reverse the refractive procedure in a and then underwent a central 3-mm trephination
rabbit model [29], and Riau et al. showed that this to obtain a donut-like shape. Intrastromal pocket
procedure was feasible also in nonhuman primate was created into the recipient cornea at 100-μm
model [30]. depth with 7.0- to 8.0-mm diameter (1 mm larger
Following these early insights, many authors than the donor lenticule) and a 4-mm superior
showed interest in this field and numerous exper- incision. The procedure was combined with
iments took place. Liu et al. [31] investigated accelerated collagen cross-linking. Improvement
autologous lenticule implantation in rabbit cor- was reported for uncorrected distance visual acu-
nea the same year of Angunawela and later, in ity (1.06 ± 0.48 to 0.38 ± 0.27 logMAR) and cor-
2015, Zhang et al. and Liu et al. showed the rected distance visual acuity (0.51 ± 0.20 to
results of allogenic lenticule implantation in rab- 0.20 ± 0.24 logMAR), and manifest spherical
bit and monkey animal models [32, 33]. The first equivalent (23.47 ± 1.15 D to 21.77 ± 1.7 D).
human case of FSL-derived allogenic lenticule Mean keratometry in 3-mm and 5-mm zones
implantation was reported by Pradhan et al. in reduced by 3.42 ± 2.09 D and 1.70 ± 1.31 D,
2013 who implanted a convex lenticule produced respectively. Mean pachymetry in the central and
by myopic SMILE of −10.50 D (127 μm central mid peripheral zones increased by 18.3 ± 7.3 mm
thickness) with 5.75 mm diameter in a lamellar and 33.0 ± 8.8 mm, respectively. All eyes had
pocket at 110 μm depth to treat a high positive reduction in higher order aberrations, specifically
refractive error due to aphakia [34]. In fact, coma with no loss of lines of corrected distance
thanks to the capability of increasing corneal cur- visual acuity. No adverse events were reported
vature with convex positive meniscus-shaped [39].
lenticule implantation, the first attempted appli- Nearly contemporary to the first lenticule
cations of this concept were in hyperopia and implantation experiments performed by many
presbyopia treatment [35–37]. authors for refractive error correction,
Mastropasqua et al. hypothesized the possibility
of treating corneal ectasia making advantage of
reatment of Keratoconus by Stromal
T the tissue addition concept [40].
Lenticule Addition
Therapeutic use of SMILE-derived lenticules is tromal Lenticule Addition

S
another possible application of this innovative Keratoplasty (SLAK)
concept of keratoplasty. Ganesh et al. proposed in
2015 the implantation of modified lenticules Taking cues from the results of early experiments
combined with corneal collagen cross-linking to with positive meniscus-shaped lenticule,
treat keratoconus in a modification of the original Matropasqua et al. hypothesized that a lenticule
technique he proposed in 2014 named femtosec- with a negative profile can induce opposite
ond intrastromal lenticular implantation (FILI) changes to corneal curvature. This concept is part
[38, 39]. of the knowledge we learned from intracorneal
Inclusion criteria were grade 1–3 keratoconus ring segment implantation (ICRS), despite the
with or without documented evidence of progres- working principle of ICRS is not fully under-
sion. The lenticule diameter was 6.0–7.0 mm stood yet. According to this hypothesis and the
with side-cut angle of 90°. The authors tried to ongoing development of a novel negative
calculate based on corneal topography the best meniscus-shaped lenticule model for application
match with available donor lenticules, but calcu- of SMILE in hyperopia treatment [41], we pro-
lations were too complex and not reliable, so they posed the use of that lenticule geometry to obtain
the desired effect in a novel procedure of kerato- All the surgical procedure is performed under
plasty [40]. The negative meniscus lenticule pres- topical anesthesia. Corneal epithelium is removed
ents a concave profile with a maximum thickness using a blunt spatula, and a modified femtosec-
in the peripheral region of the optical zone that ond laser flap-cut procedure (flocket) [43] is per-
gradually reduces toward the thinnest point formed to fashion an intrastromal pocket with a
located in the center. 500-kHz VisuMax femtosecond laser (Carl Zeiss
The first ex vivo preliminary SLAK study on Meditec, Jena, Germany) centered on the corneal
normal cornea was designed in 2015 to under- central zone (Figs. 19.1a and 19.2a). A hinge
stand the effect of negative meniscus lenticule length of 21.7 mm is set in order to produce a
addition [40]. Following the promising results circular planar plane of dissection with a single
experienced ex vivo, a human in vivo noncom- 4-mm superior opening, a diameter of 8.2 mm,
parative interventional case series was performed and depth from surface of 120 μm. After the laser
in a group of subjects affected by advanced cen- phase, the patient is moved to the surgical micro-
tral keratoconus who were already candidates for scope and flocket dissected by means of a blunt
keratoplasty in 2018 [42]. Treatment was reserved dissector (Figs. 19.1b and 19.2b) [40, 42].
to central keratoconus because only symmetrical Donor lenticule is produced by means of fem-
lenticule geometry was available. tosecond lenticule extraction procedure (FLEx)
with VisuMax femtosecond laser platform. The
Surgical Technique lenticule parameters are set as follows: flap thick-
The first phase of the surgical procedure consists ness 110 μm, flap diameter 8 mm, hyperopic cor-
of lenticule preparation in donor cornea by means rection 8.00 D, optical zone 6 mm. The maximum
of a femtosecond laser (FSL). Corneo-scleral but- lenticule thickness obtained is 148 μm with a
tons are mounted on an artificial anterior chamber central lenticule minimal thickness of 30 μm. The
and pressure is standardized with a BSS bottle at overall transition zone in the periphery of the len-
180 cm of height to inflate the anterior chamber. ticules is 0.70 mm in diameter [40, 42].
a b
c d
Fig. 19.1 Illustration of the surgical technique. inserted through the incision and spread out with a dedi-
Intrastromal pocket is created in the recipient cornea by cated forceps (c). Lenticule in its final position produces
means of femtosecond laser (a). Stromal dissection is per- peripheral thickening and central flattening (d)
formed with a blunt spatula (b) and then the lenticule is
a b
c d
Fig. 19.2 Surgical phases of stromal lenticule addition Stromal lenticule is implanted through the small incision
keratoplasty. Stromal pocket is created by means of a fem- (c) and spread out in a single maneuver with a dedicated
tosecond laser in the recipient cornea (a) and plane dissec- forceps (d)
tion is manually performed with a blunt spatula (b).
After completion of the FSL dissection phase, the operator using an intraoperative microscope,
the flap is lifted by means of a blunt spatula under AS-OCT, and topography [40, 42].
the surgical microscope and then the lenticule is
separated and transferred to the patient’s eye Results
maintaining its original orientation. The incision The results of the first human in vivo noncom-
of the recipient pocket is opened with a Seibel parative interventional case series of SLAK
spatula and the plane dissected with a blunt dis- proved that corneal flattening is possible by
section spatula (Mastropasqua SMILE kit, selective tissue addition (Fig. 19.3). Despite lack
Janach, Como, Italy). The lenticule is placed on of customization, the addition of a standardized
the patient’s cornea close to the incision opening negative meniscus lenticule centered on the cone
and while grasping its distal edge, the lenticule is apex produced a significant improvement of cor-
dragged into the pocket through the incision and neal topography along with best corrected visual
spread out (Figs. 19.1c, d and 19.2c, d). Final dis- acuity in our first case series in 2015 [42].
tention is achieved from the surface using the Uncorrected distance visual acuity (UDVA) at 6
spatula (Video 19.1). The lenticule is carefully months improved in 8 of 10 eyes (p < 0.01;
centered onto the apex of the cone and correct Fisher’s exact test), whereas all but one eye had
distension and centration should be assessed by improvement of corrected distance visual acuity
a b
Fig. 19.3 Recipient cornea before (a) and 1 month after SLAK (b). Lenticule inside the stroma was barely noticeable
on slit light direct illumination but clearly visible on retro-illumination (c)
(CDVA). UDVA improved from 1.58 ± 0.36 to corneal asphericity (Q value) was reduced indi-
1.22 ± 0.37 logMAR (p = 0.024; Wilcoxon cating a reduction of corneal irregular high pro-
signed-rank test), whereas CDVA improved from lacity (p < 0.05; Wilcoxon signed-rank test) [42].
1.07 ± 0.17 to 0.70 ± 0.23 logMAR (p = 0.007). Corneal topography after SLAK shows an
One eye gained three lines, three eyes gained two area of central curvature reduction surrounded by
lines, five eyes gained one line, and one eye had a red ring of increased curvature corresponding
no change in lines of CDVA. Spherical equivalent to the transition zone from the addition area to
significantly reduced from −7.46 ± 2.49 D to the peripheral recipient cornea. Anterior topo-
−3.61 D ± 1.99 D at 6 months (p = 0.021). graphic flattening appears similar on color-
graded map to corneal flattening pattern
Corneal Topography experienced after conventional myopic refractive
On corneal topography, a generalized flattening laser treatment [42].
of the cone can be observed after SLAK
(Fig. 19.4). We reported a mean anterior kera- I n Vivo Confocal Microscopy
tometry value reduction from 58.69 ± 3.59 to Intrastromal FSL-based refractive surgery proce-
53.59 ± 3.50 diopters at 6 months after surgery in dures proved to induce a lower level of inflamma-
the first in vivo study (p < 0.05; Wilcoxon signed- tion and apoptosis combined with a better
rank test) along with negligible variation of the preservation of anterior stromal lamellae struc-
posterior corneal mean keratometry. Anterior ture and nerve plexus compared to traditional
a b
c d
Fig. 19.4 Anterior segment OCT and tangential corneal cone flattening effect is appreciable on the anterior cor-
curvature map before (a and b, respectively) and 6 months neal curvature map (d) surrounded by a red ring of curva-
(c and d, respectively) after SLAK. Lenticule profile is ture transition zone
clearly visible inside the recipient cornea (c), while the
ablative techniques [44–49]. Interfaces produced of some cellular debris consisting of roundish
by FSL-cut appear optically clear and highly reg- appearing elements with small diameter.
ular [50]. Thanks to the low energy delivered into Interfaces remained visible up to the end of the
the recipient stroma, a low incidence of corneal follow-up of 12 months on IVCM but with
haze was reported after ReLEx treatments. decreasing reflectivity. Interface appearance was
Preservation of anterior lamellar structure proved similar to what was previously documented after
to be relevant to corneal biomechanics, while refractive lenticule extraction procedure.
subepithelial nerve sparing induces faster recov- No significant endothelial or keratocyte cell
ery of corneal sensitivity and epithelial integrity density changes nor signs of rejection were
[46, 49, 51, 52]. reported following SLAK. Lenticule microfolds
On IVCM, corneal epithelium appeared regu- were variably visible in some cases on IVCM but
larly stratified after SLAK all over the follow-up never documented on slit lamp examination [53].
[53]. Mean subbasal nerve density was restored
to preoperative values at 3 months after surgery, pithelial and Stromal Remodeling
E
and no significant variations of dendritic immune Remodeling of corneal structure is a progressive
cell density were reported. Mild anterior stromal phenomenon that requires time after surgery [54–
edema was present in all cases but rapidly sub- 58]. Epithelial compensation is a documented
sided during the first month. At 6 months postop- process that can affect final corneal curvature and
eratively, no sign of stromal reaction or keratocyte induce refractive changes after corneal surgery
activation was visible. Interface reflectivity sig- and laser treatments [59, 60]. Epithelial thickness
nificantly reduced at 3 months with persistence map 1 month after SLAK showed a central thick-
ness increase and a mid-peripheral thickness Nearly contemporary to Ganesh and Brar in
reduction corresponding to the zone of increased 2014 [39], we hypothesized the possibility of
curvature [61]. Epithelial thickness outside the making use of this principle to treat keratoconus
addition area experienced a progressive increase in 2015.
up to 6 months after surgery. No significant Ganesh and Brar proposed the implantation of
changes were reported on optical coherence donut-shaped stromal lenticules, obtained by a
tomographic analysis regarding lenticule thick- 3-mm central punching of cryopreserved myopic
ness profile and anterior or posterior recipient lenticules combined with collagen cross-linking,
stromal thicknesses throughout the follow-up in keratoconus [39]. The result of the addition of
[61]. stromal tissue in the mid-periphery and around
the cone caused a relative flattening in the center
and a reduction of hyper-prolate shape [39]. The
Discussion theoretical mechanism of action of this technique
is thought to be partially similar to the intracor-
Technical limitations of early stromal keratopha- neal ring segments (ICRS) implantation because
kia procedures (e.g. coarse mechanical dissection both techniques involve addition of volume and
of stromal bed, freezing-chiseling and subse- local elevation in the mid-periphery [64].
quent thawing of tissues, and flap suturing) were Topographical analysis after SLAK showed
linked to poor visual results [21, 22]. that all eyes had a detectable reduction of central
The introduction of femtosecond laser refrac- anterior corneal curvature, indicating a signifi-
tive lenticule extraction not only brought advan- cant relative flattening of the cone, with negligi-
tages in the refractive surgery field but also ble effects on the posterior corneal curvature
allowed novel surgical approaches to corneal (Fig. 19.5). Vision rehabilitation with current
pathologies. Stromal lenticules produced by FSL technique is limited by the lack of a customized
have accurate geometry, smooth surfaces, pre- lenticule profile. Lenticule dioptric power is
served vitality, and undamaged collagen structure technically limited to maximum +8.00 D with a
[28, 50]. Moreover, FSL can be used to fashion maximum peripheral thickness of +148 μm.
intra-stromal dissection planes in the recipient Despite the enrolled cases in the first series were
cornea minimally affecting biomechanics of advanced keratoconus with curvature of
anterior stromal layers [49]. The possibility of 58.69 ± 3.59 D, mean UDVA and CDVA signifi-
transplanting refractive lenticules was validated cantly increased: 9 of 10 eyes showed an increase
in studies conducted with animal models at first in CDVA that ranged from one to three Snellen
[29, 30, 32, 33], then stromal transplantation of lines [42].
refractive lenticules derived from myopic SMILE Interestingly, lenticule dimensions analyzed
has been proposed as a possible tissue addition by means of AS-OCT showed minimal but con-
approach to successfully steepen the cornea for sistent differences in the central lenticule thick-
treating hyperopia and aphakia [34, 35, 62, 63]. ness; on average the central lenticule thickness
Fig. 19.5 Illustration of

corneal curvature and
thickness changes
following SLAK. The
red arrows indicate
forward displacement of
the anterior surface,
while the blue arrow
indicates central
flattening. Dashed lines
represent the original
keratoconus profile
was measured to be 47 μm compared to a pro- thelial breakdown or keratoconic “pips” observed

grammed thickness of 30 μm [42, 61]. Despite in patients with thin epithelium [66].
stromal edema rapidly regressed over the first Stress redistribution and biomechanical
month, this increase was documented up to the 6 changes in lenticule addition procedures are pro-
months. Ganesh and Brar reported an increase cesses that should be evaluated in future study.
in CCT after implantation of donut-shaped We can hypothesize that Bowman Membrane
myopic lenticules, although there was no addi- (BM) could affect predictability of curvature
tion of stroma in the central area in their proce- changes after SLAK, while on the other hand the
dure [39]. They proposed that this might be procedure could affect disease progression
related to a mild lifting of the anterior corneal changing shear stress distribution on BM. In
layers and creation of a potential space in the SLAK, a significant amount of tissue is added to
center after peripheral tissue addition. They also the conus apex thus improving the corneal thin-
observed a trend toward a decrease in CCT at ning of keratoconus. Increasing thickness may be
6 months after surgery [39]. Stromal wound- not enough to halt disease progression on itself
healing process can be accounted as a possible since donor tissue integration with recipient
explanation of this phenomena in our opinion. stroma remains to be demonstrated. The contact
New extracellular matrix (ECM) may be pro- alongside the interfaces between host and donor,
duced in response to keratocyte activation and probably create at least a minimum improvement
less volume addition in central cornea concomi- to shear stress resistance but, as already proved in
tantly to a mechanical stress displacement in LASIK, tissues mechanically separated never
mid-periphery, where the lamellae become integrate completely and can be easily separated
compacted by volume addition, may be the even after years. The improved wound healing
cause of increased ECM apposition in central produced by FSL lamellar cut may positively
area. Further histological study must be per- affect this integration, as proved in corneal treph-
formed to clear this hypothesis. ination [67]. ICRS implantation does not stop the
Biomechanical redistribution of tensile disease progression and central distention of col-
strength on collagen lamellae and viscosity of lagen fibers, despite inducing flattening, theoreti-
ECM could be responsible for the slow changes cally may produce more stress on conus apex [68,
of anterior and posterior surfaces observed dur- 69]. On the contrary, the addition of tissue around
ing the first month, as already reported in other the cone base produced by SLAK may reduce the
refractive surgery procedures [65]. apex fiber stress displacing tensile stress to the
Epithelial remodeling participates in the over- peripheral intact Bowman membrane, thus affect-
all corneal thickness increase after ing the progression process and apical scarring in
SLAK. Epithelial map produced by ultrahigh- some way. These considerations need to be evalu-
resolution OCT documented that cone flattening ated in further long-term in vivo studies also
was associated with significantly increased epi- including progressive keratoconus and less
thelial thickness increase at 1 month after sur- advanced stages.
gery, stabilizing at 3-months. Observed epithelial The in vivo confocal microscopy (IVCM)
thickness values were close to normality after examination did not show a significant modifica-
1 month with an increase of about 15%. Epithelial tion of dendritic cell population after SLAK, and
remodeling occurred also in the mid peripheral patients did not experience any type of corneal
and peripheral regions, with a slight decrease in rejection during the follow-up [53]. The presence
the first and increase in the second consistent of subbasal dendritic cells has been documented
with anterior corneal profile variations [61]. This after traditional keratoplasty regardless of the
phenomenon should be taken into account in presence of a clinically evident rejection. These
future study of predictability. On the other side, cells are involved in antigen presentation and are
corneal epithelial restoration can also improve largely recruited in corneal inflammation [70].
contact lens tolerance and reduce the risk for epi- This process can have a role in stimulating the
immune reaction and onset of a later graft failure uture Perspectives: Lenticule
F
[71–73]. DALK is the current gold standard treat- Shape Customization
ment for advanced keratoconus because one of
the main advantages of this lamellar technique The main current limit of SLAK is that only eyes
over penetrating keratoplasty is the reduced affected by central keratoconus are eligible for
rejection rate with longer graft survival [74]. the treatment because only symmetrical lenticule
Although the observation period of our IVCM geometry is available.
study was limited, we suppose that, given the Ongoing studies are trying to figure out the
reduced amount of transplanted tissue and the proper way to obtain lenticules suitable for
isolation of the donor tissue inside the stromal implantation in corneas with eccentric keratoco-
pocket, the immunological stimulation in SLAK nus, which represent more than 70% of cases.
might be at least similar if not lower with respect We recently conducted an ex vivo study to
to the DALK one. assess the effects of different types of lenticule
Corneal reinnervation after keratoplasty is customizations. Human eye bank donor corneas
mainly due to subepithelial fiber proliferation not suitable for transplantation for low endothe-
rather than stromal reinnervation [75]. In vivo lial cell density were used to realize keratoconus
confocal microscopy studies revealed that lamel- models (recipients) and customized lenticules
lar cuts produced by SMILE surgery only par- (donors). In addition recent findings showed that
tially affect subbasal nerve density and nerve lenticule customization is possible by the aid for
fibers are almost recovered by 6 months after excimer laser photoablation of the FSL-prepared
surgery. Subbasal fibers are resected only at the donor lenticules [83]. This will allow to increase
level of the incision, and the stromal fibers are the precision of corneal reshaping in the various
resected only at the trespassing of the cap cut forms and stages of keratoconic eyes in treated
[46, 47, 76]. In SLAK, we observed a similar patients and to expand the indications of intra-
reduction of density shortly after surgery with stromal tissue implantation.
fiber degeneration within the first few months The encouraging results bode well for upcom-
and a recovery of the original plexus at ing advancements in corneal additive surgery.
3–6 months after surgery [53]. In conclusion, our investigations demonstrated
The regularity of the interface between donor that stromal lenticule in addition to keratoplasty is
and recipient stroma is the main factor affecting a feasible and effective technique for stromal
visual recovery in lamellar procedures [77–80]. remodeling that improves vision and corneal reg-
The femtosecond laser produces highly regular ularity in central keratoconus. Implanted tissues
stromal cuts that appear as hyperreflective planes were accepted by host stroma without rejection
in the corneal stroma [50]. Despite the interface nor opacification. Corneal remodeling should be
reflectivity, SMILE grants excellent visual out- further investigated to develop customization
comes but the degree of interface roughness can algorithms and expand indications to mild kerato-
affect the quality of vision [81]. Similarly to conus cases where this technique might ideally
SMILE [82], in SLAK, we documented moder- offer better results. Currently, no treatment can
ately reflective interfaces with the presence of outperform the results of DALK that is still the
particles in the first month that gradually reduced gold standard in keratoconus surgical treatment,
over time. Cellular and matrix debris on both the in particular for advanced cases, but SLAK
anterior and posterior interfaces significantly opened new perspectives on future developments
reduced over time but persisted until 12 months. of customized, minimally invasive, additive treat-
The presence of a double interface may affect ments for keratoconus. It may become a valuable
visual rehabilitation after SLAK but, even though option for treating those cases that have no indica-
the anterior and posterior interfaces were close to tion for traditional keratoplasty but may benefit
each other in the central area, we did not observe from a minimally invasive procedure capable of
any opacification of the corneal stroma. reshaping the corneal geometry.
Take Home Notes plasty and penetrating keratoplasty in keratoconus.

Am J Ophthalmol. 2009;148(5):744–751.e1.
• Femtosecond laser lenticule extraction proce- 7. Yu AC, Franco E, Caruso L, Myerscough J, Spena R,
dures can be used to produce transparent stro- Fusco F, et al. Ten-year outcomes of microkeratome-
mal lenticules in donor cornea suitable for assisted lamellar keratoplasty for keratoconus. Br J
additive corneal transplantation techniques. Ophthalmol. 2020;105:1651.
8. Myerscough J, Roberts H, Yu AC, Elkadim M,
• Intrastromal lenticule implantation can effec- Bovone C, Busin M. Five-year outcomes of converted
tively and predictably change curvature and mushroom keratoplasty from intended Deep Anterior
thickness of the recipient cornea according to Lamellar Keratoplasty (DALK) mandate 9-mm diam-
the lenticule design. eter DALK as the optimal approach to keratoconus.
Am J Ophthalmol. 2020;220:9–18.
• Flattening of a central keratoconus is possible 9. Liu H, Chen Y, Wang P, Li B, Wang W, Su Y, et al.
by means of negative meniscus-shaped lenti- Efficacy and safety of deep anterior lamellar kerato-
cule addition in SLAK, leading to an improve- plasty vs. penetrating keratoplasty for keratoconus: a
ment of visual acuity, thickness, and corneal meta-analysis. PLoS One. 2015;10(1):e0113332.
10. Smiddy WE, Hamburg TR, Kracher GP, Stark
curvature. WJ. Keratoconus. Contact lens or keratoplasty?
• The SLAK procedure induces mild and tran- Ophthalmology. 1988;95(4):487–92.
sient inflammation rapidly recovering during 11. Arnalich-Montiel F, Alió Del Barrio JL, Alió
the first postoperative days and can be per- JL. Corneal surgery in keratoconus: which type,
which technique, which outcomes? Eye Vis. 2016;3:2.
formed under topical anesthesia. 12. Mastropasqua L. Collagen cross-linking: when and
how? A review of the state of the art of the technique
In the future, customization of lenticule shape and new perspectives. Eye Vis. 2015;2:19.
according to corneal topography could optimize 13. Doss JD, Hutson RL, Rowsey JJ, Brown DR. Method
for calculation of corneal profile and power distribu-
the curvature regularization and make possible to tion. Arch Ophthalmol. 1981;99(7):1261–5.
treat decentered keratoconus. Lenticule harvest- 14. Olsen T. On the calculation of power from curva-
ing from living donors undergoing refractive sur- ture of the cornea. Br J Ophthalmol. 1986;70(2):
gery along with eye-banking customization and 152–4.
15. Nordan LT. Barraquer lecture. José Barraquer: father
storage could help standardizing and ease the of modern refractive keratoplasty. Refract Corneal
procedure for the surgeon. Surg. 1989;5(3):177–8.
16. Swinger CA, Barraquer JI. Keratophakia and
keratomileusis--clinical results. Ophthalmology.
1981;88(8):709–15.
References 17. Barraquer JI. Keratomileusis. Int Surg.
1967;48(2):103–17.
1. Gomes JAP, Tan D, Rapuano CJ, Belin MW, Ambrósio 18. Ainslie D. The surgical correction of refractive errors
R, Guell JL, et al. Global consensus on keratoconus by keratomileusis and keratophakia. Ann Ophthalmol.
and ectatic diseases. Cornea. 2015;34(4):359–69. 1976;8(3):349–67.
2. Belin MW, Duncan JK. Keratoconus: the ABCD 19. Troutman RC, Swinger C, Goldstein M. Keratophakia
grading system. Klin Monatsbl Augenheilkd. update. Ophthalmology. 1981;88(1):36–8.
2016;233(6):701–7. 20. Friedlander MH, Safir A, McDonald MB,
3. Parker JS, van Dijk K, Melles GRJ. Treatment options Kaufman HE, Granet N. Update on keratophakia.
for advanced keratoconus: a review. Surv Ophthalmol. Ophthalmology. 1983;90(4):365–8.
2015;60(5):459–80. 21. Friedlander MH, Rich LF, Werblin TP, Kaufman HE,
4. Olivares Jiménez JL, Guerrero Jurado JC, Bermudez Granet N. Keratophakia using preserved lenticules.
Rodriguez FJ, Serrano Laborda D. Keratoconus: Ophthalmology. 1980;87(7):687–92.
age of onset and natural history. Optom Vis Sci. 22. Maguen E, Pinhas S, Verity SM, Nesburn
1997;74(3):147–51. AB. Keratophakia with lyophilized cornea lathed at
5. Yoshida J, Murata H, Miyai T, Shirakawa R, Toyono room temperature: new techniques and experimental
T, Yamagami S, et al. Characteristics and risk fac- surgical results. Ophthalmic Surg. 1983;14(9):759–62.
tors of recurrent keratoconus over the long term after 23. Kaufman HE. The correction of aphakia. XXXVI
penetrating keratoplasty. Graefes Arch Clin Exp Edward Jackson Memorial Lecture. Am J Ophthalmol.
Ophthalmol. 2018;256(12):2377–83. 1980;89(1):1–10.
6. Han DCY, Mehta JS, Por YM, Htoon HM, Tan 24. Busin M, Cusumano A, Spitznas M. Epithelial inter-
DTH. Comparison of outcomes of lamellar kerato- face cysts after epikeratophakia. Ophthalmology.
1993;100(8):1225–9.
25. Riau AK, Liu Y-C, Yam GHF, Mehta JS. Stromal ker- opia derived from small incision lenticule extraction
atophakia: corneal inlay implantation. Prog Retin Eye (SMILE). Sci Rep. 2018;8(1):1831.
Res. 2020;75:100780. 38. Ganesh S, Brar S, Rao PA. Cryopreservation of
26. Sekundo W, Kunert K, Russmann C, Gille A, extracted corneal lenticules after small incision len-
Bissmann W, Stobrawa G, et al. First efficacy and ticule extraction for potential use in human subjects.
safety study of femtosecond lenticule extraction for Cornea. 2014;33(12):1355–62.
the correction of myopia: six-month results. J Cataract 39. Ganesh S, Brar S. Femtosecond intrastromal lenticu-
Refract Surg. 2008;34(9):1513–20. lar implantation combined with accelerated collagen
27. Sekundo W, Kunert KS, Blum M. Small incision cross-linking for the treatment of keratoconus--ini-
corneal refractive surgery using the small incision tial clinical result in 6 eyes. Cornea. 2015;34(10):
lenticule extraction (SMILE) procedure for the cor- 1331–9.
rection of myopia and myopic astigmatism: results 40. Mastropasqua L, Nubile M. Corneal thickening
of a 6 month prospective study. Br J Ophthalmol. and central flattening induced by femtosecond laser
2011;95(3):335–9. hyperopic-shaped intrastromal lenticule implantation.
28. Mohamed-Noriega K, Toh K-P, Poh R, Balehosur Int Ophthalmol. 2017;37(4):893–904.
D, Riau A, Htoon HM, et al. Cornea lenticule viabil- 41. Liu Y-C, Ang HP, Teo EPW, Lwin NC, Yam GHF,
ity and structural integrity after refractive lenticule Mehta JS. Wound healing profiles of hyperopic-
extraction (ReLEx) and cryopreservation. Mol Vis. small incision lenticule extraction (SMILE). Sci Rep.
2011;17:3437–49. 2016;6:29802.
29. Angunawela RI, Riau AK, Chaurasia SS, Tan DT, 42. Mastropasqua L, Nubile M, Salgari N, Mastropasqua
Mehta JS. Refractive lenticule re-implantation after R. Femtosecond laser-assisted stromal lenticule
myopic ReLEx: a feasibility study of stromal restora- addition keratoplasty for the treatment of advanced
tion after refractive surgery in a rabbit model. Invest keratoconus: a preliminary study. J Refract Surg.
Ophthalmol Vis Sci. 2012;53(8):4975–85. 2018;34(1):36–44.
30. Riau AK, Angunawela RI, Chaurasia SS, Lee WS, Tan 43. Konstantopoulos A, Liu Y-C, Teo EPW, Lwin NC,
DT, Mehta JS. Reversible femtosecond laser-assisted Yam GHF, Mehta JS. Early wound healing and refrac-
myopia correction: a non-human primate study of len- tive response of different pocket configurations fol-
ticule re-implantation after refractive lenticule extrac- lowing presbyopic inlay implantation. PLoS One.
tion. PLoS One. 2013;8(6):e67058. 2017;12(2):e0172014.
31. Liu H, Zhu W, Jiang AC, Sprecher AJ, Zhou 44. Slade SG. Applications for the femtosecond
X. Femtosecond laser lenticule transplantation in laser in corneal surgery. Curr Opin Ophthalmol.
rabbit cornea: experimental study. J Refract Surg. 2007;18(4):338–41.
2012;28(12):907–11. 45. Reinstein DZ, Archer TJ, Gobbe M. Small incision
32. Zhang T, Sun Y, Liu M, Zhou Y, Wang D, Chen Y, lenticule extraction (SMILE) history, fundamentals of
et al. Femtosecond laser-assisted endokeratophakia a new refractive surgery technique and clinical out-
using allogeneic corneal lenticule in a rabbit model. J comes. Eye Vis. 2014;1:3.
Refract Surg. 2015;31(11):775–82. 46. Vestergaard AH, Grønbech KT, Grauslund J, Ivarsen
33. Liu R, Zhao J, Xu Y, Li M, Niu L, Liu H, et al. AR, Hjortdal JØ. Subbasal nerve morphology, corneal
Femtosecond laser-assisted corneal small inci- sensation, and tear film evaluation after refractive
sion allogenic intrastromal lenticule implantation in femtosecond laser lenticule extraction. Graefes Arch
monkeys: a pilot study. Invest Ophthalmol Vis Sci. Clin Exp Ophthalmol. 2013;251(11):2591–600.
2015;56(6):3715–20. 47. Li M, Niu L, Qin B, Zhou Z, Ni K, Le Q, et al.
34. Pradhan KR, Reinstein DZ, Carp GI, Archer TJ, Confocal comparison of corneal reinnervation after
Gobbe M, Gurung R. Femtosecond laser-assisted small incision lenticule extraction (SMILE) and fem-
keyhole endokeratophakia: correction of hyperopia tosecond laser in situ keratomileusis (FS-LASIK).
by implantation of an allogeneic lenticule obtained PLoS One. 2013;8(12):e81435.
by SMILE from a myopic donor. J Refract Surg. 48. Mohamed-Noriega K, Riau AK, Lwin NC, Chaurasia
2013;29(11):777–82. SS, Tan DT, Mehta JS. Early corneal nerve damage and
35. Williams GP, Wu B, Liu YC, Teo E, Nyein CL, Peh recovery following small incision lenticule extraction
G, et al. Hyperopic refractive correction by LASIK, (SMILE) and laser in situ keratomileusis (LASIK).
SMILE or lenticule reimplantation in a non-human Invest Ophthalmol Vis Sci. 2014;55(3):1823–34.
primate model. PLoS One. 2018;13(3):e0194209. 49. Reinstein DZ, Archer TJ, Randleman JB. Mathematical
36. Jacob S, Kumar DA, Agarwal A, Agarwal A, Aravind model to compare the relative tensile strength of the
R, Saijimol AI. Preliminary evidence of successful cornea after PRK, LASIK, and small incision lenti-
near vision enhancement with a new technique: pres- cule extraction. J Refract Surg. 2013;29(7):454–60.
byopic allogenic refractive lenticule (PEARL) cor- 50. Riau AK, Angunawela RI, Chaurasia SS, Tan DT,
neal inlay using a SMILE lenticule. J Refract Surg. Mehta JS. Effect of different femtosecond laser-
2017;33(4):224–9. firing patterns on collagen disruption during refrac-
37. Liu Y-C, Teo EPW, Ang HP, Seah XY, Lwin NC, tive lenticule extraction. J Cataract Refract Surg.
Yam GHF, et al. Biological corneal inlay for presby- 2012;38(8):1467–75.
51. Ivarsen A, Asp S, Hjortdal J. Safety and complications 65. Kamiya K, Shimizu K, Ohmoto F. Time course of
of more than 1500 small-incision lenticule extraction corneal biomechanical parameters after laser in situ
procedures. Ophthalmology. 2014;121(4):822–8. keratomileusis. Ophthalmic Res. 2009;42(3):167–71.
52. Yang W, Li M, Fu D, Wei R, Cui C, Zhou X. A com- 66. Sorbara L, Lopez JCL, Gorbet M, Bizheva K,
parison of the effects of different cap thicknesses Lamarca JM, Pastor J-C, et al. Impact of contact lens
on corneal nerve destruction after small incision wear on epithelial alterations in keratoconus. J Opt.
lenticule extraction. Int Ophthalmol. 2020;40(8): 2021;14(1):37–43.
1905–11. 67. Alio JL, Abdelghany AA, Barraquer R, Hammouda
53. Mastropasqua L, Salgari N, D’Ugo E, Lanzini M, LM, Sabry AM. Femtosecond laser assisted deep
Alió Del Barrio JL, Alió JL, et al. In vivo confo- anterior lamellar keratoplasty outcomes and healing
cal microscopy of stromal lenticule addition kera- patterns compared to manual technique. Biomed Res
toplasty for advanced keratoconus. J Refract Surg. Int. 2015;2015:397891.
2020;36(8):544–50. 68. Warrak EL, Serhan HA, Ayash JG, Wahab CH, Baban
54. Vega-Estrada A, Mimouni M, Espla E, Alió Del TA, Daoud RC, et al. Long-term follow up of intra-
Barrio J, Alio JL. Corneal epithelial thickness intracorneal ring segment implantation in 932 keratoconus
subject repeatability and its relation with visual limi- eyes. J Fr Ophtalmol. 2020;43(10):1020–4.
tation in keratoconus. Am J Ophthalmol. 2019;200: 69. Abreu AC, Malheiro L, Coelho J, Neves MM, Gomes
255–62. M, Oliveira L, et al. Implantation of intracorneal ring
55. Franco J, White CA, Kruh JN. Analysis of com- segments in pediatric patients: long-term follow-up.
pensatory corneal epithelial thickness changes in Int Med Case Rep J. 2018;11:23–7.
keratoconus using corneal tomography. Cornea. 70. Mastropasqua L, Nubile M, Lanzini M, Carpineto
2020;39(3):298–302. P, Ciancaglini M, Pannellini T, et al. Epithelial den-
56. Luft N, Ring MH, Dirisamer M, Mursch-Edlmayr dritic cell distribution in normal and inflamed human
AS, Kreutzer TC, Pretzl J, et al. Corneal epithe- cornea: in vivo confocal microscopy study. Am J
lial remodeling induced by small incision lenticule Ophthalmol. 2006;142(5):736–44.
extraction (SMILE). Invest Ophthalmol Vis Sci. 71. Wang D, Song P, Wang S, Sun D, Wang Y, Zhang Y,
2016;57(9):OCT176–83. et al. Laser scanning in vivo confocal microscopy of
57. Hwang ES, Schallhorn JM, Randleman JB. Utility of clear grafts after penetrating keratoplasty. Biomed
regional epithelial thickness measurements in corneal Res Int. 2016;2016:5159746.
evaluations. Surv Ophthalmol. 2020;65(2):187–204. 72. Chirapapaisan C, Abbouda A, Jamali A, Müller RT,
58. Reinstein DZ, Silverman RH, Rondeau MJ, Coleman Cavalcanti BM, Colon C, et al. In vivo confocal
DJ. Epithelial and corneal thickness measurements by microscopy demonstrates increased immune cell den-
high-frequency ultrasound digital signal processing. sities in corneal graft rejection correlating with signs
Ophthalmology. 1994;101(1):140–6. and symptoms. Am J Ophthalmol. 2019;203:26–36.
59. Reinstein DZ, Archer TJ, Gobbe M. Refractive 73. Kocaba V, Colica C, Rabilloud M, Burillon
and topographic errors in topography-guided abla- C. Predicting corneal graft rejection by confocal
tion produced by epithelial compensation predicted microscopy. Cornea. 2015;34(Suppl 10):S61–4.
by 3D Artemis VHF digital ultrasound stromal 74. Giannaccare G, Weiss JS, Sapigni L, Bovone C,
and epithelial thickness mapping. J Refract Surg. Mattioli L, Campos EC, et al. Immunologic stromal
2012;28(9):657–63. rejection after deep anterior lamellar keratoplasty
60. Kanellopoulos AJ. Comparison of corneal epithelial with grafts of a larger size (9 mm) for various stromal
remodeling over 2 years in LASIK versus SMILE: diseases. Cornea. 2018;37(8):967–72.
a contralateral eye study. Cornea. 2019;38(3): 75. Al-Aqaba MA, Otri AM, Fares U, Miri A, Dua
290–6. HS. Organization of the regenerated nerves in human
61. Nubile M, Salgari N, Mehta JS, Calienno R, Erroi E, corneal grafts. Am J Ophthalmol. 2012;153(1):29–37.
Bondì J, et al. Epithelial and stromal remodelling fol- e4.
lowing femtosecond laser-assisted stromal lenticule 76. Pang A, Mohamed-Noriega K, Chan AS, Mehta
addition keratoplasty (SLAK) for keratoconus. Sci JS. Confocal microscopy findings in deep anterior
Rep. 2021;11(1):2293. lamellar keratoplasty performed after Descemet’s
62. Damgaard IB, Ivarsen A, Hjortdal J. Biological stripping automated endothelial keratoplasty. Clin
lenticule implantation for correction of hyperopia: Ophthalmol. 2014;8:243–9.
an ex vivo study in human corneas. J Refract Surg. 77. Fontana L, Parente G, Sincich A, Tassinari G. Influence
2018;34(4):245–52. of graft-host interface on the quality of vision after
63. Lazaridis A, Messerschmidt-Roth A, Sekundo W, deep anterior lamellar keratoplasty in patients with
Schulze S. Refractive lenticule implantation for keratoconus. Cornea. 2011;30(5):497–502.
correction of ametropia: case reports and literature 78. Schiano-Lomoriello D, Colabelli-Gisoldi RA, Nubile
review. Klin Monatsbl Augenheilkd. 2017;234(1): M, Oddone F, Ducoli G, Villani CM, et al. Descemetic
77–89. and predescemetic DALK in keratoconus patients: a
64. Vega-Estrada A, Alio JL. The use of intracorneal ring clinical and confocal perspective study. Biomed Res
segments in keratoconus. Eye Vis. 2016;3:8. Int. 2014;2014:123156.
79. Bhatt UK, Fares U, Rahman I, Said DG, Maharajan 82. Riau AK, Angunawela RI, Chaurasia SS, Lee WS,
SV, Dua HS. Outcomes of deep anterior lamellar kera- Tan DT, Mehta JS. Early corneal wound healing
toplasty following successful and failed “big bubble”. and inflammatory responses after refractive lenti-
Br J Ophthalmol. 2012;96(4):564–9. cule extraction (ReLEx). Invest Ophthalmol Vis Sci.
80. Abdelkader A, Kaufman HE. Descemetic versus pre- 2011;52(9):6213–21.
descemetic lamellar keratoplasty: clinical and confo- 83. Doroodgar F, Jabbarvand M, Niazi S, Karimian
cal study. Cornea. 2011;30(11):1244–52. F, Niazi F, Sanginabadi A, Ghoreishi M, Alinia C,
81. Ganesh S, Brar S, Pandey R, Pawar A. Interface heal- Hashemi H, Alió JL. Customized stromal lenti-
ing and its correlation with visual recovery and quality cule implantation for keratoconus. J Refract Surg.
of vision following small incision lenticule extraction. 2020;36(12):786–94.
Indian J Ophthalmol. 2018;66:212.
Lamellar Surgeries with SMILE
Lenticules
20
Sri Ganesh and Sheetal Brar

SMILE-derived lenticules can be used success-
fully for the potential management of hyperopia, In 1949, José Ignacio Barraquer laid the ground-
keratoconus, SMILE ectasia, and presbyopia. work for the use of natural corneal tissue to
change the refractive properties of the eye [1, 2].
• Long-term clinical outcomes of tissue addi- Subsequently, Pradhan et al., in 2013, published a
tion with SMILE-derived lenticules performed case report showing the feasibility of the use of a
in 42 eyes with moderate to high hyperopia myopic SMILE lenticule (Endokeratophakia) for
showed a mean regression of +0.66 D, at a correction of aphakia [3]. Followed by this, many
mean follow-up of 68 months (5.6 years). researchers successfully reported the use of allo-
• Addition of thick lenticule (>6 D) resulted in genic and autologous SMILE lenticules for man-
posterior curvature changes in the cornea, agement of conditions such as high hyperopia,
leading to significant under correction of keratoconus, presbyopia, and sealing corneal
hyperopia. defects [4–8]. Recently, the technique has been
• Bowmans membrane relaxation (BMR) may shown to provide satisfactory and stable results
be an easy, practical, and cost-effective tech- for managing ectasia after SMILE [9].
nique to reverse the posterior curvature This chapter aims at discussing the feasibility
changes and enhance the effect of tissue addi- of SMILE-derived lenticules for the potential
tion for high hyperopia. management of hyperopia, keratoconus, SMILE
ectasia, presbyopia and reporting the clinical out-
comes, and experience of various researchers in
this evolving field so far.

org/10.1007/978-3-031-32408-6_20.
S. Ganesh (*) · S. Brar

Nethradhama Superspeciality Eye Hospital,
Bangalore, India
https://doi.org/10.1007/978-3-031-32408-6_20
298 S. Ganesh and S. Brar
amellar Surgery with SMILE

L in detail in a previously published paper by our
Lenticules for Hyperopia group [11]. Video 20.2 shows a surgical video of
enhancement with BMR.
In the technique of femtosecond intrastromal len- For the lenticule exchange procedure, a
ticule implantation (FILI), published by our group Sinskey’s hook was used to open the old incision
in 2014, the cornea was made steeper by the addi- and enter the corneal interface. A blunt spatula was
tion of a SMILE lenticule of known thickness and then used to dissect the tissue above and below the
power, into a pocket created in the recipient’s cor- implanted lenticule and separate the same from the
nea using a femtosecond laser [4]. The concept surrounding adhesions. The free lenticule was then
was subsequently adopted by various authors, who grasped with a micro-forceps from its edge and
reported their results with certain modifications in extracted from the corneal pocket. The interface
the technique [5, 9]. Recently, Liu et al. published was washed with a balanced salt solution, fol-
their 2-year results with SMILE lenticule implan- lowed by which, the fresh lenticule was implanted
tation and suggested that allogenic lenticule trans- into the interface using the standard technique of
plantation may be a promising option for correcting FILI, described above. Postoperative regimen was
moderate to high hyperopia [10]. similar to the one published earlier [4].
We recently concluded a retrospective study of
eligible patients, who underwent FILI for correc-
tion of moderate to high hyperopia from July 2013 Results
till October 2020. Inclusion criteria included: age
18 years or older, hyperopic refractive error rang- FILI was performed on 42 eyes of 25 patients.
ing between +3.00 to +11.00 D, stable refractive Table 20.1 provides the preoperative demo-
error (change of <0.5 D within the past 12 months), graphic and baseline data of all recipient patients,
pre-op CDVA of 0.6 LogMAR or better, and a as well as the donors whose lenticules were used
strong motivation for refractive correction. Eyes for implantation. Mean follow-up was
with previous keratitis, severe dry eye disease,
cataract, glaucoma, or vitreoretinal disorders, con-
Table 20.1 Preoperative demographic data of the recipi-
comitant autoimmune diseases, pregnancy, and ent and donor eyes
patients with unrealistic expectations are excluded. Parameter Mean ± SD
Recipient details
Age (Years) 27.04 ± 5.33
Surgical Procedure UDVA (LogMAR) 1.03 ± 0.39
CDVA (LogMAR) 0.22 ± 0.23
For FILI, the donor SMILE lenticules used were Sphere (D) 5.24 ± 1.96
either cryopreserved or fresh, i.e., the extracted Cylinder (D) 0.51 ± 0.48
SE (D) 5.50 ± 1.96
lenticule was used either in the same sitting or
CCT (μm) 550.02 ± 29.68
within 48 h, when stored in balanced salt solution. Km anterior (D) 43.72 ± 1.55
Briefly, the FILI procedure involved the insertion Km posterior (D) −6.30 ± 0.26
of the donor SMILE lenticule into a femtosecond Q-value −0.34 ± 0.09
laser-enabled pocket created using VisuMax FS HOA (RMS) 0.398 ± 0.15
Laser (Carl Zeiss Meditec, Jena, Germany) at a Donor details
depth of 160 μm, as described earlier [4]. Video Age (years) 28 ± 5.33
20.1 shows a surgical video of the procedure. S.E treated (D) −6.03 ± 1.99
Optical zone (μm) 6.50 ± 0.28
For the enhancement procedure, i.e., Bowman
Lenticule thickness (μm) 114 ± 25.70
membrane relaxation (BMR), a Hessburg–Barron
Length of cryopreservation (days) 61 ± 103.61
trephine (Barron Precision Instruments, Grand
UDVA, uncorrected distant visual acuity; CDVA, corrected
blanc, Michigan) was used to trephine the distant visual acuity; SE, spherical equivalent; D, diop-
Bowman’s membrane and part of the anterior ters; RMS, root mean square; CCT, central corneal
stromal fibers. The technique has been explained thickness
20 Lamellar Surgeries with SMILE Lenticules 299
Table 20.2 Visual and refractive results post FILI (n = 42 eyes) at 2 weeks and 68 months post-op
Pre mean ± SD 2 weeks mean ± SD p-value Last follow-up mean ± SD p-value
(pre vs. (2-week vs. last
Parameter (range) (range) 2 week) (range) follow-up)
UDVA 1.03 ± 0.39 0.21 ± 0.23 <0.001 0.25 ± 0.23 0.36
(LogMAR) (0.22–1.78) (−0.10 to 0.80) (−0.10 to 0.60)
CDVA 0.22 ± 0.23 0.19 ± 0.20 0.51 0.19 ± 0.21 0.88
(LogMAR) (−0.10 to 0.80) (−0.10 to 0.70) (−0.20 to 0.60)
Sphere (D) 5.24 ± 1.96 0.57 ± 0.82 <0.001 0.56 ± 0.94 0.95
(+3 to +11) (0 to +2.25) (−1.50 to +2.25)
Cylinder (D) 0.51 ± 0.48 0.14 ± 0.65 <0.001 0.19 ± 0.67 0.71
(0 to +1.50) (−1.50 to +1.50) (−1.25 to +1.50)
SE (D) 5.54 ± 1.96 0.64 ± 1.05 <0.001 0.66 ± 1.18 0.95
(+3 to +11) (−0.625 to +4.50) (−2.00 to +2.375)
UDVA, uncorrected distant visual acuity; CDVA, corrected distant visual acuity; SE, spherical equivalent; D, diopters
68 ± 17.28 months (12–84 months). Table 20.2 residual refraction at 2 weeks post-op was
shows the postoperative visual and refractive 0.64 ± 1.05 D, which showed a nonsignificant
results at 2 weeks and at the end of the mean increase to 0.66 ± 1.17 D at 68 months post-op,
follow-up. p = 0.95 (Fig. 20.1f).
There was a significant increase in the Kmean
anterior, central corneal thickness, Q-value, and
Efficacy and Safety corneal HOAs, 2 weeks post-op compared to the
preoperative values, p < 0.05 (Table 20.3).
At 68 months, the mean efficacy index was However, no significant change was observed in
0.86 ± 0.19 (0.39–1.0). The postoperative mean these parameters at 68 months, when compared
UDVA was 0.25 ± 0.22 (−0.12 to 0.6) to 2 weeks, p > 0.05 (Table 20.3). The Kmean
LogMAR. Cumulative UDVA of 20/20 or better posterior, on the other hand, showed a significant
and 20/40 or better was seen in 38% (n = 16) and change from −6.30 ± 0.26 to −6.13 ± 0.34 D,
81% (n = 34) of eyes, respectively (Fig. 20.1a). p = 0.02 (i.e., becoming more positive), 2 weeks
The mean safety index was 1.17 ± 0.39 (0.63– post FILI, which did not change significantly
2.54). Thirty-six percent (n = 15) eyes gained one thereafter (p value, 2 weeks vs. 68 months = 0.23).
or more lines, 45% (n = 19) had no change, Figures 20.2 and 20.3, respectively, show the
whereas 19% (n = 8) eyes lost one line of 2-weeks versus pre-op difference maps of both
CDVA. No eye lost more than 2 lines of CDVA eyes of a 29 year old male, who underwent FILI
(Fig. 20.1b). for high hyperopia of +6.5 D and +7.0 D in the
right and left eye, respectively. Compared to pre-
operative, an increase in K1, K2, and thinnest
pherical Equivalent (SE),
S pachymetry by 2.7 D, 3.2 D, and 96 μ was
Astigmatism Accuracy, and Stability observed at 2 weeks in the RE (Fig. 20.2).
Similar changes were observed in the LE of the
The accuracy of SE refraction within ±0.5 D was patient, wherein the KI, K2, and thinnest
observed in 50% (n = 21) eyes; however, 71% pachymetry increased by 3.0 D, 4.1 D, and 77 μ,
(n = 30) of all the treated eyes were within respectively (Fig. 20.3). Figures 20.4 and 20.5,
±1.00 D of SE correction. A coefficient of deter- respectively, show the difference maps of both
mination value of 0.71 was obtained on the pre- eyes of the same patient at a long follow-up of
dictability curve (Fig. 20.1c, d). Sixty-four 5.8 years versus 2 weeks post FILI. Figure 20.6a1,
percent (n = 29) eyes were within 0.5 D of astig- b1 show clinical photographs of both eyes of the
matism, while 88% (n = 37) eyes were within same patient at 2 weeks post-op, showing the
±1.00 D of astigmatism (Fig. 20.1e). The mean implanted lenticule in situ. Note that, in a freshly
a 98%100%
100%
b c
100% 42 eyes (plano target)
Last follow-up postop 42 eyes (Plano target) Mean: +5.53 ± 1.96
90% 90% 93% 42 eyes (Plano target) 2 or more lines lost 68 months post-op Range: +3 to +11
81% 68 months post op 0.0%
80% 76% +8
50%
Cumulative % of Eyes
Postop UDVA 45%

+7
Preop UDVA 62% Overcorrected
Achieved SEQ (D)

60% 40% +6
52%
% of Eyes
43% +5
38% 30% 26%
40% 36%
+4
26% 20% 19% Undercorrected
20% +3
10%
10% +2
0% 2% 2% Y = 0.5673x + 1.7703
0%
0% 0.0% +1 R2 = 0.714
20/12.520/16 20/20 20/25 20/32 20/40 20/63 20/80 20/100 0%
Loss 2 or Loss 1 No Gain 1 Gain 2 or
Cumulative Snellen VA (20/x or better) 0
More Change More 0 +1 +2 +3 +4 +5 +6 +7 +8 +9 +10
Change in Snellen Lines of CDVA
Attempted SEQ (D)
d 42 eyes (Plano target)

e f
±0.50 D: 50% 50%
68 months post-op ±1.00 D: 71% +8
43% 42 eyes (Plano target) 0.50 D: 64%
30% 68 months postop 1.00 D: 88% 42 eyes (Plano target)
40% +7
40% 68 months post-op
24% +6
21%
Mean ± SD SEQ (D)

20% 19% +5
30% +5.50
% of Eyes
% of Eyes
17%
24% 24% Postop Preop +4 % changed > 0.50 D
0.5 -68 months = 2.4%
20% +3
10% 12% 11.9%
12% 12% +2
5% 5% 5%
2.4% 2.4% 10% 7% 7% +1
4.8%
0% 2%
0 +0.64 +0.66
+0.14 to
+0.51 to
+1.01 to
-1.50 to
-1.00 to
-0.50 to
-0.13 to
0%0% 0%
0%0% 0%0%
<-1.50
>+1.50
+0.13
+0.50
+1.00
+1.50
-1.01
-0.51
-0.14
0%
≤ 0.26 0.51 0.76 1.01 1.26 1.51 2.01 > -1
0.25 to to to to to to to 3.00 0.5 m 68 m
Pre
Accuracy of SEQ to Intended Target (D) 0.50 0.75 1.00 1.25 1.50 2.00 3.00
Time After Surgery (months)
Refractive Astigmatism (D)
Fig. 20.1 JRS standard graphs for n = 42 eyes treated with FILI in the series
Table 20.3 Changes in Kmean anterior, Kmean posterior, central corneal thickness, Q-value, and corneal HOAs at
2 weeks and 68 months post-op
2 weeks Last follow-up
Pre mean ± SD mean ± SD p-value mean ± SD p-value
(Pre vs. (2-week vs. last
Parameter (range) (range) 2 week) (Range) follow-up)
Km Anterior 43.72 ± 1.55 47.45 ± 1.75 <0.001 47.48 ± 2.02 0.94
(D) (41.50–46.20) (44.20–50.30) (44.30–50.90)
Km Posterior −6.30 ± 0.26 −6.13 ± 0.34 0.02 −6.19 ± 0.31 0.23
(D) (−5.70 to (−5.37 to −6.60) (−5.50 to −6.70)
−6.80)
CCT (μm) 550.02 ± 29.68 631.59 ± 37.72 <0.001 625.76 ± 41.69 0.50
(494–596) (546–717) (530–720)
Q value −0.34 ± 0.09 −0.89 ± 0.23 <0.001 −0.95 ± 0.28 0.29
(−0.13 to (−0.43 to −1.69) (−0.40 to 1.94)
−0.55)
HOA (RMS) 0.39 ± 0.15 0.83 ± 0.34 <0.001 0.96 ± 0.34 0.10
(0.07–0.97) (0.13–1.62) (0.41–1.94)
HOA, higher order aberration; CCT, central corneal thickness; RMS, root mean square
implanted lenticule, the borders are well-defined Figure 20.7 demonstrates the corresponding
and mild folds in the tissue can be observed. AS-OCT scans with clear and well-centered len-
However, at 5.8 years post-op follow-up ticules in situ.
(Fig. 20.6a2, b2), the borders of the lenticules Four eyes of three patients underwent
are merged with the s urrounding host tissue, and enhancement with Bowman membrane relax-
a very faint boundary of the lenticule is visible. ation (BMR) for a significant residual refractive
The lenticule is relatively clear and does not error. Table 20.4 depicts the visual and refractive
have any folds or interface haze of any kind. outcomes of these eyes following enhancement.
Fig. 20.2 Two weeks versus pre-op difference maps of RE eye of a 29-year-old patient, who underwent FILI for hyper-
opic refractive error of +6.5 D
Fig. 20.3 Two weeks versus pre-op difference maps of LE of a 29-year-old patient, who underwent FILI for hyperopic
refractive error of +7.0 D
Fig. 20.4 RE difference map of 5.8 years versus 2 weeks post FILI of the same patient
Fig. 20.5 LE difference map of 5.8 years versus 2 weeks post FILI of the same patient
a1 b1
a2 b2
Fig. 20.6 Clinical photographs of the same patient; (a1, b1) 2 weeks post-op and (a2, b2) 5.8 years post-op, for the
right and left eyes, respectively
Fig. 20.7 AS-OCT of both eyes of the same patient at 5.8 years follow-up, with clear and well-centered lenticules in
situ
Table 20.4 Visual and refractive results of n = 4 eyes, enhanced with Bowman membrane relaxation (BMR) in the
series
Pre-FILI mean Preenhancement mean Post-enhancement mean
Parameter (range) (range) (range)
UDVA (LogMAR) 0.80 0.55 0.33
(0.50–1.00) (0.5–0.6) (0.3–0.4)
Sphere (D) +6.88 +1.50 +0.25
(+6.50 to +7.00) (+1.00 to +2.50) (0.00 to +0.50)
Cylinder (D) +0.69 +1.50 +0.12
(+0.50 to +1.00) (+0.50 to +3.00) (−1.50 to +1.25)
SE (D) +7.22 +2.25 +0.31
(+6.75 to +7.50) (+1.75 to +2.50) (−0.50 to +1.125)
CDVA (LogMAR) 0.30 0.35 0.30
(0.2–0.4) (0.3–0.4) (0.2–0.4)
UDVA, uncorrected distant visual acuity; CDVA, corrected distant visual acuity; SE, spherical equivalent; D, diopters
a b
Fig. 20.8 Dilated clinical pictures (a) oblique illumina- cule scarring. The hazy lenticule was explanted and
tion, b (retro illumination) of left eye of a patient at 1.5 exchanged with a fresh lenticule 2 weeks later, which
±years post-op, showing interface haze due diffuse lenti- remained clear over a follow-up of 6 years
Complications Discussion
Four eyes of two patients underwent explanta- In the present study, we evaluated the long-term
tion of the lenticule due to suspected stromal clinical outcomes of FILI for the treatment of
rejection. All the lenticules used in these eyes moderate to high hyperopia in 42 eyes treated
were cryopreserved. For one patient, the lenti- with this technique. At a mean follow-up of
cules were exchanged with fresh lenticules. 68 months (5.6 years), our results were fairly
Figure 20.8 shows the dilated clinical pictures of accurate and stable showing reduction of SE
left eye of this patient at 1.5 years post-op, show- from +5.54 D to +0.64 D at 2 weeks, and +0.66 D
ing interface haze due diffuse lenticule scarring. at 5.6 years. When compared to the long-term
Post-exchange, the lenticules remained clear results of hyperopic LASIK reported by Dave
with full recovery of patient’s visual acuity. For et al., the mean SE in their study reduced from
the second patient, lenticules were explanted +3.74 D to +0.84 D at a comparable follow-up of
after 3 years of the FILI procedure, following 5 years [12]. Biscevic et al. recently evaluated the
which hyperopic LASIK was performed, safety and efficacy of laser in situ keratomileusis
2 months later. (LASIK) procedure for the correction of high
hypermetropia in a retrospective study of 160 ing the posterior curvature. Moshirfar et al.
patients (266 eyes) who underwent LASIK pro- reported a case of high hyperopia of
cedure for the correction of hypermetropia +6.00/−1.00@40 managed with lenticule intra-
between +3.00 and +7.00 D and cylinder up to stromal keratoplasty (LIKE) procedure using a
2.00 D, using the Wavelight Allegretto Eye- thick corneal lenticule of 157 μm (+7.00 D),
Q400Hzexcimer laser (Alcon, Forth Worth, TX, implanted under a flap at a depth of 100 μm. At
USA) with aberration free module and were cen- 6 months post-op, manifest refraction reduced to
tered on a corneal vertex. The authors found that 0/−1.25@71, without any noticeable change in
1 week SE was 0.03 ± 0.67 D (−0.50 to +0.63 D), the posterior curvature (0.2 D change in steep K)
while at 1 year it regressed to 0.58 ± 0.56 D [15]. Damgaard et al. evaluated changes in cor-
(+0.25 to +0.88 D) [13], which was similar to that neal tomography after stromal lenticule implan-
observed by Dave et al. at 5 years. tation ex vivo, using a combination of two
These findings may support the previously implantation depths (110 and 160 μm) and two
proposed mechanisms described in relation with lenticule thicknesses (95 μm = 4.00 D,
tissue addition, such as lesser epithelial response, 150 μm = 8.00 D). For the front curvature, a
induced abberations, and better biomechanical 110 μm implantation depth induced significantly
stability, favoring this technique over excimer more steepening than a 160 μm depth in all
laser procedures for treating higher degrees of groups [16]. These observations may suggest that
hyperopia [4, 11]. SMILE, as a treatment a relatively superficial implantation of the lenti-
modality for hyperopia was explored by Pradhan cule may result in more pronounced anterior cur-
et al., who reported a relative change in SE from vature changes.
+5.61 D to −0.19 D at 12 months follow-up [14]. In terms of the changes in front keratometry,
However, they reported an 11% loss of follow-up corneal thickness, and Q-value, we noted a sig-
at the last visit. The authors suggested SMILE to nificant increase in these parameters after FILI at
be a promising modality for high hyperopia; 2 weeks, similar to the results of Liu et al.
however, a longer follow-up is awaited to assess obtained at 1 month post allogenic lenticule
the long-term stability following this procedure. implantation [10]. However, they observed a sig-
Liu et al. recently reported their 2 years clini- nificant decrease in the anterior keratometry at
cal experience of treating 14 eyes with implanta- 2 years when compared to 3 months values
tion of allogenic SMILE lenticule for moderate (−0.36 D, p < 0.001), without a significant cor-
to high hyperopia [10]. The technique of donor responding change is the SE refraction. On the
lenticule extraction and implantation in the recip- contrary, we did not observe any significant
ient eye was similar to the one published by our change in either anterior keratometry or SE val-
group in 2014 [4]. However, all of their lenticule ues at 68 months versus 2 weeks post-op. The
implantation procedures were scheduled on the much anterior placement of the lenticule (at
same day as the myopic donor eye SMILE. In our 100 μm) may result in an acute and exaggerated
series, 24 eyes were implanted with cryopreserved change in the anterior corneal curvature and
SMILE lenticules, whereas the remaining 18 Q-value, thus, making the cornea prone to regres-
eyes received fresh lenticules. Contrary to our sion due to resultant epithelial response. On the
results, Liu et al. noted a slight overcorrection, as other hand, it may be hypothesized that when the
the pre-op SE reduced from +5.53 D to −0.60 D lenticule is implanted at a deeper depth of
at 2 years post-op. This may be explained by the (160 μm), the anterior curvature changes
fact that the depth of the femtolaser pocket at observed for the same amount of tissue may be
which the donor lenticule was implanted in their more gradual, possibly resulting in lesser epithe-
study was set at 100 μm as compared to 160 μm lial response and better refractive stability.
in our study, which may have maximized the The ideal depth at which the lenticule must be
refractive effect by mainly changing the anterior implanted and long-term stability after tissue
corneal curvature, without significantly influenc- addition for hyperopia is debatable. Based upon
the observations and discussions of the afore- protection by cryoprotectants used [18]. However,
mentioned studies, it may be proposed that the the cases wherein fresh lenticules were used, may
lenticule be implanted at around 120–130 μm, in still need to be followed up, due to the potential
order to achieve the desired effect on the anterior risk of late stromal rejection, which remains.
curvature, without inducing much posterior Pretreatment with gamma radiation has been sug-
changes. Significant over and under corrections gested to deantigenize the donor tissue and pre-
may also be avoided by potentially improving the vent future rejection [19, 20]. The feasibility of
refractive predictability. this option, however, needs to be explored. It may
Tissue additive procedures for high hyperopia be noteworthy to mention that all the four eyes
may involve insertion of natural corneal tissue or for which the lenticules were explanted achieved
SMILE lenticule under a LASIK flap (lenticule complete visual recovery following reimplanta-
intrastromal keratoplasty-LIKE) [17] or inside a tion of fresh lenticules (two eyes) and subsequent
corneal pocket created using a femtosecond laser excimer treatment (two eyes), suggesting full
(FILI and s-LIKE) [4, 9]. The creation of a flap reversibility of the procedure.
for tissue addition may pose various challenges Moshirfar et al. suggested use of CIRCLE
such as increased risk of dry eye, DLK, weaken- [21] software and the side cut only technique to
ing of biomechanics, poor adhesion and disloca- convert the cap into a LASIK flap for the purpose
tion of the flap edge, and epithelial ingrowth that of enhancement after LIKE procedure for high
may not be present when the tissue is implanted hyperopia [9]. We achieved satisfactory out-
inside a pocket [9]. Moshirfar et al. reported a comes using the BMR technique for treating
case of moderate flap necrosis with epithelial residual refractive error after FILI, by potentially
ingrowth following LIKE procedure for high reversing the posterior corneal curvature changes
hyperopia, presenting at 1-month post-op [14]. [11].
Although, the case was managed with scraping of Thus, our tissue addition technique of FILI
the epithelial ingrowth, suturing and application resulted in satisfactory visual and refractive out-
of glue at the necrotic flap edge, however, the comes with good safety, efficacy, and stability of
incidence of such complications may be mini- achieved correction. Truly reversible nature of
mized by implanting the tissue in a stromal the procedure could be verified by successful
pocket, as the incision is small, and amount of retreatments resulting in complete restoration of
surgical manipulation is less. visual acuity in eyes requiring explantation of the
Liu et al. reported good safety profile with lenticules. Enhancements with BMR resulted in
14.3% eyes gaining one line, 78.6% showing no improved refractive accuracy. However, predict-
change, and 7.1% losing one line of CDVA at ability of refractive results may be further
2 years post-op [10]. In our study, 45% (19) eyes improved by suitable nomograms and modifica-
had no change, 36% (15) eyes gained one line or tions in surgical planning and techniques.
more, and 19% (8) eyes lost 1 line of CDVA. No
eye lost two or more lines of CDVA in either
study. However, there were 4 eyes in our series, amellar Surgery with SMILE
L
which required lenticule explantation due to sus- Lenticules for Keratoconus
pected stromal rejection diagnosed at a mean and SMILE-Ectasia
period of 2.25 years. A common factor in these
four eyes was the use of a cryopreserved tissue, Recently, the feasibility of use of SMILE tissue
compared to their study, wherein all the lenti- to treat corneal ectatic conditions has been suc-
cules were harvested and implanted on the same cessfully shown by our group. The technique has
day. Cryopreservation process may alter the shown promising results when used for potential
physical properties of the stromal collagen and management of keratoconus [6] and post SMILE
keratocytes, making them susceptible to necrosis, ectasia [22]. The surgical technique and guide-
possibly due to a relative lack of cell membrane lines for treatment planning were similar for both
the scenarios. Both the donor procedure and the was allowed to soak for 60 s, after which the
FILI procedure were planned for the same day, interface was washed with a balanced salt solu-
after receiving the serology report. The power of tion. Following this, the doughnut-shaped lenti-
the lenticule to be implanted was based on the cule was inserted into the pocket in a similar way
spherical equivalent of the recipient eye that had as for keratoconus eyes (described earlier).
signs of ectasia. Exclusion criteria were advanced After the insertion of the tissue, the eye was
ectasia with significant decentration, thinnest finally exposed to ultraviolet (UV) A radiation
pachymetry less than 400 mm, maximum kera- using a power of 18 mW/cm2 for a period of
tometry more than 55 dioptre (D), central scar- 5.8 min with an accelerated system (Avedro,
ring, or corrected distance visual acuity (CDVA) Inc.), delivering a total energy of 6.3 J for both
less than 6/36. keratoconus and SMILE ectasia cases.
After extraction from the donor eye, the Postoperatively, topical steroid in the form of
SMILE lenticule was thoroughly rinsed 3 times 1% prednisolone (Predforte, Allergan, Inc.) was
in a balanced salt solution to remove any debris, prescribed for a period of 3 months in tapering
tear secretions, or metallic particles that might dosage, along with 0.5% moxifloxacin antibiotic
have attached to the lenticule surface during dis- eyedrops (Vigamox 0.5%, Allergan, Inc.) 4 times
section, extraction, and examination of the lenti- a day for 2 weeks. Lubricants were prescribed for
cule on the corneal surface. The washed lenticule 4 times a day for 1 month and as needed
was then placed on a Teflon block and stained thereafter.
with 0.23% riboflavin solution (Peschke L) to
enhance its visibility. A 3.0 mm trephine was
used to punch the center of the lenticule to create Results in Keratoconus
a doughnut-shaped lenticule tissue.
For keratoconus [6], a stromal pocket was cre- Six eyes from six patients were included in the
ated at 100 μm depth using the VisuMax FS laser study. Based on values before and 6 months after
(Carl Zeiss Meditec, Germany). The pocket was the procedure, clinical improvement was noted in
dissected using a blunt spatula followed by injec- uncorrected distance visual acuity (1.06 ± 0.48
tion of 0.25% Vibex Xtra (Avedro) or 0.23% logMAR vs. 0.38 ± 0.27 logMAR), corrected dis-
(Peschke-L) dye into the interface for 60 s, after tance visual acuity (0.51 ± 0.20 logMAR vs.
which the interface was washed with normal 0.20 ± 0.24 logMAR), and manifest spherical
saline. The corneal vertex was marked with gen- equivalent (−3.47 ± 1.15 D vs. −1.77 ± 1.7 D).
tian violet using the first Purkinje image as a ref- There was flattening of mean keratometry in
erence while asking the patient to fixate on the 3-mm and 5-mm zones by 3.42 ± 2.09 D and
microscope light. The donut-shaped lenticule 1.70 ± 1.31 D, respectively. Mean pachymetry in
with the anterior aspect facing upward was held the central and midperipheral zones increased by
with lenticule forceps and gently inserted into the 18.3 ± 7.3 μm and 33.0 ± 8.8 μm, respectively.
pocket through the 4-mm superior incision. The All eyes had reduction in higher order aberra-
lenticule was positioned around the marked cen- tions, specifically coma. No eye lost lines of cor-
ter of the cornea and ironed out from the surface rected distance visual acuity. No adverse events
using a blunt spatula. Video 20.3 shows the video such as haze, infection, or allogeneic graft rejec-
of surgical procedure of FILI for keratoconus. tion were observed.
For SMILE ectasia [22], the old SMILE inci-
sion was opened using the Sinskey side of the
Reinstein dissector (Duckworth & Kent Ltd.), Results in SMILE Ectasia
followed by dissection of the pocket using the
blunt side of the same instrument. This allowed Four eyes of three patients (mean age 25.7 years)
access to the interface, into which 0.23% ribofla- developed features of keraectasia at a mean
vin dye (Peschke-L) was then injected. The dye period of 3 years after myopic SMILE correction.
All cases were managed with insertion of heter- all eyes. Uncorrected intermediate visual acuity
ologous SMILE lenticules in the previously ranged between J3 and J5 at 67 cm and uncorrected
created pocket, followed by simultaneous accel- distance visual acuity remained 20/20 in the oper-
erated CXL. At a mean follow-up of 7.67 months, ated eye and binocularly. The patients were com-
there was improvement in corrected distance fortable and reported independence from glasses for
visual acuity and reduction in keratometry and near, intermediate, and distance for all their routine
higher order aberrations in all eyes. The visual, visual tasks for the 6-month follow-up period. There
refractive, and topographic parameters remained were no complaints of dysphotopsia or troublesome
stable at the last visit compared with the 2-week night glare/halos. All lenticules remained well cen-
follow-up visit. No eye developed haze, infec- tered during the follow-up, and no lenticule-induced
tion, or rejection requiring tissue explantation. complications were seen.
Early experience showed tissue addition with According to the authors, The PEARL inlay
simultaneous pocket CXL to be a feasible acts as a shape-changing inlay by increasing the
approach for managing ectasia after central radius of curvature and resulting in a
SMILE. However, further follow-up is required hyperprolate corneal shape. Unlike the synthetic
to establish the long-term safety and effects on implants, there is unhindered passage of oxygen
corneal stabilization. and nutrients because the PEARL inlay is made
of allogenic cornea, thus ensuring stable corneal
conditions and decreasing the risk for corneal
amellar Surgery with SMILE
L necrosis and melt. It also has the advantages of
Lenticules for Presbyopia reversibility and adjustability. However, larger
data and longer follow-ups are needed to estab-
Jacob et al., recently, described a new technique lish the long-term safety and efficacy of the pro-
called PEARL (PrEsbyopic Allogenic Refractive cedure in presbyopic individuals.
Lenticule) that uses an inlay obtained from a
small incision lenticule extraction (SMILE) lenti-
cule [7]. With regard to the methodology, the ther Uses of SMILE-Derived
O
anterior side of the stored SMILE lenticule was Lenticules
identified, and it was spread out with the anterior
side facing up, dried with a surgical sponge, and Recently, the initial clinical outcomes of the
the center marked with the inked tip of a fine small incision lenticule extraction (SMILE)-
Sinskey hook. A 1-mm trephine was centered on derived glued lenticule patch graft for manage-
the inked mark to fashion a small donor allogenic ment of micro-perforations and complicated
presbyopic corneal inlay, which was then cen- corneal tears were reported [8, 23]. In our single-
tered at co-axially sighted light reflex while it center case series, seven eyes that presented with
was inserted into a corneal pocket created using a micro-perforations, partial-thickness corneal
femtosecond laser at 120 μm depth, into the non- defect, and traumatic complicated corneal tear
dominant eye of the presbyopic individuals. were repaired with a lenticule patch graft obtained
from the SMILE procedure. The patch was
secured to the recipient eye using fibrin glue.
Results Preoperatively, anterior segment optical coher-
ence tomography was used to assess the depth of
Four emmetropic presbyopic patients underwent the defect and to decide the thickness of the len-
PEARL inlay implantation in the nondominant eye. ticule. Patients were followed up on days 1, 7,
In the operated eye, uncorrected near visual acuity and 15 and at 1 and 3 months postoperatively.
at 33 cm improved from J8 to J2 in one and from J5, Main outcome parameters measured were best-
J6, and J7, respectively, to J2 in three operated eyes corrected visual acuity, clarity of the graft, and
with improvement between three and five lines in restoration of optical and tectonic integrity.
Results pia. J Refract Surg. 2015;31(6):374–9. https://doi.

org/10.3928/1081597X-20150521-03.
6. Ganesh S, Brar S. Femtosecond intrastromal lenticu-
Significant improvement in visual acuity was lar implantation combined with accelerated collagen
seen from 15 days onward in five of seven eyes. cross-linking for the treatment of keratoconus—initial
The lenticule graft was well apposed and clinical result in 6 eyes. Cornea. 2015;34:1331–9.
7. Jacob S, Kumar DA, Agarwal A, Agarwal A, Aravind
remained clear until the last follow-up visit in all R, Saijimol AI. Preliminary evidence of success-
eyes treated. ful near vision enhancement with a new technique:
Thus, the patch graft from the SMILE-derived PrEsbyopic Allogenic Refractive Lenticule (PEARL)
lenticule using fibrin glue seems to serve as a corneal inlay using a SMILE lenticule. J Refract Surg.
2017;33:224–9.
safe, feasible, and inexpensive surgical option for 8. Bhandari V, Ganesh S, Brar S, Pandey R. Application
the management of micro perforations and com- of the SMILE-derived glued lenticule patch graft
plicated corneal tears, especially in centers that in micro perforations and partial-thickness cor-
perform the SMILE procedure in large numbers. neal defects. Cornea. 2016;35(3):408–12. https://
doi.org/10.1097/ICO.0000000000000741. PMID:
26764882.
Take Home Notes 9. Moshirfar M, Shah TJ, Masud M, Fanning T, Linn
Lamellar corneal surgeries using SMILE lenti- SH, Ronquillo Y, Hoopes PC Sr. A modified Small-
cules appear to be a feasible and exciting concept Incision Lenticule Intrastromal Keratoplasty (sLIKE)
for the correction of high hyperopia: a description of
with encouraging results for the management of a new surgical technique and comparison to Lenticule
various corneal conditions, which carries a huge Intrastromal Keratoplasty (LIKE). Med Hypothesis
potential for further research. Long-term results Discov Innov Ophthalmol. 2018;7(2):48–56.
of hyperopic tissue addition using SMILE lenti- 10. Liu S, Wei R, Choi J, Li M, Zhou X. Visual outcomes
after implantation of allogenic lenticule in a 100-
cules are available; however, further results are μm pocket for moderate to high hyperopia: 2-year
awaited for keratoconus and presbyopia. Future results. J Refract Surg. 2021;37(11):734–40. https://
research is also suggested in the areas of nomo- doi.org/10.3928/1081597X-20210730-02. Epub 2021
gram refinement, evaluating biomechanical Nov 1. PMID: 34756142.
11. Brar S, Ganesh S, Sriganesh SS, Dorennavar L. Long-
changes, epithelial and stromal remodeling, tis- term outcomes of bowman’s membrane relaxation for
sue treatments, and preservation to prevent rejec- enhancement of femtosecond intrastromal lenticule
tion following this procedure. implantation performed for the management of high
hyperopia. J Refract Surg. 2022;38(2):134–41.
12. Dave R, O’Brart DP, Wagh VK, Lim WS, Patel P,
Lee J, Marshall J. Sixteen-year follow-up of hyper-
References opic laser in situ keratomileusis. J Cataract Refract
Surg. 2016;42(5):717–24. https://doi.org/10.1016/j.
1. Barraquer JI. Queratomileusis y queratofakia. Bogota: jcrs.2016.03.028. PMID: 27255248.
Instituto Barraquer de America; 1980. p. 342. 13. Biscevic A, Pidro A, Pjano MA, Grisevic S, Ziga
2. Barraquer JI. Refractive corneal surgery: experience N, Bohac M. Lasik as a solution for high hyper-
and considerations. An Inst Barraquer. 1993;24:113–8. metropia. Med Arch. 2019;73(3):191–4. https://
3. Pradhan KR, Reinstein DZ, Carp GI, Archer doi.org/10.5455/medarh.2019.73.191-194. PMID:
TJ, Gobbe M, Gurung R. Femtosecond laser- 31402804; PMCID: PMC6643362.
assisted keyhole endokeratophakia: correction of 14. Pradhan KR, Reinstein DZ, Carp GI, Archer TJ,
hyperopia by implantation of an allogeneic len- Dhungana P. Small Incision Lenticule Extraction
ticule obtained by SMILE from a myopic donor. (SMILE) for hyperopia: 12-month refractive and
J Refract Surg. 2013;29(11):777–82. https://doi. visual outcomes. J Refract Surg. 2019;35(7):442–50.
org/10.3928/1081597X-20131021-07. Erratum in: J https://doi.org/10.3928/1081597X-2 0190529-0 1.
Refract Surg. 2015;31:60. PMID: 31298724.
4. Ganesh S, Brar S, Rao PA. Cryopreservation of 15. Moshirfar M, Hopping GC, Somani AN, et al.
extracted corneal lenticules after small incision Human allograft refractive lenticular implantation
lenticule extraction for potential use in human sub- for high hyperopic correction. J Cataract Refract
jects. Cornea. 2014;33(12):1355–62. https://doi. Surg. 2020;46(2):305–11. https://doi.org/10.1097/j.
org/10.1097/ICO.0000000000000276. jcrs.0000000000000011.
5. Sun L, Yao P, Li M, Shen Y, Zhao J, Zhou X. The 16. Damgaard IB, Ivarsen A, Hjortdal J. Biological
safety and predictability of implanting autolo- lenticule implantation for correction of hyper-
gous lenticule obtained by SMILE for hypero- opia: an ex vivo study in human corneas. J
Refract Surg. 2018;34(4):245–52. https://doi. 21. Ganesh S, Brar S, Manasa KV. CIRCLE software
org/10.3928/1081597X-20180206-01. for the management of retained lenticule tissue fol-
17. Jankov M, Mrochen M, Seiler T. Laser intra- lowing complicated SMILE surgery. J Refract Surg.
stromal keratoplasty-case report. J Refract Surg. 2019;35(1):60–5. https://doi.org/10.3928/1081597X-
2004;20:79–84. 20181120-01. PMID: 30633789.
18. Borderie VM, Laroche L. Ultrastructure of cul- 22. Ganesh S, Brar S, Bowry R. Management of small-
tured and cryopreserved human corneal keratocytes. incision lenticule extraction ectasia using tissue
Cornea. 1999;18(5):589–94. PMID: 10487434. addition and pocket crosslinking. J Cataract Refract
19. Stevenson W, Cheng SF, Emami-Naeini P, Hua J, Surg. 2021;47(3):407–12. https://doi.org/10.1097/j.
Paschalis EI, Dana R, Saban DR. Gamma-irradiation jcrs.0000000000000335. PMID: 32694305.
reduces the allogenicity of donor corneas. Invest 23. Yang H, Zhou Y, Zhao H, Xue J, Jiang Q. Application
Ophthalmol Vis Sci. 2012;53:7151–8. of the SMILE-derived lenticule in therapeutic kera-
20. Utine CA, Tzu JH, Akpek EK. Lamellar keratoplasty toplasty. Int Ophthalmol. 2020;40(3):689–95. https://
using gamma-irradiated corneal lenticules. Am J doi.org/10.1007/s10792-019-01229-y. Epub 2019
Ophthalmol. 2011;151:170–174.e1. Nov 21. PMID: 31754892.
Corneal Allogenic Intrastromal
Ring Segments (CAIRS)
21
D. Sravani and Soosan Jacob
Key Points the additional procedures may be done away

• CAIRS is a new technique that flattens and with by just CAIRS alone.
regularizes ectatic corneas in patients with ker- • CAIRS can be used to treat complications of
atoconus, post LASIK ectasia, pellucid mar- synthetic ICRS such as corneal melts.
ginal degeneration, and other ectatic disorders.
• The advantages over synthetic ICRS are many
and are discussed further in this chapter. Introduction
• CAIRS can be used even in advanced cases
where conventional synthetic ICRS would not Corneal allogenic intrastromal ring segment
be implanted. In many cases, DALK can be (CAIRS) is a minimally invasive technique first
avoided by combining CAIRS with thin- performed by one of the authors (Soosan Jacob)
cornea cross-linking techniques. in 2015 for the treatment of keratoconus and
• Customized CAIRS: CAIRS can be further other corneal ectasias with the aim of improving
customized according to individual refraction the outcome achieved and decreasing the risks
and topography to tailor results for each patient. associated with synthetic intrastromal corneal
• Bioptics: CAIRS can be implanted first to cre- ring segments (ICRS).
ate synergy with other procedures such as
topography-guided PRK, phakic IOL, refrac-
tive lens exchange, etc. by decreasing the Ectatic Corneal Disorders
amount of residual irregular astigmatism and
hence the safety and efficacy of these addi- Ectasia is an abnormal progressive thinning and
tional procedures. In many cases, the need for protrusion of the cornea due to biomechanical
failure. Keratoconus, pellucid marginal degener-
ation, keratoglobus, and post-laser vision correc-
Supplementary Information The online version con- tion ectasia are common ectatic disorders seen in
clinical practice.
org/10.1007/978-3-031-32408-6_21.
D. Sravani · S. Jacob (*) Keratoconus

Dr. Agarwal’s Refractive and Cornea Foundation
(DARCF), Chennai, India
Keratoconus is an uncommon corneal disorder
Dr. Agarwal’s Group of Eye Hospitals, with progressive thinning and steepening of
Chennai, India; http://www.dragarwal.com
https://doi.org/10.1007/978-3-031-32408-6_21
312 D. Sravani and S. Jacob
the central and paracentral cornea. This causes Mechanism of Action of ICRS
irregular astigmatism affecting the quality of
vision [1]. Abnormal curvature maps with According to the Barraquers Thickness law,
abnormal posterior elevation, with or without necessary flattening of the central cornea can be
corneal thinning, are the typical topographic achieved either by removing the tissue from the
features of keratoconus. The disease is inher- center or placing the tissue in the periphery to
ited in an autosomal dominant pattern with no achieve an arc shortening effect [5]. The flatten-
gender predilection. With the incidence rate of ing that is achieved with ICRS can be improved
one in 2000 to upto 5 per hundred, it starts in by either decreasing the optical zone diameter
puberty and progresses over a decade into the or by placing a thicker segment. The final cylin-
mid-30s [2]. Downs syndrome, Marfan’s syn- drical power correction achieved is propor-
drome, Lebers congenital amaurosis, and tional to the arc length, and the thickness and
mitral valve prolapse are some syndromes diameter of the segment determine the final
associated with the disease. Although there are myopic correction [6]. Combining collagen
no clear causative factors, postulated risk fac- cross-linking with ICRS improved the corneal
tors include chronic exposure to ultraviolet biomechanics as well as the quality of the
light, atopy and constant eye rubbing [3]. The vision achieved.
treatment of keratoconus includes medical
management such as glasses and various con-
tact lenses. The surgical treatment includes airs or Corneal Allogenic
C
corneal crosslinking, intrastromal corneal ring Intrastromal Ring Segments
segments (ICRS), deep anterior lamellar kera-
toplasty (DALK), and in advanced cases, pen- CAIRS procedure is an alternative to ICRS and
etrating keratoplasty (PK). replaces synthetic segments with a biocompat-
ible allogenic material thereby reducing the
risk of complications [7–9]. Allogenic implants
I ntrastromal Corneal Ring can be prepared from fresh or preserved donor
Segments (ICRS) stroma, processed donor stroma, or any other
source of allogenic tissue [10]. These segments
Originally developed for refractive correction of are then inserted into circular intra-stromal
high myopia and astigmatism, ICRS became a key corneal channels created within the kerato-
treatment strategy for the treatment of keratoconus. conic patient’s eye [11, 12]. The mechanism of
ICRS segments are made from synthetic material action is possibly by a combination of different
polymethylmethacrylate (PMMA). Various types mechanisms—mid-peripheral volume augmen-
of ICRS are available commercially. tation, arc shortening, and epithelial remodel-
available ICRS include Intacs, Kerarings, ling. The myopic power decreases secondary
Ferrara rings, Myorings, Bisantis segments, etc. to increased posterior corneal curvature and
Each model of ICRS has its own nomogram. The decreased anterior corneal curvature. Regular
quality of vision is improved by flattening and and irregular astigmatisms decrease because of
regularizing the corneal surface. ICRS is associ- improvement in topography. The quality of
ated with complications such as migration, intru- vision is improved secondary to regularization
sion, extrusion, and overriding. Melting and of the anterior corneal surface. CAIRS can be
stromal necrosis are common with the PMMA inserted at a more superficial depth than syn-
segments [4]. Focal edema, deposits in the chan- thetic segments and also in more advanced
nels, and neovascularization are the other com- cases. CAIRS is especially advantageous in
mon complications. Rare trouble encompasses progressive keratoconus and in eye rubbers
infectious keratitis in a few cases. since in these patients, an initially satisfacto-
21 Corneal Allogenic Intrastromal Ring Segments (CAIRS) 313
Fig. 21.1 Comparison of the pre-op and post-op topographic map and parameters after CAIRS procedure showing
improvement. Difference map shows the flattening achieved together with improvement in all parameters
rily placed synthetic segment may become

superficial with time. This fear does not exist
with CAIRS [13, 14]. Excellent biocompatibil-
ity and biointegration (follow-up range more
than 8 years), centralization and regularization
of the cone, wide applicability even in patients
with a thin cornea, easy learning curve,
improved visual quality and quantity, and the
ability to exquisitely customize the segments
on an individual basis are some of the benefits
compared to synthetic ICRS [15, 16]. Fig. 21.2 Shows the same case as Fig. 21.1 with single
Though the risk of rejection exists with CAIRS segment placed on the temporal side extending
CAIRS, it is low because of multiple reasons: superiorly and inferiorly as well
host keratocytes can quickly repopulate
CAIRS tissue due to the low volume of tissue Indications and Contraindications
that is transferred and the ensconced position
within the host stroma; lack of epithelial and • Ectatic corneal disorders like keratoconus,
endothelial transfer; distance from the limbus; pellucid marginal degeneration, post LASIK
lack of sutures, etc. The safety and efficacy of ectasias, and post-ICRS melts are common
CAIRS have been quite good till date with sig- indications.
nificant improvements noted in refraction, • Systemic contraindications include autoim-
uncorrected and best-corrected visual acuities, mune disorders, collagen vascular disorders,
topographic parameters, and quality of vision immunodeficiency, etc.
achieved. • Viral keratitis sequalae, corneal scarring espe-
The mid-peripheral location and the untouched cially in central and paracentral areas and eyes
visual axis mean that even in the rare scenario of too thin to be cross-linked are ocular contrain-
a rejection, vision remains unaffected (Figs. 21.1 dications though further studies are required
and 21.2). to ascertain these.
Preoperative Workup be customized further if required using the tech-

nique described by one of the authors (SJ). The
A detailed eye examination, automated refracto- Jacob CAIRS CutomizerTM (Epsilon Instruments,
metric reading, retinoscopy, meticulous subjec- USA) helps to exquisitely and accurately custom-
tive verification of uncorrected and spectacle ize the segments manually to any desired shape.
corrected visual acuity, and contact lens trial with Customization with the femtosecond laser is cur-
rigid gas permeable lens are done. Various devices rently being developed by Ziemer Ophthalmic
such as Pentacam® (Oculus, Wetzlar, Germany), Systems AG (Sweden) in collaboration with the
Wavelight Topolyzer® (Alcon Laboratories, Inc., author (SJ) and Shady Awwad, MD (Lebanon).
Fort Worth, Texas), Zywave aberrometer® (Bausch Under topical medication, the tunnel for the
& Lomb Zywave, Rochester, NY), MS39 ASOCT implantation of CAIRS segment in the recipient’s
and topography® (CSO Italia), and Corvis® ST eye can be made using the femtosecond laser or
(Oculus, Wetzlar, Germany) are useful for assess- can be manually dissected at the mid periphery at
ing corneal tomography, higher order aberrations, mid stromal depth without disturbing the central
epithelial mapping, corneal biomechanics, etc. optical zone of the cornea. We prefer creating
broad mid-stromal channels with an inner diame-
ter of about 4.6 mm since unlike INTACS, CAIRS
Technique does not cause glare or haloes. CAIRS are inserted
using a push-through technique using a curved
CAIRS segment is prepared from allogenic tissue Y-rod and a curved reverse Sinskey making sure
such as the stroma of an allogenic donor corneo- the segments are not twisted. Progressive cases
scleral rim. First, the epithelium and endothelium are combined with collagen cross-linking (CXL)
are removed. A 360-degree segment of stromal or thin cornea cross-linking techniques such as
tissue is trephined using the Jacob double-bladed the contact lens assisted corneal cross-linking
trephine™ (Madhu Instruments, New Delhi). (CACXL) based on the corneal thickness as men-
Finally, the segments with desired thickness, tioned already (Figs. 21.3 and 21.4; Video 21.1)
width, and arc length are prepared. CAIRS may [17, 18].
a b c d
e f g h
Fig. 21.3 Preparing CAIRS. (a, b) Scraping off the epi- sue with the special trephine; (e, f) marking of the
thelium and endothelium from the donor corneal button; Bowman’s mebrane side of the punched out donor corneal
(c) Jacob CAIRS double bladed trephine™ (Madhu stroma in the trephine; (g, h) ring of corneal stromal tissue
Instruments, New Delhi) is used to punch out the corneal and CAIRS segments are prepared by cutting the ring to
stroma with precise sharp edges; (d) punching out the tis- the appropriate sized segments
a b c d
e f g h
Fig. 21.4 Preparation of a customized CAIRS segment. ization of the segment based on the topographic need is
(a) CAIRS segment is spread out; (b) the length of the seg- done by preparing a segment with superior tapering (the
ment to be implanted is marked intraoperatively as per the area that requires less flattening); (g) the segment is pushed
preoperative topographic plan after positioning it on the through using a curved Y-rod into a channel already made
sterile Jacob CAIRS Customizer™ (Epsilon Instrumemts, using the femtosecond laser; (h) the segment is implanted
USA); (c, d) appropriate lengths of the segments are without any twists and pulled through the other end using a
marked and prepared for the length and taper; (e, f) custom- reverse Sinskey hook, and finally, the segment is ironed out
The location of CAIRS is defined by the loca- are advised for 6 months when outside. Follow-up
tion of the cone. A single segment is placed in case includes uncorrected and spectacle-corrected
of a decentered cone, the arc length being sufficient visual acuity, regular topographic evaluation to
to enclose the area of steepness. A single segment assess all parameters and to analyze stability, and
generally suffices to correct comatose aberrations, anterior segment OCT to monitor segment thick-
flatten the cone and centralize it. For already cen- ness and depth on follow-ups.
tered cones, two 150° segments are placed for flat-
tening it. Asymmetric segments may also be
placed. Smaller segments at a slightly larger optic Benefit of CAIRS in Advanced Cases
zone are placed for correcting cylindrical errors.
Thicker CAIRS segments prepared using the CAIRS has a wide range of applications from early
6.5/8 mm diameter Jacob CAIRS trephine are used to advanced stages of keratoconus. In advanced
for higher refractive errors, whereas for low refrac- cases, for instance, with steep corneal curvatures
tive errors, the 7.75/8.75 mm diameter trephines >58 D and thin pachymetry <400–450 μm, placing
are used. The 8/9.5 mm trephines are used for very a synthetic segment at 70–80% depth is associated
advanced keratoconus [19, 20]. with risks and complications. The mid-stromal
placement of CAIRS is safer without the risk of
stromal melt, migration, intrusion or extrusion. It
Postoperative Medications therefore covers a wide range of candidates who
can undergo the procedure and a wide range of
A quick and short tapering course of antibiotic thicknesses, arc lengths and customization can be
plus steroid drops in conjunction with lubricants attained by the surgeon on table with CAIRS while
are used over a period of 5 weeks. If combined synthetic implants may need FDA approval for dif-
with cross-linking, ultraviolet protective glasses ferent thicknesses and different arc lengths.
Though deep anterior lamellar keratoplasty amuch larger degree than is possible with asym-
provides satisfactory outcomes in advanced metric synthetic ICRS (Video 21.2).
cases, the risk of complications such as irregular
astigmatism, rejection, secondary glaucoma,
graft failure, Urrets Zavalia syndrome, loosening Combined Procedures
of sutures, incitement of neovascularization etc.
exist. CAIRS on the other hand has a simple CAIRS may be combined with corneal cross-
learning curve with low risk of any intra-opera- linking or thin cornea cross-linking techniques
tive complication. It is helpful in avoiding or at such as contact lens assisted collagen cross-linking
least delaying the need for corneal transplanta- for progressive cases. They may also be combined
tion. In the scenario of progression post-CAIRS as bioptics with topography- guided photorefrac-
and CXL, DALK can still be performed safely. tive keratectomy, phakic IOL placement, refrac-
tive lens exchange, conductive keratoplasty, etc.
The advantage of CAIRS is reversibility and
Customized CAIRS adjustability—they can be exchanged, adjusted, or
customized post-surgery if required.
A standard ICRS device has uniform thickness
along its length. The idea of asymmetric ICRS is
to help achieve the necessary flattening where it Stability of the CAIRS Segments
is required and to create less effect on areas
which are less severely affected. Similar to asym- Serial scans of anterior segment OCT showed no
metric synthetic ICRS, customized CAIRS can significant change in segment thickness in our
be easily fashioned with different thicknesses, initial analysis, though larger studies are needed
width, arc lengths, diameters, etc. and with spe- to confirm this. The mid stromal placement and
cific shapes in order to tailor-make the CAIRS the combined cross-linking could be some of the
according to the individual patient’s topography factors contributing to the stability of the seg-
[21]. Such customized correction for asymmetric ments (Fig. 21.5).
keratoconus targets better correction of higher
order aberrations and is more helpful for achiev-
ing a better quality of vision. Pair segments with Results
different thickness, arc lengths, or widths are also
possible as is an arc length of up to 360° To date, about 600 cases have been performed by
Customization in case of CAIRS is possible to us including customized CAIRS. In the published
Fig. 21.5 AS-OCT

showing a customised
CAIRS segment with
one end tapered and
placed at 50% depth in
the corneal stroma
pilot study conducted by Jacob et al. [7], for a 3. Gordon-Shaag A, Millodot M, Shneor E, Liu Y. The
genetic and environmental factors for keratoconus.
period of 1 year between March 2016 and March Biomed Res Int. 2015;2015:795738.
2017, 24 eyes of 20 keratoconus patients who 4. Kwitko S, Severo NS. Ferrara intracorneal ring seg-
underwent CAIRS had a significant improvement ments for keratoconus. J Cataract Refract Surg.
in almost all postoperative parameters including 2004;30(4):812–20.
5. Barraquer JI. Modification of refraction by means
the visual outcome and the topographic parame- of intracorneal inclusions. Int Ophthalmol Clin.
ters. Significant improvement was observed in 1966;6(1):53–78.
both UCVA and BCVA with reduced topographic 6. Burris TE, Baker PC, Ayer CT, et al. Flattening of
astigmatism and an improvement in the mean central corneal curvature with intrastromal cor-
neal rings of increasing thickness: an eye-bank eye
power in both 3 and 5 mm zones. study. J Cataract Refract Surg. 1993;19(Suppl):
On follow-up ranging from 6 to 18 months, 182–7.
there was an improvement in UCVA within a 7. Jacob S, Patel SR, Agarwal A, Ramalingam A,
range of 0–8 lines, mean improvement of Saijimol AI, Raj JM. Corneal Allogenic Intrastromal
Ring Segments (CAIRS) combined with cor-
2.79 ± 2.65 lines. There was also an improvement neal cross-linking for keratoconus. J Refract Surg.
in CDVA by a mean of 1.29 ± 1.33 lines. The 2018;34(5):296–303.
depth of the implantation was at an average of 8. Carpal EF, Santilli C, Maltry A. Long-term viabil-
314.4 ± 61 μm. The central corneal thickness as ity of allogenic donor stroma. Indian J Ophthalmol.
2020;68(12):3057–9.
measured by ASOCT-Pachymetry did not show 9. Dapena I, Parker JS, Melles GRJ. Potential benefits
any statistically significant change when com- of modified corneal tissue grafts for keratoconus:
pared pre and postoperatively. On follow-up, no Bowman layer ‘inlay’ and ‘onlay’ transplantation, and
significant progression was observed as mea- allogenic tissue ring segments. Curr Opin Ophthalmol.
2020;31(4):276–83.
sured with Kmax/Steep K >0.75 D. CAIRS seg- 10. Parker JS, Dockery PW, Jacob S, Parker
ments were well positioned and remained stable JS. Preimplantation dehydration for corneal allogenic
in thickness during the entire follow up. intrastromal ring segment implantation. J Cataract
Refract Surg. 2021;47(11):e37–9.
11. Parker JS, Dockery PW, Parker JS. Flattening the
Take Home Notes curve: manual method for corneal allogenic intra-
stromal ring segment implantation. J Cataract Refract
• CAIRS is superior to synthetic segments as Surg. 2021;47(11):e31–3.
biocompatible allogenic material is used, 12. Parker JS, Dockery PW, Parker JS. Trypan blue-
assisted corneal allogenic intrastromal ring seg-
thereby reducing complications. ment implantation. J Cataract Refract Surg.
• Significant improvements are seen in visual 2021;47(1):127.
and topographic parameters. 13. Jarade E, Issa M, Chanbour W, Warhekar P. Biologic
• Customization according to the topographic stromal ring to manage stromal melting after intra-
stromal corneal ring segment implantation. J Cataract
pattern of the patient can further improve Refract Surg. 2019;45(9):1222–5.
visual outcome. 14. Kozhaya K, Mehanna CJ, Jacob S, Saad A, Jabbur NS,
Awwad ST. Management of anterior stromal necrosis
after polymethylmethacrylate ICRS: explantation ver-
sus exchange with corneal allogenic intrastromal ring
Financial Disclosure Soosan Jacob has a patent for spe-
segments. J Refract Surg. 2022;38(4):256–63.
cial trephines, devices and processes used to create these
15. Gunn D. New keratoconus treatment Australia:
segments as well as for the CAIRS segments and various
CAIRS - corneal allogenic intrastromal ring seg-
types of shaped corneal segments; Madhu; Ziemer.
ments. https://www.drdavidgunn.com.au/blog/cairs-
keratoconus-treatment. Accessed 19 Jul 2022.
16. Dockery P, Parker JS, Parker JS. Corneal allo-
genic intrastromal ring segments can be implanted
References with manual technique. https://www.healio.com/
news/ophthalmology/20201029/corneal-allogenic-
1. Rabinowitz YS. Keratoconus. Surv Ophthalmol. intrastromal-ring-segments-can-be-implanted-with-
1998;42(4):297–319. manual-technique. Accessed 19 Jul 2022.
2. Millodot M, Ortenberg I, Lahav-Yacouel K, Behrman 17. Kumar DA, Jacob S, Agarwal A. Contact lens-assisted
S. Effect of ageing on keratoconic corneas. J Opt. corneal cross-linking. J Refract Surg. 2015;31(7):496.
2016;9(2):72–7.
18. Srivatsa S, Jacob S, Agarwal A. Contact lens assisted 20. Jacob S. CAIRS (Corneal Allogeneic Intrastromal
corneal cross linking in thin ectatic corneas - a review. Ring Segments) for keratoconus, ectasias &
Indian J Ophthalmol. 2020;68(12):2773–8. irregular corneas. https://www.youtube.com/
19. Jacob S. CAIRS. A new technique for keratoconus watch?v=SkxLwgP8cXA. Accessed 19 Jul 2022.
and corneal ectasias. https://youtu.be/qF 9ycSQPq8. 21. Jacob S. Customised CAIRS. https://www.youtube.com/
Accessed 19 Jul 2022. watch?v=_iiGZqNz-ig&t=9s. Accessed 19 Jul 2022.
Advanced DALK Techniques:
Mushroom Mini Big Bubble DALK
22
Kunal A. Gadhvi and Bruce D. Allan
Key Points Background

• Endothelial rejection and decompensation, the
leading cause of penetrating keratoplasty (PK) Anterior lamellar keratoplasty was first published
failure. in 1914, only 9 years after the first penetrating
• Deep anterior lamellar keratoplasty (DALK) keratoplasty (PK) [1]. It was another 50 years
in conditions with intact endothelium before a deep dissection of the stroma demon-
improves graft survival by eliminating endo- strated that Descemet membrane could be bared,
thelial rejection. giving rise to deep anterior lamellar keratoplasty
• DALK is a difficult surgery with a high intra- (DALK) in 1965 [2].
operative conversion rate to PK. DALK is a technique of corneal grafting
• Automating stages of DALK with a femtosec- whereby anterior corneal tissue is selectively
ond laser significantly reduce the risk of con- removed, down to the Descemet or pre-Descemet
version to PK. layer. This can be achieved by a layer-by-layer
• Minimizing the diameter of Descemet baring dissection or more recently, the “Big Bubble”
to within the confines of the pre-Descemet pneumatic or visco-dissection techniques. The
layer with a mushroom DALK configuration benefit of DALK over a full-thickness PK is
reduces the risk of intraoperative perforation. maintenance of the host endothelium, the rejec-
tion of which remains the leading cause of graft
failure after PK [3].
Prior to the adoption of contemporary tech-
niques like Melles air reflection or the “Big
Bubble,” the outcomes of DALK were deemed to
org/10.1007/978-3-031-32408-6_22. be inferior to PK in British and Australian regis-
try studies [4, 5]. DALK was associated with
poorer visual outcomes and higher failure rates.
K. A. Gadhvi (*) The UK graft registry demonstrates that during
Department of Corneal and External Diseases,
Liverpool University Hospitals NHS Foundation the early adoption of DALK with contemporary
Trust, Liverpool, UK techniques, failure within 3 years of surgery was
e-mail: [email protected] twice as likely compared to PK in keratoconus
B. D. Allan [4]. However, 19% of these early failures were
Department of Corneal and External Eye Diseases, reported within 30 days of surgery with equal sur-
Moorfields Eye Hospital, London, UK vival beyond this time point, demonstrating that
https://doi.org/10.1007/978-3-031-32408-6_22
320 K. A. Gadhvi and B. D. Allan
surgical technique and management of early post- A type 2 bubble is highly unpredictable and
operative complications were the dominant rea- frequently bursts intraoperatively because of the
sons. Poorer visual outcomes following DALK fragility of the Descemet layer (see Fig. 22.1b).
were related to residual pre-Descemetic stroma of The type 1 bubble is stronger because of the
greater than 80 microns [6]. presence of the pre-Descemet layer. The pre-
As a result of these factors, the uptake of Descemet layer is a robust acellular layer in the pos-
DALK has lagged compared to the uptake of terior stroma, impervious to air and conferring a
other forms of lamellar corneal surgery [7]. significant strength to the eye. Air trapped between
DALK currently still trails behind PK in the man- the pre-Descemet and posterior stroma in a type 1
agement of keratoconus worldwide. bubble has an average burst pressure of 1.45 bar
However, since the wider adoption of modern compared to 0.6 bar in a type 2 bubble [11].
Descemet baring techniques, it has become easier The pre-Descemet layer measures 10.15
to remove the stroma and bare the pre-Descemet +/−3.6 microns thick and is made of predomi-
layer [8]. With these contemporary techniques, nantly type 1 collagen bundles in 5–11 lamellae
visual outcomes and survival for DALK are now in long transverse, longitudinal, and oblique bun-
similar to PK for common etiologies such as ker- dles. This layer terminates central to the Descemet
atoconus [9]. layer, inserting into the stroma at 6-9 mm diam-
Despite these advances in technique, DALK eter. Consequently, separation of the pre-
remains a challenging operation to perform. The Descemet from stroma pneumatically is limited
risk of perforation of the Descemet membrane to a maximum of 8-9 mm, at which point a white
and intraoperative conversion from DALK to PK ring will form from stromal emphysema [11].
remain high, at 0–24% [9]. This diameter, however, is dependent on individ-
Most intraoperative perforations occur in
peripheral rather than central corneal dissection.
Understanding the anatomy of the posterior cor- a
nea may be the key to designing a more reliable
technique. In the next section, we will explore the
natural cleavage plane in the posterior cornea
with a focus on the pre-Descemet layer to dem-
onstrate why the diameter of exposure is
important.
b
ole of the Pre-Descemet layer
R
in DALK
When a pneumatic separation of the posterior

cornea with the “Big Bubble” technique is suc-
cessful, the cleavage plane develops in two poten-
Fig. 22.1 (a) Type 1 bubble. The black dome represents
tial zones, between the pre-Descemet layer and corneal stroma with air trapped between the pre-Descemet
the stroma (Type 1 bubble- approximately 80% in red and the posterior stroma representing the location
of bubbles) or the Descemet and pre-Descemet/ of a type 1 bubble. Note the abrupt termination of the red
posterior stroma (type 2 bubble—approximately Descemet layer and type 1 bubble. (b) Type 2 bubble.
Here air is trapped between the Descemet in blue and the
20% of bubbles) [10]. Type 1 bubbles propagate pre-Descemet representing the location of a type 2 bubble.
from the center outward terminating before the The Descemet extends further into the periphery than the
limbus (see Fig. 22.1a). pre-Descemet
22 Advanced DALK Techniques: Mushroom Mini Big Bubble DALK 321
information from optical coherence tomography

(OCT) mapping of the corneal dimensions either
preoperatively or intraoperatively. The automa-
tion of surgery can improve the safety and repro-
ducibility of the trephination steps of
keratoplasty.
Different techniques for the use of femtosec-
ond lasers in DALK surgery have been
described for over a decade. In 2009, Price and
Farid and Steinert et al. [12, 13] demonstrated
Fig. 22.2 Bubble rupture following visco-expansion of a
type 1 bubble that the precise determination of cut depth
using femtolaser-assisted dissection improved
canula placement for pneumatic dissection and
ual anatomy and the maximal diameter of the enhanced stromal removal in layer-by-layer
bubble and will not always match the trephina- stromal dissection, helping surgeons to achieve
tion diameter in cases of DALK. DALK is often a deep, predetermined depth in the posterior
converted to PK because the Descemet tears stroma.
when a type 1 bubble bursts or during peripheral Additionally, 3D femtosecond laser-assisted
baring of the Descemet (Fig. 22.2). cut profiles can improve tissue apposition by
Reducing the diameter of deep dissection in enhancing the surface area of contact between the
DALK to 6 mm and respecting the natural anat- donor and recipient. Alio et al. report using a
omy of the pre-Descemet layer should reduce the femtosecond laser-created mushroom configura-
risk of perforation and subsequent conversion to tion where the amount of visible scar tissue
PK. This is the theory behind the “mini-bubble” between host and donor is comparatively greater
DALK/. in this configuration compared to conventional
A 6 mm DALK would not treat wider stroma vertical trephination [14]. This suggested the
pathology such as in keratoconus adequately. presence of enhanced healing which might allow
However, the mushroom configuration (6 mm earlier suture removal and protect against late
diameter deep posterior stromal dissection, large remodeling of the graft host junction in keratoco-
diameter mid/anterior stromal dissection) com- nus (see Fig. 22.3).
bines deep central dissection with wider removal A variation of the mushroom pattern is uti-
of the anterior corneal stroma. Manually, a mush- lized for minibubble femtosecond laser-
room pattern dissection is difficult to achieve. But assisted DALK (F-DALK) combining a large
automation using femtosecond laser technology anterior cap (9 mm diameter), removing abnor-
makes mushroom pattern dissection easy and mal corneal stroma in keratoconus over a wide
safe. area, with a small diameter posterior optical
zone (6 mm) which respects the pre-Descemet
layer insertion into the anterior stroma [11]
Use of Femtosecond lasers (see Fig. 22.4). This technique more than tri-
in Keratoplasty/DALK ples the surface area of contact between the
donor and host and reduces the area of
Femtosecond lasers have the capacity to deliver Descemet bearing by a factor of 1.8 (see
complex three-dimensional cut patterns within Fig. 22.5). It was hypothesized that a larger
the cornea through the process of photo- anterior cap diameter might also reduce post-
disruption at predetermined depths based on keratoplasty astigmatism.
Fig. 22.3 (a) a

Scheimpflug image of
manual DALK in cross
section 1 year and
10 years postoperative.
The recurrence of
ectasia outside of the
host and remodeling of
the graft- host junction
are demonstrated. (b)
This corresponds to an
increase in irregular
anterior corneal
curvature over the same
period
Fig. 22.4 (a) Configuration of F-DALK in diagrammatic

cross section. (b) Postoperative anterior segment OCT of
Mushroom pattern Femto-DALK (Casia SS-1000, Tomey,
Nagoya, Japan)
Fig. 22.5 The contact

area between donor and
host corneal stroma (in
red) is over three-fold
greater with a mushroom
DALK configuration
(52mm2) compared to
conventional manual
DALK (16mm2))
echnique for Mini Bubble

T a
mushroom Femto DALK Surgery
In the following section, we will describe a tech-

nique for mini-bubble mushroom F-DALK
developed at Moorfields Eye Hospital to combine
the benefits of a wide anterior stroma removal
and a small 6 mm central zone of deep
dissection. b
524
512 504
Preoperative Planning
3 mm
478 463
Preoperatively, it is important to identify how

445 420
deep the posterior ring cut of the mushroom kera- 432
toplasty needs to be set to allow a deep dissection
without perforating the host Descemet membrane.
This is achieved by using an optical coherence Fig. 22.6 (a) OCT image of keratoconus cornea in cross
tomography (OCT) device for mapping of the section. Corneal thickness assessed 3 mm wither side of
host cornea to identify the thinnest point at a the corneal vertex at 8 different points. (b) Thinnest point
at 6 mm identified and utilized for programming of host
6 mm diameter around the corneal vertex (see femtosecond deep trephination
Fig. 22.6) .
Laser Setting
A mushroom-cut pattern is programmed in both

donor and host corneas using the desired femto-
second Laser device. Some lasers now come with
bespoke software; others require the manual pro-
gramming of parameters.
How to Prepare the Host Cornea How to Prepare the Donor Cornea
The geometric center of the host cornea needs to The donor corneal button is prepared on an artifi-
be marked as a reference for the forthcoming cial anterior chamber. A thin layer of cohesive
femtosecond cuts. All cuts are centered in refer- OVD covers the anterior surface of the artificial
ence to this mark following applanation. chamber mount to protect the endothelium.
Applanation must be performed with a firm
supporting pressure within the artificial anterior
Programming the Host Cut chamber. Filtered air is used to bring the chamber
to a firm physiological pressure once the locking
A 6 mm diameter posterior side-cut is pro- ring has been engaged symmetrically over the
grammed. This is in reference to the minimum donor corneal limbus. The use of air is important.
thickness measured on OCT at 6 mm minus Air (gas) is compressible during femtosecond
70 μm. applanation even at supraphysiological pressure,
The depth of the lamellar ring cut (horizontal whereas balanced salt solution or viscoelastics
cut between 6 and 9 mm) should set as the maxi- are not. Air support in the artificial anterior cham-
mum depth of the posterior side cut minus 50 μm. ber facilitates wide applanation without softening
The diameter of the anterior side is at 9 mm the chamber and is particularly important for
for most cases, with a reduction to 8.7 mm where clean lamellar dissection using femtosecond
the minimum white-to-white measurement is less lasers with a flat applanation interface.
than 11 mm (estimated on the slit lamp). Once the subcorneal pressure has been
A minimum cut overlap of 20 μm should be checked, the corneal epithelium can be irrigated
programmed in all directions to ensure intersec- gently with BSS, then dried carefully around the
tion of all cuts (see Fig. 22.7). edges, so that a clear image of a thin meniscus at
Fig. 22.7 (a) Schematic a

of femtosecond laser cut HOST CUT PATTERN
pattern we used in host
corneas based on 9 mm (or 8.7 mm if white <11mm)
preoperative OCT
measurements of the 6 mm
host cornea. All cuts

were programmed to
intersect by a minimum
of 20 μm. (b) Color Maximum posterior side cut depth = thinnest point minus 70µm
image of host cornea Lamellar cut depth = posterior side cut depth minus 50µm
immediately following
femtosecond cut. 6 mm b
inner and 9 mm outer
ring cuts visible
the edge of the applanation is clearly seen during After the donor cut is completed, the culture
laser docking. Excessive fluid should be avoided, medium should be infused gently through the
as fluid may become trapped by capillary action artificial anterior chamber, to expel air from
between the applanation glass and the cornea giv- beneath the donor corneal endothelium. This
ing a false meniscus—a risk factor for an incom- marks the end of the preparatory stage before the
plete anterior side cut. The meniscus should be transplantation procedure.
concentric with and just outside the guide marker
for maximum cut diameter prior to laser cutting.
Wherever possible, larger diameter donor cor- Transplantation Procedure
neas without extensive peripheral lipid deposi-
tion (arcus) should be used. Smaller diameter DALK surgery is usually performed under gen-
corneas are no problem for DMEK, but can make eral anaesthesia or local anaesthesia with seda-
dissection of a larger anterior cap in F-DALK dif- tion. When performing this surgery, the
ficult. Eye Banks should be aware of this emerg- recommended approach is to use a variation of
ing trend in tissue selection preferences. the “Big-Bubble” technique described by Anwar
and Teichman [8].
Laser Parameters for Donor 1. Identifying the Inner 6 mm Zone: The first
step is blunt dissecting the lamellar cut and
For the donor cornea, a reciprocal mushroom cut marking the inner edge of the lamellar cut
pattern is programmed into the laser. The setting 360° with gentian violet. Following the injec-
for the anterior side cut diameter is equal to the tion of air, surgical emphysema may make
host diameter plus 0.3 mm (a marginal oversiz- this 6 mm diameter posterior side cut harder
ing). The posterior side cut diameter remains to identify—the ink will help with this.
6 mm, and the lamellar ring cut depth is that of 2. Creating a channel for the air cannula: The
the host depth plus 20 μm. This allows for donor deep aspect of the posterior side cut is identi-
tissue deturgescence post-transplantation. The fied and dried with a surgical sponge. A small
posterior depth of the donor tissue should be set sharp dissection using a bent 27-gauge nee-
to maximum to ensure complete anterior cham- dle, similar to a Paufique dissector, is made at
ber penetration (see Fig. 22.8). this level as a deep stromal entry point for a
DONOR CUT PATTERN
Host anterior cap diameter = 0.3mm
6 mm
Maximum posterior side cut depth = 900µm (full thickness)
Lamellar cut depth = Host lamellar cut depth + 20µm
Fig. 22.8 Schematic of femtosecond laser cut pattern we used in donor corneas based on preoperative OCT measure-
ments of the host cornea. All cuts were programmed to intersect by a minimum of 20 μm
blunt trochar to form the track for the air

cannula.
The blunt trocar is passed, dissecting as
closely as possible to the Descemet mem-
brane (see Fig. 22.9).
3. Pneumatic dissection: A blunt 27-gauge
Fontana cannula (Surgistar, Vista, CA) is then
advanced down this channel to the center of
the cornea for air dissection. Air is injected
with the aim of forming a big bubble which Fig. 22.9 Blunt trochar used to create channel in deep
will enable dissection down to the pre- posterior stroma
Descemet layer in the 6 mm central optical
zone (see Fig. 22.10). The bubble should not
be expanded much beyond 6mm in order to a
respect the anatomy of the pre-Descemet
layer and minimize the risk of bubble
rupture.
4. Checking you have a bubble and removal of
stroma: Following attempted air-dissection,
the small-bubble technique is used to confirm
the presence of a big bubble [15]. A small
bubble is introduced to the anterior chamber
through a paracentesis and the eye is rolled to b
ensure that the small bubble remains visible in
the anterior chamber periphery. If the small
bubble moves into the center of the anterior
chamber, this indicates that a big bubble sepa-
ration of the pre-Descemet layer has not been
achieved.
Where a big bubble is present, it is recom-
mended to proceed as described by Anwar
and Teichman to expose the pre-Descemet
layer using blunt scissors to clear residual Fig. 22.10 (a) Fontana cannula passed through the chan-
nel created with the trochar. (b) Air injected and mini
posterior corneal stromal tissue within the bubble forms between posterior stroma and
6 mm zone (see Fig. 22.11). pre-Descemet
5. Where no bubble is identified: Where no big
bubble is achieved, the next stage is visco-
dissection with cohesive OVD [16]. If this
fails, or if air was injected directly into the
anterior chamber, layer-by-layer manual dis-
section is required to clear the posterior
stroma within the optical zone (see Fig. 22.12).
Various DALK blunt lamellar dissectors are
available for this process. If a micro-
perforation develops, then air can be injected
via a paracentesis and the dissection c ontinued
distal to the site of the perforation. It is often Fig. 22.11 Exposing the pre-Descemet layer following
then possible to save the posterior lamella and successful big bubble
utcomes of the Mini Bubble

O
mushroom DALK
and Contemporary Technique
Perforation and Conversion
In comparison with standard 8 mm manual

DALK techniques, we found that minibubble
F-DALK resulted in significant reductions in
intraoperative perforation of the Descemet layer
Fig. 22.12 Manual layer-by-layer dissection to clear
posterior stroma within the optical zone (25% vs. 45%) and fewer conversions to PK (3%
vs. 25%). Our review of 58 consecutive patients
found no failures with this technique after 3 years
of follow-up. Moorfields Eye Hospital is a multi-
surgeon setting in which over 50% of DALK
cases are performed by surgeons in training oper-
ating under supervision. In these circumstances,
in particular, improvements in the safety and
technical ease of DALK surgery are important.
Promising outcomes have been reported for
other F-DALK techniques. Soulouti et reported a
1/860 perforation rate over 10 years with femto-
second laser-assisted trephination and full-width
deep dissection using a variation in Melles air
reflection technique, with good visual outcomes
Fig. 22.13 Early postoperative slit lamp image of at 1 year (CDVA 0.17 ± 0.12) [17]. It remains to
F-DALK with mushroom configuration. Small white be seen whether such promising results can be
arrow indicates superficial lamellar interface. Large white replicated in a multi-surgeon setting. But this
arrow indicates deep Descemet bearing 6 mm interface extremely low perforation rate suggests that par-
tial automation using femtosecond laser assis-
tamponade with air postoperatively, as for tance may enhance safety in DALK.
endothelial keratoplasty rather than convert to
PK. If air tamponade is used, strong pupil
dilation must be maintained in the early post- Big Bubble Success
operative period to avoid pupil block.
6. Securing the donor cornea Following host dis- F-DALK gives enhanced control over trephina-
section, the predissected donor cornea is tion depth, but placement of the dissection track
peeled from the mounted corneoscleral but- for air injection remains poorly controlled with
ton. The donor Descemet membrane is the technique we describe, and we did not observe
removed with semi-dry arrow tip sponges. any benefit for F-DALK in comparison with stan-
The donor is then washed in BSS and secured dard manual DALK in relation to the rate of Type
to the host with 16 interrupted 10–0 nylon I bubble formation (61 vs. 58%). Other groups
sutures (see Fig. 22.13). Continuous sutures have reported similar rates of big bubble forma-
should be used with caution because of the tion in manual and F-DALK. Alio et al. observed
higher risk of cheese wiring of the nylon a “Big Bubble” success rate of 84% with F-DALK
through the thinner mushroom cap (typically and 80% in manual surgery utilizing a mushroom
approximately 350 μm depth). (Video 22.1). F-DALK configuration with an 8 mm cap and
Table 22.1 Astigmatic outcomes of contemporary F-DALK studies

Astigmatism CDVA
Study Laser model Applanation n Diameter (Cut pattern) (D) (LogMAR)
Espandar et al. 2016 [21] Femtec 520F Curved 24 9.25 mm (decagonal cut) 1.82 ± 0.67 0.26 ± 0.16
Salouti et al. 2019 [17] Femtec 520F Curved 391 9.3–9.5 (mushroom or 3.34 ± 1.88 0.09 ± 0.09
decagonal)
Alio et al. 2015 [14] Intralase iFS Flat 25 8 mm (mushroom) 5.43 ± NR# 0.26 ± NR
Li et al. 2016 [19] Wavelight Flat 94 8.2 mm (button) 5.35 ± 1.73 0.08 ± 0.07
FS200
Gadhvi et al. 2020 [20] Intralase iFS Flat 58 9.17 ± 0.21 (mushroom) 5.00 ± 3.76* 0.16 ± 0.20
6 mm deep optical zone reaching 80% of the grafts are usually associated with better corneal
thinnest corneal pachymetry [14]. regularization reflected by reductions in surface
Automating the deep placement of the dissec- irregularity and asymmetry. Consequently, we
tion track for the air injection cannula in big- would have expected that there would be better
bubble F-DALK may result in more consistent BSCVA in F-DALK comparative to the smaller
big bubble formation. Buzzonetti et al. demon- diameter manual DALK. We found that 87% of
strated positive results using the IntraLase femto- patients undergoing F-DALK achieved BSCVA
second laser (Intra-Lase FS Laser; Abbott of 20/40 or better, similar to that of manual
Medical Optics, Inc., Santa Ana, CA) and a metal DALK surgery at 1 year postoperative [20]. On
mask with a 0.7 mm channel to create a tunnel refraction F-DALK was associated with similar
100 microns above the thinnest pachymetric astigmatic outcomes to manual DALK with a
point. Cannula placement within this tunnel mean refractive astigmatism of −5 ± 3.76D fol-
resulted in successful big bubble formation in lowing suture removal using the technique
9/10 consecutive eyes in their series [18]. described.
Recent advances, in particular the integration The explanation for the absence of improved
of OCT imaging with femtosecond laser systems, BSCVA or reduced astigmatism with F-DALK
may make variations of this approach more likely arises from the flat applanation of an irreg-
accessible. Liu et al., utilizing the Ziemer LDV ular cornea when creating all cuts with a femto-
Z8 laser (Ziemer Ophthalmic System, Port, second laser. Flat applanation results in distortion
Switzerland) used intraoperative OCT guidance and resultant noncircular anterior side cut in
and cut programming to create a tunnel cut 3 mm irregular corneas, for example, in keratoconus
in length, 80 μm in width, at a 60° downward eyes. F-DALK studies using curved applanators
angle to the applanated horizontal plane. This show a trend toward lower astigmatism (See
tunnel terminated 50 μm from Descemet mem- Table 22.1).
brane. They reported successful big bubble for-
mation after placement of the air dissection
canula within this pre-cut dissection track in 14 Future Development
consecutive cases of F-DALK [19].
Avenues for further investigation include utiliz-
ing liquid interfaces for corneal surgery where
ision outcomes and Refractive
V applanation can avoided to reduce strain on the
Outcomes cornea and improve the cutting process. At pres-
ent, this technique is available on the Ziemer
In F-DALK with a mushroom profile visual LDV Z8 femtosecond laser (Ziemer Ophthalmic
outcomes remain similar to manual DALK sur- System, Port, Switzerland) and has only been
gery despite large graft diameters in the utilized in studies of PK [22] and is yet to dem-
F-DALK technique described. Larger diameter onstrate an improvement in circularity of the
corneal ex vivo [23] with clinical studies lacking Take Home Notes
in this area. Additionally liquid interface kerato- • Mushroom DALK with Femtosecond laser has
plasty cut profile is limited to vertical or oblique a lower perforation and conversion rate in mul-
at present with no mushroom profile tisurgeon settings compared to manual DALK.
programable. • Mushroom configurations reduce the diameter
Excimer laser keratoplasty could similarly of the Descemet baring which may contribute
help improve trephination regularity. In a to this.
mixed study of 35 eyes undergoing trephina- • “Mini” type 1 bubble avoids expansion of the
tion with an excimer laser (Zeiss Meditec bubble to the termination of the pre Descemet.
MEL70 excimer laser) and 34 eyes with a fem- This may reduce bubble rupture rate.
tosecond laser with flat applanation (60-kHz • Flat applanating Femtosecond laser may con-
IntraLase™ femtosecond laser) Tóth et al dem- tribute to higher-than-expected astigmatism
onstrated a lower post suture removal astigma- despite large graft caps. Future development in
tism and better BSCVA were achieved in the curved and liquid interfaces may mitigate this.
excimer group inferring less irregular astigma-
tism [24]. Obviously, the drawback of the
excimer laser is the absence of 3-D profiles
however this could be overcome by combining
with femtosecond assistance for lamellar cuts References
with the final anterior cap cut made with the
1. Singh NP, Said DG, Dua HS. Lamellar keratoplasty
excimer laser. techniques. Indian J Ophthalmol. 2018;66(9):1239–50.
2. Brown SI, Dohlman CH, Boruchoff SA. Dislocation
Of descemet’s membrane during keratoplasty. Am J
Surgical Pearls Ophthalmol. 1965;60:43–5.
3. Borderie VM, Sandali O, Bullet J, Gaujoux T,
Touzeau O, Laroche L. Long-term results of
Filling the artificial chamber with air during the deep anterior lamellar versus penetrating kerato-
applanation phase of donor preparation makes plasty. Ophthalmology. 2012;119(2):249–55. Feb
applanation at a supraphysiological pressure eas- [cited 2020 Feb 21]. http://www.ncbi.nlm.nih.gov/
pubmed/22054997.
ier as air is compressible while liquids or visco- 4. Jones MNA, Armitage WJ, Ayliffe W, Larkin DF,
elastics are not. Kaye SB, NHSBT Ocular Tissue Advisory Group and
The large diameter of the anterior cap in Contributing Ophthalmologists (OTAG audit study 5).
F-DALK necessitates large donor corneas with- Penetrating and deep anterior lamellar keratoplasty
for keratoconus: a comparison of graft outcomes
out peripheral arcus or opacity which may other- in the United Kingdom. Invest Ophthalmol Vis Sci.
wise interfere with the donor cutting. Do 2009;50(12):5625–9. Dec [cited 2020 Feb 21]. http://
communicate this with your eyebank. www.ncbi.nlm.nih.gov/pubmed/19661238.
Because of the larger graft with sutures being 5. Coster DJ, Lowe MT, Keane MC, Williams
KA. Australian corneal graft registry contributors.
placed close to the limbus in F-DALK, we found A comparison of lamellar and penetrating kerato-
the incidence of loose sutures to be greater. As a plasty outcomes: a registry study. Ophthalmology.
result, interrupted sutures or a double running 2014;121(5):979–87. http://www.ncbi.nlm.nih.gov/
suture provide an added level of security over a pubmed/24491643.
6. Ardjomand N, Hau S, McAlister JC, Bunce C,
single continuous. Galaretta D, Tuft SJ, et al. Quality of vision and
Although the endothelium is not transplanted graft thickness in deep anterior lamellar and pen-
in DALK and rejection more than 2 years after etrating corneal allografts. Am J Ophthalmol.
DALK is unusual there is still a significant risk 2007;143(2):228–35.
7. Keenan TDL, Jones MNA, Rushton S, Carley
within the first 2 years of surgery and we advise FM. Trends in the indications for corneal graft sur-
keeping patients on some topical immune sup- gery in the United Kingdom: 1999 through 2009.
pression, preferably a low dose steroid, during Arch Ophthalmol. 2012;130(5):621–8.
this period.
8. Anwar M, Teichmann KD. Deep lamellar kerato- (DLKP) in keratoconus patients using viscoadaptive
plasty: surgical techniques for anterior lamellar viscoelastics. Cornea. 2005;24(2):178–81.
keratoplasty with and without baring of descemet’s 17. Salouti R, Zamani M, Ghoreyshi M, Dapena I, Melles
membrane. Cornea. 2002;21(4):374–83. GRJ, Nowroozzadeh MH. Comparison between man-
9. Gadhvi KA, Romano V, Fernández-Vega Cueto L, ual trephination versus femtosecond laser-assisted
Aiello F, Day AC, Allan BD. Deep anterior lamellar deep anterior lamellar keratoplasty for keratoconus.
Keratoplasty for keratoconus: multisurgeon results. Br J Ophthalmol. 2019;103(12):1716–23.
Am J Ophthalmol. 2019;201:54–62. 18. Buzzonetti L, Petrocelli G, Valente P, Iarossi G, Ardia
10. Dua HS, Faraj LA, Kenawy MB, AlTaan S, Elalfy R, Petroni S, et al. The big-bubble full femtosecond
MS, Katamish T, et al. Dynamics of big bubble forma- laser-assisted technique in deep anterior lamellar
tion in deep anterior lamellar keratoplasty by the big Keratoplasty. J Refract Surg. 2015;31(12):830–4.
bubble technique: in vitro studies. Acta Ophthalmol. 19. Liu Y-C, Wittwer VV, Yusoff NZM, Lwin CN,
2018;96(1):69–76. Seah XY, Mehta JS, et al. Intraoperative optical
11. Dua HS, Faraj LA, Said DG, Gray T, Lowe coherence tomography-guided femtosecond laser-
J. Human corneal anatomy redefined: a novel pre- assisted deep anterior lamellar keratoplasty. Cornea.
Descemet’s layer (Dua’s layer). Ophthalmology. 2019;38(5):648–53.
2013;120(9):1778–85. 20. Gadhvi KA, Romano V, Fernández-Vega Cueto L,
12. Farid M, Steinert RF. Deep anterior lamellar kerato- Aiello F, Day AC, Gore DM, et al. Femtosecond laser-
plasty performed with the femtosecond laser zigzag assisted deep anterior lamellar Keratoplasty for kera-
incision for the treatment of stromal corneal pathol- toconus: multi-surgeon results. Am J Ophthalmol.
ogy and ectatic disease. J Cataract Refract Surg. 2020;220:191–202.
2009;35(5):809–13. 21. Espandar L, Mandell JB, Niknam S. Femtosecond
13. Price FWJ, Price MO, Grandin JC, Kwon R. Deep laser-assisted decagonal deep anterior lamellar kera-
anterior lamellar keratoplasty with femtosecond- toplasty. Can J Ophthalmol. 2016;51(2):67–70.
laser zigzag incisions. J Cataract Refract Surg. 22. Boden KT, Schlosser R, Boden K, Januschowski K,
2009;35(5):804–8. Szurman P, Rickmann A. Novel liquid Interface for
14. Alio JL, Abdelghany AA, Barraquer R, Hammouda femtosecond laser-assisted penetrating Keratoplasty.
LM, Sabry AM. Femtosecond laser assisted deep Curr Eye Res. 2020;45(9):1051–7.
anterior lamellar Keratoplasty outcomes and healing 23. Donner R, Schmidinger G. Effects of femtosecond
patterns compared to manual technique. Biomed Res laser-assisted trephination on donor tissue in liquid
Int. 2015;2015:1. interface as compared to applanated interface. Acta
15. Parthasarathy A, Por YM, Tan DTH. Use of a “small- Ophthalmol. 2022;100(2):e409–13.
bubble technique” to increase the success of Anwar’s 24. Tóth G, Szentmáry N, Langenbucher A, Akhmedova
“big-bubble technique” for deep lamellar keratoplasty E, El-Husseiny M, Seitz B. Comparison of excimer
with complete baring of descemet’s membrane. Br J laser versus femtosecond laser assisted trephination in
Ophthalmol. 2007;91(10):1369–73. penetrating Keratoplasty: a retrospective study. Adv
16. Shimmura S, Shimazaki J, Omoto M, Teruya A, Ther. 2019;36(12):3471–82.
Ishioka M, Tsubota K. Deep lamellar keratoplasty
Large Diameter Deep Anterior
Lamellar Keratoplasty
23

• Deep anterior lamellar keratoplasty (DALK)
involves selective replacement of diseased Deep anterior lamellar keratoplasty (DALK)
corneal stroma while preserving normal involves targeted replacement of diseased cor-
healthy endothelium. neal stroma while preserving healthy unaffected
• Despite several well-recognized advantages, endothelium. Although the concept of lamellar
DALK has failed to gain widespread popular- grafting was first introduced more than 150 years
ity among corneal surgeons. ago by von Walther and Mühlbauer, DALK has
• Compared to conventional DALK, large 9.0- gained renewed interest in recent years [1].
mm diameter DALK can provide superior Specifically, various surgeons have developed
visual outcomes at higher levels of Snellen several techniques for deep lamellar dissection
BSCVA and significantly lower degrees of including pneumatic dissection by Archila [2],
astigmatism without an increased risk of the big-bubble technique by Anwar and
immune rejection and graft failure. Teichmann [3], hydrodelamination by Sugita and
• Large diameter DALK with limited stromal Kondo [4], and viscoelastic-assisted DALK by
clearance within the 6.0-mm optical zone Maloney and associates [5]. Of these methods,
optimizes visual and refractive outcomes big-bubble DALK has emerged as the most popu-
while minimizing the risk of complications. lar approach [2]. Basically, the original technique
described by Anwar et al. involves partial thick-
ness trephination followed by angled centripetal
advancement of a needle and forceful intrastro-
mal injection of air to obtain the so-called “big-
bubble” [2].
One of the main challenges with the big-
bubble technique is the need to reach the central
Supplementary Information The online version con- cornea while relying on subjective signs of the
tains supplementary material available at https://doi. appropriate cannular depth [6]. However, based
org/10.1007/978-3-031-32408-6_23. on empirical evidence, what is most critical for
successful pneumatic dissection is not the radial
A. C. Yu · M. Busin (*) distance of the tip of the cannula to the center of
Department of Translational Medicine University of the cornea but the depth at which air is injected
Ferrara, Ferrara, Italy within the corneal stroma [7, 8]. Thus, when the
https://doi.org/10.1007/978-3-031-32408-6_23
332 A. C. Yu and M. Busin
tip of the cannula is inserted within 100 μm from weighed against the higher risk of immunologic
the posterior corneal surface, big-bubble forma- endothelial rejection and consequent graft failure.
tion can exceed 90% even with minimal centrip- On the other hand, large diameter 9.0-mm DALK
etal advancement of the cannula [9]. has demonstrated excellent visual outcomes, espe-
Recent advances in corneal imaging and sur- cially at higher levels of Snellen BSCVA without
gical instrumentation have driven significant an increased risk of complications.
improvements in the surgical procedure. Anterior
segment optical coherence tomography (OCT)
provides accurate measurements of the exact area Indications
of interest. Adjustable guarded trephines can be
calibrated to the desired depth. It is our prefer- The most common indication for DALK is kera-
ence to set the trephine within 100 μm from the toconus (KC), which accounts for more than half
thinnest pachymetry value at the 9 mm zone so of cases. Other surgical indications include trau-
that the base of the deep trephination can be used matic or infective corneal scarring, corneal stro-
to guide subsequent insertion of the air injection mal dystrophy and degeneration, postrefractive
cannula [7]. Instead of trying to reach the proper surgery ectasia, and scarring due to various
depth during advancement of the cannula toward causes including severe ocular surface disease
the center of the cornea, the insertion of the can- with limbal stem cell deficiency, Stevens–
nula can be initiated at the base of a deep trephi- Johnson syndrome, ocular cicatricial pemphi-
nation with minimal centripetal advancement by goid, and chemical or thermal burns. ALK can
approximately 2 mm [7]. Using deep trephination also be performed for tectonic indications includ-
during large-diameter DALK, the success rate of ing pellucid marginal degeneration and sterile
pneumatic dissection has been demonstrated to Mooren’s ulcer.
approach up to 85% [10]. Logistic regression
analyses of consecutive DALK surgeries have
likewise affirmed that both increased trephina- Keratoconus
tion diameter and depth positively influence suc-
cessful big-bubble formation [10–12]. Keratoconus is a degenerative corneal disease
Despite significant advances in the surgical characterized by progressive corneal thinning
technique and instrumentation, DALK has failed and scarring. In its early stages, keratoconus can
to gain widespread popularity among corneal be managed adequately by spectacle or contact
surgeons [3]. This is probably because the safety lens correction. However, in cases of reduced
benefits of DALK including eliminating the risk visual acuity secondary to corneal scarring and
of immune endothelial rejection and avoiding contact lens intolerance due to high degree irreg-
progressive endothelial cell loss become fully ular astigmatism, a surgical intervention would
evident only several years postoperatively. often be required. Since keratoconus cases pres-
Consequently, most surgeons do not feel suffi- ent with isolated anterior pathology, DALK is an
ciently compelled to face the learning curve in ideal procedure to retain the healthy endothelium
performing this procedure. Instead, an improve- and thereby obviate the risk of endothelial graft
ment in visual and refractive outcomes, which are rejection [13]. In big bubble DALK, advanced
directly perceived by patients soon after suture stages of keratoconus with high mean K values
removal, could provide the same surgeons with a and lower pachymetry values are associated with
greater incentive to transition to DALK. more frequent intraoperative complications [10].
It is commonly believed that increasing the The presence of scarring and older age are also
graft size can improve postoperative visual reha- significant predictors of type 2 bubble formation
bilitation by minimizing refractive astigmatism. which in turn confers an increased risk of compli-
However, with PK, an increase in graft size beyond cations such as conversion to full-thickness kera-
the conventional 8.0–8.5-mm diameter must be toplasty [14, 15].
23 Large Diameter Deep Anterior Lamellar Keratoplasty 333
orneal Scarring Secondary

C tocytes separating the overlying anterior stroma
to Trauma or Infection from a thin layer of pre-Descemet membrane
stroma, which consisted of poorly organized col-
In response to an injury or infection, the cornea lagen lamellae [24].
undergoes a complex process of wound healing
involving inflammation, cell proliferation, and
tissue fibrosis ultimately leading to stromal scar- Surgical Technique
ring. For proper preoperative assessment, ante-
rior segment optical coherence tomography In 2013, Busin and associates introduced large
(AS-OCT) is valuable for assessment of the depth diameter (9 mm) DALK with limited deep stro-
and extent of scarring. When possible, the cor- mal clearance of the 6-mm optical zone [9].
neal endothelium should also be evaluated by Standardization of the large diameter DALK
specular microscopy. Given that traumatic and technique (Fig. 23.1) has substantially simplified
post-infective corneal scars are often associated the procedure and allowed high success rates
with otherwise functional endothelium and that independent of surgical experience [25]. Initial
most of these patients frequently have an emme- partial thickness deep trephination is carried out
tropic fellow eye, DALK should be initially by a guarded trephine calibrated within 100 μm
attempted. In cases with a macroperforation or from the thinnest anterior segment OCT pachym-
unsatisfactory clearance within the optical zone etry value at the 9 mm zone [26]. In cases with
because of a full-thickness opacity, DALK can be significant asymmetry in corneal thickness,
converted to full-thickness keratoplasty. peripheral intrastromal hydration with BSS can
For cases of scarring due to HSV keratitis, be performed in zones with relative thinning to
DALK should be contemplated only after the safely allow deep trephination [27].
eyes have remained quiescent without episodes A blunt probe is then inserted 1 mm centripe-
of reactivation or inflammation for a period of tally from the base of the trephination. The blunt
6 months or longer. Initial high-dose antiviral probe is then replaced by a cannula, which is
prophylaxis with extended taper is recommended advanced 1 mm further along the same track cre-
to reduce the risk of disease recurrence [16]. ated by the probe, before attempting pneumatic
dissection. Often, significant intrastromal air
accumulation in the cornea precludes visualiza-
Other Indications tion of the circular silvery sheen of the bubble. In
such cases of extensive emphysema, one can
DALK can also be adapted for challenging cases ascertain whether big bubble formation has been
with intracorneal ring segments [17], previous achieved based on lateral displacement of an air
radial keratotomy [18], anterior lamellar kerato- bubble injected intracamerally [28].
plasty [19], and even previous DSAEK surgery Correct identification and management of the
[20, 21]. type of bubble achieved during pneumatic dissec-
Pneumatic dissection can also be achieved in tion are instrumental in minimizing complication
post-PK eyes. However, it is often challenging to rates. Depending on the movement of air injected
prevent expansion of the air bubble beyond the within the corneal stroma, pneumatic dissection
PK wound [22]. Alternatively, stromal exchange can occur along physiologic cleavage planes. A
can be accomplished through stromal peeling type 1 bubble is a well-circumscribed central
along a natural plane of separation [23]. dome-shaped elevation that results from a separa-
Ultrastructural alterations in the stromal microar- tion between the deep stroma from the pre-
chitecture allow stromal peeling of grafted cor- Descemet layer while a type 2 bubble is a large,
neas along a deep natural plane of separation thinner-walled elevation at the level of Descemet
without any type of dissection [23, 24]. The natu- membrane [29]. Since the residual bed consists
ral plane occurs along a continuous layer of kera- of a thin Descemet’s membrane-endothelium
Fig. 23.1 Intraoperative steps of large diameter (9 mm) big bubble deep anterior lamellar keraotplasty with stromal
clearance limited to the 6 mm optical zone
complex, a type 2 bubble is associated with a tion and conversion to PK [9]. The presence of a
lower bursting pressure and is therefore more peripheral stromal shoulder also protects the
fragile [30]. recipient bed during suturing, which has been
After en bloc anterior keratectomy is per- reported to cause up to 21% of intraoperative
formed starting from the base of deep trephina- Descemet’s membrane perforations during con-
tion, the central 6 mm of the bubble roof is ventional DALK surgery [33]. Over time, remod-
removed by baring the optical zone at the level of eling of the corneal stromal architecture with
the pre-Descemet’s layer or Descemet’s mem- progressive disappearance of the posterior step at
brane (DM), depending on the plane of dissection the edge of the 6 mm zone has also been observed
achieved [9]. (Figs. 23.2 and 23.3) [13].
Since a type 1 bubble is often limited within A 9-mm anterior lamellar graft is then pre-
7–8 mm optical zone, we prefer to combine pared by means of a 400-μm microkeratome head
large-diameter keratoplasty with limited stromal and sutured into place. The donor cornea is ini-
clearance within the central 6 mm [9]. Several tially fixed with four cardinal 10–0 nylon sutures
investigators have likewise employed limited at 3, 6, 9, and 12 o’clock. Interrupted or double-
stromal clearance for femtosecond laser-assisted running 10–0 nylon sutures can be used depend-
DALK [31, 32]. By respecting the anatomy of the ing on the primary pathology and presence of
pre-Descemet layer, restricting the deep dissec- neovascularization [9].
tion to the central cornea has been shown to Although baring of the Descemet’s membrane
reduce the risk of Descemet’s membrane perfora- can be attempted in cases with a type 2 bubble,
a b
Fig. 23.2 Slit lamp photos taken 1 month (a) and 5 years (b) after large diameter DALK
dent risk factor for double anterior chamber for-

a mation and conversion to PK [14, 15]. A mixed
type 1 and 2 bubble is typically managed as in a
type 1 bubble, allowing the type 2 bubble to
resorb spontaneously [13].
b
Even with the current surgical technique, the
overall success rate of pneumatic dissection still
ranges between 60 and 85% [1]. In the event of
failed big-bubble formation, other lamellar dis-
c section techniques can be employed to salvage
DALK surgery. Through the so-called microbub-
ble incision technique, additional air can be
injected into the corneal stroma. The presence of
d emphysematous tissue can then be used to guide
layer-by-layer dissection, which is aimed at
achieving a smooth recipient bed while remov-
ing as much residual corneal stroma as possible
[35].
Since manual dissection is particularly tedious
Fig. 23.3 Anterior segment optical coherence tomogra- and is a significant risk factor for conversion to
phy showing the transition between the 6-mm central PK, we prefer sequential viscoelastic-assisted
zone and the 9 mm outer zone with the residual recipient dissection in cases of failed pneumatic dissection
stroma. One day postoperatively a peripheral stromal
shoulder is clearly visible (a) but eventually disappears and consider manual dissection only when all
through spontaneous stromal thinning and remodeling of else fails [13, 36]. Even in cases of failed pneu-
posterior corneal curvature 1 month (b), 1 year (c), and matic dissection, the success of visco-bubble for-
2 years (d) after surgery mation approaches 90% [36]. Although retained
viscoelastic is initially associated with visually
manual or viscoelastic-assisted dissection tech- significant interface haze, the effect on visual
niques are often preferred in order to avoid inad- acuity is transient and final outcomes are compa-
vertent Descemet’s membrane perforation [34]. rable with those of big-bubble DALK (Video
Type 2 bubble formation is the strongest indepen- 23.1) [37].
Optimizing Visual Outcomes In our practice, should such complications occur

during large-diameter DALK, the procedure is
In the past, success following corneal transplan- converted to 2-piece mushroom PK, as first
tation was defined by the presence of a clear cor- described in 2004, instead of a full-thickness PK
neal graft [38]. Today, however, visual acuity is [9]. Minimal endothelial transplantation in mush-
considered the primary criteria for surgical suc- room PK combines excellent visual and survival
cess. Based on some earlier comparative studies, outcomes with a lower risk of immune rejection
visual outcomes were apparently worse follow- related to less antigenic load [47].
ing DALK due to inclusion of cases, which were The most frequent complication encountered
not reflective of the contemporary surgical tech- during DALK is Descemet’s membrane perfora-
nique, and, thus, were likely associated with tion [1]. If limited to a micro-perforation, the sur-
irregular or incomplete lamellar dissection [39– gery can still be successfully completed as per
41]. Currently used surgical techniques, which standard technique by filling the anterior chamber
exploit physiologic cleavage planes, result in a with air. In cases with a macroperforation or unsat-
clear graft host-interface compatible with satis- isfactory clearance within the optical zone because
factory visual outcomes, comparable with PK of a full-thickness opacity, DALK can be con-
[26, 42]. Although prospective clinical trials are verted to full-thickness keratoplasty. A mushroom-
ideally required to make statistically valid com- shaped wound configuration for full- thickness
parisons, randomizing patients to PK over DALK keratoplasty whether obtained through microkera-
could no longer be justified especially in view of tome- or femtosecond laser- assisted techniques
the disproportionately higher risk for potentially can be considered in order to combine the advan-
severe complications following PK. Nevertheless, tages of a large anterior refractive surface with
empiric evidence from large interventional case minimal endothelial transplantation [9, 32].
series has thus far validated the clinical benefits Double anterior chamber formation is another
of DALK [26, 43, 44]. complication following DALK. Inadvertent per-
From a refractive standpoint, the use of large- foration of DM, even if surgery is still success-
diameter grafts for DALK has been shown to pro- fully completed without the need for conversion
vide superior visual outcomes with lower degrees to PK, has been described to lead to an increased
of myopia and astigmatism [36, 45]. In a large risk of early postoperative detachment of the
series of 346 eyes, 9 mm DALK is associated recipient bed and consequent double anterior
with excellent outcomes with up to 94% of cases chamber formation. This has been found to occur
achieving Snellen vision ≥20/40 and up to 89% without a perforation detected intraoperatively,
with refractive astigmatism <4.5 diopters [26]. while, on the other hand, an intraoperative perfo-
Unlike PK, large-diameter DALK does not pose ration does not necessarily lead to the formation
an increased risk of immune-mediated stromal of a double anterior chamber. Corneal scarring,
rejection, which in any case can be managed perforation, and the occurrence of a type 2 bubble
medically [16]. Additionally, performing large- are known independent risk factors for double
diameter keratoplasty maximizes the removal of anterior formation following DALK. Although
the diseased stroma, thereby preventing disease there are reports of spontaneous resolution, this
recurrence [46]. complication usually requires rebubbling with air
or gas to be managed successfully. Complete AC
air fill is maintained for 2 h before release of air
anaging Complications
M at the slit lamp to achieve a fluid level at the
height of the inferior pupillary border, thereby
The use of larger grafts conceivably provokes eliminating the risk of pupil block [15].
apprehension among some surgeons due to the One of the leading causes of limited visual
potential need of conversion to full-thickness recovery after DALK is high astigmatism espe-
keratoplasty in cases of a significant perforation. cially when greater than 4.5 D. With refraction
after DALK stabilizing approximately 6 months tion groove, which in turn can be used to guide
after complete suture removal, refractive surgery subsequent placement of the air injection cannula
can be subsequently performed to reduce residual [56]. Further work is still needed to establish and
postoperative astigmatism. Residual refractive achieve the full potential of intraoperative OCT
errors after DALK can be managed through cor- for intraoperative guidance of lamellar surgery.
neal or lens-based procedures. Astigmatic kera- Several investigators have also explored fem-
totomy, either manual or femtosecond tosecond laser-assisted DALK surgery [57–63].
laser-assisted, has been found to reduce astigma- As in stepped PK wounds, customized trephina-
tism, although generally with unpredictable tion patterns created using the femtosecond laser
results [48]. Deeper, longer, and more central confer the advantage of an increased donor-host
incisions are often required to achieve a greater junction surface area, which theoretically can
treatment effect. An advantage to DALK with provide superior wound strength, induce faster
limited central stromal clearance would be the wound healing, and allow earlier suture removal
ability to create deep arcuate blunt relaxing inci- [57]. The femtosecond laser system can also be
sions within the graft–host interface which result employed to create a deep intrastromal tunnel for
in satisfactory outcomes with minimal risk of the air injection cannula for pneumatic dissection
perforation [49]. In the presence of a cataract, [58]. However, one of the main drawbacks of
lens extraction with implantation of monofocal femtosecond laser DALK is the poor laser pene-
or even toric intraocular lenses can also be con- tration through opacified and neovascularized
templated [50, 51]. For IOL power calculation, it corneal tissue, which can result in incomplete or
must be considered that a tendency toward a irregular dissection [64]. Although Li et al. dem-
myopic refractive shift is often observed in post- onstrated better visual outcomes following fem-
DALK eyes. Although the SRK/T, Kane, EVO, tosecond laser-assisted DALK, visual
and Hoffer QST formulas tend to provide more performance in the manual DALK group was
accurate outcomes, the predictability of refrac- poorer due to greater residual bed thickness
tive outcomes following cataract surgery remains obtained in diamond knife-assisted lamellar dis-
lower than in virgin eyes [52]. section [63]. Moreover, all other published stud-
Other complications of DALK include persis- ies comparing manual and femtosecond
tent epithelial defect, corneal neovascularization, laser-assisted DALK, thus far, consistently find
glaucoma, high astigmatism, cataract formation, no significant differences in terms of final visual
and interface infection [13, 53]. acuity [58–62]. In general, the internal validity of
Automated solutions have been explored to these comparative studies is affected by the retro-
further address the technical challenges associ- spective design [58–62], heterogenous study
ated with DALK. Cross-sectional imaging can populations [59, 60], unequal sample sizes [60,
reduce the surgeon’s dependence on subjective 61, 63], or varied surgical protocols for lamellar
cues during the critical depth-dependent steps. dissection [59, 62, 63], graft sizing [59, 60, 63],
Intraoperative OCT platform provides direct and even suture techniques [59]. Randomized
visualization and allows instantaneous quantita- controlled trials based on sufficient sample size
tive analysis of acquired OCT scans for intraop- and standard protocol would be necessary to
erative planning based on patient-specific corneal allow direct comparison and assessment of the
anatomy [54–56]. Thus far, however, most of the true benefit of femtosecond laser technology for
proposed applications of intraoperative OCT dur- DALK.
ing DALK are primarily qualitative [54, 55].
Based on our initial experience with a microscope- Take Home Notes
integrated intraoperative OCT with a built-in
caliper tool, trephination depth can be used to • With its superior safety profile and favorable
assist decision-making on whether to proceed postoperative outcomes, current evidence sup-
with pneumatic dissection or extend the trephina- ports that DALK outperforms PK for the
anagement of anterior corneal pathology

m 11. Borderie VM, Touhami S, Georgeon C, et al.
Predictive factors for successful type 1 big bub-
sparing the endothelium. ble during deep anterior lamellar keratoplasty.
• Recent advances in corneal imaging and avail- J Ophthalmol. 2018;2018:4685406. https://doi.
able instrumentation have led to the standard- org/10.1155/2018/4685406.
ization of the big-bubble DALK technique. 12. Feizi S, Javadi MA, Daryabari SH. Factors influenc-
ing big-bubble formation during deep anterior lamel-
• Regardless of the success of big-bubble for- lar keratoplasty in keratoconus. Br J Ophthalmol.
mation, other lamellar dissection techniques 2016;100(5):622–5.
can be employed sequentially as an alternative 13. Yu AC, Mattioli L, Busin M. Optimizing outcomes
to pneumatic dissection. for keratoplasty in ectatic corneal disease. Curr Opin
Ophthalmol. 2020;21(4):268–75.
• Since the unaffected host endothelium is 14. Myerscough J, Friehmann A, Bovone C. Evaluation of
retained, DALK provides the opportunity to the risk factors associated with conversion of intended
use large-diameter grafts, which can more deep anterior lamellar keratoplasty to penetrating ker-
reliably achieve maximum visual potential atoplasty. Br J Ophthalmol. 2020;104(6):764–7.
15. Myerscough J, Bovone C, Mimouni M, et al. Factors
without an increased risk of immune-mediated predictive of double anterior chamber formation fol-
stromal rejection. lowing deep anterior lamellar keratoplasty. Am J
Ophthalmol. 2019;205:11–6.
16. Giannaccare G, Weiss JS, Sapigni L, et al. Immunologic
stromal rejection after deep anterior lamellar kerato-
plasty with grafts of a larger size (9 mm) for various
stromal diseases. Cornea. 2018;37(8):967–72.
References 17. Ravera V, Bovone C, Scorcia V, et al. Deep anterior
lamellar Keratoplasty in eyes with intrastromal cor-
1. Yu AC, Spena R, Pellegrini M, et al. Deep anterior neal ring segments. Cornea. 2019;38(5):642–4.
lamellar keratoplasty: current status and future direc- 18. Einan-Lifshitz A, Belkin A, Sorkin N, et al. Evaluation
tions. Cornea. 2022;41(5):539–44. of big bubble technique for deep anterior lamellar ker-
2. Archila EA. Deep lamellar keratoplasty dissection atoplasty in patients with radial keratotomy. Cornea.
of host tissue with intrastromal air injection. Cornea. 2019;38(2):194–7.
1984;3(3):217–8. 19. Yu AC, Myerscough J, Galante G, et al. Pneumatic
3. Anwar M, Teichmann KD. Big-bubble technique to dissection for large-diameter (9-mm) deep anterior
bare descemet’s membrane in anterior lamellar kera- lamellar keratoplasty in eyes with previous anterior
toplasty. J Cataract Refract Surg. 2002;28(3):398. lamellar keratoplasty. Cornea. 2021;40(9):1098–103.
4. Sugita J, Kondo J. Deep lamellar keratoplasty with 20. Gutfreund S, Leon P, Graffi S, et al. Deep anterior
complete removal of pathological stroma for vision lamellar keratoplasty after descemet stripping auto-
improvement. Br J Ophthalmol. 1997;81:184–8. mated endothelial keratoplasty. Am J Ophthalmol.
5. Manche EE, Holland GN, Maloney RK. Deep lamel- 2017;175:129–36.
lar keratoplasty using viscoelastic dissection. Arch 21. Busin M, Beltz J. Deep anterior lamellar keratoplasty
Ophthalmol. 1999;117(11):1561–5. after descemet stripping automated endothelial kera-
6. Scorcia V, Lucisano A, Pietropaolo R, et al. Red toplasty. Cornea. 2011;30(9):1048–50.
reflex-guided big-bubble deep anterior lamellar kera- 22. Scorcia V, Beltz J, Busin M. Small-bubble deep
toplasty: a simple technique to judge dissection depth. anterior lamellar keratoplasty technique. JAMA
Cornea. 2015;34(9):1035–8. Ophthalmol. 2014;132:1369–71.
7. Busin M, Scorcia V, Leon P, et al. Outcomes of air 23. Bovone C, Nahum Y, Scorcia V, et al. Stromal peel-
injection within 2 mm inside a deep trephination for ing for deep anterior lamellar keratoplasty in post-
deep anterior lamellar keratoplasty in eyes with kera- penetrating keratoplasty eyes. Br J Ophthalmol.
toconus. Am J Ophthalmol. 2016;164:6–13. 2022;106(3):336–40.
8. Scorcia V, Busin M, Lucisano A, et al. Anterior segment 24. Busin M, Bovone C, Scorcia V, et al. Ultrastructural
optical coherence tomography-guided big- bubble alterations of grafted corneal buttons: the anatomic
technique. Ophthalmology. 2013;120(3):471–6. basis for stromal peeling along a natural plane of sep-
9. Busin M, Leon P, Nahum Y, et al. Large (9 mm) deep aration. Am J Ophthalmol. 2021;231:144–53.
anterior lamellar keratoplasty with clearance of a 25. Myerscough J, Bovone C, Scorcia V, et al. Deep
6-mm optical zone optimizes outcomes of keratoco- trephination allows high rates of successful pneumatic
nus surgery. Ophthalmology. 2017;124(7):1072–80. dissection for DALK independent of surgical experi-
10. Scorcia V, Giannaccare G, Lucisano A, et al. Predictors ence. Cornea. 2019;38(5):645–7.
of bubble formation and type obtained with pneumatic 26. Myerscough J, Roberts H, Yu AC, et al. Five-year
dissection during deep anterior lamellar keratoplasty outcomes of converted mushroom keratoplasty
in keratoconus. Am J Ophthalmol. 2020;212:127–33. from intended deep anterior lamellar keratoplasty
(DALK) mandate 9-mm diameter DALK as the 41. Ardjomand N, Hau S, McAlister JC, et al. Quality of
optimal approach to keratoconus. Am J Ophthalmol. vision and graft thickness in deep anterior lamellar
2020;220:9–18. and penetrating corneal allografts. Am J Ophthalmol.
27. Bovone C, Myerscough J, Friehmann A, et al. 2007;143:228–35.
Peripheral intrastromal hydration facilitates safe, deep 42. Han DC, Mehta JS, Por YM, et al. Comparison of
trephination in corneas of irregular thickness. Cornea. outcomes of lamellar keratoplasty and penetrat-
2020;39(2):207–9. ing keratoplasty in keratoconus. Am J Ophthalmol.
28. Parthasarathy A, Por YM, Tan DTH. Using a “small 2009;148:744–51.
bubble technique” to aid in success in Anwar's “big 43. Arundhati A, Chew MC, Mehta JS, et al. Comparative
bubble technique” of deep lamellar keratoplasty study of long-term graft survival between penetrating
with complete baring of descemet’s membrane. Br J keratoplasty and deep anterior lamellar keratoplasty.
Ophthalmol. 2008;92(3):422. Am J Ophthalmol. 2021;224:207–16.
29. Dua HS, Faraj LA, Kenawy MB. Dynamics of big 44. Borderie VM, Sandali O, Bullet J, et al. Long-term
bubble formation in deep anterior lamellar kerato- results of deep anterior lamellar versus penetrating
plasty by the big bubble technique: in vitro studies. keratoplasty. Ophthalmology. 2012;119(2):249–55.
Acta Ophthalmol. 2018;96(1):69–76. 45. Huang T, Hu Y, Gui M, et al. Large-diameter deep
30. AlTaan SL, Mohammed I, Said DG. Air pressure anterior lamellar keratoplasty for keratoconus:
changes in the creation and bursting of the type-1 visual and refractive outcomes. Br J Ophthalmol.
big bubble in deep anterior lamellar keratoplasty: an 2015;99:1196–200.
ex vivo study. Eye (Lond). 2018;32(1):145–51. 46. Yu AC, Franco E, Caruso L, et al. Ten-year outcomes
31. Shehadeh-Mashor R, Chan CC, Bahar I, et al. of microkeratome-assisted lamellar keratoplasty
Comparison between femtosecond laser mushroom for keratoconus. Br J Ophthalmol. 2021;105(12):
configuration and manual trephine straight-edge con- 1651–5.
figuration deep anterior lamellar keratoplasty. Br J 47. Yu AC, Spena R, Fusco F, et al. Long-term outcomes
Ophthalmol. 2014;98:35–9. of two-piece mushroom Keratoplasty for traumatic
32. Gadhvi KA, Romano V, Cueto LFV, et al. corneal scars. Am J Ophthalmol. 2022;236:20–31.
Femtosecond laser-assisted deep anterior lamellar 48. Kubaloglu A, Coskun E, Sari ES, et al. Comparison of
keratoplasty for keratoconus: multi-surgeon results. astigmatic keratotomy results in deep anterior lamel-
Am J Ophthalmol. 2020;220:191–202. lar keratoplasty and penetrating keratoplasty in kera-
33. Huang OS, Htoon HM, Chan AM, et al. Incidence and toconus. Am J Ophthalmol. 2011;151:637–43.
outcomes of intraoperative Descemet membrane per- 49. Elkadim M, Myerscough J, Bovone C, et al. A novel
forations during deep anterior lamellar keratoplasty. blunt dissection technique to treat modified deep ante-
Am J Ophthalmol. 2019;199:9–18. rior lamellar keratoplasty (DALK)-associated high
34. Myerscough J, Friehmann A, Bovone C, et al. astigmatism. Eye (Lond). 2020;34(8):1432–7.
Management of type 2 bubble formed during big bub- 50. Scorcia V, Lucisano A, Savoca V, et al. Deep anterior
ble deep anterior lamellar keratoplasty (letter to the lamellar keratoplasty followed by toric lens implan-
editor). Cornea. 2019;38:e20. tation for the treatment of concomitant anterior stro-
35. Riss S, Heindl LM, Bachmann BO, et al. Microbubble mal diseases and cataract. Ophthalmol Point Care.
incision as a new rescue technique for big-bubble 2017;1:oapoc.0000008.
deep anterior lamellar keratoplasty with failed bubble 51. Pellegrini M, Furiosi L, Yu AC, et al. Outcomes of
formation. Cornea. 2013;32(2):125–9. cataract surgery with toric intraocular lens implan-
36. Scorcia V, De Luca V, Lucisano A, et al. Results of tation after keratoplasty. J Cataract Refract Surg.
viscobubble deep anterior lamellar keratoplasty after 2022;48(2):157–61.
failure of pneumatic dissection. Br J Ophthalmol. 52. Pellegrini M, Furiosi L, Salgari N, et al. Accuracy of
2018;102(9):1288–92. intraocular lens power calculation for cataract surgery
37. Scorcia V, De Luca V, Lucisano A, et al. Comparison after deep anterior lamellar keratoplasty. Clin Exp
of corneal densitometry between big-bubble and Ophthalmol. 2022;50(1):17–22.
visco-bubble deep anterior lamellar keratoplasty. Br J 53. Pellegrini M, Scorcia V, Giannaccare G, et al. Corneal
Ophthalmol. 2019;104:336–40. neovascularisation following deep anterior lamellar
38. Cornea Donor Study Investigator Group. The effect keratoplasty for corneal ectasia: incidence, timing and
of donor age on corneal transplantation outcome risk factors. Br J Ophthalmol. 2021;106:1363. https://
results of the cornea donor study. Ophthalmology. doi.org/10.1136/bjophthalmol-2021-319339.
2008;118(4):620–6. 54. De Benito-Llopis L, Mehta JS, Angunawela RI, et al.
39. Shimazaki J, Shimmura S, Ishioka M, et al. Intraoperative anterior segment optical coherence
Randomized clinical trial of deep lamellar kerato- tomography: a novel assessment tool during deep
plasty vs penetrating keratoplasty. Am J Ophthalmol. anterior lamellar keratoplasty. Am J Ophthalmol.
2002;134:159–65. 2014;157(2):334–41.
40. Funnell CL, Ball J, Noble BA. Comparative cohort 55. Steven P, Le Blanc C, Lankenau E, et al. Optimising
study of the outcomes of deep lamellar keratoplasty deep anterior lamellar keratoplasty (DALK) using
and penetrating keratoplasty for keratoconus. Eye intraoperative online optical coherence tomography
(Lond). 2006;20:527–32. (iOCT). Br J Ophthalmol. 2014;98(7):900–4.
56. Santorum P, Yu AC, Bertelli E, Busin M. Microscope- femtosecond laser-assisted deep anterior lamel-
integrated intraoperative optical coherence lar keratoplasty for keratoconus. Br J Ophthalmol.
tomography-guided big-bubble deep anterior lamellar 2019;103(12):1716–23.
keratoplasty. Cornea. 2022;41(1):125–9. 61. Bleriot A, Martin E, Lebranchu P, et al. Comparison of
57. Farid M, Steinert RF. Deep anterior lamellar kerato- 12-month anatomic and functional results between Z6
plasty performed with the femtosecond laser zigzag femtosecond laser-assisted and manual trephination
incision for the treatment of stromal corneal pathol- in deep anterior lamellar keratoplasty for advanced
ogy and ectatic disease. J Cataract Refract Surg. keratoconus. J Fr Ophthalmol. 2017;40(6):e193–200.
2009;35(5):809–13. 62. Alio JL, Abdelghany AA, Barraquer R, et al.
58. Lucisano A, Giannaccare G, Pellegrini M, et al. Femtosecond laser assisted deep anterior lamel-
Preliminary results of a novel standardized tech- lar keratoplasty outcomes and healing patterns
nique of femtosecond laser-assisted deep ante- compared to manual technique. Biomed Res Int.
rior lamellar Keratoplasty for keratoconus. J 2015;2015:397891.
Ophthalmol. 2020;2020:5496162. https://doi. 63. Li S, Wang T, Bian J, et al. Precisely controlled
org/10.1155/2020/5496162. side cut in femtosecond laser-assisted deep lamel-
59. Shehadeh-Mashor R, Clan C, Yeung SN, et al. lar keratoplasty for advanced keratoconus. Cornea.
Long-term outcomes of femtosecond laser-assisted 2016;35(10):1289–94.
mushroom configuration deep anterior lamellar kera- 64. Yu AC, Friehmann A, Myerscough J, et al. Initial
toplasty. Cornea. 2013;32(4):390–5. high-dose prophylaxis and extended taper for mush-
60. Salouti R, Zamani M, Ghoreyshi M, et al. room keratoplasty in vascularized herpetic scars. Am
Comparison between manual trephination versus J Ophthalmol. 2020;217:212–23.
Regenerative Surgery
of the Corneal Stroma
24
for Advanced Keratoconus
Mona El Zarif, Jorge L. Alió del Barrio,

and Jorge L. Alió
Key Ponits • Advanced stem cell therapy of the corneal

• Previous studies have demonstrated, in the ani- stroma, with implantation of autologous
mal model, the capacity of human stem cells ADASCs with or without decellularized
(implanted within the corneal stroma) to alle- human corneal stroma, showed good prelimi-
viate corneal scars, improve corneal transpar- nary results for the treatment of advanced
ency, generate new organized collagen within keratoconus in the first clinical trial recently
the corneal host stroma, and with immunosup- published.
pressive and immunomodulatory properties.
• Autologous extraocular stem cells do not
require a healthy contralateral eye and they do Introduction
not involve any ophthalmic procedure for their
isolation. Mesenchymal stem cells have been Keratoconus is the most common corneal dystro-
the most widely assessed and have a high phy with a diverse prevalence in the population
potential to differentiate into functional adult from 0.05 to 2.3%. Being a relatively prevalent
keratocytes in vivo and in vitro. disease, it is more observed today than before due
to the more advanced diagnostic tools that are
available for the diagnosis of early keratoconus
Supplementary Information The online version con- [1]. Moreover, a recent meta-analysis found that
tains supplementary material available at https://doi. keratoconus is prevalent in 1.38 per 1000 of the
org/10.1007/978-3-031-32408-6_24.
world’s population (95% CI: 1.14–1.62 per 1000)
[2]. It is characterized by progressive thinning,
bulging, and distortion of the cornea and causes
progressive changes in vision with increased
myopia and myopic astigmatism, corneal irregu-
M. El Zarif larity, and visual loss [3].
Optica General, Saida, Lebanon The corneal stroma constitutes more than 90%
Division of Ophthalmology, Universidad Miguel of the corneal thickness. Many features of the cor-
Hernández, Alicante, Spain nea including its strength, morphology, and trans-
Lebanese University, Doctoral School of Sciences parency are attributable to the anatomy and
and Technology, Beirut, Lebanon properties of the corneal stroma [4]. The extracel-
J. L. A. del Barrio · J. L. Alió (*) lular matrix of the corneal stroma is composed of
Vissum Miranza, Miguel Hernández University, collagen, which forms more than 70% of the
Alicante, Spain
e-mail: [email protected] weight of the dehydrated cornea, the most abun-
https://doi.org/10.1007/978-3-031-32408-6_24
342 M. El Zarif et al.
dant being type I (75%), and proteoglycans ent alternative therapies, such as collagen cross-
including keratan sulfate which is the most abun- linking (CXL), intracorneal rings and segments,
dant (65%) whose protein nucleus is composed of corneal transplantation, and more recently,
lumican, keratocan, and mimecan [4]. Keratocan Bowman’s membrane (BM) implantation [3].
is expressed only in the corneal stroma; therefore, Meanwhile, advanced corneal ectasias require
it is considered in tissue engineering as a specific penetrating or lamellar corneal transplantation
marker of keratocytic differentiation [5]. The cel- techniques to enhance visual rehabilitation,
lular component of the corneal stroma occupies which presents several drawbacks such as fail-
only 2–3% of the stromal volume, and in it, the ure, graft rejection, and slow visual recovery due
largest cells are keratocytes, which are distributed to high levels of induced postoperative astigma-
among the collagen lamellae, mesenchymal cells tism related to the sutures [3]. Also, it should be
derived from the neural crest. In the normal cor- considered that in several countries, access to
neal keratocytes are in a quiescent state, and they donor corneal tissue is limited, approximately
are responsible for the continuous replacement of 53% of the world’s population has no access to
the stromal extracellular matrix through the pro- corneal transplantation [13]. Therefore, the
duction of collagen, which is very essential for the demand for adequate donor corneas is increasing
maintenance of corneal transparency. When kera- faster than the number of donors, leaving thou-
tocytes have an activated metabolism, they trans- sands of curable patients around the world wait-
differentiate into fibroblasts and myofibroblasts ing for possible treatment [14, 15]. The
and participate in the corneal stroma’s healing. quantification of the great shortage of corneal
The renewal capacity of stromal keratocytes is graft tissue showed that a cornea is only avail-
due to precursor cells in the anterior limbal cor- able for 70 cases needed [13].
neal stroma, which expresses adult stem cell To solve the global health problem, recent
markers [6]. research studies have focused on developing in
Keratoconus is characterized by a progressive the laboratory corneal substitutes that could
loss of keratocytes: their number decreases from mimic human cornea features in vivo, and subse-
anterior to posterior stroma [7], leading to the quently could be a substitute to human donor tis-
progressive thinning of the stroma [1, 7] and a sue, to find an alternative to classical corneal
decrease in corneal strength [8]. This definition is transplantation, but this has not yet been achieved
valid for most patients with keratoconus, although due to the extreme difficulty of mimicking the
some variations in the phenotypic expression of ultrastructure of the highly complex corneal
the disease might be present [9]. Most keratoco- stroma, obtaining substitutes that no achieve suf-
nus cases have thin corneas and a weak mechani- ficient transparency or resistance [16, 17].
cal resistance related to the progressive loss of Furthermore, synthetic scaffolds have raised
keratocyte density [10]. Apoptosis of keratocytes some important concerns, such as strong inflam-
[7, 11] or enzymes is thought to be the cause of matory responses induced in their biodegrada-
keratocyte loss and consequently loss of corneal tion, or non-specific chronic inflammatory
stroma over time [7, 11]. The proportion of the responses [18]. On the other hand, several cor-
corneal keratocytes is decreased with the pro- neal decellularization techniques have recently
gression of the disease [7]. In the end stages of been performed, which provide an acellular cor-
keratoconus, the clinical aspects of the thin and neal matrix (ECM) [19]. These scaffolds have
debilitated cornea are associated with a sharp gained great interest as they provide an ideal
decrease in the number of keratocytes. A severe natural environment for cell growth and differen-
corneal deformation is observed [7], and an alter- tiation (either transplanted donor cells or migra-
ation in the location of the corneal apex is tory host cells) [20]. Also, components of the
obtained [12], causing a severe visual loss. ECM are generally preserved among species,
The prevalence and progressive character of and the removal of all immunogenic cellular
keratoconus have led to the suggestion of differ- components could open the field of xenotrans-
24 Regenerative Surgery of the Corneal Stroma for Advanced Keratoconus 343
plantation to human recipients by using porcine [33]. It has also been newly reported that MSCs
donor tissue, which shares important similarities secrete paracrine factors such as vascular endo-
with the human cornea [21]. thelial growth factor (VEGF), platelet-derived
In the last few years, cellular therapy of the growth factor (PDGF), hepatocyte growth factor
corneal stroma has been gaining interest as a (HGF), and transforming growth factor-beta 1
potential alternative treatment option for corneal (TGFβ1). Although the precise actions of the dif-
stroma diseases such as corneal scarring, dystro- ferent growth factors for cornea wound healing
phies, and ectasias. Such diseases induce distor- are not fully understood, overall, they seem to
tion of the anatomy and physiology of the cornea promote cell migration, keratocyte survival by
and lead to loss of its transparency and subse- apoptosis inhibition, and upregulate the expres-
quent loss of vision. Using mesenchymal stem sion of the ECM component genes in kerato-
cells (MSCs) from either ocular or extraocular cytes, subsequently enhancing corneal
sources has gained a lot of importance; studies re-epithelialization and stromal wound healing
showed that MSCs are capable of differentiating [34]. MSCs can be acquired from many human
into adult keratocytes in vitro and in vivo [4]. tissues, including adipose tissue, umbilical cord,
Numerous authors, including reports from our placenta, bone marrow, dental pulp, gingiva, hair
research group, have proved [18, 20, 22] that follicle, and cornea [35, 36]:
these stem cells can not only survive and differ-
entiate into adult human keratocytes in xenoge- 1. Corneal stromal stem cells (CSSCs) are a
neic scenarios without inducing an inflammatory promising source for cellular therapy as the
reaction but also produce new collagen within isolation technique and culture methods have
the corneal stroma [22, 23], and modulated the been optimized and refined [37]; presumably,
preexisting scars [24, 25], and improving the they should be efficient in differentiating into
corneal transparency in animal models [26–29]. keratocytes as they are already committed to
MSCs have also shown immunomodulatory the corneal lineage. On the other hand, isolat-
properties in syngeneic, allogeneic, and even ing CSSCs autologously is more technically
xenogeneic scenarios [29, 30]. Early clinical demanding considering the small amount of
data on the safety and preliminary efficacy of tissue that they are obtained from.
corneal stromal cell therapy from Phase 1 human Furthermore, this technique still requires a
clinical trials are now available for up to 3-year contralateral healthy eye, which is not always
outcomes [31, 32], which may soon provide a available (bilateral disease). Therefore, these
real alternative treatment option for corneal drawbacks may limit its use in clinical prac-
diseases. tice. Allogeneic CSSC use requires living or
Considering existing scientific evidence, it cadaveric donor corneal tissue.
seems that all types of MSCs have similar behav- 2. Human adult adipose tissue is a good source
ior in vivo (Table 24.1), and thus can achieve of autologous extraocular stem cells as it sat-
keratocyte differentiation and modulate the cor- isfies many requirements: easy accessibility
neal stroma with immunomodulatory properties to the tissue, high cell retrieval efficiency, and
Table 24.1 Stem cells assessed for corneal stroma regeneration: evidence of keratocyte or keratocyte-like differentia-
tion and their potential autologous application
BM-MSCs CSSCs UMSCs ESCs ADASCs iPSCs
Keratocyte differentiation in vitro demonstrated Yes Yes Yes Yes Yes Yes
Keratocyte differentiation in vivo demonstrated Yes Yes Yes No Yes No
Possible autologous use Yes Yes/no Yes/no No Yes Yes
BM-MSCs bone marrow mesenchymal stem cells, CSSCs corneal stromal stem cells, MSCs mesenchymal stem cells,
UMSCs umbilical MSCs, (ESCs) embryonic stem cells, ADASCs adipose-derived adult stem cells, iPSCs, induced
pluripotent stem cells
the ability of its human adipose-derived adult interest; studies show that MSCs are capable of
stem cells (h-ADASCs) to differentiate into differentiating into adult keratocytes in vitro and
multiple cell types (keratocytes, osteoblasts, in vivo [4]. Several authors, including reports
chondroblasts, myoblasts, hepatocytes, neu- from our research group, have demonstrated [18,
rons, etc) [22]. This cellular differentiation 20, 22] that these stem cells can not only survive
occurs due to the effect of very specific stimu- and differentiate into adult human keratocytes in
lating factors or environments for each cell xenogeneic scenarios without inducing an inflam-
type, avoiding the mix of multiple kinds of matory reaction but also: (1) produce new colla-
cells in different niches. gen within the host stroma [22, 23], (2) modulate
3. Bone marrow MSCs (BM-MSCs) are the preexisting scars by corneal stroma remodeling
most widely studied MSCs, presenting a sim- [24, 25], and (3) improve corneal transparency in
ilar profile to adipose-derived adult stem cells animal models for corneal dystrophies by colla-
(ADASCs), but their extraction requires a gen reorganization as well as in animal models
bone marrow puncture, which is a compli- for metabolopathies by the catabolism of accu-
cated and painful procedure requiring general mulated proteins [26–29].
anesthesia. The use of autologous human keratocytes in
4. Umbilical MSCs (UMSCs) present an attrac- cell therapy of the corneal stroma is a promising
tive alternative, but their autologous use is therapeutic approach, but it has many disadvan-
currently limited as the umbilical cord is not tages, such as causing damage to the donor’s cor-
generally stored after birth. nea, insufficient cells, and inefficient cell
5. Embryonic stem cells (ESCs) have great subculture [43]. On the other hand, based on pre-
potential, but also present important ethical vious successful animal studies performed in part
issues. However, the use of induced pluripo- by our team, it has been investigated an extraocu-
tent stem cells (iPSCs) technology [38] could lar source of abundant and more accessible cells
solve such problems, and their capability to for this purpose [22, 23]. The adipose tissue has
generate adult keratocytes has already been shown to be an ideal source of autologous stem
proven in vitro [39]. cells, known as “human adipose-derived adult
stem cells” (h-ADASCs), that can differentiate
Finally, it is important to remark that the thera- into different cell lineages [22, 43]. Moreover,
peutic effect of MSCs in a damaged tissue is not these cells have shown immunomodulatory prop-
always related to the potential differentiation of erties even in xenogeneic scenarios [18, 20]. We
the MSCs in the host tissue as multiple mecha- found that human ADASCs transplanted into
nisms might contribute simultaneously to this damaged rabbit corneas were able to differentiate
therapeutic action, for example, secretion of in corneal keratocytes and produce corneal col-
paracrine trophic, and growth factors capable of lagens and keratocan that are representative of
stimulating resident stem cells, reduction of tis- the human corneal stroma [22]. Besides, the cor-
sue injury and activation of immunomodulatory neal decellularized matrices provide a more natu-
effects, in which case the direct cellular differen- ral environment for the growth and differentiation
tiation of the MSCs might not be relevant and of cells compared to synthetic scaffolds, it has
could even be non-existent [33, 40–42]. been demonstrated the efficiency of Sodium
dodecyl sulfate (SDS) decellularization on the
human cornea. In our study, we used decellular-
Background: Translational ized human corneal laminas alone, or repopu-
Regenerative Surgery lated by h-ADASCs, which have been shown
excellent results during the follow-up, the cor-
In the last few years, interest in cellular therapy neal transparency has been completely preserved
of the corneal stroma using MSCs from either without any signs of scarring [20]. We have been
ocular or extraocular sources has gained a lot of able to demonstrate that the h-ADASCs trans-
planted into a human decellularized lamina sur- the implantation of decellularized/recellularized

vives at least 12 weeks after transplantation laminas could restore the corneal thickness and
in vivo in the animal model, and they also differ- the keratometric parameters (Video 24.2).
entiate into human keratocytes, Therefore, with Nevertheless, the direct injection of stem cells
this model, we could potentially be able to obtain may provide a promising treatment modality for
an autologous graft using adipose tissue from the corneal dystrophies, corneal stroma opacifica-
patient and an allogeneic donor cornea, theoreti- tion, and the modulation of corneal scarring.
cally avoiding the risk of stromal rejection asso-
ciated with allogenic lamellar transplant options
[20]. Also, the advantage of our protocol is the Study Design, and Subjects
possibility of obtaining several grafts from a
unique allogeneic donor, increasing the availabil- This investigation was a prospective interven-
ity of donor tissue, and shortening waiting lists. tional randomized, nonmasked consecutive series
The goal of this research line is to translate the of cases. The study was conducted in strict adher-
research done by our group and apply it to human ence to the tenets of the Declaration of Helsinki
patients suffering from keratoconus. A simple and it was registered in ClinicalTrials.gov (Code:
procedure of liposuction and a nontransplantable NCT02932852).
donor cornea can provide an optically transparent Fourteen patients were enrolled in the study and
autologous stromal graft, with excellent demon- were randomly distributed into three study groups:
strated biocompatibility. With the new noninva- Group 1 (G-1) patients received ADASCs implan-
sive surgical technique, many complications tation (n = 5 patients); group 2 (G-2) received
associated with the usual techniques can be decellularized human corneal stroma implantation
avoided, improving the visual parameters and the (n = 5 patients); and group 3 (G-3) sreceived autol-
quality of life of patients. ogous ADASCs-recellularized human corneal
These experimental studies opened the trans- stroma implantation (n = 4 patients).
lational of this concept into the therapy of human Thirteen patients completed the clinical fol-
corneal diseases, using the advanced keratoconus low-up, only one patient from G-1 was lost after
disease as the model for this type of advanced the first postoperative month because of inability
therapy. This study aims to build a stem cell ther- to attend further follow-up for motives unrelated
apy alternative to the classic corneal transplanta- to the study.
tion techniques to regenerate the corneal stroma, Inclusion and exclusion criteria were defined
avoiding the complications and limitations com- in previous publications [31, 32, 44, 45]. As well
monly observed with existing techniques. clinical follow-up of the patients was established
at 1, 3, 6, 12, and 36 months for the clinical out-
comes of the investigation and to survey implant
linical Human Surgery
C safety for a long time.
in Advanced Keratoconus
Recently, our group performed the implantation Methodology

of ADASCs and decellularized/recellularized
laminas in 14 patients with advanced keratoconus utologous ADASC Isolation,
A
(Video 24.1). This clinical experience opens a Characterization, and Culture
new and exciting line of research therapy. The Patients underwent standard liposuction.
production of new ECM by the implanted MSCs Approximately 250 mL of fat mixed with local
was demonstrated in previous animal studies, anesthesia were obtained from each patient. The
although was not quantitatively enough to be able adipose tissue processing was performed accord-
to restore the thickness of a severely diseased ing to the procedures described in the previous
human cornea such as keratoconus. Meanwhile, articles [22, 31, 46–48].
Laminas mal pocket, the lamina was inserted, centered,

The human corneal stroma of donor corneas with and unfolded through gentle tapping and massag-
negative viral serology, but with nonviable endo- ing from the epithelial surface of the host. Also
thelium was used. The corneas were provided by before implantation, a temporary limbal paracen-
the eye bank. The quality and safety standards for tesis was performed to reduce intraocular pres-
donation, procurement, testing, processing, con- sure. In those cases, which received a
servation, storage, and testing of human cells and recellularized lamina (G-3), to compensate for
tissues were followed. Donor corneas were dis- the cellular damage expected by the implantation
sected with IntraLase iFS femtosecond laser process, the pocket was irrigated immediately
(AMO, Santa Ana, CA), 2–3 consecutive laminas before and after insertion with a solution contain-
of 120 (μm) thick, and 9.0 mm in diameter were ing an additional one million autologous ADASc
obtained. The decellularization protocol was in 1 mL of PBS with a 25G cannula. The incision
based on previous publications [19, 20, 32, 49]. was then closed with an interrupted 10/0 nylon
The recellularized tissue was placed 24 h before suture [19, 32, 44, 45].
implantation in tissue culture wells for recellular-
ization with autologous ADASCs (0.5 × 106 cells ostoperative Care and Follow-Up
P
per 1 mL of PBS were cultured on each side of Schedule
the laminas). Then the laminas were immersed in Postoperatively, the patients were followed at
PBS at room temperature and transported to 1 day, 1 week, and at 1, 3, 6, 12, and 36 months
implantation [32, 44, 45]. for the evaluation of clinical parameters: unaided
distance visual acuity (UDVA), corrected dis-
Surgical Procedure tance visual acuity (CDVA), rigid contact lens
distance visual acuity (CLDVA) in (decimal
Autologous ADASC Implantation equivalent to the logMar scale), refractive sphere
The method for the implantation of ADASCs has (Rx Sphr) in diopters (D), and refractive cylinder
been previously described [30]. Topical anesthe- (Rx Cyl) (D). Central corneal thickness (CCT)
sia was used. 60 kHz IntraLase iFS femtosecond (μm) was measured by (AS-OCT) (Visante, Carl
laser (AMO Inc., Irvine, CA) was used in a Zeiss). Scheimpflug corneal topography thinnest
single-pass mode for the recipient corneal lamel- point (Thinnest point) (μm), cornea volume (CV)
lar dissection. An intrastromal laminar cut of (mm3), corneal aberrometry, anterior mean kera-
9.5 mm in diameter was created at a medium tometry (anterior Km) (D), posterior mean kera-
depth of the thinnest preoperative pachymetry tometry (posterior Km) (D), maximum
point measured by the Visant OCT (Carl Zeiss, keratometry (Kmax) (D), topographic cylinder
Germany). Three million autologous ADASCs (Topo Cyl) (D), and corneal densitometry (CD)
contained in 1 mL PBS were injected into the (Pentacam; Oculus Inc., Wetzlar, Germany). Slit-
pocket. lamp biomicroscopy, fundoscopy, intraocular
pressure (Goldmann applanation tonometry IOP)
Lenticule Implantation (mmHg), and endothelial cell density (ECD)
Topical anesthesia was applied with oral sedation (cells/mm2) by specular microscopy (Nidek,
for all surgeries, and the 60-kHz IntraLase iFS Aichi, Japan). The confocal microscopy study
femtosecond laser was used in single-pass mode. was completed up to 12 months using the confo-
Assisted corneal dissection was done with a 50° cal microscope HRT3 RCM (Heidelberg) with
anterior cut. After opening the corneal intrastro- Rostock Cornea Module.
Results show any posterior stromal or predescemetic

scars and presented a fully transparent visual axis
lit Lamp Biomicroscopy and AS-OCT
S (Figs. 24.1, 24.2, 24.3, and 24.4). Full corneal
Results transparency was recovered within the first post-
operative day in all patients of G-1 (Fig. 24.1a, b)
No complications were observed during the At the level of the stromal pocket, a neo-collagen
3-year follow-up. No adverse events, such as haze production was observed as patchy hyperreflec-
or infection, were obtained. All corneas did not tive areas (Fig. 24.2a, b) [31, 44, 45]. Meanwhile,
Fig. 24.1 Biomicroscopic changes among the preopera- and 36 months (right). (c) Confocal microscopy findings
tive and up to 36 months postoperative, confocal micros- in the same Case-2: notice the high reflective deposits and
copy findings till 12 months postoperative in G-1, case-2. fibrotic tissue in the anterior stroma of the corneal at the
(a) Observe the presence of paracentral scars (blue preoperative level (blue arrows; left), that corresponded to
arrows) at the preoperative level (left) and 1 month post- the paracentral scars. At 12 months (right), an improve-
operative (right). (b) Notice the marked improvement of ment of the anterior stroma fibrotic tissue could be noticed
the paracentral scars (blue arrows), at 12 months (left), (blue arrows)
a b
c d
e f
Fig. 24.2 Corneal anterior segment OCT sections and (d) Same patient after 36 months: observe the improve-
pachymetric maps (Visante) in G-1, G-2 & G-3. (a) G-1, ment in the integration of the implanted lamina in the host
(case-2) at 1 month postoperative: observe the reflective stoma, and the stability of the pachymetric map. (e) G-3,
paracentral scar (blue arrow). (b) The same patient after (case-11) at 1 month: observe the high reflectance of the
36 months: shows low reflectance of the band of neo- implanted lamina (yellow arrows). (f) Same patient as (e)
collagen (yellow arrow), the reflective paracentral scar has at 36 months: notice the integration of the implanted tis-
disappeared (bleu arrow). (c) G-2, (case-6) at 6 months: sue in the host stroma and the enhancement in corneal
observe the high reflectance of the implanted lamina (yel- densitometry
low arrows) and the restoration of the corneal thickness.
the implanted laminas in G-2, and G-3 showed a edema. Corneal full transparency was observed
mild early haziness during the first postoperative within the third postoperative month in all patients
month. This issue was related to mild lenticular (Figs. 24.2c–f, 24.3a, b, 24.4a, b) [32, 44, 45].
Fig. 24.3 Biomicroscopic changes between the preoper- arrow; right). (c) Confocal microscopy finding at 1 month
ative and up to 36 months postoperative, confocal micros- (left) we can observe the acellular anterior surface of the
copy findings till 12 months post-op in G-2, case-5. (a) At decellularized lamina. After 12 months (right), we can
1 day postoperative (left and right): reduced transparency notice in the periphery of the lamina migrating kerato-
due to edema of the implanted lamina. (Red arrows) show cytes nuclei from the host stroma toward the posterior sur-
the borders of the lamina. (b) Improvement in the trans- face of the decellularized lamina (yellow arrow), and the
parency of the implanted tissue at 36 months (red arrows). presence of some paracentral fibrotic tissue (green arrow)
Paracentral fibrotic tissue at the surgical plane (green
Fig. 24.4 Biomicroscopy changes in G-3, case-11 from (yellow arrow) an accumulation of migrating keratocytes
preoperative until 36 months postoperative. (a) on the periphery of the posterior surface of the recellular-
Preoperative. (Blue arrows) shows the presence of a para- ized lamina is noticed (left), while the (green arrow)
central scar (right, and left). (b) Observe the improvement shows the presence of some fibrotic tissue. 12 months
of the paracentral opacification at 36 months (blue later (left), a highly reflective fibrotic tissue (green arrow)
arrows). (Green arrows) shows the presence of fibrotic in the anterior surface of the recellularized lamina is
anterior tissue. (c) Confocal microscopy in G-3 case-11: observed
Visual Parameter Results acuity (UDVA) results for all groups presented
an improvement in mean values ranging
No patient lost lines of visual acuity. All cases between [0.08 and 0.14] (Fig. 24.5a;
presented an improvement in their visuals Table 24.2). Also, corrected distance visual
parameters measured in decimal equivalent to acuity (CDVA) showed an enhancement in
LogMar scale. The unaided distance visual mean values results for all groups ranging
a
0.35
0.33
Unaided distance visual acuity
0.31
0.30
0.27 0.26
0.24
0.25 0.23 0.26
(decimal)
0.20 0.21 0.24

0.19
0.20
0.17 0.21
0.15 0.14 0.170432306

0.13
0.14
0.13
0.10
0 6 12 18 24 30 36
Time (months)
b 0.75
Corrected distance visual acuity
0.60 0.58
0.50
0.51 0.41 0.40
(decimal)
0.46
0.33
0.40 0.42 0.42 0.40
0.27
0.23 0.33
0.250.21 0.28
0.22 0.24
0.00
0 6 12 18 24 30 36
Time (months)
c 1.00
0.84
Right contact lens distance
0.80
visual acuity (decimal)
0.75
0.75 0.72 0.70
0.67
0.610.63 0.62
0.58
0.52 0.54 0.62
0.58
0.50
0.50
0.14 0.45
0.13 0.41
0.390.14
0.25
0 6 12 18 24 30 36
Time (months)
Group 1 Group 2 Group 3
Fig. 24.5 Visual outcomes in G-1, G-2, and G-3 up to ing was obtained in mean values in G-2 & G-3 compared
36 months after surgery in (decimal equivalent to LogMar to the G-1. (c) Rigid contact lens visual acuity. A statisti-
scale). (a) Unaided distance visual acuity. (b) Corrected cally significant worsening was obtained in G-2 when
distance visual acuity. A statistically significant worsen- compared to G-1 & G-3
between [0.11 and 0.2] (Fig. 24.5b; Table 24.2), volume (CV) (mm3) (Fig. 24.6b, c, 24.7, 24.8,
and rigid contact lens visual acuity (CLDVA) 24.9) presented an increase in mean values in all
improvement results were ranged between patients of all the different groups of 30 μm,
[0.1–0.23] (Fig. 24.5c; Table 24.2) at 36 months 31 μm, and 2–3 mm3, respectively, at 36 months
of follow-up [45]. of follow-up [45]. The authors found a statistical
Results of central corneal thickness (CCT) significance difference improvement at
(μm) measured by AS-OCT (Fig. 24.2, 24.6a; 36 months in CCT, thinnest point, and CV
Table 24.2), Scheimpflug corneal topography when comparing the mean values among G-2/
thinnest point (Thinnest point) (μm), and cornea G-1 and G-3/G-1 (Table 24.2) [45].
Table 24.2 Difference in Mean Values of all the vari-

ables of the study among (G-1)–(G-2), (G-1)–(G-3), and
(G-2)–(G-3)
(G-1)– (G-1)– (G-2)–
(G-2) (G-3) (G-3)
UDVA 0.07 0.07 0.00
CDVA 0.18a 0.19a 0.01
CLDVA 0.22a 0.10 −0.12a
Visante CCT (μm) −44.00a −77.00a −33.00
Thinnest point (μm) −51.00a −65.00a −14.00
CV (mm3) −5.00a −5.00a 0.00
Kmax (D) 0.00 2.00 2.00
Anterior km (D) −3.00 1.00 4.00
Posterior km (D) 1.3a 0.2 −1.1a
Topo Cyl (D) 1.1 0.9 −0.2
3rd order RMS 4.65a 3.54a −1.11
4th order RMS 1.28 −0.17 −1.45a
HOA RMS 4.9 2.78 −2.12
LOA RMS −3.87a 1.32a 5.19
Rx Sphr (D) 0.10 −0.10 −0.20
Rx Cyl (D) −1.00a −0.60a 0.40
a
Indicates a statistically significant difference between
the compared groups
a
Visante central corneal thickness (µm)
600
575
546
550
523 517
525 513
506
500487 494 490 484
473 469
475
454 461
450 446 440
430
425410 417
400
0 6 12 18 24 30 36
Time (months)
b 525
Scheimpflug corneal topography
498
500
478 480
Thinnest point (mm)
475 465 484

454
449 464 466
450
451 433
441
425435 414 416
400
400 390
385
375
0 6 12 18 24 30 36
Time (months)
c 65
62
61 61
Cornea volume (mm3)
62
60
60
59 59 61 61
60
59 59 57
56 56
54
55
53 54
50
0 6 12 18 24 30 36
Time (months) Group 1 Group 2 Group 3
Fig. 24.6 (a) Central corneal thickness was measured measured with Pentacam (mm3). CCT, thinnest point, and
with AS-OCT Visante (μm). (b) Scheimpflug corneal CV showed significant improvement in the mean values at
topography thinnest point (μm). (c) Cornea volume was 3-year in G-2 & G-3 when compared to G-1
a b
Fig. 24.7 Corneal topography (Pentacam) comparison ric parameters. (b) Four maps corneal topography with the
among preoperative, 12 and 36 months postoperative in same case at 36 months postoperative: no significant
G-1, case-1. (a) Preoperative versus 12 months postopera- increase of the pachymetric or keratometric parameters
tive: observe the minimal enhancement of the pachymet- was obtained
a b
Fig. 24.8 Corneal topography (Pentacam) comparison Preoperative versus 36 months: notice the keratometric
among preoperative, 12 and 36 months postoperative in improvement
G-2, case-7. (a) Preoperative versus 12 months, and (b)
a b
Fig. 24.9 Corneal topography (Pentacam) comparison erative, and (b) Preoperative versus 3-year postoperative:
between preoperative, 12 and 36 months postoperative in notice the keratometric improvement and the noticeable
G-3, case 11. (a) Preoperative versus 12 months postop- increase in the pachymetric parameters
a
60 60
Anterior mean keratometry (D)
60 59
59
59 59
58
58
57 57
56 56
56
56 56 56 56
56 55
56
55 55 58
0 6 12 18 24 30 36
Time (months)
b
70
Maximum keratometry (D)
68 68
68 68
68 68 67
68 67 66
66 66
66 66
66 65
65
64 64 65
62 63
0 6 12 18 24 30 36
Time (months)
Fig. 24.10 Keratometric outcomes after 3 years of fol- tening at 3-year. (b) Maximum keratometry (D): there was
low-up in G-1, G-2 & G-3. (a) Anterior mean keratometry a mean flattening of three diopters of flattening at 3-year
(D): notice the mean improvement of two diopters of flat-
Topographic Results Finally, the topographic cylinder (Topo Cyl)

(D) remained stable [45].
The refractive sphere (Rx Sphr) (D) pre-
sented an improvement of 1.1 myopic diop-
ters at 36 months. Meanwhile, the refractive Corneal Aberrometry Results
cylinder (Rx Cyl) (D) presented an increase
of 0.5D until 36 months postoperative [45]. As well, an improvement in mean values was
Also, the authors detected at 36 months post- obtained at 36 months follow-up in third-order
operative a modest improvement of 2D in the aberration RMS (third-order RMS) (μm),
anterior mean keratometry values (anterior fourth-order aberration RMS (fourth- order
Km) (D) (Fig. 24.10a), and stability in mean RMS) (μm), high-order aberration RMS (HOA
values of posterior mean keratometry (poste- RMS) (μm) (Fig. 24.11), and low-order aberra-
rior Km) (D). However, they found a flatten- tion RMS (LOA RMS) [45]. More information
ing in mean values of 3D in maximum about the comparative results among the three
keratometry (Kmax) (D) (Fig. 24.10b). groups is summarized in Table 24.2.
a b
Fig. 24.11 Corneal high-order aberration RMS (HOA RMS) (μm) comparison, notice the improvement among the
preoperative (a), and 36 months (b) in G-2 case-9
Corneal Densitometry (CD) Results Table 24.3 Average differences of corneal densitometry
among the G-1, G-2, and G-3
The authors studied the CD at the annular zones G2– G3– G3–
centered on the corneal apex (0–2 mm, 2–6 mm, G1 G1 G2
Central
and 6–10 mm), resulting in the following
0–2 mm 3.29a 0.004 6.58a 0.000 3.44a 0.002
outcomes: 2–6 mm 2.08a 0.001 3.00a 0.000 1.78a 0.000
6–10 mm 0.93a 0.019 1.18a 0.004 0.23 0.564
Anterior CD Total
In G-1, a decrease in all CD mean values was 0–2 mm 2.54a 0.030 3.59a 0.003 1.17 0.259
obtained up to 36 months regarding the preopera- 2–6 mm 1.57a 0.000 2.09a 0.000 0.70 0.010
tive values. Meanwhile, in G-2 and G-3, there 6–10 mm 0.81a 0.000 0.72a 0.000 −0.05 0.833
was an increase in the CD mean values at a
Statistically significant differences among the compared
1 month. Then, mean values reached the same groups
preoperative level (or below them) up to
36 months [50]. Posterior CD
There were slight variations in the mean values of
entral and Total CD
C the CD between preoperative and 36 months’
In G-1, the CD mean values had a slight increase postop values in all the groups [50].
at 1 month. Then, these values decreased pro-
gressively up to 36 months until they reached the
same (or below) preoperative level. However, in Confocal Microscopy Results
G-2, and G-3, the CD mean values increased
more notably than in ADASCs group at 1 month, Morphological Results
noting that the increase in G-3, was more marked The confocal microscopy resulted in morpholog-
than in G-1 and G-2. Then, the CD mean values ical findings showing that ADASCs in G-1 is
decreased progressively in G-2 and G-3 till get- more rounded, voluminous, more luminous, and
ting somehow above the preoperative mean val- refringent compared to the host keratocytes
ues (Table 24.3) [50]. (Fig. 24.12a). However, the shape of ADASCs
a b
Fig. 24.12 ADASCs morphological changes in G-1, nous than the host keratocytes (green arrows). (b) At
case-1: (a) The implanted ADASCs at 1 month have a 1 year after implantation, all corneal stromal cells assimi-
rounded shape, larger, more refringent, and more lumi- late a similar morphology to the host keratocytes
changed from round to fusiform after 6 months G-2, and G-3 a year after the surgery in com-
from implantation (Fig. 24.12b) [31, 44, 45, 51]. parison with the preoperative cell density mean
Meanwhile, the decellularized laminas appeared values. In G3, the results in cell density were the
acellular within the first few months (Fig. 24.13a), highest, followed by G-2 and then G-1. Also,
unlike the recellularized ones that showed in results in cellular densities at the mid-corneal
some determined areas similar structures to cor- stroma in G-1 showed a significant increase a
neal keratocytes (Fig. 24.13b). The number of 12 months postoperative. Similar results were
cells increased gradually during the 12 months obtained at the anterior and posterior surfaces
follow-up in the decellularized and recellularized and within the implanted laminas in G-2 and
laminas in G-2, and G-3 and became more colo- G-3 [51].
nized by keratocyte-type cells until they showed On the other hand, we detected the formation
similar morphology of normal corneal kerato- of a few fibrotic tissue areas in some cases of
cytes (Fig. 24.13c, d) [31, 44, 45, 51]. G-1, while a somehow stronger formation of
such fibrotic tissue areas was observed in almost
Statistical Results all cases of G-2 and G-3. Nevertheless, we did
The confocal microscopy statistical density not find a direct and significant association
mean values showed a gradual and statistically between the recellularization of the implanted
significant increase in the cellularity in the ante- laminas and the presence of such fibrotic tissue
rior and posterior stroma of patients in G-1, [51].
a b
c d
Fig. 24.13 Morphological changes in decellularized and same case-9, the decellularized lamina is observed with
recellularized laminas in G-2, and G-3. (a) G-2, case-9, the presence of cells assimilate to the host 1 year after the
the anterior surface of a decellularized lamina appears surgery. (d) The posterior surface of the recellularized
without cells 1 month after surgery. (b) The posterior sur- lamina with the same case-13 at 12 months postoperative,
face of the recellularized lamina in G-3, case-13 at one- notice the high number of cells similar to the host corneal
month postoperative, notice the presence of few ADASCs, stroma
similar in morphology to keratocytes-type cells. (c) In the
Discussion and Conclusions the corneal dystrophies, scars and increase

slightly the corneal thickness, but this improve-
Our research group demonstrated for the first ment is not enough to reestablish the corneal dis-
time the feasibility of the implantation of ease in advanced keratoconus [31, 44, 45].
ADASCs into the corneal stromal pocket in cases Also, we confirmed for the first time that
of advanced keratoconus, they confirmed the decellularized human corneal stromal laminas,
appearance of new collagen in the injected areas, colonized or not by autologous ADASCs, can be
this new collagen could be useful for repairing implanted for a clinical basis in the corneal
stroma for therapeutic purposes. Also, such clini- duced by the interaction between the seeded
cal application showed the safety and the feasi- ADASCs and the collagenous tissue of the lam-
bility of the use of femtosecond laser to dissect ina, obtaining in this group permanent increase in
the cornea in the middle of the corneal stroma CD values up to 36 postoperative months
with advanced keratoconus, even when a large (Table 24.3) [50].
9.5 mm corneal pocket was performed [32, 44, Confocal microscopy was a necessary tool for
45]. After 3-year postoperative, no patient the evaluation and monitoring in “in vivo” of the
showed inflammation, rejection, or any evidence evolution of the ADASCs nuclei and their mor-
of scarring or haze (Figs. 24.1, 24.2, 24.3 and phological changes, being rounded cells and
24.4). Moreover; there was an improvement with highly refractive up to 6 months, changing to
all the cases after 3-year follow-up in all the fusiform shaped structures, and less refringent
visual parameters with 1–2 lines in decimal nuclei up to 12 months. These findings demon-
equivalent to LogMar scale (Fig. 24.5). A signifi- strate in the human clinical model that the
cant increase in thinnest corneal point, in central ADASCs implanted in the corneal human pocket
corneal thickness, and corneal volume was have survived and have been able to differentiate
obtained (Fig. 24.6), and the mean value results into keratocytes (Fig. 24.12) [51, 52]. Such find-
were statistically significantly better with the ings were confirmed previously in animal studies
groups with implanted decellularized/recellular- in which the post-mortem analysis demonstrated
ized laminas with ADASCs when comparing the survival of these human cells, and their capac-
with the group of implanted ADASCs alone ity to produce human collagen in the corneal rab-
(Table 24.2). There was also an improvement in bit [18, 22]. Also, confocal microscopy allowed
all the corneal aberrations (Fig. 24.11) and stabil- to monitor the morphological changes that
ity or improvement of corneal topographic occurred in the decellularized and recellularized
parameters (Figs. 24.7, 24.8, 24.9 and 24.10) laminas, it assisted in determining the change in
[45]. the cell density in the implanted tissue, as well as
Besides, our study showed that the CD behav- in all the corneal stroma [51, 52]. The ADASCs
ior significantly differed among the different implantation increased significantly the cellular-
studied groups (Table 24.3) [50]. Clinically, in ity at 12 months (Figs. 24.12b). In addition, the
G-1, corneal transparency fully recovered within implantation of decellularized or recellularized
24 h postoperatively and was maintained through- laminas increased significantly the level of the
out the follow-up period. The changes in CD stromal cells when compared with the groups of
were obtained at the central cornea, increasing implanted ADASCs. Besides, this increase was
slightly the mean values at 1 month, this fact is larger with corneal laminas impregnated with
possibly related to the surgical intervention itself, ADASCs than that observed when acellular cor-
the implantation of ADASCs increased cell den- neal laminas are implanted (Fig. 24.13c, d) [51,
sity in the surgical interface, or with the subtle 52].
deposit of neo-collagen [50]. While, with G-2, In conclusion, cellular therapy of the corneal
and G-3 patients clinically showed an early hazi- stroma is a novel treatment modality for stromal
ness during the first postoperative month, this diseases, which even though further studies are
fact was related with a mild lenticule edema, then still mandatory with larger sample sizes to estab-
progressively improving throughout the follow- lish its safety and efficacy for different corneal
up period, demonstrating total corneal transpar- stromal diseases, the initial results obtained from
ency by the third postoperative month [32, 44, the pilot clinical trials are encouraging. Corneal
45, 50], this clinical result was directly related to thickness improvement in the corneas only with
the increase produced in central DC, presenting a ADASCs implantation seems to be insufficient to
spike between months 1 and 3, and then improv- restore normal corneal thickness, while the
ing during the 36-month follow-up. This increase implantation of corneal laminas demonstrated the
was greater in G-3 than in G-2 and may be pro- best result regarding corneal thickness restora-
tion with a significant increase in cell density in JL, et al. Frontiers in regenerative medicine for cor-
nea and ocular surface. In: Rahman A, Anjum S, edi-
the group implanted with recellularized laminas. tors. Frontiers in stem cell and regenerative medicine
The total absence of complications, the corneal research. 1st ed. Sharjah: Bentham Publisher; 2015.
transparency restoration, the reestablishment of p. 92–138.
the CD, and the modest, but significant visual and 5. Carlson EC, Liu C-Y, Chikama T, Hayashia Y, Kao
CW-C, Birk DE, et al. Keratocan, a cornea-specific
refractive enhancements obtained confirm the keratan sulfate proteoglycan, is regulated by lumican.
achievability of this therapeutic approach as a J Biol Chem. 2005;280:25541–7.
possible novel technique for the treatment of ker- 6. Du Y, Funderburgh M, Mann M, Sundar Raj N,
atoconus and other corneal dystrophies. Future Funderburgh J. Multipotent stem cells in human cor-
neal stroma. Stem Cells. 2005;23(9):1266–75.
studies with larger samples in less advanced ker- 7. Ku J, Niederer R, Patel D, Sherwin T, McGhee
atoconus cases will guarantee the therapeutic C. Laser scanning in vivo confocal analysis of ker-
potential of this new regenerative surgery. atocyte density in keratoconus. Ophthalmology.
2008;115(5):845–50.
8. Piñero DP, Alió JL, Barraquer RI, Michael R,
Take Home Notes Jiménez R. Corneal biomechanics, refraction,
and corneal aberrometry in keratoconus: an inte-
• Cellular therapy of the corneal stroma is a grated study. Invest Ophthalmol Vis Sci. 2010;51:
novel treatment modality for corneal stroma 1948–55.
9. Alió J, Piñero D, Alesón A, Teus M, Barraquer R,
diseases, which initial results obtained from a Murta J, et al. Keratoconus-integrated characteriza-
pilot clinical trial are encouraging. tion considering anterior corneal aberrations, internal
• Nevertheless, further studies with larger sam- astigmatism, and corneal biomechanics. J Cataract
ples and other treated diseases different from Refract Surg. 2011;37(3):552–68.
10. Mastropasqua L, Nubile M. Normal corneal morphol-
keratoconus are still necessary in order to sta- ogy. In: Mastropasqua L, Nubile M, editors. Confocal
blish its efficacy and safety profiles and so its microscopy of the cornea. Thorofare, NJ: SLACK;
clinical usefulness. 2002. p. 7–16.
• ADASC corneal thickness improvement 11. Ali Javadi M, Kanavi M, Mahdavi M, Yaseri M, Rabiei
H, Javadi A, et al. Comparison of keratocyte density
seems to be insufficient to restore severely between keratoconus, post-laser in situ keratomileu-
thinned corneas, while the implantation of sis keratectasia, and uncomplicated post-laser in situ
decellularized corneal stroma laminas keratomileusis cases. A confocal scan study. Cornea.
enhances such thickness restoration. 2009;28(7):774–9.
12. Edmund C. Assessment of an elastic model in the
• Enhanced corneal stromal remodeling is pathogenesis of keratoconus. Acta Ophthalmol.
observed by confocal microscopy in those 1987;65(5):545–50.
cases receiving ADASC within the corneal 13. Gain P, Jullienne R, He Z, Aldossary M, Acquart
stroma. S, Cognasse F, et al. Global survey of corneal
transplantation and eye banking. JAMA Ophthalmol.
2016;134(2):167–73.
14. Griffith M, Alarcon EI, Brunette I. Regenerative
approaches for the cornea. J Intern Med.
2016;280(3):276–86.
References 15. Fagerholm P, Lagali N, Merrett K, Jackson W,
Munger R, Liu Y, et al. A biosynthetic alternative to
1. Alio J. In: Alió JL, editor. Keratoconus: recent human donor tissue for inducing corneal regeneration:
advances in diagnosis and treatment. Cham: Springer; 24-month follow-up of a phase 1 clinical study. Sci
2017. Transl Med. 2010;2(46):46ra61.
2. Hashemi H, Heydarian S, Hooshmand E, Saatchi M, 16. Isaacson A, Swioklo S, Connon CJ. 3D bioprint-
Yekta A, Aghamirsalim M, et al. The prevalence and ing of a corneal stroma equivalent. Exp Eye Res.
risk factors for keratoconus: a systematic review and 2018;173:188–93.
meta-analysis. Cornea. 2020;39(2):263–70. 17. Ruberti J, Zieske J. Prelude to corneal tissue engineer-
3. Arnalich-Montiel F, Alió Del Barrio J, Alió J. Corneal ing—gaining control of collagen organization. Prog
surgery in keratoconus: which type, which technique, Retin Eye Res. 2008;27(5):549–77.
which outcomes? Eye Vis (Lond). 2016;3:2. 18. Alió del Barrio J, Chiesa M, Ferrer GG, Garagorri N,
4. De Miguel MP, Casaroli-Marano RP, Nieto-Nicolau Briz N, Fernandez-Delgado J, et al. Biointegration of
N, Martínez-Conesa EM, Alió del Barrio JL, Alió corneal macroporous membranes based on poly(ethyl
acrylate) copolymers in an experimental animal 33. Harkin D, Foyn L, Bray L, Sutherland A, Li F, Cronin
model. J Biomed Mater Res A. 2015;103(3):1106–18. B. Concise reviews: can mesenchymal stromal cells
19. Lynch A, Ahearne M. Strategies for developing differentiate into corneal cells? A systematic review
decellularized corneal scaffolds. Exp Eye Res. of published data. Stem Cells. 2015;33(3):785–91.
2013;108:42–7. 34. Jiang Z, Liu G, Meng F, Wang W, Hao P, Xiang Y,
20. Alio del Barrio J, Chiesa M, Garagorri N, Garcia- et al. Paracrine effects of mesenchymal stem cells
Urquia N, Fernandez-Delgado J, Bataille L, et al. on the activation of keratocytes. Br J Ophthalmol.
Acellular human corneal matrix sheets seeded with 2017;101(11):1583–90.
human adipose-derived mesenchymal stem cells inte- 35. Hendijani F. Explant culture: an advantageous method
grate functionally in an experimental animal model. for isolation of mesenchymal stem cells from human
Exp Eye Res. 2015;132:91–100. tissues. Cell Prolif. 2017;50(2):e12334.
21. Hara H, Cooper DKC. Xenotransplantation— 36. Górski B. Gingiva as a new and the most accessible
the future of corneal transplantation? Cornea. source of mesenchymal stem cells from the oral cavity
2011;30(4):371–8. to be used in regenerative therapies. Postepy Hig Med
22. Arnalich-Montiel F, Pastor S, Blazquez-Martinez Dosw (Online). 2016;70(0):858–71.
A, Fernandez-Delgado J, Nistal M, Alio J, De 37. Basu S, Hertsenberg AJ, Funderburgh ML, Burrow
Miguel M. Adipose-derived stem cells are a source MK, Mann MM, Du Y, Lathrop KL, Syed-Picard FN,
for cell therapy of the corneal stroma. Stem Cells. Adams SM, et al. Human limbal biopsy-derived stro-
2008;26(2):570–9. mal stem cells prevent corneal scarring. Sci Transl
23. Espandar L, Bunnell B, Wang G, Gregory P, McBride Med. 2014;6(266):266ra172.
C, Moshirfar M. Adipose-derived stem cells on hyal- 38. Takahashi K, Yamanaka S. Induction of pluripotent
uronic acid-derived scaffold: a new horizon in bioengi- stem cells from mouse embryonic and adult fibroblast
neered cornea. Arch Ophthalmol. 2012;130(2):202–8. cultures by defined factors. Cell. 2006;126(4):663–76.
24. Mittal SK, Omoto M, Amouzegar A, Sahu A, 39. Naylor RW, Charles NJM, Cowan CA, Davidson
Alexandra R, Katikireddy KR, et al. Restoration of AJ, Holm TM, Sherwin T. Derivation of corneal
corneal transparency by mesenchymal stem cells. keratocyte-like cells from human induced pluripotent
Stem Cell Rep. 2016;7(4):583–90. stem cells. PLoS One. 2016;11(10):e0165464.
25. Demirayak B, Yüksel N, Çelik O, Subaşı C, Duruksu 40. Yao L, Bai H. Review: mesenchymal stem cells and
G, Unal Z, et al. Effect of bone marrow and adipose corneal reconstruction. Mol Vis. 2013;19:2237–43.
tissue-derived mesenchymal stem cells on the natu- 41. Caplan AI. Mesenchymal stem cells: time to change
ral course of corneal scarring after penetrating injury. the name! Stem Cells Transl Med. 2017;6(6):1445–51.
Exp Eye Res. 2016;151:227–35. 42. Alió JL, El Zarif M, Alió del Barrio JL. Cellular
26. Du Y, Carlson E, Funderburgh M, Birk D, Pearlman therapy of the corneal stroma: a new type of corneal
E, Guo N, et al. Stem cell therapy restores trans- surgery for keratoconus and corneal dystrophies a
parency to defective murine corneas. Stem Cells. translational research experience. 1st ed. Amsterdam:
2009;27(7):1635–42. Elsevier; 2020.
27. Liu H, Zhang J, Liu C-Y, Wang I-J, Sieber M, Chang 43. De Miguel M, Alio J, Arnalich-Montiel F, Fuentes-
J, et al. Cell therapy of congenital corneal diseases Julian S, de Benito-Llopis L, Amparo F, Bataille
with umbilical mesenchymal stem cells: lumican null L. Cornea and ocular surface treatment. Curr Stem
mice. PLoS One. 2010;5(5):e10707. Cell Res Ther. 2010;5(2):195–204.
28. Coulson Thomas VJ, Caterson B, Kao 44. Alió J, Alió Del Barrio J, El Zarif M, Azaar A,
W. Transplantation of human umbilical mesen- Makdissy N, Khalil C, et al. Regenerative surgery of
chymal stem cells cures the corneal defects of the corneal stroma for advanced keratoconus: 1-year
mucopolysaccharidosis VII mice. Stem Cells. outcomes. Am J Ophthalmol. 2019;203:53–68.
2013;31(10):2116–26. 45. El Zarif M, Alió J, Alió del Barrio J, Abdul Jawad K,
29. Winston W-YK, Vivien J. CT Cell therapy of corneal Palazón-Bru A, Abdul Jawad Z, et al. Corneal stromal
diseases. Cornea. 2016;35(Suppl 1):S9–S19. regeneration therapy for advanced keratoconus: long-
30. De Miguel M, Fuentes-Julián S, Blázquez-Martínez A, term outcomes at 3 years. Cornea. 2021;40(6):741–54.
Pascual C, Aller M, Arias J, et al. Immunosuppressive 46. Zuk P, Zhu M, Mizuno H, Huang J, Futrell J, Katz
properties of mesenchymal stem cells: advances and A, et al. Multilineage cells from human adipose tis-
applications. Curr Mol Med. 2012;12(5):574–91. sue: implications for cell-based therapies. Tissue Eng.
31. Alió Del Barrio J, El Zarif M, De Miguel M, Azaar 2001;7(2):211–28.
A, Makdissy N, Harb W, et al. Cellular therapy with 47. Zuk PA, Zhu M, Ashjian P, De Ugarte D, Huang
human autologous adipose-derived adult stem cells for J, Mizuno H, et al. Human adipose tissue is a
advanced keratoconus. Cornea. 2017;36(8):952–60. source of multipotent stem cells. Mol Biol Cell.
32. Alió Del Barrio J, El Zarif M, Azaar A, Makdissy N, 2002;13(12):4279–95.
Khalil C, Harb W, et al. Corneal stroma enhancement 48. Bourin P, Bunnell B, Casteilla L, Dominici M, Katz
with decellularized stromal laminas with or without A, March K, et al. Stromal cells from the adipose
stem cell recellularization for advanced keratoconus. tissue-derived stromal vascular fraction and culture
Am J Ophthalmol. 2018;186:47–58. expanded adipose tissue-derived stromal/stem cells:
a joint statement of the International Federation for stroma cell density evolution in keratoconus corneas
Adipose Therapeutics and Science (IFATS) and the following the implantation of adipose mesenchymal
international Society for Cellular Therapy (ISCT). stem cells and corneal laminas: an in vivo confo-
Cytotherapy. 2013;15(6):641–8. cal microscopy study. Invest Ophthalmol Vis Sci.
49. Ponce Márquez S, Martínez V, McIntosh Ambrose W, 2020;61(4):22.
Wang J, Gantxegui N, Schein O, et al. Decellularization 52. El Zarif M, Abdul Jawad K, Alió JL. Confocal micros-
of bovine corneas for tissue engineering applications. copy of the cornea in a clinical model of corneal stro-
Acta Biomater. 2009;5(6):1839–47. mal expansion using adipose stem cells and corneal
50. El Zarif M, Alió del Barrio J, Mingo D, Abdul Jawad decellularized laminas in patients with keratoconus.
K, Alió J. Corneal stroma densitometry evolution in a In: Alió JL, Alió del Barrio JL, Arnalich-Montiel F,
clinical model of cellular therapy for advanced kera- editors. Corneal regeneration therapy and surgery.
toconus. Cornea. 2022;42:332. 1st ed. Essentials in ophthalmology. Cham: Springer;
51. El Zarif M, Abdul Jawad K, Alió del Barrio J, Abdul 2019. p. 363–86.
Jawad Z, Palazón-Bru A, De Miguel M, et al. Corneal
Part V
Endothelial Keratoplasty
Endothelial Keratoplasty.
Historical Review and Current
25
Outcomes
Farideh Doroodgar, Hassan Hashemi, Sana Niazi,

Sepehr Feizi, and Mohammad Ali Javadi
Key Points corneal transplantation surgery, with surgical

• Corneal endothelial dysfunction is treated processes and clinical outcomes that are con-
using EK, which has been demonstrated to be stantly evolving. Because past review studies
more effective than PK. have fallen short of giving the most up-to-date
• DSEK and its variations became the most information, we examine the history of EK evo-
popular EK approach because to its good sur- lution with a focus on the primary forms of EK,
gical results. which include (1) Descemet stripping endothe-
• The new suggested details for DSEK, DSAEK, lial keratoplasty (DSEK) and its improvements,
and DMEK need to be thoroughly evaluated. such as Descemet stripping automated endothe-
lial keratoplasty (DSAEK) and ultrathin
DSAEK, (2) Descemet membrane endothelial
Introduction keratoplasty (DMEK), and (3) pre-Descemet’s
endothelial keratoplasty (PDEK) and address
After more than a century of success with pen- the indications, surgical techniques, risks, and
etrating keratoplasty (PK), endothelial kerato- prospects of various surgical treatments.
plasty (EK) ushered in a paradigm shift in (Fig. 25.1).

org/10.1007/978-3-031-32408-6_25.
F. Doroodgar (*)
Negah Aref Ophthalmic Research Center, Shahid
Beheshti University of Medical Science, Tehran, Iran
S. Niazi
Translational Ophthalmology Research Center, Negah Aref Ophthalmic Research Center, Shahid
Tehran University of Medical Sciences, Tehran, Iran Beheshti University of Medical Science, Tehran, Iran
Translational Ophthalmology Research Center,
H. Hashemi Tehran University of Medical Sciences, Tehran, Iran
Noor Ophthalmology Research Center, Noor Eye e-mail: [email protected]
Hospital, Tehran, Iran
S. Feizi · M. A. Javadi
Eye Research Center, Farabi Eye Hospital, Tehran Department of Opthalmology, Labbafinezhad
University of Medical Sciences, Tehran, Iran Hospital, Shahid Beheshti University of Medical
e-mail: [email protected] Sciences, Tehran, Iran
https://doi.org/10.1007/978-3-031-32408-6_25
Corneal Transplantation (Keratoplasty)
Posterior Lamellar Keratoplasty (PLK)

Anterior Lamellar
or Penetrating Keratoplasty (PKP)
Keratoplasty (ALK) Endothelial Keratoplasty
Superficial Anterior Deep Anterior

Lamellar Keratoplasty Lamellar Keratoplasty Indications:
(SALK) (DALK) Fuchs endothelial corneal dystrophy (FECD)
Corneal edema after cataract surgery
Bullous keratopathy (BK)
Posterior polymorphous dystrophy
Iridocorneal endothelial syndrome
Endothelial decompensation from trauma
Complications of previous intraocular surgeries
Pre-Descemet’s Descemet Membrane

Endothelial Keratoplasty Endothelial Keratoplasty
(PDEK) (DMEK)
Descemet Stripping Descemet Stripping

Endothelial Keratoplasty Automated Endothelial
(DSEK) Keratoplasty (DSAEK)
Fig. 25.1 The types of corneal transplantations
Evolution of EK formed the first posterior lamellar keratoplasty

(PLK) using full-thickness grafts from donor
In 1906, the first successful human corneal trans- endothelium, which was the first attempt to use
plant took place [1]. As the suggested procedures the cornea’s inner layer. The transplant failed
became more successful, the demand for cornea because of glaucoma, endothelial deterioration,
from cadavers grew, leading in the establishment and poor donor–recipient apposition [3]. The next
of eye banking in the 1940s, which greatly attempts, such as the anterior approach, were like-
expanded corneal transplantation in the United wise unsuccessful. Because the use of sutures for
States and around the world [2]. Surgical proce- attachment of the donor graft to the recipient cor-
dures improved dramatically as knowledge of nea was the main technical difficulty of these pro-
basic principles of the cornea, such as anatomy, cedures, Melles and colleagues (1998) introduced
physiology, immunology, tissue biology, and a new modification in which air was used instead
anesthetic, grew [2]. of sutures to attach the donor lenticule to the pos-
EK’s origins can be traced back to studies on terior recipient cornea, resulting in a better visual
the selective replacement of damaged posterior outcome [4]. They next lowered the corneal inci-
corneal layers. Tillett used full-thickness grafts sion from 9 to 5 millimeters and placed a folded
from donor endothelium to accomplish posterior donor lenticule into the anterior chamber [5].
lamellar keratoplasty (PLK) in 1956, which was Deep lamellar endothelial keratoplasty (DLEK)
the first effort to employ the cornea’s inner layer. was introduced to the United States by Terry and
Glaucoma, endothelial damage, and poor donor– Ousley [6]. Unfortunately, the clinical outcomes
recipient apposition all contributed to the trans- were poor, and the surgical approach was difficult
plant’s failure [3]. EK was developed as a result of to duplicate, preventing the technique from
research into the selective replacement of injured becoming widely used and prompting the
posterior corneal layers. In 1956, Tillett per- researchers to seek new alternatives.
25 Endothelial Keratoplasty. Historical Review and Current Outcomes 367
Melles and colleagues (2004) simplified the ultrathin DSAEK (UT-DSAEK), which led to
procedure known as DSEK by removing the most superior postoperative visual results than normal
difficult element of PLK/DLEK (i.e., dissection DSAEK [21] in search of a more uniform surface
and excision of the recipient stroma) and replac- [22]. The surgical outcome and best-corrected
ing it with descemetorhexis. DSEK takes an inter- visual acuity (BCVA) of UT-DSAEK have been
nal method to remove the host’s Descemet’s shown to be equivalent to [23, 24], or better than
membrane (DM) and endothelium, leaving a that of DMEK in later trials [25, 26]. However, a
smooth surface [7]. Because of the smaller inci- 32-month follow-up of patients who had DMEK,
sion, this procedure quickly gained popularity, DSAEK, UT-DSAEK, nano-thin DSAEK (15–
resulting in improved safety, a lower risk of post- 49 m), ultrathin DSAEK (50–99 m) revealed
operative complications, and faster visual reha- identical visual acuity, ruling out the influence of
bilitation [8]. Manual dissection of the donor graft thickness and regularity on patient visual
cornea, like PLK/DLEK, resulted in a donor lenti- outcomes [27]. The contradictory findings high-
cule with a variable thickness profile; thus, light the need for additional standardized studies
Gorovoy (2006) modified this technique by dis- and meta-analysis in this area.
secting donor corneas with a microkeratome and Agarwal and colleagues (2013) reported
introducing DSAEK, which improved the visual PDEK, a surgical method for EK in which the
outcome by smoothing the donor-recipient inter- pre-Descemet stromal layer (Dua’s layer; sepa-
face [9]. When compared to PK, the early research rated by an air bubble) is additionally trans-
on DSAEK showed it to be a successful surgical planted with DM and endothelium [28]. The
technique with less problems, including as graft following publications all showed positive out-
failure, and faster visual recovery with less astig- comes, suggesting that PDEK tissue preparation
matism [10–12]. With the preparation of the donor and unrolling is straightforward and repeatable
tissue by the eye banks and the use of innovative [29]. In comparison to DMEK, PDEK grafts are
techniques for tissue delivery, DSAEK acquired 25–30 mm thick, which reduces postoperative
greater popularity; nonetheless, the visual acuity graft-host interface haze and improves intraop-
following DSAEK remained unsatisfactory [13]. erative tissue handling [30]. Furthermore, PDEK
Melles and colleagues (2006) presented overcomes another drawback of DMEK by
another improvement of EK to improve visual allowing surgeons to use corneal tissue from
acuity, known as DMEK, in which rolled DM young donors [28]. In DMEK, the presence of a
and endothelium from the donor are inserted into large number of hexagonal cells in the stromal
the recipient’s anterior chamber (AC) through a layer of donors younger than 40 raises the likeli-
3-mm incision and unfolded using air and bal- hood of implant rejection, whereas PDEK lacks
anced salt solution (BSS) [14]. When opposed to a full stromal layer and hence has no such con-
DSAEK, DMEK preserves corneal architecture cerns [30].
and delivers superior visual results by replacing
the defective layer of the recipient cornea with
matching donor tissue [15–17]. Even after pro- Indications
viding better tissue preparation procedures, such
as ‘no-touch’ harvesting technique, the funda- Patients with endothelial dysfunction, such as
mental downside of this technology that has Fuchs endothelial corneal dystrophy (FECD),
resulted in limited expansion of its application is corneal edema after cataract surgery, bullous
technical difficulties in preparing donor tissue keratopathy (BK), posterior polymorphous dys-
[18–20]. According to the Eye Bank Association trophy, iridocorneal endothelial syndrome, and
of America’s data, the number of DSAEK still endothelial decompensation from trauma, as well
outnumbers the number of DMEK (Fig. 25.1). as complications of previous intraocular surger-
Neff and colleagues (2011) created ultrathin ies, are candidates for EK [31, 32]. DSAEK
DSAEK grafts, thinner than 130 m, known as became the most popular form of EK utilized
globally with the introduction of DSEK and sub- DSAEK and DMEK following failed PK was
sequent modifications. When compared to equal [50].
DSAEK and PK, the next procedure presented, Patients who have had previous glaucoma sur-
DMEK, resulted in better and quicker visual gery, such as trabeculectomy and implantation of
recovery as well as greater endothelium survival a glaucoma shunt device, have an additional hur-
for patients with FECD and BK [33, 34]. Patients dle with EK, which can result in bleb-related
with FECD who had no corneal scar, preserved problems, tissue loss, and graft displacement.
AC anatomy, and an intact lens/iris diaphragm Furthermore, owing to mechanical strain of the
had favorable results; additionally, patients with glaucoma device [43], the DMEK donor graft
BK had a higher rate of endothelial damage than may be injured during unfolding [51]. Although
those with FECD, with endothelial failure occur- additional considerations are necessary during
ring only in patients with concomitant ocular surgery to ensure a full air fill at the conclusion of
pathology [35, 36]. Despite these limitations, surgery and relocating the glaucoma device,
DMEK was not frequently utilized after its incep- DSAEK is regarded a suitable surgical approach
tion, and DSAEK remains the most common sur- for patients with concurrent glaucoma or a his-
gical approach of EK, especially in difficult tory of glaucoma surgery [52]. Additionally, nar-
situations [37], as discussed below. row angels in glaucoma patients and Asians,
The state of the crystalline lens is a crucial resulting in peripheral anterior synechiae and
factor to consider when choosing a surgical pro- shallow ACs, provide a significant problem that
cedure; individuals with cataracts require both necessitates EK method changes [53].
cataract surgery and EK, which may be done In aphakic individuals and those with iris
simultaneously or in stages depending on the AC anomalies such aniridia, EK might be difficult. In
depth and the condition of the anterior corneal such circumstances, DSAEK can be done with
surface [38]. In phakic eyes left without cataract additional surgical modifications, such as donor
surgery for 2 years, DSEK has been reported to lenticule suturing to the recipient cornea, to
hasten cataract formation, but DMEK has been lessen the chance of posterior graft dislodgment
proven to be safe with great visual results in these [54]. Obviously, this change cannot be made dur-
circumstances without removing the crystalline ing DMEK. In situations with corneal endothe-
lens [39–41]. lium dysfunction necessitating EK, the presence
A history of unsuccessful keratoplasty, either of an AC IOL poses an additional obstacle; for
PK or EK, is another reason for EK. However, in whom the IOL to be replaced, this operation can
this situation, the choice of surgical technique is be done in stages or all at once. Although visual
debatable; some suggest the safety and favorable acuity was poorer than with DSAEK alone, evi-
outcomes of DSAEK, with graft dislocation rates dence suggests that concurrent IOL exchange
comparable to primary DSAEK [42], while oth- with DSAEK does not increase the risks of dislo-
ers have reported favorable visual outcomes for cation, primary graft failure, donor endothelial
DMEK after failed PK, with relatively good cell loss, or pupillary block [55]. In patients with
long-term outcomes [43, 44]; however, in cases an AC IOL, DMEK is recommended as a viable
complicated by glaucoma or AC intraocular lens procedure, but IOL removal is recommended in
(IOL), DMEK was associated with higher graft individuals with a high risk of postoperative
detachment and reb [45, 46]. DMEK following a problems [46]. Overall, DSAEK is preferred over
failed DSAEK has also been proven to improve DMEK in difficult cases due to its greater adapt-
the visual quality of patients to a level equivalent ability and predictability, which includes a wider
to initial DMEK [47, 48]. When comparing the range of graft insertion techniques, the ability to
results of DMEK for failed PK with a repeat PK, secure the lenticule to the overlying stroma, and
it was discovered that DMEK resulted in greater direct interface fluid evacuation, as well as the
wound stability and fewer suture-related prob- more robust nature of the DSAEK lenticule itself
lems [49]. Furthermore, the total failure rate of [56]. However, the present literature has a poor
level of evidentiary certainty, and further research DSEK/DSAEK

is needed to establish the best surgical approach
for these purposes [57]. DSEK, as well as its subsequent modifications
It’s also been claimed that UT-DSAEK can be DSAEK, is the most frequent technique per-
done on any eye that’s been diagnosed as needing formed today, and it is suitable for practically all
DSAEK. In terms of postoperative BCVA, endo- patients. On the recipient eye, a 3–5 mm incision
thelial cell density, and survival rate, UT-DSAEK (corneal/scleral or limbal) is made. Then, utiliz-
and DMEK had similar outcomes [58, 59]. ing air, fluid, or viscoelastic, descemetorhexis is
However, there is a paucity of data about the conducted, and recipient DM is removed. In
superiority of surgical procedures in certain situ- DSEK, donor preparation is done by hand, but in
ations, and additional research is needed before DSAEK, lamellar dissection is done with a
definitive conclusions can be drawn. PDEK suf- microkeratome, which makes donor preparation
fers from a greater lack of data, since most studies easier and results in a smoother interface (Video
have focused on surgical procedures and overall 25.1). The clinical results of precut tissues from
results rather than particular purposes or com- eye banks are positive and comparable to those of
parisons with other techniques [30]. Despite the surgeon-cut tissues [60, 61]. To avoid AC angle
fact that the surgical approaches are comparable closure, the donor cornea is sliced with a trephine
to those used in DMEK, data shows that surgeons 3 mm smaller than the recipient horizontal cor-
prefer DSAEK in the majority of cases. nea diameter after lamellar dissection [62]. The
recipient’s AC receives the trephined donor pos-
terior lenticule. To limit endothelial cell loss,
Surgical Procedures many procedures for donor insertion have been
created, including forceps (taco technique),
Table 25.1 summarizes the four endothelial kera- glides (including Busin glide, Sheets glide, Tan
toplasty surgical procedures, with more specific EndoGlide), and inserters [9, 43]. To avoid pupil-
information provided in the following sections. lary obstruction, a peripheral iridotomy may be
(Also, in Fig. 25.2, you can see the procedures required before donor implantation in some situ-
together.) ations [43]. The air bubble is utilized to approxi-
Table 25.1 The distinctions between the four endothelial keratoplasty surgical procedures
DSAEK/DSEK/
UT-DSAEK DMEK PDEK
Surgical layers Stroma, DM, and DM and Predescemet, DM, and
endothelium endothelium endothelium
Microkeratome Yes No No
Artificial anterior chamber Required N/A N/A
Type of procedure Tissue additive Tissue neutral Minimal tissue additive
Technical difficulty Easy Difficult Moderate
Graft unrolling Easy Difficult Moderate
Tissue handling Good Difficult Good
Eye bank prepared donor tissue Yes Yes Yes
Induced hyperopia Yes No No
Corneal thickness Increased Normal Minimal
Intrastromal interface Yes No Minimal
Type of big bubble N/A Type 2 Type 1
Donor tissue loss Negligible Yes Yes
Cost Costly Cost effective Cost effective
Visual recovery Slow Fast Fast
Fig. 25.2 Illustration of Endothelial Keratoplasty tech- layer prepared using pneumodissection. (e) DWEK/DSO
niques. (a) In DLEK, a posterior stromal pocket in the is descemetorhexis alone of the patient’s cornea, relying on
patient’s cornea is created for placement of a partial- primary intention healing. (f) DMET is a proposed tech-
thickness graft. (b) In DSEK/DSAEK, only the pathologic nique involving a focally attached graft that acts as a
tissue is removed via a descemetorhexis, and a graft with a source for endothelial cells. (Courtesy of Moshirfar &
thin layer of stroma, created manually or with a microkera- Thomson This figure is distributed under the terms of the
tome, is used to replace the removed tissue. (c) DMEK is Creative Commons Attribution 4.0 International License
the replacement of pathologic Descemet’s membrane and (http://creativecommons.org/licenses/by/4.0/), which per-
endothelium with a graft prepared by a descemetorhexis mits use, duplication, adaptation, distribution, and repro-
technique. (d) PDEK grafts include the pre-Descemet duction in any medium or format)
mate the donor lenticule to the recipient cornea’s UT-DSAEK was compared to DMEK, the results
posterior stroma after correct unfolding. In com- were mixed; although some studies found similar
plex circumstances, such as aphakic eyes, full- results [23, 24], others found that DMEK resulted
thickness sutures can also be utilized to secure in better visual outcomes [25, 26]. A microkera-
donor corneas. tome can be used to prepare the UT-DSAEK tis-
sue in a single- or double-pass procedure. The
double-pass approach has been linked to a greater
UT-DSAEK risk of endothelial injury and donor tissue perfo-
ration in several studies. To lessen the danger of
UT-DSAEK is a DSAEK variant in which the donor perforation, hydration of the grafts after
central donor thickness is less than 130 m. An the first incision and the use of a low-pulse high-
RCT found that UT-DSAEK had better visual frequency femtosecond laser have been recom-
outcomes than DSAEK [22, 23]. When mended [63–65].
DMEK [73]. After unfolding the donor tissue in the

proper orientation, such as with DSAEK, air is
By removing the stromal layer, DMEK was able injected to attach the graft to the recipient cornea;
to overcome the key drawbacks of DSEK, such as some surgeons remove the air after a few hours to
interface haziness and hyperopic shift. As a avoid pupillary block, while others maintain a
result, DMEK produces a better visual result than complete air fill and routinely perform a periph-
DSEK/DSAEK. Donor DM is removed in order eral iridotomy.
to prepare DMEK donor tissue (Video 25.2).
Several strategies for tissue preparation have
been introduced. In the SCUBA technique, DM PDEK
is grasped (using a nontoothed forceps) and
slowly stripped away from the stroma approxi- In addition to DM and endothelium, EK’s recent
mately halfway to the center; then, the central invention advises retaining the prelayer
part is separated by central partial-thickness Descemet’s composite, resulting in a graft with a
trephination (SCUBA technique), while other diameter of 7.5–8.5 mm, which has demonstrated
techniques such as manual peeling, forceps peel- good outcomes. PDEK tissue is simpler to handle
ing, combined manual delamination, and hydro and unroll in AC since it scrolls less. It also
dissection, as well as a trend toward a no-touch enables for the harvesting of transplant tissue
technique, have also been suggested [66, 67]. from young donors [28, 74, 75]. Intrastromal
Another suggested approach for DMEK tissue injection of air is used to generate a type 1 large
production is pneumatic dissection, in which the bubble (BB) with a diameter of 7–8 mm for donor
pre-Descemet stromal layer remains linked to the tissue preparation [30]. The donor tissue is next
DM was, in reality, supplying PDEK tissue trephined or manually incised at the intersection
(described below). Eye banks also supply pre- of the bubble wall and the corneal stroma with
stripped tissues for DMEK, which have proved to MicroVannas scissors [30]. After stripping the
have similar cell viability to surgeon-prepared recipient’s DM and endothelium, the graft is
tissues [68], and preloaded DMEK tissues have injected or pulled into the AC and unfurled fol-
also exhibited superior visual results than pre- lowing the procedures outlined before. PDEK tis-
loaded DSAEK tissues [69]. sue is simpler to handle than DMEK tissue,
The insertion of the donor tissue, which is the making the treatment easier to complete [29].
most difficult aspect of all types of EK, comes More research is needed to discover the optimum
next after tissue preparation and recipient cut. insertion and unfolding strategies, as well as
Melles and colleagues first utilized a glass pipette endothelial cell loss and visual results following
to implant DMEK tissue [14], but other the use of this EK approach [62, 76–78].
approaches, such as the use of IOL injectors such
as the STAAR microinjector, Alcon B cartridge,
Jones tube, and Bonfadini-Todd injector, have DWEK
also been proposed. Each insertion approach
causes endothelial cell death at a variable rate Descemetorhexis without endothelial kerato-
[70–72], and there is no consensus on the best plasty (DWEK) is a proposed term to describe
and safest procedure. After insertion, the donor the surgical removal of Descemet membrane
tissue should be unfolded using a variety of tech- (DM) without subsequent endothelial transplan-
niques, including a combination of air and BSS, tation (Video 25.3) [79]. DWEK was based on
the Yoeruek no-touch technique with a double several case reports of spontaneous resolution of
roll, the Dirisamer technique (carpet unrolling corneal edema after iatrogenic (during intraocu-
while fixing graft edge), the Dapena maneuver lar surgery) and deliberate removal of DM in
(small air bubble-assisted unrolling), mechanical patients with FECD [80–83]. This potential for
tapping, and the single sliding cannula maneuver endothelial “rejuvenation” in FECD contrasts
with bullous keratopathy, thought to be an endo- When comparing DMEK to DSAEK, Zhu and
thelial depletion disease [84]. Presence of central colleagues found that BCVA was superior, with
guttae, clear peripheral cornea with an endothe- more patients having a BCVA of 20/25 and 20/20
lial cell count >1000 cells/mm2 on confocal or [26]. Other meta-analysis studies have also dem-
specular microscopy, phakic or pseudophakic, onstrated that DMEK improves visual accuracy
and patient’s preference are indications of FECD, over DSEK/DSAEK [96, 97].
while advanced corneal stromal edema, periph- Clinical results following UT-DSAEK and
eral endothelial cell count <1000 cells/mm2, PDEK were investigated and compared in a
presence of secondary corneal pathology, and smaller number of trials. BSCVA of 20/20 in
history of herpes simplex virus or cytomegalovi- 36.3%, 37.4%, 46.4%, and 53.4% of eyes 1 year,
rus keratitis are the contraindications. 2 years, 3 years, and 5 years after UT-DSAEK,
Complications include descemetorhexis and 20/40 in 95.5%, 95.3%, 96.0%, and 96.6% of
decentration, descemet membrane detachment,
eyes 1 year, 2 years, 3 years, and 5 years after
posterior stromal opacities at the margin of the UT-DSAEK, respectively [58]. With improved
descemetorhexis, abnormal corneal topography results following UT-DSAEK vs. DSAEK, the
and irregular corneal astigmatism [85], and per- quicker and better visual recovery has been
sistent corneal edema (apart from Arbelaez [86], observed [22]. Droutsas and colleagues, on the
which had issues with DMEK detachments, suc- other hand, found that DSAEK had much better
cessful DMEK and DSAEK procedures were results than UT-DSAEK [98]. Similarly, compar-
described in these cases [87, 88]). ing the visual outcomes of UT-DSAEK and
DMEK revealed mixed findings; some research-
ers claim that UT-DSAEK is equal to DMEK [23,
SEK, DSAEK, UT-DSAEK, DMEK,
D 24, 99], while others claim that BCVA after
and PDEK Visual Results DMEK is higher [25, 26].
Despite the technical benefits of PDEK, few
DSEK has been proven to provide rapid and con- research have looked into patients’ postoperative
sistent visual recovery, with BCVAs of 20/40 or visual acuity [30]. As a result, additional research
better reported in 38–100% of patients 3 months is needed to establish the results of various EK
after surgery, and 20/33 to 20/60 3–30 months approaches after surgery.
after surgery [89, 90]. However, when compared
to DSAEK, a larger number of patients had a
BCVA of 20/200 following PK [43]. BCVA var- raft Survival of DSEK, DSAEK,
G
ies according on the reason for EK, with eyes UT-DSAEK, DMEK, and PDEK
with PK having a better visual result than those
with FECD [91, 92]. DMEK has been demon- The 5-year graft survival after DSEK has been
strated to produce the quickest and finest visual compared to that after PK [100, 101]; however,
recovery, with 82% and 89% of patients, respec- DSAEK was shown to be superior than PK
tively, reaching a BCVA of 20/25, 5, and 10 years (79.4% vs. 66.5%, respectively) [102].
after surgery [93]. Unsatisfactory visual outcomes following
One year following DSAEK and DMEK, a DSAEK, on the other hand, lead to a greater prob-
single surgeon’s experience with more than 200 ability of repeat EK [103]. DSEK/DSAEK sur-
patients revealed BCVAs of 20/50 and 20/125, vival rates have been reported to range from 55 to
respectively [94]. DMEK demonstrated a supe- 100% [43]. This large range reflects a variety of
rior BCVA (mean difference of 20/16) after variables that affect graft survival, including the
1 year [57] and a better BSCVA (mean differ- severity of intraoperative trauma, which is directly
ence of 20/13.5) after 6 months in comparative related to the surgeon’s experience. In addition,
meta- analysis studies compared to DSEK/ the patient’s underlying ocular disorders have a
DSAEK [95]. role; 5-year graft survival is 95% in FECD, 76%
in pseudophakic or aphakic bullous keratopathy, nect, whereas the complete dislocated graft can
and 40% in eyes with a glaucoma shunt or a his- be reattached using air injection or rebubbling.
tory of trabeculectomy [101]. After 1, 2, 3, and However, this technique may result in a greater
5 years, UT-DSAEK graft survival was reported loss of endothelial cells [109]. DMEK and
to be 99.1%, 96.2%, 94.2%, and 94.2%, respec- DSAEK done by competent surgeons had identi-
tively [58], and endothelial cell loss was observed cal graft dislocation rates, according to Philips
to be equivalent to DSAEK [22]. and colleagues [110]. A review of nonrandom-
Despite the fact that DMEK is associated with ized trials found that DMEK had a greater risk of
a decreased graft survival rate, some authors have graft dislocations and rebubbling than DSAEK,
observed greater 5- and 10-year survival rates fol- but the evidence was of insufficient quality to
lowing DMEK [93, 104]. Price and colleagues draw firm conclusions [57].
found that DSEK and DMEK graft survival was Endothelial graft rejection is less common
equivalent after 5 years [105]. The authors ascribed after EK than it is after PK. At the same time,
this result to DMEK’s reduced rate of immuno- race, glaucoma history, and corticosteroid-
logic rejection, which compensates for the tech- induced ocular hypertension have all been found
nique’s higher rate of intraoperative cell loss [105]. as significant variables to transplant failure [111].
As a result, additional research is needed to dem- After 2 years, the rejection rates of DSEK,
onstrate DMEK’s advantage over DSEK/DSAEK DMEK, and PK were significantly different
in terms of graft failure [106]. More research on (12%, 1%, and 18%, respectively), indicating
the long-term effects of PDEK is needed. that DMEK is a key priority in this regard [112].
Other research verified that DMEK has the low-
est rejection rate when compared to DSAEK and
Complications PK [113]. Modern lamellar procedures, which
result in a lower immune response, are thought to
Graft displacement, graft rejection, and idio- be the most important element in lowering the
pathic primary graft failure are some of the most graft rejection rate following DSAEK and DMEK
common postoperative complications described [114]. Only one incidence of graft rejection was
for various kinds of EK (IPGF). According to recorded in a study of 72 individuals (144 eyes)
Maier and colleagues, graft dislocation was who received DSAEK in one eye followed by
0–82% in DSAEK and 31–81% in DMEK, IPGF DMEK in the other, indicating that this complica-
was 0–29% in DSAEK, 0–9% in DMEK, and tion is not a worry with these surgical techniques
0–3% in PK, and the probability of rejection was [57]. More research on the long-term effects of
4% in DSAEK, 1–3% in DMEK, and 0.5–23.3% PDEK and UT-DSAEK is needed.
in PK [107]. The broad range of complication IPGF has documented a rare PK consequence
rates alludes to the variations in attributing vari- (3%), but DSAEK and DMEK have reported
ables between experiments. To identify the greater rates (0–29% and 0–9%, respectively)
pooled analysis findings of the pure rates of com- [107]. IPGF is caused by the donor’s endothelial
plications, as well as a comparison of the compli- function being inadequate, which might be related
cation rates between the EK techniques, to poor donor quality or surgical stress. A retro-
meta-analysis studies are necessary. spective comparison between DSAEK and DMEK
Graft dislocation can occur if donor tissue is revealed that DMEK had a higher IPGF rate [115],
not adhered properly. To reduce postoperative which might be attributable to surgeons’ lack of
wound leakage and hypotony, precise wound competence with the newer procedure [116].
construction is essential. Complete fluid removal Glaucoma, cataract, endophthalmitis, micro-
from the interface and full AC air fill are two fur- bial keratitis, suture-related problems, supracho-
ther methods. It is also a good idea to tell the roidal hemorrhage, and DM separation from the
patient not to touch his or her eyes [108]. Partially donor graft are some of the less frequent conse-
separated lenticules may spontaneously recon- quences [117].
Future of EK Literature Search
EK techniques are always changing. Several inno- We used PubMed, MEDLINE, EMBASE, ISI,
vative surgical approaches are proposed, includ- and the Cochrane Central Register of Controlled
ing the use of a hydrogel scaffold [118], a Trials (CENTRAL) to conduct an electronic
preloaded transport cartridge [119], and a DMEK database search, following the PRISMA
surgery marking technique [120]. For improved (Preferred Reporting Items for Systematic
vision [121] and predicting the visual prognosis Review and Meta-Analyses) statement’s widely
following surgery, new technologies and proce- accepted technique standards. “Endothelial ker-
dures are being developed [122]. As a result, the atoplasty,” “corneal“ and “transplantation,”
new recommended details for the aforementioned “Descemet stripping endothelial keratoplasty,”
surgical kinds, such as DSEK, DSAEK, and “Descemet membrane endothelial kerato-
DMEK, must be extensively evaluated for their plasty,” “Descemet stripping automated endo-
reliability and application. Furthermore, the two thelial keratoplasty,” “Descemet stripping
EK procedures examined in this study, automated endothelial keratoplasty,” “Descemet
UT-DSAEK and PDEK, are relatively novel tech- stripping automated endothelial keratoplasty,”
niques, with few major randomized clinical trials “Desc “DSEK“ vs. “DMEK,” “DSEK” vs.
examining their long-term surgical results. PDEK, “DSAEK,” “ultrathin Descemet stripping auto-
which is simpler and more adaptable than DMEK, mated endothelial keratoplasty,” “UT-DSAEK”
may be the next generation of the most often used vs. “DSAEK,” “UT-DSAEK” vs. “DMEK,”
EK. More research is needed to compare the long- “pre-endothelial descemet’s keratoplasty,” “eye
term surgical results of PDEK with other EK bank,” or “eye banking” and “endot.” From
methods. 2010 to 2021, one researcher examined the
journals and retrieved studies published in
English. This review did not include any non-
Conclusion English articles. Conference abstracts, com-
plete texts without raw data retrieval, duplicate
EK has been found to be superior to PK in treat- publications, case reports, and letters were all
ing corneal endothelial dysfunction and is now removed. The whole text of the publications
the therapy of choice. DSEK and its variations chosen by the first researcher was then evalu-
became the most extensively utilized EK ated by all writers, who meticulously scruti-
approach after its inception, owing to its positive nized the articles. Any relevant references listed
surgical results. The eye banks’ precut tissues in the publications were also reviewed and
play an important role in the widespread usage of incorporated in the research during the careful
this technology. However, because of the long examination of the complete text of the articles.
visual recovery and danger of problems like as The final form of the literature review described
graft rejection, another approach, DMEK, was above was agreed upon by all authors. Articles
developed, which demonstrated a quicker and were chosen for inclusion depending on their
better visual recovery. Despite this, DMEK’s sur- importance and limits.
gical problems prevented it from being widely
used. PDEK’s innovation may be able to over- Take Home Notes
come DMEK’s constraints, such as technological • PDEK may be the next generation of the most
difficulties and the age restriction for donor widely used EK because it is easier and more
selection. However, because PDEK is a novel applicable than DMEK.
approach, further research is needed to assess • More research is needed to compare the long-
long-term clinical outcomes and compare it to term surgical outcomes of PDEK with other
other procedures. EK methods.
References 16. Dapena I, Ham L, Melles GR. Endothelial kerato-

plasty: DSEK/DSAEK or DMEK-the thinner the
better? Curr Opin Ophthalmol. 2009;20(4):299–307.
1. Zirm E. Eine erfolgreiche totale Keratoplastik.
17. Bhandari V, Reddy JK, Relekar K, Prabhu
Graefes Arhiv für Ophthalmologie.
V. Descemet’s stripping automated endothelial kera-
1906;64(3):580–93.
toplasty versus Descemet’s membrane endothelial
2. Güell JL, El Husseiny MA, Manero F, Gris O, Elies
keratoplasty in the fellow eye for fuchs endothelial
D. Historical review and update of surgical treat-
dystrophy: a retrospective study. Biomed Res Int.
ment for corneal endothelial diseases. Ophthalmol
2015;2015:1.
Therapy. 2014;3(1–2):1–15.
18. Price MO, Feng MT, Price FW Jr. Endothelial
3. Tillett CW. Posterior lamellar keratoplasty. Am J
Keratoplasty update 2020. Cornea.
Ophthalmol. 1956;41(3):530–3.
2021;40(5):541–7.
4. Melles G, Eggink F, Lander F, Pels E, Rietveld
19. Eye OAS. Standardized “no-touch” technique for
F, Beekhuis WH, Binder PS. A surgical tech-
Descemet membrane endothelial keratoplasty. Arch
nique for posterior lamellar keratoplasty. Cornea.
Ophthalmol. 2011;129(1):88–94.
1998;17(6):618–26.
20. Kruse FE, Laaser K, Cursiefen C, Heindl LM,
5. Melles GR, Lander F, Nieuwendaal C. Sutureless,
Schlötzer-Schrehardt U, Riss S, Bachmann BO. A
posterior lamellar keratoplasty: a case report of a
stepwise approach to donor preparation and inser-
modified technique. Cornea. 2002;21(3):325–7.
tion increases safety and outcome of Descemet
6. Terry MA, Ousley PJ. Deep lamellar endothelial
membrane endothelial keratoplasty. Cornea.
keratoplasty in the first United States patients: early
2011;30(5):580–7.
clinical results. Cornea. 2001;20(3):239–43.
21. Neff KD, Biber JM, Holland EJ. Comparison
7. Melles GR, Wijdh RH, Nieuwendaal CP. A tech-
of central corneal graft thickness to visual acu-
nique to excise the Descemet membrane from
ity outcomes in endothelial keratoplasty. Cornea.
a recipient cornea (descemetorhexis). Cornea.
2011;30(4):388–91.
2004;23(3):286–8.
22. Dickman MM, Kruit PJ, Remeijer L, van Rooij J,
8. Price FW, Price MO. Descemet's stripping with
Van der Lelij A, Wijdh RH, van den Biggelaar FJ,
endothelial keratoplasty in 50 eyes: a refrac-
Berendschot TT, Nuijts RM. A randomized multi-
tive neutral corneal transplant. J Refract Surg.
center clinical trial of ultrathin Descemet stripping
2005;21(4):339–45.
automated endothelial keratoplasty (DSAEK) versus
9. Gorovoy MS. Descemet-stripping automated
DSAEK. Ophthalmology. 2016;123(11):2276–84.
endothelial Keratoplasty. Cornea. 2006;25(8):
23. Busin M, Madi S, Santorum P, Scorcia V, Beltz
886–9.
J. Ultrathin Descemet's stripping automated endo-
10. Koenig SB, Covert DJ, Dupps WJ Jr, Meisler
thelial keratoplasty with the microkeratome double-
DM. Visual acuity, refractive error, and endothelial
pass technique: two-year outcomes. Ophthalmology.
cell density six months after Descemet stripping
2013;120(6):1186–94.
and automated endothelial keratoplasty (DSAEK).
24. Mencucci R, Favuzza E, Marziali E, Cennamo
Cornea. 2007;26(6):670–4.
M, Mazzotta C, Lucenteforte E, Virgili G, Rizzo
11. Bahar I, Sansanayudh W, Levinger E, Kaiserman
S. Ultrathin Descemet stripping automated endothe-
I, Srinivasan S, Rootman D. Posterior lamel-
lial keratoplasty versus Descemet membrane endo-
lar keratoplasty—comparison of deep lamellar
thelial keratoplasty: a fellow-eye comparison. Eye.
endothelial keratoplasty and Descemet stripping
2020;7:1–9.
automated endothelial keratoplasty in the same
25. Chamberlain W, Lin CC, Austin A, Schubach N,
patients: a patient’s perspective. Br J Ophthalmol.
Clover J, McLeod SD, Porco TC, Lietman TM,
2009;93(2):186–90.
Rose-Nussbaumer J. Descemet endothelial thick-
12. Covert DJ, Koenig SB. Descemet stripping and
ness comparison trial: a randomized trial comparing
automated endothelial keratoplasty (DSAEK) in
ultrathin Descemet stripping automated endothe-
eyes with failed penetrating keratoplasty. Cornea.
lial keratoplasty with Descemet membrane endo-
2007;26(6):692–6.
thelial keratoplasty. Ophthalmology. 2019;126(1):
13. Chen ES, Shamie N, Terry MA, Hoar KL. Endothelial
19–26.
keratoplasty: improvement of vision after healthy
26. Zhu L, Zha Y, Cai J, Zhang Y. Descemet stripping
donor tissue exchange. Cornea. 2008;27(3):279–82.
automated endothelial keratoplasty versus descemet
14. Melles GR, San Ong T, Ververs B, van der Wees
membrane endothelial keratoplasty: a meta-analysis.
J. Descemet membrane endothelial keratoplasty
Int Ophthalmol. 2018;38(2):897–905.
(DMEK). Cornea. 2006;25(8):987–90.
27. Tourabaly M, Chetrit Y, Provost J, Georgeon C,
15. Price MO, Giebel AW, Fairchild KM, Price FW Jr.
Kallel S, Temstet C, Bouheraoua N, Borderie
Descemet's membrane endothelial keratoplasty: pro-
V. Influence of graft thickness and regularity on
spective multicenter study of visual and refractive
vision recovery after endothelial keratoplasty. Br J
outcomes and endothelial survival. Ophthalmology.
Ophthalmol. 2020;104(9):1317–23.
2009;116(12):2361–8.
28. Agarwal A, Dua HS, Narang P, Kumar DA, Agarwal 42. Clements JL, Bouchard CS, Lee WB, Dunn SP,
A, Jacob S, Agarwal A, Gupta A. Pre-Descemet's Mannis MJ, Reidy JJ, John T, Hannush SB, Goins
endothelial keratoplasty (PDEK). Br J Ophthalmol. KM, Wagoner MD. Retrospective review of graft
2014;98(9):1181–5. dislocation rate associated with descemet strip-
29. Ross AR, Said DG, Gisoldi RAC, Nubile M, ping automated endothelial keratoplasty after
El-Amin A, Gabr AF, Abd M, Mencucci R, Pocobelli primary failed penetrating keratoplasty. Cornea.
A, Mastropasqua L. Optimizing pre-Descemet endo- 2011;30(4):414–8.
thelial keratoplasty technique. J Cataract Refract 43. Anshu A, Price MO, Tan DT, Price FW Jr.
Surg. 2020;46(5):667–74. Endothelial keratoplasty: a revolution in evolution.
30. Narang P, Agarwal A. Pre-Descemet's endothelial Surv Ophthalmol. 2012;57(3):236–52.
keratoplasty. Indian J Ophthalmol. 2017;65(6):443. 44. Schrittenlocher S, Schlereth SL, Siebelmann S,
31. Park CY, Lee JK, Gore PK, Lim C-Y, Chuck Hayashi T, Matthaei M, Bachmann B, Cursiefen
RS. Keratoplasty in the United States: a 10-year C. Long-term outcome of descemet membrane
review from 2005 through 2014. Ophthalmology. endothelial keratoplasty (DMEK) following failed
2015;122(12):2432–42. penetrating keratoplasty (PK). Acta Ophthalmol.
32. Mohamed A, Chaurasia S, Murthy SI, Ramappa M, 2020;98(7):e901–6.
Vaddavalli PK, Taneja M, Garg P, Chinta S, Basu 45. Liarakos VS, Satué M, Livny E, van Dijk K, Ham
S, Rathi VM. Endothelial keratoplasty: a review of L, Baydoun L, Dapena I, Melles GR. Descemet
indications at a tertiary eye care Centre in South membrane endothelial keratoplasty for a decom-
India. Asia Pac J Ophthalmol. 2014;3(4):207–10. pensated penetrating keratoplasty graft in the pres-
33. Woo J-H, Ang M, Htoon HM, Tan D. Descemet ence of a long glaucoma tube. Cornea. 2015;34(12):
membrane endothelial keratoplasty versus Descemet 1613–6.
stripping automated endothelial keratoplasty 46. Liarakos VS, Ham L, Dapena I, Tong CM,
and penetrating keratoplasty. Am J Ophthalmol. Quilendrino R, Yeh R-Y, Melles GR. Endothelial
2019;207:288–303. keratoplasty for bullous keratopathy in eyes with an
34. Green M, Wilkins MR. Comparison of early surgi- anterior chamber intraocular lens. J Cataract Refract
cal experience and visual outcomes of DSAEK and Surg. 2013;39(12):1835–45.
DMEK. Cornea. 2015;34(11):1341–4. 47. Weller JM, Tourtas T, Kruse FE, Schlötzer-
35. Baydoun L, Ham L, Borderie V, Dapena I, Hou J, Schrehardt U, Fuchsluger T, Bachmann
Frank LE, Oellerich S, Melles GR. Endothelial BO. Descemet membrane endothelial keratoplasty
survival after Descemet membrane endothe- as treatment for graft failure after descemet strip-
lial keratoplasty: effect of surgical indication ping automated endothelial keratoplasty. Am J
and graft adherence status. JAMA Ophthalmol. Ophthalmol. 2015;159(6):1050–7. e2
2015;133(11):1277–85. 48. Agha B, Shajari M, Slavik-Lencova A, Kohnen T,
36. Birbal RS, Baydoun L, Ham L, Miron A, Van Dijk Schmack I. Outcome of Descemet membrane endo-
K, Dapena I, Jager MJ, Böhringer S, Oellerich S, thelial keratoplasty for graft failure after Descemet
Melles GR. Effect of surgical indication and pre- stripping automated endothelial keratoplasty. Clin
operative lens status on Descemet membrane endo- Ophthalmol. 2019;13:553.
thelial keratoplasty outcomes. Am J Ophthalmol. 49. Mitry D, Bhogal M, Patel AK, Lee BS, Chai SM,
2020;212:79–87. Price MO, Price FW, Jun AS, Aldave AJ, Mehta
37. Veldman PB, Terry MA, Straiko MD. Evolving JS. Descemet stripping automated endothelial kera-
indications for Descemet's stripping automated toplasty after failed penetrating keratoplasty: sur-
endothelial keratoplasty. Curr Opin Ophthalmol. vival, rejection risk, and visual outcome. JAMA
2014;25(4):306–11. Ophthalmol. 2014;132(6):742–9.
38. Price FW Jr, Price MO. Combined cataract/ 50. Einan-Lifshitz A, Mednick Z, Belkin A, Sorkin
DSEK/DMEK: changing expectations. Asia Paci J N, Alshaker S, Boutin T, Chan CC, Rootman
Ophthalmol. 2017;6(4):388–92. DS. Comparison of Descemet stripping auto-
39. Chaurasia S, Price FW Jr, Gunderson L, Price mated endothelial keratoplasty and Descemet
MO. Descemet’s membrane endothelial kera- membrane endothelial keratoplasty in the treat-
toplasty: clinical results of single versus triple ment of failed penetrating keratoplasty. Cornea.
procedures (combined with cataract surgery). 2019;38(9):1077–82.
Ophthalmology. 2014;121(2):454–8. 51. Bonnet C, Ghaffari R, Alkadi T, Law SK, Caprioli J,
40. Parker J, Dirisamer M, Naveiras M, Tse WHW, van Yu F, Deng SX. Long-term outcomes of Descemet
Dijk K, Frank LE, Ham L, Melles GR. Outcomes of membrane endothelial keratoplasty in eyes with
Descemet membrane endothelial keratoplasty in pha- prior glaucoma surgery. Am J Ophthalmol.
kic eyes. J Cataract Refract Surg. 2012;38(5):871–7. 2020;218:288–95.
41. Trindade BLC, Eliazar GC. Descemet membrane 52. Aldave AJ, Chen JL, Zaman AS, Deng SX, Yu
endothelial keratoplasty (DMEK): an update F. Outcomes after DSEK in 101 eyes with previ-
on safety, efficacy and patient selection. Clin ous trabeculectomy and tube shunt implantation.
Ophthalmol. 2019;13:1549. Cornea. 2014;33(3):223–9.
53. Mehta JS, Parthasarthy A, Por Y-M, Cajucom-Uy H, for Descemet membrane endothelial keratoplasty: an
Beuerman RW, Tan D. Femtosecond laser-assisted updated review. Cornea. 2018;37(1):128–35.
endothelial keratoplasty: a laboratory model. 67. Maharana PK, Sahay P, Singhal D, Sharma N, Titiyal
Cornea. 2008;27(6):706–12. JS. Donor preparation in Descemet membrane
54. Eguchi H, Miyamoto T, Hotta F, Tomida M, Inoue endothelial keratoplasty. New Front Ophthalmol.
M, Mitamura Y. Descemet-stripping automated 2018;5:1–6.
endothelial keratoplasty for vitrectomized cases with 68. Romano V, Kazaili A, Pagano L, Gadhvi KA, Titley
traumatic aniridia and aphakic bullous keratopathy. M, Steger B, Fernández-Vega-Cueto L, Meana A,
Clin Ophthalmol. 2012;6:1513. Merayo-Lloves J, Diego P. Eye bank versus surgeon
55. Shah AK, Terry MA, Shamie N, Chen ES, prepared DMEK tissues: influence on adhesion and
Phillips PM, Hoar KL, Friend DJ, Davis-Boozer re-bubbling rate. Br J Ophthalmol. 2020;106:177.
D. Complications and clinical outcomes of Descemet 69. Romano V, Pagano L, Gadhvi KA, Coco G, Titley
stripping automated endothelial keratoplasty with M, Fenech MT, Ferrari S, Levis HJ, Parekh M,
intraocular lens exchange. Am J Ophthalmol. Kaye S. Clinical outcomes of pre-loaded ultra-
2010;149(3):390–7. e1 thin DSAEK and pre-loaded DMEK. BMJ Open
56. Sharma N, Maharana PK, Singhi S, Aron N, Patil Ophthalmol. 2020;5(1):e000546.
M. Descemet stripping automated endothelial kera- 70. Ighani M, Karakus S, Eghrari AO. Clinical outcomes
toplasty. Indian J Ophthalmol. 2017;65(3):198. of Descemet membrane endothelial keratoplasty
57. Stuart AJ, Romano V, Virgili G, Shortt AJ. Descemet's using the Bonfadini-Todd injector for graft insertion.
membrane endothelial keratoplasty (DMEK) versus Clin Ophthalmol. 2019;13:1869.
Descemet's stripping automated endothelial kera- 71. Shen E, Fox A, Johnson B, Farid M. Comparing the
toplasty (DSAEK) for corneal endothelial failure. effect of three Descemet membrane endothelial ker-
Cochrane Database Syst Rev. 2018;6(6):CD012097. atoplasty injectors on endothelial damage of grafts.
58. Madi S, Leon P, Nahum Y, D'Angelo S, Giannaccare Indian J Ophthalmol. 2020;68(6):1040.
G, Beltz J, Busin M. Five-year outcomes of ultrathin 72. Droutsas K, Lazaridis A, Kymionis G,
Descemet stripping automated endothelial kerato- Chatzistefanou K, Moschos M, Koutsandrea C,
plasty. Cornea. 2019;38(9):1192–7. Sekundo W. Comparison of endothelial cell loss
59. Dunker SL, Dickman MM, Wisse RP, Nobacht S, and complications following DMEK with the use
Wijdh RH, Bartels MC, Tang ML, van den Biggelaar of three different graft injectors. Eye. 2018;32(1):
FJ, Kruit PJ, Nuijts RM. Descemet membrane 19–25.
endothelial keratoplasty versus ultrathin Descemet 73. Liarakos VS, Dapena I, Ham L, van Dijk K,
stripping automated endothelial keratoplasty: a Melles GR. Intraocular graft unfolding techniques
multicenter randomized controlled clinical trial. in Descemet membrane endothelial keratoplasty.
Ophthalmology. 2020;127(9):1152–9. JAMA ophthalmol. 2013;131(1):29–35.
60. Yamazoe K, Yamazoe K, Shinozaki N, Shimazaki 74. Dua HS, Termote K, Kenawy MB, Said DG,
J. Influence of the precutting and overseas transpor- Jayaswal R, Nubile M, Mastropasqua L, Holland
tation of corneal grafts for Descemet stripping auto- S. Scrolling characteristics of pre-Descemet endo-
mated endothelial keratoplasty on donor endothelial thelial keratoplasty tissue: an ex vivo study. Am J
cell loss. Cornea. 2013;32(6):741–4. Ophthalmol. 2016;166:84–90.
61. Price MO, Baig KM, Brubaker JW, Price FW Jr. 75. Pereira NC, dos Santos FA, Maluf RCP, Dua
Randomized, prospective comparison of precut vs HS. Pre-Descemet’s endothelial keratoplasty: a
surgeon-dissected grafts for descemet stripping auto- simple, Descemet’s membrane scoring technique
mated endothelial keratoplasty. Am J Ophthalmol. for successful graft preparation. Br J Ophthalmol.
2008;146(1):36–41. e2 2021;106:786–9.
62. Singh NP, Said DG, Dua HS. Lamellar keratoplasty 76. Dua H, Said D. Pre-descemets endothelial kerato-
techniques. Indian J Ophthalmol. 2018;66(9):1239. plasty: the PDEK clamp for successful PDEK. Eye.
63. Phillips PM, Phillips LJ, Saad HA, Terry MA, 2017;31(7):1106–10.
Stolz DB, Stoeger C, Franks J, Davis-Boozer D. 77. Saint-Jean A, Soper M, Den Beste K, Iverson S,
“Ultrathin” DSAEK tissue prepared with a low– Price MO, Price FW. Technique for ensuring type I
pulse energy, high-frequency femtosecond laser. bubble formation for pre-descemet endothelial kera-
Cornea. 2013;32(1):81–6. toplasty preparation. Cornea. 2019;38(10):1336–8.
64. Sikder S, Nordgren RN, Neravetla SR, Moshirfar 78. Altaan SL, Gupta A, Sidney LE, Elalfy MS, Agarwal
M. Ultra-thin donor tissue preparation for endothe- A, Dua HS. Endothelial cell loss following tissue
lial keratoplasty with a double-pass microkeratome. harvesting by pneumodissection for endothelial
Am J Ophthalmol. 2011;152(2):202–8. e2 keratoplasty: an ex vivo study. Br J Ophthalmol.
65. Boynton GE, Woodward MA. Eye-bank prepara- 2015;99(5):710–3.
tion of endothelial tissue. Curr Opin Ophthalmol. 79. Kaufman AR, Nosé RM, Roberto Pineda
2014;25(4):319. I. Descemetorhexis without endothelial keratoplasty
66. Birbal RS, Sikder S, Lie JT, Groeneveld-van Beek (DWEK): proposal for nomenclature standardiza-
EA, Oellerich S, Melles GR. Donor tissue preparation tion. Cornea. 2018;37:e20–1.
80. Moloney G, et al. Descemetorhexis without grafting geon’s more than 200 initial consecutive cases. Clin
for Fuchs endothelial dystrophy—supplementation Ophthalmol. 2021;15:909.
with topical ripasudil. Cornea. 2017;36:642–8. 95. Singh A, Zarei-Ghanavati M, Avadhanam V, Liu
81. Dirisamer M, Ham L, Dapena I, van Dijk K, Melles C. Systematic review and meta-analysis of clinical
GR. Descemet membrane endothelial transfer:“free- outcomes of descemet membrane endothelial kera-
floating” donor Descemet implantation as a toplasty versus Descemet stripping endothelial kera-
potential alternative to “keratoplasty”. Cornea. toplasty/descemet stripping automated endothelial
2012;31:194–7. keratoplasty. Cornea. 2017;36(11):1437–43.
82. Shah RD, Randleman JB, Grossniklaus 96. Pavlovic I, Shajari M, Herrmann E, Schmack I,
HE. Spontaneous corneal clearing after des- Lencova A, Kohnen T. Meta-analysis of postop-
cemet's stripping without endothelial replacement. erative outcome parameters comparing descemet
Ophthalmology. 2012;119:256–60. membrane endothelial keratoplasty versus Descemet
83. Moloney G, et al. Descemetorhexis for fuchs’ dys- stripping automated endothelial keratoplasty.
trophy. Can J Ophthalmol. 2015;50:68–72. Cornea. 2017;36(12):1445–51.
84. Borkar DS, Veldman P, Colby KA. Treatment of 97. Li S, Liu L, Wang W, Huang T, Zhong X, Yuan J,
Fuchs endothelial dystrophy by Descemet strip- Liang L. Efficacy and safety of Descemet’s membrane
ping without endothelial keratoplasty. Cornea. endothelial keratoplasty versus Descemet’s stripping
2016;35:1267–73. endothelial keratoplasty: a systematic review and
85. Iovieno A, Neri A, Soldani AM, Adani C, Fontana meta-analysis. PLoS One. 2017;12(12):e0182275.
L. Descemetorhexis without graft placement for 98. Droutsas K, Petrelli M, Miltsakakis D, Andreanos
the treatment of fuchs endothelial dystrophy: pre- K, Karagianni A, Lazaridis A, Koutsandrea C,
liminary results and review of the literature. Cornea. Kymionis G. Visual outcomes of ultrathin-descemet
2017;36:637–41. stripping endothelial keratoplasty versus Descemet
86. Arbelaez JG, Price MO, Price FW Jr. Long-term stripping endothelial keratoplasty. J Ophthalmol.
follow-up and complications of stripping des- 2018;2018:1.
cemet membrane without placement of graft in 99. Torras-Sanvicens J, Blanco-Domínguez I, Sánchez-
eyes with fuchs endothelial dystrophy. Cornea. González J-M, Rachwani-Anil R, Spencer J-F,
2014;33:1295–9. Sabater-Cruz N, Peraza-Nieves J, Rocha-de-Lossada
87. Bleyen I, Saelens IE, van Dooren BT, van Rij C. Visual quality and subjective satisfaction in ultra-
G. Spontaneous corneal clearing after Descemet’s thin descemet stripping automated endothelial kera-
stripping. Ophthalmology. 2013;120:215. toplasty (UT-DSAEK) versus descemet membrane
88. Koenig SB. Planned descemetorhexis without endo- endothelial Keratoplasty (DMEK): a fellow-eye
thelial keratoplasty in eyes with Fuchs corneal endo- comparison. J Clin Med. 2021;10(3):419.
thelial dystrophy. Cornea. 2015;34:1149–51. 100. Li ALW, Kwok RPW, Kam KW, Young AL. A 5-year
89. Heinzelmann S, Shanab WA, Böhringer D, Maier analysis of endothelial vs penetrating keratoplasty
P, Reinhard T. Prediction of visual acuity after graft survival in Chinese patients. Int J Ophthalmol.
Dsaek from oct images. Invest Ophthalmol Vis Sci. 2020;13(9):1374.
2012;53(14):43. 101. Price MO, Fairchild KM, Price DA, Price FW Jr.
90. Li JY, Terry MA, Goshe J, Davis-Boozer D, Descemet's stripping endothelial keratoplasty:
Shamie N. Three-year visual acuity outcomes after five-year graft survival and endothelial cell loss.
Descemet’s stripping automated endothelial kerato- Ophthalmology. 2011;118(4):725–9.
plasty. Ophthalmology. 2012;119(6):1126–9. 102. Ang M, Soh Y, Htoon HM, Mehta JS, Tan D. Five-
91. Lanza M, Boccia R, Ruggiero A, Melillo P, year graft survival comparing DESCEMET stripping
Bifani Sconocchia M, Simonelli F, Sbordone automated endothelial keratoplasty and penetrating
S. Evaluation of donor and recipient characteristics keratoplasty. Ophthalmology. 2016;123(8):1646–52.
involved in Descemet stripping automated endo- 103. Letko E, Price DA, Lindoso EM, Price MO, Price
thelial Keratoplasty outcomes. Front Med. 2021; FW Jr. Secondary graft failure and repeat endothe-
8:421. lial keratoplasty after Descemet’s stripping auto-
92. Wacker K, Bourne WM, Patel SV. Effect of graft mated endothelial keratoplasty. Ophthalmology.
thickness on visual acuity after Descemet stripping 2011;118(2):310–4.
endothelial keratoplasty: a systematic review and 104. Birbal RS, Dhubhghaill SN, Bourgonje VJ, Hanko J,
meta-analysis. Am J Ophthalmol. 2016;163:18–28. Ham L, Jager MJ, Böhringer S, Oellerich S, Melles
93. Vasiliauskaitė I, Oellerich S, Ham L, Dapena I, GR. Five-year graft survival and clinical outcomes
Baydoun L, van Dijk K, Melles GR. Descemet of 500 consecutive cases after Descemet membrane
membrane endothelial keratoplasty: ten-year graft endothelial keratoplasty. Cornea. 2020;39(3):290–7.
survival and clinical outcomes. Am J Ophthalmol. 105. Price DA, Kelley M, Price FW Jr, Price MO. Five-
2020;217:114–20. year graft survival of Descemet membrane endothe-
94. Jansen C, Zetterberg M. Descemet membrane lial keratoplasty (EK) versus Descemet stripping EK
endothelial keratoplasty versus descemet stripping and the effect of donor sex matching. Ophthalmology.
automated keratoplasty–outcome of one single sur- 2018;125(10):1508–14.
106. Patel SV. Graft survival and endothelial outcomes 115. Hamzaoglu EC, Straiko MD, Mayko ZM, Sáles CS,
in the new era of endothelial keratoplasty. Exp Eye Terry MA. The first 100 eyes of standardized des-
Res. 2012;95(1):40–7. cemet stripping automated endothelial keratoplasty
107. Maier P, Reinhard T, Cursiefen C. Descemet strip- versus standardized descemet membrane endothelial
ping endothelial keratoplasty—rapid recovery of keratoplasty. Ophthalmology. 2015;122(11):2193–9.
visual acuity. Dtsch Arztebl Int. 2013;110(21):365. 116. Dapena I, Ham L, van Luijk C, van der Wees J,
108. Deshmukh R, Nair S, Ting DSJ, Agarwal T, Beltz J, Melles GR. Back-up procedure for graft failure
Vajpayee RB. Graft detachments in endothelial kera- in descemet membrane endothelial keratoplasty
toplasty. Br J Ophthalmol. 2021;106:1. (DMEK). Br J Ophthalmol. 2010;94(2):241–4.
109. Gerber-Hollbach N, Baydoun L, López EF, Frank 117. Zafar S, Wang P, Woreta FA, Aziz K, Makary M,
LE, Dapena I, Liarakos VS, Schaal S-C, Ham L, Ghous Z, Srikumaran D. Postoperative complica-
Oellerich S, Melles GR. Clinical outcome of rebub- tions in Medicare beneficiaries following endo-
bling for graft detachment after descemet membrane thelial keratoplasty surgery. Am J Ophthalmol.
endothelial keratoplasty. Cornea. 2017;36(7):771–6. 2020;219:1–11.
110. Phillips PM, Phillips LJ, Muthappan V, Maloney 118. Daniell M, Brown KD, Gurr P, Scheerlinck J-P,
CM, Carver CN. Experienced DSAEK surgeon's Dusting G, Sawant O, Titus M, Qiao G. Use of novel
transition to DMEK: outcomes comparing the last hydrogel scaffold to assist Descemet’s membrane
100 DSAEK surgeries with the first 100 DMEK sur- endothelial keratoplasty (DMEK) surgery. Invest
geries exclusively using previously published tech- Ophthalmol Vis Sci. 2020;61(7):3601.
niques. Cornea. 2017;36(3):275–9. 119. Rickmann A, Wahl S, Hofmann N, Knakowski J,
111. Price MO, Jordan CS, Moore G, Price FW. Graft Haus A, Börgel M, Szurman P. Comparison of pre-
rejection episodes after Descemet stripping with loaded grafts for Descemet membrane endothelial
endothelial keratoplasty: part two: the statisti- keratoplasty (DMEK) in a novel preloaded transport
cal analysis of probability and risk factors. Br J cartridge compared to conventional precut grafts.
Ophthalmol. 2009;93(3):391–5. Cell Tissue Bank. 2020;21:1–9.
112. Anshu A, Price MO, Price FW Jr. Risk of corneal 120. Or L, Krakauer Y, Sorkin N, Knyazer B, Ashkenazy
transplant rejection significantly reduced with Z, Gushansky K, Dubinsky-Pertzov B, Gazit I, Einan-
Descemet's membrane endothelial keratoplasty. Lifshitz A. A novel marking technique for descemet
Ophthalmology. 2012;119(3):536–40. membrane endothelial graft using an ophthalmic vis-
113. Heinzelmann S, Böhringer D, Eberwein P, Reinhard coelastic device. Cornea. 2021;40(4):529–32.
T, Maier P. Outcomes of Descemet membrane endo- 121. Fogla R. A novel device to visualize Descemet mem-
thelial keratoplasty, Descemet stripping automated brane during donor preparation for descemet mem-
endothelial keratoplasty and penetrating kerato- brane endothelial keratoplasty. Indian J Ophthalmol.
plasty from a single Centre study. Graefes Arch Clin 2021;69(6):1609–13.
Exp Ophthalmol. 2016;254(3):515–22. 122. Khalid M, Khan FS. Nonlinear DSEK model: a
114. Hos D, Matthaei M, Bock F, Maruyama K, Notara novel mathematical model that predicts stability in
M, Clahsen T, Hou Y, Le VNH, Salabarria A-C, ocular parameters after Descemet’s stripping endo-
Horstmann J. Immune reactions after modern lamel- thelial Keratoplasty. Punjab Univ J Mathematics.
lar (DALK, DSAEK, DMEK) versus conventional 2020;52(4):1–14.
penetrating corneal transplantation. Prog Retin Eye
Res. 2019;73:100768.
Endothelial Keratoplasty: Current
State of the Art
26
Anjulie Gang, Francis W. Price,
and Marianne O. Price

• EK offers a significantly better risk/benefit
ratio than PK. The advent of selective endothelial replacement
• The primary methods of EK are DSAEK/ (EK) has revolutionized treatment of corneal
DSEK and DMEK. endothelial dysfunction, a leading indication for
• DMEK provides the most rapid visual corneal transplantation, because EK can be per-
improvement, least refractive change, and formed through a small (2-mm) incision, is a
least risk of rejection and the need for topical closed eye procedure, and provides faster recov-
corticosteroids. ery and more predictable visual outcomes than
• DSEK is more advantageous in eyes with penetrating keratoplasty (PK). Moreover, just as
large iris defects or aphakia. in cataract surgery, the smaller incision provides
• Trifolded EK grafts unfold easier and are protection against wound/eye rupture from minor
advantageous in the complicated eye with pre- trauma which is a life-long risk after PK or other
vious pars planna vitrectomy, aphakia, or iris large incision surgery like intracapsular cataract
defects. surgery. The smaller EK incisions also preserve
corneal innervation and therefore do not produce
the neurotrophic issues and secondary dry eye
problems often see after PK. The improved
benefit-to-risk ratio with EK allows earlier inter-
vention in the disease process.
The two EK techniques in widespread use are
Descemet membrane endothelial keratoplasty
tains supplementary material available at https://doi. (DMEK) and Descemet stripping automated
org/10.1007/978-3-031-32408-6_26. endothelial keratoplasty (known as DSAEK or
DSEK). Both typically involve the removal of
dysfunctional endothelium and Descemet mem-
A. Gang · F. W. Price
Price Vision Group, Indianapolis, IN, USA brane from the central 8 to 9 mm of the recipient
e-mail: [email protected]; cornea and replacement with healthy donor tis-
[email protected] sue. With DMEK the donor tissue consists of
M. O. Price (*) endothelium and Descemet membrane, thus
Cornea Research Foundation, Price Vision Group, exactly replacing the layers removed from the
Indianapolis, IN, USA host cornea. DSAEK includes a thin layer of pos-
https://doi.org/10.1007/978-3-031-32408-6_26
382 A. Gang et al.
terior donor stroma along with Descemet mem- Penetrating keratoplasty can be a better option
brane and endothelium. Although the thinner than EK for visual rehabilitation of eyes with
DMEK tissue was initially more challenging to endothelial dysfunction accompanied by visually
handle when the technique was first introduced significant stromal scarring or opacification
by Melles in 2006 [1], with subsequent advances, because PK replaces all layers of the cornea.
DMEK has become the preferred EK technique However, if the goal is simply to alleviate edema
worldwide and is a viable option even in difficult and painful bullae, EK is a less invasive option.
eyes with various ocular comorbidities. In eyes with a dislocated intraocular lens
Compared with PK and DSAEK, DMEK has the (IOL) or anterior chamber (AC) IOL, we only
lowest rejection rate [2], requires the smallest perform EK after the lens anomaly is treated first
incision, and provides the fastest visual recovery as either a staged or combined procedure. In eyes
[3]. However, DSAEK can be advantageous in with significant iris abnormalities, aniridia and/or
certain situations, particularly in eyes with chal- aphakia, DSAEK is generally a better choice than
lenging ocular co-morbidity. DMEK because the thicker DSAEK tissue can be
PDEK is a variation of EK with a slightly suture-fixated to the recipient cornea to prevent
thicker donor than DMEK [4]. The PDEK donor dislocation into the posterior chamber, and the
tissue is produced by performing a type-1 bubble pressure changes in the unicameral eye when
and then excising the bubble (DMEK is essen- DMEK donors are manipulated can easily lead to
tially a thinner layer as achieved with a type-2 loss of the donor into the posterior portion of the
bubble). Current limitations of PDEK are the eye. For both DMEK or DSAEK, pull-through
donor prep is more difficult and the size of the techniques are advantageous in aphakic eye or
bubble is limited to 7–8 mm in diameter. eyes with large iris defects.
However, recent studies have suggested tech- Among patient characteristics to consider,
niques to produce 8–9 mm donor diameters for essential blepharospasm or chronic eye rubbing
PDEK and that may allow more widespread use could impede EK attachment because indenting
of this technique. the cornea can lead to graft detachment in the
early post-operative period. Obesity or thyroid
orbitopathy could increase the posterior pressure
Indications during surgery. Finally, it was long thought that
supine positioning for at least the initial 24 h was
Endothelial keratoplasty is the gold standard for necessary to ensure DMEK attachment, but that
treating all types of visually significant corneal is problematic for some patients. Recent studies
endothelial dysfunction. The leading indications have suggested that prolonged supine positioning
in Europe, America, and Australia are Fuchs may not be necessary after all, especially if the
endothelial corneal dystrophy (FECD), corneal patient is discharged with a large residual air or
decompensation following cataract and/or glau- gas bubble to hold the graft in place.
coma surgery, and failure of a prior corneal trans-
plant. In Asia, FECD is less prevalent and
endothelial dysfunction is more often associated Surgical Technique
with trauma from surgery or laser procedures in
eyes with anatomically narrow angles and a shal- DSAEK Tissue Preparation
low anterior chamber. Less common conditions
amenable to treatment with EK include iridocor- Both DMEK and DSAEK donor tissue can be
neal endothelial syndrome (ICE) syndrome, pos- prepared by the surgeon or by an eye bank techni-
terior polymorphous dystrophy (PPMD), and cian, and it can be prepared at the time of surgery
corneal endotheliitis (in the absence of visually or up to several days ahead of time. Occasionally,
significant stromal scarring). the endothelium may be damaged excessively
26 Endothelial Keratoplasty: Current State of the Art 383
during tissue preparation, so preparing the tissue age, higher failure rates, or with type-1 bubbles
ahead of time can prevent uncertainty and save limitations on donor diameter (PDEK).
time on the day of surgery. Multiple variations of direct peeling exist. In
DSAEK/DSEK tissue is usually prepared by the SCUBA technique [7, 8], a Y- hook, or other
mounting a donor corneal-scleral rim on an artifi- instrument, with a blunt smooth tip is used to
cial anterior chamber and dissecting it with score the peripheral Descemet membrane just
microkeratome. Thinner tissue is associated with inside the trabecular meshwork and Schwalbe’s
better visual outcomes so various single- and line for the full circumference of the donor cor-
double-pass techniques have been devised to pro- nea. Trypan blue dye appropriate for ocular use
duce very thin tissue while minimizing the risk of (i.e. Vision Blue 0.6 mg/mL, DORC) is applied
perforation. A microkeratome is expensive, so it for about 30 s to mark the scored edges and reveal
can make sense for an eye bank to purchase one any areas incompletely scored. Trypan blue stains
and prepare tissue for multiple surgeons. This exposed stroma and DM but not intact endothe-
approach also allows the eye bank to absorb the lium. Prolonged exposure to trypan blue at high
cost of any tissue lost in preparation. The rate of concentrations should be avoided because it can
donor loss increases as thinner cuts are attempted be toxic to the endothelium.
with a microkeratome. Manual dissection can be The donor tissue is submerged in corneal stor-
done instead of using a microkeratome and was age solution, and an olive-tipped micro-finger
the original way the donor was prepared, but a (Moria, Antony, France) is glided under the outer
microkeratome generally produces a more uni- edge of DM to lift and free it from the underlying
form dissection plane. We are not aware of any- stroma. As the DM edge is lifted, any tags or tears
one who has tried an anterior peel technique (as are removed with small tiers to prevent tears from
used in deep anterior lamellar keratoplasty extending further centrally. After completely lift-
(DALK) for donor preps, which if feasible would ing the edges, the DM is grasped with forceps
provide a very thin uniform dissection plane [5]. and carefully peeled in quadrants towards the
Femtosecond laser dissection has been tried, center. It is important to watch for localized areas
but tissue applanation is usually required and this of strong adhesion, which appear as horse-shoe
produces compression folds in the posterior cor- shaped tears. These start small and, when detected
nea, resulting in undulating dissection plane after promptly, can be gently lifted free with the micro-
the applanation is released. Also, the laser dissec-finger instrument. Alternatively, peeling can be
tion plane is not as regular or smooth as that attempted from the opposite direction to free
achieved with a microkeratome because of the localized adhesions without propagating larger
collagen fiber arrangement in the posterior cor- tears. After each quadrant is freed, it is floated
nea. Thus, the visual outcomes tend to be back down onto the stroma. After peeling all four
disappointing. quadrants, the donor cornea is transferred to a
cutting block. An 8- to 9-mm diameter trephine is
used to cut through DM and partially into the
DMEK Tissue Preparation stroma with the donor endothelial side up. The
trephine is gently tapped to perforate DM with
Direct peeling from the posterior surface of the only shallow penetration into the stroma. We typ-
donor is the most widely used method of separat- ically leave the donor attached in a small area
ing the donor endothelium and Descemet mem- centrally, and one of three options can be taken.
brane (DM) from the stroma [6–8]. This approach The first traditional option is to peel the donor
does not require any expensive instrumentation completely and load it endothelial side out in the
and can be readily implemented in any part of the desired inserting instrument. The second option
world. Alternatives include hydro- or pneumo- is to leave it attached and place the whole donor
dissection producing either type-1 or 2 bubbles, corneal scleral rim back into the storage solution
but these can result in more endothelial cell dam- to be peeled the next day at surgery. The final
384 A. Gang et al.
option is to fold it into a trifold c onfiguration and on the non-endothelial side during donor tissue
then peel it completely and load it into an IOL preparation, although this can cause some endo-
injector (Fig. 26.1a–g). thelial damage [9]. Asymmetric orientation
marks can also be cut along the edge of the tissue
during preparation. Orientation marks help the
Tissue Orientation Marks surgeon ensure that the tissue is correctly ori-
ented inside the recipient eye with the donor
The final orientation of the donor graft in the eye endothelium facing toward the host iris. However,
is critical, and the endothelium needs to be facing such marks are not needed if the surgeon has
the iris; if it is facing the stroma the graft will be access to intraoperative optical coherence tomog-
nonfunctional. Therefore, with both DSAEK and raphy (OCT) because it provides a cross-sectional
DMEK, many surgeons like to have a Gentian image of the graft configuration, and in our prac-
violet orientation mark, such as an “S”, stamped tice, we have not used orientation marks [10].
a b
c d
Fig. 26.1 DMEK tissue preparation and insertion using inward, and the tissue was pulled to the edge of the scleral
the trifold configuration. (a) After the donor tissue was button. (e) The blue-stained DMEK trifold was pulled into
punched to the desired diameter, typing forceps were used IOL cartridge. (f) The IOL cartridge tip was filled with
to grasp one edge of the tissue and fold it over with the fresh storage solution, and the DMEK trifold was pulled
endothelium facing inward. (b) An arrow shows where into the tip using 23-gage intraocular forceps. (g) The
one-third of the tissue was folded over; also, the corneal- blue-stained DMEK trifold was inserted into the recipient
scleral rim was rotated 180°. (c) The opposite side was eye; the inset at the lower right shows the corresponding
folded over to create a trifold. (d) Both sides of the tissue intraoperative OCT image
were folded into a trifold with the endothelium facing
e f
Fig. 26.1 (continued)
Recipient Preparation caine and intravenous sedation, thereby avoiding

the risks associated with a local block or general
We prescribe topical nonsteroidal anti-anesthesia. We place 4–0 silk scleral traction
inflammatory eye drops for 3 days before sur- sutures superiorly and inferiorly to aid with eye
gery. At the time of surgery, pilocarpine nitrate positioning.
2% eye drops are instilled to constrict the pupil, With both DMEK (Video 26.1) and DSAEK
or carbachol 0.01% can be instilled if EK is not (Video 26.2), we mark the recipient epithelium
being combined with either cataract surgery or with a trephine of the same diameter as that used
pars planna vitrectomy. Pupil constriction helps for the donor graft to provide a reference for
prevents contact between the graft endothelium removal of host DM and endothelium. When EK
and the crystalline or intraocular lens during graft is combined with cataract surgery, we use a tem-
unfolding and positioning and decreases the like- poral 2.0–2.8 mm clear corneal incision. With
lihood of the graft being inserted posterior to the non-cataract cases we use a 2.0-mm scleral tun-
iris. Also, it can help minimize the risk of vitre- nel incision temporally. All cases have 2 short
ous prolapse in pseudophakic eyes with previous self-sealing paracenteses 45° to each side of the
YAG-laser capsulotomies. main incision made with 15° super sharp para-
We almost always perform EK with topical centesis blades.
anesthesia supplemented with intracameral lido-
386 A. Gang et al.
Next, the dysfunctional endothelium and the eye with forceps with single- or multiuse
Descemet membrane are removed from the cen- devices specifically designed for this purpose,
tral recipient cornea. Use of air or viscoelastic in pulled in with sutures, pushed/inserted with a
the anterior chamber (AC) improves visualiza- needle or forceps, or inserted with specially
tion and helps prevent corneal edema from designed injectors. Certain bimanual techniques
increasing during the DM stripping process. In involve pulling the tissue through a device that
cases with combined cataract surgery, stripping causes it to curl, using micro-forceps introduced
occurs after the IOL is in place and viscoelastic from the opposite side of the eye (Busin glide
fills the anterior chamber. A reverse Price-Sinskey (Moria); Endoglide (Coronet)). This allows the
hook (Moria, Antony, France) is introduced surgeon to hold onto the tissue until an air bubble
through the paracentesis and used first to score can be injected beneath it to press it against the
DM along the epithelial reference mark, then to recipient cornea, which can be helpful in cases of
strip DM from the planned graft area and discard aphakia with large pupils, large iris defects, or
it. If viscoelastic was used in the AC it must be aniridia.
thoroughly evacuated at this point with phaco- DMEK tissue naturally tends to curl into a
emulsification bi-manuals, because retained vis- single or double scroll configuration with the
coelastic at the graft/host interface impedes endothelium facing outward when submerged in
attachment and impairs vision. Trypan blue dye fluid (whereas in air it crumples up). The scrolled
is injected into the AC and then irrigated out after tissue can be sucked into a glass tube or loaded
about 30 s to help show any residual strands of into a plastic intraocular lens cartridge or other
DM, loose stromal fibers, or areas of incomplete insertion device for injection into the recipient
stripping. Exposed DM stains dark blue, exposed eye. Alternatively, the tissue can be folded into a
stroma stains light blue, and areas with attached trifold configuration (like a pamphlet folded into
DM will have little to no staining. Loose residual thirds) with the endothelium facing inward to
tags of DM can be removed with phacoemulsifi- facilitate unfolding after insertion.
cation bi-manuals and loose stromal fibers with During graft insertion, it is important to be
intraocular 23-gage scissors. aware of the graft orientation and preferable to
With DMEK it is common to create an inferior insert the tissue with the endothelium facing
peripheral iridotomy (PI) to minimize the risk of downward (toward the recipient iris). Graft orien-
pupillary block with prolonged use of an air or tation can be ascertained by looking at an orienta-
gas bubble to hold the graft in place. The PI can tion mark that was added during donor
be created with micro-scissors, followed by aspi- preparation, or it can be determined by viewing
ration of posterior residual iris pigment using the the graft configuration in cross-section using a
bi-manual aspiration tip of the bi-manuals to hand-held slit beam or intraoperative OCT [10,
ensure patency. Alternatively, a PI can be created 11], because a DMEK graft always curls with
ahead of time with a laser. Hydration is used to endothelium facing outward. If the graft is not
seal the paracenteses. correctly oriented in the eye, it can be flipped or
rotated with short bursts of balanced salt solution
(BSS) aimed slightly under the graft to create a
Graft Insertion and Positioning fluid wave.
An advantage of the trifold configuration is
DSAEK grafts are typically folded into a 60/40 that it tends to open naturally as the AC is grad-
taco or a trifold endothelium-inward configura- ually deepened by injection of BSS to provide
tion for insertion. The tissue can be pulled into space for the leaflets to unfold, and for a short
time there is often some memory to the trifold, Postoperative Management

so the graft does not immediately revert to an
endothelium- outward scroll configuration. If the eye is left nearly full of air/gas at the end of
Deepening the AC too quickly or too much the case, as is common with DMEK, the patient
should be avoided as this can allow the trifold should have a slit lamp exam and intraocular
to revert to a scroll or double scroll pressure (IOP) measurement about an hour later,
configuration. to ensure there is no pupillary block [13]. Patients
To uncurl scrolled DMEK tissue short bursts should be advised not to rub or push on the eye
of BSS are used. Long continuous injections of because that could dislocate the graft.
fluid are avoided, because that would raise the Topical antibiotic eye drops are typically pre-
pressure in the eye and extrude the graft out of scribed for a week and topical corticosteroids
the incision. As the graft is uncurled, the orien- (beginning 4 times daily and tapering to once
tation is checked to make sure it is unscrolling daily) are prescribed indefinitely to prevent graft
endothelium down and DM up. If the graft is rejection. It is common to start patients on a rela-
upside down, then short bursts of BSS under the tively strong topical corticosteroid, such as pred-
graft will flip it over. Once correct orientation is nisolone acetate 1% or dexamethasone. Studies
confirmed, the AC is shallowed, and a small air have shown that after DMEK it is safe to switch
bubble is placed beneath the tissue with a Caucasian patients to a weaker topical steroid,
30-gauge needle to hold the graft open and in such as fluorometholone 0.1% or loteprednol eta-
the correct orientation while it is centered on the bonate 0.5% after the first month. This signifi-
area of recipient stripping. A cannula is used to cantly reduces the risk of steroid-induced ocular
tap the recipient corneal surface at the edges of hypertension or glaucoma [14, 15]. Patients with
the scroll to open it. The graft is centered by darkly pigmented irises may be more prone to
gently stroking the corneal surface with a can- inflammation and graft rejection, and therefore,
nula (like swinging a golf club) to induce fluid we recommend tapering topical corticosteroids
waves that shift the tissue in the desired more slowly [16].
direction. Resumption of daily activities depends on
After the tissue is centered and unfolded, air incision size and contour. Like cataract surgery,
or gas, such as 10% C3F8 or 20% SF6 [12], is normal activities can be resumed within a couple
injected at the limbus with a 30-gauge needle, of weeks with small corneal incisions or self-
bevel facing up, to achieve a 90% air/gas fill. sealing scleral tunnel incisions.
Care should be taken not to increase the pressure Visual rehabilitation is so rapid with DMEK
to the point of compromising the optic nerve or that patients with bilateral endothelial dysfunc-
its blood supply. This can be confirmed by asking tion can have the second eye treated within a
the patient if they can see the microscope light week of the first eye, with or without combined
come and go as the surgeon waves a hand beneath cataract surgery [17]. Rapid sequential DMEK
it, and if they can see any photopsia. Fluid and air has not been found to increase the risk of immu-
levels in the AC can be adjusted as needed to nologic rejection. The typical postoperative
achieve the appropriate fill. appearance at 1 day, 5 days, and 1 month after
DMEK is shown in (Fig. 26.2a–e).
388 A. Gang et al.
a b c
d e
Fig. 26.2 DMEK postoperative images. (a) A postopera- sponding mild inferior corneal edema. The uncorrected
tive day 1 slit lamp image showed that a residual air bub- distance visual acuity was 20/50 at day 5. (d) A preopera-
ble filled approximately 50% of the anterior chamber with tive specular microscopy image showed dark areas of gut-
the patient sitting upright. (b) A postoperative day 5 slit tae, characteristic of Fuchs dystrophy. (e) At 1-month,
lamp image showed no residual air remaining in the ante- specular microscopy showed the central DMEK endothe-
rior chamber. (c) A postoperative day 5 slit beam image lial cell density was 2518 cells/mm2; also, the corrected
showed the central cornea was compact and clear, and distance visual acuity was 20/20 Snellen
there was a small inferior edge detachment with corre-
Complications tamponade active intraocular bleeding. Next, the

AC should be thoroughly washed out with BSS to
I ntraoperative Bleeding and Fibrin remove any remaining heme, fibrin, viscoelastic,
Formation or air. If bleeding recurs, the above steps should
be repeated. It is important to wait until the bleed-
Heme in the AC can occur at any step of the case: ing is controlled before inserting the graft.
during surgical peripheral iridotomy creation, Fibrin can sometimes develop in the anterior
from iatrogenic iris damage/iridodialysis during chamber making it difficult to unfold or manipu-
insertion of instruments or insertion of the donor late the graft, as can blood. Fibrin is not visible.
tissue, or from blood entering from the main Use of an anterior chamber maintainer helps
wound. To address blood in the AC, the first step washout any heme as well as minimizing the
is to stop active bleeding by temporarily increas- chance of fibrin accumulation in the eye with
ing IOP for several minutes. Filling the AC with insertion of the graft. We remove the anterior
viscoelastic or a full air fill is an alternative way to chamber maintainer once the graft is in the eye.
Graft Detachment failure is corneal decompensation after initial

clearing and can be caused by immunologic
Graft detachment is the most common early post- rejection or endothelial decompensation.
operative complication. It is usually detected by It is straightforward to remove and replace a
slit lamp examination and can be confirmed with failed EK graft and best to do so promptly, before
anterior segment OCT imaging. Strategies to long-term edema results in stromal changes that
reduce the risk of DMEK or DSAEK detachment could impair vision [18].
include ensuring a firm gas or air tamponade,
careful wound construction to preclude wound
leaks, and reminding patients to avoid rubbing Immunologic Rejection
the eye.
Strategies specifically designed to promote Immunologic rejection of an EK graft is charac-
DSAEK attachment include massaging the sur- terized by AC cell reaction, corneal edema, a
face of the host cornea to remove residual fluid rejection line, or keratic precipitates.
from the host/graft interface, creation of small Immunologic rejection rates vary with the amount
venting incisions in the mid-periphery of the host of donor tissue implanted. Rejection rates are
cornea to allow fluid escape, and scraping the lowest with DMEK (1% within 2 years), some-
periphery of the host corneal bed. Intraoperative what higher with DSAEK, and highest with PK
OCT aids in detecting pockets of residual inter- (20% within 2 years) [2]. Compared with PK
face fluid. With DSAEK, partial detachments can rejection episodes, EK rejection episodes tend to
be watched and usually seal down over time be milder and less likely to result in graft failure
without intervention. [19]. Treatment consists of increased topical cor-
DMEK grafts do not adhere well to host DM, ticosteroid use.
so the graft is typically slightly undersized rela-
tive to the host stripped area. Partial edge
detachments are more common with DMEK Less Common Complications
than DSAEK. Partial detachments can be treated
by reinjecting an air or gas bubble at the slit Other complications may include cystoid macu-
lamp or in a minor procedure room. It is easiest lar edema, subcapsular lens opacities, cataract
to add air or gas to an existing residual bubble progression, posterior synechiae, pupillary
because this ensures that the new bubble is not abnormalities, iris ischemia, fungal infections
being injected between the graft and host cor- from donor tissue, and calcification of hydro-
nea. Determining when to reinject air is very philic IOLs.
subjective with different surgeons. Typical crite-
ria for DMEK re-bubbling are when the detach-
ment affects more than one-third of the graft Anatomy-Related Challenges
area, is increasing in size, or seems to be affect- and Technique Adaptations
ing vision. Full DMEK detachments are usually
taken back to the operating room for reposition- Crystalline Lens
ing and attachment.
In patients with endothelial dysfunction and visu-
ally significant lens opacity, EK can be combined
Graft Failure with or staged before or after cataract surgery.
Many patients prefer the convenience of a single,
Early graft failure is defined as initial failure to combined procedure. To achieve the best refrac-
clear and can be caused by surgical trauma, graft tive outcomes and uncorrected vision, EK can be
detachment, upside-down positioning, or be combined with implantation of an IOL that can
attributable to the donor tissue. Secondary graft be adjusted postoperatively, or EK can be staged
390 A. Gang et al.
first to eliminate the corneal edema before IOL Post-Vitrectomy

selection [20].
Patients under 50 years of age without a visu- In eyes with a posterior vitrectomy, an AC air
ally significant cataract and an adequately deep bubble has a tendency to move posteriorly into
anterior chamber are generally good candidates the large vitreous cavity, in accordance with
for phakic DMEK. A shallow anterior chamber Laplace’s law. This is especially a problem in
makes donor positioning and manipulation diffi- eyes with large iris defects or aniridia. It is impor-
cult and increases the risk of iatrogenic cataract tant to have the eye relatively firm prior to inject-
formation with intraocular surgery. Pilocarpine ing air to decrease the tendency of the air going
eye drops are instilled preoperatively to provide posteriorly.
miosis and to protect the crystalline lens. Scoring Also, for DMEK where we often rely on ante-
and stripping are done under viscoelastic to pro- rior chamber shallowing to hold the donor open
vide increased anterior chamber stability and after it is unfolded, post-vitrectomy eyes may not
decrease the risk of premature cataract formation. shallow at all. Therefore, pull-in techniques are
Patients over 50 years of age are more likely to helpful and with a trifold, air can be injected
experience cataract formation and progression under the graft before letting go of the graft once
after EK or any other type of intraocular surgery, it is pulled in.
so the pros and cons of a staged vs. combined Figure 26.3a–c show DMEK in a 63-year-old
approach should be carefully discussed with them. female with a history of aphakia after cataract
surgery. A pars plana victrectomy with placement
of a secondary IOL using the Yamane intrascleral
Post YAG-Capsulotomy haptic fixation technique was staged 1 month
before DMEK.
Preoperative examination should include evalua-
tion of the posterior capsule for signs of a capsu-
lotomy. An open capsule, especially a large Severe Host Corneal Edema
posterior capsulotomy, can increase the chance of
vitreous prolapse into the AC during surgery. If Severe corneal edema or severe anterior stromal
pupil peaking or vitreous to the wounds is or subepithelial scarring can limit the view dur-
observed during EK and vitreous prolapse is sus- ing surgery increasing the risk of complications.
pected, a thorough anterior vitrectomy should be Intraoperative OCT is particularly helpful
completed prior to graft insertion because vitre- because it can image through a cloudy cornea.
ous in the AC will interfere with graft manipula- Removing the epithelium can help improve the
tion and unfolding. view into the eye, but we typically do not remove
a b c
Fig. 26.3 Intraoperative images of DMEK in an eye with DMEK tissue. The previous victrectomy made it difficult
bullous keratopathy and a previous pars plana victrec- to sufficiently shallow the anterior chamber, so the central
tomy. (a) Image showing the hazy cornea at the start of the cornea was indented to help unfold and hold the DMEK
case. (b) Image showing insertion of the blue-stained tissue open; the indentation of the central cornea is more
DMEK trifold. (c) Image showing unfolding of the readily apparent in the OCT image (inset lower right)
a b
c d
Fig. 26.4 Intraoperative images of a DMEK procedure in beneath the graft to help hold it open in the correct orien-
an eye with severe corneal edema. (a) Image showing the tation. The inset OCT image was used to reconfirm cor-
severe corneal edema and diffuse bullae at the start of the rect orientation and centration before increasing the
case. (b) Image taken immediately after insertion of the bubble size. (d) The shadow cast on the OCT image, by
DMEK trifold into the host anterior chamber. The corre- the cannula used to inject balanced salt solution, was used
sponding intraoperative OCT image (inset lower right) to help check whether the graft edges were centered on the
showed that the left side of the graft was scrolled up previously scored and stripped area of the recipient cor-
toward the host cornea and the right third was still folded nea. Centration could be adjusted by gently stroking the
under in the trifold configuration, indicating that the graft corneal surface with the cannula to induce fluid waves to
was correctly oriented with the donor endothelium facing shift the tissue in the desired direction
the host iris. (c) Next, a small air bubble was injected
any epithelium until the donor is in the eye to tion marks on the graft and impedes assessment
minimize the chance of dragging some recipient of graft orientation. With DMEK, use of a trifold
epithelium into the eye during graft insertion. configuration with a pull-through technique is
Poor visualization while scoring and stripping helpful in such eyes because this approach mini-
the host DM increases the risk of disrupting the mizes the need for graft manipulation in the
posterior stroma, which may impair DMEK AC. Ensuring that the tissue is thoroughly stained
attachment, whereas roughened posterior stroma with trypan blue is helpful, and intraoperative
does not affect DSAEK attachment. Poor visual- OCT is helpful for confirming graft orientation
ization also makes it difficult to discern orienta- and centration as shown in (Fig. 26.4a–d).
392 A. Gang et al.
ens, Iris, or Anterior Chamber

L eyes after DMEK and especially after DSAEK
Abnormalities because it is thicker.
In eyes with an AC IOL, some surgeons prefer
Tamponade with an intracameral bubble is to leave the lens in place if it is correctly posi-
required for EK adherence. The bubble and graft tioned and the AC depth is sufficient. However,
are normally retained in the anterior chamber by other surgeons prefer to exchange the lens for an
the lens/iris diaphragm. When this diaphragm is iris- or scleral-fixated lens either ahead of time or
compromised, it may not be possible to ade- in combination with EK.
quately create or maintain the bubble, and pre-
cautions should be taken to prevent escape of the
stripped host DM and/or graft to the posterior Previous Failed PK
chamber. Replacement or repositioning of a dis-
located IOL, treatment of aphakia with implanta- EK provides significantly faster visual rehabilita-
tion of a sulcus- or scleral-fixated IOL, and tion than a PK regraft and is the preferred
treatment of iris abnormalities with iridoplasty, approach if the refractive outcomes were accept-
pupilloplasty, or iris reconstruction should be able with the original PK. The 6-month median
considered before performing EK. corrected distance vision was 20/30 Snellen with
DSAEK is usually preferable to DMEK in DMEK and 20/50 Snellen with DSAEK in series
aphakic eyes or eyes with significant iris open- performed to rescue failed PK grafts [22].
ings. With DSAEK, host DM does not need to be The principal considerations for EK planning
stripped unless it has guttae or scarring. The are that the PK incision has limited wound
DSAEK graft can be inserted over a glide that strength, and the host posterior corneal surface
covers the iris or it can be pulled into the eye with may be irregular along the PK incision.
micro-forceps, which allow the surgeon to hold Preoperative assessment with anterior segment
onto the graft until it is secured with a bubble. OCT helps reveal irregularities and step offs that
Also, a temporary fixation suture should be may affect graft unfolding and attachment.
placed in a DSAEK graft at 12 o’clock to secure With DSAEK, it is not necessary to strip the
it to the host cornea during the early postopera- host DM unless irregularities that could affect
tive period when there are large iris defects or vision, such as guttae or scarring, are present
fixed pupils. Finally, pupillary block is not a con- because DSAEK will easily adhere to host
cern in eyes with large iris defects, so the AC can DM. With DMEK in primary cases, it has been
be left completely full of air or gas at the end of observed that the DMEK graft does not adhere
the case to promote attachment. well to unstripped areas of DM, so in early
In eyes with a shallow or crowded AC, the AC series for treatment of failed PK, it was stan-
can be filled with viscoelastic to improve stabil- dard practice to strip DM when placing a
ity prior to stripping host DM. DMEK grafts DMEK graft. However, stripping DM in post
tend to be easier to unfold in a relatively shallow PK eyes can be difficult, leaving shreds of DM
AC. On the other hand, with DSAEK, continu- that must be removed with forceps and/or dis-
ous infusion with an AC maintainer helps keep turbing the posterior stroma. One option is to
the AC formed during graft insertion, and use of use a femtosecond laser to do the scoring, so
glide insertion techniques have been associated there is a more precise demarcation between
with better outcomes in Asian eyes, which are the area of DM removed and that which is left
more likely to have a shallow AC and iris pro- [23]. Another option is to not strip DM from the
lapsing [21]. prior PK; this approach should leave a smoother
Any preoperative anterior synechiae should be posterior surface and works well when com-
lysed before inserting EK tissue. Postoperative bined with use of long-acting gas to help the
anterior synechiae may develop in overly shallow donor stay in contact with the PK DM until it
attaches [12, 24, 25]. In fact, we recommend Conclusions

always using a long-acting gas with DMEK
after failed PK, whereas we use air for most In conclusion, EK has revolutionized treatment
other DMEK cases. We have also found that of corneal endothelial dysfunction and is cur-
keeping the DMEK graft diameter somewhat rently the gold standard treatment. Appropriate
smaller than that of the previous PK facilitates technique modifications allow successful use in
attachment [22]. eyes with various ocular comorbidities.
Take Home Notes

Previous Glaucoma Surgery • Topical anesthesia is the preferred method for
EK surgery.
EK provides better visual outcomes and faster • The risk of rejection is least with DMEK.
visual rehabilitation than PK in eyes with prior • The need for topical corticosteroids is least
glaucoma surgery. In eyes with a glaucoma tube with DMEK leading to less steroid-induced
shunt, the proximity of the tube to the cornea glaucoma.
should be assessed preoperatively because a long • EK works well for failed PK, leading to faster
tube could pose challenges during EK manipula- visual recovery, less tissue damage, and less
tion and centration and damage the donor endo- risk of rejection compared to repeat PK, if the
thelium. The tube can be trimmed to an original PK has acceptable ocular surface
appropriate length with micro-scissors during EK characteristics.
surgery if the tube placement is acceptable.
However, in cases where the tube needs to be
repositioned, we prefer to do that a month ahead References
of time to give the eye time to heal and get past
the risk of hypotony that sometimes occurs with 1. Melles GR, Ong TS, Ververs B, van der Wees
repositioning tubes to posterior chamber or pars J. Descemet membrane endothelial keratoplasty
(DMEK). Cornea. 2006;25:987–90.
planna insertions. 2. Anshu A, Price MO, Price FW Jr. Risk of cor-
The presence of a superior glaucoma tube neal transplant rejection significantly reduced with
shunt or trabeculectomy can allow air to escape Descemet’s membrane endothelial keratoplasty.
from the eye as the AC is being filled, and further Ophthalmology. 2012;119:536–40.
3. Chamberlain W, Lin CC, Austin A, Schubach N,
outflow can occur when the patient sits up. To Clover J, McLeod SD, Porco TC, Lietman TM, Rose-
help promote graft attachment, the AC can be left Nussbaumer J. Descemet endothelial thickness com-
almost completely full of air or gas in eyes with parison trial: a randomized trial comparing ultrathin
an aqueous shunt. Trabeculectomies are a bit Descemet stripping automated endothelial kerato-
plasty with Descemet membrane endothelial kerato-
more complicated, because sometimes the air can plasty. Ophthalmology. 2019;126:19–26.
block the filtration site leading to increased IOP 4. Agarwal A, Dua HS, Narang P, Kumar DA, Agarwal
or loss of the bleb. The IOP should be checked A, Jacob S, Agarwal A, Gupta A. Pre-Descemet’s
1–2 h afterward to ensure adequate filtration and endothelial keratoplasty (PDEK). Br J Ophthalmol.
2014;98:1181–5.
the bubble should be reduced if needed. Use of a 5. Malbran E, Stefani C. Lamellar keratoplasty in cor-
long-acting gas, such as 10% C3F8 instead of air, neal ectasias. Ophthalmologica. 1972;164:59–70.
can also help promote attachment. When neces- 6. Lie JT, Birbal R, Ham L, van der Wees J, Melles
sary to prevent bubble escape, an aqueous tube GR. Donor tissue preparation for Descemet mem-
brane endothelial keratoplasty. J Cataract Refract
can be temporarily plugged with viscoelastic, Surg. 2008;34:1578–83.
which will spontaneously fall away when the 7. Price MO, Giebel AW, Fairchild KM, Price FW Jr.
patient sits up. Descemet’s membrane endothelial keratoplasty: pro-
394 A. Gang et al.
spective multicenter study of visual and refractive 17. McKee Y, Price MO, Gunderson L, Price FW Jr.
outcomes and endothelial survival. Ophthalmology. Rapid sequential endothelial keratoplasty with and
2009;116:2361–8. without combined cataract extraction. J Cataract
8. Tenkman LR, Price FW, Price MO. Descemet mem- Refract Surg. 2013;39:1372–6.
brane endothelial keratoplasty donor preparation: 18. Letko E, Price DA, Lindoso EM, Price MO, Price
navigating challenges and improving efficiency. FW Jr. Secondary graft failure and repeat endothe-
Cornea. 2014;33:319–25. lial keratoplasty after Descemet’s stripping auto-
9. Veldman BP, Dye PK, Holiman JD, Mayko ZM, mated endothelial keratoplasty. Ophthalmology.
Sales CS, Straiko MD, Galloway JD, Terry MA. The 2011;118:310–4.
S-stamp in Descemet membrane endothelial kerato- 19. Price MO, Price FW Jr. Descemet stripping endo-
plasty safely eliminates upside-down graft implanta- thelial keratoplasty: fifteen-year outcomes. Cornea.
tion. Ophthalmology. 2016;123:161–4. 2022;42:449–55.
10. Price FW Jr. Intraoperative optical coherence 20. Price MO, Pinkus D, Price FW Jr. Implantation
tomography: game-changing technology. Cornea. of presbyopia-correcting intraocular lenses staged
2021;40:675–8. after Descemet membrane endothelial kerato-
11. Burkhart ZN, Feng MT, Price MO, Price plasty in patients with Fuchs dystrophy. Cornea.
FW. Handheld slit beam techniques to facilitate 2020;39:732–5.
DMEK and DALK. Cornea. 2013;32:722–4. 21. Mehta JS, Por YM, Poh R, Beuerman RW, Tan
12. Guell JL, Morral M, Gris O, Elies D, Manero D. Comparison of donor insertion techniques for
F. Comparison of sulfur hexafluoride 20% versus air Descemet stripping automated endothelial kerato-
tamponade in Descemet membrane endothelial kera- plasty. Arch Ophthalmol. 2008;126:1383–8.
toplasty. Ophthalmology. 2015;122:1757–64. 22. Pasari A, Price MO, Feng MT, Price FW Jr. Descemet
13. Gonzalez A, Price FW Jr, Price MO, Feng membrane endothelial keratoplasty for failed pen-
MT. Prevention and management of pupil block etrating keratoplasty: visual outcomes and graft sur-
after Descemet membrane endothelial keratoplasty. vival. Cornea. 2019;38:151–6.
Cornea. 2016;35:1391–5. 23. Sorkin N, Mimouni M, Santaella G, Trinh T,
14. Price MO, Price FW Jr, Kruse FE, Bachmann BO, Cohen E, Einan-Lifshitz A, Chan CC, Rootman
Tourtas T. Randomized comparison of topical pred- DS. Comparison of manual and femtosecond laser-
nisolone acetate 1% versus fluorometholone 0.1% in assisted Descemet membrane endothelial keratoplasty
the first year after Descemet membrane endothelial for failed penetrating keratoplasty. Am J Ophthalmol.
keratoplasty. Cornea. 2014;33:880–6. 2020;214:1–8.
15. Price MO, Feng MT, Scanameo A, Price FW Jr. 24. Alió Del Barrio JL, Montesel A, Ho V, Bhogal
Loteprednol etabonate 0.5% gel vs. prednisolone M. Descemet membrane endothelial keratoplasty
acetate 1% solution after Descemet membrane endo- under failed penetrating keratoplasty without host
thelial keratoplasty: prospective randomized trial. Descemetorhexis for the management of secondary
Cornea. 2015;34:853–8. graft failure. Cornea. 2020;39:13–7.
16. Price MO, Jordan CS, Moore G, Price FW Jr. Graft 25. Alió Del Barrio JL, Bhogal M, Ang M, Ziaei M,
rejection episodes after Descemet stripping with Robbie S, Montesel A, Gore DM, Mehta JS, Alió
endothelial keratoplasty: part two: the statistical anal- JL. Corneal transplantation after failed grafts: options
ysis of probability and risk factors. Br J Ophthalmol. and outcomes. Surv Ophthalmol. 2021;66:20–40.
2009;93(93):391–5.
Corneal Endothelial Cell Transfer
27
Shigeru Kinoshita, Morio Ueno,
and Chie Sotozono
istorical Pathway Leading

H neal dystrophy, a non-inflammatory sporadic, or
to Endothelial Cell Transfer autosomal dominant disorder that can ultimately
lead to blindness if left untreated [8]. Moreover, a
Due to recent developments in corneal endothe- clinical trial was recently initiated to investigate
lial transplantation strategies, the methods the safety and efficacy of topically applying Rho-
applied for the treatment of corneal endothelial associated protein kinase (ROCK)-inhibitor eye
dysfunction and failure are evolving and are now drops for the promotion of corneal endothelial
beginning to shift away from penetrating corneal wound healing post DSO [9, 10].
transplantation to corneal endothelial transplan- As an alternative to corneal transplantation or
tation [1, 2]. Current transplantation procedures DSO, groundbreaking corneal regenerative-
include Descemet stripping automated endothe- medicine strategies have now introduced the next
lial keratoplasty (DSAEK) [3–5] and Descemet generation of state-of-the-art therapeutic path-
membrane endothelial keratoplasty (DMEK) [6, ways for the treatment of corneal endothelial dys-
7], and more recently, Descemet stripping only function and endothelial failure. To that end, and
(DSO). Of those, DSO is a novel procedure that in concert with the latest biological knowledge of
strips away the Descemet membrane with corneal human corneal endothelial cells (CECs) (HCECs)
guttae in the central region of the cornea and has and cutting-edge cell-culture technology, cell-
been found to be somewhat effective for treating based therapy has been developed for corneal
cases afflicted with mild Fuchs endothelial cor- endothelial dysfunction and failure. One such
innovative pathway involves the surgical transfer
of cultured HCECs (cHCECs), termed “HCEC-
Injection Therapy,” a regenerative-medicine ther-
org/10.1007/978-3-031-32408-6_27. apeutic concept that provides several advantages,
such as being minimally invasive, the ability to
S. Kinoshita (*) intraoperatively supply possibly less-damaged
Department of Frontier Medical Science and nonaged CECs to the posterior surface of the cor-
Technology for Ophthalmology, Kyoto Prefectural nea, and a vastly improved optical quality of the
University of Medicine, Kyoto, Japan cornea post surgery due to no additional corneal
M. Ueno · C. Sotozono tissue being implanted [11, 12]. Since the CECs
Department of Ophthalmology, Kyoto Prefectural obtained from just one donor eye can be used to
University of Medicine, Kyoto, Japan
e-mail: [email protected]; subsequently create enough cHCECs to treat a
[email protected] large number of patients, another benefit of the
https://doi.org/10.1007/978-3-031-32408-6_27
396 S. Kinoshita et al.
procedure is that it holds the promise of ultimately Since the transplantation of cHCECs in vivo does
ending the worldwide shortage of donor corneas, not require constant cell proliferation, surgical
a problem that persists annually [13]. strategies for treating corneal endothelial dys-
function must be considered differently from
those used for the treatment of limbal deficiency
CEC Failure and Repair that require an adequate supply of corneal epithe-
lial stem cells [27].
Corneal endothelium comprises of a single layer
of CECs that lines the posterior surface of the
cornea [14]. In normal healthy eyes, CECs are HCEC-Injection Therapy Concept
arrested in the G1-phase of the cell cycle and
rarely proliferate due to cell-to-cell contact inhi- Currently, there are at least two different
bition and the high concentration of transforming regenerative-medicine cell-based therapeutic
growth factor beta 2 (TGF-β2) in the aqueous concepts applied for the treatment of damaged
humor [15]. Thus, in cases in which CECs corneal tissue. One is the release of beneficial
migrate and/or enlarge, it is more likely that the small molecules via the transfer of cHCECs to
migration and enlargement is in response to the damaged tissue, which promotes functional
wound healing than cell division [16]. In corneal restoration of the healthy tissue via cell regenera-
endothelium, the CEC density slowly decreases tion and reorganization. The other is a “true
with age, even in normal healthy subjects [17]. replacement” of damaged CECs via the surgical
Corneal endothelial dysfunction occurs due to an transfer of cHCECs. The novel HCEC-injection
impairment of pump and barrier function result- therapy that we recently developed aims at the
ing from the abnormality of CECs with guttae latter concept of a regenerative-medicine cell-
formation and/or a decrease of CEC density in based therapeutic approach [11, 12]. One funda-
response to a wide variety of diseases, such as mental question that needs to be addressed is
Fuchs endothelial corneal dystrophy [18], pseu- whether or not constant cell proliferation is
doexfoliation syndrome [19], and cytomegalovi- essential for maintaining healthy corneal endo-
rus corneal endothelitis [20], as well as thelial function following the transfer of the
intraoperative corneal endothelial trauma that cHCECs. To our surprise, the answer to that
can occur and ultimately lead to complete cor- question is “NO!,” as corneal endothelial dys-
neal endothelial failure in cases undergoing laser- function results from the loss of the physiological
iridotomy ophthalmic surgery [21], cataract function of the CEC layer due to a depletion or
surgery [22], glaucoma surgery [23], and vitreo- malfunction of the CECs themselves, and not due
retinal surgery [24]. Thus, in cases afflicted with to endothelial stem-cell deficiency. Thus, our
corneal endothelial dysfunction and failure investigations have revealed that there is no need
resulting from corneal endothelial disease, there for stem or progenitor cells once the cHCECs are
are several different “soils” in the anterior cham- surgically transferred to the posterior surface of
ber environment [25, 26]. Although the existence the cornea.
of stem/progenitor cells in in vivo HCECs has yet
to be proven, it is theorized that these endothelial
cells likely retain their proliferative ability cHCECs for Clinical Use
in vitro [15]. Furthermore, studies have shown
that in cases of corneal endothelial failure, cor- Numerous previous studies have been con-
neal endothelial transplantation using a central- ducted to investigate the laboratory procedures
cornea donor graft can sufficiently restore corneal used for the culture of HCECs, as they are
transparency for a substantial period of time. known to be difficult to proliferate in vitro. For
27 Corneal Endothelial Cell Transfer 397
Fig. 27.1 Schematic presentation of the procedures from cell culture to cell transfer in HCEC-Injection Therapy
example, for the cHCECs used in our novel For practical application, donor corneas
HCEC-injection therapy, we found that in obtained from young donors are used in in vitro
order to closely mimic HCECs in vivo, it is culture material and are cultured at a cell pro-
essential to create cells with a high (i.e., 90%) cessing center under the standard operating pro-
purity rate of mature- differentiated cells to cedure that conforms to the good manufacturing
obtain the optimal surgical outcome and a practices guidelines. Cell lots for clinical appli-
higher postoperative CEC density [28]. For the cation are examined to verify that they meet the
creation of such cells, merely the expression of strict criteria for clinical application, and
(Na+- K+)-ATPase and ZO-1 is not enough, as enzyme-linked immunosorbent assay (ELISA)
that is fundamental for defining cell maturation and/or flow cytometric analysis are performed to
by cell density and several cell surface mark- verify the above-described biological character-
ers, etc. [29–40]. Thus, the mature-istics of the cells. A sterile suspension of
differentiated cHCECs created for use in our cHCECs is then prepared into a small container
HCEC-injection therapy must express distinct several hours prior to the HCEC-injection ther-
cell-surface markers such as CD166+, CD44−/ apy being performed. Currently, the number of
dull, CD24-, CD26-, and CD105−/dull to cHCECs to be injected in 300 μL of modified
closely mimic in vivo HCECs [28]. Opti-MEM™ I Reduced Serum Medium
In addition, in order to verify that the (Thermo Fisher Scientific) is supplemented with
cHCECs are suitable for clinical application, it ROCK-inhibitor Y-27632 is 1.0 × 106 cells
must be confirmed via bacterial testing, viral (Fig. 27.1) [11, 12, 28].
testing, and the findings of a low endotoxin con-
centration that there is no mycoplasma contami-
nation of the cells and that the final culture Surgical Procedure Used for HCEC-
media is completely sterile [11]. Since many Injection Therapy
cultured cells are prepared from a single donor
cornea and are provided to many patients as For patients undergoing HCEC-injection therapy,
“one lot,” that single lot can bring not only a 1.6-mm incision at the corneal limbus is first
effective clinical results but also adverse events created under local anesthesia. Next, a silicone
if those strict safety guidelines are not followed. needle is used to remove abnormal extracellular
Thus, a great deal of care is required to assure matrix on the patient’s Descemet membrane and/
safety. Incidentally, mature- differentiated or degenerated CECs in an 8-mm-diameter area
cHCECs are not tumorigenic and have no chro- of the posterior surface of the cornea. Post
mosomal abnormalities [30, 41]. removal and full collapse of the anterior cham-
ber, all of the prepared cHCECs in the suspension 601 to 2067 cells/mm2), a decrease in the coef-
are injected into the anterior chamber using a ficient of variation, and an increase of cell hex-
26-gauge needle with a dead-spaced free syringe agonality, thus indicating that at 5-years
(Video 27.1). Immediately after post injection, postoperative, the CECs at the posterior corneal
the patients are placed in a “face-down” position surface tended to be more biophysically stable
for 3 h to enhance the adhesion of the injected than those observed at the early postoperative
cells [11]. period. Compared with the previously published
Following surgery, all patients receive both data regarding the surgical outcomes of DSAEK
systemic and topical administrations of steroids and DMEK at 5-years postoperative, the find-
to inhibit acute innate immunity-related inflam- ings in our pilot study showed that our novel
mation and/or immunological reaction, with anti- HCEC-injection therapy seems to be equivalent,
microbial agents also being administered as a or even a bit superior, to the various previously
prophylaxis to prevent infection under the drug reported clinical outcomes, including the find-
regimen administered in standard corneal trans- ings related to corneal graft survival rate, immu-
plantation procedures [11]. nological rejection rate, CEC density, and
best-corrected visual acuity post surgery
(Fig. 27.3) [11, 12].
Clinical Results The findings in our interventional study con-
firmed that although the number of cells injected
Based on the 5-year postoperative findings of into the anterior chamber, the surgical procedure
the 11 cases treated in the initial clinical trial, applied, and the postoperative care administered
HCEC-injection therapy has been found to be an were identical between the two groups in the
overall safe and effective treatment for complete study, HCEC-injection therapy using the second-
corneal restoration in patients afflicted with generation cHCECs (i.e., a cell suspension with
severe corneal endothelial failure (Fig. 27.2). an over 90% higher population of the mature-
Those clinical findings revealed that normal differentiated cells) resulted in an even better cor-
corneal thickness was achieved in 10 of the 11 neal restoration in terms of CEC density than that
treated eyes during the 5-year-postoperative resulting from the use of the first-generation cul-
follow-up period, with complete disappearance tured HCECs at both 24-weeks and 3-years post-
of corneal edema. The cHCECs produced by the operative (Figs. 27.4 and 27.5). In addition, use of
first-generation culture protocol were success- the second-generation cHCECs resulted in a
fully repopulated on the Descemet membrane faster recovery of corneal thinning compared with
and/or the bare posterior surface of the corneal the findings obtained using the first-generation
stroma, thus illustrating that they are biologi- cHCECs, probably due to the rapid functional
cally functional with excellent longevity. recovery for corneal dehydration (Fig. 27.6).
Specular microscopy imaging performed at Specular microscopy images obtained at 3-years
5-years postoperative revealed a relatively high postoperative confirmed a higher CEC density at
CEC density at the center of the posterior cor- the central area of the cornea in the eyes treated
neal surface in 10 of the 11 treated eyes (range, with the second-generation cHCECs (range, 2182
Fig. 27.2 Slit-lamp microscopy images (upper) and 5-years post injection (right column). The color maps
Scheimpflug camera images (lower) of two representative shown below each slit-lamp microscopy image (a) and
patients obtained at prior to surgery and at 3- and 5-years Scheimpflug camera image (b) illustrate the corneal thick-
post HCEC-Injection Therapy with the first-generation ness at each representative area of the corneal image
cHCECs. (a) A patient with Fuchs endothelial corneal above. The color bar located below b indicates the approx-
dystrophy (FECD), (a) a patient with argon-laser- imate corneal thickness of each of the colors shown in the
iridotomy induced bullous keratopathy. Pre-surgery (left maps (cited from Fig. 2, Numa K, et al. Ophthalmology
column), 3-years post injection (middle column), and 2021;128:504–514)
Fig. 27.3 Contact specular microscopy images of the not clear, thus suggesting that the eye was borderline cor-
central cornea in each of the 11 cases at 5-years postop- neal edema. FECD cases (Patients 2, 3, 5, 7, 8, 10, and 11)
erative. Scale bar: 100 μm. The endothelial cells are still show corneal guttae, however, the density of corneal
clearly visible, and a reasonable corneal endothelial cell guttae in those cases was found to have tended to be
density (ECD) can be seen in 10 of the 11 treated eyes. decreased (cited from Fig. 4, Numa K, et al. Ophthalmology
The image of the eye of Patient 4 shows some cells, yet 2021;128:504–514)
to 4417 cells/mm2) than in the eyes treated with In regard to safety, all eyes that underwent
the first-generation cHCECs (range, 746 to 2104 our HCEC-injection therapy showed no immu-
cells/mm2). Surprisingly, at 3-years post-injec- nological rejection, uveitis, infection, or
tion, the CEC density of the eyes treated with the increase of intraocular pressure directly related
second-generation cHCECs was very high with to the cell product. It should be noted that
only mild decay compared with the outcomes blood tests, including blood cell and blood bio-
obtained from ordinary corneal endothelial trans- chemistry tests, were performed at 1 month
plantation. Those findings may suggest not only post injection, and that the findings in those
the long-term stability and integrity of the examinations were almost within a normal
cHCECs post surgery but also excellent rejuvena- range. Moreover, our doctor-initiated clinical
tion of the CEC layer when using the well-differ- trial of HCEC-injection therapy in Japan has
entiated (i.e., mature) cHCECs obtained from now been completed with favorable clinical
young-age donor corneas [28]. results [11, 28].
a b
c d
Fig. 27.4 Representative data of the cultured human cor- images a and c based on CD44 and CD105 to identify
neal endothelial cells (hCECs) in Group 1 (Gr1) cell lots subpopulations (SPs). Mature-differentiated SP are indi-
used for Patient 9 and Group 2 (Gr2) cell lots used for cated in blue and the E-ratio (the proportion of these blue
Patient 15. (a, c) Phase contrast microscopy images of the cells) was calculated. (a, b) Cell Lot 23 in Gr1 showed
cultured hCECs used for the hCEC-injection therapy. 76.3% of E-ratio. (c, d) Cell Lot 33–2 in Gr2 showed
Scale bars, 200 μm. (b, d) Fluorescence-activated cell 99.2% of E-ratio (cited from Fig. 1, Ueno M, et al. Am J
sorting (FACS) analysis graphs of the cells shown in Ophthalmol 2022;237:267–277)
Fig. 27.5 Clinical data of the corneal endothelial cell

Corneal Endothelial Cell Density (CECD)
(cells/mm2)
density (CECD) obtained from the contact specular
4500 * *
microscopy images. Box and whisker plot of the
CECD of the patients at 24-weeks and 3-years 4000
postoperative in Group 1 (Gr1) and Group 2 (Gr2). In
3500
the Gr1 treated eyes, a relatively lower proportion (0.1
to 76.3%) of mature cell SPs was administered, while 3000
in the Gr2 eyes, a relatively higher proportion (>90%) 2500
of mature cell SPs was administered. Box plots
demonstrate the median (line) as well as lower and 2000
upper interquartile range (IQR; box), whiskers show 1500
the highest and lowest CECD values. There were
marked significant differences in CECD at 24-weeks 1000
and 3-years postoperative between the two groups 500
(*P < 0.001) (Wilcoxon rank-sum test) (cited from
24 Wk 3 Yr
Fig. 2, Ueno M, et al. Am J Ophthalmol
Group 1 Group 2
2022;237:267–277)
Group 1 (Patient 11)

a
Pre Cell Injection 4 Wk 12 Wk
CCT CCT CCT

725 μm 605 μm 586 μm
300 340 380 420 460 500 540 580 620 660 700 740 780 820 860 900
Group 2 (Patient 16)

b
Pre Cell Injection 4 Wk 12 Wk
CCT CCT CCT

792 μm 569 μm 554 μm
300 340 380 420 460 500 540 580 620 660 700 740 780 820 860 900
Fig. 27.6 Slit-lamp microscopy images (Upper) and patient (Patient 16). Pre-surgery (Left column), 4-weeks
Scheimpflug images (Lower) of representative patients in post injection (Middle), and 12-weeks post injection
Group 1 (Gr1) and Group 2 (Gr2) obtained at prior to sur- (Right column). The central corneal thickness (CCT) is
gery and at 4- and 12-weeks post cultured human corneal shown in each panel. The rapid decrease of CCT in the
endothelial cell (hCEC)-injection therapy. (a) Fuchs Gr2 patient is apparent by the two-dimensional corneal
endothelial corneal dystrophy in a Gr1 patient (Patient pachymetry images (cited from Fig. 6, Ueno M, et al. Am
11). (b) Pseudophakic corneal endothelial failure in a Gr2 J Ophthalmol 2022;237:267–277)
ntigenicity of Allogeneic Cells

A for acquiring an abundant cell expansion, an easy
and Fate of Escaped Cells cell delivery procedure, and cryopreservation
from the Eye will flourish. In order to attain that groundbreak-
ing apex, the cHCECs must be of the highest
Since cHCECs are created from allogeneic donor quality to be deemed ideal for use in the clinical
corneas, strict and thorough consideration must setting. When this formidable barrier is eventually
be made toward immunological reactions. In fact, overcome, the clinical use of cHCECs will surely
corneal endothelial reaction occurs more often in become the global standard for treating patients
cases undergoing penetrating keratoplasty [42], afflicted with severe corneal disorders. Now that
followed by cases undergoing DSAEK [43] and it has been confirmed that cHCECs injected as a
DMEK [44]. However, none of the eyes that have cell suspension into the anterior chamber can
undergone HCEC-injection therapy have shown self-organize and reconstitute the endothelial cell
immunological reaction in the clinical setting. In layer with full corneal restoration, cutting-edge
fact, in mouse experiments, an immune tolerance advancements in medical science and clinical
was obtained by injecting allogeneic CECs into applications are sure to arise in the near future.
the anterior chamber, and similar results occurred
when administered intravenously [45]. Moreover,
Streilein and colleagues reported that an immu- References
nological privilege could be maintained via ante-
rior chamber-associated immune deviation to 1. Tan DT, Dart JK, Holland EJ, Kinoshita S. Corneal
transplantation. Lancet. 2012;379(9827):1749–61.
protect allogeneic cells in the anterior chamber 2. Dana R. A new frontier in curing corneal blindness. N
[46, 47]. Thus, it can be assumed that a minimal Engl J Med. 2018;378(11):1057–8.
amount of immunological reaction occurs post 3. Gorovoy MS. Descemet-stripping automated endo-
cell injection therapy. thelial keratoplasty. Cornea. 2006;25(8):886–9.
4. Price MO, Calhoun P, Kollman C, Price FW Jr, Lass
Furthermore, the expression of human leuko- JH. Descemet stripping endothelial keratoplasty: ten-
cyte antigen class 1 in the cHCECs diminished in year endothelial cell loss compared with penetrating
the cells with a higher degree of cell maturation keratoplasty. Ophthalmology. 2016;123(7):1421–7.
[34], thus indicating that from the aspect of 5. Lass JH, Benetz BA, Patel SV, et al. Donor, recipi-
ent, and operative factors associated with increased
immunogenicity, mature-differentiated cells are endothelial cell loss in the cornea preservation time
ideal for clinical use. Moreover, the findings in study. JAMA Ophthalmol. 2019;137(2):185–93.
the animal-model experiments have shown that a 6. Melles GR, Ong TS, Ververs B, van der Wees
few cultured CECs or cellular components that J. Preliminary clinical results of Descemet mem-
brane endothelial keratoplasty. Am J Ophthalmol.
escape from the anterior chamber into the circu- 2008;145(2):222–7.
latory system post injection produce no adverse 7. Tourtas T, Laaser K, Bachmann BO, Cursiefen
events [11, 28, 41]. C, Kruse FE. Descemet membrane endothe-
lial keratoplasty versus descemet stripping auto-
mated endothelial keratoplasty. Am J Ophthalmol.
2012;153(6):1082–90.e2.
Future Directions 8. Weinstein JE, Weiss JS. Descemet membrane and
endothelial dystrophies. In: Mannis JM, Holland EJ,
In the future, clinical feedback is needed to fully editors. Cornea. 4th ed. Philadelphia, PA: Elsevier;
2017. p. 800–17.
elucidate the relationship between the “seeds,” 9. Macsai MS, Shiloach M. Use of topical rho kinase
i.e., in terms of cHCEC quality, and the “soil,” inhibitors in the treatment of Fuchs dystrophy after
i.e., in terms of corneal endothelial diseases and Descemet stripping only. Cornea. 2019;38(5):529–34.
the anterior chamber environment, in order to 10. Moloney G, Garcerant Congote D, Hirnschall N,
Arsiwalla T, Luiza Mylla Boso A, Toalster N, et al.
optimize the indicative diseases for HCEC- Descemet stripping only supplemented with topical
injection therapy [25]. ripasudil for fuchs endothelial dystrophy 12-month
Over time, the understanding of CEC biology outcomes of the sydney eye hospital study. Cornea.
will broaden, and interest in the methods applied 2021;40(3):320–6.
11. Kinoshita S, Koizumi N, Ueno M, Okumura N, Imai Foundations of corneal disease. Springer: Cham;
K, Tanaka H, et al. Injection of cultured cells with 2020. p. 215–24.
a ROCK inhibitor for bullous keratopathy. N Engl J 28. Ueno M, Toda M, Numa K, Tanaka H, Imai K, Bush
Med. 2018;378(11):995–1003. J, et al. Superiority of mature differentiated cultured
12. Numa K, Imai K, Ueno M, Kitazawa K, Tanaka H, human corneal endothelial cell injection therapy
Bush JD, et al. Five-year follow-up of first 11 patients for corneal endothelial failure. Am J Ophthalmol.
undergoing injection of cultured corneal endothelial 2021;237:267–77.
cells for corneal endothelial failure. Ophthalmology. 29. Ueno M, Asada K, Toda M, Schlötzer-Schrehardt
2021;128(4):504–14. U, Nagata K, Montoya M, et al. Gene signature-
13. Gain P, Jullienne R, He Z, Aldossary M, Acquart based development of elisa assays for reproduc-
S, Cognasse F, Thuret G. Global survey of corneal ible qualification of cultured human corneal
transplantation and eye banking. JAMA Ophthalmol. endothelial cells. Invest Ophthalmol Vis Sci.
2016;134(2):167–73. 2016;57(10):4295–305.
14. Dawson DG, Ubels JL, Edelhauser HF. Cornea and 30. Hamuro J, Ueno M, Toda M, Sotozono C, Montoya
sclera. In: Adler’s physiology of the eye. 11th ed. M, Kinoshita S. Cultured human corneal endothe-
Philadelphia, PA: Elsevier; 2011. p. 96–104. lial cell aneuploidy dependence on the presence of
15. Joyce NC. Proliferative capacity of corneal endothe- heterogeneous subpopulations with distinct differ-
lial cells. Exp Eye Res. 2012;95(1):16–23. entiation phenotypes. Invest Ophthalmol Vis Sci.
16. Matsuda M, Sawa M, Edelhauser HF, Bartels SP, 2016;57(10):4385–92.
Neufeld AH, Kenyon KR. Cellular migration and 31. Ueno M, Asada K, Toda M, Nagata K, Sotozono C,
morphology in corneal endothelial wound repair. Kosaka N, et al. Concomitant evaluation of a panel
Invest Ophthalmol Vis Sci. 1985;26(4):443–9. of exosome proteins and mirs for qualification of
17. Ono T, Mori Y, Nejima R, Iwasaki T, Miyai T, Miyata cultured human corneal endothelial cells. Invest
K. Corneal endothelial cell density and morphology Ophthalmol Vis Sci. 2016;57(10):4393–402.
in ophthalmologically healthy young individuals in 32. Hamuro J, Ueno M, Asada K, Toda M, Montoya M,
Japan: an observational study of 16842 eyes. Sci Rep. Sotozono C, et al. Metabolic plasticity in cell state
2021;11(1):18224. homeostasis and differentiation of cultured human
18. Krachmer JH, Purcell JJ Jr, Young CW, Bucher corneal endothelial cells. Invest Ophthalmol Vis Sci.
KD. Corneal endothelial dystrophy. A study of 64 2016;57(10):4452–63.
families. Arch Ophthalmol. 1978;96(11):2036–9. 33. Toda M, Ueno M, Yamada J, Hiraga A, Tanaka H,
19. Miyake K, Matsuda M, Inaba M. Corneal endothe- Schlötzer-Schrehardt U, et al. The different binding
lial changes in pseudoexfoliation syndrome. Am J properties of cultured human corneal endothelial cell
Ophthalmol. 1989;108(1):49–52. subpopulations to Descemet’s membrane components.
20. Koizumi N, Yamasaki K, Kawasaki S, Sotozono C, Invest Ophthalmol Vis Sci. 2016;57(11):4599–605.
Inatomi T, Mochida C, et al. Cytomegalovirus in 34. Hamuro J, Toda M, Asada K, Hiraga A, Schlötzer-
aqueous humor from an eye with corneal endotheli- Schrehardt U, Montoya M, et al. Cell homogeneity
itis. Am J Ophthalmol. 2006;141(3):564–5. indispensable for regenerative medicine by cultured
21. Pollack IP. Current concepts in laser iridotomy. Int human corneal endothelial cells. Invest Ophthalmol
Ophthalmol Clin. 1984;24(3):153–80. Vis Sci. 2016;57(11):4749–61.
22. Ho JW, Afshari NA. Advances in cataract surgery: 35. Ueno M, Asada K, Toda M, Hiraga A, Montoya
preserving the corneal endothelium. Curr Opin M, Sotozono C, et al. MicroRNA profiles qual-
Ophthalmol. 2015;26(1):22–7. ify phenotypic features of cultured human cor-
23. Realini T, Gupta PK, Radcliffe NM, Garg S, Wiley neal endothelial cells. Invest Ophthalmol Vis Sci.
WF, Yeu E, et al. The effects of glaucoma and glau- 2016;57(13):5509–17.
coma therapies on corneal endothelial cell density. J 36. Toda M, Ueno M, Hiraga A, Asada K, Montoya M,
Glaucoma. 2021;30(3):209–18. Sotozono C, et al. Production of homogeneous cul-
24. Matsuda M, Tano Y, Inaba M, Manabe R. Corneal tured human corneal endothelial cells indispensable
endothelial cell damage associated with intraocular for innovative cell therapy. Invest Ophthalmol Vis Sci.
gas tamponade during pars plana vitrectomy. Jpn J 2017;58(4):2011–20.
Ophthalmol. 1986;30(3):324–9. 37. Yamamoto A, Tanaka H, Toda M, Sotozono C,
25. Forbes SJ, Rosenthal N. Preparing the ground for tis- Hamuro J, Kinoshita S, et al. A physical biomarker
sue regeneration: from mechanism to therapy. Nat of the quality of cultured corneal endothelial cells and
Med. 2014;20(8):857–69. of the long-term prognosis of corneal restoration in
26. Yamaguchi T, Higa K, Suzuki T, Nakayama N, Yagi- patients. Nat Biomed Eng. 2019;3(12):953–60.
Yaguchi Y, Dogru M, et al. Elevated cytokine levels in 38. Hamuro J, Numa K, Fujita T, Toda M, Ueda K,
the aqueous humor of eyes with bullous Keratopathy Tokuda Y, et al. Metabolites interrogation in cell fate
and low endothelial cell density. Invest Ophthalmol decision of cultured human corneal endothelial cells.
Vis Sci. 2016;57(14):5954–62. Invest Ophthalmol Vis Sci. 2020;61(2):10.
27. Kinoshita S, Ueno M. Cultivated cells in the treatment 39. Hamuro J, Deguchi H, Fujita T, Ueda K, Tokuda Y,
of corneal diseases. In: Colby K, Dana R, editors. Hiramoto N, et al. Polarized expression of ion chan-
nels and solute carrier family transporters on hetero- endothelial keratoplasty: safety and outcomes: a
geneous cultured human corneal endothelial cells. report by the American academy of ophthalmology.
Invest Ophthalmol Vis Sci. 2020;61(5):47. Ophthalmology. 2009;116(9):1818–30.
40. Numa K, Ueno M, Fujita T, Ueda K, Hiramoto N, 44. Deng SX, Lee WB, Hammersmith KM, Kuo AN,
Mukai A, et al. Mitochondria as a platform for dictat- Li JY, Shen JF, et al. Descemet membrane endo-
ing the cell fate of cultured human corneal endothelial thelial keratoplasty: safety and outcomes: a report
cells. Invest Ophthalmol Vis Sci. 2020;61(14):10. by the American academy of ophthalmology.
41. Toda M, Yukawa H, Yamada J, Ueno M, Kinoshita Ophthalmology. 2018;125(2):295–310.
S, Baba Y, et al. In vivo fluorescence visualization of 45. Yamada J, Ueno M, Toda M, Shinomiya K, Sotozono
anterior chamber injected human corneal endothelial C, Kinoshita S, et al. Allogeneic sensitization and
cells labeled with quantum dots. Invest Ophthalmol tolerance induction after corneal endothelial cell
Vis Sci. 2019;60(12):4008–20. transplantation in mice. Invest Ophthalmol Vis Sci.
42. Vickers LA, Foulks GN, Gupta PK. Diagnosis and 2016;57(11):4572–80.
management of corneal allograft rejection. In: Cornea. 46. Niederkorn JY. Immune privilege and immune regula-
4th ed. Philadelphia, PA: Elsevier; 2011. p. 1687–96. tion in the eye. Adv Immunol. 1990;48:191–226.
43. Lee WB, Jacobs DS, Musch DC, Kaufman SC, 47. Streilein JW. Immune regulation and the eye: a dan-
Reinhart WJ, Shtein RM. Descemet’s stripping gerous compromise. FASEB J. 1987;1:199–208.
Ultrathin DSAEK
28
Key Points increased frequency of postoperative complica-

• Ultrathin Descemet stripping automated endo- tions both during and after surgery [2–4].
thelial keratoplasty (UT-DSAEK) offers the Moreover, the learning curve of DMEK is accom-
potential to achieve the visual results of panied by a high rate of tissue loss (up to 16%), a
DMEK with the ease of handling and tissue high graft detachment rate of up to 63%, and a
preparation of conventional DSAEK. graft failure rate of up to 8% [5–7]. In eyes with
• Quality control during donor tissue prepara- complex anterior segment anatomy such as abnor-
tion for UT-DSAEK is mandatory in order to malities of the iris-lens diaphragm or in eyes with
optimize postoperative outcomes. previous glaucoma surgery or pars plana vitrec-
tomy, poor control of the DMEK graft within the
anterior chamber during unfolding increases the
Introduction technical complexity of the procedure and often
results in excess graft manipulation [8–11].
Over the past two decades, endothelial kerato- Currently, DSAEK is still the most popular EK
plasty (EK) has become the gold standard for the technique performed worldwide [2, 3].
surgical management of endothelial decompensa- DSAEK involves replacement of Descemet
tion [1]. Modern EK procedures can broadly be membrane and the diseased endothelium with
divided into Descemet stripping automated endo- donor tissue composed of a thin layer of posterior
thelial keratoplasty (DSAEK) and Descemet stroma, Descemet membrane, and endothelium
membrane endothelial keratoplasty (DMEK) [1]. [1]. In 2006, Holland reported that the postopera-
While DMEK provides faster visual rehabilitation tive best-corrected visual acuity of DSAEK grafts
and lower rate of immune rejection in comparison thinner than 131 μm compared favorably to that
to DSAEK, DMEK is associated with technical of thicker grafts and even those DMEK. This data
challenges in graft preparation and delivery and supported the correlation of postoperative vision
to the morphologic characteristics of the DSAEK
graft [12, 13]. Graft regularity has increasingly
Supplementary Information The online version con- been recognized as a key determinant of the qual-
ity of DSAEK grafts [14]. Aberrations derived
org/10.1007/978-3-031-32408-6_28.
from substantial irregularities in lenticule shape
and graft thickness have been also found to play
A. C. Yu · M. Busin (*) a significant role in determining the final visual
Department of Translational Medicine, University of outcome [15, 16].
Ferrara, Ferrara, Italy
https://doi.org/10.1007/978-3-031-32408-6_28
In an attempt to improve the postoperative explored. A history of previous operations or

outcomes of DSAEK, Busin introduced the con- prior infections may require modifications of the
cept of ultrathin DSAEK (UT-DSAEK) in 2009. surgical technique and perioperative
UT-DSAEK employed grafts within 100 μm in management.
thickness, thereby combining the visual out- Complete ophthalmologic examination
comes of DMEK with the technical ease of including slit-lamp examination, best spectacle-
DSAEK. Since then, double- and single-pass corrected visual acuity (BSCVA), manifest
techniques have been used by surgeons and eye refraction, applanation tonometry, and fundus-
banks to reproducibly obtain DSAEK grafts of a copy must also be performed. Endothelial cell
predetermined thickness and planar profile, function is assessed by specular or confocal
which have substantially improved the outcomes microscopy of the central and peripheral cornea.
of DSAEK [17, 18]. Additionally, anterior segment OCT is useful to
assess corneal curvature and pachymetry.
The decision to perform either combined or
Indications sequential EK and cataract surgery is dependent
on several preoperative factors. A combined pro-
UT-DSAEK shares the same indications of con- cedure is often considered in patients aged 50
ventional DSAEK. Conditions requiring the pro- years and older with signs of corneal decompen-
cedure include patients with any type of sation, as the retained crystalline lens will other-
endothelial dysfunction such as endothelial dys- wise invariably develop a cataract as a consequence
trophies including Fuchs endothelial dystrophy of both surgical trauma and postoperative topical
or posterior polymorphous dystrophy, pseudo- steroid treatment [21]. Phakic UT-DSAEK may
phakic or aphakic bullous keratopathy, iridocor- be alternatively performed in younger patients
neal endothelial syndrome and endothelial with otherwise clear crystalline lens.
decompensation secondary to previous trauma, If a modern flexible open-loop anterior cham-
intraocular surgery, or failed previous grafts [19]. ber intraocular lens (ACIOL) or an iris-fixated
While DMEK is also feasible in complex eyes IOL is well-positioned and of appropriate size, it
[20], UT-DSAEK can be more easily performed may be left in place [22, 23]. Otherwise, poorly
in the presence of ocular comorbidities such as positioned IOLs can be exchanged for a posterior
aniridia, aphakia, extensive iris trauma, anterior chamber IOL (PCIOL) using transscleral suture
chamber intraocular lenses, and previous glau- fixation.
coma surgery [21–24]. However, when subse- Finally, phakic IOLs causing progressive endo-
quent surgery of the host cornea can be thelial cell loss (ECL) may be removed at the time
anticipated, either because of stromal ulceration of a combined UT-DSAEK. Phacoemulsification
or fibrosis or because of high-degree refractive and posterior chamber IOL implantation can then
errors, a thicker DSAEK graft may be considered be sequentially performed to minimize the time
to avoid inadvertent penetration into the anterior required for visual rehabilitation [24].
chamber while performing the secondary anterior
procedure.
Anesthesia Considerations
Preoperative Planning Selection of anesthesia should be individualized

for every patient. In most cases, UT-DSAEK can
The first step of the preoperative evaluation be performed with local anesthesia such as perib-
involves eliciting a careful and detailed general ulbar or retrobulbar anesthesia. Adjunct intrave-
and ophthalmic history. The presence of concom- nous sedation can also be considered.
itant eye diseases such as amblyopia, glaucoma, Decompression with a Honan balloon or similar
optic neuropathy, and retinal disease must be devices for at least 10 min allows vitreous dehy-
28 Ultrathin DSAEK 409
dration and softening of the eye, thereby reduc- The improved predictability of the current ALTK
ing the risk of excessive posterior pressure and systems has allowed the use of single-pass tech-
anterior chamber shallowing. niques with microkeratome heads even up to
450 μm cutting depth. An additional advantage of
the single-pass technique is that the dissection
Surgical Technique yields an anterior lamella thicker than that cut
with the first dissection of the double-pass tech-
Donor graft preparation is a very crucial step in nique and can be more properly used for tectonic
the UT-DSAEK procedure. The first attempts at keratoplasty or also for deep anterior lamellar
standardizing UT-DSAEK graft preparation by keratoplasty (DALK) (Video 28.1).
means of a pivoting microkeratome (Carriazo- Before removing the tissue from the AAC, the
Barraquer, Moria SA, Antony, France) have led to stromal side is marked to facilitate correct intra-
the development of the “double-pass” technique. operative orientation of the graft. The posterior
The technique was based on the observation that donor lamella is then placed on a punch with the
the predictability of the dissection depth was endothelial side up and cut to the desired diame-
inversely proportional to the width of the micro- ter (8.0–9.0 mm).
keratome head slit. In the double-pass technique,
the donor cornea is mounted on an artificial ante-
rior chamber (AAC) of the ALTK system (Moria, Recipient Preparation and Graft
Antony, France). The central corneal thickness of Delivery
the donor is measured using ultrasound pachym-
etry (SP-3000; Tomey GmbH). An initial debulk- The initial steps of UT-DSAEK (Fig. 28.1) do not
ing step is performed using 300 μm microkeratome differ from those of conventional DSAEK. Prior
head which would debulk the donor tissue to to commencing surgery, loose and edematous
around 180–250 μm in thickness. After turning epithelium is removed from the recipient cornea
the dovetail of the AAC by 180°, a second micro- to allow better intraoperative visualization and
keratome-assisted dissection (refinement cut) is postoperative epithelialization. A small aliquot of
carried out from the direction opposite to the one aqueous is aspirated, and air is injected intracam-
of the first cut. As microkeratome dissection is erally. Descemet membrane-endothelium com-
deepest at the beginning of the cut, dissecting plex is scored and stripped using a 25-gauge
twice from opposite directions not only prevents needle or cannula, Descemet stripper, or reverse
perforation but also equalizes peripheral graft Sinskey hook. Gentle pressure is applied to the
thickness, thereby producing a regular lenticule inner cornea, taking care not to press into corneal
with planar configuration and less optical aberra- stroma in order to avoid creation of stromal tissue
tions. A thinner microkeratome head (90, 110, or strands. Descemetorhexis under air improves
130 μm) is used in the refinement cut based on the visualization and obviates the need for viscoelas-
Busin nomogram which was optimized to obtain tic or trypan blue. In cases with poor anterior
a final central graft thickness within 100 μm. The chamber visualization, Descemet membrane may
pressure of the system is standardized to an ideal be left in situ in the absence of central guttae.
level of 80–90 mmHg by raising the infusion bot- If not present, a peripheral iridotomy (PI) is
tle to a height of 120 cm above the level of the created to avoid postoperative pupillary block.
AAC and then clamping the tubing at 50 cm from The authors prefer to perform a surgical inferior
the entrance into the AAC. In order to obtain regu- PI using guillotine micro-incision scissors under
lar graft thickness, care must be taken in main- continuous irrigation from the anterior chamber
taining a slow uniform movement during manual maintainer.
microkeratome-dissection. While several glides have been developed, our
More recently, linear microkeratomes have preferred delivery device for UT-DSAEK is the
been used for the creation of UT-DSAEK grafts. modified Busin glide (mini-glide or Mini Busin
Fig. 28.1 Intraoperative steps of ultrathin Descemet stripping automated endothelial keratoplasty
Spatula, Moria SA, Antony, France). The modi- graft with minimal graft manipulation. Gentle
fied Busin glide ensures proper graft delivery tapping onto the surface of the anterior cornea
with correct orientation. The graft can also be tri- allows spontaneous unfolding and centration of
folded before pulling it into the glide funnel, the UT-DSAEK graft.
which can easily be inserted along the main inci- Both the clear cornea tunnel and the side entry
sion at the nasal cornea. No viscoelastic sub- are sutured with interrupted 10–0 nylon sutures.
stance is required during insertion. Using The graft is attached to the posterior corneal sur-
micro-incision forceps inserted through the tem- face by filling the anterior chamber with air
poral paracentesis, the UT-DSAEK graft is deliv- injected. The authors prefer intracameral injec-
ered bimanually under continuous, low-flow tion of air avoid due to concerns of potential
irrigation via an anterior chamber maintainer. endothelial toxicity of SF6 and conflicting evi-
Delivering the graft using the pull through tech- dence of the latter’s efficacy. Triamcinolone ace-
nique provides total control throughout the pro- tonide and gentamicin sulfate, 0.3%, are injected
cedure and allows spontaneous unfolding of the subconjunctivally at the end of the procedure.
Combined Surgeries Postoperative Care
UT-DSAEK can be combined with other intra- Subconjunctival antibiotic and corticosteroid
ocular procedures such as phacoemulsification, injections may be administered immediately after
IOL implantation, IOL exchange, secondary IOL completion of surgery. A fixed combination of
implantation, pupilloplasty, and vitrectomy. topical antibiotic and steroid drops is initiated
These procedures are often performed immedi- every 2 h daily and tapered off to 4 times daily
ately before insertion of the UT-DSAEK graft. over the first postoperative month. Subsequently,
Viscoelastic is also preferably avoided even in topical antibiotics is discontinued, while topical
combined procedures. Continuous curvilinear steroid is slowly tapered to once daily indefi-
capsulorhexis can be performed using a bent nee- nitely. Steroid-induced ocular hypertension is
dle mounted on a syringe filled with saline to treated with intraocular pressure lowering agents.
maintain a closed system, while a foldable IOL After the surgery, patients are instructed to lie
can be injected under continuous irrigation from supine for at least 2 h. Slit lamp exam is then per-
an anterior chamber maintainer. IOLs should be formed to check for graft attachment. Air can be
of the hydrophobic type, as recent reports have removed from one of the side entries using a
pointed out the possibility of opacification of blunt cannula, in order to avoid pupillary block.
hydrophilic IOLs after DSAEK [25, 26]. When the graft is detached, a rebubbling proce-
Intracameral acetylcholine chloride is used to dure is performed.
constrict the pupil after IOL implantation and
prior to UT-DSAEK.
If use of viscoelastic is preferred, careful Results
removal from the anterior chamber must be per-
formed because retained material may hinder Visual Outcomes
graft attachment or may result in interface
opacities that can interfere with vision months Visual recovery is faster and the proportion of
after surgery. Intracameral acetylcholine is eyes with final Snellen visual acuity of 20/20 is
used to constrict the pupil after IOL implanta- higher after UT-DSAEK than after conventional
tion and prior to UT-DSAEK surgery DSAEK, while no substantial difference is found
(Fig. 28.2). with post-DMEK outcomes.
Fig. 28.2 Preoperative and postoperative images from a patient who underwent ultrathin Descemet stripping auto-
mated endothelial keratoplasty for Fuchs endothelial dystrophy
Recently published randomized controlled tri- comparing UT-DSAEK and DMEK likewise did
als comparing provide discordant results with not find significant differences in ECD and ECL
regard to visual outcomes. Unlike the DETECT [19, 27].
study, Dunker et al. found no significant differ- At 1 year, the ECL was comparable with that
ences in visual acuity between DMEK and ultra- of conventional DSAEK, despite the the use of
thin DSAEK as early as 3 months and up to smaller (3.0-mm) incisions. After UT-DSAEK,
1 year after surgery [19, 27]. No significant dif- ECD is significantly higher in glaucoma patients,
ferences were also observed in terms of com- as was previously reported after DSAEK [38].
bined higher order aberrations, contrast
sensitivity, straylight and vision-related quality
of life [28–30]. Although 2-year results of Graft Survival
DETECT seemed to suggest that DMEK pro-
vides superior visual outcomes, the results were One-year graft survival for conventional DSAEK
inconclusive as the loss to follow-up may have in series excluding the initial learning curve has
affected the study findings [31]. A larger random- been reported to vary between 94% and 100%
ized clinical trial is warranted to further clarify [39, 40]. In our recent series on UT-DSAEK,
differences in visual acuity after DMEK and Kaplan–Meier graft survival probability at 1, 2,
ultrathin DSAEK. 3, and 5 years was 99.1% (212/214 eyes), 96.2%
(167/172 eyes), 94.2% (144/147 eyes), and
94.2% (105/105 eyes), respectively [41]. Price
Refractive Outcomes reported recently a conventional DSAEK graft
survival rate at 5 years of 93%, with a signifi-
Similar to DMEK, no significant change in astig- cantly higher value for Fuchs patients (95%) than
matism is seen after UT-DSAEK [17]. While a for patients with pseudophakic bullous keratopa-
significant induced cylinder (up to 0.6 D) has thy (76%) or previous glaucoma surgery (40%)
been reported by several authors after conven- [36]. Vasiliauskaite et al. reported a cumulative
tional DSAEK, this is most probably a conse- survival probability of 83% (95% confidence
quence of the different wound sizes employed for interval [CI], 0.75–0.92) 5 years following
graft delivery [32]. UT-DSAEK grafts can be DMEK.
delivered through a 3 mm incision using the mod- Immunologic rejection occurs less frequently
ified Busin glide. after UT-DSAEK than after DSAEK, but still
Mild hyperopic shifts (0.78 ± 0.59 D) occur occurs more frequently over DMEK. Gender
after UT-DSAEK procedures. Reversal of cor- matching does not affect rates of immune rejec-
neal edema after surgery can result in significant tion or graft failure [42].
alterations in corneal curvature. Taking into
account this shift during (IOL) calculations can
minimize refractive surprises. Complications
The most common complication following

Endothelial Cell Density UT-DSAEK is transient cystoid macular edema,
which responds well to conservative treatment
Five years after UT-DSAEK, endothelial cell loss [43]. Graft detachment and graft failure are much
averaged 52%. This is similar to the values less frequent than after conventional DSAEK or
reported after conventional DSAEK by Price DMEK. In particular, the rate of rebubbling fol-
et al. (53%) [33], Wacker et al. (55%) [34], and lowing UT-DSAEK is within 4%, which is lower
Ang et al. (48.7%) [35] and with that of DMEK than any DMEK statistic published to date. Total
by Ham et al. (55%) [36] and by Vasiliauskaite detachment is seen in most cases and generally
et al. (59%) [37]. Randomized controlled trials successfully managed by rebubbling (with a sin-
gle or double injection). Recent randomized con- 5. Dapena I, Ham L, Droutsas K, van Dijk K,
Moutsouris K, Melles GR. Learning curve in
trolled trials have shown higher rebubbling rates Descemet’s membrane endothelial keratoplasty: first
for DMEK compared to ultrathin DSAEK [19, series of 135 consecutive cases. Ophthalmology.
26]. Both studies report markedly similar rates of 2011;118(11):2147–54.
graft detachment (24%) in DMEK versus 4% in 6. Rodríguez-Calvo-de-Mora M, Quilendrino R, Ham
L, et al. Clinical outcome of 500 consecutive cases
ultrathin DSAEK. Considering their absolute dif- undergoing Descemet’s membrane endothelial kera-
ferences, the graft detachment rates after DMEK toplasty. Ophthalmology. 2014;20(14):161–4.
are appreciably higher than after ultrathin 7. Gorovoy MS. DMEK complications. Cornea.
DSAEK. 2014;33(1):101–4.
8. Aravena C, Yu F, Deng SX. Outcomes of Descemet
Other complications of the procedure include membrane endothelial keratoplasty in patients with
pupillary block, persistent epithelial defect, per- prior glaucoma surgery. Cornea. 2017;36/3:284–9.
sistent interface haze, interface infections, and 9. Sorkin N, Mimouni M, Kisilevsky E, et al. Four-year
cataract formation. Such adverse events do not survival of Descemet membrane endothelial kerato-
plasty in patients with previous glaucoma surgery. Am
seem to occur more or less frequently than after J Ophthalmol. 2020;218:7–16.
conventional DSAEK. 10. Santaella G, Sorkin N, Mimouni M. Outcomes
of Descemet membrane endothelial kerato-
Take Home Notes plasty in aphakic and aniridic patients. Cornea.
2020;39(11):1389–93.
• UT-DSAEK still remains a valuable tool in the 11. Spaniol K, Holtmann C, Schwinde JH, Deffaa S,
surgical armamentarium of any corneal Guthoff R, Geerling G. Descemet-membrane endo-
specialist. thelial keratoplasty in patients with retinal comor-
• Similar to conventional DSAEK, UT-DSAEK bidity-a prospective cohort study. Int J Ophthalmol.
2016;9(3):390–4.
can be more easily performed in all eyes with 12. Busin M, Albé E. Does thickness matter: ultrathin
complex anatomy and poor anterior chamber Descemet stripping automated endothelial kerato-
visualization. plasty. Curr Opin Ophthalmol. 2014;25(4):312–8.
• Unlike DMEK, the complication rates espe- 13. Busin M, Yu A. The ongoing debate: Descemet
membrane endothelial keratoplasty versus ultrathin
cially in terms of graft detachment are signifi- Descemet stripping automated endothelial kerato-
cantly lower following UT-DSAEK. plasty. Ophthalmology. 2020;127(9):1160–1.
• Standardized graft preparation yields consis- 14. Dickman MM, Cheng YY, Berendschot TT, et al.
tent graft quality, regularity, and thickness Effects of graft thickness and asymmetry on visual
gain and aberrations after Descemet stripping auto-
which in turn translates to excellent visual mated endothelial keratoplasty. JAMA Ophthalmol.
outcomes. 2013;131:737e744.
15. Rudolph M, Laaser K, Bachmann BO, Cursiefen C,
Epstein D, Kruse FE. Corneal higher-order aberra-
tions after Descemet's membrane endothelial kerato-
plasty. Ophthalmology. 2012;119(3):528–35.
References 16. Ruzza A, Parekh M, Ferrari S, et al. Preloaded donor
corneal lenticules in a new validated 3D printed
1. Deng SX, Barry Lee W, Hammersmith KM, et al. smart storage glide for Descemet stripping auto-
Descemet membrane endothelial keratoplasty: safety mated endothelial keratoplasty. Br J Ophthalmol.
and outcomes. Ophthalmology. 2018;125(2):295–310. 2015;99(10):1388–95.
2. 2019 Eye banking statistical report. Washington, DC: 17. Busin M, Madi S, Santorum P, Scorcia V, Beltz
Eye Bank Association of America; https://restore- J. Ultrathin descemet's stripping automated endo-
sight.org/wp-content/uploads/2020/04/2019-EBAA- thelial keratoplasty with the microkeratome double-
Stat-Report-FINAL.pdf. Accessed 23 May 2020. pass technique: two-year outcomes. Ophthalmology.
3. Suryan DL, Armitage JW, Armitage M, et al. 2013;120(6):1186–94.
Outcomes of corneal transplantation in Europe: report 18. Villarrubia A, Cano-Ortiz A. Development of
by the European cornea and cell transplantation regis- a nomogram to achieve ultrathin donor corneal
try. J Cataract Refract Surg. 2021;47(6):780–5. disks for Descemet-stripping automated endo-
4. Zafar S, Parker JS, de Kort C, Melles G, Sikder thelial keratoplasty. J Cataract Refract Surg.
S. Perceived difficulties and barriers to uptake of 2015;41(1):146–51.
Descemet’s membrane endothelial keratoplasty 19. Dunker SL, Dickman MM, Wisse RPL, et al.
among surgeons. Clin Ophthalmol. 2019;13:1055–61. DMEK versus ultrathin DSAEK: a multicenter ran-
domized controlled clinical trial. Ophthalmology. 31. Rose-Nussbaumer J, Lin CC, Austin A, Liu Z, Clover
2020;127(9):1152–9. J, McLeod SD, et al. Descemet endothelial thickness
20. Yu AC, Myerscough J, Spena R, et al. Three-year comparison trial: two-year results from a random-
outcomes of tri-folded endothelium-in Descemet ized trial comparing ultrathin Descemet stripping
membrane endothelial keratoplasty with pull-through automated endothelial keratoplasty with Descemet
technique. Am J Ophthalmol. 2020;219:121–31. membrane endothelial keratoplasty. Ophthalmology.
21. Burkhart ZN, Feng MT, Price FW Jr, Price MO. One- 2020;128:1238. https://doi.org/10.1016/j.oph-
year outcomes in eyes remaining phakic after tha.2020.12.021; Epub ahead of print.
Descemet membrane endothelial keratoplasty. J 32. Koenig SB, Covert DJ. Early results of small-incision
Cataract Refract Surg. 2014;40(3):430–4. Descemet’s stripping and automated endothelial kera-
22. Elderkin S, Tu E, Sugar J, Reddy S, Kadakia A, toplasty. Ophthalmology. 2007;114(2):221–6.
Ramaswamy R, Djalilian A. Outcome of descemet 33. Price MO, Fairchild KM, Price DA. Descemet’s
stripping automated endothelial keratoplasty in stripping endothelial keratoplasty: five-year graft
patients with an anterior chamber intraocular lens. survival and endothelial cell loss. Ophthalmology.
Cornea. 2010;29(11):1273–7. 2011;118:725–9.
23. Beltz J, Busin M. Descemet stripping automated endo- 34. Wacker K, Baratz KH, Maguire LJ, et al. Descemet
thelial keratoplasty in a case with a posteriorly fixated stripping endothelial keratoplasty for Fuchs’ endothe-
iris-claw intraocular lens. Cornea. 2012;31(1):96–7. lial corneal dystrophy: five-year results of a prospec-
24. Nahum Y, Busin M. Quadruple procedure for visual tive study. Ophthalmology. 2016;123:154–60.
rehabilitation of endothelial decompensation fol- 35. Ang M, Soh Y, Htoon HM, et al. Five-year graft
lowing phakic intraocular lens implantation. Am J survival comparing Descemet stripping automated
Ophthalmol. 2014;158(6):1330–4. endothelial keratoplasty and penetrating keratoplasty.
25. Ahad MA, Darcy K, Cook SD, Tole DM. Intraocular Ophthalmology. 2016;123:1646–52.
lens opacification after descemet stripping 36. Ham L, Dapena I, Liarakos VS, et al. Midterm results
automated endothelial keratoplasty. Cornea. of Descemet membrane endothelial keratoplasty:
2014;33(12):1307–11. 4 to 7 years clinical outcome. Am J Ophthalmol.
26. Werner L, Wilbanks G, Nieuwendaal CP, Dhital 2016;171:113–21.
A, Waite A, Schmidinger G, Lee WB, Mamalis 37. Vasiliauskaite I, Oellerich S, Ham L, Dapena I,
N. Localized opacification of hydrophilic acrylic Baydoun L, van Dijk K, et al. Descemet mem-
intraocular lenses after procedures using intracam- brane endothelial keratoplasty: ten-year graft sur-
eral injection of air or gas. J Cataract Refract Surg. vival and clinical outcomes. Am J Ophthalmol.
2015;41(1):199–207. 2020;217:114–20.
27. Chamberlain W, Lin CC, Austin A, et al. Descemet 38. Quek DT, Wong T, Tan D, Mehta JS. Corneal graft
endothelial thickness comparison trial: a random- survival and intraocular pressure control after des-
ized trial comparing ultrathin Descemet stripping cemet stripping automated endothelial keratoplasty in
automated endothelial keratoplasty with Descemet eyes with pre-existing glaucoma. Am J Ophthalmol.
membrane endothelial keratoplasty. Ophthalmology. 2011;152(1):48–54.
2019;126:19e26. 39. Busin M, Bhatt PR, Scorcia V. A modified technique
28. Duggan MJ, Rose-Nussbaumer J, Lin CC, Austin for descemet membrane stripping automated endo-
A, Labadzinzki PC, Chamberlain WD. Corneal thelial keratoplasty to minimize endothelial cell loss.
higher-order aberrations in Descemet membrane Arch Ophthalmol. 2008;126(8):1133–7.
endothelial keratoplasty versus ultrathin DSAEK 40. Price MO, Gorovoy M, Benetz BA, Price FW Jr,
in the Descemet endothelial thickness comparison Menegay HJ, Debanne SM, Lass JH. Descemet’s
trial: a randomized clinical trial. Ophthalmology. stripping automated endothelial keratoplasty out-
2019;126(7):946–57. comes compared with penetrating keratoplasty
29. Ang MJ, Chamberlain W, Lin CC, Pickel J, Austin from the cornea donor study. Ophthalmology.
A, Rose-Nussbaumer J. Effect of unilateral endo- 2010;117(3):438–44.
thelial keratoplasty on vision-related quality-of-life 41. Madi S, Leon P, Nahum Y, et al. Five-year outcomes
outcomes in the Descemet endothelial thickness of ultrathin Descemet stripping automated endothelial
comparison trial (DETECT): a secondary analysis keratoplasty. Cornea. 2019;38:1192–7.
of a randomized clinical trial. JAMA Ophthalmol. 42. Romano V, Parekh M, Virgili G, et al. Gender match-
2019;137(7):747–54. ing did not affect 2-year rejection or failure rates
30. Dunker SL, Dickman MM, Wisse RPL, Nobacht following DSAEK for Fuchs endothelial corneal dys-
S, Wijdh RHJ, Bartels MC, Mei-Lie Tang NE, trophy. Am J Ophthalmol. 2022;235:204–10.
et al. Quality of vision and vision-related quality of 43. Myerscough J, Roberts HW, Yu AC, et al. Factors
life after Descemet membrane endothelial kerato- predictive of cystoid macular oedema following
plasty: a randomized clinical trial. Acta Ophthalmol. endothelial keratoplasty: a single-Centre review of
2021;99:e1127. https://doi.org/10.1111/aos.14741; 2233 cases. Br J Ophthalmol. 2021;107:24. https://
Epub ahead of print. doi.org/10.1136/bjophthalmol-2020-318076.
Innovations in Descemet
Membrane Endothelial
29
Keratoplasty (DMEK)
Darren S. J. Ting and Marcus Ang
Key Points DMEK possible for complex eyes such as

• DMEK is emerging as an important lamellar those with shallow anterior chamber, aphakia
keratoplasty technique for selective endothe- and/or significant iris defect.
lial replacement to treat end-stage corneal
endothelial diseases.
• Innovations in donor insertion for DMEK Introduction
have led to development of devices that use
injection (endothelium-out) or pull-through In recent decades, there has been a paradigm shift
(endothelium-in) approaches. from penetrating keratoplasty to endothelial ker-
• The use of intraoperative optical coherence atoplasty in treating end-stage corneal endothe-
tomography (iOCT) may enhance intraopera- lial diseases, including Fuchs endothelial corneal
tive visualisation of graft unfolding and dystrophy (FECD) and pseudophakic bullous
attachment during DMEK surgery. keratopathy (PBK) [1–7]. Although the concept
• Innovations in techniques, including safety- of endothelial keratoplasty (EK) for selective
net suture, phakic collamer lens implantation replacement of diseased corneal endothelium
and artificial iris implantation, have rendered was first introduced as far back as 1950s by
Charles Tillet [8], further modifications have led
Supplementary Information The online version con- to the development of various techniques [8].
org/10.1007/978-3-031-32408-6_29.
Currently, Descemet stripping automated endo-
thelial keratoplasty (DSAEK) and Descemet
D. S. J. Ting
membrane endothelial keratoplasty (DMEK)
Birmingham and Midland Eye Centre, Birmingham, UK represent the two most widely performed EK
Academic Unit of Ophthalmology, Institute of
techniques [9, 10].
Inflammation and Ageing, University of Birmingham, DSAEK involves removal of the diseased
Birmingham, UK Descemet membrane (DM) and endothelium,
Academic Ophthalmology, School of Medicine, followed by grafting of a healthy donor cornea
University of Nottingham, Nottingham, UK consisting of posterior stroma, DM and endothe-
M. Ang (*) lium [11–13]. The preparation of the donor cor-
Singapore National Eye Center, Singapore Eye nea is performed with the assistance of an
Research Institute, Singapore, Singapore automated microkeratome [6, 14]. This is cur-
Department of Ophthalmology and Visual Sciences, rently a popular EK technique, supported by
Duke-NUS Medical School, Singapore, Singapore donor preparation by eye banks, though it may
https://doi.org/10.1007/978-3-031-32408-6_29
416 D. S. J. Ting and M. Ang
result in donor stroma and graft–host interface The DMEK surgery was original described
irregularity, which can negatively impact on using an endothelium-out, injection technique
visual outcomes [14–16]. In 2006, Melles et al. [17], though recent innovations have made both
[17] introduced Descemet membrane endothelial endothelium-out and endothelium-in techniques
keratoplasty (DMEK)—a like-to-like replace- feasible with comparable clinical outcomes. For
ment of the diseased DM-endothelium with donor graft harvesting and marking, the steps are
healthy donor DM-endothelium only. This tech- similar for both techniques. DMEK graft harvest-
nique helps minimise donor stroma–host inter- ing is usually initiated by a 360° peripheral scor-
face irregularities, postoperative hyperopic shift ing and stripping of the donor peripheral DM
and higher order aberrations associated with from the posterior stroma. Since the first descrip-
DSAEK. Most studies also report a faster recov- tion of its original technique using manual scor-
ery, lower graft rejection rate and reduced need ing with a Sinskey hook or fine non-toothed
for topical steroids (hence a lower risk of glau- forceps, various techniques have been described
coma) compared to DSAEK [10, 14, 18–20]. for facilitate this step, including big and small
In addition, the preparation of the DMEK trephines, big bubble technique and liquid bubble
donor does not require a microkeratome unlike in technique using a DMEK graft preparation
DSAEK, which makes donor preparation more device, DescePrep [9, 17, 24–27]. Femtosecond
accessible. However, DMEK donor preparation, laser (FSL)-assisted graft preparation has also
insertion and unfolding have a steeper learning been described by McKee and Jhanji in three
curve than DSAEK [9, 19], which may account patients [28]. In this technique, the donor cornea
for the slower adoption of this technique in some is mounted on an artificial anterior chamber and a
centres. Thus, some centres have reported a partial deep circular cut is fashioned with the use
higher rate of complications such as graft detach- of FSL through the posterior stroma, DM and
ment requiring postoperative re-bubbling and pri- corneal endothelium.
mary graft failure in DMEK compared to DSAEK During graft preparation, several vital dyes,
[14, 19, 21–23]. including trypan blue or VisionBlue (D.O.R.C.,
In view of the above-described challenges Zuidland, The Netherland), Membrane Blue
faced with DMEK, various improvement and Dual (D.O.R.C., Zuidland, The Netherland) and
modifications have been described to further Brilliant Blue G (Sigma-Aldrich, St. Louis,
refine the surgical techniques of DMEK, with an USA), can be used to enhance the visualisation of
aim to reduce complications and improve clinical the thin DM-endothelium tissue [8, 9, 29, 30].
outcomes. Thus, we aim to provide an overview After the peripheral donor DM is completely
of recent innovations in DMEK, including donor detached from the posterior stroma, the DMEK
insertion (endothelium-out) and pull-through graft is then peeled and harvested using the “sub-
(endothelium-in) devices, and novel techniques merged cornea using backgrounds away
for DMEK in complex eyes. In each section, we (SCUBA)” technique [31–33]. By using this
also describe the original techniques and high- technique, the DM can be peeled more easily
light the innovative measures that have been without any tear or break, with >95% success
introduced. rate in DMEK graft preparation.
In the original technique, the graft orientation
was primarily ascertained based on the inherent
Graft Preparation endothelium-out scrolling pattern of the donor
DM tissue. However, eyes with poor intraocular
In this section, we describe the main techniques view or donor tissues with less scrolling (observed
and innovations related to DMEK donor prepara- in older donors) may cause difficulty in deter-
tion, which involves donor stripping and mining the graft orientation, which can lead to
marking. inadvertent implantation of an upside-down graft
29 Innovations in Descemet Membrane Endothelial Keratoplasty (DMEK) 417
and consequent primary DMEK graft failure [34, prepared pre-stripped and pre-loaded DMEK
35]. To address this, several innovations have graft for clinical use [41]. Preparation of the
been proposed and implemented for marking the donor tissue by the eye bank can also increase the
graft to ensure its correct orientation [34, 36]. consistency and quality of the graft as the eye
One of the most common methods is the use of a bank technicians will have more access and expe-
stromal window in marking the anterior part of rience in donor tissue preparation (compared to
the DM. In this technique, the donor the surgeons who are likely to perform 1–2
DM-endothelium is partially peeled away from DMEK per week or even less in some centres).
the stroma, followed by the creation of a small The DMEK graft can be pre-loaded in the chosen
stromal window using a 3–4 mm diameter skin injector either in an endothelium-out or endothe-
punch. The DM-endothelium is then completely lium-in fashion. Chen et al. [42] compared the
placed back onto the stroma. After drying the endothelial cell viability between preloaded
excess fluid from the graft preparation bed (to scrolled DM-endothelium (endothelium-out) and
ensure complete attachment between tri-folded DM-endothelium (endothelium-in)
DM-endothelium and stroma), the donor cornea technique and demonstrated similar cell viability
is then flipped to face epithelium-side up. The between the two techniques (86.3% vs. 85.2%) at
anterior part of the DM is then accessed via the 4 days post-loading into the injector. In a recent
stromal window and marked with a violet ink- multi-centre study, Parekh et al. [43] reported the
stained “S” or “F” stamp [37]. The marked, clinical outcomes of preloaded DMEK for FECD
incompletely detached DM-endothelium is then or PBK and found that the rate of graft detach-
punched with a 7.5–8.0 mm trephine and com- ment was around 40% and the mean endothelial
pletely released from the stroma, in preparation cell loss was 46% and 48% at 3-month and 1-year
for graft loading. post-DMEK surgery, respectively. Therefore,
While this technique produces consistent while pre-loaded DMEK has its inherent advan-
marking of the DMEK graft, the creation of a tages, this needs to be balanced with the slightly
stromal window negates the possibility of using higher rate of graft detachment and endothelial
the remaining anterior portion of donor cornea cell loss when compared to non-preloaded-
for other types of transplantation such as anterior DMEK [18, 29, 44, 45].
lamellar keratoplasty (ALK) and deep anterior
lamellar keratoplasty (DALK). To increase the
utility of the donor corneas for simultaneous Graft Insertion and Unfolding
DMEK and ALK/DALK (one donor cornea for
two different recipients), a number of innovations After the DMEK graft is marked and harvested,
have been described, which include asymmetry the free-floating graft has an inherent tendency to
marking/cutting of the edge of the DMEK graft, scroll with endothelium on the outer surface due
use of ophthalmic viscoelastic device (OVD), to a relatively higher elastin content in the ante-
and bandage contact lens interface technique [8, rior part of the DM [46]. The scroll tightness of
36, 38, 39]. These marking techniques are par- the donor DM-endothelium appears to increase
ticularly useful in the current era of corneal trans- with the decrease in donor age [47], which may
plantations where there is persistent shortage of lead to difficulties in unfolding the graft in the
donor corneas globally, for which the issue has transplanted eye. That said, reduced scroll tight-
been further exacerbated by the recent COVID-19 ness observed in older donor may cause issue in
pandemic [40]. identifying the correct orientation of the DMEK
To eliminate the risk of unsuccessful harvest- graft, highlighting the importance of graft
ing of the DMEK graft intraoperatively and to marking during the harvesting process. Once the
reduce the intraoperative time, there has been an donor DM-endothelium graft has been prepared,
increasing shift towards the use of eye bank- it can be loaded into an injector or an insertor to
a b c
d e f
Fig. 29.1 An animated video demonstrating the surgical (c) Unfolding of the graft using the tapping method. (d)
technique of an endothelium-out Descemet membrane Complete unfolding of the DMEK graft in a correct orien-
endothelial keratoplasty (DMEK) surgery using a Geuder tation (confirmed by the “S” stamp). (e) Centration of the
glass tube with injection technique. (a) A DMEK graft DMEK graft using the tapping method. (f) Air or gas tam-
(stained with vital blue dye) is loaded into the Geuder ponade of the DMEK graft against the recipient’s poste-
glass tube in an endothelium-out fashion. (b) Insertion of rior stroma
the DMEK graft into the anterior chamber of the recipient.
facilitate an endothelium-out (Fig. 29.1a–f) caused less iatrogenic damage to the donor endo-
(Video 29.1) or endothelium-in DMEK thelium (24%) compared to the other two injec-
(Fig. 29.2a–i). tors (37–38%). On the other hand, Droutsas et al.
[53] conducted a large comparative clinical study
evaluating the amount of endothelial cell loss of
onor Injection (Endothelium-out)
D the DMEK graft among three commercially
Technique available glass injectors, namely the Melles
DMEK injector (D.O.R.C., Zuidland, The
Different types of injectors, including glass injec- Netherlands), the Szurmann DMEK injector
tion devices and IOL cartridges, have been used (Geuder, Germany) and the Pasteur pipette, and
for loading an endothelium-out DMEK graft [8, found no statistically significant difference
34, 45, 48–51]. As the endothelium-out technique among them.
is associated with an inevitable contact between Once the graft is loaded into an injector, it is
the donor endothelium and the luminal wall of inserted into the anterior chamber of the recipi-
the insertion device, it can cause undesirable and ent’s eye and unfolded using a combination of
irreversible loss of corneal endothelial cells. Shen techniques and manoeuvres. To facilitate the
et al. [52] conducted an ex vivo study evaluating unfolding of DMEK graft, the majority of the
three different commercially available injectors described techniques require a relatively flat-
for DMEK, namely the Geuder glass injector, tened anterior chamber. A series of controlled
modified Jones tube and the STAAR intraocular taps on the corneal surface or short burst of intra-
(IOL) injector. It was shown that Geuder cannula cameral injection of water is then performed to
a b c
d e f
g h i
Fig. 29.2 An animated video demonstrating the surgical DMEK graft is pulled into the DMEK Endoglide. (e)
technique of an endothelium-in Descemet membrane Securing of the DMEK graft in the loader with the “load-
endothelial keratoplasty (DMEK) surgery using a DMEK ing” mark facing up. (f) The DMEK Endoglide is turned
Endoglide (Network Medical Products, North Yorkshire, around with the “insertion” mark facing up. (g) Insertion
UK) with pull-through technique. (a) A DMEK graft of the DMEK graft using a pull-through technique. (h)
(stained with vital blue dye) with complete detachment Complete unfolding of the DMEK graft while holding the
from the posterior stroma. (b) Preparation of the DMEK graft with the curved forceps. (i) Air or gas tamponade of
graft in a tri-folded, endothelium-in fashion. (c, d) After the DMEK graft against the recipient’s posterior stroma
drying the excess fluid from the graft preparation bed, the
centralise and unfold the graft. Double-roll no- cient manipulation of the graft within the anterior
touch technique, Dirisamer technique (carpet chamber.
unrolling while fixating 1 graft edge), Dapena To overcome this, Hayashi et al. [56] and
manoeuvre (small air bubble-assisted unrolling) Parker et al. [57] described an innovative double-
and single sliding cannula manoeuvre are some bubble technique to help unfold the graft in eyes
of the innovative techniques described in the lit- with deep anterior chamber and/or previous vit-
erature [8, 54, 55]. Once the graft is fully unfolded rectomy. By using a small bubble over the graft
and centred, intracameral injection of air or gas is to help partially unfold the graft, a slightly big-
performed to tamponade and attach the DMEK ger bubble is injected under the graft to help float
graft. However, such technique may be infeasible and fully unfold the graft. On the other hand,
in eyes with deep anterior chamber, previous vit- Saad et al. [58] described a relatively simple and
rectomy or aphakia where the anterior chamber reproducible technique (named Cornea Press or
cannot be sufficiently flattened to achieve effi- C-Press) to help unfolding the DMEK graft in
vitrectomised eyes. During the graft unfolding, athelium-in manner (with DM side up) will allow
cannula is inserted within the scrolled graft (DMthe graft to unfold naturally. As this technique
side) and moved left and right to open the graft obviates the need for excessing tapping and
while irrigating with balanced salt solution. At manipulation of the graft within the AC, it helps
the same time, another cannula is used to press reduce the risk of graft misorientation, the intra-
on the corneal surface to artificially induce theoperative time and the technical difficulty in
shallowing of anterior chamber. After the graft is
challenging eyes (e.g., eyes with deep anterior
fully unfolded, intracameral injections of air/gas
chamber, previous vitrectomy or poor corneal
are performed underneath to tamponade the clarity obscuring the intraoperative view;
graft. The cannula over the corneal surface is Fig. 29.3a–l).
released. Various devices, including the DMEK
Furthermore, Kobayashi et al. [59] described EndoGlide (Network Medical Products, North
an innovative technique using a 25-gauge graft Yorkshire, UK) [45] and IOL cartridges [61],
manipulator to assist an endothelium-out DMEK have been used to insert an endothelium-in
surgery. After the graft was inserted into the ante-
DMEK graft using a pull-through technique. The
rior chamber using an injector, the graft was graft is usually loaded into the devices in an
grasped throughout the unfolding and centration endothelium tri-folded inward manner and
process, until the graft is fully attached and tam-
inserted into the eye using a bimanual pull-
ponaded with air or gas. This technique was through technique, similar to the technique used
shown to help reduce the intraoperative surgical in DSAEK. A hybrid-DMEK technique has also
time by around 16 min, with comparable clinical been described, which involves using EndoGlide
outcomes. Other technique such as the use of Ultrathin DSAEK pull-through donor insertion
pars plana infusion to stabilise the anterior cham-
device and donor stroma as carrier [62]. In this
ber during DMEK surgery in previously vitrect- technique, the DM is partially peeled from a thin,
omised eyes has also been described [60]. pre-cut DSAEK donor tissue graft (~150 μm
thickness). After trephining through the donor
DM, stroma and anterior cap, the remaining
Pull-Through (Endothelium-in) Donor attached part of DM is then completely separated
Insertion Technique from the stroma. Subsequently, the anterior cap is
removed and the DM-endothelium and thin
In recent years, the endothelium-in DMEK tech- stroma are transferred together into the EndoGlide
nique has been gaining increasing popularity in and pulled towards the anterior opening of the
view of the perceived advantages over the glide to achieve a “double-coil” configuration,
endothelium-out techniques [45, 61]. First, by similar to a DSAEK graft. The donor
having an endothelium-in graft, it reduces any DM-endothelium (without the donor stroma) is
undesirable touch of the donor corneal endothe- subsequently pulled through the corneal or cor-
lium against the luminal wall of the injector. neoscleral tunnel into the anterior chamber, with
Second, as the scrolled donor DM-endothelium endothelium side down. The graft is then allowed
has a natural tendency to roll endothelium- to open spontaneously, followed by a relatively
outward, inserting the DMEK graft in an endo- full air or gas tamponade.
a b c d
e f g h
i j k l
Fig. 29.3 Intraoperative snapshots of an endothelium-in the peripheral DM strip from the central DMEK graft. (g)
Descemet membrane endothelial keratoplasty (DMEK) Marking of the DMEK graft orientation using an asym-
surgery in an eye with difficult view. (a) A decompensated metrical cut. (h) Preparation of the DMEK graft in a tri-
cornea with significant corneal oedema and haziness, folded, endothelium-in fashion. (i, j) Loading of the
obscuring the view of the intraocular structures. (b) DMEK graft into the DMEK Endoglide. (k) Insertion of
Removal of the swollen and hazy corneal epithelium to the DMEK graft into the anterior chamber. (l) Complete
improve the intraoperative view. (c) Manual descemetor- graft unfolding while securing the DMEK graft with a
rhexis. (d) Staining of the pre-stripped DMEK graft with curved forceps, followed by a complete air or gas
vital blue dye. (e) Trephination of the donor central poste- tamponade
rior cornea using an 8-mm donor punch. (f) Removal of
Innovative DMEK Techniques transplantation. In the current standard practice

of DMEK surgery, one donor cornea is utilised
Splitting of graft to address shortage for only one recipient. To increase the utilisation
of donor corneas of the donor corneas, other variants of DMEK
such as hemi-DMEK and quarter-DMEK have
Currently, there is a global shortage of donor cor- also been proposed to increase the use of one
neal tissues in both developed and developing donor cornea for two and four patients, respec-
countries [63]. Despite many initiatives have tively [67, 68]. The technique is similar to a
been introduced to increase the eye donation rate DMEK but only differs in the size and shape of
and the utilisation of the donor corneas [63–66], the graft. Three-quarter DMEK technique has
such issue remains a persistent barrier to corneal also been described in eyes with glaucoma drain-
age device to avoid secondary graft failure [69]. and endothelial cell loss at 5 years (13.6% differ-
Both hemi-DMEK and quarter-DMEK have ence at 5 years). The low rate of graft detachment
demonstrated comparable results to DMEK up to observed in FSL-assisted DMEK surgery is likely
2 years in terms of visual outcome (with 40–60% attributed to the precise sizing and cutting of the
eyes achieving a corrected-distance-visual-acuity descemetorrhexis, which helps reduce the overlap-
of 6/6 or better), albeit the central endothelial cell ping of the DMEK graft and host peripheral DM
density was shown to be lower than and minimise the area of denuded stroma uncov-
DMEK. Further studies are required to confirm ered by the DMEK graft (which can lead to bul-
the long-term clinical outcomes in a larger patient lous keratopathy) [70, 78, 79].
sample size. It is also noteworthy to mention that
all these DMEK variants have only been per-
formed in a single centre; therefore, the general- se of Intraoperative Optical
U
isability of this technique remains to be Coherence Tomography (iOCT)
elucidated.
The advent of optical coherence tomography
(OCT) has significantly revolutionised the clini-
Descemetorrhexis cal diagnosis and management of many ophthal-
mic conditions [80, 81]. Recently, there has been
Studies showed that having a descemetorrhexis an emerging interest of employing intraoperative
larger than the graft size (to avoid peripheral host OCT (iOCT) for assisting ophthalmic surgeries,
DM-graft overlap) is associated with a lower risk including both anterior and posterior segments
of postoperative graft detachment [70]. In surgeries [82, 83]. Ehlers et al. [82] previously
standard DMEK, descemetorrhexis is often per- conducted the DISCOVER study examining the
formed manually using various types of DM feasibility and utility of iOCT during ophthalmic
scorers and strippers, which may sometimes lead surgeries. They demonstrated that iOCT was able
to inconsistency in the size and shape of the des- to assist and augment the decision-making pro-
cemetorrhexis, or incomplete tear/flap of the host cess during lamellar keratoplasty by ~40%. In
DM at the periphery, which may interfere with addition, iOCT has been shown to be a useful
the attachment of the DMEK graft. tool in facilitating various steps during DMEK
In the past decade, femtosecond laser (FSL) has surgery, including graft preparation, orientation,
gained popularity in a number of ophthalmic sur- graft–host apposition, and tissue interface fluid
gical procedures, including cataract surgery [71, dynamics [84].
72], refractive surgery [73], penetrating kerato- Based on our personal experience, we have
plasty [74], DALK [75], pterygium surgery [76] found iOCT to be particularly useful in eyes
and removal of conjunctival neoplasia [77]. Pilger with difficult view (secondary to significant
et al. [78] previously explored the feasibility of corneal haze/oedema) as it facilitates the visu-
FSL in performing descemetorrhexis and found alisation of the Descemet membrane during
that FSL was able to achieve highly consistent size descemetorrhexis, graft unfolding and attach-
and shape of descemetorrhexis. Recently, Sorkin ment (Fig. 29.4a–f) (Video 29.2), and the pres-
et al. [79] reported the 5-year outcome of a novel ence of any graft–host interface fluid (Fig. 29.5).
FSL-assisted technique in performing descemetor- Additional measure such as intraoperative ret-
rhexis in DMEK (known as F-DMEK). Comparing roillumination using a light pipe could also help
with the manual descemetorrhexis technique, improve the visualisation of the DMEK graft in
F-DMEK was shown to result in substantially eyes with difficult view.
lower rates of graft detachment (33.3% vs. 6.3%)
a b
c d
e f
Fig. 29.4 The use of intraoperative optical coherence a complete unfolding of the DMEK graft. (d) The use of
tomography (iOCT) in Descemet membrane endothelial an illuminating light pipe demonstrating the position of
keratoplasty in an eye with difficult view. (a) Visualisation DMEK graft (the edge of graft is highlighted by the “red
of the Descemet membrane (DM; yellow arrows) during arrows”). (e, f) Intracameral injection of air/gas to achieve
the descemetorrhexis. (b) Demonstration of DMEK graft complete tamponade of the graft against the recipient’s
in a tight scroll with endothelium-out. (c) Confirmation of posterior cornea
Fig. 29.5 The use of intraoperative optical coherence DMEK graft to the recipient’s corneal stroma. In view of
tomography (iOCT) in Descemet membrane endothelial this intraoperative finding, further manoeuvres such as
keratoplasty in visualising the graft-host interface. The tapping of the cornea and venting incision were performed
iOCT demonstrates an incomplete attachment of the to facilitate a complete attachment of the DMEK graft
Techniques in complex eyes or yes with Aphakia or Iris Defects

E
patients Performing DMEK in eyes with aphakia or sig-
nificant iris defects can be particularly challeng-
In certain cases, DMEK techniques may require ing during donor insertion, donor unfolding and
modifications or adjunctive steps in order to achieving an effective air/gas tamponade.
improve the outcomes or reduce potential com- Recently, several innovative adjunctive tech-
plications. With the increasing experience in niques have been described to overcome this.
DMEK, many surgeons have started performing
combined DMEK-cataract surgery (or “triple Safety-Net Suture
DMEK”) in eyes with co-existing cataract and Berger et al. [90] recently described a low-cost,
corneal endothelial disease. However, there has accessible “safety-net suture” method to create a
been concern that combined cataract surgery and temporary, partial barrier between the anterior
DMEK may lead to increased risk of postopera- and posterior chambers, which enables an effec-
tive complications, particularly graft detachment tive air tamponade and reducing the risk of poste-
requiring re-bubbling [23, 85, 86]. In a recent rior graft dislocation in DMEK for eyes with
meta-analysis of 11,401 eyes, Tey et al. [87] aphakia and large iris defects. The safety-net
demonstrated that staged DMEK-cataract sur- suture is performed using a continuous 10–0
gery and combined DMEK-cataract surgery polypropylene suture placed across the anterior
achieve similar clinical outcomes and safety, chamber in a cat’s-cradle pattern anterior to the
including corrected-distance-visual-acuity, post- trabecular meshwork. The suture is left in placed
operative re-bubbling rate, endothelial cell loss, during the air tamponade of the DMEK graft
graft failure and cystoid macular oedema. intraoperatively and is removed at the conclusion
However, there are several scenarios that increase of the surgery after a period of air tamponade.
the complexity of combining DMEK with cata-
ract surgery and/or anterior segment surgery: hakic IOL Implantation
P
Shweikh et al. [91] described another inventive
yes with Angle Closure
E technique of using an implantable collamer pha-
Eyes with angle closure commonly seen in Asian kic IOL (ICL) to temporarily create an anterior
eyes can pose considerable challenges to per- and posterior chamber to enable the DMEK in an
forming a successful DMEK [88, 89]. Angle clo- aphakic eye with fixed and dilated pupil. In this
sure eyes may have floppy iris, iris damage from technique, an ICL is inserted into the eye, unfolded
previous iridotomy or iridoplasty, and high vitre- over the iris and supported over the anterior cham-
ous pressures—all of which can cause difficulty ber angle. The DMEK graft is then inserted into
during graft insertion and unfolding. In view of the anterior chamber and unfolded. The anterior
these potential issues, Hayashi et al. [88] chamber is then filled with 100% air for 5 min at
described an innovative preoperative strategy to a supraphysiological IOP. After 5 min of air tam-
increase the success of DMEK in Asian eyes with ponade, the air is partially removed and the ante-
angle closure, using a combination of preopera- rior chamber is filled with cohesive viscoelastic,
tive diuretic (e.g., mannitol or acetazolamide) followed by removal of the ICL with a pair of for-
and core vitrectomy (to reduce positive vitreous ceps. The viscoelastic is gradually removed using
pressure). Comparing with the standard tech- low-pressure irrigation and aspiration with a
nique, a similar optimised DMEK technique for Simcoe cannula, alternating with air injection to
Asian eyes was shown to reduce the surgical time ensure attachment of the DMEK graft. The sur-
(by 7 min) and corneal endothelial cell loss at gery is then concluded with an intracameral air
6 months (by 14%) [89]. injection to achieve a firm physiological IOP.
rtificial Iris Implantation

A supine posturing can now benefit from the sur-
Customised artificial iris implant serves as a use- gery. In addition, this will reduce patient’s
ful device in managing congenital and acquired inconvenience (as it obviates the need for 48 h
iris defects [92]. Recently, Ang and Tan [93] of postoperative supine posturing) and will
described a staged anterior segment reconstruc- place less demand on the limited hospital space
tion technique to facilitate DMEK in complex and resources as the patients do not need to a
eyes, including those with significant iris abnor- hospital bed postoperatively to perform supine
malities [e.g., fixed-dilated pupil, significant posturing.
peripheral anterior synechiae (PAS), and partial
to near-total aniridia]. The surgery involves a
4-step approach, starting with synechiolysis of Summary
PAS and excision of iris remnants (to widen the
anterior chamber angle), ensuring a stable and Since its inception, various innovative modifica-
well-positioned posterior chamber IOL, and tions or adjunctive techniques have been
implantation of a CustomFlex Artificial Iris (CAI; described in an attempt to improve outcomes and
HumanOptics, Erlangen, Germany). The CAI is reduce complications in DMEK. The develop-
first trephined to 1 mm less than the corneal ment of various DMEK devices has facilitated
white-to-white diameter. After inserting and both endothelium-out and endothelium-in donor
unfolding the CAI in the anterior chamber, the insertion. In addition, eyes that were once thought
edge of the CAI is secured to the anterior sclera to be less suitable, including those with angle
via four pre-placed mattress sutures. Following closure or aphakia with iris defects can now
the recovery from the initial anterior segment undergo DMEK using modified surgical tech-
reconstruction (usually around 2 months), the nique. Advances in imaging such as iOCT may
DMEK is then performed. be useful in enhancing intraoperative visualisa-
tion during graft unfolding and attachment dur-
ifficulty in Supine Posturing
D ing DMEK surgery, especially in eyes with
Traditionally with EK, the patient is instructed advanced cornea decompensation. Finally, pre-
to lie in a supine position during the immediate stripped and preloaded DMEK donors prepared
postoperative period to maximise the air tam- by eye banks may help to reduce intraoperative
ponade of the graft against the posterior aspect surgical time and lessen risk of donor preparation
of the host cornea [86, 94]. However, two failure, which may improve uptake and practice
recent studies [95, 96] have highlighted the of DMEK surgeries.
possibility of achieving good DMEK graft
attachment and outcomes without the need for
supine posturing intraoperatively or postopera- Take Home Notes
tively. These patients’ eyes were filled with • Both endothelium-out and endothelium-in
99–100% air or SF6 gas intraoperatively, and techniques serve as effective and safe DMEK
an inferior PI is performed to reduce the risk of techniques.
pupillary block. Clinically significant graft • Various recent innovations have helped refine
detachment (>30% area) was observed in only the original DMEK technique, leading to bet-
4–22%, with 5–6% cases requiring regrafting ter clinical outcome, improved graft survival
due to primary graft failure or persistent graft rate, and lower risk of complications such as
detachment. Furthermore, one of the studies graft detachment and pupillary block.
demonstrated a similar graft detachment rate • With suitable modifications to the original
when compared to a historical cohort of technique, DMEK can lead to good clinical
patients who underwent 48 h of postoperative outcomes in difficult eyes.
supine posturing [96]. These findings suggest • Pre-stripped and pre-loaded DMEK graft has
that patients who have difficulty in maintaining been shown to reduce intraoperative surgical
time and risk of graft harvesting failure, thelial keratoplasty versus penetrating keratoplasty.
though long-term outcomes remain to be Ophthalmology. 2013;120(3):464–70.
13. Ang M, Saroj L, Htoon HM, Kiew S, Mehta JS,
determined. Tan D. Comparison of a donor insertion device
to sheets glide in Descemet stripping endothelial
keratoplasty: 3-year outcomes. Am J Ophthalmol.
Funding/Support None. 2014;157(6):1163–9 e3.
14. Marques RE, Guerra PS, Sousa DC, Gonçalves AI,
Quintas AM, Rodrigues W. DMEK versus DSAEK
Conflict of Interest None. for Fuchs’ endothelial dystrophy: a meta-analysis.
Eur J Ophthalmol. 2019;29(1):15–22.
15. Ang M, Lim F, Htoon HM, Tan D, Mehta JS. Visual
acuity and contrast sensitivity following Descemet
References stripping automated endothelial keratoplasty. Br J
Ophthalmol. 2016;100(3):307–11.
1. Park CY, Lee JK, Gore PK, Lim CY, Chuck 16. Fuest M, Ang M, Htoon HM, Tan D, Mehta JS. Long-
RS. Keratoplasty in the United States: a 10-year term visual outcomes comparing Descemet stripping
review from 2005 through 2014. Ophthalmology. automated endothelial keratoplasty and penetrating
2015;122(12):2432–42. keratoplasty. Am J Ophthalmol. 2017;182:62–71.
2. Tan DT, Dart JK, Holland EJ, Kinoshita S. Corneal 17. Melles GRJ, Ong TS, Ververs B, van der Wees
transplantation. Lancet. 2012;379(9827):1749–61. J. Descemet membrane endothelial keratoplasty
3. Soh YQ, Mehta JS. Regenerative therapy for (DMEK). Cornea. 2006;25(8):987–90.
Fuchs endothelial corneal dystrophy. Cornea. 18. Deng SX, Lee WB, Hammersmith KM, Kuo AN,
2018;37(4):523–7. Li JY, Shen JF, et al. Descemet membrane endo-
4. Ting DS, Sau CY, Srinivasan S, Ramaesh K, Mantry thelial keratoplasty: safety and outcomes: a report
S, Roberts F. Changing trends in keratoplasty in the by the American academy of ophthalmology.
west of scotland: a 10-year review. Br J Ophthalmol. Ophthalmology. 2018;125(2):295–310.
2012;96(3):405–8. 19. Stuart AJ, Romano V, Virgili G, Shortt AJ. Descemet’s
5. Ang M, Soh Y, Htoon HM, Mehta JS, Tan D. Five- membrane endothelial keratoplasty (DMEK) versus
year graft survival comparing Descemet stripping Descemet’s stripping automated endothelial kera-
automated endothelial keratoplasty and penetrating toplasty (DSAEK) for corneal endothelial failure.
keratoplasty. Ophthalmology. 2016;123(8):1646–52. Cochrane Database Syst Rev. 2018;6(6):CD012097.
6. Ang M, Mehta JS, Lim F, Bose S, Htoon HM, Tan 20. Ang M, Sng CCA. Descemet membrane endothelial
D. Endothelial cell loss and graft survival after keratoplasty and glaucoma. Curr Opin Ophthalmol.
Descemet’s stripping automated endothelial kerato- 2018;29(2):178–84.
plasty and penetrating keratoplasty. Ophthalmology. 21. Heinzelmann S, Maier P, Böhringer D, Hüther S,
2012;119(11):2239–44. Eberwein P, Reinhard T. Cystoid macular oedema fol-
7. Ting DSJ, Deshmukh R, Ting DSW, Ang M. Big lowing Descemet membrane endothelial keratoplasty.
data in corneal diseases and cataract: Current appli- Br J Ophthalmol. 2015;99(1):98–102.
cations and future directions. Front Big Data. 22. Myerscough J, Roberts HW, Yu AC, Mimouni M,
2023;6:1017420. Furiosi L, Mandrioli M, et al. Factors predictive of
8. Ong HS, Ang M, Mehta J. Evolution of therapies cystoid macular oedema following endothelial kera-
for the corneal endothelium: past, present and future toplasty: a single-centre review of 2233 cases. Br J
approaches. Br J Ophthalmol. 2021;105(4):454–67. Ophthalmol. 2021;107(1):24–9.
9. Ang M, Wilkins MR, Mehta JS, Tan D. Descemet 23. Deshmukh R, Nair S, Ting DSJ, Agarwal T, Beltz J,
membrane endothelial keratoplasty. Br J Ophthalmol. Vajpayee RB. Graft detachments in endothelial kera-
2016;100(1):15–21. toplasty. Br J Ophthalmol. 2022;106(1):1–13.
10. Woo JH, Ang M, Htoon HM, Tan D. Descemet 24. Muraine M. Techniques for graft preparation in
membrane endothelial keratoplasty versus Descemet DMEK. Acta Ophthalmol. 2019;97:S263.
stripping automated endothelial keratoplasty 25. Solley KD, Berges A, Diaz C, Ostrander BT, Ding AS,
and penetrating keratoplasty. Am J Ophthalmol. Larson SA, et al. Evaluation of efficacy, efficiency,
2019;207:288–303. and cell viability of a novel Descemet membrane
11. Ang M, Mehta JS, Anshu A, Wong HK, Htoon HM, endothelial keratoplasty graft preparation device,
Tan D. Endothelial cell counts after Descemet’s strip- DescePrep, in nondiabetic and diabetic human donor
ping automated endothelial keratoplasty versus pen- corneas. Cornea. 2022;41(4):505–11.
etrating keratoplasty in Asian eyes. Clin Ophthalmol. 26. Bhogal M, Balda MS, Matter K, Allan BD. Global
2012;6:537–44. cell-by-cell evaluation of endothelial viability after
12. Bose S, Ang M, Mehta JS, Tan DT, Finkelstein two methods of graft preparation in Descemet mem-
E. Cost-effectiveness of Descemet’s stripping endo- brane endothelial keratoplasty. Br J Ophthalmol.
2016;100(4):572–8.
27. Tan TE, Devarajan K, Seah XY, Lin SJ, Peh GSL, math of the COVID-19 pandemic: an international
Cajucom-Uy HY, et al. Lamellar dissection technique perspective. Br J Ophthalmol. 2020;104(11):1477–81.
for Descemet membrane endothelial keratoplasty 41. Parekh M, Romano V, Hassanin K, Testa V,
graft preparation. Cornea. 2020;39(1):23–9. Wongvisavavit R, Ferrari S, et al. Delivering endo-
28. McKee HD, Jhanji V. Femtosecond laser-assisted thelial keratoplasty grafts: modern day transplant
graft preparation for Descemet membrane endothelial devices. Curr Eye Res. 2022;47(4):493.
keratoplasty. Cornea. 2018;37(10):1342–4. 42. Chen C, Solar SJ, Lohmeier J, Terrin S, Baliga S,
29. Rodríguez-Calvo-de-Mora M, Quilendrino R, Ham Wiener BG, et al. Viability of preloaded Descemet
L, Liarakos VS, van Dijk K, Baydoun L, et al. membrane endothelial keratoplasty grafts with
Clinical outcome of 500 consecutive cases undergo- 96-hour shipment. BMJ Open Ophthalmol.
ing Descemet’s membrane endothelial keratoplasty. 2021;6(1):e000679.
Ophthalmology. 2015;122(3):464–70. 43. Parekh M, Pedrotti E, Viola P, Leon P, Neri E, Bosio L,
30. Hayashi T, Yuda K, Oyakawa I, Kato N. Use of bril- et al. Factors affecting the success rate of pre-loaded
liant blue G in Descemet’s membrane endothelial DMEK with endothelium-inwards technique: a multi-
keratoplasty. Biomed Res Int. 2017;2017:9720389. centre clinical study. Am J Ophthalmol. 2022;241:272.
31. Maharana PK, Sahay P, Singhal D, Sharma N, https://doi.org/10.1016/j.ajo.2022.03.009.
Titiyal JS. Donor preparation in descemet membrane 44. Birbal RS, Ni Dhubhghaill S, Bourgonje VJA, Hanko
endothelial keratoplasty. New Front Ophthalmol. J, Ham L, Jager MJ, et al. Five-year graft survival
2019;5:2–6. and clinical outcomes of 500 consecutive cases
32. Price MO, Giebel AW, Fairchild KM, Price FW Jr. after Descemet membrane endothelial keratoplasty.
Descemet’s membrane endothelial keratoplasty: pro- Cornea. 2020;39(3):290–7.
spective multicenter study of visual and refractive 45. Tan TE, Devarajan K, Seah XY, Lin SJ, Peh GSL,
outcomes and endothelial survival. Ophthalmology. Cajucom-Uy HY, et al. Descemet membrane endothe-
2009;116(12):2361–8. lial keratoplasty with a pull-through insertion device:
33. Schlötzer-Schrehardt U, Bachmann BO, Tourtas surgical technique, endothelial cell loss, and early
T, Cursiefen C, Zenkel M, Rössler K, et al. clinical results. Cornea. 2020;39(5):558–65.
Reproducibility of graft preparations in Descemet’s 46. Mohammed I, Ross AR, Britton JO, Said DG, Dua
membrane endothelial keratoplasty. Ophthalmology. HS. Elastin content and distribution in endothelial
2013;120(9):1769–77. keratoplasty tissue determines direction of scrolling.
34. Veldman PB, Dye PK, Holiman JD, Mayko ZM, Sáles Am J Ophthalmol. 2018;194:16–25.
CS, Straiko MD, et al. The S-stamp in Descemet 47. Bennett A, Mahmoud S, Drury D, Cavanagh HD,
membrane endothelial keratoplasty safely eliminates McCulley JP, Petroll WM, et al. Impact of donor
upside-down graft implantation. Ophthalmology. age on corneal endothelium-Descemet mem-
2016;123(1):161–4. brane layer scroll formation. Eye Contact Lens.
35. Bardan AS, Goweida MB, El Goweini HF, Liu 2015;41(4):236–9.
CS. Management of upside-down Descemet mem- 48. Romano V, Kazaili A, Pagano L, Gadhvi KA, Titley
brane endothelial keratoplasty: a case series. J Curr M, Steger B, et al. Eye bank versus surgeon prepared
Ophthalmol. 2020;32(2):142–8. DMEK tissues: influence on adhesion and re-bubbling
36. Narayanan N, Dubey A. Resource maximization dur- rate. Br J Ophthalmol. 2022;106(2):177–83.
ing COVID-19 crunch—a novel graft marking tech- 49. Kim EC, Bonfadini G, Todd L, Zhu A, Jun
nique to use one cornea for two recipients for either AS. Simple, inexpensive, and effective injector for
Descemet membrane endothelial keratoplasty or deep descemet membrane endothelial keratoplasty. Cornea.
anterior lamellar keratoplasty. Indian J Ophthalmol. 2014;33(6):649–52.
2022;70(3):1037–41. 50. Arnalich-Montiel F, Muñoz-Negrete FJ, De
37. Veldman PB, Dye PK, Holiman JD, Mayko ZM, Sáles Miguel MP. Double port injector device to reduce
CS, Straiko MD, et al. Stamping an S on DMEK endothelial damage in DMEK. Eye (Lond).
donor tissue to prevent upside-down grafts: laboratory 2014;28(6):748–51.
validation and detailed preparation technique descrip- 51. Ang M, Mehta JS, Newman SD, Han SB, Chai J, Tan
tion. Cornea. 2015;34(9):1175–8. D. Descemet membrane endothelial keratoplasty: pre-
38. Or L, Krakauer Y, Sorkin N, Knyazer B, Ashkenazy liminary results of a donor insertion pull-through tech-
Z, Gushansky K, et al. A novel marking technique using a donor mat device. Am J Ophthalmol.
nique for Descemet membrane endothelial graft 2016;171:27–34.
using an ophthalmic viscoelastic device. Cornea. 52. Shen E, Fox A, Johnson B, Farid M. Comparing the
2021;40(4):529–32. effect of three Descemet membrane endothelial kera-
39. Basak SK, Basak S. Marking DMEK grafts using toplasty injectors on endothelial damage of grafts.
bandage contact lens Interface technique: doubling Indian J Ophthalmol. 2020;68(6):1040–3.
the utilization during the acute shortage of donor cor- 53. Droutsas K, Lazaridis A, Kymionis GD,
neas. Cornea. 2022;41(4):512–7. Chatzistefanou K, Moschos MM, Koutsandrea
40. Ang M, Moriyama A, Colby K, Sutton G, Liang L, C, et al. Comparison of endothelial cell loss and
Sharma N, et al. Corneal transplantation in the after- complications following DMEK with the use
of three different graft injectors. Eye (Lond). 67. Birbal RS, Hsien S, Zygoura V, Parker JS, Ham L, van
2018;32(1):19–25. Dijk K, et al. Outcomes of hemi-Descemet membrane
54. Vasquez-Perez A, Phylactou M, Din N, Liu C. “The endothelial keratoplasty for Fuchs endothelial corneal
spinning technique” for unfolding tightly scrolled dystrophy. Cornea. 2018;37(7):854–8.
DMEK grafts. Cornea. 2022;41(1):130–4. 68. Birbal RS, Ni Dhubhghaill S, Baydoun L, Ham L,
55. Liarakos VS, Dapena I, Ham L, van Dijk K, Melles Bourgonje VJA, Dapena I, et al. Quarter-Descemet
GR. Intraocular graft unfolding techniques in des- membrane endothelial keratoplasty: one- to two-year
cemet membrane endothelial keratoplasty. JAMA clinical outcomes. Cornea. 2020;39(3):277–82.
Ophthalmol. 2013;131(1):29–35. 69. Oganesyan O, Makarov P, Grdikanyan A, Oganesyan
56. Hayashi T, Kobayashi A. Double-bubble tech- C, Getadaryan V, Melles GRJ. Three-quarter
nique in Descemet membrane endothelial kerato- DMEK in eyes with glaucoma draining devices to
plasty for vitrectomized eyes: a case series. Cornea. avoid secondary graft failure. Acta Ophthalmol.
2018;37(9):1185–8. 2021;99(5):569–74.
57. Parker JS, Parker JS, Melles GRJ. “Double-bubble” 70. Tourtas T, Schlomberg J, Wessel JM, Bachmann BO,
Descemet membrane endothelial keratoplasty Schlötzer-Schrehardt U, Kruse FE. Graft adhesion in
unfolding in eyes with deep anterior chambers descemet membrane endothelial keratoplasty depen-
and anterior chamber intraocular lenses. Cornea. dent on size of removal of host’s descemet membrane.
2020;39(7):919–23. JAMA Ophthalmol. 2014;132(2):155–61.
58. Saad A, Awwad ST, El Salloukh NA, Panthier C, 71. Schweitzer C, Brezin A, Cochener B, Monnet D,
Bashur Z, Gatinel D. C-press technique to facilitate Germain C, Roseng S, et al. Femtosecond laser-
Descemet membrane endothelial keratoplasty sur- assisted versus phacoemulsification cataract sur-
gery in vitrectomized patients: a case series. Cornea. gery (FEMCAT): a multicentre participant-masked
2019;38(9):1198–201. randomised superiority and cost-effectiveness trial.
59. Kobayashi A, Yokogawa H, Mori N, Masaki T, Lancet. 2020;395(10219):212–24.
Sugiyama K. Development of a donor tissue hold- 72. Liu YC, Setiawan M, Chin JY, Wu B, Ong HS,
ing technique for Descemet’s membrane endothelial Lamoureux E, et al. Randomized controlled trial
keratoplasty using a 25-gauge graft manipulator. Case comparing 1-year outcomes of low-energy femto-
Rep Ophthalmol. 2018;9(3):431–8. second laser-assisted cataract surgery versus conven-
60. Sorkin N, Einan-Lifshitz A, Ashkenazy Z, Boutin T, tional phacoemulsification. Front Med (Lausanne).
Showail M, Borovik A, et al. Enhancing Descemet 2021;8:811093.
membrane endothelial keratoplasty in postvitrectomy 73. Ang M, Farook M, Htoon HM, Mehta JS. Randomized
eyes with the use of pars plana infusion. Cornea. clinical trial comparing femtosecond LASIK and
2017;36(3):280–3. small-incision lenticule extraction. Ophthalmology.
61. Busin M, Leon P, D'Angelo S, Ruzza A, Ferrari 2020;127(6):724–30.
S, Ponzin D, et al. Clinical outcomes of preloaded 74. Wade M, Muniz Castro H, Garg S, Kedhar S,
Descemet membrane endothelial keratoplasty Aggarwal S, Shumway C, et al. Long-term results of
grafts with endothelium tri-folded inwards. Am J femtosecond laser-enabled keratoplasty with zig-zag
Ophthalmol. 2018;193:106–13. trephination. Cornea. 2019;38(1):42–9.
62. Woo JH, Htoon HM, Tan D. Hybrid Descemet 75. Gerten G, Oberheide U, Thiée P. Clear cornea femto
membrane endothelial keratoplasty (H-DMEK): DALK: a novel technique for performing deep ante-
results of a donor insertion pull-through technique rior lamellar keratoplasty. Graefes Arch Clin Exp
using donor stroma as carrier. Br J Ophthalmol. Ophthalmol. 2022;260(9):2941–8.
2020;104(10):1358–62. 76. Ting DSJ, Liu YC, Lee YF, Ji AJS, Tan TE, Htoon HM,
63. Gain P, Jullienne R, He Z, Aldossary M, Acquart S, et al. Cosmetic outcome of femtosecond laser-assisted
Cognasse F, et al. Global survey of corneal trans- pterygium surgery. Eye Vis (Lond). 2021;8(1):7.
plantation and eye banking. JAMA Ophthalmol. 77. Dimacali VG, Liu YC, Ong HS, Ting DSJ, Mehta
2016;134(2):167–73. JS. Femtosecond laser-assisted excision of con-
64. Gupta N, Vashist P, Ganger A, Tandon R, Gupta junctival melanocytic lesions: cosmetic and
SK. Eye donation and eye banking in India. Natl Med long-term outcomes. Clin Exp Ophthalmol.
J India. 2018;31(5):283–6. 2021;49(3):312–5.
65. Ting DS, Potts J, Jones M, Lawther T, Armitage 78. Pilger D, von Sonnleithner C, Bertelmann E,
WJ, Figueiredo FC. Changing trend in the utilisa- Joussen AM, Torun N. Femtosecond laser-assisted
tion rate of donated corneas for keratoplasty in the Descemetorhexis: a novel technique in Descemet
UK: the North East England study. Eye (Lond). membrane endothelial keratoplasty. Cornea.
2016;30(11):1475–80. 2016;35(10):1274–8.
66. Ting DS, Potts J, Jones M, Lawther T, Armitage 79. Sorkin N, Gouvea L, Din N, Mimouni M, Alshaker
WJ, Figueiredo FC. Impact of telephone con- S, Weill Y, et al. Five-year safety and efficacy of fem-
sent and potential for eye donation in the UK: tosecond laser-assisted Descemet membrane endo-
the newcastle eye centre study. Eye (Lond). thelial keratoplasty. Cornea. 2022;42:145. https://doi.
2016;30(3):342–8. org/10.1097/ICO.0000000000003019.
80. Huang D, Swanson EA, Lin CP, Schuman JS, Stinson 88. Hayashi T, Oyakawa I, Kato N. Techniques for learn-
WG, Chang W, et al. Optical coherence tomography. ing Descemet membrane endothelial keratoplasty for
Science. 1991;254(5035):1178–81. eyes of Asian patients with shallow anterior chamber.
81. Ang M, Ting DSJ, Sng CCA, Schmetterer L. Anterior Cornea. 2017;36(3):390–3.
segment OCT: angiography. In: Alió JL, Alio del 89. Ang M, Ting DSJ, Kumar A, May KO, Htoon HM,
Barrio JL, editors. Atlas of anterior segment optical Mehta JS. Descemet membrane endothelial kerato-
coherence tomography essentials in ophthalmology. plasty in Asian eyes: intraoperative and postoperative
Cham: Springer; 2021. p. 159–69. complications. Cornea. 2020;39(8):940–5.
82. Ehlers JP, Goshe J, Dupps WJ, Kaiser PK, Singh RP, 90. Berger O, Kriman J, Vasquez-Perez A, Allan
Gans R, et al. Determination of feasibility and utility BD. Safety-net suture for Aphakic Descemet
of microscope-integrated optical coherence tomog- membrane endothelial keratoplasty. Cornea.
raphy during ophthalmic surgery: the DISCOVER 2022;41(6):789–91.
study RESCAN results. JAMA Ophthalmol. 91. Shweikh Y, Vasquez-Perez A, Allan BD. Phakic
2015;133(10):1124–32. intraocular lens as a temporary barrier in aphakic
83. Ang M, Baskaran M, Werkmeister RM, Chua J, Descemet’s membrane endothelial keratoplasty. Eur J
Schmidl D, Aranha Dos Santos V, et al. Anterior seg- Ophthalmol. 2019;29(5):566–70.
ment optical coherence tomography. Prog Retin Eye 92. Srinivasan S, Ting DS, Snyder ME, Prasad S, Koch
Res. 2018;66:132–56. HR. Prosthetic iris devices. Can J Ophthalmol.
84. Cost B, Goshe JM, Srivastava S, Ehlers 2014;49(1):6–17.
JP. Intraoperative optical coherence tomography- 93. Ang M, Tan D. Anterior segment reconstruction
assisted descemet membrane endothelial kerato- with artificial iris and Descemet membrane endo-
plasty in the DISCOVER study. Am J Ophthalmol. thelial keratoplasty: a staged surgical approach. Br J
2015;160(3):430–7. Ophthalmol. 2022;106(7):908–13.
85. Röck D, Bartz-Schmidt KU, Röck T, Yoeruek 94. Marques RE, Guerra PS, Sousa DC, Ferreira NP,
E. Air bubble-induced high intraocular pressure Gonçalves AI, Quintas AM, et al. Sulfur hexa-
after Descemet membrane endothelial keratoplasty. fluoride 20% versus air 100% for anterior cham-
Cornea. 2016;35(8):1035–9. ber tamponade in DMEK: a meta-analysis. Cornea.
86. Leon P, Parekh M, Nahum Y, Mimouni M, Giannaccare 2018;37(6):691–7.
G, Sapigni L, et al. Factors associated with early graft 95. Roberts HW, Kit V, Phylactou M, Din N, Wilkins MR.
detachment in primary Descemet membrane endothe- ‘Posture-less’ DMEK: is posturing after Descemet
lial keratoplasty. Am J Ophthalmol. 2018;187:117–24. membrane endothelial keratoplasty actually neces-
87. Tey KY, Tan SY, Ting DSJ, Mehta JS, Ang M. Effects sary? Am J Ophthalmol. 2022;240:23–9.
of combined cataract surgery on outcomes of 96. Parker JS, Parker JS, Tate H, Melles GRJ. DMEK
Descemet’s membrane endothelial keratoplasty: a without postoperative supine posturing.
systematic review and meta-analysis. Front Med. Cornea. 2022;42:32. https://doi.org/10.1097/
2022;9:857200. ICO.0000000000003000.
Descemet Stripping Only (DSO)
30
Maryam Eslami and Greg Moloney
Key Points sensitivity [2]. In the late stages, vision loss

• The history of the Descemet stripping-only becomes constant as corneal oedema worsens,
procedure is outlined in this chapter. and the patient may experience intermittent pain
• The suspected mechanism of corneal clear- from ruptured bullae and epithelial defects [2].
ance and the role of Rho Kinase inhibitors in Prior to the advent of endothelial keratoplasty
this procedure is explained. (EK), penetrating keratoplasty (PK) was the only
• The importance of patient selection and other surgical option to treat FECD [3]. With the refine-
predictors of successful outcomes are ment of the surgical technique in the preparation
described in detail. of both the host and donor cornea in the past two
• Surgical steps for a successful corneal clear- decades, EK has become the gold standard surgi-
ance are summarized, and common pitfalls are cal treatment of FECD and has outnumbered PK
highlighted. in the United States since 2012 [3]. The two com-
monly utilized techniques of EK are Descemet
Fuchs’ endothelial corneal dystrophy (FECD) stripping automated endothelial keratoplasty
is the most common posterior corneal dystrophy (DSAEK) and Descemet membrane endothelial
[1]. It is characterized by progressive endothelial keratoplasty (DMEK).
cell loss and Descemet membrane (DM) excres- Posterior lamellar keratoplasty was first per-
cences called guttae, which may lead to gradual formed by Tillett in 1956 using lamellar posterior
and fluctuating decreased vision and contrast dissection and suture fixation of donor cornea to
the recipient [4]. Melles et al. outlined the poste-
Supplementary Information The online version con- rior surgical approach to EK in 1998 [5] and fur-
tains supplementary material available at https://doi. ther refined it in 2004 [6] using descemetorhexis
org/10.1007/978-3-031-32408-6_30. to prepare the host cornea. Additional surgical
M. Eslami
Ophthalmology and Visual Sciences, University of
British Columbia, Vancouver, BC, Canada
G. Moloney (*)
University of British Columbia,
Vancouver, BC, Canada
Sydney Eye Hospital, Sydney University,
Sydney, Australia
https://doi.org/10.1007/978-3-031-32408-6_30
432 M. Eslami and G. Moloney
advancements, including the use of an automated reported studies. Iovieno et al. also remarked on
microkeratome to prepare donor lenticule [6], the induced postoperative astigmatism from pos-
gave rise to the DSAEK technique utilized today. terior elevation in the patients who achieved
In 2006, Melles et al. introduced DMEK, which, complete corneal clarity; these results were pos-
unlike DSAEK, involved selective transplanta- sibly affected by posterior stromal trauma, the
tion of donor DM and endothelial layer only [7]. significance of which has become better under-
In both techniques, the donor cornea is approxi- stood over time [23].
mated to the host stroma using air or gas bubbles Early in this surgical journey, the mechanism
[8]. of spontaneous corneal clearance after DMEK/
The most common early postoperative com- DSAEK graft detachment was debated [24].
plication of DSAEK and DMEK is partial or Some authors hypothesized that the repopulation
complete graft detachment [8]. Interestingly, of endothelial cells is a result of the migration of
multiple cases were reported on spontaneous transplanted donor endothelial cells from the
clearance of cornea and repopulation of endothe- attached donor–host area in a partially detached
lial cells following complete graft detachment graft or from the anterior chamber in a com-
[9–16]. Additionally, multiple authors noted sim- pletely detached graft [25]. This procedure was
ilar findings after inadvertent central descemeto- therefore named DMET, Descemet membrane
rhexis during cataract surgery [16, 17]. These endothelial transfer. However, it is now believed
findings prompted experimentation with desce- that the repopulated endothelial cells originate
metorhexis only as the primary intervention for from the stimulated host endothelium [24, 26,
corneal endothelial disease. This technique was 27]. Prior studies have shown that endothelial
first proposed as an intervention decades earlier cells have mitotic regeneration capabilities
by Paufique, who described this surgery, with in vitro, which are arrested in vivo in the G1
accompanying illustration in 1955 [18]. The first phase of the cell cycle [3, 28]. This mitotic quies-
modern case report of intentional stripping with- cence is thought to be secondary to TGF-β inhibi-
out graft placement was in 2012. Shah et al. tion in aqueous humour, lack of effective growth
reported on the intentional Descemet stripping factor stimulation and cell–cell contact inhibition
procedure after the fellow eye’s “unusual pattern [28]. It is hypothesized that the loss of cell–cell
of healing” following DSAEK graft detachment contact inhibition in DSO promotes peripheral
[19]. In 2013, Bleyen et al. reported on 8 eyes endothelial cells to migrate and possibly prolifer-
with phacoemulsification combined with 8 mm ate in the denuded area of descemetorhexis [24,
DSO [20]. They observed some corneal clearing 27]. A study of cell counts over time demonstrat-
in 3/8 eyes with 7/8 eyes requiring grafting within ing a reduction in overall counts post-DSO would
18 months [20]. Subsequently, Arbelaez et al. suggest that without additive stimulation of mito-
tried smaller 6–6.5 mm DSOs in 3 eyes but with sis, corneas achieve clearance post-DSO primar-
variable and disappointing results [21]. ily via migration of existing cells [29]. This
In a report of a successful corneal clearance hypothesis along with other medical treatments
on 2/2 eyes using a 4 mm descemetorhexis in for in-vivo reactivation of regenerative capabili-
2015, Moloney et al. emphasized the importance ties of endothelial cells is currently the subject of
of patient selection, hypothesized higher success intense research. Notable among these is the use
with smaller descemetorhexis and recommended of Rho-Kinase (ROCK) inhibitors in corneal
the procedure for patients with dense central gut- endothelial disease.
tae but adequate peripheral endothelial reserve ROCK is a serine/threonine protein kinase
[16]. Borkar et al. [21], Malyugin et al. [22] and with two isoforms, ROCK1 and ROCK2, that
Iovieno et al. [23] also reported a higher success phosphorylate several targets with multiple
rate at clearing cornea using 4 mm descemeto- downstream cellular effects but primarily result
rhexis. However, the variability of response rate in alteration of the internal cell cytoskeleton [30,
and time to resolution remained a challenge in all 31]. This results in altered cellular adhesion,
30 Descemet Stripping Only (DSO) 433
membrane permeability, motility, proliferation, postoperative day [42]. They also reported the
differentiation, apoptosis and extracellular matrix best spectacle-corrected visual acuity gain of
dynamics [30, 31]. In ophthalmology, two ROCK 0.16 LogMAR post-DSO that was statistically
inhibitors, netarsudil and ripasudil, are under significant [42].
investigation for their potential use in lowering The true impact of DSO on visual acuity and
intraocular pressure in glaucoma, intravitreal refractive change is currently unclear in the liter-
injection for diabetic retinopathy and endothelial ature owing to some reports being combined with
wound healing in cornea [30]. Ripasudil 0.4% cataract surgery confounding the results. The
(Glanatec; Kowa Co. Ltd., Tokyo, Japan) was systematic review and meta-analysis mentioned
approved in Japan in 2014 for glaucoma [32]. above found statistically significant improvement
Netarsudil 0.02% is a ROCK inhibitor with added in visual acuity in both DSO only and DSO com-
norepinephrine transport inhibition, marketed as bined with cataract surgery [41]. Some authors
Rhopressa (Aerie Pharmaceuticals, Bedminster, have postulated that the observed induced poste-
NJ), which was approved by the United States rior astigmatism mentioned earlier may be due to
Food and Drug Administration in 2017 for simi- stromal fibrosis resulting from stromal scoring
lar indication [33]. Currently, their use in corneal intraoperatively [43]. Interestingly, in their 5-year
endothelial disease is considered off-label. follow-up report, Iovieno et al. observed an
ROCK inhibitors promote corneal wound improvement in their previously reported poste-
healing through three mechanisms: inhibiting rior stromal opacities and irregular astigmatism
apoptosis, promoting migration and increasing [44]. In a comparative cohort study of DSO ver-
intercellular adhesions [30]. Stimulation of pro- sus DMEK, Huang et al. reported similar visual
liferation is also postulated. Extensive work done outcomes in mild to moderate FECD with a
by Okumura, Kinoshita and Koizumi et al. in this higher rate of adverse events in the DMEK group
area has demonstrated a high safety profile and [45]. However, similar to other studies of DSO,
has shown promise in an array of corneal endo- the authors emphasized the importance of patient
thelial cell diseases including FECD [34–38]. selection to achieve their reported outcomes.
Moloney et al. were the first to use ROCK Due to its reliance on an existing cell popula-
inhibitors as a salvage treatment for non-clearing tion to migrate, Descemet stripping only is not
corneas after DSO [39]. They reported rapid and suitable in cases of pseudophakic bullous kera-
complete clearance of residual corneal edema topathy or other endothelial cell diseases with a
within 2 weeks from initiation of ripasudil in 3 more diffuse endotheliopathy as reported by mul-
eyes that failed to clear initially. In a comparative tiple authors [3, 25, 34]. However, it can be con-
study of patients undergoing DSO with or with- sidered the primary procedure of choice for
out ripasudil, Macsai and Shiloch report higher FECD in isolation or in combination with cata-
endothelial cell count and faster resolution of ract surgery in patients with more central guttata
symptoms and visual rehabilitation in the group affecting visual acuity who have a higher endo-
treated with ROCK inhibitors [40]. In a recent thelial cell reserve peripherally. For instance,
systematic review and meta-analysis in 2021 Moloney et al. only included patients with a
including 127 eyes, the overall rate of DSO fail- superior cell density of 1000 or higher in their
ure was reported at 17% [41]. This figure included study mentioned above [42]. However, no corre-
patients with or without ROCK inhibitors and all lation has yet been found between peripheral cell
descemetorhexis sizes and decreased to 4% with count and the rate or speed of corneal clearance.
a descemetorhexis size of 4 mm. In their study Age was also not a contributing factor in patient
published in 2020, Moloney et al. reported a sim- selection [39, 46]. Nevertheless, the healing
ilar failure rate of 4.3% (1/23) and an average response was reported to be strikingly similar
resolution time of 4.1 weeks in their cohort of 23 between the two eyes of the same patient, imply-
patients who were all treated with 4 mm desce- ing that there are patient factors beyond age and
metorhexis and started on ripasudil from the first cell count that are yet to be uncovered [43, 46].
Local aqueous factors or underlying patient

genetics arise as areas of future study.
On the other hand, the surgical technique and
the size of descemetorhexis were found to be
important factors [42, 43, 46]. Davies et al.
reported 3 surgical techniques to remove the cen-
tral 4 mm Descemet membrane in their cohort of
17 eyes [46]. They had 3 eyes that failed to clear;
in all 3 eyes, the Descemet membrane was
removed using a reverse Sinskey hook to score
the posterior cornea 360° followed by Descemet
membrane stripping. The remainder of the eyes Fig. 30.2 16x magnification slit lamp image of DSO at
that successfully cleared within 3 months had year 2
their Descemet membrane removed with only 2
clock hours scored or with a descemetorhexis To this end, Garcerant et al. outline their pro-
technique instead [46]. The scoring may also lead posed surgical technique; after creating a 2 mm
to posterior stromal scarring, even in cases that wound and instillation of cohesive viscoelastic,
did clear eventually [43, 46]. This abnormal heal- they propose creating a small Descemet mem-
ing response may even result in posterior stromal brane tag via small side-to-side movements using
nodules, as illustrated by Garcerant et al. [43]. a reverse Sinskey at the edge of the 4 mm central
There is also evidence that endothelial cell migra- circle. This tag can then be peeled in a circular
tion is improved if the endothelial layer alone is fashion using grasping forceps [43]. Care is
stripped, leaving the Descemet membrane behind advised not to leave any Descemet tags behind.
[38]. These findings suggest that bare or rough- This technique is highlighted by the supplemen-
ened stroma may not be the most suitable bed for tal Video 30.1 attached.
endothelial cell migration, and care must be taken In summary, DSO is a safe and effective inter-
to leave the stroma undisturbed intraoperatively vention in the treatment of select cases of FECD,
[43]. Figures 30.1 and 30.2 demonstrate the cor- especially if supplemented with ripasudil or pos-
neal clarity and expected descemetorhexis sibly other ROCKi. It is more accessible, quicker,
appearance at postoperative year 2. and easier to perform than alternatives such as
DMEK, which requires a donor graft placement.
Careful patient selection is key to success, as
highlighted in this review. The long-term out-
comes of this operation still must be reported to
ensure that this is a viable surgical option for
patients in the medium to long term.
Take Home Notes

• DSO is slowly transitioning from an experi-
mental procedure to an acceptable surgical
option for select patients with Fuchs’
Dystrophy.
• It remains suitable only for patients with cen-
tral guttae but adequate peripheral endothelial
reserve.
Fig. 30.1 Intraoperative view of DSO at year 2, return to • Surgical factors of importance include limit-
theatre for phacoemulsification ing the size of the descemetorhexis to 4.5 mm
30 Descemet Stripping Only (DSO) 435
or less and avoiding stromal injury with a lium imaged over time by in vivo confocal micros-
copy. Clin Experiment Ophthalmol. 2004;32:539–42.
peeling technique. 15. Choo SY, Zahidin AZ, Then KY. Re: spontane-
• A failure rate still exists with conversion to ous corneal clearance despite graft detachment in
DMEK required in some patients, but success Descemet endothelial keratoplasty. Am J Ophthamol.
rates are improved with emerging supplemen- 2010;149:531.
16. Moloney G, Chan UT, Hamilton A, et al.
tal medical therapies. Descemetorhexis for Fuchs’ dystrophy. Can J
Ophthalmol. 2015;50:68–72.
17. Pan JC, Au Eong KG. Spontaneous resolution of
corneal oedema after inadvertent “descemetorhexis”
during cataract surgery. Clin Experiment Ophthalmol.
References 2006;34:896–7.
18. Paufique. Lamellar keratoplasty. In: Rycroft BW,
1. Franceschino A, Dutheil F, Pereira B, Watson S, editor. Corneal grafts. London: Butterworth and Co;
Chiambaretta F, Navel V. Descemetorhexis without 1955. p. 132–3.
endothelial keratoplasty in Fuchs endothelial corneal 19. Shah RD, Randleman JB, Grossniklaus
dystrophy. Cornea. 2021;41(7):815. HE. Spontaneous corneal clearing after Descemet’s
2. Elhalis H, Azizi B, Jurkunas UV. Fuchs endothelial stripping without endothelial replacement.
corneal dystrophy. Ocul Surf. 2010;8(4):173–84. Ophthalmology. 2012;119:256–60.
https://doi.org/10.1016/s1542-0124(12)70232-x. 20. Bleyen I, Saelens I, Van Dooren B, et al. Spontaneous
3. Blitzer AL, Colby KA. Update on the surgical man- corneal clearing after Descemet’s stripping.
agement of Fuchs endothelial corneal dystrophy. Ophthalmology. 2013;120:215.
Ophthalmol Ther. 2020;9:757–65. 21. Borkar DS, Veldman P, Colby KA. Treatment of Fuchs
4. Melles GR, Eggink FA, Lander F, Pels E, Rietveld FJ, endothelial dystrophy by descemet stripping without
Beekhuis WH, et al. A surgical technique for posterior endothelial keratoplasty. Cornea. 2016;35:1267–73.
lamellar keratoplasty. Cornea. 1998;17:618–26. 22. Malyugin BE, Izmaylova SB, Malyutina EA, et al.
5. Melles GR, Wijdh RH, Nieuwendaal CP. A tech- Clinical and functional results of one-step phaco
nique to excise the descemet membrane from surgery and central descemetorhexis for cataract and
a recipient cornea (descemetorhexis). Cornea. Fuchs primary endothelial corneal dystrophy. Vestn
2004;23:286–8. Oftalmol. 2017;133:16–22.
6. Gorovoy MS. Descemet-stripping automated endo- 23. Iovieno A, Neri A, Soldani AM, et al.
thelial keratoplasty. Cornea. 2006;25:886–9. Descemetorhexis without graft placement for the
7. Melles GRJ, Ong TS, Ververs B, van der Wees treatment of Fuchs endothelial dystrophy: prelimi-
J. Descemet membrane endothelial keratoplasty nary results and review of the literature. Cornea.
(DMEK). Cornea. 2006;25:987–90. 2017;36:637–41.
8. Price MO, Gupta P, Lass J, Price FW. EK (DLEK, 24. Dirisamer M, Yeh RY, van Dijk K, et al. Recipient
DSEK, DMEK): new frontier in cornea surgery. Annu endothelium mayrelate to corneal clearance in
Rev Vis Sci. 2017;3:69–90. Descemet membrane endothelial transfer. Am J
9. Watson SL, Abiad G, Coroneo MT. Spontaneous Ophthalmol. 2012;154:290–6.
resolution of corneal oedema following Descemet’s 25. Dirisamer M, Ham L, Dapena I, et al. Descemet
detachment. Clin Experiment Ophthalmol. membrane endothelial transfer: “free-floating” donor
2006;34:797–9. Descemet implantation as a potential alternative to
10. Balachandran C, Ham L, Verschoor CA, et al. “keratoplasty”. Cornea. 2012;31:194–7.
Spontaneous corneal clearance despite graft detach- 26. Birbal R, Parker J, Dirisamer M, et al. Descemet
ment in Descemet membrane endothelial kerato- membrane endothelial transfer: ultimate outcome.
plasty. Am J Ophthalmol. 2009;148:227–34. Cornea. 2018;37(2):141–4. https://doi.org/10.1097/
11. Zafirakis P, Kymionis GD, Grentzelos MA, et al. ICO.0000000000001395.
Corneal graft detachment without corneal edema after 27. Anitha V, Swarup R, Ravindran M. Descemet mem-
Descemet stripping automated endothelial kerato- brane endothelial transfer (DMET) in pseudophakic
plasty. Cornea. 2010;29:456–8. bullous keratopathy after DSEK—a case report and
12. Ziaei M, Barsam A, Mearza AA. Spontaneous corneal review of literature. Cornea. 2002;41:1179. https://
clearance despite graft removal in Descemet stripping doi.org/10.1097/ICO.0000000000002942; Publish
endothelial keratoplasty in Fuchs endothelial dystro- Ahead of Print.
phy. Cornea. 2013;32:e164–6. 28. Joyce NC, Harris DL, Mello DM. Mechanisms of
13. Braunstein RE, Airiani S, Chang MA, et al. Corneal mitotic inhibition in corneal endothelium: contact
edema resolution after “descemetorhexis”. J Cataract inhibition and TGF-beta2. Invest Ophthalmol Vis Sci.
Refract Surg. 2003;29:1436–9. 2002;43:2152–9.
14. Patel DV, Phang KL, Grupcheva CN, et al. Surgical 29. Artaechevarria Artieda J, Wells M, Devasahayam RN,
detachment of Descemet’s membrane and endothe- Moloney G. 5-year outcomes of Descemet stripping
only in Fuchs dystrophy. Cornea. 2020;39(8):1048– for treating Fuchs endothelial corneal dystrophy in
51. https://doi.org/10.1097/ICO.0000000000002270. rabbit model. PloS One. 2018;13:e0191306.
30. Syed ZA, Rapuano CJ. Rho kinase (ROCK) inhibi- 39. Moloney G, Petsoglou C, Ball M, et al.
tors in the management of corneal endothelial disease. Descemetorhexis without grafting for Fuchs endo-
Curr Opin Ophthalmol. 2021;32:268–74. thelial dystrophy—supplementation with topical ripa-
31. Meekins LC, Rosado-Adames N, Maddala R, Zhao sudil. Cornea. 2017;36:642–8.
JJ, Rao PV, Afshari NA. Corneal endothelial cell 40. Macsai MS, Shiloach M. Use of topical rho kinase
migration and proliferation enhanced by rho kinase inhibitors in the treatment of Fuchs dystrophy after
(ROCK) inhibitors in in vitro and in vivo models. Descemet stripping only. Cornea. 2019;38:529–34.
Invest Ophthalmol Vis Sci. 2016;57(15):6731–8. 41. Franceschino A, Dutheil F, Pereira B, Watson SL,
https://doi.org/10.1167/iovs.16-20414. Chiambaretta F, Navel V. Descemetorhexis without
32. Garnock-Jones KP. Ripasudil: first global approval. endothelial keratoplasty in Fuchs endothelial corneal
Drugs. 2014;74:2211–5. dystrophy: a systematic review and meta-analysis.
33. Hoy SM. Netarsudil ophthalmic solution 0.02%: first Cornea. 2021;41(7):815; Publish Ahead of Print.
global approval. Drugs. 2018;78:389–96. 42. Moloney G, Garcerant Congote D, Hirnschall N, et al.
34. Okumura N, Koizumi N, Kay EP, et al. The ROCK Descemet stripping only supplemented with topical
inhibitor eye drop accelerates corneal endothe- ripasudil for Fuchs endothelial dystrophy 12-month
lium wound healing. Invest Ophthalmol Vis Sci. outcomes of the sydney eye hospital study. Cornea.
2013;54(4):2493–502. 2020;2021(40):320–6.
35. Okumura N, Inoue R, Okazaki Y, et al. Effect of 43. Garcerant D, Hirnschall N, Toalster N, Zhu M,
the rho kinase inhibitor Y-27632 on corneal endo- Wen L, Moloney G. Descemetʼs stripping without
thelial wound healing. Invest Ophthalmol Vis Sci. endothelial keratoplasty. Curr Opin Ophthalmol.
2015;56(10):6067–74. 2019;30:275–85.
36. Okumura N, Okazaki Y, Inoue R, et al. Effect of the 44. Iovieno A, Moramarco A, Fontana L. Descemet strip-
Rhoassociated kinase inhibitor eye drop (ripasudil) on ping only in Fuchs’ endothelial dystrophy without
corneal endothelial wound healing. Invest Ophthalmol use of topical rho-kinase inhibitors: 5-year follow-up.
Vis Sci. 2016;57(3):1284–92. Can J Ophthalmol. 2021;57(6):402.
37. Koizumi N, Okumura N, Ueno M, Nakagawa H, 45. Huang MJ, Kane S, Dhaliwal DK. Descemetorhexis
Hamuro J, Kinoshita S. Rho-associated kinase without endothelial keratoplasty versus DMEK for
inhibitor eye drop treatment as a possible medi- treatment of Fuchs endothelial corneal dystrophy.
cal treatment for Fuchs corneal dystrophy. Cornea. Cornea. 2018;37:1479–83.
2013;32(8):1167–70. 46. Davies E, Jurkunas U, Pineda R. Predictive factors
38. Okumura N, Matsumoto D, Fukui Y, et al. Feasibility for corneal clearance after Descemetorhexis without
of cell-based therapy combined with descemetorhexis endothelial keratoplasty. Cornea. 2018;37:137–40.
Pre-Descemets Endothelial
Keratoplasty (PDEK): Science
31
and Surgery
Harminder Singh Dua
Key Points making selective lamellar keratoplasty the

• DSAEK, DMEK and PDEK are examples of accepted norm replacing penetrating keratoplasty
endothelial keratopathy. for many indications [1–3]. For all conditions
• DMEK is a true like-for-like replacement of affecting the corneal stroma, like dystrophies,
diseased Descemet’s membrane (DM) and degenerations, scars and ectasias; deep anterior
endothelial cells (EC) with healthy donor DM lamellar keratoplasty (DALK), wherein the host
and EC. endothelium is retained, became the preferred
• PDEK involves the transplantation of the pre- option [4]. For endothelial diseases causing per-
Descemet’s layer (Dua’s layer), DM and EC. sistent corneal edema, endothelial keratoplasty
• PDEK tissue can be obtained by separating (EK), wherein the host endothelial cells are
the layers from the stroma with air or replaced, is the first choice procedure [5]. In
viscoelastic. DALK, the risk of graft failure due to endothelial
• PDEK tissue is easier to handle and unscroll rejection is eliminated, and in EK, avoidance of
in the eye as it scrolls less. A big advantage is induced astigmatism and other suture-related
that it can be obtained from very young donors problems and rapid visual recovery are distinct
as well. advantages over PK. Descemet’s stripping auto-
• PDEK gives similar visual outcomes as mated endothelial keratoplasty (DSAEK) and
DMEK without any induced refractive change. Descemet’s membrane endothelial keratoplasty
(DMEK) are EK procedures, to which Pre-
Descemet’s endothelial keratoplasty (PDEK) is
Introduction the later addition [4, 6].
For more than 100 years after the first successful

corneal transplant was performed in the human The Science Behind PDEK Surgery
eye, penetrating keratoplasty (PK) remained the
gold standard. Then occurred a paradigm shift Though improved patient outcome was the major
benefit of selective lamellar keratoplasty, another
H. S. Dua (*) significant consequential benefit was our
Professor of Ophthalmology, University of improved understanding of corneal anatomy, of
Nottingham, Nottingham, UK
concepts in lamellar corneal surgery and corneal
Consultant Ophthalmologist, Eye ENT Centre, pathology. DALK was shown to be not a
Queens Medical Centre, University Hospitals NHS
Trust, Nottingham, UK
Descemet’s baring procedure in most cases but
https://doi.org/10.1007/978-3-031-32408-6_31
438 H. S. Dua
rather that the plane of cleavage was between the central to the termination of the DM, air escaping
posterior stroma and the anterior surface of the through them accesses the plane between the
pre-Descemet’s layer (PDL) (Dua’s layer) [7]. Ex posterior surface of the PDL and anterior surface
vivo simulation of DALK by pneumo-dissection of the DM and lifts the DM off to form a type 2
in human sclero-corneal discs provided insights BB. When the fenestrations are located periph-
into posterior corneal anatomy that inform our eral to the termination of the DM, air escaping
concepts on corneal surgery today. Three types of through them enters the anterior chamber, as is
big bubbles (BB) can form; Type 1, where the often seen in DALK [8] (Fig. 31.1).
posterior wall of the BB is made of the PDL, Characterization of the PDL has demonstrated
Descemet’s membrane (DM) and endothelium; that it is made of 5 to 11 lamellae of compact type
Type 2 where the posterior wall of the BB is 1 and long spacing collagen; the lamellae are
formed by the DM and endothelium; and Mixed thinner than corresponding layers of the posterior
BB (also referred to as type 3), where both types corneal stroma and are essentially devoid of kera-
1 and 2 occur together and each may be partial or tocytes. The PDL has the highest concentration
complete [7]. A type 2 BB usually starts at the of elastin fibres relative to the rest of the cornea
periphery and spreads across the posterior sur- [7, 9] (Fig. 31.2). At the periphery, the fibres of
face of the cornea, though at times, especially in the PDL separate and fan out to continue as the
advanced keratoconus eyes, it can start centrally core of the trabecular beams of the trabecular
and remain localised. Examination of the periph- meshwork [10, 11]. The fenestrations described
ery of the PDL at the point of commencement of above are located in this part of the PDL. The
a type 2 BB after reflecting the DM revealed tiny compactness and arrangement of the lamellae,
fenestrations in the PDL, which is essentially the proteoglycan content and the lack of kerato-
impervious to air. There are 15–20 such fenestra- cyte spaces could explain why it is impervious to
tions with an average size of 20 microns present air. There is anecdotal evidence to suggest that air
singly or in clusters, and distributed randomly passes along the keratocyte spaces in the stroma
along the circumference of the periphery of the to fill the stroma and reach the plane anterior to
PDL [8]. When these fenestrations are located the PDL, where its progress is arrested, and it
a b
Fig. 31.1 Features of the periphery of the pre-Descemet’s beams of the trabecular meshwork (TM). Some fenestra-
layer (Dua’s Layer. DL). (a) The Descemet’s membrane tions are also seen. (b) The black arrows point to escaping
(DM) has been peeled off the periphery of DL to reveal a bubbles or air at the periphery. These are emerging periph-
fenestration (white arrow). Air escaping through this fen- eral to the attachment of the DM; hence, in vivo during
estration will access the plane between DM and DL and deep anterior lamellar keratoplasty they enter the anterior
create a type 2 big bubble. The area enclosed in the white chamber
circle shows the periphery of the DL fanning out as the
31 Pre-Descemets Endothelial Keratoplasty (PDEK): Science and Surgery 439
a b
Fig. 31.2 Histology and elastin staining of the pre- compact lamellae of collagen. (b) Immunostaining of the
Descemet’s layer (Dua’s layer, DL). (a) The pre- PDL for elastin (green). The entire thickness of the PDL is
Descemet’s layer (PDL/DL) is seen anterior to the staining positive for elastin
Descemet’s membrane (DM). This is made of thin and
accumulates to build enough pressure to spread 17]. The PDL on it own scrolls the least when
along the cleavage plane anterior to the PDL. The compared to DMEK tissue and PDEK tissue.
imperviousness of the PDL to air is a key feature This relates to the distribution of elastin in the
that enables the PDL to be lifted off the deep components of the respective tissues. In the PDL,
stroma when air reaches the plane between it and elastin is distributed uniformly across its entire
the posterior surface of the deep stroma. It has thickness, unlike in the DM, where it is concen-
been shown that if the DM is stripped off the trated as a band on the anterior surface, most
PDL, air injection can still produce a type 1 BB likely corresponding to the banded layer [18, 19].
[7]. However, if the PDL is ablated by excimer (Fig. 31.3). The PDL therefore can stretch and
laser treatment, a type 1 BB can never be formed return to its original shape and position without
[12]. There are therefore several features that much, if any scrolling. The anterior concentration
illustrate the uniqueness of the PDL, which bear of elastin on the DM on the other hand, forces the
on lamellar corneal surgery, including DALK and isolated DM to form a scroll with the posterior
PDEK and corneal pathology. (endothelial) surface on the outside of the scroll.
In PDEK, the posterior wall of a type 1 BB, Experimental data has confirmed that when iso-
made of the PDL, DM and endothelium, is lated scrolled DM is treated with elastase enzyme,
excised, and the composite (PDEK tissue) the scroll spontaneously unscrolls, and the DM
(Fig. 31.3) is transplanted in the host eye to disc becomes flat. Histology of the DM so treated
restore endothelial cell function. PDEK was first shows the degradation of the anterior elastin band
proposed by Dua et al. [7, 13] PDEK tissue, like [15]. In PDEK tissue where the two components
DMEK tissue (DM and endothelium only), are together, the PDL splints the DM to reduce its
scrolls with the endothelial cells on the outside of scrolling effect, and conversely, almost all the
the scroll. The scrolling, however, is less tight scrolling of PDEK tissue can be attributed to the
compared to DMEK tissue (Fig. 31.4), making DM [14].
unscrolling in the eye easier [14]. Hence, com- When air is injected in the corneal stroma to
pared to DMEK, endothelial cell loss related to create a type 1 BB, to harvest PDEK tissue, a
manoeuvres performed to open the scroll in the fairly consistent pattern of movement of air is
host anterior chamber should be less. Endothelial observed [8]. Upon emerging from the tip of the
cell loss using pneumodissection to create PDEK needle, depending on the state of hydration of the
and DMEK tissue is similar, if not slightly less donor stroma, the air may spread diffusely or as
for PDEK tissue compared to DEMK tissue [15– fine lines like cracks in glass. It then almost
440 H. S. Dua
a b
Fig. 31.3 Distribution of elastin in pre-Descemet’s endo- ing. The DM shows elastin only on its anterior surface.
thelial keratoplasty (PDEK) tissue and in Descemet’s The white arrows point to the endothelial surface of DM.
membrane (DM) shown by immunofluorescent staining. (b) The DM shows a distinct anterior band of elastin
Elastin stains green. (a) PDEK tissue shows the anterior (stars) with the rest of the DM showing no elastin stain
layer made of Dua’s layer (DL) with diffuse elastin stain-
always, tracks towards the limbus as radial

a
streaks. Upon reaching the limbus, it moves cir-
cumferentially in a clockwise and counterclock-
wise direction, as narrow white bands of 1.5 to
2 mm, until the two bands meet. Air then moves
centripetally filling the stroma. The aerated
stroma of the circumferential bands is thicker
than the central aerated stroma. As air is continu-
ally injected, multiple small pockets of air appear
b
in the central cornea, lifting the PDL. These
coalesce to form a type 1 BB that expands cen-
trifugally to reach a maximum diameter of around
8.5 mm [8]. During this process, several points of
air leakage can be seen along the circumference
of the sclero-corneal disc. For a type 1 BB to
form, a critical intra tissue pressure of air has to
be created to force air along the cleavage plane
c anterior to the PDL. If air leakage at the periph-
ery is excessive, more air has to be injected with
greater force to compensate for the leaking air
and attain the critical pressure required. This pro-
cess is difficult to control and balance. If the criti-
cal intratissue pressure is not attained, a type 1
BB will not form, and conversely, if too much
force is used, a type 1 BB could form very rap-
idly and burst with a popping sound [8].
In a series of experiments, we [20] demon-
Fig. 31.4 Scrolling characteristics of pre-Descemet’s strated that the maximum pressure reached to cre-
endothelial keratoplasty (PDEK) tissue and its component
layers. (a) PDEK tissue shows a grade 2 scroll. The pre-
ate a type 1 BB was 96.25+/−21.61 kpa, and the
Descemet’s layer/Dua’s layer (PDL/DL) were peeled mean bursting pressure for a type 1 BB was
apart from the tissue in ‘A’. (b) The PDL/DL shows mini- 66.65+/−18.65 kpa. The pressure in the type 1 BB
mal scrolling. (c) The Descemet’s membrane shows a measured 10.16+/−3.65 kpa, and the volume of a
grade 4 scroll. Almost all the scrolling of PDEK tissue is
conferred by the attached DM
type 1 BB was 0.1 mL. In these experiments, a
clamp (see below) was used to prevent any periph- side-port in the lower ring through which the
eral escape of air. The mean volume of air required needle, attached to an air-filled syringe, is inserted
to create type1 BB was 0.54 mL, but this can be in the corneal stroma to inject air (Fig. 31.5).
variable as the injected air does not always follow Occluding the fenestrations prevents any escape
the course of radial, circumferential and centripe- of injected air and stops air reaching the plane
tal movement. At any point during the injection, between DL and DM avoiding a type 2
most likely depending on the depth of the needle BB. Injected air is retained in the stroma and
in the stroma, air can access the cleavage plane and eventually makes its way to the plane along the
create a type 1 BB. The sooner this happens, the anterior surface of the PDL creating a type-1 BB,
less is the amount of air required to create the BB. which can be created in donor discs of all ages,
The formation of a type 2 BB during the pro- hence even young donor eyes with higher endo-
duction of PDEK tissue is an undesirable event. thelial cell counts can be used for PDEK [21, 22].
A type 2 BB would generate tissue suitable for Another strategy to avoid the formation of a
DMEK but not for PDEK. To avoid this the type 2 BB is adopted in the scoring technique
PDEK clamp [21] was developed by exploiting [23, 24]. Here, the extreme periphery of the DM
the knowledge of the peripheral fenestrations in along the entire circumference is scored by a
the PDL (Fig. 31.5). The PDEK clamp is manu- reverse Sinski hook. Any air that reaches the
factured by e. Janach of Italy. The clamp is made plane between the PDL and DM escapes through
of two opposing rings of 1 mm width and 9 mm the cut edge of the DM without lifting it off as a
in diameter. The sclerocorneal disc is clamped in type 2 BB. In this technique, however, the diffi-
this instrument, which compresses the peripheral culty in maintaining a balance between escaping
tissue occluding the fenestrations. There is a air and attainment of the critical intra-tissue pres-
b
c
Fig. 31.5 The Pre-Descemet’s endothelial keratoplasty grip. (c) The lower ring has a notch for inserting the nee-
(PDEK) clamp. (a) The PDEK-clamp is made of two arms dle for injecting air in the stroma, on the right of the han-
with a 9 mm ring attached at the end of each. The two dles (or on the left for left handed individuals). The
arms can be tightened by rotating the screw. (b) The con- position of the notch is indicated by a mark on the upper
tour of the rims of the two apposing rings conform to the ring
periphery of the cornea and are roughened to allow a firm
442 H. S. Dua
sure to create a type 1 BB remains, with the Descemet’s plane [30]. The PDL is also impli-
potential consequences mentioned above. cated in macular corneal dystrophy and in Lewis
It is evident from the above account that the syndrome, where there is a biclonal gammopathy
new scientific knowledge on posterior corneal with paraproteinemic keratopathy manifesting in
anatomy gained with the advent of lamellar kera- the form of a central discoid yellow-brownish
toplasty has enabled the development of adapta- discoloration in the PDL [31]. A thickened
tions, modifications and innovations in opaque PDL that could be peeled off was noted
instrumentation and surgical steps, including the during DSEK [32].
innovation of PDEK itself. As stated above, it Lamellar corneal surgery thus revealed novel
has also considerably enhanced our understand- anatomical features that in turn informed and
ing of corneal pathology, challenging concepts improved our understanding of corneal pathology
that were held as gospel for almost a century. and corneal surgery, leading to the innovation of
It is now clearly understood that Descemet’s three procedures, namely suture management of
membrane detachment (DMD) is not a detach- acute hydrops, DALK-Triple and PDEK.
ment of the DM alone. Often, the PDL is also
detached. In fact, DMD follows the exact same
patterns as three types of BB and has been clas- PDEK: Surgical Principles
sified as type 1 DMD, wherein the PDL and DM
are detached together and on optical coherence Tissue Harvesting
tomography (OCT) appear as a straight line, like
the cord of a circle; type 2 DMD, which repre- PDEK tissue is usually obtained by pneumo-
sents a detachment of the DM alone appearing as dissection. Fresh or preserved donor sclero-
a fine double contour wavy line on OCT; and corneal discs can be used. Corneal stromal
Mixed DMD (type 3) where both the PDL and swelling postmortem, or induced by storage in
DM are detached and also separated from each different media for up to 4 weeks, may be an
other, with the anteriorly located PDL detach- advantage as it would make it easier to insert the
ment appearing as a straight line and the posteri- needle or canula in the stroma with reduced risk
orly located DM detachment as a wavy line [25]. of perforation and also allow intrastromal dis-
Acute hydrops in keratoconus, hitherto con- persion of air. The principle of air injection and
sidered to be due to a tear in the DM, is due to a excising PDEK tissue is the same with the use
tear in the DM and the PDL in the background of of the PDEK clamp, scoring technique or direct
the abnormal collagen and proteoglycans of a injection [21, 23, 24] (Figs. 31.6 and 31.7). For
keratoconus cornea. DM detachment and tear in reasons explained above (under ‘science behind
keratoconus eyes do not result in acute hydrops PDEK surgery’), the use of the clamp is recom-
unless associated with a tear in the PDL as well. mended. Under an operating microscope, the
Furthermore, loss or degradation of elastin in the sclero-corneal disc, endothelial side up, is
PDL has been implicated in the pathogenesis of mounted on the lower ring with the clamp in the
keratoconus [8, 10, 26–28]. fully open position (Fig. 31.7). The disc is care-
Descemetoceles too are of type 1, protrusion fully centered on the lower ring and clamped
of the DM covered with the PDL; or type 2, pro- into position by tightening the screw on the han-
trusion of the DM alone; or type 3 a protrusion of dle. During this manoeuvre, the disc can slip
the DM covered with the PDL and a variable and become decentered. Centering is important
amount of deep corneal stroma [8, 10, 29]. as an eccentrically clamped donor disc may
Intracorneal hypopyon is due to the accumulation leave some peripheral fenestrations un-
of inflammatory debris (pus) in the plane between occluded, resulting in the escape of air and/or
a detached PDL and posterior stroma. The spread the formation of a type 2 BB. One strategy to
of infecting microbes, especially fungi, has been ensure centration is to place four ink marks on
shown to preferentially occur along the pre- the epithelial surface at 12, 3, 6, and 9 o’clock
a b c
d e f
Fig. 31.6 Harvesting Pre-Descemet’s endothelial kerato- the BB. (d) A trephine of appropriate size is placed on the
plasty (PDEK) tissue. (a) Air is injected into a sclero- area of the collapsed BB and the PDEK-tissue is tre-
corneal disc stained with vision blue. (b) A type 1 big phined. (e) The cut PDEK tissue is gently peeled off. (f)
bubble (BB), has formed. (c) The needle tip is advanced The PDEK tissue shows only mild to moderate scrolling
into the cavity of the BB, and air is aspirated to collapse giving the tissue a triangular shape
a b c
d e f
Fig. 31.7 Preparing pre-Descemet’s keratoplasty to the mark, and advanced in the posterior stroma, towards
(PDEK) tissue using the PDEK clamp. (a) The clamp is the centre of the cornea. (c) Air is injected to fill the
centred on a sclero-corneal disc, endothelial side up, and stroma and create a type 1 big bubble (BB). (d) The clamp
screwed tight. The white arrow points to the mark on the is removed and the wall of the BB is incised at the edge.
upper ring that indicates the position of the notch in the (e) Vision blue dye is injected into the BB to delineate the
inferior ring. (b) A 30 gauge needle, bent to 135 degrees circumference. (f) PDEK tissue is excised by cutting the
with the bevel up, attached to a 5 mL luer lock syringe attachment of the BB to the stroma
filled with air, is inserted in the scleral rim corresponding
444 H. S. Dua
positions, 1.5 mm inside the limbus. During the then be gently advanced into the cavity of the
tightening of the clamp, the sclero-corneal disc BB, and a little more air is injected to enlarge the
is held such that all four ink marks are visible BB further. Once a BB is obtained, it can be col-
within the diameter of the clamp ring [33]. lapsed by sucking the air out before removing
A 30 gauge needle, bent to approximately the needle, or it can be left inflated. The needle is
135° bevel up, is mounted on a 5 mL luer lock withdrawn, the clamp is opened and the tissue
syringe filled with air. It helps to move the removed. PDEK tissue is then either cut with an
plunger back and forth a few times to ensure appropriate size trephine, or the edge of the BB
ease of movement. Most plastic syringes present punctured with a lance blade. Vision Blue® is
considerable resistance before the plunger starts injected in the cavity of the BB to enhance visu-
to move. This can build up a high-pressure head alization of the circumference of the BB and the
and risk rapid formation and bursting of the PDEK tissue is excised with a pair of scissors. It
BB. The needle is inserted into the mid-corneal is left on the stromal bed, covered with balanced
stroma through the side-port of the clamp, start- salt solution (BSS) or culture medium until
ing in the rim of sclera around the cornea ready for insertion in the recipient eye.
(Fig. 31.7). A black metal plate is provided with
the clamp and can be used as a background to
improve contrast and enhance depth perception Recipient Bed Preparation
while inserting and advancing the needle. A
guiding principle is that the tip of the needle The recipient eye is prepared in a manner similar
should be positioned close to the endothelial sur- to any EK procedure. A corneal or sclero-corneal
face without risking perforation. When the tip is main entry tunnel and side ports are created. The
advanced to the centre of the disc, air is slowly area from which the recipient DM is to be
and firmly injected in a continuous manner until removed, usually larger than the size of the PDEK
a type 1 BB is formed. The tip of the needle can tissue, is marked and the DM stripped (Fig. 31.8).
a b c
d e f
Fig. 31.8 Pre-Descemet’s endothelial keratoplasty tissue is injected into the anterior chamber (AC) where it
(PDEK). (a) The host diseased Descemet’s membrane lies with the scrolls up. (e) The scrolls are gently tapped to
(DM) is scored and stripped. (b) The prepared PDEK tis- an open position, endothelial cells down, and centred. (f)
sue is stained with vision blue and loaded in an injector Air is injected between the iris and the endothelial surface
cartridge. (c) The scrolled PDEK tissue, endothelial cells of the PDEK tissue to fill the AC
outside, is nudged into the nozzle of the injector. (d) The
PDEK Tissue Preparation facing the cornea, which is the correct

and Insertion orientation.
PDEK tissue can be inserted by injecting the

scroll in the recipient eye or by the pull-through Maneuvers in the Anterior
technique. The former is the preferred option. Chamber (AC)
The PDEK tissue, whether excised with scis-
sors or trephined is carefully lifted off the stro- Once in the AC, the aim is to unfold the tissue,
mal bed. At this point, some attachments along ensure correct orientation, centre it on the pupil
the circumference or fine strands extending and appose it to the posterior surface of the cor-
between the anterior surface of the PDEK tis- nea from which the DM has been stripped. Key
sue and the stromal bed may be noted. These principles here are (a) to move the scroll and
are cut to release the PDEK tissue. The tissue unscrolled tissue into the desired position, there
can be folded with the endothelium inside, the should be some fluid in the AC; (b) to ascertain
PDL dried with a surgical sponge and an ‘F’ or orientation, the rolls of the scroll should be fac-
‘S’ mark placed if required. The tissue is then ing the cornea or the ‘F’/‘S’ marks should read
covered with BSS and allowed to form a scroll. correctly; and (c) to unscroll the tissue, the AC
It is then stained with Vision Blue® or should be kept shallow during the tapping
Membrane Blue® for 1–3 min and washed. A maneuvers. As the tissue unscrolls, the narrow
standard lens injector or glass pipette is used to space and contact with iris and cornea keeps the
introduce the tissue in the recipient’s anterior tissue from re-scrolling. Once in the correct
chamber. The injector cartridge is filled with position, orientation and fully open, air or 20%
BSS, the scroll of PDEK tissue is placed in the SF6 gas is injected under the tissue. The canula
groove and gently nudged into the nozzle. The is inserted through one of the side ports and
cartridge is mounted on a 2 mL syringe filled advanced along the iris plane to the centre of the
with BSS, and the scroll is inspected and the tissue before commencing injection. If air is
syringe is rotated such that the overlapping injected at one edge, the bubble can displace the
edges of the scroll are anteriorly located (fac- tissue and cause decentration. Once the tissue is
ing up). The nozzle is inserted in the anterior attached to the back of the cornea, ensure that
chamber regardless of the direction of the the marks, if used, are correct. The ‘F’/‘S’ marks
bevel, keeping the scroll facing up (Fig. 31.8). are best seen when the tissue is attached to the
Alternatively, a fine forceps can be inserted cornea. Air fill is then completed and maintained
through a paracentesis wound opposite to the for 10 min during which the entry sites are
main entry incision, advanced to the tip of the secured if needed (Fig. 31.8). Antibiotic, steroid
nozzle and the PDEK tissue grasped in the noz- and mydriatic either by subconjunctival injec-
zle and pulled in (pull through technique). tion or topically are administered. Some air/gas
Before loading, PDEK tissue can alternatively is released to maintain normal pressure whilst
be folded, whilst lying on the stromal bed, with retaining a full fill.
the endothelium inside and the PDL outside, in
three one-thirds parts. This is then loaded into the
cartridge as described, but the overlapping edges Post-operative Management
of the fold should be facing down (not up) before
injecting. Once in the anterior chamber, the natu- Post-operative management with topical steroids,
ral tendency of the tissue to scroll with the endo- antibiotics and mydriatic drops is continued. Pupil
thelium outside causes the folded tissue to open, block glaucoma is an issue whenever air is left in
with the endothelium facing the iris and the PDL the AC. The patient is rested in a supine position
446 H. S. Dua
and the pressure checked approximately 1 h later. Discussion

High pressure, depending on the measurement, is
managed with pupil dilation, air release, oral or When Dua et al. demonstrated the presence of the
intravenous acetazolamide or intravenous manni- PDL they also were the first to expound the con-
tol injection. An inferior peripheral iridectomy cept of using what is now termed “PDEK-tissue”
(PI) is very useful in mitigating the risk of post in endothelial keratoplasty thus: “Knowledge of
operative high pressure. The timing of the PI is the existence of this layer and its characteristics
important. It can be done preoperatively with a will influence corneal surgery; for example, the
YAG laser or intraoperatively. The latter has a plane between the DL and stroma can be exploited
major risk of inducing bleeding or fibrin release in in generating tissue for endothelial transplant,
the AC. If this were to happen, it can seriously allowing easier handling and insertion of the tis-
comprise the procedure as the fibrin traps the sue because it does not tend to scroll as much as
PDEK tissue and prevents it from unscrolling. the DM, with the DL splinting the DM [14]. This
Intraoperative PI should always be followed by concept was realized in collaboration with Dr.
careful observation for any blood or fibrin and this Amar Agarwal, and the first report on PDEK was
cleared by washing, aspiration or viscoelastic tam- published in 2014 [13].
ponade, before inserting the tissue. PDEK has several advantages as an EK proce-
Once the graft attaches to the host bed, clear- dure. The thickness of the tissue is around thirty
ing starts within a couple of days and any epithe- microns, which allows visual acuity improve-
lial bull® settle with complete stromal clearing ment to a normal level as with DMEK. The tissue
and visual improvement occurring between 1 and is easier to handle, insert and unscroll in the eye.
3 weeks (Fig. 31.9). This is partly related to the splinting of the DM
a b
c d
Fig. 31.9 Post-operative outcome after pre-Descemet’s bullae are clearly visible. (c) At 4 weeks post-operative
endothelial keratoplasty (PDEK). (a) Day one post- the cornea has completely cleared. (d) Optical coherence
operative. A small residual bubble of air is visible in the tomography at 4 weeks post-operative showing the epithe-
anterior chamber. (b) With fluorescein stain, the epithelial lium, stroma and the attached PDEK tissue
by the PDL, which restricts the amount of scroll- that the long-term outcomes of PDEK might
ing, which in turn makes it easier to unscroll in surpass those of DMEK on account of the abil-
the eye, reducing endothelial cell loss from tissue ity to use younger donor tissue and the accom-
handling. Donors of any age can be used, espe- panying high endothelial cell density coupled
cially young donors, as a type 1 BB can be cre- with less loss of cells related to preparation
ated in corneas of all ages, even in infants [22]. techniques and the limited intraocular manipu-
Younger donors come with higher endothelial lations required.
cell counts. This contrasts with DMEK tissue
which become more difficult to obtain, the Take Home Notes
younger the donor due to the stronger attachment • Pre-Descemet’s endothelial keratoplasty
of the DM with the overlying PDL. Young DM is (PDEK) is a viable option for endothelial
thinner and scrolls tightly, requiring greater keratoplasty.
manipulation inside the AC. The PDL in PDEK • PDEK tissue is made of the endothelium,
tissue contributes to the ease of handling, as it Descemet’s membrane and the pre-Descemet’s
can be stroked with a blunt spatula when center- layer (Dua’s layer).
ing the PDL in the AC. Experience has shown • PDEK tissue is obtained by injecting air into
that when younger donors are used, corneas with the donor tissue and creating a type 1 big
gross edema and opacification clear remarkably bubble.
well [22]. On the other hand, the maximum diam- • The PDEK clamp can be used for the con-
eter of PDEK is usually less than 8 mm, which is trolled creation of a type 1 BB.
the commonest diameter of DMEK tissue used. • PDEK tissue can be obtained from donor eyes
This disadvantage is more than offset by the of any age, including infant eyes.
advantages described. • Support provided by the PDL/DL limits the
EK procedures have a steep learning curve scrolling of PDEK tissue.
related to both the preparation of tissue and • PDEK tissue is easier to unscroll, centre and
transplantation. Once the techniques are mas- attach, in the recipient eye.
tered both are relatively simple but invariably
some tissue loss does occur in the process. Pre-
prepared DMEK and PDEK tissues are provided
by eye banks. PDEK can be a good procedure References
for beginners to gain confidence and refine
skills, which are transferrable to DMEK. Details 1. Moffatt SL, Cartwright VA, Stumpf TH. Centennial
review of corneal transplantation. Clin Experiment
of the technique and surgical tips for successful Ophthalmol. 2005;33(6):642–57.
PDEK have been variously published [3, 21, 2. Terry MA. The evolution of lamellar graft-
33–38]. ing techniques over twenty-five years. Cornea.
Though publications on PDEK are compara- 2000;19(5):611–6.
3. Singh NP, Said DG, Dua HS. Lamellar kera-
tively few, good outcomes comparable to other toplasty techniques. Indian J Ophthalmol.
EK procedures have been reported [39, 40]. 2018;66(9):1239–50.
PDEK is compatible with other simultaneous 4. Dua HS, Said DG. Deep anterior lamellar keratoplasty
surgical interventions performed to maximize (DALK) science and surgery. In: Albert DM, et al., edi-
tors. Albert and Jakobiec’s principles and practice of
outcomes. Combined PDEK and cataract ophthalmology. Switzerland: Springer Nature; 2021.
extraction, pupilloplasty and glued-on lens https://doi.org/10.1007/978-3-319-90495-5_218-1.
implantation have been reported [31, 41, 42] 5. Anshu A, Price MO, Tan DT, Price FW Jr. Endothelial
and also offers some advantage in phakic eyes, keratoplasty: a revolution in evolution. Surv
Ophthalmol. 2012;57(3):236–52.
especially in younger patients preserving 6. Lee WB, Jacobs DS, Musch DC, et al. Descemet’s
accommodation [43]. As PDEK becomes more stripping endothelial keratoplasty: safety and out-
acceptable and experience with it grows, more comes: a report by the American academy of ophthal-
data on outcomes will emerge. It is anticipated mology. Ophthalmology. 2009;116:18–30.
448 H. S. Dua
7. Dua HS, Faraj LA, Said DG, et al. Human corneal 22. Agarwal A, Agarwal A, Narang P, et al. Pre-Descemet
anatomy redefined: a novel pre-Descemet’s layer endothelial keratoplasty with infant donor corneas: a
(Dua’s layer). Ophthalmology. 2013;120:1778–85. prospective analysis. Cornea. 2015;34:859–65.
8. Dua HS, Faraj LA, Kenawy MB, AlTaan S, Elalfy MS, 23. Pereira NC, Forseto ADS, Maluf RCP, Dua
Katamish T, Said DG. Dynamics of big bubble forma- HS. Pre-Descemet’s endothelial keratoplasty: a
tion in deep anterior lamellar keratoplasty by the big simple, Descemet’s membrane scoring technique
bubble technique: in vitro studies. Acta Ophthalmol. for successful graft preparation. Br J Ophthalmol.
2018;96(1):69–76. 2022;106(6):786–9.
9. Bizheva K, Haines L, Mason E, MacLellan B, 24. Saint-Jean A, Soper M, Den Beste K, Iverson S, Price
Tan B, Hileeto D, Sorbara L. In vivo imaging and MO, Price FW. Technique for ensuring type I bubble
morphometry of the human pre-Descemet’s layer formation for pre-Descemet endothelial keratoplasty
and endothelium with ultrahigh-resolution optical preparation. Cornea. 2019;38(10):1336–8.
coherence tomography. Invest Ophthalmol Vis Sci. 25. Dua HS, Sinha R, D'Souza S, Potgieter F, Ross A,
2016;57(6):2782–7. Kenawy M, Scott I, Said DG. "Descemet membrane
10. Dua HS, Faraj LA, Said DG. Dua’s layer: discov- detachment": a novel concept in diagnosis and clas-
ery, characteristics, clinical applications, controversy sification. Am J Ophthalmol. 2020;218:84–98.
and potential relevance to glaucoma. Expert Rev 26. Parker J, Birbal RS, van Dijk K, Oellerich S, Dapena
Ophthalmol. 2015;10:531–47. I, Melles GRJ. Are Descemet membrane ruptures the
11. Dua HS, Faraj LA, Branch MJ, et al. The collagen root cause of corneal hydrops in keratoconic eyes?
matrix of the human trabecular meshwork is an exten- Am J Ophthalmol. 2019;205:204–5.
sion of the novel pre-Descemet’s layer (Dua’s layer). 27. Ting DSJ, Said DG, Dua HS. Are Descemet mem-
Br J Ophthalmol. 2014;98:691–7. brane ruptures the root cause of corneal hydrops in
12. Dua HS, Mastropasqua L, Faraj L, et al. Big bubble keratoconic eyes? Am J Ophthalmol. 2019;205:204.
deep anterior lamellar keratoplasty: the collagen 28. Parker JS, Birbal RS, van Dijk K, Oellerich S, Dapena
layer in the wall of the big bubble is unique. Acta I, Melles GRJ. Are Descemet membrane ruptures the
Ophthalmol. 2015;93:427–30. root cause of corneal hydrops in keratoconic eyes?
13. Agarwal A, Dua HS, Narang P, et al. Pre-Descemet’s Am J Ophthalmol. 2019;205:147–52.
endothelial keratoplasty (PDEK). Br J Ophthalmol. 29. Narang P, Agarwal A, Kumar DA. Predescemetocele:
2014;98:1181–5. a distinct clinical entity. Indian J Ophthalmol.
14. Dua HS, Termote K, Kenawy MB, et al. Scrolling 2017;65(11):1224–6.
characteristics of pre-Descemet endothelial kera- 30. Liu Z, Zhang P, Liu C, Zhang W, Hong J, Wang
toplasty tissue: an ex vivo study. Am J Ophthalmol. W. Split of Descemet’s membrane and pre-Descemet’s
2016;166:84–90. layer in fungal keratitis: new definition of corneal
15. Altaan SL, Gupta A, Sidney LE, at al. Endothelial cell anatomy incorporating new knowledge of fungal
loss following tissue harvesting by pneumodissection infection. Histopathology. 2015;66(7):1046–9.
for endothelial keratoplasty: an ex vivo study. Br J 31. Lisch W, Vossmerbaeumer U. Corneal opacity and
Ophthalmol. 2015;99:710–3. copper levels of the Lewis syndrome after sys-
16. Gamal El Din SA, Salama MM, El Shazly MI. Seven- temic chemotherapy. Am J Ophthalmol Case Rep.
day storage of pneumatically dissected Descemet’s 2020;20:100926; eCollection 2020 Dec.
endothelial grafts with and without Dua’s layer. Acta 32. Sharma VK, Sinha R, Sati A, Agarwal M. Was it
Ophthalmol. 2016;94:e130–4. Characterization of thickened Dua’s layer? clinical, tomographical, and
Endothelial, 94, e130. histopathological correlation: a case report. Eur J
17. Bedard P, Hou JH. Cell loss in pre-Descemet endo- Ophthalmol. 2020:1120672120974271.
thelial keratoplasty graft preparation. Cornea. 33. Ross AR, Said DG, Colabelli Gisoldi RAM,
2021;40(3):364–9. Nubile M, El-Amin A, Gabr AF, Abd Ed-Moniem
18. Mohammed I, Ross AR, Britton JO, Said DG, Dua M, Mencucci R, Pocobelli A, Mastropasqua L,
HS. Elastin content and distribution in endothelial Dua HS. Optimising pre-Descemet endothelial
keratoplasty tissue determines direction of scrolling. keratoplasty technique. J Cataract Refract Surg.
Am J Ophthalmol. 2018;194:16–25. 2020;46(5):667–74.
19. Lewis PN, White TL, Young RD, Bell JS, Winlove CP, 34. Jacob S. Use of pressurized air infusion for pre
Meek KM. Three-dimensional arrangement of elas- Descemet’s endothelial keratoplasty (PDEK)—the air
tic fibers in the human corneal stroma. Exp Eye Res. pump assisted PDEK technique. Open Ophthalmol J.
2015;146:43–53. 2018;12:175–80.
20. AlTaan SL, Mohammed I, Said DG, Dua HS. Air pres- 35. Agarwal A. PDEK in 15 steps. Ophthalmol. 2017.
sure changes in the creation and bursting of the type-1 https://theophthalmologist.com/subspecialties/
big bubble in deep anterior lamellar keratoplasty: an pdek-in-15-steps.
ex vivo study. Eye (Lond). 2018;32(1):146–51. 36. Tsatsos M, Mironidou M, Jacob S, Ziakas N. Factors
21. Dua HS, Said DG. Pre-Descemets endothelial kera- influencing corneal predescemetic endothelial kerato-
toplasty: the PDEK clamp for successful PDEK. Eye plasty (PDEK) graft creation: it’s all in a bubble. Hell
(Lond). 2017;31(7):1106–10. J Nucl Med. 2019;22(Suppl 2):42–6.
37. Jacob S, Narasimhan S, Agarwal A, Agarwal A, A 40. Kumar DA, Dua HS, Agarwal A, Jacob S. Postoperative
I S. Air pump-assisted graft centration, graft edge spectral-domain optical coherence tomography evalu-
unfolding, and graft uncreasing in young donor ation of pre-Descemet endothelial keratoplasty grafts.
graft pre-Descemet endothelial keratoplasty. Cornea. J Cataract Refract Surg. 2015;41:1535–6.
2017;36(8):1009–13. 41. Narang P, Mehta K, Agarwal A. Phacoemulsification
38. Studeny P, Netukova M, Hlozanek M, Bednar J, with single-pass four-throw pupilloplasty and pre-
Jirsova K, Krizova D. Frequency of complica- Descemet’s endothelial keratoplasty for manage-
tions during preparation of corneal lamellae used ment of cosmetic iris implant complication. Indian J
in posterior lamellar keratoplasty using the pneu- Ophthalmol. 2018;66(6):841–4.
modissection technique (big bubble). Cornea. 42. Narang P, Agarwal A, Kumar DA. Single-pass
2018;37(7):904–8. 4-throw pupilloplasty for pre-Descemet endothelial
39. Huang T, Jiang L, Zhan J, Ouyang C. Pre-descement keratoplasty. Cornea. 2017;36(12):1580–3.
membrane endothelial keratoplasty for treatment of 43. Dítě J, Netuková M, Klimešová KD, Studený
patients with corneal endothelial decompensation. P. Results of posterior lamellar keratoplasties in pha-
Zhonghua Yan Ke Za Zhi. 2018;54(2):105–10. kic eyes. Cesk Slov Oftalmol. 2022;78(1):20–3.
Pre-Descemet’s Endothelial
Keratoplasty (Pdek): Clinical
32
Considerations and Surgical
Details
Priya Narang and Amar Agarwal

• The PDEK graft comprises of Pre-Descemet’s
layer (Dua’s layer) along with Descemet’s Posterior lamellar keratoplasty (PLK) has revolu-
membrane and endothelium. tionized the management of corneal endothelial
• Type-1 bubble is formed for performing disorders. The widespread adoption of PLK tech-
PDEK surgery. niques has led to better visual outcomes and
• The PDEK graft is less flimsy than the DMEK enhanced globe stability due to closed chamber
graft due to the additional splinting effect of manipulation, unlike penetrating keratoplasty
Pre-Descemet’s layer. that leads to refractive instability and delayed
• Young donor grafts can be used for PDEK recovery.
surgery. Dua et al. [1] discovered and described a
novel, well-defined, acellular layer known as Pre-
Descemet’s layer (PDL; Dua’s layer)—a distinct
Take Home Notes layer that is considered to have considerable
The essential for performing a PDEK surgery is impact on posterior corneal surgery. PDL is doc-
the creation of Type-1 bubble. The graft can then umented to be an acellular layer composed of 5–8
be harvested, and endothelial keratoplasty can be lamellae of predominantly type-1 collagen bun-
performed. dle that measured approximately 10.15 ± 3.6
microns [1]. Pre-Descemet’s endothelial kerato-
plasty (PDEK) [2] originated after the descrip-
tion of PDL in 2013. Type-1 bubble achieved in
PDEK is a well-circumscribed, central dome-
shaped elevation up to 8.5 mm in diameter. The
org/10.1007/978-3-031-32408-6_32.
P. Narang
Narang Eye Care and Laser Centre,
Ahmedabad, India
A. Agarwal (*)
Dr. Agarwal’s Eye Hospital and Research Centre,
Chennai, India
https://doi.org/10.1007/978-3-031-32408-6_32
452 P. Narang and A. Agarwal
PDEK graft involves implanting a graft that corneo-scleral junction. The needle is intro-
involves a PDL with an endothelial layer. The duced up to the mid-periphery and the air is
first PDEK surgery was performed by Dr. Amar injected. This forms a Type-1 bubble that is
Agarwal in 2013. characterized by a dome-shaped elevation, is
around 7–8 mm in diameter and typically
spreads from the center to the periphery with a
Technique distinct edge all around. Trypan blue is injected
inside the bubble with a 26 G needle introduced
Donor Graft Preparation from the edge of the bubble. A corneo-scleral
scissor is used to cut the graft all around the
An air-filled 5 mL syringe attached to a 30 G edges of the bubble, and the graft is harvested
needle is introduced in a bevel-up position, with and stored in the McCarey Kaufman tissue cul-
the endothelial side up, from the rim of the ture media (Fig. 32.1).
a b
c d
Fig. 32.1 Donor graft preparation. (a) An air-filled 30G periphery with the help of a side port blade. (c) Trypan
needle is introduced from the corneo-scleral rim up to the blue is injected to stain the bb. (d) The graft is cut along
mid-periphery, and air is injected to create a type-1 big the peripheral edge of the bb with the corneo-scleral
bubble (bb). (b) The bb is punctured at the extreme scissor
32 Pre-Descemet’s Endothelial Keratoplasty (Pdek): Clinical Considerations and Surgical Details 453
Recipient Bed Preparation et al. [3], the spring of the injector is removed to
prevent any damage to the graft. The graft is
This step is essentially the same as in DMEK. slowly injected inside the eye, centered and ori-
Descemetorhexis is performed, and the scored ented, with the rolls of the scroll facing upward.
edges of DM are grasped with non-toothed for- It is gradually unrolled using air and fluidics. In
ceps and slowly stripped away from the stroma cases of extremely hazy cornea, an endoillumina-
(Fig. 32.2). tor can be used to direct the light obliquely on the
cornea so as to check the correct orientation of
the graft [4]. Once the graft is uncurled, the air is
Graft Insertion injected beneath the graft to enhance the adher-
ence to the posterior corneal stroma of the recipi-
The harvested donor graft is loaded onto the ent tissue (Fig. 32.3, Video 32.1).
injector of a foldable IOL. As described by Price
a b
c d
Fig. 32.2 Pseudophakic bullous keraopathy (one-eyed removal—note the trocar AC maintainer. (c) Single pass 4
patient) with silicone oil in the eye- Part 1. (a) throw pupilloplasty. (d) Continuous air infusion and
Pseudophakic bullous keratopathy. (b) Silicone oil descemtorhexis
a b
c d
Fig. 32.3 Pseudophakic bullous keraopathy (one-eyed patient) with silicone oil in the eye—Part 2. (a–d) PDEK graft
injected inside the Ac, unrolled and fixed
Post-op Regimen tion of moxifloxacin eye drops (4 times a day)

and prednisolone acetate 1% (6 times a day) for
The patient is advised to lie supine for the most the initial 2 weeks, 4 times daily for 1 month,
part during the first postoperative day. The stan- twice daily for 2 months and once daily thereafter
dard postoperative protocol includes the applica- for 3 months (Fig. 32.4).
a b c
Fig. 32.4 Pseudophakic bullous keraopathy (one-eyed image of a case with silicon oil in the eye and pseudophakic
patient) with silicone oil in the eye. Preop on the left and post bullous keratopathy. (b) Postoperative image of the case, (c)
op on the middle and fellow eye on the right. (a) Preoperative Dense keraopathy in the other eye of the case.
Donor Tissue achieving a graft for the PDEK procedure than a

DMEK procedure.
Donor graft characteristics form a crucial part of
tissue selection for endothelial keratoplasty pro-
cedure. Harvesting a donor graft for the DMEK I mportance of Performing an Iris
procedure from individuals less than 40 years of Repair for Endothelial Keratoplasty
age is difficult as the tissue gets rolled and curls Procedure in Complicated Cases
up against itself. Surgeons therefore prefer to use
tissue from older donors. This situation does not In cases with traumatic/iatrogenic iris defects, the
seem to be a limitation for the PDEK procedure importance of iris reconstruction with pupillo-
as the graft is comparatively thicker, and the plasty method cannot be undermined [7, 8]. An
additional layer of PDL provides a splinting inadvertent opening in the iris can act as a
effect to the tissue. potential source of air leakage into the retro-

The authors have a wide experience with the iridial space or the vitreous cavity depending
usage of young donor corneas for the PDEK pro- upon the clinical situation. This eventually
cedure. In a prospective study, the application decreases the air-tamponade effect on the donor
and usage of infant donor corneas where the lenticule and may contribute to decreased graft
donor age ranged from 9 months to 1 year have adhesion to the host interface, eventually leading
also been documented [5, 6]. to graft detachment.
In a young donor, strong adhesions are present The authors usually perform Single pass
between the DM and the PDL. Therefore, it is 4-throw (SFT) pupilloplasty [9] procedure for
easy to create a Type-1 bubble as compared to iris repair in cases associated with dislocated/
type-2 bubble, which is a pre-requisite for subluxated IOL with endothelial decompensation
DMEK. These adhesions therefore facilitate (Figs. 32.5, 32.6, 32.7 and 32.8).
a b
Fig. 32.5 Failed PK case treated with PDEK and single pass 4 throw pupilloplasty. Preop on the left and post op on the
right
Fig. 32.6 Pseudophakic bullous keratopathy with iridodialysis treated with PDEK and trocar assisted iridodialysis
repair and single pass 4 throw pupilloplasty. Preop on the left and post op on the right
Fig. 32.7 Pseudophakic bullous keratopathy in a one-eyed patient treated with PDEK and single pass 4 throw
pupilloplasty
a b
Fig. 32.8 Pseudophakic bullous keratopathy treated with PDEK and single pass 4 throw pupilloplasty. (a) Pre-op, (b)
Post-op
Combined Procedures Optimizing PDEK Graft
Double-Infusion Cannula Technique Microscope-integrated optical coherence

(DICT) tomography (i-OCT) [11] can optimize the
PDEK donor graft preparation by providing
In double infusion cannula technique (DICT) direct visualization and guiding the surgeon by
[10], two infusion cannulas are introduced: Trocar providing details of the appropriate depth of
cannula for fluid infusion is introduced at the pars corneal tissue. I-OCT has been demonstrated to
plana site and either an anterior chamber main- facilitate Type-1 bubble formation, thereby
tainer (ACM) or a trocar ACM (T-ACM) is intro- decreasing the chances of donor tissue loss.
duced in the anterior segment for air infusion I-OCT navigates the surgeon through all the
while performing a combined procedure of glued stages of PDEK surgery in addition to serving
IOL with an endothelial keratoplasty procedure. as an indicator for appropriate graft adhesion
Aphakic, prior vitrectomized eyes, and compli- and wound apposition.
cated IOL cases pose a definite challenge to graft PDEK clamp described by Dua et al. enables
unfolding and its adherence to the recipient's eye. appropriate handling of the sclero-corneal donor
In the absence of posterior segment infusion, tissue and allows consistently obtaining a PDEK
these eyes are more prone to global collapse. This graft [12, 13]. The clamp prevents the air to
automatically translates into poor outcomes for escape from the fenestration around the PDL,
the EK surgery. The infusion fluid in the vitreous thereby decreasing the chances of the formation
cavity prevents globe collapse and helps to main- of Type-2 bubble. The mean size of type-1 bubble
tain adequate pressure in the posterior segment. achieved with the PDEK clamp, as documented
This also prevents seepage of air from the anterior in a study, was 7.255 ± 0.535 × 6.745 ± 0.668 m
to the posterior segment, thereby maintaining an m, and the volume of air required to obtain type-1
adequate air tamponade anteriorly that facilitates bubble is 0.14–0.37 mL.
graft adherence intraoperatively as well as in the
postoperative period too.
PDEK Results and Analyses Summary
In our study [14], we analyzed the postoperative To summarize, PDEK is a technique that is repro-
graft thickness, graft configuration and ducible and is easy to perform. It scores over
detachment using spectral-domain optical coher- DMEK in ways that it allows the usage of donors
ence tomography (SD-OCT) at 1, 7, 30 and of any age group. This can be a huge advantage,
90 days postoperatively. especially in situations where there is a shortage
The donor age ranged from 1 to 65 years, and of pool of donor corneas.
the graft size ranged from 7.5 to 8.0 mm. The
mean graft thickness observed at 1 day was Dislosures No financial disclosures.
37.3 G3.5 mm (range 32–44 mm), at 7 days,
30 days and 90 days was 35.5 ± 3.4 mm (range
32–40 mm), 33 ± 1.8 mm (range 32–36 mm) and References
30.3 ± 2.6 mm (range 28–36 mm), respectively. A
statistically significant difference was observed 1. Dua HS, Faraj LA, Said DG, et al. Human corneal
anatomy redefined: a novel pre-Descemet’s layer
in the graft thickness over the time period. The (Dua’s layer). Ophthalmology. 2013;120:1778–85.
graft was well attached in all eyes except one eye 2. Agarwal A, Dua HS, Narang P, et al. Pre-Descemet’s
that had grade 3 detachment. Total graft detach- endothelial keratoplasty (PDEK). Br J Ophthalmol.
ment or lenticular drop was not observed in any 2014;98:1181–5.
3. Price FW Jr, Price MO. Descemet’s stripping with
case. endothelial keratoplasty in 200 eyes: early challenges
Interface haze was minimal (1 eye/12 eyes) and technique to enhance donor adherence. J Cataract
that receded over a period of 1 month postop- Refract Surg. 2006;32:411–8.
eratively. The 1-day postoperative corneal 4. Jacob S, Agarwal A, Agarwal A, et al. Endoilluminator-
assisted transcorneal illumination for Descemet
edema showed significant resolution over membrane endothelial keratoplasty: enhanced
1-week follow- up. A statistically significant intraoperative visualization of the graft in corneal
improvement in visual acuity from preopera- decompensation secondary to pseudophakic bullous
tively to postoperatively. The final corrected keratopathy. J Cataract Refract Surg. 2014;40:1332–6.
5. Agarwal A, Agarwal A, Narang P, Kumar DA,
distance visual acuity (CDVA) was 0.61 G 0.2. Jacob S. Pre-Descemet endothelial keratoplasty
In our analysis, no significant correlation with infant donor corneas: a prospective analysis.
between the graft thickness and the CDVA at Cornea. 2015;34(8):859–65.
1 day was observed. 6. Agarwal A, Narang P, Kumar DA, Agarwal A. Young
donor-graft assisted endothelial keratoplasty (PDEK/
As per Busin et al., the mean central graft DMEK) with epithelial debridement for chronic pseu-
thickness in the ultrathin DSAEK graft was dophakic bullous keratopathy. Can J Ophthalmol.
78.28 ± 28.89 mm at 3 months postoperatively 2017;52(5):519–26.
[15]. The PDEK grafts are thicker than DMEK 7. Narang P, Agarwal A. Triple procedure for pseudo-
phakic bullous keratopathy in complicated cataract
grafts but thinner than those observed in ultrathin surgery: glued IOL with single-pass four-throw pupil-
DSAEK. loplasty with pre-Descemet’s endothelial keratoplasty.
Raising the pressure in the anterior chamber J Cataract Refract Surg. 2019;45(4):398–403.
might be helpful by increasing the intracorneal 8. Narang P, Agarwal A, Kumar DA. Single-pass
4-throw pupilloplasty for pre-Descemet endothelial
pressure, as negative imbibition pressure of both keratoplasty. Cornea. 2017;36(12):1580–3.
the donor and recipient corneal tissues might add
to tissue adherence [16].
9. Narang P. Agarwal a single-pass four-throw tech- R, Pocobelli A, Mastropasqua L, Dua HS. Optimizing
nique for pupilloplasty. Eur J Ophthalmol. pre-Descemet endothelial keratoplasty technique. J
2017;27(4):506–8. Cataract Refract Surg. 2020;46(5):667–74.
10. Narang P, Agarwal A. Double-infusion can- 14. Kumar DA, Dua HS, Agarwal A, Jacob S. Postoperative
nula technique for glued fixation of intraocular lens spectral-domain optical coherence tomography evalu-
with endothelial keratoplasty. Can J Ophthalmol. ation of pre-Descemet endothelial keratoplasty grafts.
2018;53(5):503–9. J Cataract Refract Surg. 2015;41(7):1535–6. https://
11. Sharma N, Devi C, Agarwal R, Bafna RK, Agarwal doi.org/10.1016/j.jcrs.2015.05.015.
A. I-PDEK: microscope-integrated OCT-assisted pre- 15. Busin M, Madi S, Santorum P, Scoria V, Beltz
Descemet endothelial keratoplasty. J Cataract Refract J. Ultrathin Descement’s stripping automated endo-
Surg. 2021;47(12):e44–8. thelial keratoplasty with the microkeratome double-
12. Dua HS. Said DG pre-Descemets endothelial kerato- pass technique; two-year outcomes. Ophthalmology.
plasty: the PDEK clamp for successful PDEK. Eye 2013;120:1186–94.
(Lond). 2017;31(7):1106–10. 16. Dapena I, Ham L, Melles GRJ. Endothelial kerato-
13. Ross AR, Said DG, Colabelli Gisoldi RAM, Nubile M, plasty: DSEK/DSAEK or DMEK—the thinner the
El-Amin A, Gabr AF, Abd Ed-Moniem M, Mencucci better? Curr Opin Ophthalmol. 2009;20:299–307.
Descemet Membrane
Transplantation
33
Hon Shing Ong and Jodhbir S. Mehta

• Descemet Membrane Transplantation is a
form of regenerative therapy. The metabolically active corneal endothelium
• It is indicated for patients with localized endo- plays an important function in maintaining opti-
thelial dysfunction caused by FECD. mal corneal hydration, which is required to keep
• It requires acellular Descemet Membrane to the cornea transparent and essential for good
be used as a scaffold for cellular migration. vision [1, 2]. In corneal endothelial diseases,
• Adjuvant topical Rock Inhibitors are indicated there is an accelerated loss of healthy corneal
in older patients. endothelial cells (CEnCs). Below a certain
threshold, the corneal endothelium loses its abil-
ity to regulate corneal hydration. The cornea
becomes oedematous with a consequent loss in
its transparency [3–5]. This is corneal endothelial
failure, which results in visual impairment.
As human CEnCs are arrested in a dormant,
non-proliferative G1 phase of the cell cycle, their
Supplementary Information The online version con- regenerative potential in vivo is thought to be
org/10.1007/978-3-031-32408-6_33.
limited [6–9]. The restoration of vision in patients
with corneal endothelial failure thus relies on a
H. S. Ong J. S. Mehta (*)

Corneal and External Diseases Department, Corneal and External Diseases Department,
Singapore National Eye Centre, Singapore, Singapore Singapore National Eye Centre, Singapore, Singapore
Tissue Engineering and Cell Therapy Department, Tissue Engineering and Cell Therapy Department,
Singapore Eye Research Institute, Singapore, Singapore Eye Research Institute, Singapore,
Singapore Singapore
Ophthalmology and Visual Sciences Academic Ophthalmology and Visual Sciences Academic
Clinical Programme, Duke-NUS Medical School, Clinical Programme, Duke-NUS Medical School,
Singapore, Singapore Singapore, Singapore
School of Material Science and Engineering and
School of Mechanical and Aerospace Engineering,
Nanyang Technological University, Singapore,
Singapore
https://doi.org/10.1007/978-3-031-32408-6_33
462 H. S. Ong and J. S. Mehta
replacement of an exogenous source of donor corneal verticillate, ocular surface inflammation

endothelial cells through corneal transplantation. and discomfort, and reticular bullous epithelial
Corneal transplantation is thus currently the main oedema can occur [19]. Furthermore, by defini-
treatment to restore vision impaired by corneal tion, a regenerative approach to the treatment of
endothelial diseases. corneal endothelial disease requires some
With significant developments in corneal remaining healthy CEnCs in the patient’s cornea.
transplantation surgeries to treat corneal diseases, In diseases with widespread damage and loss of
good outcomes can often be achieved when such CEnCs (e.g. pseudophakic bullous keratopathy),
advanced techniques are applied. However, limi- the utility of regenerative medicine will thus be
tations do exist with all types of allogenic corneal limited. Instead, a cell-based therapeutic
transplantations. In addition to the need for a approach in these conditions which involves the
considerable level of subspecialised surgical injection of CEnCs expanded in culture may be
expertise and the cost of surgery, inherent risks of more appropriate [20]. In this chapter, we intro-
corneal transplantations include allogenic graft duce a regenerative therapeutic approach of
rejection and associated graft failure [10–12]. Descemet membrane transplantation (DMT) as a
Furthermore, performing donor-reliant corneal novel surgical technique to treat corneal endothe-
transplantations is limited by a shortage of suit- lial diseases.
able transplantable donor corneal tissues [13]. In
a 2012 report, the current donor availability only
meets 1.4% of the global requirements for cor- he Concept of Centripetal Host
T
neal transplantations, with 50% of individuals Corneal Endothelial Cell Migration
worldwide without access to suitable donor cor-
neas [13]. The DMT technique is based on the concept of
Thus, there is an impetus to pursue alternative, centripetal host corneal endothelial cell migra-
less donor-reliant treatment approaches for dis- tion. Reports using sex-mismatched transplant
eases of the corneal endothelium. One such area tissues have indicated that recipient corneal
approach is regenerative medicine. Regenerative endothelial cells in the periphery can exhibit cen-
medicine involves the reparation and restoration tripetal migration across the graft–host junction
of diseased cells or the re-distribution of remain- to populate and integrate with the endothelium of
ing healthy cells to replace lost cells in an attempt donor tissues [21]. Furthermore, superior graft
to restore physiological function. Various poten- survival is often observed in transplantations per-
tial
regenerative strategies to restore corneal formed for corneal endothelial diseases where
endothelial function have been reported [14]. the peripheral corneal endothelium is relatively
One approach to corneal endothelial regeneration preserved (e.g. Fuchs’ Endothelial Corneal
is the induction of CEnCs to leave the dormant Dystrophy), compared to conditions where there
G1 phase of the cell cycle and enter the prolifera- is more widespread endothelium damage (e.g.
tive S phase. Pharmacological agents, such as pseudophakic bullous keratopathy) [22]. Such
rho-associated protein kinase (ROCK) inhibitors, observations suggest that migration of peripheral
have been widely reported to promote cell adhe- host corneal endothelial cells to the central cor-
sion, inhibit apoptosis and enhance cell prolifera- nea contributes towards post-transplantation sta-
tion in cultured primate and human CEnCs [15, bility of the corneal endothelium.
16]. Clinical reports have described the recovery Using this concept of centripetal host corneal
of corneal endothelial function following trans- endothelial cell migration, the first surgical tech-
corneal freezing of patients with corneal endo- nique introduced to treat patients with Fuchs’
thelial dysfunction and the administration of Endothelial Corneal Dystrophy (FECD) is
ROCK inhibitor drops [17, 18]. Nevertheless, ‘Descemet Stripping Only’ (DSO), previously
side effects of ROCK inhibitors, such as conjunc- known as ‘Descemet Stripping Without
tival hyperaemia, subconjunctival haemorrhage, Endothelial Keratoplasty’ (DWEK) [23]. In
33 Descemet Membrane Transplantation 463
DSO, the central diseased Descemet membrane ‘Submerged Cornea Using Backgrounds Away’
of the patient is removed (descemetorhexis) with- (SCUBA) technique designed for Descemet
out transplantation, with the anticipation that membrane endothelial keratoplasty (DMEK)
healthy peripheral corneal endothelial cells will graft harvesting. A small acellular Descemet
migrate centrally to establish a functional corneal membrane graft (4–5 mm) was harvested using a
endothelium. Nonetheless, the evidence of DSO free-hand corneal trephine. It was also marked to
is based mostly on non-comparative clinical case ensure the correct orientation was maintained
series, where varying results have been reported during graft insertion.
[24–28].
Our group observed that an intact Descemet
membrane enhances the migration of endothelial Descemet Membrane Graft Insertion
cells and facilitates the formation of a corneal
endothelial monolayer [29]. The presence of a The technique of insertion of the acellular
Descemet membrane has also been shown to Descemet membrane graft was similar to that
minimize endothelial–mesenchymal transition used for standard DMEK surgery. Figure 33.1d–f
(EMT) when compared to endothelial cell migra- illustrate the surgical steps of DMT in a patient
tion across a bare posterior corneal surface [30]. who had undergone a DMT procedure. A desce-
Based on such observations, a technique of DMT metorhexis was performed to remove the central
was introduced [31]. In DMT, an acellular diseased recipient Descemet membrane and
Descemet membrane is transferred and made to endothelium. The diameter to be stripped was
adhere to the patient’s posterior cornea over the 0.5 mm larger than the size of Descemet’s mem-
central area of descemetorhexis [29, 31]. We brane graft that was to be transplanted. The har-
observed the restoration of functional integrity of vested Descemet membrane graft was stained
the corneal endothelium through the central using vital dyes (Membrane Blue Dual®,
migration of peripheral endothelial cells [29, 31]. D.O.R.C, The Netherlands), loaded and injected
This technique also allowed us to remove a larger into the eye using a standard glass injector used
area of diseased central Descemet membrane. In for DMEK (Geuder, Heidelberg, Germany)
2018, we reported a first-in-man clinical trial of through a 2.8-mm temporal corneal incision. An
using DMT as a regenerative therapeutic anterior chamber maintainer was placed to avoid
approach for FECD [31]. anterior chamber collapse during graft insertion.
Within the anterior chamber, the graft was
unfolded through various tapping manoeuvers
Surgical Technique over the corneal surface, whilst maintaining a
correct graft orientation. It was then made to
Descemet Membrane Graft adhere to the posterior corneal surface within the
Preparation area of descemetorhexis using a non-expansile
concentration of sulphur hexafluoride gas (SF6
A cadaveric donor cornea of low corneal endo- 20%) (Video 33.1).
thelial cell density, unsuitable for penetrating or Post-operatively, patients were instructed to
endothelial keratoplasty was procured from the adopt a face-up posture for at least 3 h. They were
eye bank. Corneal endothelial cells were removed prescribed topical antibiotics (levofloxacin 0.5%,
from the donor cornea using a double freeze- Santen Pharmaceutical, Osaka, Japan) and ste-
thaw cycle, followed by denudation with a cus- roids (dexamethasone 0.1%, Alcon, TX USA),
tomized silicone soft-tip cannula (catalogue which were tapered off after 1 month. For patients
number: SP-125053, ASICO, USA). (Fig. 33.1a– over 50 years of age, a topical ROCK inhibitor
c) A DMT graft was then harvested using the was also prescribed in the post-operative period.
a b c
d e f
Fig. 33.1 Surgical technique of the Descemet Membrane donor DM showing that it is completely acellular; (c, d)
Transplantation. (a) Following a double freeze-thaw Descemet Membrane graft is loaded into a glass injector
cycle, donor Descemet membrane (DM) is scraped with (Geuder, Heidelberg, Germany) and inserted into the
custom-made silicone tip cannula (SP-125053, ASICO, recipient’s anterior chamber; (e) the Descemet Membrane
USA) to remove endothelial cells; (b) donor DM is scored graft is unfolded using various tapping techniques on the
and peeled and trephined in a similar technique to recipient’s corneal surface; (f) gas bubble is injected into
Descemet Stripping Endothelial Keratoplasty (DMEK); the anterior chamber to provide tamponade of the
note trypan blue stains the entire posterior surface of the Descemet’s membrane graft
the severity of corneal guttata and oedema (Konan

Case Series Medical Corp, Hyogo, Japan). All patients had
relatively preserved peripheral corneal endothe-
At the time of writing, we have performed a total lium, as illustrated in Fig. 33.2.
of seven cases of DMT at the Singapore National As all patients had significant lenticular opac-
Eye Centre (Table 33.1). All patients were ities, a combined triple procedure of phacoemul-
recruited under a clinical trial. Ethical approval sification with an intraocular lens implant
of the study protocol was granted by the (phaco/IOL) for the management of cataracts
SingHealth Centralized Institutional Review and DMT for the management of FECD was
Board (IRB reference number: R1366/52/2016). offered. Post-operative topical ROCK inhibitor
All surgeries and evaluations were performed in (Netarsudil, Aerie Pharmaceuticals, New Jersey,
accordance with the tenets of the Declaration of United States) was administered to five of the
Helsinki. patients, starting with a four times daily dosing
The diagnosis of all patients was Fuchs’ and tapering over the course of four to 6 months,
Endothelial Corneal Dystrophy (FECD). The depending on the clinical course. Two patients
baseline preoperative visual acuities of the patients experienced conjunctival hyperaemia, and one
ranged from 6/12 to 6/21. Preoperative ultrasound patient (Case 6) developed a reticular bullous
measured central corneal thickness ranging from epithelial oedema (Fig. 33.3), thought to be asso-
576 μm to 778 μm (Sonogage Inc., Cleveland, ciated with the topical administration of
USA). For all patients, the preoperative central Netarsudil drops.
corneal endothelial cell densities assessed through During the 6 month post-operative period, six
specular microscopy were not recordable due to out of the seven patients responded favourably
Table 33.1 Demographic data on patients included in the study
Pre-operative Pre-operative Post-operative
Age, Sex, Best corrected CCT Resolution of Best corrected Post-operative ROCK inhibitor
Case Ethnicitya Diagnosis Procedure performed visual acuity (μm) corneal oedema visual acuityb CCT (μm)b (Netarsudil)
1 56, F, C FECD Phaco/IOL/DMT 6/18 603 Yes 6/7.5 483 No
2 37, F, P FECD Phaco/IOL/DMT 6/12 673 Yes 6/9 632 No
3 51, M, A FECD Phaco/IOL/DMT 6/21 645 Yes 6/12 620 Yes
33 Descemet Membrane Transplantation
4 53, F, C FECD Phaco/IOL/DMT 6/12 778 Yes 6/12 519 Yes

5 71, F, C FECD Phaco/IOL/DMT 6/15 645 Yes 6/15 437 Yes
6c 66, M, C FECD Phaco/IOL/DMT 6/12 576 No CF 785 Yesd
7 64, F, C FECD Phaco/IOL/DMT 6/12 643 Partial 6/18 467 Yes
M Male, F Female, C Chinese, P Filipino A Arab, FECD fuchs endothelial corneal dystrophy, Phaco phacoemulsification, IOL intraocular lens implantation, DMT descemet
membrane transplantation, CCT central corneal thickness, ROCK rho-associated protein kinase, NA not available (data), CF counting fingers
a
Age in years
b
Outcomes at 6 months
c
Subsequently underwent a successful Descemet membrane endothelial keratoplasty (DMEK) at month 8
d
Developed ROCK inhibitor-associated reticular bullous keratopathy
465
Fig. 33.2 Representative

pre-operative specular
microscopy (Konan
Medical Corp, Hyogo,
Japan) images of a
patient included in this
case series illustrating
severe corneal guttata and
corneal endothelial cell
loss in the central cornea
with relatively preserved
peripheral corneal
endothelium
(Fig. 33.4, Table 33.1). There was a progressive

resolution of central corneal oedema, with full
resolution in five patients and partial resolution
of oedema in one patient. All DMT grafts
remained attached. Post-operative visual acuities
ranged from 6/7.5 to 6/18 in these six patients.
By 6 months, all six patients had successfully
weaned off all topical medications. One patient’s
cornea failed to clear following DMT (Case 6).
This is the same patient who developed the
ROCK inhibitor-associated reticular bullous epi-
thelial oedema, where the drops had to be stopped
prematurely. He subsequently underwent a suc-
cessful DMT removal with Descemet’s mem-
brane endothelial keratoplasty at month 8
following his original DMT (Table 33.1).
Fig. 33.3 Slit-lamp images and corresponding anterior

segment optical coherence tomography (RTVue, Optovue,
Fremont, USA) of Case 6 illustrating reticular epithelial
bullous oedema associated with the use of rho-associated
protein kinase (ROCK) inhibitor
Fig. 33.4 Representative images illustrating the post- thickness as observed on anterior segment optical coher-
operative recovery of a patient who underwent Descemet’s ence tomography (RTVue, Optovue, Fremont, USA).
Membrane Transplantation. (Top row) Slit lamp images BCVA best corrected visual acuity, CCT central corneal
showing progressive resolution of central corneal oedema thickness, POW post-operative week, POM post-operative
over 6 months with improved clarity of the cornea. month
(Bottom row) There is corresponding reduction in corneal
Discussion lation of remaining healthy host CEnCs in the

periphery following the central removal of dis-
Through DMT, we showed that appropriately eased endothelium. It has been shown that a
selected patients with endothelial dysfunction larger population of peripheral CEnCs after
could circumvent the need for allogenic trans- Descemet membrane removal contributes to bet-
plantation, thus avoiding the risks of graft rejec- ter endothelial stability following centripetal
tion and the need for long-term migration and a faster recovery of endothelial
immunosuppressive therapies. Furthermore, a function [32]. We thus propose that a small
significant proportion of donor corneas har- Descemet membrane graft is used (4–5 mm).
vested worldwide are not utilized due to low cor- Given that the size of the Descemet membrane
neal endothelial cell densities [13]. Thus, the graft is small, multiple grafts can also be har-
utilization of acellular tissues allows for the vested from one donor cornea. Indeed, one donor
repurposing of donor corneas that are currently corneal tissue could be potentially used to treat
not suitable for standard corneal multiple patients with endothelial diseases.
transplantations. To ensure the success of DMT, the selection of
In such a regenerative therapeutic approach, patients is also important. We have previously
one factor to take into consideration is the popu- reported that increasing age is associated with a
decline in the migration potential of human cor- sufficient follow-up are required to determine the
neal endothelial cells [30]. For older patients longer term safety and efficacies of these novel
above the age of 50 years, we propose that cor- therapies.
neal endothelial cellular migration should be
enhanced with the application of topical ROCK Take Home Notes
inhibitors (e.g. ripasudil, netarsudil). The role of • Graft orientation of the acellular membrane is
these agents and the longer term outcomes of important and must not overlap with the recip-
DMT requires further evaluation. It is also worth ient DM.
highlighting that in such regenerative therapeutic • DMT allows a larger area of guttata to be
approaches, the CEnCs that migrate centripetally removed.
from the peripheral cornea may be the patient’s • Membrane adhesion is achieved through the
diseased cells (e.g. FECD). Thus, clinical fea- full gas fill.
tures such as the presence of guttata may still • Key benefits are the lack of long-term steroid
develop within the migrated cells in the trans- use and the use of otherwise unusable tissue
planted Descemet membrane graft. Nonetheless, hence increasing the donor pool.
such regenerative approaches would have delayed • Early physiological response can be seen by
the timing at which corneal transplantation is 6 weeks on ASOCT, if there is no response,
needed. rescue DMEK can be performed with good
Compared to other regenerative techniques outcomes.
such as DSO, clinical studies assessing DSO for • More cases are required for a full evaluation
FECDs have reported inconsistent outcomes of this technique.
[24–28]. A recent systematic review of 11 publi-
cations indicated that the results of DSO are
likely to be more predictable if ≤4 mm of the References
patient’s central diseased Descemet membrane is
removed [33]. Other groups have also reported 1. Bonanno JA. Molecular mechanisms underlying the
corneal endothelial pump. Exp Eye Res. 2012;95:2–7.
better outcomes of DSO when topical ROCK https://doi.org/10.1016/j.exer.2011.06.004.
inhibitors are applied [27, 28]. However, ade- 2. Edelhauser HF. The balance between corneal trans-
quately powered randomized controlled studies parency and edema: the proctor lecture. Invest
with sufficient follow-up are still needed to estab- Ophthalmol Vis Sci. 2006;47:1754–67. https://doi.
org/10.1167/iovs.05-1139.
lish the efficacy of DSO as an intervention. 3. Tuft SJ, Coster DJ. The corneal endothelium. Eye.
The presence of a Descemet membrane graft 1990;4(Pt 3):389–424. https://doi.org/10.1038/
in DMT to support CEnC migration appears to eye.1990.53.
reduce the risk of endothelial-to-mesenchymal 4. McCartney MD, Wood TO, McLaughlin BJ. Freeze-
fracture label of functional and dysfunctional human
transition (EMT) compared to CEnC migration corneal endothelium. Curr Eye Res. 1987;6:589–97.
across a denuded corneal stroma in DSO [29, 30]. 5. Mahdy MA, Eid MZ, Mohammed MA, Hafez A,
Furthermore, following primary descemetorhexis Bhatia JR. elationship between endothelial cell
in DSO may predispose the activation of stromal loss and microcoaxial phacoemulsification param-
eters in noncomplicated cataract surgery. Clinical
keratocytes in the posterior cornea to fibroblastic Ophthalmol. 2012;6:503–10. https://doi.org/10.2147/
transformation through the exposure of media- OPTH.S29865.
tors (e.g. TGFß) found in the aqueous humour, 6. Joyce NC, Meklir B, Joyce SJ, Zieske JD. Cell
which may ultimately lead to fibrosis [34]. The cycle protein expression and proliferative status in
human corneal cells. Invest Ophthalmol Vis Sci.
presence of a Descemet membrane graft in DMT 1996;37:645–55.
may act as a barrier between the aqueous humour 7. Joyce NC, Navon SE, Roy S, Zieske JD. Expression
and posterior stromal keratocytes to avoid such of cell cycle-associated proteins in human and rabbit
pro-fibrotic phenotypic transformations. corneal endothelium in situ. Invest Ophthalmol Vis
Sci. 1996;37:1566–75.
Nevertheless, despite these theoretic advantages 8. Murphy C, Alvarado J, Juster R, Maglio M. Prenatal
of DMT over DSO, further clinical trials with and postnatal cellularity of the human corneal endo-
thelium, a quantitative histologic study. Investig Eye Res. 1999;68:341–6. https://doi.org/10.1006/

Ophthalmol Vis Sci. 1984;25:312–22. exer.1998.0611.
9. Edelhauser HF. The resiliency of the corneal endo- 22. Coster DJ, Lowe MT, Keane MC, Williams
thelium to refractive and intraocular surgery. Cornea. KA. Australian corneal graft Registry, C. a compari-
2000;19:263–73. son of lamellar and penetrating keratoplasty outcomes:
10. Australian Corneal Graft Registry, C. The Australian a registry study. Ophthalmology. 2014;121:979–87.
graft registry 2018 report. 2018. <https://dspace. https://doi.org/10.1016/j.ophtha.2013.12.017.
flinders.edu.au/xmlui/bitstream/handle/2328/37917/ 23. Kaufman AR, Nose RM, Pineda R 2nd.
ACGR%202018%20Report.pdf?sequence=3&isAllo Descemetorhexis without endothelial keratoplasty
wed=y>. (DWEK): proposal for nomenclature standardiza-
11. Ang M, Soh Y, Htoon HM, Mehta JS, Tan D. Five- tion. Cornea. 2018;37:e20–1. https://doi.org/10.1097/
year graft survival comparing Descemet stripping ICO.0000000000001528.
automated endothelial keratoplasty and penetrating 24. Borkar DS, Veldman P, Colby KA. Treatment of
keratoplasty. Ophthalmology. 2016;123:1646–52. fuchs endothelial dystrophy by descemet strip-
https://doi.org/10.1016/j.ophtha.2016.04.049. ping without endothelial keratoplasty. Cornea.
12. Woo JH, Ang M, Htoon HM, Tan D. Descemet 2016;35:1267–73. https://doi.org/10.1097/
membrane endothelial keratoplasty versus Descemet ICO.0000000000000915.
stripping automated endothelial keratoplasty 25. Soh YQ, Peh GS, Mehta JS. Evolving therapies for
and penetrating Keratoplasty. Am J Ophthalmol. fuchs' endothelial dystrophy. Regen Med. 2018;13:97–
2019;207:288–303. https://doi.org/10.1016/j. 115. https://doi.org/10.2217/rme-2017-0081.
ajo.2019.06.012. 26. Artaechevarria Artieda J, Wells M, Devasahayam RN,
13. Gain P, et al. Global survey of corneal trans- Moloney G. 5-year outcomes of descemet stripping
plantation and eye banking. JAMA Ophthalmol. only in fuchs dystrophy. Cornea. 2020;38(8):1048.
2016;134:167–73. https://doi.org/10.1001/ https://doi.org/10.1097/ICO.0000000000002270.
jamaophthalmol.2015.4776. 27. Garcerant D, et al. Descemet's stripping without
14. Kumar A, Yun H, Funderburgh ML, Du endothelial keratoplasty. Curr Opin Ophthalmol.
Y. Regenerative therapy for the cornea. Prog Retin 2019;30:275–85. https://doi.org/10.1097/
Eye Res. 2022;87:101011. https://doi.org/10.1016/j. ICU.0000000000000579.
preteyeres.2021.101011. 28. Macsai MS, Shiloach M. Use of topical rho kinase
15. Okumura N, et al. Enhancement on primate corneal inhibitors in the treatment of fuchs dystrophy after
endothelial cell survival in vitro by a ROCK inhibitor. descemet stripping only. Cornea. 2019;38:529–34.
Invest Ophthalmol Vis Sci. 2009;50:3680–7. https:// https://doi.org/10.1097/ICO.0000000000001883.
doi.org/10.1167/iovs.08-2634. 29. Bhogal M, Lwin CN, Seah XY, Peh G, Mehta
16. Okumura N, et al. Involvement of cyclin D and JS. Allogeneic descemet's membrane transplantation
p27 in cell proliferation mediated by ROCK inhibi- enhances corneal endothelial monolayer formation
tors Y-27632 and Y-39983 during corneal endo- and restores functional integrity following descemet's
thelium wound healing. Invest Ophthalmol Vis stripping. Invest Ophthalmol Vis Sci. 2017;58:4249–
Sci. 2014;55:318–29. https://doi.org/10.1167/ 60. https://doi.org/10.1167/iovs.17-22106.
iovs.13-12225. 30. Soh YQ, et al. Predicative factors for corneal endo-
17. Okumura N, et al. The ROCK inhibitor eye drop accel- thelial cell migration. Invest Ophthalmol Vis
erates corneal endothelium wound healing. Invest Sci. 2016;57:338–48. https://doi.org/10.1167/
Ophthalmol Vis Sci. 2013;54:2493–502. https://doi. iovs.15-18300.
org/10.1167/iovs.12-11320. 31. Soh YQ, Mehta JS. Regenerative therapy for fuchs
18. Koizumi N, et al. Rho-associated kinase inhibitor eye endothelial corneal dystrophy. Cornea. 2018;37:523–
drop treatment as a possible medical treatment for 7. https://doi.org/10.1097/ICO.0000000000001518.
Fuchs corneal dystrophy. Cornea. 2013;32:1167–70. 32. Jullienne R, et al. Corneal endothelium self-healing
https://doi.org/10.1097/ICO.0b013e318285475d. mathematical model after inadvertent descemeto-
19. Syed ZA, Rapuano CJ. Rho kinase (ROCK) inhibi- rhexis. J Cataract Refract Surg. 2015;41:2313–8.
tors in the management of corneal endothelial disease. https://doi.org/10.1016/j.jcrs.2015.10.043.
Curr Opin Ophthalmol. 2021;32:268–74. https://doi. 33. Franceschino A, et al. Descemetorhexis without
org/10.1097/ICU.0000000000000748. endothelial keratoplasty in fuchs endothelial corneal
20. Ong HS, Ang M, Mehta J. Evolution of therapies dystrophy: a systematic review and meta-analysis.
for the corneal endothelium: past, present and future Cornea. 2022;41:815–25. https://doi.org/10.1097/
approaches. Br J Ophthalmol. 2021;105:454–67. ICO.0000000000002855.
https://doi.org/10.1136/bjophthalmol-2020-316149. 34. Chen J, et al. Descemet's membrane supports
21. Wollensak G, Green WR. Analysis of sex-mismatched corneal endothelial Cell regeneration in rabbits.
human corneal transplants by fluorescence in Sci Rep. 2017;7:6983. https://doi.org/10.1038/
situ hybridization of the sex-chromosomes. Exp s41598-017-07557-2.
Femtosecond Laser-Assisted Deep
Lamellar Endothelial Keratoplasty
34
Jorge L. Alió del Barrio, Verónica Vargas,
and Bruce D. Allan
Key Points recent advances in femtosecond laser technol-

• Endothelial keratoplasty is not able to restore ogy and donor corneal tissue preparation.
the corneal transparency if well-established
stromal scars are already present.
• However, if the corneal opacity is limited Introduction
exclusively to the posterior cornea, the
removal of the deeper stromal layers, together Descemet’s Stripping Automated Endothelial
with the DM and endothelium, could be a via- Keratoplasty (DSAEK) and Descemet’s
ble alternative to PK, avoiding all well-known Membrane Endothelial Keratoplasty (DMEK)
drawbacks of full-thickness corneal have become the gold standard treatment for cor-
transplantation. neal endothelial diseases [1, 2]. In DMEK, the
• Classical deep lamellar endothelial kerato- recipient dysfunctional Descemet membrane
plasty (DLEK) is a challenging and time- (DM) and endothelial layers are excised and
consuming surgery as it was initially replaced by a healthy DM-endothelium from a
described. donor cornea (same for DSAEK, but with the
• In the current chapter, we are showing a sim- exception that the donor lenticule also contains a
plified DLEK technique that integrates the thin layer of stroma), restoring corneal transpar-
ency by a resolution of the pre-existing corneal
oedema.
Posterior Lamellar Keratoplasty (PLK) was
tains supplementary material available at https://doi. originally described by Melles in 1998 [3]. The
org/10.1007/978-3-031-32408-6_34. surgery was later modified by Terry in 2001 and
was termed Deep Lamellar Endothelial
Keratoplasty (DLEK) [4–6]. In DLEK, the failed
J. L. A. del Barrio (*)
Vissum Miranza, Miguel Hernández University, endothelium, DM and a thin lamella of posterior
Alicante, Spain corneal stroma were replaced by a healthy disc of
V. Vargas donor corneal tissue which was positioned in the
Cornea, Cataract and Refractive Surgery Unit, reciprocal host dissection. Manual host lamellar
Vissum (Miranza Group), Alicante, Spain dissection in DLEK was time-consuming, and
B. D. Allan the posterior side cut, in particular, was techni-
Cornea and External Disease Service, Moorfields Eye cally challenging. Although femtosecond laser
Hospital, London, UK assisted tissue preparation was utilised by a num-
ber of investigators to simplify corneal tissue dis-
https://doi.org/10.1007/978-3-031-32408-6_34
section [7, 8], DLEK was largely abandoned in for treating corneal endothelial diseases associ-
favour of Descemet’s stripping techniques which ated with posterior stromal opacities [10, 11].
do not require host stromal dissection (DSEK,
DSAEK and DMEK), and it is now a largely for-
gotten endothelial keratoplasty modality. Indications
DMEK and other modalities of endothelial
keratoplasty are not able to restore transparency The main indication for Femto-DLEK is a visual
if well-established stromal scars are already pres- impairment resulting from pathologies involving
ent. Penetrating keratoplasty (PK) is the normal exclusively the deeper posterior corneal stroma
elective procedure for such cases. However, if and DM, with or without endothelial failure and
corneal opacity is limited exclusively to the pos- corneal oedema (Fig. 34.1). Obviously, for this
terior cornea, the removal of the deeper stromal technique to be successful, it is critical to have a
layers together with the DM and endothelium, as healthy epithelium and anterior/mid stroma.
in DLEK, would be a viable alternative to PK, Alternative indications are as follows:
avoiding all well-known drawbacks of full-
thickness corneal transplantation (long visual • Endothelial failure or retained deep scarring
rehabilitation, suture-related complications, sig- after deep anterior lamellar keratoplasty
nificant residual ametropia, risk of rejection and (DALK) (Fig. 34.2).
glaucoma) [9]. For such cases, with posterior • Revision of a failed DSAEK in cases of pseu-
stromal opacity but a normal anterior stroma, we dophakic bullous keratopathy with an anterior
have developed a simplified DLEK technique chamber intraocular lens (AC-IOL). If a pseu-
that integrates the recent advances in femtosec- dophakic AC-IOL is properly placed within
ond laser technology and donor corneal tissue the AC, Femto-DLEK can be used to enhance
preparation [10, 11]. This simplified form of the anterior chamber depth and so increase the
DLEK (Femto-DLEK) combined with conven- distance between the IOL and the new donor
tional DSAEK or DMEK donor preparation and tissue, thereby reducing the risk of repeat fail-
insertion techniques has been safe and effective ure due to endothelial touch (Fig. 34.3c).
Fig. 34.1 AS-OCT image of a pseudophakic bullous keratopathy associated with a severe deep stromal scar without
involving other corneal layers. We can measure the precise depth of the scar (yellow arrows)
34 Femtosecond Laser-Assisted Deep Lamellar Endothelial Keratoplasty 473
a b
Fig. 34.2 Femto-DLEK (b) for retained deep stromal opacity after DALK (a)
firmed with AS-OCT prior to the EK surgery

Preoperative Considerations (Fig. 34.4).
Then, under local anesthesia with sedation,
Before surgery, the extension and depth of the the IntraLase iFS femtosecond laser (AMO Inc.,
deep corneal opacity should be identified and Irvine, CA) is used to create an intersecting pos-
measured with anterior segment optical coher- terior side cut of 7.0–8.5 mm of diameter (depths
ence tomography (AS-OCT) (Fig. 34.1). should be determined by AS-OCT depending on
the exact depth of the lamellar dissected plane;
femtosecond laser energy settings as previously
Surgical Technique described) [10]. Subsequently, the dissected pos-
terior recipient disc is removed from the anterior
After mapping the extension and depth of the chamber through a 3.2 mm limbal incision placed
posterior opacification, an overlying manual temporally (Fig. 34.5). A 7.0–8.5 mm DSAEK or
lamellar dissection plane is created 1 week prior DMEK graft is then inserted into the AC and
to the endothelial keratoplasty (EK) surgery at attached to the recipient stroma using a standard
approximately 75–85% depth through a 5.0 mm DSAEK or DMEK technique (following each
superior scleral incision, using a combination of surgeon’s preference) (Figs. 34.3a, b).
blunt dissection with the Morlet lamellar dissec- In cases of failure or scarring of the recipient
tor (Duckworth & Kent, England) and sharp dis- bed after DALK, the posterior side cut is per-
section with a Crescent knife (Alcon, Fort Worth, formed without preliminary lamellar dissection,
TX), assisted by an anterior chamber air fill to as the deep lamellar plane already exists from the
help judge dissection depth using Melles’ air original lamellar surgery, and the unwanted pos-
reflection technique. The accuracy of the manu- terior host tissue can simply be peeled away after
ally dissected pocket location should be con- the posterior side-cut (Video 34.1).
Fig. 34.3 (a) AS-OCT

image of Femto
a
DL-DSAEK. Observe
how the DSAEK
lenticule fills the space
left after the recipient’s
posterior cornea
removal; (b) AS-OCT
image of Femto
DL-DMEK. Observe
how the DMEK
lenticule does not fill the b
space left after the
recipient’s posterior
cornea removal,
remaining a step in
between the central and
peripheral posterior
corneal surfaces; (c)
Scheimpflug image of a
Femto DL-DSAEK after
previous failed DSAEK
providing enhanced
clearance between the
donor endothelium and c
the anterior chamber
lens in a case of
pseudophakic bullous
keratopathy
Fig. 34.4 AS-OCT image in which the manually dissected pocket location is confirmed (yellow arrows)
tory step and restricting femtosecond laser assis-

tance to the posterior side cut at the time of
DLEK surgery. Our early experience was that
posterior stromal femtosecond dissection was
incomplete, leaving problematic tissue bridges
that were difficult to dissect manually after the
posterior side cut. The posterior side cut removes
the anchoring counter traction required for easy
lamellar dissection, so these tissue bridges usu-
ally have to be divided using sharp dissection and
Fig. 34.5 With a reverse sinsky hook, the diseased
endothelium-descemet membrane-posterior stroma com- counter traction with vitrectomy forceps. This is
plex is stripped out from the cornea through a 3.2 mm a challenging manoeuvre that may compromise
limbal incision to host interface smoothness. The reason for this
is due to the fact that deep stromal femtosecond
laser dissections leave a relatively rougher and
Postoperative Management more irregular interface compared with the more
superficial ones, probably in relation to the scat-
Postoperative care and patient positioning are tering of the energy and the anatomical differ-
exactly as for standard DSAEK or DMEK ences between the anterior and the less compacted
surgery. posterior stromal collagen fibers [13–15].
In Femto-DLEK, after removing the recipient
diseased lamellae of the posterior cornea, there
Complications will remain a step in between the central and
peripheral posterior corneal surfaces, space that
Probably, the most critical step in Femto-DLEK will be filled with the endothelial donor tissue
is the manual dissection of the lamellar plane just (Fig. 34.3a, b). Because of this, correct centration
over the deep stromal opacification (Fig. 34.4). It of the DSAEK or DMEK graft into the recipient
is not a technically demanding step and its per- bed becomes more important than for a standard
formance is as described for manual DALK by EK case, as graft decentration may increase the
Melles [12]. However, an unsatisfactory dissec- risk for graft detachment.
tion plane, either because of a too superficial
plane or an incorrect one that leaves a part of the
opacity on the anterior cornea in relation to the Outcomes
dissection, will force us to either try a second
more accurate lamellar cut (starting from a differ- Existing published evidence (n = 8; six cases
ent pocket) or to abort this technique and to con- associated with a DSAEK graft and two associ-
vert into a PK. If available, intraoperative corneal ated with a DMEK graft) is insufficient to deter-
OCT imaging systems could simplify this step by mine the Femto-DLEK specific rate of other
allowing an easy “live” assessment of the correct postoperative complications such as rejection,
dissection depth. Alternatively, the femtosecond primary failure or graft detachment. Probably,
laser can be used in single pass mode for the rate of such complications in Femto-DLEK is
recipient posterior corneal lamellar dissection by similar to the observed for standard DSAEK and
creating a posterior intrastromal lamellar cut of DMEK surgeries provide the EK graft is well
7.5–9.0 mm diameter at 400–450 μm depth centred within the posterior stromal recipient
(depending on the preoperative corneal pachym- bed. In our series, two eyes required re-bubbling
etry as measured by AS-OCT). However, we rec- of the graft for partial detachment 1 week after
ommend manual dissection of this posterior surgery, and no cases of graft rejection, primary
lamellar plane one week in advance as a prepara- or secondary failure or glaucoma have been
observed with more than 1 year follow-up in all of AC IOL exchange include intraoperative
cases. Regarding efficacy, aside from two cases haemorrhage, iris trauma, cyclodialysis, pupil
with ocular comorbidity affecting the visual out- distortion, chronic cystoid macular oedema and
come, all published eyes achieved a spectacle- late failure of fixation with posterior IOL disloca-
corrected distance vision of 6/9 or better. tion [17, 18]. Femto-DL-DMEK deepens the
The use of a DMEK donor graft after Femto- anterior chamber and may reduce the risk of graft
DLEK host dissection (Femto DL-DMEK) may failure with a stable AC IOL left in situ
enhance the outcomes by adding the advantages (Fig. 34.3c).
of DMEK over DSAEK (better visual potential,
lower rejection rate and faster visual rehabilita- Take Home Notes
tion), although a larger sample is still necessary • Femto DL-DMEK or Femto DL-DSAEK is a
to confirm this statement (Fig. 34.3b) [11, 16]. successful and alternative treatment option for
The potential refractive impact of removing a the treatment of posterior corneal stromal
thick lamella of posterior stroma, particularly problems which are causing impaired vision.
when a DMEK donor is used, leaving a counter- • As far as the anterior and mid-stroma are not
sunk posterior corneal profile (Fig. 34.3a, b) affected, opacifications of the posterior cornea
remains uncertain, and further studies will have with or without endothelial dysfunction can
to explore this. However, considering published benefit from endothelial keratoplasty advan-
evidence and excluding those cases of Femto- tages through a Femto-DLEK procedure.
DLEK after DALK, no significant impact on the • We recommend a manual dissection of the
refractive sphere or cylinder has been observed: posterior lamellar plane 1 week prior to the
mean spherical equivalent of +2.85D and mean transplant as a preparatory step and restricting
refractive cylinder of −0.8D (n = 4) [10, 11]. femtosecond laser assistance to the posterior
side cut at the time of DLEK surgery.
Conclusions Conflict of Interest None of the authors have any con-

flict of interest to disclose.
DLEK was a technically demanding and chal-
lenging surgical technique, but the addition of a
femtosecond laser has made the most difficult References
step – the posterior side cut in host dissection –
much easier. Femto DL-DMEK (Fig. 34.3b) or 1. Gorovoy MS. Descemet-stripping automated endo-
Femto DL-DSAEK (Fig. 34.3a) is a successful thelial keratoplasty. Cornea. 2006;25:886–9.
2. Melles GR, Ong TS, Ververs B, et al. Descemet mem-
and alternative treatment option for treating pos- brane endothelial Keratoplasty (DMEK). Cornea.
terior corneal stromal problems which are caus- 2006;25:987–90.
ing impaired vision. As far as the anterior and 3. Melles GR, Eggink FA, Lander F, et al. A surgical
mid stroma are not affected, opacifications of the technique for posterior lamellar keratoplasty. Cornea.
1998;17:618–26.
posterior cornea with or without endothelial dys- 4. Terry MA, Ousley PJ. Endothelial replacement with-
function can benefit from endothelial kerato- out surface corneal incisions or sutures: topography of
plasty advantages through a Femto-DLEK the deep lamellar endothelial keratoplasty procedure.
procedure, avoiding all well-known drawbacks of Cornea. 2001;20:14–8.
5. Terry MA. Deep lamellar endothelial keratoplasty
PK [9]. The question of whether to exchange AC (DLEK): pursuing the ideal goals of endothelial
IOLs for suture-fixated posterior chamber IOLs replacement. Eye (Lond). 2003;17:982–8.
in cases of pseudophakic bullous keratopathy is 6. Terry MA, Ousley PJ. Deep lamellar endothelial kera-
unresolved. The goal of exchange for a scleral or toplasty: visual acuity, astigmatism, and endothelial
survival in a large prospective series. Ophthalmology.
iris-fixated posterior chamber IOL, where the 2005;112:1541–8.
existing AC IOL is stable, is to reduce the risk of 7. Terry MA, Ousley PJ, Will B. A practical femtosec-
optic/endothelial touch. Potential complications ond laser procedure for DLEK endothelial transplan-
tation: cadaver eye histology and topography. Cornea. with optomised FS laser settings. Br J Ophthalmol.
2005;24:453–9. 2012;96:122–7.
8. Lee DH, Chung TY, Chung ES, et al. Case report: 14. Vetter JM, Butsch C, Faust M, et al. Irregularity of the
femtosecond laser-assisted small incision deep lamel- posterior corneal surface after curved interface femto-
lar endothelial keratoplasty. Korean J Ophthalmol. second laser-assisted versus microkeratome-assisted
2008;22:43–8. descemet stripping automated endothelial kerato-
9. Maier AKB, Gundlach E, Gonnermann J, et al. Fellow plasty. Cornea. 2012;32:118–24.
eye comparison of descemet membrane endothelial 15. Phillips PM, Phillips LJ, Saad HA, et al. “Ultrathin”
keratoplasty and penetrating keratoplasty. Cornea. DSAEK tissue prepared with a low-pulse energy,
2013;32:1344–8. high-frequency femtosecond laser. Cornea.
10. Alió del Barrio JL, Ziaei M, Bhogal M, Allan 2013;32:81–6.
BD. Femtosecond laser-assisted deep lamellar endo- 16. Goldich Y, Showrail M, Avni-Zauberman N, et al.
thelial keratoplasty: a new approach to a forgotten Contralateral eye comparison of descemet membrane
technique. Cornea. 2015;34:1369–74. endothelial keratoplasty and descemet stripping auto-
11. Alió Del Barrio JL, Vargas V. Femtosecond laser- mated endothelial keratoplasty. Am J Opthalmol.
assisted deep lamellar descemet membrane endo- 2015;159:155–9.
thelial keratoplasty for the treatment of endothelial 17. Hayashi K, Hirata A, Hayashi H. Possible predis-
dysfunction associated with posterior stromal scar- posing factors for in-the-bag and out-of-the-bag
ring. Cornea. 2019;38(3):388–91. intraocular lens dislocation and outcomes of intra-
12. Melles GR, Lander F, Rietveld FJ, Remeijer L, ocular lens exchange surgery. Ophthalmology.
Beekhuis WH, Binder PS. A new surgical technique 2007;114:969–75.
for deep stromal, anterior lamellar keratoplasty. Br J 18. Gonnermann J, Klamann MK, Maier AK, et al. Visual
Ophthalmol. 1999;83(3):327–33. outcome and complications after posterior iris-claw
13. Rousseau A, Bensalem A, Garnier V, et al. Interface aphakic intraocular lens implantation. J Cataract
quality of endothelial keratoplasty buttons obtained Refract Surg. 2012;38:2139–43.
Femtosecond Descemet
Membrane Endothelial
35
Keratoplasty
Nir Sorkin, David S. Rootman,

and Michael Mimouni
Key Points Rationale

• Despite advances in postoperative DMEK
graft detachment, subsequent cell loss is still a Descemetorhexis and removal of the recipient’s
matter of concern. Descemet membrane (DM) from the transplant
• In femtosecond-assisted DMEK, a femtosec- bed is an important step in Descemet membrane
ond laser is used to perform the Endothelial Keratoplasty (DMEK). Proper
descemetorhexis. Descemet removal facilitates DMEK graft attach-
• Femtosecond-assisted DMEK technical ment, thereby reducing rates of graft detachment –
parameters and adjustments in technique are the most common complication following DMEK
reviewed. surgery. Graft detachment can affect surgical out-
• An outline and comparison of femtosecond comes and may require a rebubbling procedure
assisted DMEK compared to manual DMEK (repeat injection of air into the anterior chamber).
outcomes is provided. Graft detachment requiring rebubbling occurs in
12.8% of DMEK cases [1]. Any Descemet tags or
islands remaining in the transplant bed can pro-
duce spatial interference, which could prevent
Supplementary Information The online version con- attachment of the delicate DMEK graft [2–4]. On
tains supplementary material available at https://doi. the other hand, excess removal of the recipient’s
org/10.1007/978-3-031-32408-6_35. DM peripheral to the location of the graft is also
undesirable since this area may contain viable
N. Sorkin endothelium (especially in eyes with Fuchs’
Department of Ophthalmology, Tel Aviv Medical endothelial dystrophy where the peripheral DM
Center and Sackler Faculty of Medicine, Tel Aviv may contain a good number of viable endothelial
University, Tel Aviv, Israel cells) which will be unnecessarily removed and
D. S. Rootman won’t be covered later by the graft. This would
Department of Ophthalmology and Vision Sciences, necessitate excess migration of endothelial cells
University of Toronto, Toronto, ON, Canada
e-mail: [email protected] of the graft to repopulate the denuded area,
thereby reducing the effective endothelial cell
M. Mimouni (*)
Department of Ophthalmology, Rambam Health Care density (ECD) [5, 6].
Campus, Bruce and Ruth Rappaport Faculty of In femtosecond DMEK, a femtosecond laser
Medicine, Technion-Israel Institute of Technology, is used to outline the descemetorhexis [7].
Haifa, Israel Femtosecond descemetorhexis is accurate in size
https://doi.org/10.1007/978-3-031-32408-6_35
480 N. Sorkin et al.
and shape [5]. Also, in contrast to the mechanical can be measured using an ultrasound probe at 8
scoring of DM, which inevitably affects DM points along the planned incision location.
located peripheral to the descemetorhexis inci- Pachymetry measurements can be confirmed
sion, the femtosecond laser performs a non- using either optical coherence tomography
mechanical incision that does not disturb DM (OCT) or Scheimpflug tomography with the con-
peripheral to the incision site [8]. Further, the sideration that corneal opacities may produce
femtosecond incision creates a physical barrier measurement artifacts in optical imaging. The
that prevents over stripping of DM beyond the use of femtosecond platforms incorporating
planned stripping diameter. intraoperative OCT may obviate the need for pre-
operative pachymetry measurements.
The iFS IntraLase (J&J Vision, Santa Ana,
Surgical Technique California, USA) is used with the following cut-
ting parameters: an energy of 2.29 μJ, a spot sep-
Since femtosecond descemetorhexis is precise aration of 3 μm, a layer separation of 2 μm and a
and does not affect DM beyond the incision site, sidecut angle of 90°.
its diameter can be same-sized (or just slightly The femtosecond laser creates a vertical cylin-
oversized) compared to DMEK graft itself [6]. If drical cut whose depth is set from 100 μm below
performed under a failed penetrating keratoplasty the thickest measured pachymetry (into the ante-
(PKP) graft, femtosecond descemetorhexis diam- rior chamber) to 100 μm above the thinnest mea-
eter should be at least 0.25 mm smaller than the sured pachymetry (into the stroma) (Fig. 35.1). In
PKP graft to avoid the irregularity and opacities OCT-guided femtosecond platforms, incision
around the PKP graft-host interface [9]. In order height and location can be determined according
to determine incision depth settings, pachymetry to intraoperative OCT imaging. Next, DM is
Epithelium
Cornea
Endothelium
Anterior Chamber
Epithelium
(Corneal thickness) - 100µm
Endothelium
(Cornea thickness) + 100µm
Anterior Chamber
Fig. 35.1 Femtosecond Descemet membrane endothe- N, Boutin T, Showail M, Borovik A, Alobthani M, Chan
lial keratoplasty incisions: a cylindrical cut at a depth CC, Rootman DS (2017) Comparison of femtosecond
extending from 100 μm below the thickest measured laser-enabled descemetorhexis and manual descemeto-
pachymetry to 100 μm above the thinnest measured rhexis in descemet membrane endothelial keratoplasty.
pachymetry. [Reproduced from: Einan-Lifshitz A, Sorkin Cornea 36:767–770]
35 Femtosecond Descemet Membrane Endothelial Keratoplasty 481
stripped using a blunt instrument such as a reverse a

Sinskey hook. Stripping of DM should not begin
at the location of the femtosecond incision itself
since the incision extends into the stroma, and
therefore, initial separation of DM from the
stroma at this location may be difficult due to
increased stromal mobility and lack of stromal
resistance around the incision. Rather, stripping
should be initiated slightly more central to
achieve an initial DM flap which can then be nor-
mally peeled (Video 35.1). The rest of the DMEK b
procedure is identical to standard DMEK.
Safety
An ex-vivo study evaluating endothelial vitality

and stromal integrity following femtosecond des-
cemetorhexis found a minimal impact of both the
laser incision and DM peeling on the vitality of c
surrounding endothelial cells. A dense amount of
vital endothelial cells was seen even very close to
the incision edge both after the laser incision and
after DM peeling. Phase contrast microscopy and
scanning electron microscopy evaluation showed
that the laser produced precise, clear-cut edges,
leaving no stromal tissue bridges with minimal
stromal damage (Fig. 35.2) [8].
The use of a femtosecond laser to create cor-
neal incisions carries the risk of an incomplete Fig. 35.2 A magnified view of a femtosecond desceme-
cut. In the case of DM incisions, an incomplete trohexis incision after Descemet stripping. (a) Phase con-
incision may lead to radial tears extending to trast imaging showing the exposed corneal stroma
following descemet stripping (top part of the image), the
the peripheral DM in a manner similar to an continuous laser incision and the area of untouched des-
incomplete femtosecond capsulorhexis in fem- cemet peripheral to the incision (bottom part of the
tosecond laser-assisted cataract surgery image). (b) Propidium iodide staining shows a minimal
(FLACS), causing an anterior capsular tear. In amount of devitalized stromal cells (red dots) near the
incision and within the exposed stroma. (c) Hoechst
contrast to FLACS, where an anterior capsular 33342 staining shows a high density of vital endothelial
tear may have serious surgical implications, a cells even close to the edge of the cut. (Reproduced from:
radial tear of the recipient’s DM, although Feldhaus L, Dirisamer M, Ohlmann A, Luft N, Kassumeh
undesirable, should not be detrimental to the S, Shajari M, Priglinger SG, Mayer WJ (2022)
Femtosecond laser-assisted descemetorhexis for
course of the DMEK procedure. There are no Descemet membrane endothelial keratoplasty: cell-based
reported cases of an incomplete femtosecond and tissue-based ex vivo analysis of precision and safety.
descemetorhexis in femtosecond DMEK publi- J Cataract Refract Surg 48:89–94)
cations where the cylindrical incision height
extended 100 μm into the stroma [6, 7, 10, 11]. Their results showed that while incisions
Pilger et al. looked at reducing the stromal extending 100 μm into the posterior stroma
extension depth of the femtosecond descemeto- yielded a completely separated descemeto-
rhexis. They performed femtosecond desce- rhexis cut in all cases, incisions extending only
metorhexis incisions with varying stromal 75 or 60 μm into the stroma were associated
extension depths of 100 μm, 75 μm and 60 μm. with incomplete cut edges, DM bridges over
482 N. Sorkin et al.
the incision and the occurrence of small radial ment and rebubble rates, as well as reduced endo-
tears when the DM was removed [5]. This was thelial cell loss (ECL) compared with standard
attributed in part to thickness changes through- DMEK [5, 6, 10]. Rates of ECL following femto-
out the cornea and corneal positioning in rela- second DMEK were reduced by 5.8–13.6% com-
tion to the laser, as well as to the presence of pared with standard DMEK over 5 years of
Descemet folds which reduce tissue clarity and follow-up [12]. The reduction in ECL following
deform the posterior surface of the cornea. femtosecond DMEK may extend graft survival
Therefore, we would recommend keeping the by several years.
stromal incision depth at 100 μm. DMEK performed to replenish a failed PKP
Performing a femtosecond incision that graft is associated with high rates of postopera-
extends 100 μm into the posterior stroma raises tive graft detachment, ranging between 26 and
the question of its influence on corneal biome- 100% [13–15]. We found that femtosecond
chanics, corneal curvature and the refractive sta- DMEK performed under a failed PKP graft has
bility of the cornea. In a study evaluating 3-year low detachment rates, [11] significantly lower
outcomes of femtosecond DMEK, our group than standard DMEK performed in this setting
found postoperative spherical equivalent to be [9]. Rates of ECL in this setting did not differ
stable, decreasing by just 0.28 ± 0.54 D between femtosecond and standard DMEK. The
(range − 0.75 to + 0.75 D) throughout follow-up. advantages of femtosecond DMEK relating to
There are no reported cases of corneal perfo- ECL seem not to be realized in eyes with a failed
ration following femtosecond descemetorhexis. PKP. This may be because the descemetorhexis
area is limited by the PKP graft size and there-
fore, there may not be a big difference in the
Efficacy denuded stromal area between femtosecond and
manual descemetorhexis (Fig. 35.3). Additionally,
In femtosecond DMEK, the accurate and com- in the setting of graft failure, there are few
plete removal of the recipient’s DM can improve remaining viable recipient endothelial cells, and
attachment of the graft and preserve more viable therefore, preserving more of the recipient’s DM
recipient endothelial cells. Pilger et al. have
shown that the diameter of a descemetorhexis
performed using a femtosecond laser varies in
size by just 1% compared with 7–8% size vari-
ability of a standard descemetorhexis (p = 0.001).
They also found that the accuracy of the femto-
second incisions was associated with a smaller
area of denuded stroma around the area of
planned descemetorhexis and graft location. For
example, for a planned 8.0 mm descemetorhexis,
they found the surrounding denuded area to be
2.5 mm2 in femtosecond DMEK and 11.6 mm2 in
standard DMEK – a difference of 9.1 mm2
(p < 0.001) [5]. Considering that an 8.0 mm
DMEK graft has a total area of 50.3 mm2, the
addition of 9.1 mm2 of surrounding denuded
stroma in standard DMEK equals roughly 18% Fig. 35.3 A postoperative slit-lamp photograph of a
patient with a femtosecond Descemet membrane endothe-
of the graft area. The denuded area needs to be lial keratoplasty (DMEK) graft performed under a failed
repopulated by endothelial cells migrating off the penetrating keratoplasty (PKP). The larger circle repre-
graft, thereby causing a more significant decrease sents the PKP graft edge, and the smaller circle represents
in ECD in manual DMEK. the femtosecond descemetorhexis and DMEK graft diam-
eters which were the same sized in this case. PKP func-
In Fuchs’ dystrophy patients, femtosecond tionality and clarity were restored following femtosecond
DMEK has been found to have reduced detach- DMEK
35 Femtosecond Descemet Membrane Endothelial Keratoplasty 483
in femtosecond DMEK does not substantially in descemet membrane endothelial keratoplasty.

BMJ Open Ophthalmol. 2018;3:e000148. https://doi.
increase the total postoperative endothelial cell org/10.1136/BMJOPHTH-2018-000148.
count. 6. Sorkin N, Mednick Z, Einan-Lifshitz A, Trinh T,
Santaella G, Telli A, Chan CC, Rootman DS. Three-
year outcome comparison between femtosecond laser-
assisted and manual Descemet membrane endothelial
Conclusion Keratoplasty. Cornea. 2019;38:812–6.
7. Pilger D, von Sonnleithner C, Bertelmann E, Joussen
Femtosecond DMEK is safe and effective, with AM, Torun N. Femtosecond laser-assisted desce-
the advantages of better graft attachment and metorhexis: a novel technique in descemet membrane
endothelial Keratoplasty. Cornea. 2016;35:1274–8.
improved endothelial cell viability. 8. Feldhaus L, Dirisamer M, Ohlmann A, Luft N,
Kassumeh S, Shajari M, Priglinger SG, Mayer
Take Home Notes WJ. Femtosecond laser-assisted descemeto-
• Femtosecond-assisted DMEK may be safely rhexis for Descemet membrane endothelial kera-
toplasty: cell-
based and tissue-based ex vivo
and efficiently performed in routine DMEK analysis of precision and safety. J Cataract Refract
cases. Surg. 2022;48:89–94.
• Femtosecond-assisted DMEK has advantages 9. Sorkin N, Mimouni M, Santaella G, Trinh T,
over manual DMEK in patients with a prior Cohen E, Einan-Lifshitz A, Chan CC, Rootman
DS. Comparison of manual and femtosecond laser–
failed penetrating keratoplasty. assisted descemet membrane endothelial keratoplasty
• Using the provided femtosecond parameters for failed penetrating keratoplasty. Am J Ophthalmol.
and adjustments in technique transition to 2020;214:1–8.
femtosecond-assisted DMEK is feasible. 10. Einan-Lifshitz A, Sorkin N, Boutin T, Showail
M, Borovik A, Alobthani M, Chan CC, Rootman
DS. Comparison of femtosecond laser-enabled des-
cemetorhexis and manual descemetorhexis in des-
cemet membrane endothelial keratoplasty. Cornea.
References 2017;36:767–70.
11. Sorkin N, Trinh T, Einan-Lifshitz A, Mednick
1. Deng SX, Lee WB, Hammersmith KM, Kuo AN, Li Z, Santaella G, Telli A, Belkin A, Chan CC,
JY, Shen JF, Weikert MP, Shtein RM. Descemet mem- Rootman DS. Outcomes of femtosecond laser-
brane endothelial keratoplasty: safety and outcomes. assisted descemet membrane endothelial kerato-
Ophthalmology. 2018;125:295–310. plasty for failed penetrating keratoplasty. Can J
2. Müller TM, Verdijk RM, Lavy I, Bruinsma M, Parker Ophthalmol. 2019;54:741. https://doi.org/10.1016/j.
J, Binder PS, Melles GRJ. Histopathologic features jcjo.2019.04.003.
of descemet membrane endothelial keratoplasty graft 12. Sorkin N, Gouvea L, Din N, Mimouni M, Alshaker S,
remnants, folds, and detachments. Ophthalmology. Weill Y, Gendler S, Slomovic AR, Chan CC, Rootman
2016;123:2489–97. DS. Five-year safety and efficacy of femtosecond
3. Baydoun L, Ham L, Borderie V, Dapena I, Hou J, laser—assisted descemet membrane endothelial
Frank LE, Oellerich S, Melles GRJ. Endothelial Keratoplasty. Cornea. 2022;42:145–9.
survival after descemet membrane endothe- 13. Einan-Lifshitz A, Belkin A, Sorkin N, Mednick
lial keratoplasty: effect of surgical indication Z, Boutin T, Gill I, Karimi M, Chan CC, Rootman
and graft adherence status. JAMA Ophthalmol. DS. Descemet membrane endothelial Keratoplasty
2015;133:1277–85. after penetrating Keratoplasty: features for success.
4. Dapena I, Ham L, Moutsouris K, Melles Cornea. 2018;37:1093–7.
GRJ. Incidence of recipient Descemet membrane 14. Pasari A, Price MO, Feng MT, Price FW. Descemet
remnants at the donor-to-stromal interface after membrane endothelial Keratoplasty for failed pen-
descemetorhexis in endothelial keratoplasty. Br J etrating keratoplasty. Cornea. 2019;38:151–6.
Ophthalmol. 2010;94:1689–90. 15. Kemer ÖE, Karaca EE, Oellerich S, Melles
5. Pilger D, Von Sonnleithner C, Bertelmann E, Maier G. Evolving techniques and indications of Descemet
AKB, Joussen AM, Torun N. Exploring the preci- membrane endothelial Keratoplasty. Turk J
sion of femtosecond laser-assisted descemetorhexis Ophthalmol. 2021;51:381–92.
Cultured Cells for Corneal
Endothelial Therapy
36
M. P. De Miguel, M. Cadenas Martín, A. Moratilla,
and F. Arnalich-Montiel
Key Points content. The corneal endothelium, the inner layer,

• Corneal endothelial therapeutics has been is in charge of maintaining the cornea in a rela-
transformed by lamellar endothelial tively dehydrated state and therefore transparent.
transplants. The endothelial cell layer failure leads to corneal
• Recent developments in endothelial cell cul- swelling, loss of transparency, and blindness. In
ture techniques make it possible to expand the past, penetrating keratoplasty (PK) had been
ex vivo the corneal endothelial cells. the gold standard surgical treatment of corneal
• Expanded cells can be delivered subsequently diseases for any layer, including primary endo-
by direct injection into the anterior chamber or thelial diseases [1]. Central endothelial cell den-
in sheet constructs made up of different sity (ECD, expressed in cells per mm2) decreases
materials. at an average rate of about 0.6% per year in nor-
• Recent advances have been achieved in differ- mal corneas throughout adult life [2]. In a normal
entiation protocols from extraocular cells individual, this decline in endothelial cells (EC)
capable of differentiating into corneal endo- does not impair corneal transparency, even in
thelial cells such as embryonic stem cells and centenarians, and only if the density falls below
adipose-derived mesenchymal stem cells. the threshold of 300–500 cells per mm2, irrevers-
ible corneal edema can lead to blindness [3]. This
event can occur following intraocular surgeries,
I ntroduction to Corneal Endothelial traumas, or dystrophies. In fact, blindness due to
Transplant corneal edema is the indication of corneal graft-
ing of one in every three recipients.
The cornea is a five-layered tissue that provides Human corneal endothelium is held in a non-
two-thirds of the total refractive power of the eye, replicative state within the eye [4]. It has been a
and it is the first barrier protecting the intraocular common belief that in vivo, corneal endothelium
has limited wound-healing capacity, mainly by
using residual EC which, by enlargement and
M. P. De Miguel (*) · M. Cadenas Martín
A. Moratilla migration, covers the space left by the lost cells
Cell Engineering Laboratory, La Paz University without division [5]. Joyce [6] demonstrated that
Hospital Health Research Institute, IdiPAZ, hCECs are arrested in the G1-phase of the cell
Madrid, Spain cycle in vivo. Mitotic inhibition has been sug-
F. Arnalich-Montiel gested to be due to contact-dependent inhibition
IRYCIS, Ophthalmology Department, Ramón y Cajal and the transforming growth factor beta (TGF-ß)
University Hospital, Madrid, Spain
https://doi.org/10.1007/978-3-031-32408-6_36
486 M. P. De Miguel et al.
found within the aqueous humor [4]. However, a a lower immune rejection rate. It is the gold
series of clinical observations suggest the ability standard in the treatment of endothelial dis-
of endothelial regeneration in vivo from the eases, although it has not been adopted every-
human corneal periphery after implanting free where yet, due to the higher surgical skills
floating Descemet membrane in the anterior needed. In settings with the scarcity of donor
chamber or in the newly described technique tissue, this technique has evolved to hemi-
known as Descemet stripping only in selected DMEK [18] or quarter-DMEK [19], allowing
cases [3, 7, 8]. one donor to provide tissue for several recipi-
Currently, the only effective and proven way ents by dividing the graft into two or four
to restore endothelial function universally is to pieces, respectively.
perform an allogenic graft. Since Melles [9] revo- 3. In DSO, or Descemet stripping only, there is
lutionized the field in 2004 describing desce- no grafting, only a descemetorrhexis, and
metorrhexis, a method to dissect only Descemet relies on primary healing of the peripheral
Membrane (DM) from the recipient eye, leaving endothelium [8]. There is a need for longer
the posterior lamella intact, and after Price [10] term comparison studies, but it has several
and Gorovoy [11] pioneered the procedure known advantages over the other two procedures, it
as Descemet stripping endothelial keratoplasty requires only basic skills, it does not need
(DSEK) [10], a variety of endothelial kerato- donor tissue, there is no risk of rejection, and
plasty techniques have taken over PK as the elec- there are no early postoperative complications
tive procedure in endothelial keratoplasty. such as DM detachment. On the other hand, a
Nowadays, all the different approaches include good peripheral endothelial cell count is
“descemetorrhexis,” and the difference lies in the needed, the disease must be limited to the
tissue grafted: 5 mm-central part, and although it may pro-
vide similar visual outcomes to DMEK, it
1. In DSAEK, or Descemet stripping automated requires longer periods to achieve transpar-
endothelial keratoplasty, the graft is prepared ency with lower endothelial cell counts as a
using a microkeratome and includes not only baseline point. The instillation of ROCK
DM and endothelium but also part of the pos- inhibitors has been used to speed up recovery
terior stroma [11]. It has been widely adopted, and to salvage failing cases [20].
and the eye bank produces precut tissue which
is used directly by the surgeon [12]. Although
the correlation between preoperative graft Cultured Corneal Endothelial Cells
thickness and clinical outcomes has been dis-
puted [13, 14], there is a tendency to believe Human corneal endothelial cells (hCECs) are
that thinner grafts are associated with better arrested at G1 phase of the cell cycle, and do not
visual acuity. Ultra-thin DSAEK is a variant proliferate in vivo, in part due to contact inhibi-
of the technique where grafts are around 100 tion but also presumably because of lack of
microns to improve the visual acuity of stan- growth factor stimulation even when damage to
dard DSAEK [15]. the endothelial layer occurs [21]. Therefore, the
2. In DMEK, or Descemet membrane endothe- supply of human corneal tissue is limited; there-
lial keratoplasty, a step forward in the endo- fore, in vitro CEC culture is an option to increase
thelial keratoplasty developed by Melles [16, the number of cells for potential therapeutic pur-
17], the graft consists of endothelium and DM poses. However, this is challenging by the very
without any stroma, around 10–15 microns biology of CECs, and it is important to consider
thick. Compared to DSAEK, DMEK has bet- several factors:
ter visual outcomes, faster recovery time, and
36 Cultured Cells for Corneal Endothelial Therapy 487
Donor Factors Cell Isolation Protocols
Age of Donors Isolation of hCECs is one of the most critical

For cell culture of corneal endothelial cells, it is steps for a successful culture. The most com-
essential to start from a source of viable and pro- monly used method is the peel-and-digest proto-
liferating cells, i.e., young human corneal tissue. col by Peh’s Laboratory [23]. The endothelium
Most human corneas are used for transplantation, along with the Descemet membrane is separated
leaving those of old donors with less endothelial from the rest of the cornea, and this is enzymati-
cell count for research. It has been shown that in cally digested by collagenase. This enzyme gen-
the corneas of young donors (<30 years old), the tly digests the junctions of the endothelial cells to
mean cell density is 3000 cells/mm2, while in old the Descemet membrane (DM), consisting
donors (> 50 years old), it is 2700 cells/mm2 [22]. mainly of ECM proteins like collagen IV. The
There are also differences in cellular morphol- intercellular junctions mediated by ZO-1 are
ogy within these two groups, young donor endo- maintained as well as cell-to-basement mem-
thelial cells show homogeneous hexagonal brane interactions [23]. Other enzymatic meth-
morphology, while older donor cells were poly- ods have also been tried, such as trypsin, causing
morphic. Proliferative capability was maintained complete degradation of the CECs when too
from young donor CECs, maintaining their mor- aggressive, or separation by EDTA and pipetting,
phology and characteristics until the third pas- a technique by which the CECs did not maintain
sage, while old donor CECs were senescent viability either [27, 28].
earlier during the culture [23] hence culturing Cell viability is checked routinely in eye banks
cells from old donors is more challenging. using Trypan blue positive cell count, using
Trypan Blue staining (0.25%), and counting blue
Tissue Preservation stained cells as dead cells. Using this method and
The quality of donor corneas also depends on tis- a hemocytometer for counting, viability and plat-
sue preservation conditions. There are two funda- ing density can also be checked after cell isola-
mental methods, maintenance in Optisol-GS and tion [26].
organ culture. Optisol-GS corneal storage
medium (Bausch & Lomb, Irvine, California), a
hybrid of K-sol and DexSol media containing Coatings
chondroitin sulfate and dextran, is stored at 2 °C
to 8 °C for 14 days [24]. Meanwhile, organ cul- In vivo, endothelial cells adhere to the Descemet
ture maintains the corneas between 31 °C and membrane via extracellular matrix proteins. The
37 °C for up to 28 days, using different culture extracellular matrix is composed of different
media. Most of these media are supplemented collagens, laminin, and fibronectin among oth-
with serum such as CorneaMax, but serum-free ers. With the idea of creating a biomimetic envi-
media such as Human Endothelial-SFM is also ronment, these and other cell adhesion coatings
used [25]. Viability comparison studies showed a have been evaluated for culturing endothelial
dead cell percentage of 9.34% ± 4% and cells. Comparison of wells precoated with
0.46% ± 0.3% in Optisol-GS and organ culture, Fibronectin, Poly-D-Lysine, Collagen I,
respectively [26]. Nevertheless, successful cell Fibronectin/Collagen I, or FNC Coating Mix
culture was obtained from tissue preserved in [29] showed that the coating with the higher
both conditions. Although the viability is higher adhesion with almost 100% of cells attached
with organ culture, in both cases proliferation, after rinsing while maintaining cell morphology
hexagonal morphology and expression of typical was FNC Coating Mix, followed by Collagen I
CECs markers are achieved. and Fibronectin/Collagen I (with 90% of cells
attached). On the downside, FNC Coating Mix

is a commercially formulated reagent contain-
ing bovine fibronectin and bovine collagen I
among other components so it is useful for cell
culture but not suitable for clinical studies [29,
30]. Other studies have shown collagen IV as an
optimal coating for the culture of CECs for tis-
sue engineering as it is part of the endothelial
basement membrane [31].
Media Fig. 36.1 Phase contrast microphotograph of a human

CEC culture showing the typical polygonal cell morphol-
Different culture media have been used for the ogy. Bar: 200 μm
expansion of CECs, usually with a dual approach,
with a proliferation medium followed by a main- SFM 4% FBS, 50 μg/mL gentamicin, and 1%
tenance medium. antibiotic/antimycotic (Fig. 36.1).
For proliferation, combinations of one or two
media with external growth factors have been
used. The media include DMEM, DMEM/F12, Carriers for DSAEK and DMEK
Opti-MEM-I, and Ham’s F12/M199, compared
by Peh [23]. CECs cultured with DMEM or Following the isolation of hCECs, the next step is
DMEM/F12 do not go beyond the first or second to engineer a scaffold mimicking DM and use it
passage, while using Opti-MEM-I or Ham’s F12/ as a graft. This scaffold needs to provide a favor-
M199, the cells start to show typical endothelial able environment for endothelial cell expansion
markers such as Na+/K+ ATPase or ZO-1 from and maintenance as well as a robust tissue that
passage 3 [23]. As human CECs do not prolifer- can be handled easily for transplantation. In
ate, external factors and supplements have been recent decades, studies have been carried out
used to overcome the cell cycle arrest such as using both natural and synthetic materials that
serum, ascorbic acid, FGF, or insulin [32]. can serve as grafts with CECs. Today, in addition
However, hexagonal morphology was not to using biomaterials as scaffolds, their use is
achieved by culturing in proliferation media being studied to increase cell viability and long-
alone. For the maintenance of CECs five media term transplantation success [39].
were compared, including HCEC growth medium
(F99), MEM with FCS, and humanized endothe-
lial SFM, the latter being the one with the best Natural Scaffolds
results in terms of lower endothelial cell apopto-
sis [33, 34]. Natural scaffolds can be obtained from different
Parekh [35] cultured CECs using only a pro- animal sources, which mimic components of
liferation medium based on Ham’s F12/M199. DM, improving biocompatibility, proliferation,
Other groups used Opti-MEM-I with 8% FBS and maintaining the phenotype of CECs.
and supplemented with ROCK inhibitor However, since they are derived from animals,
(Y-27632) [36–38]. One of the most effective their composition is not well defined, and the
protocols is Peh’s Laboratory [23], which uses a resulting scaffolds show little optical transpar-
proliferation media with Ham’s F12/M199 with ency and weak mechanical properties.
5% FBS, 20 μg/mL ascorbic acid, 1% ITS, 10 ng/ Initially, as with coatings, the use of natural
mL FGF2 and 1% antibiotic/antimycotic com- polymer from the extracellular matrix such as
bined a maintenance media Human endothelial- collagen was considered because of its biocom-
patibility properties, low immunogenicity, and The human amniotic membrane is a collagen-
degradability. However, the laminas were not based scaffold that can be used intact, decellular-
consistent, difficult to handle, and easily degraded ized or lyophilized and possesses
by proteases. To solve this, different hardening anti-inflammatory, anti-fibrosis, and anti-
techniques have been used, such as chemical angiogenic properties that reduce potential graft
crosslinking or physical crosslinking by ultravio- rejection and have been used in other ocular
let light, rendering suboptimal results [40, 41]. applications [27]. The main problems are avail-
Over time, technologies have appeared that allow ability and lack of mass manufacturing, sub-
for the creation of plastic compressed collagen optimal transparency with a low biodegradation
films, based on Real Architecture for 3D Tissues rate in long-term transplantation, and risk of con-
(RAFT) that allow rapid production of grafts tamination and transmission of infectious dis-
with improved mechanical properties without eases [48].
compromising biosafety; however, transparency Decellularized corneas provide the perfect
is not adequate, and there are no in vivo studies substrate for CECs to grow while maintaining
yet [42]. optimal transparency and ultrastructure.
Other natural polymers have been tried such Decellularization removes native cells and other
as gelatin or chitosan. Gelatin has great poros- immunogenic compounds while preserving the
ity, permeability to water, helps cell adhesion, structural and functional proteins of the stroma
and is widely available [43]. However, gelatin [49]. Different corneal scaffolds have been used,
hydrogels do not provide stability as a graft, from porcine corneas to human. Due to a low
and there is a risk of carrying bovine spongi- number of donated corneas and a lengthy decel-
form encephalopathy due to the source of gela- lularization process, obtaining various lamellae
tin [44]. Chitosan is a biomimetic polysaccharide per cornea with the femtosecond laser method is
derived from chitin and has great biocompati- vital for the usage of this material as a scaffold
bility but low strength. To create a hard con- [30]. There are various studies with clinical
struct, it was combined with other natural applications leading to corneal edema relief [50,
materials, and a graft consisting of hydroxy- 51].
ethyl chitosan, gelatin, and chondroitin sulfate Human crystalline lens capsule is composed
was created and tested on CECs, showing of collagen IV and sulfated glycosaminoglycans.
promising results but causing in vivo inflamma- The anterior lens capsule is a byproduct of cata-
tion in animal models [45]. ract surgery and presents biomechanical proper-
An approach using silk fibroin precoated with ties similar to DMEK grafts, can be used
collagen type IV, has also been evaluated for decellularized with good biocompatibility and
human CEC culture [46]. Silk fibroin a natural inherent transparency; however, there are limita-
fibrin derived from silk has low immunogenicity tions due to their small diameter and high depen-
and good transparency but on its own cannot dency on the supply of cadaveric eye donors [52].
maintain a CEC culture, lacks elasticity and In addition, a natural material xenograft using
mechanical strength, and can cause hypersensi- decellularized fish scales is being assessed. It
tivity. Using non-mulberry silk combined with presents a collagen I pattern similar to the human
other materials shows better biocompatibility, but cornea and provides a cost-effective available
further studies need to be done [47]. substrate for corneal grafts. CECs adhesion is
Other biologically derived scaffolds are mem- adequate but can be improved with FNC coating,
branes such as amniotic membrane, decellular- and proliferation is irregular, but post-
ized cornea, and human anterior lens capsule. In modification fish scale scaffolds show some
both, the high dependency on the human donor is promise due to their inherent transparency being
a limitation. similar to DSAEK grafts [53].
Synthetic Scaffolds function, as the corneas maintained their optical

transparency.
There are interesting materials because their Synthetic hydrogels of poly-ε-lysine cross-
properties such as the structure, shape, chemical linked 60% with octanedioic-acid to a polymer
composition, mechanical strength, and durability density of 0.066 g/mL using
can be customized. Therefore, many authors try N-hydroxysulfosuccinimide (NHS) and 1-ethyl-
to find the best synthetic scaffold-based to regen- 3-(3-dimethylaminopropyl) carbodiimide (EDC),
erate the corneal endothelium. produced a thin, transparent, porous, and robust
Kruse [54] compared scaffolds of poly substrate for corneal endothelial cells culture
(methyl-methacrylate) (PMMA), poly (lactic-co- [56]. Their results demonstrated that functional-
glycolic acid) (PLGA) and polycaprolactone ization of the poly-ε-lysine hydrogel with
(PCL) for the culture of hCECs. PLGA fibers arginine-glycine-aspartic acid (RGD) provides a
were spun from a solution with a mass concentra- suitable surface for 5-week culture of primary
tion of 5 w/v% in 75% chloroform and 25% porcine hCECs and facilitates the generation of
methanol. PCL fibers were spun from a 14 w/v% the confluent monolayer with of ZO-1 and Na+/
solution in 75% CHCl3 and 25% MeOH. PMMA K+ ATPase expression.
fibers were produced from a 16 w/v% solution of
75% CHCl3 and 25% MeOH. Even using identi-
cal production parameters, the three scaffolds Combination of Natural
differed significantly in terms of viscosity, pore and Synthetic Materials
size, thickness, and light transmittance. Then,
40,000 cells/cm2 of human corneal endothelial Other authors combine natural and synthetic
cell line (HCEC-12) were seeded onto the scaf- polymers to create a biomaterial with the advan-
folds and cultured for a week. The results revealed tages of both. The mechanical properties of the
that HCEC-12 mainly grew on the surface and synthetic scaffolds and the extracellular matrix
retained physiological morphology, but the for- (ECM) proteins of natural ones.
mation of a uniform monolayer was not evident Kim [57] created the Col-I-PLGA scaffold by
in PLGA. The PCL scaffold maintained high cell combining the appropriate mechanical strength
viability, while PMMA showed cytotoxicity. In of the 5 w/v% PLGA films as a substrate with 5
conclusion, PLGA and PCL electro-spun scaf- μg/cm2 collagen I coating to enhance its biocom-
folds showed similar biocompatibility, but only patibility. This polymer adequately resembled the
PLGA maintained the characteristic polygonal required surface properties to facilitate adhesion,
shape of hCECs. migration, and proliferation of primary rabbit
Poly (ethylene glycol) (PEG)-based hydrogel corneal endothelial cells, as well as roughness,
films containing sebacoyl chloride (SebCl) and appropriate hydrophilicity, stability, and water
5 w/v% of α, ω-dihydroxy-poly (ε-caprolactone) uptake, compared to bare PLGA films. Also, the
(PCL) dissolved in dichloromethane showed cultured cells on Col I-PLGA scaffolds showed
similar tensile strengths to human corneal tissue significant enhancement in the expression of cor-
and more than 98% optical transparency activity neal endothelial cell-associated marker genes
[55]. In vitro analysis performed with sheep such as aquaporin and Na+/K+ ATPase, along
CECs on hydrogel films resulted in 100% conflu- with well-maintained cell morphology.
ence with natural morphology after 7 days. In Palchesko [58] demonstrated that bovine
vivo studies revealed that the cell-free hydrogel CECs cultured in vitro on a polydimethylsiloxane
implanted on the inner surface of ovine corneas surface with an elastic modulus of 50 kPa previ-
for 28 days showed no toxicity or inflammatory ously coated with collagen IV grew in monolayer
response and did not compromise the native CEC with a polygonal morphology and positive stain-
ing for the characteristic endothelial marker cell monolayers, are highly fragile and techni-
ZO-1. cally difficult to transplant into the anterior
Rizwan [59] produced an improved gelatin chamber. To overcome this problem, some
methacrylate hydrogel named GelMA+ and UV researchers have transplanted cultured corneal
crosslinking. GelMA+ showed an eight-fold endothelial sheets with a carrier, but they have
increase in mechanical strength and slower deg- adhered only temporarily before eventually
radation compared to regular GelMA. In addi- detaching, with the exception of corneal stromal
tion, primary human CECs at passage 3 from laminas, which is a limited source and whose
donor corneas reached confluence in a monolayer necessity hinders the advantages of transplanta-
with rise ZO-1 expression, higher cell density tion of cultured CEC [65]. However, this ther-
and cell size homogeneity on GelMA+ carrier moresponsive polymer has been used for patient
compared to GelMA. therapy to enable corneal epithelial reconstruc-
Wang [60] hybridized chitosan and polycap- tion [66].
rolactone (PCL) and cultured bovine corneal
endothelial cells on this scaffold and reported
that the cells reached confluence on day 11, dis- tem Cells Induced Differentiation
S
played a normal polygonal morphology and to Human Corneal Endothelial Cells
showed ZO-1, Na+/K+ ATPase expression after
14 days of incubation on the 25% PCL and 75% Since the corneal endothelium was shown to be
chitosan blend membrane. derived from neural crest [67, 68], most
An alternative method to is cell sheet engi- approaches to induce corneal endothelial cell dif-
neering. Cells were cultured on the surface of a ferentiation from stem cells in vitro started mim-
stimuli-sensitive polymer that allows controlled icking the developmental process. The strategy
cell adhesion and detachment without using pro- consisted of a first phase in which stem cells were
teolytic enzymes. differentiated into neural crest cells and a second
Several studies have shown that Poly-N- stage in which corneal endothelial cells were fur-
isopropylacrylamide (PIPAAm) is a good ther differentiated from these neural crest cells.
temperature-responsive polymer for generating Three labs, McCabe [69], Ali [70], and
hCEC sheets. Their chains display hydrophobic Wagoner [71], independently derived corneal
properties at 37 °C so the cultured cells could endothelium from pluripotent stem cells under
adhere and proliferate on the polymer. In con- chemically defined conditions with a first step
trast, by lowering the culture temperature to called “dual inhibition” to promote neural crest
20 °C, the polymer turns into a hydrophilic state cell induction, either embryonic stem cells (ESC)
with fully extended chains, so the formed cell or induced pluripotent stem cells (iPSC). McCabe
sheets spontaneously detach from the surface [69] and Ali [70] used 10 μM TGF beta signaling
with intact ECM proteins. The harvested hCECs, inhibitor SB431542 and 500 ng/mL BMP signal-
which exhibit hexagonal morphology with the ing inhibitor Noggin in a basal medium of
presence of microvilli and cellular interconnec- DMEM-F12, knock out serum replacement, non-
tions, were transferred to gelatin disc supports essential AA, and 8 ng/mL fibroblast growth fac-
for transplantation into the anterior chamber of tor 2 (FGF2). However, Wagoner [71] used 3 μM
rabbit models. After 2 weeks, the hCEC film GSK-3 inhibitor CHIR99021 instead of a BMP
was attached to the denuded surface of signaling blocker in a basal medium of DMEM/
Descemet’s membrane with tight junction for- F12, bovine serum albumin (BSA), 50 μg/ml
mation (ZO-1) between cells [61–64]. This (+)-sodium L-ascorbate, 10 μg/mL transferrin,
approach has not gone clinically forward 10 ng/mL Heregulin β-1, 200 ng/mL IGF-I, and
because cultured corneal endothelial sheets, as 8 ng/mL FGF2. After a minimum of 3 days, the
Fig. 36.2 Confocal images of Na+/K+ ATPase immunofluorescence (in green) in human ADSC-derived CEC using
Wagoner et al. (left) or Ali et al. (right) differentiation media. DAPI nuclear staining in blue. Bars: 100 μm.
dual inhibitors were replaced by 10 ng/mL

platelet-
derived growth factor B (PDGF-BB),
10 ng/mL Dickkopf-related protein 2 (DKK-2),
and 0.1 × B27 supplement for at least 7 days to
generate hexagonal corneal endothelial-like cells.
Their analyses revealed increased expression of
corneal endothelial cell-associated markers such
as ZO-1 and Na+/K+ ATPase α1 (ATP1A1) as
well as the key Descemet’s membrane protein,
Collagen type VIII (COL8A1 and COL8A2).
At the same time, Zhao and Afshari [72] used
a three-step chemical method. A first dual inhibi-
tion step like previous researchers with 5 μM
Fig. 36.3 Confocal image of N-Cadherin immunofluo-
SB431542 and 50 nM BMP signaling inhibitor rescence in human ADSC-derived CEC using Wagoner
LDN193189, adding a Wnt inhibitor 1 μM IWP2 et al. differentiation media. DAPI nuclear staining in blue.
to raise eye field stem cell development in a prim- Bar: 50 μm
ing medium of DMEM/F12, N2, B27, BSA, non-
essential AA for 6 days. Next, they derived neural
crest cells from these stem cells using an induc- adult patients. This would be advantageous
tion medium of DMEM/F12 50:50, N2, B27, because the risk of rejection may be reduced
0.3 mM 2-phospho-l-ascorbic acid supplemented when patient-specific autologous cells are used
with 3 μM CHIR99021. Lastly, they were able to for the treatment of corneal endothelial disorders.
differentiate neural crest cells into corneal Among them, Ali et al. [70] show the highest
endothelial-like cells, which expressed Na+/K+ advantage because with only 20 days of proce-
ATPase, ZO-1, and N-cadherin, with human dure, they generated CECs with 90.82% pro-
endothelial-SFM, 5% FBS, 0.3 mM 2-phosphate teome similarity to a human corneal endothelium
ascorbic acid, 1 μM SB431542, and a 2.5 μM (Figs. 36.2 and 36.3).
ROCK inhibitor H-1125. On the other hand, three other labs derived
In our opinion, both Ali [70] and Wagoner corneal endothelial-like cells from stem cells
[71] are better protocols than the others men- using different cell-conditioned media.
tioned above because both have been able to Obviously, these approaches with conditioned
achieve the generation of CECs using cells from media are less applicable to clinical practice, as
undefined factors and concentrations of the mol- Clinical Studies on CEC

ecules present in the conditioned media prevent Transplantation
their safe and reproducible use:
Zhang [73] derived corneal endothelial-like There are some alternative procedures that are
cells from human ESCs by co-culture for 5 days currently evaluated under clinical trials and study
with human corneal stroma cells in a basal the use of carriers and endothelial cells in
medium contained DMEM/F12 supplemented culture:
with 10% FBS, B27, 20 ng/ml EGF and 40 ng/ml
bFGF to generate an outgrowth of precursors of 1. CECs migrate much more efficiently over
neural crest cells which expressed CD73 and intact DM rather than bare corneal stroma in
FoxC1. Next, the medium was changed to SV-40 DSO, leading to the idea that for the treatment
transformed human lens epithelial cell- of FECD, DSO could potentially be improved
conditioned medium for 14 additional days to by increasing the size of the descemetorrhexis
obtain a monolayer of corneal endothelial-like to incorporate most of the large guttas, but
cells with positive signals for Na+/K+ ATPase, providing a cell-free Descemet’s membrane
ZO-1, vimentin, and N-cadherin. graft afterwards to complete a descemetor-
Chen [74] promoted neural crest cell differen- rhexis. This way it acts as a support for endo-
tiation from mouse ESC and mouse iPSC by cul- thelial cells favoring their proliferation and
turing them in a first stage with embryonic body centripetal migration. This technique is known
differentiation medium adding 1 μM all-trans as Descemet membrane transfer (DMT) [76].
retinoic acid during 4 days. Then, they induced Unlike endothelial keratoplasty, it has the
differentiation towards corneal endothelial cells advantage of using an acellular graft that is
by exposing them for 14–17 days to conditioned widely available and avoiding problems
medium collected from rabbit lens epithelial cell related to postoperative graft rejection due to
culture medium. The differentiated cells pre- the absence of allogeneic endothelium. A
sented an up-regulation of corneal endothelial clinical trial is currently underway to evaluate
cell-associated marker genes as Aquaporin-1, the efficacy of DMT for the treatment of
ZO-1, Na+/K+ ATPase, N-cadherin, and Collage FECD in a larger cohort of patients and for
type VIII compared with undifferentiated cells. longer-term monitoring of its safety and effi-
In search for adult stem cells capable of CEC cacy (ClinicalTrials.gov; identifier:
differentiation, Bosch [75] used dental pulp stem NCT03275896).
cells. They transdifferentiated these stem cells 2. Cell culture techniques make it possible to
into neural crest stem cells with an induction expand ex vivo the CEC to subsequently inject
medium consisting of DMEM-F12 supplemented a cell solution into the anterior chamber [37],
with 1× B-27, 1× N-2, 20 ng/mL EGF, 20 ng/ml or else to manufacture constructs made up of
FGF2, 5 ng/mL heparin, and 2 mM L-alanyl-l- acellular corneal stroma, acellular Descemet
glutamine. On day 4, an adequate number of cells membrane or material manufactured by tissue
showed up-regulation of neural crest stem cells bioengineering [51], and colonized by
markers such as AP2, Nestin, and p75; therefore, expanded CEC. These grafts could then be
these cells were cultured in the human corneal transplanted onto the recipient in the same
endothelial conditioned medium for a further way as in the previously seen endothelial kera-
15 days to derive corneal endothelial cells. At the toplasties. In both approaches, a single popula-
end of the differentiation process, gene expres- tion of endothelial cells can be amplified many
sion of typical CEC markers like ZO-1, Na+/K+ times for distribution to large numbers of
ATPase pump ATP1A1 and extracellular matrix patients. Currently, within the framework of a
components COL4A2 and COL8A2 were signifi- clinical trial that included 11 patients, it has
cantly increased compared to undifferentiated been found that the injection of cells in sus-
dental pulp stem cells. pension is capable of effectively treating cor-
neal edema secondary to various conditions, Respect to carriers, so far, the most advanta-
including Fuchs’ Dystrophy and pseudophakic geous carrier is a corneal stroma decellularized
bullous keratopathy, in addition to secondary lamina [30, 51, 82]. However, this carrier still
corneal edema, argon laser peripheral iri- depends on donors; new advances in biomi-
dotomy (LPI) or pseudoexfoliation syndrome metic materials and manufacturing protocols
[37]. At 2 years after cell injection, corneal such as electrospinning, electrogradient trans-
thickness was less than 600 μm in 10 eyes, and port, shear flow, nano-lithography, flow-induced
the cornea was thinner than the baseline mea- crystallization, vitrification, and advances in
sure in all 11 eyes. The same study [37], how- novel 3D printing techniques such as LIFT,
ever, also found a relatively broad range of laser-assisted bioprinting, and fused filament
endothelial counts among trial participants fabrication, and other methods of achieving
2 years after treatment (mean CEC density, lamellar parallel bundles of collagen, such as
1534 cells per square millimeter [95% CI, molecular crowding and densification to a liq-
1213 to 1855]). Each of the 11 eyes main- uid crystalline state [83–86] will aid in the
tained corneal transparency. Regarding the search for a donor-independent biocompatible
efficacy of tissue bioengineered constructs, carrier.
there are no human data yet, although a clini- Further development of these and previous
cal trial is currently underway (ClinicalTrials. approaches by defining the growth factors, the
gov; identifier: NCT04319848). signaling pathways implicated in directed differ-
entiation, the use of more practical cells to derive
hCECs, and the in vivo demonstration of func-
oncluding Remarks and Future
C tionality are urgently needed.
Perspectives
Take Home Notes
Nowadays, we are in an outstanding position to • Recently, a variety of endothelial kerato-
develop corneal endothelial cell sheets for endo- plasty techniques to restore endothelial
thelial keratoplasty: With respect to culture con- function have taken over the classical allo-
ditions, reproducible and well-defined culturing genic graft. However, there is a scarcity of
methods, and conditions have been achieved in donors to adequate to high and increasing
the last decades [23, 33, 34, 36, 77–79]. demand.
Regardless of advances in promoting hCEC • Nowadays, we are in an outstanding posi-
proliferation, the achieved capacity for expand- tion to develop corneal endothelial cell
ing human CECs is still highly limited; new sheets for endothelial keratoplasty: repro-
sources of CECs are therefore sought. The use of ducible and well-defined culturing methods
extraocular cells capable of differentiating into and conditions have been achieved in the
corneal endothelial cells is highly desirable. last decades.
Recent advances have been achieved in differen- • The use of extraocular cells capable of dif-
tiation protocols from adipose-derived mesen- ferentiating into corneal endothelial cells
chymal stem cells (ADSC) from our lab [80]. from embryonic stem cells and adipose-
Our results broaden the type of cells of autolo- derived mesenchymal stem cells is readily
gous extraocular origin that could be employed in available.
the clinical setting for corneal endothelial defi- • New advances in biomimetic materials and
ciency. In addition, recent in vivo demonstration manufacturing protocols such as electrospin-
of the functionality of hESC-derived hCEC ning, nanolithography, vitrification, and
together with nicotinamide [81] provides experi- advances in novel 3D printing techniques and
mental evidence for a potential approach for others will aid in the search for a donor-
treating corneal endothelial dysfunction. independent biocompatible carrier.
• Further development of these and previous regularity on vision recovery after endothelial kerato-
plasty. Br J Ophthalmol. 2020;104(9):1317–23.
approaches by defining the growth factors, the 14. Perone J-M, Goetz C, Zevering Y, Derumigny A, Bloch
signaling pathways implicated in directed dif- F, Vermion J-C, et al. Graft thickness at 6 months
ferentiation, the use of more practical cells to postoperatively predicts long-term visual acuity out-
derive hCECs, and the in vivo demonstration comes of descemet stripping automated endothelial
keratoplasty for fuchs dystrophy and moderate phakic
of functionality are urgently needed. bullous keratopathy. Cornea. 2021;41(11):1362.
15. Busin M, Madi S, Santorum P, Scorcia V, Beltz
J. Ultrathin descemet’s stripping automated endo-
thelial keratoplasty with the microkeratome double-
pass technique: two-year outcomes. Ophthalmology.
References 2013;120(6):1186–94.
16. Melles GRJ, Ong TS, Ververs B, van der Wees
1. Moshirfar M, Thomson AC, Ronquillo Y. Corneal J. Preliminary clinical results of descemet mem-
endothelial transplantation. Treasure Islad, FL: brane endothelial keratoplasty. Am J Ophthalmol.
StatPearls Publishing; 2022. 2008;145(2):222.
2. Bourne W, Nelson L, Hodge D. Central corneal endo- 17. Melles GRJ, Ong TS, Ververs B, van der Wees
thelial cell changes over a ten-year period. Invest J. Descemet membrane endothelial keratoplasty
Ophthalmol Vis Sci. 1997;38(3):779–82. (DMEK). Cornea. 2006;25(8):987–90.
3. He Z, Campolmi N, Gain P, Ha Thi BM, Dumollard 18. Lam FC, Baydoun L, Dirisamer M, Lie J, Dapena I,
JM, Duband S, et al. Revisited microanatomy of the Melles GRJ. Hemi-descemet membrane endothelial
corneal endothelial periphery: new evidence for con- keratoplasty transplantation: a potential method for
tinuous centripetal migration of endothelial cells in increasing the pool of endothelial graft tissue. JAMA
humans. Stem Cells. 2012;30(11):2523–34. Ophthalmol. 2014;132(12):1469–73.
4. Joyce NC, Harris DL, Mello DM. Mechanisms of 19. Birbal RS, Ni Dhubhghaill S, Baydoun L, Ham L,
mitotic inhibition in corneal endothelium: contact Bourgonje VJA, Dapena I, et al. Quarter-descemet
inhibition and TGF-β2. Investig Ophthalmol Vis Sci. membrane endothelial keratoplasty: one- to two-year
2002;43(7):2152–9. clinical outcomes. Cornea. 2020;39(3):277–82.
5. Sherrard ES. The corneal endothelium in vivo: 20. Moloney G, Garcerant Congote D, Hirnschall N,
its response to mild trauma. Exp Eye Res. Arsiwalla T, Luiza Mylla Boso A, Toalster N, et al.
1976;22(4):347–57. Descemet stripping only supplemented with topical
6. Joyce NC, Meklir B, Joyce SJ, Zieske JD. Cell Ripasudil for Fuchs endothelial dystrophy 12-month
cycle protein expression and proliferative status in outcomes of the Sydney eye hospital study. Cornea.
human corneal cells. Investig Ophthalmol Vis Sci. 2021;40(3):320–6.
1996;37(4):645–55. 21. Joyce NC. Proliferative capacity of corneal endothe-
7. Dirisamer M, Yeh RY, Van Dijk K, Ham L, Dapena lial cells. Exp Eye Res. 2012;95(1):16–23.
I, Melles GRJ. Recipient endothelium may relate to 22. McGlumphy EJ, Margo JA, Haidara M, Brown CH,
corneal clearance in descemet membrane endothelial Hoover CK, Munir WM. Predictive value of corneal
transfer. Am J Ophthalmol. 2012;154(2):290–296.e1. donor demographics on endothelial cell density.
8. Garcerant D, Hirnschall N, Toalster N, Zhu M, Cornea. 2018;37(9):1159–62.
Wen L, Moloney G. Descemet’s stripping without 23. Peh GSL, Toh KP, Wu FY, Tan DT, Mehta
endothelial keratoplasty. Curr Opin Ophthalmol. JS. Cultivation of human corneal endothelial cells
2019;30(4):275–85. isolated from paired donor corneas. PLoS One.
9. Melles GRJ, Wijdh RHJ, Nieuwendaal CP. A tech- 2011;6(12):e28310.
nique to excise the descemet membrane from 24. Kitazawa K, Inatomi T, Tanioka H, Kawasaki S,
a recipient cornea (Descemetorhexis). Cornea. Nakagawa H, Hieda O, et al. The existence of dead
2004;23(3):286–8. cells in donor corneal endothelium preserved with stor-
10. Price FW, Price MO. Descemet’s stripping with endo- age media. Br J Ophthalmol. 2017;101(12):1725–30.
thelial keratoplasty in 50 eyes: a refractive neutral cor- 25. Valtink M, Donath P, Engelmann K, Knels L. Effect
neal transplant. J Refract Surg. 2005;21(4):339–45. of different culture media and deswelling agents on
11. Gorovoy MS. Descemet-stripping automated endo- survival of human corneal endothelial and epithelial
thelial keratoplasty. Cornea. 2006;25(8):886–9. cells in vitro. Graefes Arch Clin Exp Ophthalmol.
12. Chen ES, Terry MA, Shamie N, Hoar KL, Friend 2016;254(2):285–95.
DJ. Precut tissue in descemet’s stripping automated 26. Parekh M, Peh G, Mehta JS, Ahmad S, Ponzin D,
endothelial keratoplasty. Donor characteristics and Ferrari S. Effects of corneal preservation conditions
early postoperative complications. Ophthalmology. on human corneal endothelial cell culture. Exp Eye
2008;115(3):497–502. Res. 2019;179:93–101.
13. Tourabaly M, Chetrit Y, Provost J, Georgeon C, Kallel 27. Ishino Y, Sano Y, Nakamura T, Connon CJ, Rigby
S, Temstet C, et al. Influence of graft thickness and H, Fullwood NJ, et al. Amniotic membrane as a
carrier for cultivated human corneal endothelial 41. Navaratnam J, Utheim T, Rajasekhar V, Shahdadfar
cell transplantation. Investig Ophthalmol Vis Sci. A. Substrates for expansion of corneal endothelial
2004;45(3):800–6. cells towards bioengineering of human corneal endo-
28. Engelmann K, Bednarz J, Valtink M. Prospects thelium. J Funct Biomater. 2015;6(3):917–45.
for endothelial transplantation. Exp Eye Res. 42. Levis H, Kureshi A, Massie I, Morgan L,
2004;78(3):573–8. Vernon A, Daniels J. Tissue engineering the cor-
29. Engler C, Kelliher C, Speck CL, Jun AS. Assessment nea: the evolution of RAFT. J Funct Biomater.
of attachment factors for primary cultured human cor- 2015;6(1):50–65.
neal endothelial cells. Cornea. 2009;28(9):1050–4. 43. Watanabe R, Hayashi R, Kimura Y, Tanaka Y,
30. He Z, Forest F, Bernard A, Gauthier AS, Montard R, Kageyama T, Hara S, et al. A novel gelatin hydrogel
Peoc’h M, et al. Cutting and decellularization of mul- carrier sheet for corneal endothelial transplantation.
tiple corneal stromal lamellae for the bioengineering Tissue Eng Part A. 2011;17(17–18):2213–9.
of endothelial grafts. Investig Ophthalmol Vis Sci. 44. Lai JY, Cheng HY, Hui-Kang MD. Investigation of
2016;57(15):6639–51. overrun-processed porous hyaluronic acid carriers
31. Zhu YT, Tighe S, Chen SL, John T, Kao WY, Tseng in corneal endothelial tissue engineering. PLoS One.
SCG. Engineering of human corneal endothelial 2015;10(8):1–20.
grafts. Curr Ophthalmol Rep. 2015;3(3):207–17. 45. Liang Y, Liu W, Han B, Yang C, Ma Q, Zhao W, et al.
32. Møller-Pedersen T, Hartmann U, Ehlers N, Engelmann Fabrication and characters of a corneal endothelial
K. Evaluation of potential organ culture media for cells scaffold based on chitosan. J Mater Sci Mater
eye banking using a human corneal endothelial cell Med. 2011;22(1):175–83.
growth assay. Graefes Arch Clin Exp Ophthalmol. 46. Madden PW, Lai JNX, George KA, Giovenco T,
2001;239(10):778–82. Harkin DG, Chirila TV. Human corneal endothelial
33. Jäckel T, Knels L, Valtink M, Funk RHW, Engelmann cell growth on a silk fibroin membrane. Biomaterials.
K. Serum-free corneal organ culture medium (SFM) 2011;32(17):4076–84.
but not conventional minimal essential organ culture 47. Ramachandran C, Gupta P, Hazra S, Mandal BB. In
medium (MEM) protects human corneal endothelial vitro culture of human corneal endothelium on non-
cells from apoptotic and necrotic cell death. Br J mulberry silk fibroin films for tissue regeneration.
Ophthalmol. 2011;95(1):123–30. Transl Vis Sci Technol. 2020;9(4):1–15.
34. Bednarz J, Doubilei V, Wollnik PCM, Engelmann 48. Mimura T, Yamagami S, Amano S. Corneal endothe-
K. Effect of three different media on serum free culture lial regeneration and tissue engineering. Prog Retin
of donor corneas and isolated human corneal endothe- Eye Res. 2013;35:1–17.
lial cells. Br J Ophthalmol. 2001;85(12):1416–20. 49. Bayyoud T, Thaler S, Hofmann J, Maurus C, Spitzer
35. Parekh M, Romano V, Ruzza A, Kaye SB, Ponzin MS, Bartz-Schmidt KU, et al. Decellularized bovine
D, Ahmad S, et al. Culturing discarded peripheral corneal posterior lamellae as carrier matrix for culti-
human corneal endothelial cells from the tissues vated human corneal endothelial cells. Curr Eye Res.
deemed for preloaded DMEK transplants. Cornea. 2012;37(3):179–86.
2019;38(9):1175–81. 50. Chakraborty J, Roy S, Murab S, Ravani R, Kaur K,
36. Okumura N, Sakamoto Y, Fujii K, Kitano J, Nakano Devi S, et al. Modulation of macrophage phenotype,
S, Tsujimoto Y, et al. Rho kinase inhibitor enables maturation, and graft integration through chondroitin
cell-based therapy for corneal endothelial dysfunc- sulfate cross-linking to Decellularized cornea. ACS
tion. Sci Rep. 2016;6(January):1–11. Biomater Sci Eng. 2019;5(1):165–79.
37. Kinoshita S, Koizumi N, Ueno M, Okumura N, Imai 51. Arnalich-Montiel F, Moratilla A, Fuentes-Julián S,
K, Tanaka H, et al. Injection of cultured cells with Aparicio V, Martin MC, Peh G, et al. Treatment of
a ROCK inhibitor for bullous keratopathy. N Engl J corneal endothelial damage in a rabbit model with a
Med. 2018;378(11):995–1003. bioengineered graft using human decellularized cor-
38. Okumura N, Matsumoto D, Fukui Y, Teramoto M, neal lamina and cultured human corneal endothelium.
Imai H, Kurosawa T, et al. Feasibility of cell-based PLoS One. 2019;14(11):1–16.
therapy combined with descemetorhexis for treating 52. Yoeruek E, Saygili O, Spitzer MS, Tatar O, Bartz-
Fuchs endothelial corneal dystrophy in rabbit model. Schmidt KU. Human anterior lens capsule as carrier
PLoS One. 2018;13(1):1–14. matrix for cultivated human corneal endothelial cells.
39. Parekh M, Romano V, Hassanin K, Testa V, Cornea. 2009;28(4):416–20.
Wongvisavavit R, Ferrari S, et al. Biomaterials for 53. Parekh M, Van den Bogerd B, Zakaria N, Ponzin
corneal endothelial cell culture and tissue engineer- D, Ferrari S. Fish scale-derived scaffolds for cultur-
ing. J Tissue Eng. 2021;12:2041731421990536. ing human corneal endothelial cells. Stem Cells Int.
40. Koizumi N, Sakamoto Y, Okumura N, Okahara 2018;2018:8146834.
N, Tsuchiya H, Torii R, et al. Cultivated cor- 54. Kruse M, Walter P, Bauer B, Rütten S, Schaefer K,
neal endothelial cell sheet transplantation in a Plange N, et al. Electro-spun membranes as scaffolds
primate model. Investig Ophthalmol Vis Sci. for human corneal endothelial cells. Curr Eye Res.
2007;48(10):4519–26. 2018;43(1):1–11.
55. Ozcelik B, Brown KD, Blencowe A, Ladewig K, human embryonic stem cell-derived corneal endo-
Stevens GW, Scheerlinck JPY, et al. Biodegradable thelial cells by directed differentiation. PLoS One.
and biocompatible poly(ethylene glycol)-based 2015;10(12):1–24.
hydrogel films for the regeneration of corneal endo- 70. Ali M, Kabir F, Raskar S, Renuse S, Na CH, Delannoy
thelium. Adv Healthc Mater. 2014;3(9):1496–507. M, et al. Generation and proteome profiling of PBMC-
56. Kennedy S, Lace R, Carserides C, Gallagher AG, originated, iPSC-derived lentoid bodies. Stem Cell
Wellings DA, Williams RL, et al. Poly-ε-lysine based Res. 2020;46:101813.
hydrogels as synthetic substrates for the expansion of 71. Wagoner MD, Bohrer LR, Aldrich BT, Greiner MA,
corneal endothelial cells for transplantation. J Mater Mullins RF, Worthington KS, et al. Feeder-free differ-
Sci Mater Med. 2019;30(9):1–13. entiation of cells exhibiting characteristics of corneal
57. Kim EY, Tripathy N, Cho SA, Lee D, Khang endothelium from human induced pluripotent stem
G. Collagen type I–PLGA film as an efficient sub- cells. Biol Open. 2018;7(5):1–10.
stratum for corneal endothelial cells regeneration. J 72. Zhao JJ, Afshari NA. Generation of human corneal
Tissue Eng Regen Med. 2017;11(9):2471–8. endothelial cells via in vitro ocular lineage restriction
58. Palchesko RN, Lathrop KL, Funderburgh JL, Feinberg of pluripotent stem cells. Investig Ophthalmol Vis Sci.
AW. In vitro expansion of corneal endothelial cells on 2016;57(15):6878–84.
biomimetic substrates. Sci Rep. 2015;5:32–4. 73. Zhang K, Pang K, Wu X. Isolation and transplanta-
59. Rizwan M, Peh GS, Adnan K, Naso SL, Mendez tion of corneal endothelial cell-like cells derived from
AR, Mehta JS, et al. In vitro topographical model of in-vitro-differentiated human embryonic stem cells.
Fuchs dystrophy for evaluation of corneal endothe- Stem Cells Dev. 2014;23(12):1340–54.
lial cell monolayer formation. Adv Healthc Mater. 74. Chen P, Chen JZ, Shao CY, Li CY, Zhang YD, Lu
2016;5(22):2896–910. WJ, et al. Treatment with retinoic acid and lens epi-
60. Wang YH, Young TH, Wang TJ. Investigating thelial cell-conditioned medium in vitro directed
the effect of chitosan/ polycaprolactone blends the differentiation of pluripotent stem cells towards
in differentiation of corneal endothelial cells and corneal endothelial cell-like cells. Exp Ther Med.
extracellular matrix compositions. Exp Eye Res. 2015;9(2):351–60.
2019;185:107679. 75. Bosch BM, Salero E, Núñez-Toldrà R, Sabater AL,
61. Hsiue GH, Lai JY, Chen KH, Hsu WM. A novel Gil FJ, Perez RA. Discovering the potential of dental
strategy for corneal endothelial reconstruction pulp stem cells for corneal endothelial cell produc-
with a bioengineered cell sheet. Transplantation. tion: a proof of concept. Front Bioeng Biotechnol.
2006;81(3):473–6. 2021;96:17724.
62. Sumide T, Nishida K, Yamato M, Ide T, Hayashida Y, 76. Soh YQ, Mehta JS. Regenerative therapy for
Watanabe K, et al. Functional human corneal endothe- fuchs endothelial corneal dystrophy. Cornea.
lial cell sheets harvested from temperature-responsive 2018;37(4):523–7.
culture surfaces. FASEB J. 2006;20(2):392–4. 77. Peh GSL, Chng Z, Ang HP, Cheng TYD, Adnan
63. Ide T, Nishida K, Yamato M, Sumide T, Utsumi M, K, Seah XY, et al. Propagation of human corneal
Nozaki T, et al. Structural characterization of bio- endothelial cells: a novel dual media approach. Cell
engineered human corneal endothelial cell sheets Transplant. 2015;24(2):287–304.
fabricated on temperature-responsive culture dishes. 78. Peh GSL, Ang HP, Lwin CN, Adnan K, George
Biomaterials. 2006;27(4):607–14. BL, Seah XY, et al. Regulatory compliant tissue-
64. Lai JY, Chen KH, Hsiue GH. Tissue-engineered engineered human corneal endothelial grafts restore
human corneal endothelial cell sheet transplanta- corneal function of rabbits with bullous keratopathy.
tion in a rabbit model using functional biomaterials. Sci Rep. 2017;7(1):1–17.
Transplantation. 2007;84(10):1222–32. 79. Numa K, Imai K, Ueno M, Kitazawa K, Tanaka H,
65. Okumura N, Koizumi N. Regeneration of the corneal Bush JD, et al. Five-year follow-up of first 11 patients
endothelium. Curr Eye Res. 2020;45(3):303–12. undergoing injection of cultured corneal endothelial
66. Burillon C, Huot L, Justin V, Nataf S, Chapuis F, cells for corneal endothelial failure. Ophthalmology.
Decullier E, et al. Cultured autologous oral muco- 2021;128(4):504–14.
sal epithelial cell sheet (CAOMECS) transplanta- 80. Marta CM, Adrian M, Jorge FD, Francisco AM,
tion for the treatment of corneal limbal epithelial De Miguel MP. Improvement of an effective pro-
stem cell deficiency. Investig Ophthalmol Vis Sci. tocol for directed differentiation of human adi-
2012;53(3):1325–31. pose tissue- derived adult mesenchymal stem
67. Tuft SJ, Coster DJ. The corneal endothelium. Eye. cells to corneal endothelial cells. Int J Mol Sci.
1990;4(3):389–424. 2021;22(21):11982.
68. Zavala J, López Jaime GR, Rodríguez Barrientos CA, 81. Li Z, Duan H, Jia Y, Zhao C, Li W, Wang X, et al.
Valdez-Garcia J. Corneal endothelium: developmental Long-term corneal recovery by simultaneous deliv-
strategies for regeneration. Eye. 2013;27(5):579–88. ery of hPSC-derived corneal endothelial precur-
69. McCabe KL, Kunzevitzky NJ, Chiswell BP, Xia sors and nicotinamide. J Clin Invest. 2022;132(1):
X, Goldberg JL, Lanza R. Efficient generation of 1–11.
82. Wongvisavavit R, Parekh M, Ahmad S, Daniels a review of stromal replacements. Acta Biomater.
JT. Challenges in corneal endothelial cell culture. 2018;69:31–41.
Regen Med. 2021;16(9):871–91. 85. Lagali N. Corneal stromal regeneration: current sta-
83. Brunette I, Roberts CJ, Vidal F, Harissi-Dagher M, tus and future therapeutic potential. Curr Eye Res.
Lachaine J, Sheardown H, et al. Alternatives to eye 2020;45(3):278–90.
bank native tissue for corneal stromal replacement. 86. Alió del Barrio JL, Arnalich-Montiel F, De
Prog Retin Eye Res. 2017;59:97–130. Miguel MP, El Zarif M, Alió JL. Corneal stroma
84. Matthyssen S, Van den Bogerd B, Dhubhghaill regeneration: preclinical studies. Exp Eye Res.
SN, Koppen C, Zakaria N. Corneal regeneration: 2021;202:108314.
What Is New in Keratoprostheses
37
Saif Bani Oraba and Christopher Liu

• Cadaveric Keratoplasty is the preferred option
for the treatment of corneal opacity, scarring Corneal disease is a major cause of blindness in
and deformation but not in cases of end-stage the world, both in developed and developing
and ocular surface diseases. countries. The preferred option for treatment of
• Keratoprosthesis is the last resort for end- corneal opacity, scarring and deformed cornea is
stage corneal and ocular surface disease and keratoplasty. However, in situations where the
may, in the future, offer an alternative to ocular surface is keratinised, lids and blinking are
cadaveric keratoplasty. defective, or the corneal vascularisation is signifi-
• The main keratoprostheses currently in use are cant, keratoplasty is not an option anymore, and
the Boston KPro Type 1 (BKPro1) and the an alternative to the cornea as an optical system is
osteo-odonto-keratoprosthesis (OOKP). necessary. The last resort in such end-stage cor-
• BKPro1 is indicated for wet blinking eyes, neal and ocular surface diseases is keratoprosthe-
while the OOKP is indicated for dry eyes and ses. Keratoprosthesis surgery and its long-term
those with defective or absent blink or lids. management are very complex and require broad
• Although they may be visually devastating, and extensive multidisciplinary team involve-
the rate of complications of keratoprostheses ment. Although several devices have been devel-
has been significantly reduced by improving oped and trialled, very few have had successful
the design of the devices and the development long-term results and continue in regular clinical
of prevention and management protocols for use [1]. The main current keratoprostheses are
the complications. Boston KPro Type 1 (BKPro1) for wet blinking
eyes and the osteo-odonto-keratoprosthesis
(OOKP) for dry eyes and those with defective or
absent blink or lids.
S. Bani Oraba
Sussex Eye Hospital, Brighton, UK
Ibra Hospital, Ibra, Sultanate of Oman
C. Liu (*)
Brighton and Sussex Medical School, Brighton, UK
Brighton and Tongdean Eye Clinic, Sussex Eye
Hospital, Hove, UK
https://doi.org/10.1007/978-3-031-32408-6_37
500 S. Bani Oraba and C. Liu
Limitations of Keratoplasty treated with debridement of the abnormal epithe-

lial cells, optimizing the ocular surface, use of a
Keratoplasty, both penetrating and lamellar, pro- scleral contact lens or in advance cases, limbal epi-
vides the standard treatment for most corneal thelial transplantation. Although there have been
opacities that jeopardize vision. It is considered several advances in the field in the last 20–30 years,
one of the most successful organ transplanta- the management of LSCD remains a challenge.
tions. It provides an excellent outcome and high These advances include transplantation of the lim-
success rate in low-risk patients. The success rate bal epithelium (autografts and allografts); culti-
of keratoplasty depends on the type, indication vated limbal epithelial transplantation (CLET),
and the period after the procedure. Outcome cultivated oral mucosal epithelial transplantation
analyses for tens of thousands of full-thickness (COMET) and simple limbal epithelial transplan-
and lamellar corneal transplants have consis- tation (SLET) [5]. Systemic immunosuppression
tently demonstrated that long-term functional is required to maintain the survival of allo-
graft survival rates are high for recipients of first transplanted cells, with attendant side effects.
transplants with non-inflammatory corneal dis- Some of the aforementioned limitations of
ease such as keratoconus and corneal dystro- keratoplasty and limbal stem cell transplantation
phies. However, other recipient subgroups can be addressed by keratoprostheses (KPros).
experience substantially poorer long-term out- KPros are last resort operations and should only
comes. Conditions not amenable to corneal trans- be offered to cases not amenable to conventional
plant include cases of chronic ocular surface cadaveric keratoplasty.
inflammation, extensive corneal vascularisation
and multiple graft failure.
The success of keratoplasty is however not History and Development
fully sustained in the long term. In over one thou-
sand penetrating keratoplasty procedures per- The first description of a keratoprosthesis is
formed over 20 years, the transplants remained attributed to the French surgeon Guillaume
clear in only 55.4% of patients at 10 years, 52% Pellier de Quengsy (1789) [6]. He proposed
at 15 years and 44% at 20 years post-surgery [2]. replacing the opaque cornea with a transparent
Another limitation of keratoplasty is the scar- material to restore vision. This was followed by a
city of corneal donor tissues due to multiple rea- number of attempts to develop an ideal kerato-
sons like cultural, religious, and economic prosthesis. Different attempts with various tech-
barriers. The availability of donor tissues contin- niques and designs of keratoprostheses were
ues to be a challenge. A recent global survey of made, all of which failed to have sustainable suc-
eye banking and corneal transplantation quanti- cess. Interest in keratoprostheses faded after the
fied the drastic mismatch between the supply and catastrophic complications due to the absence of
demand of donor corneas worldwide, finding biotechnology and has been replaced by the
only 1 cornea available for every 70 needed [3]. introduction of corneal transplantation. The field
The unavailability of corneal donor tissues has of keratoplasty continued to develop with the
become even more evident in the post-COVID-19 introduction of steroids, fine needles, and suture
pandemic. materials. However, it was eventually realised
Limbal stem cell deficiency (LSCD) is an ocu- that corneal transplantation alone is not a perma-
lar surface disease caused by a decrease in the nent solution for all corneal blindness, and kera-
population and/or function of limbal epithelial toprostheses came into consideration again.
stem cells (LSCs), which leads to the inability to Nussbaum described KPro prototypes manu-
sustain the normal homeostasis of the corneal epi- factured from quartz crystal. It was large and rap-
thelium [4]. This deficiency results in persistent idly extruded. This was followed by the
epithelial defects and conjunctivalisation of the development of smaller devices that were suc-
corneal surface. In mild cases, LSCD may be cessfully implanted in animals and tried in
37 What Is New in Keratoprostheses 501
humans. These initial KPros had a high failure the patient selection process. Often the patient
rate due to infection, leakage, and extrusion of has undergone different treatment modalities
the device [7]. After several years of attempts to which failed before considering a BKPro1. The
develop an ideal KPro, the principle of assem- surgeon should acquire details of the underlying
bling a device where a central optical cylinder diagnosis, current ocular condition, medical sys-
and two plates are assembled into a corneal graft temic and ocular treatment received including
carrier has emerged. This formed the basis to use of steroids or immunosuppression, and types
develop the Boston KPro, until the early twenti- and number of any ocular surgical interventions.
eth century, when Salzer implanted a quartz disc The initial eye examination aims to identify eyes
bounded by a platinum ring with prongs into with good visual potential, with healthy optic
human eyes which lasted a number of years. The nerve function and normal retina. Also, it focuses
next development was to use a lighter, biocom- on the overall health of the ocular surface, the
patible material polymethylmethacrylate amount of ocular surface scarring and keratinisa-
(PMMA). Multiple further stages of design tion and forniceal shortening. Any degree of
refinements and developments in the surgical keratinisation, either bulbar or tarsal, would lead
procedure improved outcomes with reduced to poor results for the BKPro1. The anatomy and
complications. function of the lids, the blinking mechanism and
the quality of tear film should be carefully
assessed along with the ability to apply and
BKPro1 retaining a soft contact lens and the compliance
to topical antibiotics. BKPro1 surgery should not
The BKPro1, made of polymethylmethacrylate be offered for patients in whom keratoplasty car-
(PMMA) and titanium, is currently a widely used ries a good chance of success. Also, it is contrain-
device. It is composed of 2 plates; an anterior dicated in patients with end-stage glaucoma,
5 mm diameter PMMA plate with a 3.5 mm cen- retinal or optic nerve pathology, and when there
tral optical stem and an 8.5 mm titanium back is a seeing fellow eye.
plate with 16 holes to facilitate access of aqueous Evaluation of visual potential is necessary
humour to the sandwiched corneal graft carrier. before offering the option of keratoprostheses.
The corneal graft carrier is then sutured to the This can be started by assessing the light percep-
host cornea akin to a full-thickness corneal graft. tion and projection in all quadrants. Poor accu-
This KPro device does not eliminate the need of racy of light projection may be due to media
corneal tissue, either fresh or frozen. To help opacity rather than retina or optic nerve pathol-
maintain the complex and prevent complications, ogy. B-scan ultrasonography must be performed
an extended-wear therapeutic soft contact lens is to exclude retinal pathology. Electrophysiological
worn, and daily broad-spectrum antibiotics eye tests like electroretinogram and visual-evoked
drops are used. potential could be beneficial in doubtful situa-
tions, but they may also not be precise in quanti-
fying the visual potential in the presence of media
Indications and Pre-operative opacities [8].
Assessment Full glaucoma assessment is crucial as it is
frequently associated with ocular surface dis-
BKPro1 is indicated in cases of multiple graft eases, either as a result of the underlying pathol-
failure and vascularized corneae, with or without ogy with damaged trabeculum, anterior chamber
limbal stem cell failure. The device can only be angle and episcleral venous drainage, or as a side
used successfully when there is an intact blink effect of chronic use of steroids. The diagnosis
mechanism and adequate tear secretion (so-called and management of glaucoma with associated
“wet blinking eye”). Thus, a detailed ocular his- ocular surface diseases may be challenging. A
tory and meticulous eye examination is crucial in thorough clinical examination should be carried
out to identify signs and risk factors, glaucoma- ally presents, especially if the patient has multi-
tous changes and any signs of previous glaucoma system involvement condition. This team may
surgery. It is difficult to accurately measure the consist of anaesthetists, physicians, oculoplas-
intraocular pressure (IOP) and assess the optic tics, vitreoretinal, and glaucoma surgeons. Also,
nerve due to poor fundal view. Hence, the sur- ensuring easy access and clear clinical pathway
geon should use available ancillary tests such as for patients and health care providers in case of
anterior segment optical coherence tomography emergency. Patient support groups, leaflets and
(AS-OCT) or ultrasound biomicroscopy to deter- written information, clear instructions, and edu-
mine the status of anterior segment structures cation of the patients and carers should form part
including irido-corneal angle before planning for of the care pathway [9].
surgery. Signs of poor visual prognosis may be
part of the underlying disease, for example, nys-
tagmus in case of aniridia. On the other hand, Surgical Technique
improvement of vision following previous surgi-
cal intervention may be considered a good sign of The BKPro1 complex is assembled by sand-
visual potential. Although BKPro1 surgery is wiching a double trephine corneal graft carrier
offered usually to bilaterally blind patients, only with the front plate and the locking back plate
one eye should be operated on, leaving the other (Fig. 37.1a), prior to host cornea trephination. In
as spare, given the inherent instability of the more detail, the donor cornea is trephined
device. The other eye should not be neglected, 0.25 mm to 0.5 mm larger than the back plate
and care should be taken to treat glaucoma or any diameter, followed by a punch out of a 3 mm
other condition to preserve the potential vision. central opening using the disposable dermato-
Before taking the decision to offer the patient logical trephine supplied. The doughnut-shaped
or to perform the surgery, it is important to corneal tissue is then placed onto to front plate,
emphasize the need of long-term commitment which itself is resting upside down on an adhe-
from both the patient and surgeon. A holistic sive tape, with the optical cylinder passing
view should be carried out to understand and through the central opening. The fenestrated tita-
address the visual needs, the psychological and nium back plate is then placed on top of the
general health status. Involvement of a clinical donor tissue posteriorly. A titanium locking ring
psychologist is recommended to assess the is snapped around the stem portion of the front
patient’s adaptation to the blindness, current life- plate, which protrudes posteriorly through the
style and coping mechanisms, current employ- cornea and the back plate, thus locking the
ment, and social support in place [9]. The assembly. The latest design of a locking titanium
counselling process will provide the patient the back plate replaces the PMMA back plate and
required information regarding the preparation separate titanium locking ring. The recipient bed
for surgery, the surgery itself and its stages, and is then prepared with trephine smaller than the
the post-surgery treatment and follow-up plan. carrier graft by 0.5 to 1 mm. The KPro assembly
Finally, the team will be able to determine is then sutured to the recipient bed using 16
whether the patient is a good candidate for the interrupted 10–0 nylon sutures (Fig. 37.1b). A
procedure and to ensure that life-long manage- hydrophilic therapeutic (bandage) contact lens is
ment plan is sustainable. The candidates should placed to keep a certain amount of tear film
have realistic expectations and full insight of the intact on the KPro and prevent Dellen formation.
whole process; this includes continued adherence A concurrent glaucoma tube implant can be
to long-term treatment plan with regular life-long implanted if indicated, which could also be done
follow-up with multidisciplinary care. The holis- pre- or post-keratoprosthesis surgery. If indi-
tic approach may include multi-specialty team to cated, pars plana vitreoretinal surgery can be
address the patient’s comorbidities, which usu- performed after KPro surgery.
a b c
Fig. 37.1 Boston Keratoprosthesis type 1: (a) front and back plates, (b) in situ, (c) anterior segment OCT
Results There are very few reports on medium-term

follow-up (2 to 5 years) and almost none on long-
The majority of short-term (0–2 years follow-up) term results (over 5 years). As the lifespan of
outcomes of the BKPro1 is favourable [10]. In patients is usually significantly longer than just
eyes with successful implants, vision is affected by five years, and there is the possibility of losing
a number of factors in the postoperative period visual potential when a device fails, long-term
such as retro-prosthetic membrane (RPM), glau- results are very important.
coma, retinal detachment, endophthalmitis and The incidence of an RPM ranges from 25%–
stromal melts [9]. In one study, the percentage of 65% with the BKPro1s [15]. This proliferation of
eyes with post-operative visual acuity (VA) of fibrovascular tissue over the internal surface of
20/100 or better was 67% (30/45) of patients at the device can occlude the optical portion leading
6 months and 75% (21/28) at 1 year with 90% to visual obstruction and make the ocular exami-
retention rate at 1 year [11]. In a comparative case nation difficult and impossible in some cases.
study involving various international centres with Nearly 45% required treatment with YAG laser or
a cumulative number of 113 procedures, against surgical membranectomy [15]. On histological
110 procedures performed in one of the USA cen- study, it is hypothesized that RPM is derived
tres, 2% of the patients from the international from corneal stromal downgrowth from the host
group and 6% of the patients from the USA group side due to the gaping of the posterior wound
had preoperative visual acuity level of 20/200, beyond the back plate. Also, metaplastic lens epi-
whereas six months postoperatively, 70% of the thelium and native iris stroma contribute to the
international patients and 69% of the USA patients development of RPM [15]. Risk factors for the
have regained a VA level of 20/200 [12]. However, development of RPM include anterior segment
the number of patients maintaining the same level inflammation, previous keratitis, and simultane-
of vision gradually declined over 2 years to 59% ous performance of other intraocular surgery at
and 60%, respectively. Interestingly, the percent- the time of BKPro1 implantation [16].
age of patients with pre- and postoperative VA of BKPro1 implantation is associated with the
less than or equal to light perception did not development of glaucoma and progression of pre-
change significantly in both groups (international: existing glaucoma. Pathophysiology of glaucoma
50% preoperative versus 60% postoperative; USA may include distortion of anterior chamber angle
group: 9% preoperative versus 10% postopera- structures, occurrence of RPM, and peripheral
tive). A device retention rate of 80% at a mean anterior synechiae [17]. The prevalence of glau-
follow-up period of 14 months in the international coma ranges from 36% to 76% in BKPro1 patients
group against a similar retention rate of 80% at an and de novo glaucoma developed in 2%–28% of
average of 24 months in the USA group [12]. In the patients after the device implantation [18]. As
two other studies, retention rates of 100% at previously mentioned, glaucoma detection, moni-
16 months and 95% at an average of 8 months, toring and treatment are considered as a significant
were reported [13, 14]. challenge in BKPro1 patients. The most useful
modality to diagnose and monitor glaucoma pro- ence of a limited field of vision through the optic
gression in the BKPro1 patient may be optic disc make vitreoretinal surgery a daunting task.
photography and OCT imaging with OCT or Despite the existence of various problems,
Heidelberg retinal tomography (HRT). Measuring there has been a steady increase (more than three-
the IOP and conducting a visual field examination fold) in the number of BKPro1 implantations
may be difficult and less accurate. Treatment performed in the USA and the rest of the world
options for glaucoma in BKPro1 patients include [24]. This may be largely due to the increase in
topical and oral glaucoma medications and glau- device retention rates and awareness of the
coma surgery. Out of 45 eyes after Boston BKPro1 procedure.
implantation, 17 eyes needed glaucoma drainage The current BKPro1 design uses a titanium
tube insertion, with an incidence of 59% of con- instead of PMMA back plate (Fig. 37.1). The
junctival erosions following glaucoma tube inserts advantages of using a titanium plate include tak-
in BKPro1 patients [19]. 60% of eyes without ing up less space in the anterior chamber, possi-
“glaucoma device-associated conjunctival ero- bly inducing less inflammation, and a larger
sions” retained a VA of 20/200 and only a 25% of diameter to stem the migration of keratocytes to
the eyes that suffered erosions could retain a VA of form RPM. Moreover, titanium can be coloured
20/200 at 1-year follow-up [19]. The presence of by anodisation to improve cosmesis. Also, the
glaucoma is associated with poor visual prognosis newer click-on design replaced the need for a
but can be ameliorated by prompt pre- and post- locking ring, making the surgery easier. The
operative glaucoma management. Where a glau- Boston KPro team developed another less expen-
coma drainage device is used, the development of sive device, the Lucia. Lucia has a single titanium
erosions and subsequent complications, such as back plate with radial petaloid-shaped holes and
hypotony, endophthalmitis, and choroidal and reti- may be anodised to improve the cosmesis.
nal detachments, may adversely affect the visual LVP keratoprosthesis is a modification of
potential of the eye [17]. Cyclophotocoagulation BKPro1 implanted under buccal mucosal graft in
can be useful in those who do not respond to drain- patients with severely affected ocular surface like
age tubes [20]. Stevens-Johnson syndrome and chemical burns.
Another complication of BKPro1 is endo- In this modification, the optical cylinder is elon-
phthalmitis, with an incidence ranging between gated to protrude through the buccal mucous
0% and 25% with an estimated prevalence of membrane. The initial outcomes of its use,
5.4% in the last 10 years with BKPro1 [21]. The including in paediatric patients are promising
risk of endophthalmitis is generally considered [25]. Boston KPro2 has been similarly modified
higher with inflammatory conditions like to be open through oral mucosal graft instead of
Stevens-Johnson syndrome (SJS), mucous mem- upper lid skin. The main changes are an elon-
brane pemphigoid (MMP), and burns [22]. gated PMMA optical cylinder and titanium sleeve
Although the current standard practice of daily around the cylinder [26].
administration of topical vancomycin has reduced
the incidence of Gram-positive endophthalmitis,
an increased incidence of Gram-negative bacte- OOKP
rial and fungal endophthalmitis is observed by
some investigators [23]. Should endophthalmitis The osteo-odonto keratoprosthesis was first
develops, device explantation followed by vitrec- described in 1963 in Italy by Strampelli, who
tomy and intravitreal injection of broad-spectrum used a donor tooth root and alveolar bone to sup-
antibiotics is advised in view of the high inci- port a PMMA optical cylinder [27]. Falcinelli
dence of posterior segment complications [21]. improved this design by adding certain modifica-
Posterior segment complications like retinal tions such as using a larger biconvex optic and
detachment have been reported in the range of performing cryo-extraction of the lens. This led
3–12% [14]. Altered eye anatomy and the pres- to the modified technique now known as modi-
fied osteo-odonto-keratoprosthesis (MOOKP) Preoperative Assessment

[28]. The central optical cylinder is supported by
the alveo-dental lamina of a single tooth, usually Preoperative assessment of patients for OOKP is
canine. The complex is covered with a buccal similar to that for BKPro1 (please see above) but
mucosa to provide protection and nourishment. there are further specific aims to confirm the suit-
ability of the patient according to the selection
criteria, identifying any risk factors that may
Indications affect the outcome of OOKP and planning to
optimise the eye by managing these risk factors,
The OOKP is indicated in patients with bilat- and to prepare the patient and the family for a
eral blindness from severe, end-stage corneal mostly irreversible and life-changing decision.
and ocular surface diseases with intact retinal The assessment is conducted by a multidisci-
and optic nerve function. Dry eye, keratinisa- plinary team comprising of ophthalmologist, oro-
tion, and any defective blink or lid preclude maxillary surgeon, radiologist, anaesthetist,
success with conventional keratoplasty or ocu- nurses, and a clinical psychologist.
lar surface reconstruction but can be withstood During the ophthalmic assessment, the under-
by the OOKP. Examples of these conditions lying pathology and the indication of the surgery
include eyes with severe SJS, severe ocular is determined, the current status of the eyes is
MMP, severe chemical and thermal burns, and evaluated and any procedure to optimise the eye
eyes that have unsuccessfully undergone ocular is performed before the surgery. A patient-centred
surface or stem cell transplantation. Usually, approach should be followed to choose the eye
the procedure is performed only in one eye, for the proposed surgery, it is generally offered
with the other eye reserved as a spare in case of for bilaterally blind patients and the worse eye is
procedure failure. OOKP is not suitable in chil- usually selected, except when the visual potential
dren due to high bone turnover that may lead to is doubtful.
complete laminar resorption. It is contraindi- The oromaxillary surgeon performs a clinical
cated in phthisis bulbi and eyes without light and radiological assessment, usually by way of
perception. Patients must have intact canine or an orthopantomogram, of the oral cavity and den-
premolar teeth, minimal gum disease, and, tition and based on that, selects the tooth. Patients
preferably, reasonably good dental hygiene in who have poor dental health and oral hygiene are
order to enable a suitable tooth to be harvested counselled towards improving dental health and
[29]. Before the surgery, the patients should be to stop smoking if applicable. For patients who
aware of the gravity of their condition, have an are edentulous or do not have any appropriate
insight of the complexity of the surgery and its teeth, related or unrelated tooth donors may be
potential severe complications, and be prepared considered and screened as required [9].
for life-long follow-up. Also, the issue of Temprano introduced another variation when he
altered cosmetic appearance should be dis- used a fragment of the tibia in a patient who
cussed adequately with the patients and their lacked teeth. The so-called osteo-keratoprosthesis
relatives before the surgery. Patients should be (OKP) presented comparable anatomical and
highly motivated to comply with the long-term visual outcomes, although reabsorption of the
management plan. To establish this, a multidis- bone occurred more frequently, resulting in an
ciplinary approach is ideal, and a clinical psy- increased rate of extrusion of the device [30].
chologist should be part of the team. For The patients’ education about the surgery
patients who are psychologically unstable, do starts as soon as the option of OOKP is discussed
not wish or cannot come for follow-up visits with the patients, their family members, and their
and patients with defective light perception carers. The clinical psychologist will then assess
possibly due to end-stage glaucoma, OOKP is the patients’ perception of the surgery and further
considered relatively contraindicated [8]. investigate their psychological and mental health.
The ability, willingness and motivation for life- rounding jawbone are shaped into a 3 mm thick
long follow up is ensured and stressed on during and up to 15 mm long rectangular lamina. This
the psychological session. Patients may need is used as a skirt or frame to surround an optical
multiple visits to arrive at a decision. Once the cylinder, which is accommodated in a tight, cen-
decision has been taken the patient is then tral and perpendicular tunnel through dentine.
referred for anaesthetic assessment. Dental cement is used to fill any small gaps
between the dentine and optical cylinder; the fit
needs to be snug as the cement is a filler and not
Surgical Technique an adhesive. The PMMA optical cylinder is
made up of an anterior stem that ranges in diam-
OOKP is a multi-staged, complex, and rather eter from 3.5 to 4 mm and a posterior section
invasive surgical programme, which requires ranging from 4.5 to 5.25 mm in width. The ante-
highly subspecialised ophthalmic, dental, clinical rior stem protrudes 2–3 mm beyond the alveolar
psychologist and nursing expertise. Details of the side while the posterior projects into the ante-
OOKP stages and technique are described in the rior chamber [31]. The implant is inserted in a
Rome-Vienna protocol [8]. The OOKP procedure submuscular pouch in the orbito-zygomatic area
is performed in two stages; first preparing the on the contralateral side with the dentine facing
globe, the buccal mucous membrane graft and the orbit and the bone facing the periorbital
the osteo-odonto-acrylic lamina, and second muscles. It is kept there for about 3 months to
implanting the lamina. enhance revascularisation of the implant, pro-
Stage 1 OOKP surgery (Fig. 37.2) is further mote growth of connective tissue and remaining
divided into two stages. Initially, the ocular sur- periosteum.
face is prepared by performing a 360-degree In Stage 2 OOKP surgery (Fig. 37.3), the
conjunctival peritomy, followed by a superficial osteo-odonto-acrylic lamina is retrieved and
keratectomy to remove the corneal scars and examined for signs of absorption and infection,
epithelium. The eye is then covered with a buc- and excess fibrovascular and connective tissue is
cal mucous membrane graft which is sutured to removed. The healthy lamina is then implanted at
recti insertion sites (deriving a blood supply the anterior surface of the globe under the muco-
from the anterior ciliary arteries) and sclera. A sal membrane. This is achieved by creating a
single-rooted tooth (usually a canine, alterna- large buccal mucous membrane flap to expose
tives are incisors and premolars) with the largest the cornea. A Flieringa ring is used to support the
and straightest root is harvested en bloc with the sclera. The centre of the cornea is marked and
surrounding jawbone. The tooth root and sur- trephined with the same diameter as the posterior
Fig. 37.2 Harvesting the tooth root and surrounding jawbone, and preparation of the lamina
Fig. 37.3 Stage 2 OOKP: retrieval and implantation of the lamina
part of the optical cylinder, avoiding decentration lar pressure. The clinical examination aims to
as it may result in visual distortion. Total iridodi- evaluate the visual acuity, estimate the intraocu-
alysis, cryoextraction of lens and a generous lar pressure by palpation with a cotton-tip and or
open sky vitrectomy are performed. The lamina fingertip, check the stability and clarity of the
is sutured to the sclera and the remaining cornea optic cylinder and the state of the mucous mem-
and is covered by the flap of the oral mucosa. brane and the thickness of the lamina, and assess
Postoperative treatment includes topical and sys- the retina and the optic nerve. A B-scan is carried
temic antibiotics, five days of oral steroids and out as necessary. A CT-scan of the lamina with or
oral antiglaucoma medications. Topical broad- without 3-D rendition is done soon after Stage 2
spectrum antibiotics must be applied for the as a baseline and then every few years, as guided
patient’s lifetime with a rotation of antibiotics by clinical examination, to estimate the bone and
every few months, whereas oral antibiotics can dentine.
be stopped after five days. Patients may be
advised to wear dark glasses to improve cosmesis
and reduce glare, some may wear a hat too to Results
reduce glare. Long-term follow-up requires the
OOKP patient to be examined by an OOKP- The visual acuity of patients following OOKP
experienced ophthalmologist every 3 months. surgery can be as good as 6/4. A systematic
The patient is assessed for any signs or symptoms review of eight different case studies found VA of
of infection, retinal detachment or high intraocu- ≥6/18 in 52% of patients after OOKP [32].
Another study recorded a visual acuity of Oral acetazolamide and sublingual timolol eye
≥6/12 in 53% of all OOKP patients, and 78% of drops is the mainstay of glaucoma management
patients achieved a VA of ≥6/60 [33]. in OOKP eyes. Surgical treatment is usually by
Long-term anatomical retention of the OOKP way of drainage tubes. Glaucoma remains a chal-
is excellent. The probability of retaining laminar lenging condition to manage in OOKP.
autografts over 5 years is found to be 81% [33]. Eyelid and mucosal complications are com-
In 85 patients, the retention of the lamina over a mon after OOKP and make up the bulk of the
20-year follow-up was reported to be 98% [34]. surgical revisions after each stage of the proce-
10-year anatomical survival of 145 OOKP and 82 dure. Mucosal thinning and ulcerations were
tibial KPro implants was 66% and 47%, respec- common after both Stage 1 and Stage 2 due to
tively [30]. The main factor resulting in anatomi- inadequate vascularisation and lubrication. This
cal failure was the resorption of the OOKP can be managed by mucosal grafting, which itself
lamina. Resorption leads to decreased thickness may create relative eyelids shortening and mal-
and defects in the lamina, which may in turn lead position necessitating surgical repair [35]. On the
to optical cylinder tilt, aqueous leak and endo- other hand, mucosal overgrowth concealing the
phthalmitis. Lamina resorption can be detected optic cylinder is a common complication that
even in its early stages by clinical palpation in requires excision (Fig. 37.4), with the use of
experienced hands. Radiological studies to detect mitomycin-C in case of recurrence [35].
minor reduction of the laminar dimensions and A retroprosthetic membrane (RPM) is a fibro-
early laminar resorption may include electron vascular proliferation behind the lamina that can
beam tomography (EBT) and CT-scan with or grow across the optic, obscuring vision. A RPM
without 3-D rendition. Radiological studies is usually treatable with an Nd-YAG laser, similar
should be correlated with the clinical assessment to capsulotomy in the early stages. However, it
for a full evaluation of resorption. If progressive may recur and can be difficult to laser, which
or pathological resorption is detected, alendronic may carry risks of optic spoliation and intraocu-
acid is prescribed for the patient. Bone morpho- lar haemorrhage.
genetic protein (BMP) and bone graft can be used New vitreous haemorrhage long after Stage 2
independently. However, in severe cases of is a worrying complication. It can denote a retinal
resorption, they are combined. If the above meth- tear due to a posterior vitreous detachment. Such
ods fail, then laminar replacement is required, eyes must be examined with B-Scan ultrasonog-
and in cases of imminent danger of endophthal- raphy by an experienced operator. Patients must
mitis, the lamina should be explanted and the cor- be very closely monitored with serial B-scans in
neal opening closed with a small full-thickness the absence of a retinal detachment. Endoscopic
corneal graft. If another suitable canine is avail- vitrectomy should be performed in case of sus-
able, a new lamina can be created and exchanged pected or confirmed retinal detachment.
three months later after it has gained soft tissues Endophthalmitis may develop secondary to
in a submuscular pocket (as above). laminar resorption or intraocular surgery. It may
The main complication that affects the visual also result from loose optical cylinder and leak-
outcome in anatomically successful OOKP is age due to laminar resorption. This is managed
glaucoma. Glaucoma was observed in 26% of the by taking samples for culture and sensitivity fol-
eyes before OOKP, and in 60% of the eyes after lowed by injection of intravitreal and or systemic
OOKP [1]. Digital (fingertip) estimation is the antibiotic and or antifungal, and in some cases
only usable method for the estimation of IOP, vitrectomy.
which requires user experience and training. Potential complications during and after Stage
Clinical optic nerve head assessment by fundos- 1 may include impending and inadvertent perfo-
copy, serial photography, optical coherence ration of thin cornea requiring a lamellar or full-
tomography and periodic visual field testing are thickness graft from a donor cornea and buccal
useful in glaucoma evaluation and monitoring. mucosa graft necrosis requiring the additional
a b
Fig. 37.4 Overgrowth of mucous membrane (a) pre- and (b) post-surgical excision
grafting procedure. During Stage 1 oromaxillary Although the general concept of the surgical
fistulae, fractures of the mandible, and damage to technique of OOKP has not changed, certain
adjacent teeth may occur. Also, excessive force surgical aspects have been refined. The optical
or overheating during drilling can break the den- cylinder has been modified to improve the visual
tine or damage the dentoalveolar ligament. When field and reduce the glare. Also, the use of bone
the implant is in the submuscular pouch, absorp- morphogenetic protein for the augmentation of
tion of the dentine, bone infection, or loosening the resorbed lamina. There is emerging evidence
of the optical cylinder can occur. that surgical management of glaucoma, namely
The rare devastating expulsive suprachoroidal tube shunt operations in OOKP patients is more
haemorrhage may occur during Stage 2. The risk effective than topical and systemic medical
of this complication may be reduced by control- treatment. The endoscopic vitrectomy is the
ling high intraocular ocular pressure and blood modality of choice in repairing retinal
pressure prior to and during surgery. Positioning detachment.
the patient in a head-up position, using deep Future areas of keratoprotheses development,
anaesthesia with full paralysis and hyperventila- aiming to prolong the survival time of kerato-
tion are some intraoperative strategies to reduce prostheses and improve the quality and field of
the risk of suprachoroidal haemorrhage. vision, may include biosynthetic or totally syn-
Intravitreous haemorrhage may occur, and it is thetic versions. 3D printing is promising in man-
usually self-limiting. Early postoperative compli- ufacturing corneal stromal tissue equivalent with
cations include low IOP causing choroidal embedded human endothelial cells, with the
detachment. potential of producing full thickness, multilayer
corneal model in future. The OOKP procedure
will evolve to utilise a lamina made of synthetic
The Future materials that would allow the integration of buc-
cal mucous membrane but be more resistant to
The BKPro1 needs to be more affordable, espe- resorption. Biosynthetic or synthetic OOKP
cially for developing countries. More attention would offer a solution, especially for patients
should be paid to the long-term outcome of this who lack suitable teeth.
device. This will help guide its use with improve- On the contrary, advancements in other fields of
ments in retention and prevention, diagnosis, and ophthalmology may decrease the need for any KPro.
management of complications. BKPro2 may Community-based prevention implementation will
have reached its end point. It has been modified decrease corneal blindness because of infectious
to be implanted under buccal mucosa rather than keratitis and chemical burn. A better understanding
that under lid skin [26]. of corneal immunological mechanisms may lead to
more effective strategies for increased survival of 8. Hille K, Grabner G, Liu C, et al. Standards for modi-
fied osteo-odonto-keratoprosthesis (OOKP) surgery
multiple graft failure and control of immunological according to Strampelli and Falcinelli: the Rome-
eye diseases such as SJS and MMP. Vienna protocol. Cornea. 2005;24:895–908.
9. Avadhanam VS, Smith HE, Liu C. Keratoprostheses
Take Home Notes for corneal blindness: a review of contemporary
devices. Clin Ophthalmol. 2015;9:697–720. Published
• In unilateral corneal blindness, keratoprosthe- 2015 Apr 16. https://doi.org/10.2147/OPTH.S27083.
ses should not be offered, and in the case of 10. Holland G, Pandit A, Sánchez-Abella L, Haiek A,
bilateral corneal blindness, only one eye Loinaz I, Dupin D, Gonzalez M, Larra E, Bidaguren
should be rehabilitated, keeping the other eye A, Lagali N, Moloney EB, Ritter T. Artificial cor-
nea: past, current, and future directions. Front
as spare. Med. 2021;8:770780. https://doi.org/10.3389/
• Keratoprostheses should only be offered by fmed.2021.770780.
multidisciplinary teams in regional and 11. Aldave AJ, Kamal KM, Vo RC, Yu F. The Boston type
national centres, which can also provide emer- I keratoprosthesis: improving outcomes and expand-
ing indications. Ophthalmology. 2009;116(4):640–51.
gency access 24/7/365. 12. Aldave AJ, Sangwan VS, Basu S, et al. International
• While OOKP eliminates the need for any cor- results with the Boston type I keratoprosthesis.
neal tissue, BKPro requires a full thickness of Ophthalmology. 2012;119(8):1530–8.
corneal tissue. 13. Chew HF, Ayres BD, Hammersmith KM, et al.
Boston keratoprosthesis outcomes and complications.
• There has been a steady increase in BKPro1 Cornea. 2009;28:989–96.
implantation due to a high device retention 14. Zerbe BL, Belin MW, Ciolino JB. Boston type 1 kera-
rate and awareness of the procedure. toprosthesis study group results from the multicenter
• The visual acuity outcome and the long-term Boston type 1 keratoprosthesis study. Ophthalmology.
2006;113:1779–84.
retention rate of OOKP are excellent. 15. Stacy RC, Jakobiec FA, Michaud NA, Dohlman CH,
Colby KA. Characterization of retrokeratoprosthetic
membranes in the Boston type 1 keratoprosthesis.
Arch Ophthalmol. 2011;129(3):310–6.
References 16. Magalhães FP, Sousa LB, Oliveira LA. Boston
type I keratoprosthesis: review. Arq Bras Oftalmol.
1. Gomaa A, Comyn O, Liu C. Keratoprostheses 2012;75(3):218–22.
in clinical practice—a review. Clin Exp 17. Kamyar R, Weizer JS, de Paula FH, et al. Glaucoma
Ophthalmol. 2010;38:211–24. https://doi. associated with Boston type I keratoprosthesis.
org/10.1111/j.1442-9071.2010.02231.x. Cornea. 2012;31(2):134–9.
2. Anshu A, Li L, Htoon HM, de Benito-Llopis L, 18. Banitt M. Evaluation and management of glau-
Shuang LS, Singh MJ, et al. Long-term review of coma after keratoprosthesis. Curr Opin Ophthalmol.
penetrating keratoplasty: a 20-year review in Asian 2011;22:133–6.
eyes. Am J Ophthalmol. 2020;224:254–66. https:// 19. Li JY, Greiner MA, Brandt MC, Lim MC, Mannis
doi.org/10.1016/j.ajo.2020.10.014. MJ. Long-term complications associated with glau-
3. Gain P, Jullienne R, He Z, et al. Global survey of coma drainage devices and Boston keratoprosthesis.
corneal transplantation and eye banking. JAMA Am J Ophthalmol. 2011;152(2):209–18.
Ophthalmol. 2016;134(2):167–73. 20. Rivier D, Paula JS, Kim E, Dohlman CH, Grosskreutz
4. Deng SX, Borderie V, Chan CC, et al. And the inter- CL. Glaucoma and keratoprosthesis surgery: role
national Limbal stem cell deficiency working group. of adjunctive cyclophotocoagulation. J Glaucoma.
Global consensus on definition, classification, diagno- 2009;18:321–4.
sis, and staging of limbal stem cell deficiency. Cornea. 21. Robert MC, Moussally K, Harissi-Dagher M. Review
2019;38(3):364–75. of endophthalmitis following Boston keratoprosthesis
5. Fernandez-Buenaga R, Aiello F, Zaher SS, et al. type 1. Br J Ophthalmol. 2012;96(6):776–80.
Twenty years of limbal epithelial therapy: an update 22. Nouri M, Terada H, Alfonso EC, Foster CS, Durand
on managing limbal stem cell deficiency. BMJ ML, Dohlman CH. Endophthalmitis after keratopros-
Open Ophthalmol. 2018;3:e000164. https://doi. thesis: incidence, bacterial causes, and risk factors.
org/10.1136/bmjophth-2018-000164. Arch Ophthalmol. 2001;119(4):484–9.
6. de Quengsy GP, Des HD. Pellier de Quengsy 23. Durand ML, Dohlman CH. Successful prevention
Sammlung von Aufsätzen und Wahrnehmungen of bacterial endophthalmitis in eyes with the Boston
sowohl über die Fehler der Augen, als der Theile, die keratoprosthesis. Cornea. 2009;28(8):896–901.
sie umgeben Junius. Leipzig: Junius; 1789. 24. Boston KPro News. Fall. 2011. 8. Accessed 23 Jan
7. Lam FC, Liu C. The future of keratoprostheses (arti- 2013. www.masseyeandear.org.
ficial corneae). Br J Ophthalmol. 2011;95:304–5. 25. Basu S, Nagpal R, Serna-Ojeda JC, Bhalekar S,
https://doi.org/10.1136/bjo.2010.188359. Bagga B, Sangwan V. LVP keratoprosthesis: ana-
tomical and functional outcomes in bilateral end- 31. Zarei-Ghanavati M, Avadhanam V, Vasquez Perez A,
stage corneal blindness. Br J Ophthalmol. 2018. pii: Liu C. The osteo-odonto-keratoprosthesis. Curr Opin
bjophthalmol-2017-311649. Ophthalmol. 2017;28:397–402.
26. Mehran Z-G, Liu C. Keratoprosthesis: current choices 32. Tan A, Tan DT, Tan XW, Mehta JS. Osteo-odonto ker-
and future development. Asia Pac J Ophthalmol. atoprosthesis: systematic review of surgical outcomes
2019;8(6):429–31. https://doi.org/10.1097/ and complication rates. Ocul Surf. 2012;10:15–25.
APO.0000000000000268. https://doi.org/10.1016/j.jtos.2012.01.003.
27. Strampelli B. Keratoprosthesis with osteodontal tis- 33. Liu C, Okera S, Tandon R, Herold J, Hull C,
sue. Am J Ophthalmol. 1963;89:1029–39. Thorp S. Visual rehabilitation in end-stage inflam-
28. Falcinelli GC. Personal changes and innovations in matory ocular surface disease with the osteo-
Strampelli's osteo-odonto-keratoprosthesis. An Inst odontokeratoprosthesis: results from the UK. Br
Barraquer. 1998;28:47–8. J Ophthalmol. 2008;92:1211–7. https://doi.
29. Tan A, Tan DT, Tan X-W, Mehta JS, org/10.1136/bjo.2007.130567.
Keratoprosthesis O-o. Systematic review of sur- 34. Marchi V, Ricci R, Pecorella I, Ciardi A, Di Tondo
gical outcomes and complication rates. Ocul Surf. U, Osteo-odonto-keratoprosthesis. Description of sur-
2012;10(1):15–25. gical technique with results in 85 patients. Cornea.
30. Michael R, Charoenrook V, de la Paz MF, Hitzl W, 1994;13:125–30.
Temprano J, Barraquer RI. Long-term functional 35. Avadhanam VS, Vasquez-Perez A, Chervenkoff
and anatomical results of osteo- and osteoodonto- JV, El-Zahab S, Liu C. Mucosal complications in
keratoprosthesis. Graefes Arch Clin Exp Ophthalmol. osteo-odonto keratoprosthesis (OOKP) surgery. J
2008;246:1133–7. https://doi.org/10.1007/ EuCornea. 2020;6:13–23. https://doi.org/10.1016/j.
s00417-008-0850-3. xjec.2020.01.001.
Intraoperative OCT in Anterior
Segment Surgery
38
Francis W. Price Jr, Anjulie Gang,
and Marianne O. Price
Key Points Prior to its introduction, fluorescein angiography

• Intraoperative OCT allows direct visualization was needed to diagnose cystoid macular edema;
of DMEK graft orientation, eliminating the this was much more invasive than taking an OCT
need for orientation marks on the donor tissue. image of the macula. The in-office OCT also
• Intraoperative OCT can evaluate graft orienta- allows us to easily screen candidates for multifo-
tion and placement through cloudy corneas. cal intraocular lenses (IOL) and to rule out epiret-
• Intraoperative OCT allows direct visualization inal membrane or other macular pathology,
of the depth and uniformity of a DALK dis- because the OCT shows cellular structures in
section plane. exquisite detail. Intraoperative OCT has been an
• In complex DSEK or DMEK cases, important advance because it provides the sur-
Intraoperative OCT allows the surgeon to scan geon with detailed cross-sectional images of ocu-
the interface for fluid pockets and loose resid- lar structures not well visualized with the coaxial
ual tags of stroma, Descemet membrane, or microscope. This chapter will describe the uses
other tissue. of intraoperative OCT in anterior segment
surgeries.
We know of three companies that market or
Introduction have sold intraoperative OCT units with their
operating microscopes. Each company’s device
In-office optical coherence tomography (OCT) only works with its own microscope, which lim-
has dramatically changed the diagnosis and treat- its adoption, because one has to purchase the
ment of cataracts and other ocular conditions. OCT device plus the operating microscope that is
made to utilize it. This limitation has provided a
Supplementary Information The online version con- barrier to entry for those who would like to utilize
tains supplementary material available at https://doi. this technology. The three companies that have
org/10.1007/978-3-031-32408-6_38.
OCT units in practice are Carl Zeiss Meditec,
Leica, and Haag-Streit, but Haag-Streit has dis-
F. W. Price Jr (*) · A. Gang
Price Vision Group, Indianapolis, IN, USA continued its OCT unit.
e-mail: [email protected]; There are three ways the OCT images can be
[email protected] viewed in the operating room: on a monitor
M. O. Price hooked up to the OCT device, as an inset in the
Cornea Research Foundation of America, video recording from the operating microscope,
Indianapolis, IN, USA and through the oculars of the operating micro-
https://doi.org/10.1007/978-3-031-32408-6_38
514 F. W. Price Jr et al.
a b
Fig. 38.1 Three ways to view intra-operative OCT geon through the oculars of the operating microscope (a
images in the operating room. (a) With the Haag-Streit unique feature of the Haag-Streit device). (b) The image
iOCT™, the OCT image can be viewed on a television seen by the surgeon through the oculars of the operating
screen as an inset within the surgical video feed (top left; microscope. This image was captured with an iPhone
this view is provided by all commercially available intra- camera, which made the OCT image appear to be in color,
operative OCT devices), or it can be viewed on the moni- although it is actually in greyscale. The surgeon can use
tor used to program the OCT which is attached to the the foot switch of the microscope to turn on or off the
microscope (lower right), or it can be viewed by the sur- OCT image superimposed into the oculars
scope (Fig. 38.1). To our knowledge, only the desired by the surgeon. We use the iOCT™ most
Haag-Streit device provides a sufficiently high- frequently in DMEK cases for the treatment of
resolution image superimposed in the oculars to Fuchs’ dystrophy. In these cases, the cornea is
allow the surgeon to operate without having to relatively clear, so we turn the OCT image on
look away from the surgical field to view a moni- when the donor tissue is being inserted into the
tor. With the other two devices, the surgeon either recipient's anterior chamber, and after confirming
has to look away from the surgical field to view that the tissue is correctly oriented with the donor
the monitor or they could ask someone else in the endothelium facing the recipient iris, we turn the
room to relay to them what is visible on the moni- OCT image off. By using the iOCT™ to deter-
tor. These later two devices provide very nice mine whether the tissue is correctly oriented, we
images for teaching and documentation but are do not have to place orientation marks on the tis-
not as efficient in surgery as the unit that allows sue, which can cause endothelial cell damage.
the surgeon to view the image superimposed in Gentian violet marks on a DMEK graft typically
the oculars. We have the Haag-Streit device, so cause about 5% cell loss, and any notches or slits
our images and discussion are based on experi- made along the edge of the tissue to show orien-
ence with that device, the iOCT™. tation also cause cell loss.
In eyes with more advanced corneal decom-
pensation, which makes it more difficult for the
Intraoperative OCT and EK surgeon to visualize the posterior cornea, we use
the iOCT™ while stripping the host Descemet
The iOCT™ image superimposed in the oculars membrane (DM) to look for loose strands of
(Fig. 38.2) can be turned on and off with the foot DM, stroma, or scar tissue in the anterior cham-
switch, so the OCT image is only visible when ber or for abnormalities on the back surface of
38 Intraoperative OCT in Anterior Segment Surgery 515
a b
c d
Fig. 38.2 Use of intraoperative OCT with the DALK taken while peeling, the residual bed appears thinner on
peeling technique. The intraoperative OCT helps the sur- the right and thicker on the left, but the left still has areas
geon visualize the dissection depth and adjust as needed of injected air and stretched stromal fibers indicating that
before starting the peel technique. Typically the desired it has not been completely peeled. After peeling is com-
dissection depth is almost down to the Descemet mem- pleted, the bed will be uniformly thin. It is important for
brane, although a thicker residual bed is safer in cases of the surgeon to only grasp one spot on the cornea while
previous hydrops or perforation. (a)The dissection plane pulling the stroma away from the bed. (d) In a different
is not deep enough and has not reached the desired depth case, note the uniform and very thin residual bed achieved
of the residual bed. (b) The appropriate dissection depth is with the peeling technique. This residual bed is very simi-
now reached, as indicated by the arrow. (c) In this image lar to that achieved with a Type 1 big bubble
the cornea which could interfere with donor

placement or attachment. We can also use the Intraoperative OCT and DALK
iOCT™ to help center the graft if it is hard to see
the graft edges through the recipient cornea. This The iOCT™ greatly improved the efficiency of
improved visualization allows us to successfully our EK surgery and has been even more impact-
use thinner DMEK grafts instead of thicker ful with DALK. The big bubble (BB) technique
DSEK grafts in cases with very cloudy corneas. introduced by Anwar was a key DALK advance
Once the EK graft is in place, the OCT allows [1], because it provided a consistent and uni-
the surgeon to evaluate the apposition of the formly thin dissection bed that produced better
donor against the recipient cornea to make sure visual outcomes than manual dissection, which
that there are no areas of detachment or obstruc- often resulted in inconsistent bed depth and irreg-
tions preventing attachment, such as loose DM, ularities across the dissection plane. Malbran had
stromal strands, iris, or scar tissue. The chapter earlier introduced manual peeling techniques [2,
entitled Endothelial Keratoplasty: Current State 3], but determining the dissection and peeling
of the Art includes a video demonstrating the use depth was difficult with a coaxial microscope.
of the iOCT™ while positioning a DMEK graft Melles et al. introduced the use of an air bubble
and confirming its orientation in an eye with a in the anterior chamber to determine dissection
very cloudy cornea. depth [4], but it was difficult to keep the manual
516 F. W. Price Jr et al.
dissection depth uniform with his technique, and the anterior chamber and venting fluid from the
in cases of advanced keratoconus, it was chal- interface, and the OCT is helpful for assessing
lenging to get the dissecting blade across and when the double anterior chamber has fully col-
over the apex of the large cone. Video 38.1 shows lapsed. It is surprising how much larger and
Type 1 and Type 2 big bubbles. Video 38.2 shows extensive the fluid pockets appear when viewed
an irregular bed produced with manual dissec- with the OCT intraoperatively.
tion. In both videos, the ability to scan the bed As we do more DALK procedures and the
with the intraoperative OCT is demonstrated. patients get older and develop cataracts, we have
Because the intraoperative OCT allows direct found that occasionally a double anterior cham-
viewing of the lamellar structures of the cornea ber may form between the donor graft and the
and the ability to estimate the depth of dissection, residual stromal bed when cataract incisions are
we now have the ability to perform DALK dis- hydrated to close them at the end of the case. It is
sections without a BB that postoperatively are difficult to appreciate the extent of the double
indistinguishable from BB cases. This is very anterior chamber or its resolution with the coax-
helpful for cases where either a BB is difficult to ial microscope. Video 38.3 shows such a case and
achieve or is not desirable to try. In eyes with pre- demonstrates the use of the iOCT™ to diagnose
vious hydrops, penetrating scars, or previous and treat the separation of a DALK graft from the
cataract or anterior segment surgery, it can be dif- host residual bed during subsequent cataract
ficult to successfully create a BB without rupture surgery.
of the previous perforations. Our current tech-
nique is to make a side cut with a calibrated tre-
phine or femtosecond laser zig-zag incision. Intraoperative OCT and Lens
After making the side cut, we begin a lamellar Implants
dissection inward from the side cut and check the
depth of dissection, then gradually deepen the During cataract surgery, the iOCT™ can be used
depth until we reach the desired residual bed to assess capsular bag anatomy, lens tilt, and
thickness, which is typically close to a BB depth placement of the IOL relative to the capsule.
unless we want a thicker bed because of scarring. Also, we routinely use the iOCT™ to measure
Once the desired depth is reached, we dissect the vault of an ICL implant over the crystalline
centrally for 1–2 mm, 360 ° around the side cut, lens. The final vault after the viscoelastic has dis-
and then perform a peel as described by Malbran. sipated often varies from the vault measured
Figure 38.1 shows the peeling technique with the intraoperatively, but the intraoperative assess-
use of the iOCT™ to guide the depth of the dis- ment is helpful for identifying whether the vault
section; the residual bed after a successful peel is substantially over or under what was planned.
resembles that achieved with a type 1 big We have used the vault measurements to decide
bubble. whether to rotate a lens with too much vault from
We have not found any difference in the inci- a horizontal position to a vertical position or
dence of stromal rejection episodes between when deciding whether to change the size of the
DALK procedures performed with BB vs. peel lens for the second eye when doing bilateral
techniques, but we did find that the rejection epi- simultaneous cases. It is important to perform all
sode rate was significantly lower with the use of measurements at the same magnification.
femtosecond laser zig-zag side cut incisions
compared with standard metal trephination [5].
A double anterior chamber can form during Retained Nuclear Fragments
DALK cases from perforation during dissection
or suturing. Intraoperative OCT is very helpful Video 38.4 shows the use of the iOCT™ to facili-
for identifying when a double anterior chamber tate the removal of a nuclear fragment. This
has formed. Treatment involves injecting air into patient presented to our clinic with inferior cor-
38 Intraoperative OCT in Anterior Segment Surgery 517
neal decompensation a few months after alerting them to potential problems in DMEK
undergoing uncomplicated cataract surgery with or DSEK.
the placement of an intraocular lens. Notes from • The OCT reveals the depth and uniformity of
the patient’s surgeon indicated that he had the dissection plane in DALK and shows
removed a nuclear fragment from the eye several whether there is a double anterior chamber or
days after the cataract surgery. Upon taking the separation of the donor and recipient bed.
patient to surgery, the corneal edema and iris • IOL placement relative to the bag and residual
color made it difficult to detect a residual nuclear capsule can be evaluated.
fragment, but we were able to find it and properly • ICL vault can be measured, allowing the sur-
direct the phaco tip to remove it with the use of geon to identify cases with unusually high or
the iOCT™. low vaults.
Other Uses References
Intraoperative OCT can be used, just like in- 1. Anwar M, Teichmann KD. Big-bubble technique to
bare Descemet’s membrane in anterior lamellar kera-
office OCT, to evaluate the depth of scars in the toplasty. J Cataract Refract Surg. 2002;28:398–403.
cornea or lesions on the conjunctiva. 2. Malbran E, Stefani C. Lamellar keratoplasty in corneal
ectasias. Ophthalmologica. 1972;164:50–8.
Take Home Notes 3. Malbran E. Lamellar keratoplasty in keratoconus. Int
Ophthalmol Clin. 1966;6:99–109.
• Intraoperative OCT opens up a whole new 4. Melles G, Lander F, Rietveld F, et al. A new surgical
view of the anatomical structures of the ante- technique for deep stromal, anterior lamellar kerato-
rior segment and cornea. plasty. Br J Ophthalmol. 1999;83:327–33.
• It allows a more efficient assessment of graft 5. Price FW Jr, Price MO, Grandin JC, Kwon R. Deep
anterior lamellar keratoplasty with femtosecond-
orientation with DMEK surgery. laser zigzag incisions. J Cataract Refract Surg.
• Irregularities on the posterior corneal surface 2009;35:804–8.
are easily visualized real-time by surgeons
Epilogue: Corneal Graft Surgery,
a Glance to the Future
39
Jorge L. Alió and Jorge L. Alió del Barrio
Throughout this book, we have witnessed the quent, with reported levels of survival from 52%
major evolution that corneal graft surgery has to 98.8% for penetrating keratoplasty at 10 years,
experienced over the last two decades. A better from 77% to 99.3% for deep anterior lamellar
understanding of the corneal anatomy and physi- keratoplasty at 5 years, from 56% to 94.1% for
ology, technical improvements in the manage- Descemet stripping endothelial keratoplasty at
ment of corneal bank tissue, improvements in 5 years and from 90% to 97.4% for Descemet
surgical instrumentations (such as the availability membrane endothelial keratoplasty at 5 years [1].
of femtosecond laser), new surgical techniques The main pitfalls are immune graft rejection,
that have emerged and have finally been consoli- comorbidities and relapse of the previous dis-
dated as better options to the classical penetrating ease. In addition, functional failures, not fre-
keratoplasty with better results, medical educa- quently estimated as real ones, happen in a
tion and, above all, the skills and the talent of cor- considerable number of patients, especially in
neal surgeons, have made corneal surgery enter a PKP (for example, 10% recurrence of the ectatic
final stage of development since its early begin- disease at the host remnant peripheral cornea
nings with the description of PKP by Zirm in after 20 years), leading to a lack of adequate gain
1906 and popularized by Castroviejo in 1936. of visual acuity [2]. Targeting the control and
Over all these years, the evolution has been con- resolution of these problems, especially immune
stant and always in the benefit of better tech- graft rejection, is mandatory and one of the real
niques, better results and better solutions to challenges that the modern corneal graft surgeons
corneal blindness. face. However, this is not going to be enough, as
However, even though the results have been functional failures still influence the outcomes,
widely implemented, we have seen in the early and they are not always within the surgeon’s con-
chapters of this book how corneal graft still offers trol. So, it seems mandatory to move to a totally
a challenge. Anatomical success does not always different model, a paradigm shift. The philoso-
happen and anatomic failures are relatively fre- pher Thomas Kuhn defined a paradigm shift as
needing to happen first in the mind of the
decision-makers in that particular topic [3]. This
is exactly what has to happen now in corneal sur-
J. L. Alió (*) · J. L. A. del Barrio gery; we need a paradigm shift.
Vissum Miranza, Miguel Hernández University, The new paradigm will be, instead of tissue
Alicante, Spain replacement, tissue restoration by regeneration.
e-mail: [email protected] Corneal regeneration has also been targeted in
https://doi.org/10.1007/978-3-031-32408-6_39
520 J. L. Alió and J. L. A. del Barrio
this book, and it is still in its early stages. The Corneal graft surgery, as it is practiced today, it
possibility of restoring the ocular surface has will never disappear and indeed will continue
been a highlight. The challenges that are involved evolving, but it may be finally largely substituted
are clear and there is still time ahead to achieve by corneal regeneration procedures.
this on a consistent and repeatable basis. We hope that the reader of this book has fore-
Translational research in this area is evolving, seen the present and the future of corneal surgery,
and in the coming years, it will probably take an and this book will contribute to the medical edu-
important role in our practice once costs associ- cation of those corneal surgeons interested in cor-
ated with it fade, allowing their availability to be neal graft surgery and stimulate them to search
more general. Corneal stroma regeneration, for new innovative and better solutions for cor-
according to our recent pioneer clinical studies, neal blindness in the benefit of our patients.
seems to be feasible and even more easily appli-
cable with the use of xenogenic tissue (instead of
human corneal tissue, that is so scarce and expen- References
sive) [4–7]. Corneal endothelial substitution
seems to be feasible even though the lack of bio- 1. Alio JL, Montesel A, El Sayyad F, Barraquer RI,
Arnalich-Montiel F, Alio Del Barrio JL. Corneal
logical productivity of the corneal endothelium graft failure: an update. Br J Ophthalmol.
seems to be a limitation during the culture of 2021;105(8):1049–58. https://doi.org/10.1136/
these cells. Future clinical research, both basic bjophthalmol-2020-316705.
and translational, will likely increase the effi- 2. de Toledo JA, de la Paz MF, Barraquer RI, Barraquer
J. Long-term progression of astigmatism after pen-
ciency and involved costs of such procedures in etrating keratoplasty for keratoconus: evidence of late
order to succeed in real clinical practice. We can recurrence. Cornea. 2003;22(4):317–23.
clearly foresee that it, in the future eye cellular 3. Kuhn T. The structure of scientific revolutions. 2nd ed.
eye banks, containing the best donor lineages for Chicago, IL: University of Chicago Press; 1970. ISBN
978-0-226-45804-5.
each cell type, will centralize the production and 4. Alió Del Barrio JL, Arnalich-Montiel F, De Miguel
delivery of the different stem cell regenerative MP, El Zarif M, Alió JL. Corneal stroma regeneration:
products among all clinical centers, making these preclinical studies. Exp Eye Res. 2021;202:108314.
new procedures cost-effective and available for https://doi.org/10.1016/j.exer.2020.108314.
5. Zarif ME, Alió del Barrio JL, Arnalich-Montiel
all ophthalmology clinics, without the need of F, De Miguel MP, Makdissy N, Alió JL. Corneal
investing in expensive facilities. These cells and stroma regeneration: new approach for the treat-
derived tissues may no longer depend on ocular ment of cornea disease. Asia Pac J Ophthalmol
human sources at some point and start coming (Phila). 2020;9(6):571–9. https://doi.org/10.1097/
APO.0000000000000337.
from human bioengineering extraocular sources 6. El Zarif M, Alió JL, Alió Del Barrio JL, Abdul Jawad
or even xenogenic tissues. Stem cells, that con- K, Palazón-Bru A, Abdul Jawad Z, De Miguel MP,
tain proven immunomodulatory properties, will Makdissy N. Corneal stromal regeneration therapy for
unlikely be provided from the same patient, advanced keratoconus: long-term outcomes at 3 years.
Cornea. 2021;40(6):741–54. https://doi.org/10.1097/
affected by the same genetic imbalance that made ICO.0000000000002646.
the corneal disease to happen initially (except for 7. Alió del Barrio JL, De la Mata A, De Miguel MP,
those abnormalities caused by external aggres- Arnalich-Montiel F, Nieto-Miguel T, El Zarif M,
sions), but rather from considered genetically Cadenas-Martín M, López-Paniagua M, Galindo
S, Calonge M, Alió JL. Corneal regeneration
“optimal” donors in which immortal stem cells using adipose-derived mesenchymal stem cells.
constantly proliferating will provide an easy and Cell. 2022;11(16):2549. https://doi.org/10.3390/
cheap way to restore the biology of the cornea. cells11162549.
Index
A Best-corrected visual acuity (BCVA), 149

Acanthamoeba keratitis (AK), 271, 272 Big bubble (BB) technique, 23, 163, 261, 325, 331, 438,
Active storage machine (ASM), 33 515
Adipose tissue-derived mesenchymal stem cells Big bubble (BB) type 1, 261, 320
(AT-MSCs), 214 Big bubble (BB) type 2, 262
Adrenaline, 206 Bilateral LSCD, 179–181
Adult stem cells, 226 Biopsy, 205
Advanced keratoconus, 345–346 BKPro1, 501–504
autologous ADASC implantation, 346 Blunt trochar, 326
autologous ADASC isolation, 345 Bone marrow MSCs (BM-MSCs), 344
characterization, 345 Bone morphogenetic protein (BMP), 508
confocal microscopy, 355–356 Boston type 2 keratoprosthesis (BKPro), 148
corneal aberrometry, 354–355 Bowman layer (BL) transplantation, 246, 253
corneal densitometry, 355 clinical outcomes, 252
lenticule implantation, 346 complications, 252, 253
postoperative care and follow-up schedule, 346 indications, 246
Air-viscobubble (AVB), 265 surgical technique, 246, 248, 251
Allogeneic bone marrow-derived MSCs, 217 Bowman layer graft preparation, 246–248
Allogeneic cells, 403 Bowman layer inlay graft, 248
Allogeneic CLET, 206, 207 clinical outcomes, 252
Allo simple limbal epithelial transplantation complications, 252, 253
(ALLOSLET), 192, 195, 198, 199 Bowman layer onlay graft, 251
American Academy of Ophthalmology (AAO), 164 clinical outcomes, 252
Amniotic membrane, 489 complications, 253
Anterior chamber (AC), 445 Bowman membrane (BM), 291, 342
Anterior chamber intraocular lens (ACIOL), 408 Bowman membrane relaxation (BMR), 298, 300
Anterior chamber maintainer (ACM), 457 Bubble test, 263–264
Anterior donor lamella, 163
Anterior lamellar keratoplasty (ALK), 258–259
Anterior segment optical coherence tomography C
(AS-OCT), 163, 164, 192, 333, 335, 473 CAIRS procedure, 312
Applanation, 324, 328 Cannula big bubble technique, 262
Artificial intelligence, 30, 31 Cataract surgery, 4
Artificial iris implantation, 425 Cell culture techniques, 493
Astigmatism, 85, 160 Central corneal thickness (CCT), 351
Autologous CLET, 204, 205 Chitosan, 489
Autologous human keratocytes, 344 Chronic GVHD, 217
Cohesive OVD, 326
Collagen cross-linking (CXL), 314
B Confocal microscopy, 355–356, 358
Balanced salt solution (BSS), 367, 444 Conjunctival and Limbus Autograft (CLAU), 79
Bandage contact lens (BCL), 195 Conjunctival autograft, 181
Barraquer’s “thickness law”, 284 Conjunctival cicatrization, 193
© The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature 521
Switzerland AG 2023
https://doi.org/10.1007/978-3-031-32408-6
522 Index
Conjunctival limbal autograft /allograft (CLAu/CLAL), amniotic membrane transplantation (AMT), 79

191 anti-herpes medication, 80
Conjunctival migration, 203 Conjunctival and Limbus Autograft (CLAU), 79
Conjunctival symblepharon, 204 Cyclosporine A, 79
Conjunctival tissue for transplantation, 173 Fasudil, 79
Contact-lens corrected visual acuity, 252 FD006, 79
Conventional mechanical trephination, 162 Tacrolimus, 79
Conventional therapeutic PK (TPK), 271 Verteporfin photodynamic therapy, 79
Cornea, 214, 215 Corneal neovascularization (CNV), 337
Cornea guttata (CG), 29–31 epidemiology, 59, 60
Corneal aberrometry, 354–355 surgical and medical treatment
Corneal allogenic intrastromal ring segment (CAIRS) anti-VEGF local therapy, 62, 63
minimally invasive technique, 311 cauterization, 60
Corneal allograft survival, 215, 216 clinical considerations, 60, 61
Corneal collagen cross-linking (CXL), 284 fine needle diathermy (FND), 61, 62
Corneal damage, 212 laser treatment, 62
Corneal decellularization techniques, 342 scleral lens devices, 62–64
Corneal densitometry, 355 Corneal scarring, 336
Corneal disease, 499 Corneal stroma, 341
Corneal edema, 398 Corneal stroma regeneration, 520
Corneal endothelial cell transfer, 395 Corneal stromal stem cells (CSSCs), 343
See also HCEC-injection therapy Corneal topography, 288
Corneal endothelial cells (CEnCs), 461 Corneal transplant, 472
Corneal endothelial dysfunction, 396 Corneal transplantation, 159, 215–217, 258, 366
Corneal endothelial failure, 396, 398 anatomical success, 4
Corneal endothelial therapy classification, 1
cell isolation protocols, 487 complications, 4, 5
coatings, 487–488 DALK, 2, 3
culture media, 488 functional outcomes, 6–9
donor factors, 487 functional success, 4
human corneal endothelial cells, 491–493 graft failure
natural and synthetic materials, 490–491 anatomical failure, 5, 6
natural scaffolds, 488 functional failure, 5
synthetic scaffolds, 490 graft rejection
Corneal epithelial cell markers, 214 anti-angiogenic and lymphangiogenic therapy, 46
Corneal epithelial cells, 196, 227 lymphangiogenesis, 45, 46
conditioned medium, 231 histocompatibility matching, 46, 47
feeder cells, 232 immunosuppression
natural scaffolds, 232 limbal stem cell transplantation, 51, 52
xeno-free methods, 234 systemic therapy, 49–51
Corneal epithelial stem cells (CESC), 170, 171, 174–176, topical therapy, 47–49
178, 179, 181, 183, 185–187 penetrating keratoplasty, 2
Corneal epithelium, 212, 214 posterior lamellar keratoplasty techniques, 3, 4
Corneal graft failure risk stratification
classification, 146 active inflammation, 43
definition, 146 age, 45
DMEK, 151–153 aphakia, 44
DSEK, 150, 151 atopy, 44
incidence, 147 glaucoma, 43
repeat keratoplasty herpetic infection, 44
comorbidities, 147 limbal stem cell deficiency, 45
management, 147, 148 neovascularization, 44
outcomes, 149 pathogenic mechanisms, 42
overview, 147 primary diagnosis, 42
preoperative imaging, 147 risk factors, 42
surgical considerations, 148, 149 smoking, 45
risk factors, 147 survival rate, 9, 10
Corneal graft surgery, 520 Corneoscleral explantation, 19, 20
Corneal neovascularization (angiogenesis) Corrected distance visual acuity (CDVA), 307, 350, 458
recipient-related factors, 82 Corvis®, 314
treatment COVID-19 pandemic, 10, 20
Index 523
Crystalline lens, 389–390 Descemet membrane endothelial keratoplasty (DMEK),

Cultivated limbal epithelial transplantation 23, 24, 151–153, 163, 164, 258, 365, 381, 416,
(CLET), 172, 191, 198, 212, 213, 437, 463, 471, 486, 519
228, 500 descemetorrhexis, 422
allogeneic CLET, 206, 207 difficulty in supine position, 425
autologous CLET, 204, 205 donor injection technique, 418–420
surgical procedure donor preparation, 416
biopsy, 205 eyes with angle closure, 424
grafting, 205, 206 eyes with aphakia, 424–425
post-operative management, 206 graft insertion and unfolding, 417–420
stem-cell expansion, in culture, 205 graft preparation, 416–417
Cultivated oral mucosal epithelial transplantation innovations in complex eyes, 424–425
(COMET), 80, 500 innovative techniques, 421–422
Cultured endothelial cells, 396 intraoperative optical coherence tomography, 422,
Customized CAIRS, 316 423
pull-through donor insertion technique, 420
Descemet membrane endothelial transfer (DMET), 432
D Descemet membrane transfer (DMT), 493
DALK peeling technique, 515 Descemet or pre-Descemet layer, 319
Dapena maneuver, 371 Descemet stripping automated endothelial keratoplasty
Decentration, 160 (DSAEK), 3, 24, 258, 365, 381, 471
Deep anterior lamellar keratoplasty (DALK) techniques, Descemet stripping endothelial keratoplasty (DSEK),
2, 3, 6, 23, 163, 164, 246, 253, 258, 261, 283, 150, 151, 258, 365, 486, 519
292, 312, 316, 409, 437 Descemet stripping only (DSO), 395
big bubble, 261–262 impact on visual acuity, 433
bubble test, 263–264 intraoperative view, 434
cannula big bubble technique, 262 pseudophakic bullous keratopathy, 433
classifications of, 260–261 stromal injury, 434
complications, 274–275, 475 Descemet stripping procedure, 432
contemporary techniques, 319 Descemetocele, 272
donor corneal button, 324 Descemetorhexis, 479, 481, 482
donor cornea preparation, 267 Descemetorhexis Without Endothelial Keratoplasty
femtosecond laser, 132, 133 (DWEK), 371
host cornea, 324 Descemetorrhexis, 422, 486
indications, 268–273, 472 Descemet’s membrane ("naked Descemet’s"), 163, 164
laser parameters for donor, 325 Direct trans-differentiation, 230
layer by layer manual dissection, 266–267 Dirisamer technique, 371
long term outcomes, 267–268 DMEK, 3
manual DALK techniques, 265–267 Donor cornea preparation, 267
mini bubble mushroom, 323 Donor graft preparation, 452
needle big bubble technique, 262, 264 Donor injection (endothelium-out) technique, 418–420
outcomes of mini bubble mushroom, 327–328 Donor lenticule, 286
pachy-bubble, 262 Double-bubble technique, 419
posterior side-cut, 324 Double infusion cannula technique (DICT), 457
postoperative management, 475 Double-pass technique, 409
pre-descemet layer, 320–321 Downs syndrome, 312
pre-Descemet layer, 320 Dry eye disease (DED), 74, 212, 217
pre-operative planning, 323 treatment of, 215, 217
refractive outcomes, 328 Dua’ s layer (DL), 260, 451
surgical technique, 473 Dysfunctional endothelium, 386
transplantation procedure, 325–327
use of femtosecond lasers, 321–323
variation of mushroom pattern, 321 E
visco-bubble, 264 Ectasia, 311
vision outcomes, 328 Ectatic corneal disorders, 313
visual outcomes, 268 Ectatic disorder, 269
Deep lamellar endothelial keratoplasty EK include iridocorneal endothelial syndrome (ICE)
(DLEK), 366 syndrome, 382
Descemet baring techniques, 320 Electron beam tomography (EBT), 508
Descemet membrane (DM), 85, 334, 371, 438, 439, Embryonic stem cells (ESCs), 344
444, 486 Endophthalmitis, 508
524 Index
Endothelial cell density (ECD), 26 microbiological testing, 25

Endothelial cells (EC), 485 slit-lamp biomicroscopy, 25, 26
Endothelial immune reaction, 164 sterile donor tomography, 27–29
Endothelial keratoplasty (EK), 3, 6, 415, 437, 455 Eyelid malposition, 204
anatomy-related challenges, 389–393
complications, 373, 388–389
DLEK, 370 F
DMEK, 151–153, 371, 372 Feeder-free monolayer differentiation method, 234
DMEK tissue preparation, 383–384 Femtosecond Descemet membrane endothelial
DSAEK, 372 keratoplasty (DMEK), 479
DSAEK tissue preparation, 382–383 efficacy, 482–483
DSEK, 150, 151, 372 postoperative slit-lamp photograph of a patient with,
DSEK/DSAEK, 369–370 482
evolution of, 366–367 safety, 481–482
future of, 374 surgical technique, 480–481
graft insertion and positioning, 386–387 Femtosecond intrastromal lenticular implantation (FILI),
graft survival of, 372–373 285, 298
indications, 367–369, 382 Femtosecond laser (FSL), 119, 120, 263, 286, 471–473,
PDEK, 371, 372 475
postoperative management, 387–388 corneal tissue, 127–129
recipient preparation, 385–386 DALK, 132, 133
tissue orientation marks, 384 healing patterns, 134, 135
UT-DSAEK, 370, 372 incision design software, 129, 130
Endothelial–mesenchymal transition (EMT), 463, 468 keratoplasty incision profiles
Enzyme-linked immunosorbent assay (ELISA), 397 mushroom based profile, 131
Epikeratophakia, 284 tophat profile, 130
Epikeratoplasty (EPK), 258 zigzag graft, 131
Epithelial and stromal rejection, 268 penetrating keratoplasty, 131, 132
Epithelial thickness, 289 Femtosecond laser assisted cataract surgery (FLACS),
European Medicine Agency (EMA), 204 481
Excimer laser-assisted deep anterior lamellar Femtosecond laser-assisted DALK, 337
keratoplasty (excimer-DALK), 163, 164 Femtosecond laser-assisted DMEK surgery, 422
donor, 163 Femtosecond laser-assisted tuck-in PKP, see
donor lamella, 163 Keratoglobus (KTG)
recipient cornea, 163 Femtosecond laser dissection, 383
surgical technique, 163–165 Focal edema, 312
Excimer laser-assisted penetrating keratoplasty Fontana cannula, 326
(excimer-PKP), 159–164 Fuchs’ Dystrophy, 494
Excimer laser keratoplasty, 329 Fuchs endothelial corneal dystrophy (FECD), 23, 29,
Eye bank 367, 382, 395, 396, 399, 415, 417, 431, 462,
amniotic membrane, 25 464
anterior lamellar keratoplasty, 23 Fuchs marginal keratitis, 259
ASM, 33 Full-thickness keratoplasty, 333
cryopreservation, 20, 21
donor’s screening, 18–20
German keratoplasty register, 17, 18 G
hypothermic storage, 21 Gelatin, 489
moist chamber preservation, 20 General anesthesia, 194
organ culture preservation, 21, 22 Glaucoma, 104, 337, 366, 373, 508
PKP, 22, 23 Glueless SLET, 196
posterior lamellar keratoplasty, 23, 24 Goldmann applanation tonometry (GAT), 85
pre-cut tissues, 24, 25 Graft detachment, 389
pre-loaded tissues, 24, 25 Graft dislocation, 373
pre-stripped tissues, 24 Graft failure, 267, 389
QMS, 31–33 Graft rejection
scleral tissue, 25 angiogenesis suppression, 78
tissue collection, 18–20 astigmatism, 85
tissue selection characteristics
cornea guttata, 29–31 donor’s age, 80, 81
endothelial evaluation, 26, 27 donor’s sex, 80
Index 525
eye banking practices, 81 Human adipose-derived adult stem cells"

graft size, 80, 81 (H-ADASCs), 344
corneal angiogenesis treatment Human adult adipose tissue, 343
amniotic membrane transplantation (AMT), 79 Human amniotic membrane (HAM), 195
anti-herpes medication, 80 Human corneal endothelial cells (HCECs), 395–397,
Conjunctival and Limbus Autograft 401, 486, 491–493
(CLAU), 79 Human corneal endothelium, 485
Cyclosporine A, 79 Human crystalline, 489
Fasudil, 79 Human leukocyte antigen (HLA) matching, 46, 47, 148
FD006, 79 Human skin fibroblasts, 230
Tacrolimus, 79
Verteporfin photodynamic therapy, 79
DED, 74 I
diagnosis, 78 Iatrogenic LSCD, 200
endothelial cell failure, 73, 74 iFS IntraLase, 480
endothelial cell loss (ECL), 77 Immunosuppression, 216
functional and refractive outcomes, 75–77 limbal stem cell transplantation, 51, 52
immunological rejection, 73 systemic therapy
keratoconus cyclosporine, 49–51
decentered grafts, 84 mycophenolate, 50, 51
suturing technique, 84–86 tacrolimus, 51
OSD, 74 topical therapy
postoperative care cyclosporine, 48
clinical factors, 92–95 steroids, 47, 48
corticosteroids, 87, 88 tacrolimus, 48, 49
immunosuppressive therapies, 88–91 Inappropriate mechanical trephination, 161
intraocular pressure, 85 Indocyanine Green (ICG), 118, 119
stages and approach, 91, 92 Induced pluripotent stem cells (iPSCs), 344
preoperative precautions, 78 advantages, 230
for prevention and treatment, 77 human skin fibroblasts, 230
recipient-related factors molecular markers of, 227
corneal neovascularization, 82 murine-derived feeder cells, 230
glaucoma, 82, 83 OSKM factors, 229
infections, 82 Intra corneal ring segments (ICRS) implantation, 284,
inflammation, 83 285, 290
ocular surface diseases, 83 Intracameral acetylcholine chloride, 411
primary diagnosis, 81, 82 IntraLase enabled keratoplasty (IEK) module, 129, 130
primary/repeat PK performance, 82 IntraLase femtosecond laser, 328
systemic chronic diseases, 83 Intraocular lens (IOL), 24
rejection rates, 77 Intraocular pressure (IOP)
surgery-related factors, 83, 84 postoperative care, 85
Graft survival, 267 Intraoperative optical coherence tomography (IOCT),
Graft-versus-host disease (GVHD), 215, 217 422
cystoid macular edema, 513
and DALK, 515–516
H and EK, 514–515
Haag-Streit device, 514 and lens implants, 516
Haag-Streit iOCT™, 514 retained nuclear fragments, 516–517
HCEC-injection therapy, 395, 396 Intraoperative trephination, 159
clinical results, 398–400 Intrastromal corneal ring segments (ICRS), 312
future research, 403 allogenic tissue, 314
surgical procedure for, 397 benefit of, 315–316
Hemi-DMEK, 421 biocompatibility and biointegration, 313
Herpes simplex virus (HSV), 269 combined procedures, 316–317
Hessburg-Barron trephine, 298 contraindications, 313
High order abberations (HOA), 5 indications, 313
Histocompatibility matching, 46, 47 mechanism of action of, 312
Homburg/Erlangen technique, 161 post operative medications, 315
Horizontal torsion, 159–163 stability of, 316
HSV related neurotrophic keratopathy, 273 Intra-stromal FSL-based refractive surgery, 288
526 Index
Intrastromal viscoelastic injection, 265 Laser peripheral iridotomy (LPI), 494

In-vivo confocal microscopy (IVCM), 288–289, 291 Laser welding
Iris defects, 424 laser closuring
Irregular astigmatism, 159 biocompatible patches, 123
Isolated BL grafts, 246 clinical setting, 118
Indocyanine Green, 118, 119
photothermal effects
K AlGaAs diode laser, 119
Keratocan, 342 all-laser corneal transplantation, 121, 122
Keratoconus (KC), 246, 248, 251–254, 269, 283, 307, donor flap, 119–121
311, 332, 341, 342 femtosecond laser, 119, 120
decentered grafts, 84 monitoring, 122, 123
suturing technique, 84–86 protocol, 119
Kerato-epithelioplasty, 170 LASIK flap, 306
Keratoglobus (KTG), 269 Lebers congenital amaurosis, 312
complications, 142 Lenticule customization, 284
indications, 140 Lenticule intrastromal keratoplasty (LIKE) procedure, 305
outcomes, 142, 143 Leukoma, 269
surgical technique Lid abnormalities, 104
donor cornea preparation, 140, 141 Lidocaine, 206
donor-recipient apposition, 141 Limbal allograft, 79, 171, 178–181, 185, 186
postoperative management, 142 Limbal autograft, 175–178, 184
preoperative considerations, 140 Limbal biopsy, 195, 205
recipient cornea preparation, 141 Limbal epithelial stem cells (LESCs), 192, 196, 212,
Keratolimbal allograft (KLAL), 181, 191 213, 216, 500
Keratolimbal grafts Limbal graft, 194
combined techniques, 181–184 Limbal peritomy, 206
donor options, 174, 175 Limbal stem cell deficiency (LSCD), 45, 169, 171,
historical background, 169–172 173–175, 178–181, 184, 187, 188, 191–193,
limbal allograft techniques, 178–181 197–200, 203, 204, 206–208, 212–214, 217,
limbal autograft techniques, 175–178 227–229, 270, 500
ocular surface transplantation procedures, treatment of, 213–217
classification of, 172, 173 Limbal stem cells, 226, 227
postoperative management, outcomes and Limbal stem-cell transplantation (LSCT), 203, 204, 228
complications, 184–186 Linear microkeratomes, 409
SLET, 186, 187 Living-related conjunctivo-limbal autograft (lr-CLAL), 171
tissue options, 173, 174 LogMAR, 298
Keratometric astigmatism, 161, 163 Long-term immunosuppressive therapy, 228
Keratoplasty, 169, 178, 183, 184, 186, 215 Lymphangiogenesis, 45, 46
Keratoprosthesis
BKPro1, 501–504
future of, 509 M
history and development, 500–501 Macroperforation, 336
limitations of, 500 Major Histocompability Complex (MHC), 73
OOKP, 504–508 Malocclusion, 204
Kmean anterior, 299 Marfan’s syndrome, 312
Kmean posterior, 299 Mechanical trephination, 161
Mesenchymal stem cells (MSCs), 229, 343, 344
Mesenchymal stem cells (MSCs), ocular surface
L regeneration, 212, 213, 218
Lamellar keratoplasty, 207, 258, 422 clinical evidence of MSC efficacy, 216
Lamellar patch grafts, 259 DED, treatment of, 217
Large-diameter DALK, 332 LSCD, treatment of, 216, 217
complications, 336–338 preclinical evidence of MSC efficacy, 213
keratoconus, 332 corneal allograft survival, 215, 216
postoperative visual rehabilitation, 332 DED, treatment of, 215
standardization of, 333 treatment of LSCD and corneal epithelial damage,
surgical technique, 333–335 213–215
Laser-induced optical breakdown (LIOB), 127 Microbubble incision technique, 335
Laser in situ Keratomileusis (LASIK) procedure, 304 Minor ipsilateral SLET (Mini-SLET), 196
Index 527
Mooren’s ulcer, 259 post-operative endophthalmitis, 105

Morphological graft failure, 5, 146 scleral fixation ring, 105
Multi-zone ocular cells (MZOC), 234 suprachoroidal haemorrhage, 105
Murine-derived feeder cells, 230 suture modifications, 107
suture material, 107, 108
suture sizes, 108
N suturing
Natural scaffolds, 488 continuous double suturing, 110–112
Needle big bubble technique, 262, 264 continuous single suturing, 109, 110
New big bubble (BB) opening, 264 interrupted suturing, 108, 109
N-hydroxysulfosuccinimide (NHS), 490 intraoperative adjustment, 113
Non-inflammatory degenerative ectatic disorder, 283 postoperative adjustment, 113
Non-mechanical excimer laser-assisted trephination, 161 principles, 108
Non-MHC antigens matching, 73 suture knot, 112
Numerical aperture (NA), 127, 128 wound apposition, 112, 113
trephination, 160
Penetrating or perforating corneal wounds, 270
O Pentacam®, 314
Ocular surface, 228, 231, 233, 234 Peripheral anterior synechiae (PAS), 149
Ocular surface disease (OSD), 74, 204 Peripheral iridotomy, 409
pre-operative considerations, 104 Peripheral ulcerative keratitis (PUK), 259
Ocular surface regeneration Persistent epithelial defect, 337
MSCs (see Mesenchymal stem cells (MSCs), ocular Phakic IOL implantation, 424
surface regeneration) Photodisruption, 127
Optical coherence tomography (OCT) mapping, 197, Photorefractive keratectomy (PTK), 284
198, 321, 332 Phototherapeutic keratectomy (PTK), 259
Optimal trephination, 160 Plasma-induced ablation, 127
Optimizing visual outcomes, 336 Platelet-derived growth factor (PDGF), 343
Orientation notches, 161, 162 Pneumatic dissection, 333
Orientation teeth, 161–163 Poly (lactic-co-glycolic acid) (PLGA), 490
Osteo-odonto keratoprosthesis (OOKP), 499, 504–508 Polymethylmethacrylate (PMMA), 312, 490, 501
Post-mitotic cells (PMCs), 226
Post-vitrectomy, 390
P Post YAG-capsulotomy, 390
Pachy-bubble, 262 Posterior lamellar keratoplasty (PLK), 3, 4, 366, 431,
Pannectomy, 206 451, 471
Paradigm shift, 519 Posterior polymorphous dystrophy (PPMD), 382
Paufique dissector, 325 Postoperative astigmatism, 159
Paufique knife, 265 Postoperative keratometric astigmatism, 161
Peeling off DALK technique, 266 Pre-Descemet’s endothelial keratoplasty (PDEK), 365,
Peeling off technique, 265 437, 439, 441–443, 456
Pellucid marginal degeneration (PMD), 259, 269 anterior chamber, 445
Penetrating keratoplasty (PK), 2, 5, 6, 22, 23, 159, 169, donor graft preparation, 452
203, 246, 253, 283, 312, 319, 381, 382, 431, donor tissue, 455
437, 485 endothelial keratoplasty procedures, 455
corneal graft failure (see Corneal graft failure) graft insertion, 453
excimer-PKP, 159–163 post-op regimen, 454
femtosecond laser, 131, 132 post-operative management, 445, 446
graft rejection (see Graft rejection) recipient bed preparation, 444, 453
history, 103, 104 tissue harvesting, 442, 444
intraoperative conversion rate, 164 tissue preparation, 445
pre-operative considerations, 104, 105 Pre-Descemet’s layer (PDL), 260, 438, 440, 442
severe irregular astigmatism, 159 Primary graft failure (PGF), 5, 146
surgical techniques Proliferative capability, 487
donor cornea trephination, 105 Pseudoexfoliation syndrome, 494
host cornea sizing, 105 Pseudophakic bullous keraopathy, 453, 454
host cornea trephination, 105, 106 Pseudophakic bullous keratopathy, 415, 494
instrument set, 107 Pull-through (endothelium-in) donor insertion technique,
operating time, 105 420, 421
Paton spatula, 107 Pyogenic granulomas, 200
528 Index
Q primary failure, 200

Quality management system (QMS), 31–33 rejection, 199
Quarter-DMEK, 421 mechanism of action, 196
modifications
Glueless SLET, 196
R Mini-SLET, 196
Randomized controlled clinical trial, 217 sandwich technique, 196
Real Architecture for 3D Tissues (RAFT), 489 outcomes, 194
Recurrence of keratoconus, 269 clinical efficacy, 198
Refractive Lenticule Extraction (ReLEx) techniques, 284 monitoring outcomes, 197
Refractive sphere (Rx Sphr), 354 postoperative care, 195
Refractive surgery, 4 preoperative workup
Regenerative-medicine, 395 case selection, 192
Reinstein dissector, 307 presurgical care, 194
Retinal detachment procedures, 4 sequence of surgeries, 193
Retinal pigmented epithelium (RPE), 232 technique
Retroprosthetic membrane (RPM), 508 anesthesia, 194
Rho-associated protein kinase (ROCK), 462 donor eye, 194–195
Rho-associated protein kinase (ROCK)-inhibitor, 395 recipient eye, 195
Rho-Kinase (ROCK) inhibitors, 432, 433 Sinskey’s hook, 298
Risk stratification Sjögren’s syndrome-associated DED, 217
active inflammation, 43 Slit lamp biomicroscopy, 164
age, 45 Small-bubble technique, 326
aphakia, 44 Small incision lenticule extraction (SMILE), 232
atopy, 44 Small incision refractive lenticule extraction (SMILE),
glaucoma, 43 284
herpetic infection, 44 SMILE-derived lenticules, 308–309
limbal stem cell deficiency, 45 SMILE lenticules
neovascularization, 44 complications, 304
pathogenic mechanisms, 42 efficacy and safety, 299
primary diagnosis, 42 keratoconus, 307
risk factors, 42 keratoconus and SMILE-ectasia, 306–308
smoking, 45 Lamellar surgery with, 298
LASIK flap, 306
presbyopia, 308
S spherical equivalent, 299–304
Safety-net suture method, 424 surgical procedure, 298
Scheimpflug tomography, 480 Spectral-domain optical coherence tomography
Schirmer test, 204 (SD-OCT), 458
SCHWIND-AMARIS Excimer-Laser, 162 Spherical equivalent (SE), 299–304
Secondary graft failure (SGF), 5, 146 Stem cell niche, 227
Seibel spatula, 287 Stevens-Johnson syndrome, 183
Self-formed ectodermal autonomous multi-zone Stromal cell-derived inducing activity (SDIA) method,
(SEAM), 233 232
Sequential keratoplasty, 193 Stromal dystrophyes, 271
Severe corneal edema, 390 Stromal edema, 78
Severe irregular astigmatism, 159 Stromal lenticule addition keratoplasty (SLAK)
Severe limbal stem-cell deficiency, 203 advancements in corneal surgery, 284–292
Severe thermal/chemical injuries, 270 and biomechanical changes, 291
Silk fibroin, 489 biomechanical redistribution, 291
Simple limbal epithelial transplantation (SLET), 80, corneal reinnervation, 292
186–188, 191, 193, 197, 198, 200, 500 corneal topography, 288
alloSLET, 192 epithelial and stromal remodeling, 289–290
approach to keratoplasty, 199 epithelial remodeling, 291
auSLET, 192 history of, 284
early postoperative complications, 199 in-vivo confocal microscopy, 288–289
indications, 192 IVCM, 292
intraoperative complications, 199 lenticule shape customization, 292–293
late complications surgical procedure, 286, 287
focal recurrence, 199 treatment of, 285
Index 529
Stromal lenticules, 290 postoperative care, 411

Sub-epithelial infiltrates (SEIs), 78 preoperative planning for, 408
Sub-Total Anterior Lamellar Keratoplasty (STALK), 261 recipient preparation and graft delivery, 409–410
Superficial Anterior Lamellar Keratoplasty (SALK), 259 refractive outcomes, 412
Suprachoroidal haemorrhage, 105 surgical technique, 409
Surface regularity index (SRI), 161 visual outcomes, 411
Ultraviolet (UV), 307
Umbilical MSCs (UMSCs), 344
T Unaided distance visual acuity (UDVA), 350
Tamponade, 392 Uncorrected distance visual acuity (UDVA), 287
Terminally differentiated cells, 226 Unilateral LSCD, 192
Terrien’s marginal degeneration, 259
Therapeutic DALK (TDALK), 271
Thicker CAIRS segments, 315 V
3D femtosecond laser, 321 Vascular endothelial growth factor (VEGF), 343
Tissue additive procedures, 306 Vertical tilt, 159, 160, 162, 163
Tissue-specific stem cells, 226 Visco-bubble, 264
Topical anaesthesia, 205, 346 Viscodissection, 264–265
Topical lidocaine, 205 Viscoelastic-assisted dissection, 335
Topographic astigmatism, 160 Visual acuity, 159, 160
Total Anterior Lamellar Keratoplasty (TALK), 261 Visual rehabilitation, 283
Transforming growth factor beta (TGF-ß), 485 VisuMax FS laser, 298, 307
Transforming growth factor-beta 1 (TGFβ1), 343
Transient amplifying cell (TAC), 226
Translational regenerative surgery, 344, 345 W
Trephination, 160, 161 Wavelight Topolyzer®, 314
25-gauge graft manipulator, 420 Whole eye globe collection, 19
Wound dehiscence, 284
U
Ultrathin Descemet stripping automated endothelial Y
keratoplasty (UT-DSAEK), 24, 408 Yoeruek no-touch technique, 371
anesthesia considerations, 408
combined surgeries, 411
complications, 412–413 Z
endothelial cell density, 412 Ziemer LDV Z8 laser, 328
graft survival, 412 Zywave aberrometer®, 314
indications of, 408

Modern Keratoplasty

Uploaded by

Copyright:

Available Formats

Modern Keratoplasty

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Modern Keratoplasty

Uploaded by

Copyright:

Available Formats

Essentials in Ophthalmology

Series Editor: Arun D. Singh

ISSN 1612-3212 ISSN 2196-890X (electronic)

Corneal surgery has become today a highly specialized discipline in oph-

Alicante, Spain Jorge L. Alio

After cataract, corneal blindness is the commonest cause of curable sight

Pre-Descemet’s endothelial keratoplasty (PDEK) wherein the donor trans-

endothelial cells encouraged to migrate on to the surface thus exposed, restor-

Nottingham, UK Harminder S. Dua

Part I Preparing the Patient

2 Modern Eye Banking: Preservation, Type of Tissues,

Part II Penetrating Keratoplasty

5 Main Issues to Overcome in Modern Penetrating

Part III Epithelial Lamellar Keratoplasty

Part IV Stromal Lamellar Keratoplasty

17 Bowman Layer Transplantation���������������������������������������������������� 245

Part V Endothelial Keratoplasty

Jorge L. Alió, Dominika Wróbel-Dudzińska,

Key Points Introduction

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 1

According to the Spanish study, the main reason

Table 1.2 Distribution of BCVA according to the surgical procedure performed

Table 1.2 (continued)

refractive surgery (EUREQUO): a database study of Int J Ophthalmol. 2017;10(3):385–90. https://doi.

Key Points Introduction

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 17

 onor’s Screening and Tissue

Preservation Techniques Cryopreservation

Moist Chamber Preservation Cryopreservation has long been a goal in the

protocols may emerge. Nevertheless, the absence Organ Culture Preservation

theory—already be implanted after only 12 h of Penetrating Keratoplasty

biomechanics of DSAEK grafts suggests that Tissue Selection

Our department succeeded, in collaboration

Benefits of a Highly-Structured Organization for Standardization) 9001:2015

Ophthalmic Surg. 1988;19:817–20. https://doi. preservation on DMEK (Descemet membrane endo-

anniversary - Video article]. Ophthalmologe. keratoplasty versus Descemet stripping auto-

Paula W. Feng and Guillermo Amescua

Key Points been considered to have immune privilege, and

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 41

Herpetic Infection Atopy

for postoperative immunosuppression over a Selecting Immunosuppression

Cultivated limbal epithelial transplantation is angiogenic and lymphangiogenic therapy

Nadim S. Azar, Matias Soifer, and Victor L. Perez

Key Points corneal transplantation, namely fine-needle

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 59

necessary to study corneal neovascularization  linical Considerations for Treatment

Anti-VEGF Local Therapy bevacizumab have been shown to decrease neo-

Farideh Doroodgar, Sana Niazi, Hassan Hashemi,

Key Points raising eye pressure and hastening infectious

Supplementary Information The online version con-

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 71

Introduction improved outcomes in PK and other surgical

Immunological Rejection are the presence of stromal blood vessels in one or

4. After surgery, immunosuppressive drops such Following transplantation, treatments such

 he Short-Term and Long-Term

Mechanical Trephines Nonmeehanieal Trephines

The 193-nm Excimer

Fig. 5.1 Different types of trephines

 linical Aspects of Graft Failure

Treatment of Corneal Angiogenesis Various studies have suggested several meth-

Alicante, Spain Jorge L. Alio

Nottingham, UK Harminder S. Dua

Part I Preparing the Patient

2 Modern Eye Banking: Preservation, Type of Tissues,

Part II Penetrating Keratoplasty

5 Main Issues to Overcome in Modern Penetrating

Part III Epithelial Lamellar Keratoplasty

Part IV Stromal Lamellar Keratoplasty

17 Bowman Layer Transplantation�� 245

Part V Endothelial Keratoplasty

Key Points Introduction

Key Points Introduction

onor’s Screening and Tissue

Preservation Techniques Cryopreservation

Moist Chamber Preservation Cryopreservation has long been a goal in the

protocols may emerge. Nevertheless, the absence Organ Culture Preservation

theory—already be implanted after only 12 h of Penetrating Keratoplasty

biomechanics of DSAEK grafts suggests that Tissue Selection

Benefits of a Highly-Structured Organization for Standardization) 9001:2015

Herpetic Infection Atopy

for postoperative immunosuppression over a Selecting Immunosuppression

necessary to study corneal neovascularization linical Considerations for Treatment

Anti-VEGF Local Therapy bevacizumab have been shown to decrease neo-

Introduction improved outcomes in PK and other surgical

Immunological Rejection are the presence of stromal blood vessels in one or

he Short-Term and Long-Term

linical Aspects of Graft Failure

Treatment of Corneal Angiogenesis Various studies have suggested several meth-

nylon sutures: single continuous suture (SCS), Management of Postpenetrating

Corneal transplant rejection usually starts with Better Management

Non-absorbable important to emphasise some principles that are

Continuous Suturing (Double) Spadea et al. compared the outcomes of a

junction that may lead to delayed graft-host Postoperative Suture Adjustment

Key Points Introduction

All-Laser Transplantation cornea. The pre-operative BCVA was 1/20, and

Key Points emtosecond Laser Action

Incision Design Software

Keratoplasty Incision Profiles

Mushroom-Based Profile Zigzag slanted incisions increase the wound

ealing Patterns in Femtosecond

Key Points Introduction

Since KTG is a rare ectasia, diverse algo- Surgical Technique

ecipient Cornea Preparation

Postoperative Management tion. Probably, the most challenging step is the

Conclusions and it is more accurate than a free-hand manual

Key Points Introduction

K Graft Failure: Incidence and Risk

escemet Stripping Endothelial

Conclusion 4. Alino AM, Perry HD, Kanellopoulos AJ, Donnenfeld

Key Points xcimer Laser-Assisted Penetrating