Direct or Indirect Restoration of Endodontically Treated

Maxillary Central Incisors with Class III Defects?

Composite vs Veneer or Crown Restoration
Manja von Stein-Lausnitza / André Mehnertb / Maria Bruhnkec / Guido Sterzenbachd / Martin Rosentritte /
Benedikt C. Spiesf / Kerstin Bitterg / Michael Naumannh

Purpose: The aim of this ex-vivo study was to evaluate the load capacity of direct or indirect endodontically re-
stored maxillary central incisors with Class III defects, with or without glass-fiber posts.
Materials and Methods: Seventy-two extracted human maxillary central incisors were endodontically treated and bi-
proximal Class III cavities were prepared. Specimens were randomly allocated to six groups (n = 12): direct restor-
ation with composite (C); direct restoration with composite and additional glass-fiber post (CP); ceramic veneer
restoration (V), ceramic veneer restoration and additional glass-fiber post (VP), ceramic crown restoration (Cr), ce-
ramic crown restoration and additional glass-fiber post (CrP). Specimens were exposed to thermomechanical load-
ing (TML: 1.2 million cycles, 1 to 50 N; 6000 thermal cycles between 5°C and 55°C for 1 min each), and
subsequently linearly loaded until failure (Fmax [N]) at an angle of 135 degrees 3 mm below the incisal edge on the
palatal side. Statistical tests were performed using the Kruskall-Wallis and Mann-Whitney U-Test.
Results: During dynamic loading by TML, one early failure occurred in group C, CP, and CrP. Subsequent linear
loading resulted in mean fracture load values [N] of C = 483 ± 219, CP = 536 ± 281, V = 908 ± 293,
VP = 775 ± 333, Cr = 549 ± 258, CrP = 593 ± 259. The Kruskal-Wallis test showed significant differences of load
capacity between groups (p < 0.05). Mann-Whitney U-test revealed significantly lower maximum fracture load val-
ues of group C compared to group V (p = 0.014), after Bonferroni-Holm correction. Non-restorable root fracture was
the most frequent type of failure.
Conclusion: Endodontically treated maxillary central incisors with Class III defects directly restored with composite
are as loadable as indirect crown restorations. Compared to full-coverage restorations, less invasive veneers ap-
pear to be more beneficial. Additional placement of glass-fiber posts shows no positive effect.
Keywords: all-ceramic crown, Class III restorations, direct composite restorations, endodontically treated teeth, in-
direct restoration, post and core, veneers.

J Adhes Dent 2018; 20: 519–526. Submitted for publication: 02.08.18; accepted for publication: 30.10.18
doi: 10.3290/j.jad.a41635

a Assistant Professor, Charité – Universitätsmedizin Berlin, corporate member e Professor and Engineer, Department of Prosthetic Dentistry, Regensburg Uni-
of Freie Universität Berlin, Humboldt-Universität zu Berlin, and Berlin Institute versity Medical Center, Regensburg, Germany. Performed part of experiments,
of Health, Department of Prosthodontics, Geriatric Dentistry and Cranioman- chewing simulation.
dibular Disorders, Berlin, Germany. Idea, hypothesis, wrote the manuscript. f Associate Professor, Charité – Universitätsmedizin Berlin, corporate member
b PhD Student and Assistant Professor, Charité – Universitätsmedizin Berlin, of Freie Universität Berlin, Humboldt-Universität zu Berlin, and Berlin Institute
corporate member of Freie Universität Berlin, Humboldt-Universität zu Berlin, of Health, Department of Prosthodontics, Geriatric Dentistry and Cranioman-
and Berlin Institute of Health, Department of Prosthodontics, Geriatric Den- dibular Disorders, Berlin, Germany. Contributed substantially to discussion
tistry and Craniomandibular Disorders, Berlin, Germany; Tooth Prophylaxis and figure design.
Center Berlin, Schönefeld, Germany. Performed experiments in partial fulfill- g
ment of requirements of PhD degree. Associate Professor, Department of Operative and Preventive Dentistry,
Charité – Universitätsmedizin Berlin, corporate member of Freie Universität
c PhD Student and Assistant Professor, Charité – Universitätsmedizin Berlin, Berlin, Humboldt-Universität zu Berlin, and Berlin Institute of Health, Berlin,
corporate member of Freie Universität Berlin, Humboldt-Universität zu Berlin, Germany. Consulted on statistical evaluation, proofread the manuscript.
and Berlin Institute of Health, Department of Prosthodontics, Geriatric Den- h
tistry and Craniomandibular Disorders, Berlin, Germany. Performed experi- Professor, Charité – Universitätsmedizin Berlin, corporate member of Freie Univer-
ments in partial fulfillment of requirements of PhD degree. sität Berlin, Humboldt-Universität zu Berlin, and Berlin Institute of Health, Depart-
ment of Prosthodontics, Geriatric Dentistry and Craniomandibular Disorders, Berlin,
d Associate Professor, Charité – Universitätsmedizin Berlin, corporate member Germany. Idea, hypothesis, proofread the manuscript, interpretation of data.
of Freie Universität Berlin, Humboldt-Universität zu Berlin, and Berlin Institute
of Health, Department of Prosthodontics, Geriatric Dentistry and Cranioman- Correspondence: Manja von Stein-Lausnitz, Department of Prosthodontics, Ge-
dibular Disorders, Berlin, Germany. Performed a part of experiments, static riatric Dentistry and Craniomandibular Disorders, Charité – Universitätsmedizin
load testing. Berlin, Aßmannshauser Str 4-6, 14197 Berlin. Tel: +49-30-45-056-2724; e-mail:
[email protected]

Vol 20, No 6, 2018 519

Stein-Lausnitz et al

an equivalent performance of both treatment options.19

However, a transfer of these results to direct and indirect
C no post composite restorations of anterior ETT is limited.
To date, the in vitro performance of anterior ETT with
Class III cavities has been solely evaluated when finally re-
CP post composite
stored with full-coverage crowns,15 composite or ceramic
veneers,4,5,36 and composite only.2,35 Regrettably, the re-
V no post veneer spective studies differ in defect design, endodontic treat-
ment procedure, endodontic post material, and protocols of
VP post veneer loading, making comparison of the data impossible. To the
best of our knowledge, there is currently no comparative in
Cr no post crown vitro study that tested the load capacity of anterior ETT with
either direct or indirect restorative approaches.
Furthermore, it has been frequently described that glass-
CrP post crown fiber posts (GFPs) increase fracture strength of ETT,1,4 and
decrease the clinical failure risk compared to restorations
without a post.12,29 However, other studies found no posi-
Fig 1 Experimental group design. tive effect of a post on fracture resistance of ETT with Class
III cavities restored either with composite36 or with crown
restoration.15 Thus, it remains uncertain whether GFPs
have a strengthening, positive effect on load capacity in an-

R estoration of endodontically treated teeth (ETT) is a chal-

lenging procedure. To date, the selection of the type of
final restoration mainly depends on the remaining tooth
terior ETT with Class III defects.
Hence, the null hypotheses were: 1. there is no differ-
ence regarding load capacity of endodontically treated max-
structure.20 However, evidence for the best type of post- illary central incisors restored either directly with compos-
endodontic restoration considering the dimension of coronal ite, or indirectly with veneer or crown restorations; 2. there
tooth destruction is scarce to nonexistent.16,30,31 In particu- is no difference regarding load capacity between post-re-
lar, endodontically treated anterior teeth were found to have stored and post-free restorations, irrespective of the final
a higher risk of biomechanical failure due to shear forces,9, restoration, ie, composite, veneer, or crown restorations.
21 and are of specific interest in the literature.7,9,23,26 While
severely damaged, decoronated ETT with and without ferrule
design have been well documented regarding post place- MATERIALS AND METHODS
ment and full crown coverage,22,23,25 recommendations to
restore anterior ETT with Class III cavities are underrepre- Specimens and Allocation
sented in the literature. A systematic review on anterior Seventy-two human maxillary central incisors, stored in
composite restorations concluded that Class III restorations 0.5% chloramine solution, fulfilled the following criteria: car-
generally have lower annual failure rates than other restor- ies- and defect-free, no coronal defects, and endodontic
ations in the anterior region.9 One treatment option to re- treatment. Using a mechanical caliper, specimens’ dimen-
store anterior Class III cavities is to apply composite restor- sions were measured as the distance between the highest
ations directly without extensive preparation and removal of point of the cementoenamel junction (CEJ) and apex for root
tooth structure,14 avoiding the need for additional work in length, as well as the buccolingual and mesiodistal exten-
the dental laboratory. Hence, direct restorations allow a less sions at the CEJ. The minimum required root length was
invasive and time-consuming treatment procedure. Further- 15 mm. Considering the ratio of the buccolingual and me-
more, a range of different types of adhesively bonded ve- siodistal distance, specimens were allocated randomly to 6
neer restorations are available for anterior teeth.10 The indi- experimental groups with 12 specimens each (Fig 1).
cation for restoring anterior ETT with veneer restorations
ranges from solely esthetic indications, ie, masking highly Pretreatment of Specimens
discolored teeth, to highly functional restorations saving re- An endodontic access cavity was prepared (diamond bur FG
sidual coronal tooth structure.10,17,38 Nevertheless, full-cov- 290-015C Komet CCLC, Gebr Brasseler; Lemgo, Germany)
erage crown restorations on ETT have a wide acceptance in and root canals were instrumented with NiTi files up to size
restorative dentistry.6 A literature review concluded that ETT F2 (ProTaper System, Dentsply Maillefer; Ballaigues, Swit-
can reliably serve as abutments for dental crown restor- zerland). The root canals were irrigated with 1 ml of 1% so-
ations.13 In addition, a systematic review on the restoration dium hypochlorite solution (Hedinger; Stuttgart, Germany)
of ETT showed acceptable 10-year survival rates of 81% for after every change of file size. Root canals were obturated
ETT restored with crowns, and found reduced survival rates with a vertical obturation system (Calamus Dual, Dentsply
of 63% for ETT restored with direct restorations.32 A com- Maillefer) using a sealer (AH Plus Jet, Dentsply DeTrey; Kon-
parative study on endodontically treated premolars with ei- stanz, Germany) and a thermoplasticized obturation mater-
ther dental crowns or direct composite restorations showed ial (Gutta-percha, Dentsply Maillefer).

520 The Journal of Adhesive Dentistry

 Stein-Lausnitz et al

Restorative Procedure According to Group Allocation

Group C, direct restoration with composite
Before preparation of Class III cavities, the mesiodistal and
cervico-incisal extent of the tooth crown was measured with
a mechanical caliper. The vestibular surface was divided
into four equal parts on the horizontal axis and three equal
parts on the vertical axis. Mesial and distal Class III cavi-
ties were prepared, and included one-third of the vertical
and one-quarter of the horizontal crown dimension, respect-
ively (Fig 2).36 A minimum distance of 2 mm to the CEJ was
defined for the cervical extension of the Class III cavity. In
order to connect the mesial cavity with the endodontic ac-
cess cavity, a perforation was prepared with a round bur in
the mesioproximal to central direction. Gutta-percha was
removed 3 mm apically to the CEJ. The cavity walls and the
coronal part of the root canal were etched with 37% phos-
phoric acid, rinsed with water, and air dried. Dentin was re-
wetted for 30 s with 2% chlorhexidine solution using a mi-
crobrush (Root Canal Applicator Tips, Dentsply DeTrey). An
adhesive (Prime&Bond XP, Dentsply DeTrey) was applied to
all internal surfaces of the cavity for 20 s using a micro-
brush, and air dried for 5 s to ensure volatilization of adhe-
sive solvent. Subsequently, the adhesive was light cured for
10 s (1200 mW/cm2; Elipar Freelight 2, 3M Oral Care; St
Paul, MN, USA). A bulk-fill composite (SDR, Dentsply De-
Trey) was filled into the first 3 mm of the root canal and
Fig 2 Dimensions of Class III cavities. Mesial and distal cavities
light cured for 20 s. Access and Class III cavities were in- included one-third of the vertical and one-quarter of the horizontal
crementally filled with a composite (CERAM X Mono+, crown dimension.
Dentsply DeTrey), followed by light curing in increments
(20 s each) and a final polishing procedure.
cavity, pre-treatment was performed with aluminum oxide
Group CP, direct restoration with endodontic glass-fiber particles (50 μm, 2 bar, 5 s, Microetcher, Zest Dental Solu-
post and composite tions; Carlsbad, CA, USA). Adhesive and composite were
Preparation of cavities was performed as described for applied into the cavity and on the post, ensuring a compos-
group C. During the down-pack phase of vertical obturation, ite layer of at least 2 mm composite on top of the post. Fi-
gutta-percha was applied 10 mm apically to the mesial CEJ nally, the composite restoration was finished and polished.
to leave post space open, ensuring at least 4 mm of gutta- Further information and results on groups C and CP are
percha for apical sealing. Glass-fiber posts were tried in to described elsewhere.37
verify a passive fit and correct length of post space. Post
space was irrigated with 5 ml of 1% sodium hypochlorite Group V, veneer restoration
(NaOCl) for 1 min with passive ultrasonic irrigation for 30 s, Cavity preparation and composite restoration was per-
followed by 5 ml of distilled water. Access cavity and post formed as described for group C. Specimens were fixed
space were simultaneously etched for 15 s (phosphoric with the root into a polyether impression material (Imp-
acid 37%), and the post space was dried with ethanol regum Penta Soft, 3M Oral Care). A pre-scan of the clinical
(99%). Drying was finally verified with paper points. The den- crown was performed with an intraoral scanner (Trios color,
tin walls of the cavity were rewetted with 2% chlorhexidine 3Shape DentalDesigner; Copenhagen, Denmark) before ve-
with a microbrush. Adhesive was mixed with Self-Cure-Acti- neer preparation. Teeth were prepared considering a long-
vator (SCA, Dentsply DeTrey) according to the manufactur- wrap veneer removing proximal contact points. This in-
er’s instructions and applied into the root canal and cavi- cluded a facial preparation of 0.6 mm, an incisal reduction
ties. The glass-fiber post (X-Post Size 2, Dentsply DeTrey) of 1.5 mm, interproximal reduction of 1 mm, a cervical re-
was cleaned with ethanol (99%). Post cementation was per- duction of 0.4 mm (ending 1 mm incisal from CEJ), and a
formed with a dual-curing composite (CoreXFlow, Dentsply palatal chamfer preparation.3 Proximal Class III cavities
DeTrey), light cured for 20 s using the same light-emitting were not completely removed, ie, the veneer was intended
diode (LED) curing unit (Elipar Freelight 2, 3M Oral Care). to be luted to both tooth structure and composite. A prepar-
The access and Class III cavities were filled with a compos- ation scan was performed with the same intraoral scanner.
ite in layers, and stepwise light-curing sequences were per- Based on the pre-scan data set, the veneer restoration was
formed. The post was shortened 2 mm crestal to the top of designed with CAD Software (3Shape DentalDesigner). Ce-
the cavity with a diamond bur. To finally seal the access ramic veneers (zirconia-reinforced lithium silicate, Celtra

Vol 20, No 6, 2018 521

Stein-Lausnitz et al

a b c

d e f g

Fig 3 CAD of the crown restoration. a-c: vestibular (a), incisal (b) and oral (c) sites; d-g: crown restoration vestibular (d,e), oral site (f,g).

Duo, Dentsply DeTrey) were milled according to the manu- Group VP, veneer restoration and additional glass-fiber post
facturer’s guidelines. Veneers were manually polished with Cavity preparation, post placement and restorative proce-
diamond polishing bodies, cleaned with a steam cleaner, dure with composite were performed as described for group
and dried. Afterwards, veneers were placed on a carrier with CP. Veneers were fabricated and luted as described for
flexible inlay pins and glaze-fired according to the manufac- group V.
turers’ guidelines. Prior to the cementation procedure, the
internal surface of the veneer was cleaned with 34% phos- Group Cr, crown restoration
phoric acid. A 5% hydrofluoric acid gel (Ivoclar Vivadent) was Cavity preparation was performed as described for group C.
applied to the inner surface of the veneer for 20 s, rinsed Teeth were fixed with the root into a polyether impression
with water, and air dried. Silane coupling agent (Calibra Si- material (Impregum, ESPE; Seefeld, Germany). A pre-scan
lane, Dentsply DeTrey) was applied with a needle tip to of the clinical crown was performed with an intraoral scan-
avoid pooling of the silane agent, then gently air dried. Prox- ner (Trios color, 3Shape DentalDesigner) before crown prep-
imal Class III composite fillings were pre-treated with alumi- aration. Crown preparation was then performed with a par-
num oxide particles (50 μm, 2 bar, Microetcher, Zest Den- allel-sided cylindric diamond bur, size 012 1/10 mm
tal Solutions). The prepared tooth was etched with 34% (Komet Dental, Gebr Brasseler) considering anatomical re-
phosphoric acid for 15 s, rinsed with water spray for 15 s, duction of hard tissues and ensuring the minimum wall
and air dried for 5 s. Silane coupling agent was applied on thickness of the artificial all-ceramic crown, according to the
exposed composite areas. Adhesive (Prime&Bond XP, manufacturers’ guidelines (2 mm incisal reduction, 1.0 mm
Dentsply DeTrey) was mixed on a mixing dish with Self- chamfer, 1.5 mm circumferential reduction, angle of ap-
CureActivator (SCA) using an applicator tip, and uniformly proximately 6 degrees to the anatomical vertical axis of
applied to the prepared surface. Subsequently, surfaces tooth). The finishing line of the preparation was set 0.5 to
remained undisturbed for 20 s. A dual-curing composite ce- 1.0 mm coronal to the CEJ. Based on the pre-scan data
ment (Calibra Esthetic Resin Cement, Dentsply DeTrey) was set, crown restorations (zirconia-reinforced lithium silicate)
applied at the veneer, and dispensed with an application tip were designed and fabricated based on the CAD/CAM work-
on the entire inner surface. The veneers were seated with flow reported in the manufacturer’s guidelines (Fig 3). The
gradual pressure and excess cement was removed from internal crown surface was cleaned with 34% phosphoric
marginal areas. Glycerol gel (AirBlock, Dentsply DeTrey) was acid. A 5% hydrofluoric acid gel (Ivoclar Vivadent) was ap-
applied at the restoration margins, and marginal areas were plied to the inner surface of the crown for 20 s, rinsed with
light cured for 20 s from each direction. water and air dried. Silane coupling agent (Calibra Silane,

522 The Journal of Adhesive Dentistry

 Stein-Lausnitz et al

Dentsply DeTrey) was applied with a needle tip and gently (4) fracture diagonal at crown-root level including more than
air dried. The prepared tooth was etched with 34% phos- 2 mm of the root; (5) root fracture more than 2 mm below
phoric acid for 15 s. It was then rinsed with water spray for the CEJ; (6) root fracture in the cervical third. For statistical
15 s and air dried for 5 s. Adhesive (Prime&Bond XP) was analysis, failure modes 1 to 3 were classified as re-restor-
mixed on a mixing dish with SCA using an applicator tip, able failures. Failure modes 4 to 6 were classified as cata-
and uniformly applied to the prepared tooth surfaces. Expo- strophic failures.
sure time was 20 s. A dual-curing composite cement (Cali-
bra Esthetic, Dentsply DeTrey) was applied into the crown Statistical Analysis
using an application tip. Crowns were seated with gradual The number of samples per group was defined to collect
pressure and excess cement was removed from circumfer- data for further studies; therefore no sample size calcula-
ential restoration margins. Glycerol gel (AirBlock, Dentsply tion was performed. Non-parametric Kruskal-Wallis and the
DeTrey) was applied at the margin, and marginal areas were post-hoc Mann-Whitney U-test with Bonferroni-Holm correc-
light cured for 20 s from each direction. tion were applied for analysis of maximum load capacity
(Fmax). Differences in the frequency of the failure modes
Group CrP, crown restoration and additional glass-fiber post (re-restorable and catastrophic) between the groups and
Cavity preparation, post placement, and restorative proce- impact of post material were evaluated by by Pearson’s chi-
dure with composite and endodontic post were performed squared test (p = 0.05).
as described for group CP. Crown fabrication and placement
were performed as described for group C.
RESULTS
Embedding of Specimens
Roots were coated with a layer of wax (0.3 mm casting wax, During TML, one fracture of the incisal edge occurred in
veined green, Dentaurum; Pforzheim, Germany), and a group C, one loss of composite restoration in group CP was
round wax wire was coated 2.5 mm apically to the CEJ to observed, and one crown fracture occurred in group CrP.
simulate biological width. Teeth were retained parallel to Specimens with TML failure were assigned a load capacity
the tooth axis using a parallelometer and roots were em- of Fmax = 0.28.
bedded into an acrylic resin (Technovit 4004, Heraeus Kul- Group V showed highest mean fracture load
zer; Wehrheim, Germany). After polymerization of the em- (Fmax = 908 N), while group C presented the lowest mean
bedding material, teeth were removed and cleaned. fracture load (Fmax = 483 N) (Table 1, Fig 4). The Kruskal-
Thereafter, roots were coated with a thin layer of acrylic Wallis test indicated significantly different load capabilities
resin (Paladur, Heraeus Kulzer), and an adhesive (Mollosil between test groups (p = 0.002). The post-hoc Mann-Whit-
Adhäsiv, DETAX; Ettlingen, Germany) was applied on the ney U-test showed significant differences of load capacity
acrylic resin layer. A polysiloxane soft lining material (Mol- between groups C and V (p = 0.001), C and VP (p = 0.019),
losil, DETAX) was inserted into the root space of the acrylic CP and V (p = 0.005), CP and VP (p = 0.014), V and C
block. Specimens were finally placed back into the acrylic (p = 0.008), V and CP (p = 0.015), VP and CrP (p = 0.023).
block. After Bonferroni-Holm correction only Fmax values of group C
were significantly different from group V (p = 0.014).
Loading of Specimens Non-restorable root fracture with fracture lines more than
TML of specimens was carried out in distilled water at 5°C 2 mm below the CEJ was the most frequent type of failure
and 55°C, with 3000 thermal cycles per temperature for in all groups (68%) (Fig 5). The overall chi-squared test
2 min, and 1.2 x 106 mastication cycles of 1 and 50 N at a among all groups indicated no statistically significant differ-
loading angle of 135 degrees to the horizontal (chewing ence regarding failure type (p = 0.227). The subgroup com-
simulator, EGO; Regensburg, Germany). The contact point parison of re-restorable and catastrophic failures among
was set 3 mm below the incisal edge on the palatal surface test groups revealed a significant difference between
of the tooth crown. groups CrP and CP (p = 0.035), and between groups V and
Specimens that survived TML without any failures were CP (p = 0.035), with a higher number of catastrophic fail-
again loaded at 135 degrees in a universal testing machine ures in group PC and group V. A further chi-squared test
(Zwick 1446; Ulm, Germany) at v = 1 mm/min until failure showed no impact of the variable “post” on the failure
occurred. Failure detection was defined at 10% loss of max- mode (p = 0.221).
imum load force (Fmax).

Analysis of Failure Modes DISCUSSION

After testing maximum load capacity, each specimen was
observed to determine the failure mode. Failure modes This study investigated direct, ie, composite fillings, and
were classified as follows: (1) fracture of crown or veneer indirect treatment procedures, ie, veneers or crowns, to re-
restoration; (2) combined crown-tooth fracture at the ana- store endodontically treated maxillary central incisors with
tomic crown level (located coronal to the CEJ); (3) fracture Class III defects. Indirect veneer-restored teeth showed a
diagonal at crown-root level including first 2 mm of the root; significantly higher load capacity after dynamic and subse-

Vol 20, No 6, 2018 523

Stein-Lausnitz et al

Table 1 Results of experimental groups including survival after TML (dynamic loading), and load values after linear
loading (Fmax)

Group n Preliminary Invalid measurements during Mean ± SD Restorable / catastrophic

TML failure (n) linear loading Fmax [N]* failure (n)
C 12 1 0 483 ± 219 4/8

CP 12 1 0 536 ± 281 3/9

V 12 0 0 908 ± 293 2 / 10

VP 12 0 1** 775 ± 333 3/8

Cr 12 1 0 549 (258) 7/ 5

CrP 12 0 1*** 593 ± 259 2 / 10

* Linear loading; [N] = Newtons; SD = standard deviation; **tooth was repeatedly pressed out of the embedment mass without fracture or failure; ***tooth
was loaded until failure, but software did not record Fmax. C = direct restoration with composite; CP = direct restoration with endodontic glass-fiber post and
composite; V = veneer restoration, VP = veneer restoration and additional glass-fiber post; Cr = crown restoration; CrP = crown restoration and additional
glass-fiber post.

of this defect size. Valdivia et al36 reported contrary re-

sults. Endodontically treated incisors with composite res-
1400
torations with or without GFP showed higher fracture resis-
1200
tance compared to crown restorations. However, only linear
loading was performed. Earlier studies demonstrated that
1000 dynamic loading by TML prior to linear loading is crucial.34
Load capability [N]

Maximum load capacity values of directly restored ETT in

800 the present study were lower than those found in other
studies on direct Class III restorations.2,36 This can be ex-
600
plained by the absence of dynamic load simulation prior to
400
static loading, as described above. Likewise, load capaci-
ties of crown-restored teeth were lower than those reported
200 p = 0.014 by Heydecke et al.15 In that experimental design, ETT with
Class III cavities were restored with metal crowns and tita-
0 nium or ceramic posts. Median loads were 1038 N and
Composite Composite Veneer Veneer Crown Crown 1057 N, respectively. The same study included an addi-
+ post + post + post tional group with a composite core inserted 3 mm into the
root canal instead of a post. Maximum load capacity of
Fig 4 Box-and-whisker plots showing the distribution of Fmax values this group (750 N) was roughly comparable to the present
[N] in different test groups after TML (1.2 million cycles, 1 to 50 N) results (600 N).
and subsequent linear loading until failure. The bold printed bar rep-
An in vitro study by D’Arcangelo et al4 investigated load
resents the median value of maximum load capacity. О: mild outli-
ers; *: extreme outliers of maximum load capacity . capacity and deflection of ETT restored with composite and
porcelain veneers, with and without GFPs.4 The authors
concluded that veneer restorations appear to be an optimal
treatment option for anterior ETT. The explanation given
was that veneers allow a vestibular reinforcing effect sup-
quent linear loading compared to teeth directly restored ported by the non-prepared cevicopalatal trunk of the tooth,
with composite. There was no post effect, either for direct which is confirmed by the results of this study. In contrast
or for indirect restorative approaches. Therefore, both null to the present study, those authors4 tested ETT with only
hypotheses were rejected. an endodontic access cavity prepared.
The methods used in the present study are well docu- The present results of veneer restorations show that the
mented,8,22,34 and the technical procedures of TML and lin- veneer, a less invasive treatment option, is as loadable as
ear loading are frequently performed in in vitro studies.24,27,33 a crown restoration. Since there was no statistically signifi-
The results show that anterior ETT with Class III defects cant difference between groups of veneer and crown restor-
restored with composite present no differences in load ca- ations, veneers would appear to be the preferred treatment
pacity compared to indirect ceramic crown restorations. option in the tested scenario. An entirely non-prepared cer-
This supports direct restorations as appropriate and pos- vicopalatal part of the tooth as reinforcing structure can be
sibly equivalent treatment compared to crown restorations left untouched and seems beneficial as it is less invasive.

524 The Journal of Adhesive Dentistry

 Stein-Lausnitz et al

a b c

Fig 5 Nonrestorable fracture patterns. a: root fracture of tooth restored directly with composite and post; b: root fracture of a tooth restored
with veneer; c: root fracture below tip of glass fiber post, additional infraction of the crown restoration.

The current findings are in accordance with Valdivia et al,36 fined as catastrophic, in contrast to five catastrophic fail-
showing higher load capacities for veneer restorations com- ures in the post-free group with crown restorations, which
pared to crown restorations. was a statistically significant difference. This is in accor-
A point of discussion is the veneer material used in the dance with a previous study evaluating metal crowns, where
present study. To specifically compare the different tech- fewer catastrophic failures without posts were reported.15 A
niques, it would have been advantageous to choose a com- recent study on extensively damaged teeth also described
posite veneer and composite crown material. However, re- that the presence of a post negatively affected the failure
garding the results, the present authors assume that mode.18
composite veneers can achieve values similar to those of Based on the present findings and depending on the
ceramic veneers. The study by d’Arcangelo et al4 included specific clinical circumstances, one might conclude that di-
composite and ceramic veneers, and found no statistically rect composite restorations are less invasive and more time
significant difference between the two material types of ve- saving due to the need for only one treatment appointment,
neers, with maximum loads of 1303 N (composite veneer) and last but not least because they are a more economical
and 1168 N (porcelain veneer). alternative to any indirectly manufactured restoration from
The maximum load capacities of the present study show the dental laboratory, ie, veneer, incisal edge-up, or crown.
high values of standard deviation. This is a typical problem Veneer restorations have the advantages of good esthetics
in testing extracted teeth, although specimen preparation and a demonstrated strengthening effect under functional
was standardized as far as possible. Thus, it is a confound- loading condition. Crown restorations involve the highest
ing factor, which means that the results should be inter- amount of hard tissue loss during preparation11 and show
preted with caution. Furthermore, the study is limited by the load capacities similar to those of direct restorations. Thus,
moderate dimension of Class III defects. From the clinical they may only play a role only as a third-line intervention
perspective, a simulated higher loss of coronal hard tissue when restoring anterior ETT with Class III cavities.
would be of interest.
There was no significant effect of a post, either for direct
or indirect restoration tested. This is in accordance with
findings from the study mentioned above, where no effect
CONCLUSION
of GFP placement was shown for ETT with Class III cavities
restored with direct composite and veneer restoration.36 Endodontically treated maxillary central incisors with Class
Another study on anterior ETT with Class III restorations III defects restored with direct composite restorations pres-
with composite and different post systems showed that the ent load capacities comparable to indirect crown restor-
load capacity was not affected by the presence of an end- ations. Veneer restorations appear more advantageous
odontic post.2 The lack of a post effect can be explained by than crowns. Placement of glass-fiber posts has no positive
relatively well-preserved tooth structure of anterior ETT with effect.
bi-proximal Class III cavities compared to decoronated
teeth, which show a proven effect of post placement. In
contrast, one study evaluating anterior ETT with composite ACKNOWLEDGEMENTS
and ceramic veneer restorations showed that the place- The study was funded by Dentsply DeTrey. The authors thank
ment of a GFP significantly increased the load capacity with Rübeling+Klar dental lab, Berlin, especially Mr. Sven Schober, mas-
porcelain veneers.4 The presence or absence of an end- ter dental technician, and his colleague Mrs. Anne Friebel, dental
odontic post had no effect on failure modes in ETT restored technician, for their support of the present study. Furthermore
either with composite fillings or with veneer restorations. thanks go to Dr. René Tunjan, DDS, for his support with the embed-
Among crown restorations, ten of twelve failures were de- ding procedure. The authors deny any conflict of interest.

Vol 20, No 6, 2018 525

Stein-Lausnitz et al

REFERENCES 22. Naumann M, Preuss A, Frankenberger R. Load capability of excessively

flared teeth restored with fiber-reinforced composite posts and all-ce-
1. Abduljawad M, Samran A, Kadour J, Al-Afandi M, Ghazal M, Kern M. Ef- ramic crowns. Oper Dent 2006;31:699-704.
fect of fiber posts on the fracture resistance of endodontically treated an- 23. Naumann M, Preuss A, Rosentritt M. Effect of incomplete crown ferrules
terior teeth with cervical cavities: An in vitro study. J  Prosthet Dent on load capacity of endodontically treated maxillary incisors restored with
2016;116:80-84. fiber posts, composite build-ups, and all-ceramic crowns: an in vitro
2. Abduljawad M, Samran A, Kadour J, Karzoun W, Kern M. Effect of fiber evaluation after chewing simulation. Acta Odontol Scand 2006;64:31-36.
posts on the fracture resistance of maxillary central incisors with class III 24. Naumann M, Sterzenbach G, Proschel P. Evaluation of load testing of
restorations: An in vitro study. J Prosthet Dent 2016. postendodontic restorations in vitro: linear compressive loading, gradual
3. Castelnuovo J, Tjan AH, Phillips K, Nicholls JI, Kois JC. Fracture load and cycling loading and chewing simulation. J Biomed Mater Res B Appl Bio-
mode of failure of ceramic veneers with different preparations. J Prosthet mater 2005;74:829-834.
Dent 2000;83:171-180. 25. Peroz I, Blankenstein F, Lange KP, Naumann M. Restoring endodontically
4. D’Arcangelo C, De Angelis F, Vadini M, D’Amario M, Caputi S. Fracture re- treated teeth with posts and cores--a review. Quintessence Int
sistance and deflection of pulpless anterior teeth restored with compos- 2005;36:737-746.
ite or porcelain veneers. J Endod 2010;36:153-156. 26. Rodrigues MP, Soares PBF, Valdivia A, Pessoa RS, Verissimo C, Versluis
5. D’Arcangelo C, De Angelis F, Vadini M, Zazzeroni S, Ciampoli C, D’Amario A, Soares CJ. Patient-specific Finite Element Analysis of Fiber Post and
M. In vitro fracture resistance and deflection of pulpless teeth restored Ferrule Design. J Endod 2017;43:1539-1544.
with fiber posts and prepared for veneers. J Endod 2008;34:838-841. 27. Rosentritt M, Behr M, van der Zel JM, Feilzer AJ. Approach for valuating
6. Dammaschke T, Nykiel K, Sagheri D, Schafer E. Influence of coronal res- the influence of laboratory simulation. Dent Mater 2009;25:348-352.
torations on the fracture resistance of root canal-treated premolar and 28. Roulet JF, Van Meerbeek B. Editorial: Statistics: a nuisance, a tool, or a
molar teeth: a retrospective study. Aust Endod J 2013;39:48-56. must? J Adhes Dent 2007;9:287-288.
7. Dejak B, Mlotkowski A. The influence of ferrule effect and length of cast 29. Scotti N, Eruli C, Comba A, Paolino DS, Alovisi M, Pasqualini D, Berutti E.
and FRC posts on the stresses in anterior teeth. Dent Mater Longevity of class 2 direct restorations in root-filled teeth: A retrospective
2013;29:e227-237. clinical study. J Dent 2015;43:499-505.
8. DeLong R, Douglas WH. Development of an artificial oral environment for 30. Sequeira-Byron P, Fedorowicz Z, Carter B, Nasser M, Alrowaili EF. Single
the testing of dental restoratives: bi-axial force and movement control. crowns versus conventional fillings for the restoration of root-filled teeth.
J Dent Res 1983;62:32-36. Cochrane Database Syst Rev 2015:CD009109.
9. Demarco FF, Collares K, Coelho-de-Souza FH, Correa MB, Cenci MS, 31. Shu X, Mai QQ, Blatz M, Price R, Wang XD, Zhao K. Direct and Indirect
Moraes RR, Opdam NJ. Anterior composite restorations: A systematic re- Restorations for Endodontically Treated Teeth: A Systematic Review and
view on long-term survival and reasons for failure. Dent Mater Meta-analysis, IAAD 2017 Consensus Conference Paper. J  Adhes Dent
2015;31:1214-1224. 2018;20:183-194.
10. Edelhoff D, Prandtner O, Saeidi Pour R, Liebermann A, Stimmelmayr M, 32. Stavropoulou AF, Koidis PT. A systematic review of single crowns on end-
Guth JF. Anterior restorations: The performance of ceramic veneers. Quin- odontically treated teeth. J Dent 2007;35:761-767.
tessence Int 2018;49:89-101.
33. Sterzenbach G, Kalberlah S, Beuer F, Frankenberger R, Naumann M. In-vi-
11. Edelhoff D, Sorensen JA. Tooth structure removal associated with various tro simulation of tooth mobility for static and dynamic load tests: a pilot
preparation designs for posterior teeth. Int J  Periodontics Restorative study. Acta Odontol Scand 2011;69:316-318.
Dent 2002;22:241-249.
34. Sterzenbach G, Rosentritt M, Frankenberger R, Paris S, Naumann M.
12. Ferrari M, Cagidiaco MC, Goracci C, Vichi A, Mason PN, Radovic I, Tay F. Loading standardization of postendodontic restorations in vitro: impact of
Long-term retrospective study of the clinical performance of fiber posts. restorative stage, static loading, and dynamic loading. Oper Dent
Am J Dent 2007;20:287-291. 2012;37:71-79.
13. Goga R PD. The use of endodontically treated teeth as abutments for 35. Vadini M, De Angelis F, D’Amario M, Marzo G, Baldi M, D’Arcangelo C.
crowns, fixed partial dentures, or removable partial dentures: A literature Conservative restorations of endodontically compromised anterior teeth
review. Quintessence Int 2007;38:e106-e111. in paediatric patients: physical and mechanical considerations. Eur J Pae-
14. Heintze SD, Rousson V, Hickel R. Clinical effectiveness of direct anterior diatr Dent 2012;13:263-267.
restorations--a meta-analysis. Dent Mater 2015;31:481-495. 36. Valdivia AD, Raposo LH, Simamoto-Junior PC, Novais VR, Soares CJ. The
15. Heydecke G, Butz F, Strub JR. Fracture strength and survival rate of end- effect of fiber post presence and restorative technique on the biomech-
odontically treated maxillary incisors with approximal cavities after restor- anical behavior of endodontically treated maxillary incisors: an in vitro
ation with different post and core systems: an in-vitro study. J  Dent study. J Prosthet Dent 2012;108:147-157.
2001;29:427-433. 37. von Stein-Lausnitz M, Bruhnke M, Rosentritt M, Sterzenbach G, Bitter K,
16. Koelpin M, Sterzenbach G, Naumann M. Composite filling or single Frankenberger R, Naumann M. Direct restoration of endodontically
crown? The clinical dilemma of how to restore endodontically treated treated maxillary central incisors: post or no post at all? Clin Oral Investig
teeth. Quintessence Int 2014;45:457-466. 2018.
17. Kreulen CM, Creugers NH, Meijering AC. Meta-analysis of anterior veneer 38. Wiedhahn K, Kerschbaum T, Fasbinder DF. Clinical long-term results with
restorations in clinical studies. J Dent 1998;26:345-353. 617 Cerec veneers: a nine-year report. Int J  Comput Dent 2005;8:
18. Lazari PC, de Carvalho MA, Del Bel Cury AA, Magne P. Survival of exten- 233-246.
sively damaged endodontically treated incisors restored with different
types of posts-and-core foundation restoration material. J  Prosthet Dent
2018;119:769-776.
19. Mannocci F, Bertelli E, Sherriff M, Watson TF, Ford TR. Three-year clinical
Clinical relevance: Endodontically treated incisors with
comparison of survival of endodontically treated teeth restored with ei- Class III defects achieve an adequate load capacity irre-
ther full cast coverage or with direct composite restoration. J  Prosthet spective of direct (composite filling) or indirect ap-
Dent 2002;88:297-301.
proaches (veneer or crown) of restoration. An endodontic
20. Naumann M. Restorative procedures: effect on the mechanical integrity
of root-filled teeth. Endodontic Topics 2015;33:73-86. post does not seem necessary. Veneers enable hard tis-
21. Naumann M, Koelpin M, Beuer F, Meyer-Lueckel H. 10-year survival sue conservation, and are a highly esthetic and func-
evaluation for glass-fiber-supported postendodontic restoration: a pro- tional treatment approach.
spective observational clinical study. J Endod 2012;38:432-435.

526 The Journal of Adhesive Dentistry