Otology & Neurotology

29:953Y960 Ó 2008, Otology & Neurotology, Inc.

Mastoid and Epitympanic Bony Obliteration in

*Jean-Philippe Vercruysse, †Bert De Foer, *Thomas Somers,

*University Department of ENT; ÞDepartment of Radiology, A.Z. St-Augustinus Hospital, Antwerp, and
þDepartment of Radiology, A.Z. St-Jan AV, Bruges, Belgium

Objective: The primary goal of cholesteatoma surgery is com- the malleus handle and a sculpted allograft malleus or incus for
plete eradication of the disease. To lower the recurrence rate in columellar reconstruction.
the pediatric population in canal wall up techniques and to Main Outcome Measures: Recurrent rate; residual rate; func-
avoid the disadvantages of canal wall down techniques, the tional outcome; hygienic status of the ear; long-term safety
bony obliteration technique with epitympanic and mastoid issues.
obliteration has been developed. The objective of this study Results: The mean follow-up time was 49.5 months (range,
was to evaluate the long-term surgical outcome and recurrence 12Y101.3 mo). Recurrent cholesteatoma occurred in 1.9%
rate of this technique in children. (n = 1). Residual cholesteatoma was detected in 15.4% (n =
Study Design: Retrospective case review. 8) of the cases. Postoperative hearing results revealed a median
Setting: Tertiary referral center. gain on pure-tone averages of 14.3 dB and a median postopera-
Patients: Fifty-two children (G16 yr) were operated on in tive air-bone gap of 25.6 dB.
90.4% (n = 47) for a primary or recurrent cholesteatoma and Conclusion: The mastoid and epitympanic BOT is an effec-
in 9.6% (n = 5) for an unstable cavity. tive technique to lower the recurrence rate of cholesteatoma in
Intervention: In all cases, we closed the tympanoattical barrier the pediatric population. Follow-up by magnetic resonance
and the posterior tympanotomy with sculpted cortical bone and imaging provides a safe, noninvasive method for postopera-
then completed obliteration of the epitympanum and mastoid tive detection of residual cholesteatoma. Key Words: Bone
with bone pâté. A reconstruction of the middle ear was per- pâtéVCholesteatomaVMastoid obliterationVMastoidectomy.
formed by means of an allograft tympanic membrane including Otol Neurotol 29:953Y960, 2008.

The primary goal of the surgical treatment of chronic technique preserves the normal bony anatomy, avoids
otitis media with cholesteatoma is the complete eradica- the disadvantages associated with cavities, and has
tion of the disease (no residual disease), whereas second- shown better hearing results, but it has a significantly
ary goals are the prevention of recurrent disease, the higher recurrence rate. Retraction pocket formation in
improvement of the hygienic status of the ear, and the CWU ears can lead to recurrent cholesteatoma. The
preservation or improvement of hearing (1). Due to higher bony obliteration technique (BOT) consists of the meti-
rates of recurrent and residual disease, children present a culous bony reconstruction of the canal wall and of seal-
greater challenge than adults (2Y6). Various techniques ing off the middle ear by means of sculpted cortical bone
have been advocated in order to reach these goals. The chips and then by a complete obliteration of the drilled-
canal wall down (CWD) mastoidectomy provides lower out epitympanum and mastoid with bone pâté. This seems
recurrence rates, but it often requires regular cavity clean- to dramatically lower the incidence of recurrent disease.
ing and is associated with recurrent infection, water intol- In this report, we describe our experience with the BOT in
erance, caloric-induced vertigo, and the diminished ability a pediatric series of cholesteatoma cases and in cases with
to wear a hearing aid (7,8). The canal wall up (CWU) an unstable cavity after CWD surgery for cholesteatoma.
The indications, the surgical outcome including the recur-
rence rate, the residual rate, the functional results, and the
Address correspondence and reprint requests to Jean-Philippe otoscopic and imaging follow-up, are discussed. Our
Vercruysse, M.D., University Department of ENT, A.Z. St-Augustinus results should be considered preliminary until the com-
Hospital, Antwerp, Belgium; E-mail: [email protected] plete 5- and 10-year follow-up results become available.

953

Copyright @ 2008 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.
954 J.-P. VERCRUYSSE ET AL.

MATERIAL AND METHODS surgery. The second group comprised cases with an unstable
cavity. In both groups, the same surgical principal was applied.
We retrospectively evaluated a series of 52 consecutive chil- Surgery was performed under general hypotensive anesthesia
dren younger than 16 years. All children underwent a CWU- using facial nerve monitoring. A classic retroauricular incision
BOT operation at the University ENT Department of the was followed by the elevation of anteriorly based dermal and
Antwerp St-Augustine Hospital between September 1997 and musculoperiosteal flaps. Cortical bone chips were harvested
June 2006. Forty-seven children presented with a primary using a flat chisel and put aside. A bone pâté collector (Bess,
acquired (n = 16) or recurrent (n = 31) cholesteatoma and Berlin-Zehlendorf, Germany) and a cutting burr were used to
five children with a problematic or draining cavity. All surgery collect healthy bone pâté from the cortex of the mastoid and if
was performed by the two senior authors (E. O. and T. S.). needed from the squama of the temporal bone. Care was taken
The following outcome measures were analyzed: recurrent not to damage the soft tissues and not to harvest diseased bone.
rate, residual rate, functional outcome, hygienic status of the The bone pâté was mixed with an antibiotic solution (rifamycin
ear, and long-term safety issues. For this purpose, a database solution, 500 mg/10 ml) forming a semisolid paste. A cortical
was created, including age at surgery, sex, side, history of mastoidectomy and a wide posterior tympanotomy using the
surgery, surgical findings, otoscopic follow-up, and audiologi- CWU technique were performed. In contrast to the original
cal testing results with preoperative and postoperative air con- Mercke technique (9) and the technique described by Gantz
duction (AC), bone conduction (BC), air-bone gaps (ABGs), et al. (10), the posterior canal wall was left intact during the
and pure-tone averages (PTAs). Audiological assessment was whole procedure. Our aim was to preserve maximum vitality of
conducted every 3 months in the first postoperative year and the remaining bony canal wall, thus speeding up the healing
once yearly in the following postoperative years in a sound- process. The cholesteatoma, diseased soft tissue, ossicular rem-
treated room using a Madsen Electronics OB 822 and Inter- nants (eroded incus/malleus), and unhealthy bone were com-
acoustics AC33 Clinical Audiometer, calibrated according to pletely removed, and all cell tracts were cleaned.
ISO standards. No response to air-conducted sound was If the malleus handle was favorably positioned for columellar
coded as 120 dB, and no response to bone-conducted sound reconstruction to the head of the stapes or to the footplate, the
was coded as 80 dB. Missing values were coded as such. The head of the malleus was removed using a malleus nipper. Bone
postoperative anatomical status of the external auditory canal chips were sculpted and placed at the tympanoattical barrier and
(EAC) and tympanic membrane (TM) was evaluated by yearly posterior tympanotomy to completely seal off the epitympanum
otoscopic control, thus controlling for the presence of retraction and mastoid from the middle ear cavity. Lesions of the scutum
pockets, canal wall breakdown, or recurrent cholesteatoma. The and the bony canal wall were carefully reconstructed with
presence of residual cholesteatoma was visually evaluated dur- sculpted solid cortical bone. In case of a radical cavity, the
ing planned or functional second stage surgery or by the com- epithelial lining and all pathological remnants were first
bination of high-resolution computed tomography (HRCT) and removed, and the remaining mastoid cavity was adequately
magnetic resonance imaging (MRI). Recurrent cholesteatoma is checked and drilled to remove mucosal remnants. Harvested
defined as a new cholesteatoma developing from an unsafe, cortical bone was sculpted to form a new bony canal wall in
nonYself-cleaning retraction pocket. Residual cholesteatoma is continuity with the complete closure of the tympanoattical bar-
defined as keratinizing squamous epithelium left behind during rier. The paratympanic space was thus completely isolated from
first-stage surgery, which has regrown into a visually identifi- the middle ear cavity by a solid bony partition. It was then
able cholesteatoma. Second-look surgery using a retroauricular progressively and completely filled up with bone pâté, up to
(n = 23) or transmeatal approach (n = 19) was executed after the level of the cortex. An M-meatoplasty according to Mirck
12 months. The decision to stage was taken by the surgeon (11) was often performed to optimize the size of the external
during the first-stage surgery based on the extent and character- meatus, thus stimulating the self-cleaning capacity of the outer
istics of the cholesteatoma and on the surgical complexity of the ear canal. The middle ear reconstruction was performed using a
anatomy. Although the bony reconstruction of the canal wall tympano-ossicular allograft (TOA) (12). The allograft consisted
and of the tympanoattical barrier was always performed at the of a meatal periosteal cuff in continuity with the TM and mal-
first stage, in 23 cases, the obliteration of the mastoid and attic leus handle. The malleus head was removed with a malleus
space with bone pâté was postponed until the second stage for nipper at the level of the lateral process of the malleus. The
safety reasons, taking into account the possibility of residual allograft TM (with malleus handle) was rotated clockwise (left
disease. During second-look surgery, subsequent bony oblitera- ear) or counterclockwise (right ear) to place the malleus handle
tion and, if needed, functional correction were performed. in an advantageous position, perpendicularly centered above
When we considered the odds for residual disease negligible, the oval window. This allows for the most effective columellar
no second-stage surgery was performed. However, all patients energy transduction between the implanted malleus handle and
were regularly followed up by yearly micro-otoscopy and by the stapes or stapes footplate. The ossicular reconstruction was
HRCT and MRI including nonYecho-planar diffusion-weighted executed using a remodeled allograft incus or malleus. If
imaging (nonYEPI DWI) sequence at 1 and 5 years after sur- needed, a thin silastic sheet (0.5 mm) was placed extending
gery. Adequate long-term imaging follow-up of obliterated from the protympanum to the retrotympanum, to avoid fibrous
mastoids is compulsory to prevent late complications due to adhesions and to promote the regrowth of healthy middle ear
residual cholesteatoma. Counts, percentages, histograms, and mucosa during postoperative healing. Perioperative intravenous
box plots were used to describe nominal data. Statistical anal- antibiotics (cefazoline) were continued for 24 hours. Patients
ysis was performed using a t test, and significance was defined were sent home with amoxicillin-clavulanate or cefuroxim in
as p G 0.05. case of penicillin allergy for 5 days. TOA are harvested and
prepared at the tissue bank of the St-Augustine Hospital,
Surgical Technique according to the standards of the Belgian law (Belgisch Staats-
The BOT was applied in two subgroups of the study popula- blad 13.6.86). Immediately after removal from the cadaver, the
tion. The first group comprised cases with either primary cho- grafts are fixed for at least 2 weeks in a solution of 4.5%
lesteatoma or recurrent cholesteatoma after previous CWU buffered formaldehyde. After dissection, tissues are preserved

Otology & Neurotology, Vol. 29, No. 7, 2008

Copyright @ 2008 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.
 BONY OBLITERATION IN PEDIATRIC CHOLESTEATOMA 955

TABLE 1. General patients characteristics (n = 52) cholesteatoma in the third year of otoscopic follow-up
(Figs. 1 and 2). In this case, revision surgery revealed the
Mean age 11.6 years (range, 5Y16 yr; SD, 2.81)
Sex (M/F) 34:18
presence of a mesotympanic atelectatic TM, partial
History of surgery 36 patients (69.2%) resorption of the tympanoattical barrier with extension
Mean postoperative 49.6 months (range, 12Y101 mo) of the retraction into the attic, and presence of cholestea-
follow-up toma (Fig. 3). Reclosure of the epitympanum by means of
bone chips and pâté, in combination with cartilage tympa-
noplasty, was performed. During subsequent follow-up,
in Cialit (15,000 aqueous solution of a sodium salt of an orga-
nomercuric compound) for a period of 3 weeks to 2 months. no new recurrence developed. In three other patients, the
This study does not address the alleged risk of transfer of self-cleaning retraction was surgically corrected using
infectious diseases, such as Creutzfeldt-Jacob disease (CJD) cartilage reinforcement during planned functional sur-
and human immunodeficiency virus (HIV) infections. Trans- gery. The fifth retraction pocket remained self-cleaning
mission of CJD has been reported after implantation of dura and stable during follow-up. The percentage of ears that
mater (13) or corneal grafts (14). It has, however, never been remained without recurrent disease was 98.1% (n = 51). In
reported after transplantation of tissues other than brain, cada- 52% (n = 27) of all cases, a preliminary (n = 6), first-stage
veric dura matter, or corneal grafts. In addition, the incidence of (n = 20) or second-stage (n = 1) M-meatoplasty has been
CJD is extremely low (1: 1,000,000), and the applied stringent performed to widen the external meatus, thus enhancing
criteria for donor selection exclude donors at risk for CJD. No
the self-cleaning capacity of the outer ear canal. Second
reports of transmission of HIV by nonvital allograft material
have appeared in the literature. Formaldehyde, used in the pre- staging by means of second look tympanotomy (n = 42) or
servation of allografts, is known to inactivate HIV readily (15). imaging follow-up (HRCT and nonYEPI DWI; n = 31)
revealed in eight cases the presence of a residual choles-
RESULTS teatoma pearl, each of which was present in the tympanic
cavity (15.4%). No residual cholesteatoma within the
Fifty-two children underwent the BOT. Thirty-four bony obliteration was detected to date.
(65%) were males and 18 (35%) were females. One One case of the studied population needed readmission
child had a cleft palate. The mean age was 11.6 years for wound infection, which was treated with intravenous
(range, 5Y15 yr). Thirty-six patients (69.2%) had a his- antibiotics. This case was one of the first cases in our
tory of ear surgery. The mean postoperative follow-up series and did not receive preoperative and postoperative
period was 49.6 months (range, 12Y101.3 mo; Table 1). antibiotics. Following this event, antibiotics became the
Twenty-one (40.4%) of the children had a follow-up rule. No other complications (facial nerve, SNHL, bone
period of 5 years. Three patients were followed for resorption, or canal wall breakdown) occurred in this
more than 8 years. At latest follow-up, a safe, dry, studied population. A list of complications is summa-
and trouble-free graft was present in 46 children rized in Table 2.
(88.5%). One patient developed a perforation after an A reconstruction of the ossicular chain was attempted
acute otitis media during the sixth postoperative year. during primary surgery in all patients. One of the recon-
In five patients, otoscopic follow-up revealed the pre- structive advantages of TOAs is that they allow a stable
sence of a self-cleaning mesotympanic retraction pocket, positioning of the TM. The fibrous annulus is anchored
one of which progressively evolved toward a recurrent in the patient’s bony annulus, and the periosteal cuff of

FIG. 1. Histogram showing population follow-up data (n): Aspect of the tympanic membrane (TM); normal, retraction, perforation TM, or
cholesteatoma. An asterisk shows the occurrence of a recurrent cholesteatoma after 3 years of follow-up (see text for further details).

Otology & Neurotology, Vol. 29, No. 7, 2008

Copyright @ 2008 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.
956 J.-P. VERCRUYSSE ET AL.

FIG. 2. Histogram showing population folow-up data (% of total population): Aspect of the tympanic membrane (TM): normal TM,
retraction TM, perforation TM, or cholesteatoma. An asterisk shows the occurrence of a recurrent cholesteatoma (see text for further
details).

the graft is securely glued in place in the bony EAC. The detected at second stage or by follow-up MRI, the colu-
implanted malleus handle forms an integral part of the mella can be removed with the pearl and a new ossicular
graft and allows for stable anchoring of the remodeled reconstruction can be made.
allograft ossicle as a columellar reconstruction to stapes At primary surgery, the stapes superstructure was
or stapes footplate. If a residual cholesteatoma pearl is absent in 61.5% (n = 32) due to destruction by primary

FIG. 3. Postoperative imaging performed after primary bony obliteration. A, Coronal reformation HRCT image of a patient showing the
presence of an atelectatic tympanic membrane with an attical soft tissue lesion causing a partial destruction of the attico-tympanal barrier,
suggestive for the presence of a recurrent cholesteatoma (arrows) of the right ear. B, Coronal reformation HRCT images of a patient with
normal postoperative bony obliteration with an homogeneous obliteration (asterisk) and middle ear aeration (arrow). C, Coronal non echo-
planar diffusion weighted image (same patient as Fig A) showing a nodular hyperintense lesion on the right side (arrow), characteristics for the
presence of a recurrent cholesteatoma. D, Coronal non echo-planar diffusion weighted image (same patient as Fig D) showing no clear
nodular hyperintensity.

Otology & Neurotology, Vol. 29, No. 7, 2008

Copyright @ 2008 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.
 BONY OBLITERATION IN PEDIATRIC CHOLESTEATOMA 957

TABLE 2. Postoperative complications after using the BOT TABLE 3. Detailed summary of the postoperative hearing
(see text for further details) levels of 52 children for whom full audiometric data were
available 1 year postoperatively
Postoperative wound infection 1
Perforation 1 Percentage ABG closures (n = 52)
Mesotympanic self-cleaning retraction 4 (5)
Residual cholesteatoma 8 0Y10 dB 13.4% (n = 7)
Recurrent cholesteatoma 1 11Y20 dB 30.8% (n = 16)
21Y30 dB 25% (n = 13)
931 dB 30.8% (n = 16)
The hearing results (PTA) are expressed in percentages of ABG
or recurrent cholesteatoma. A fixed footplate was found in closures.
5.7% of the patients (n = 3). The median preoperative
PTA-AC was 51.67 dB with a median preoperative
PTA-ABG of 43.32 dB. Postoperative hearing results persisted because of the lack of middle ear aeration or
were assessed after 1 year and revealed a median gain because of a fixed footplate (n = 3).
on PTA of 15 dB and median postoperative ABG of
25.6 dB. The postoperative ABG closure after ossicular
reconstruction is presented in Figure 4 and Table 3. No DISCUSSION
statistically significant difference was found between pre-
operative and postoperative bone conduction ( p 9 0.05; The goals of the surgical treatment of middle ear cho-
t test). In 44% (n = 23/52) of the patients, a functional lesteatoma are the complete eradication of the disease,
correction was attempted during second-look surgery, the prevention of recurrent cholesteatoma, the restoration
using a remodeled malleus (n = 18) or incus allograft of the hygienic status of the ear, and the preservation or
(n = 5). In several children, a conductive hearing loss improvement of the hearing (1). Two basically different

FIG. 4. Pre- and postoperative hearing levels in 52 children for whom full audiometric data were available 1 year postoperatively. Box and
Whisker plots showing pre- and postoperative hearing thresholds (PTA). Preop BC, preoperative bone conduction threshold; Preop AC,
preoperative air conduction threshold; Postop AC: postoperative air conduction threshold; Postop BC, postoperative bone conduction
threshold; Gain AC, difference between Preop AC and Postop AC. Whiskers, minimum values; Large rectangles, 25thY75th percentile;
Small horizontal black bars, median values; Dots, outlying value.

Otology & Neurotology, Vol. 29, No. 7, 2008

Copyright @ 2008 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.
958 J.-P. VERCRUYSSE ET AL.

techniques have been advocated to reach these goals: Moreover, our results are not inferior to the recurrence
CWU and CWD techniques. The advantages of the rate reported by authors using the CWD technique
CWD technique are as follows: no need for staging, a (3,6Y8). It must be emphasized that our results must
lower rate of residual cholesteatoma, and a lower rate of be considered preliminary until the full 5- and 10-year
recurrent cholesteatoma. On the other hand, the advan- follow-up results of the presented series have become
tages of the CWU technique are as follows: a better available. Therefore, it is obvious that long-term, yearly
hygienic status of the ear and a better functional out- otoscopic follow-up remains the criterion standard in
come (2Y5). The disadvantages of the CWD technique cholesteatoma surgery. The residual rate in our study
are associated higher morbidity, such as the need for was 15.4%, as evaluated by the combination of staging
regular cleaning, recurrent infections, water intolerance, and postoperative imaging. It is lower in comparison
caloric-induced vertigo, and the difficulty to wear a with the results of our previously published conven-
hearing aid if needed; and a worse functional outcome. tional CWU study on pediatric cholesteatoma, which
The disadvantages of the CWU technique are the need showed a residual rate of 27% (16). To date, no residual
for second and potentially further staging, the need for cholesteatoma has been detected within the obliterated
long-term follow-up to detect recurrent cholesteatoma, paratympanic spaces. However, given the nonnegligi-
higher rates of residual disease, and higher rates of ble rate of residual disease, it remains compulsory to
recurrent disease (1Y8). apply a very strict follow-up. In all our cases nowadays,
In 1997, we decided to find out whether it is possible a nonYEPI DWI control is executed or planned at 5 years
to combine the advantages and, at the same time, avoid after surgery (Long-Term Safety Issues).
the disadvantages of both the CWU and CWD tech-
niques: the mastoid and epitympanic bony obliteration.
As a comparison basis for our study, we used our pre- Functional Outcome
viously published results of the conventional CWU with The functional outcome at 1 year of this series seems
TOA reconstruction (16) and the results of the CWU and somewhat poorer or similar than those reported by
CWD techniques available in the literature. some others in CWU surgery (1,3,5,7,16) but compara-
Many materials have been used to obliterate the mas- ble to those reported by Gantz et al. (10) after pediatric
toid cavity (17) including autologous material as fascia, bony mastoid obliteration. Despite the fact that these
fat, vascularized musculoperiosteal flaps, cortical bone children experienced poor hearing with a median pre-
chips, cartilage and bone pâté, and other biocompat- operative PTA-AC of 51.67 dB and median preopera-
ible materials such as hydroxyapatite granules/cement tive PTA-ABG of 43.3 dB, postoperative measurements
(18Y19) and demineralized bone matrix (20). Preserva- revealed a marked improvement with acceptable hear-
tion and/or reconstruction of a normal anatomy of the ing results with a median postoperative PTA-ABG
EAC and TM are possible in most patients. The presence of 25.6 dB, a median postoperative PTA-AC of 32.5
of a solid bony barrier and obliteration of the mastoid dB, and PTA-ABG closure 20 dB or less in 30.9%
with bone pâté close off the epitympanum and mastoid (Table 3).
from the middle ear cavity and seem to lower the inci- The improvement is probably due to the fact that we
dence of new retractions of the TM. now position the malleus handle of the TOA in a more
advantageous position, as compared to the conventional
position of the monobloc TOA. In the latter case, a full-
Rate of Recurrent and Residual Cholesteatoma monobloc TOA, consisting in a TM, malleus, incus,
In our study, the recurrence rate was 1.9% after a and the anterior stapedial crus with half a footplate, is
mean follow-up of 49.6 months (range, 12Y101.3 mo). used for the reconstruction of the middle ear. In our
This compares very favorably to the outcome of our actual series, we used a columellar reconstruction bet-
previously published series of the CWU technique in ween the malleus handle and the stapes head or foot-
a pediatric population, which showed a recurrence rate plate. However, by a slight rotation of the graft, the
of 18% after a mean follow-up of 54 months (range, malleus handle is placed in a more favorable position,
12Y191 mo) (16). It seems safe to infer that bony oblit- perpendicular to the center of the oval window. This
eration enhances the biological stability of the ear, allows for the most effective energy transduction.
probably by reducing the size of the middle ear cell Although a columellar reconstruction seems to achieve
system, thus decreasing the total surface of mucosal better functional results, the conventional drawbacks are
lining and diminishing its capacity for gas absorption its long-term instability and its tendency toward extru-
and/or inflammation. Our results confirm the lower sion. The use of TOAs for columellar reconstruction
recurrence rates of the BOT in a pediatric population offers a distinct advantage. The malleus handle is firmly
reported by Gantz et al. (10). Other reports on complete incorporated in the allograft TM and, as such, forms a
bony obliteration showed similar results in an adult and stable anchor point for the columella, while at the same
childhood cholesteatoma population (9,10,17Y23). The time, it protects against columellar extrusion. The func-
results also compare favorably to the outcome of other tional outcome of our series compares poorly to the re-
papers reporting on the results of the conventional sults of noncholesteatoma chronic otitis surgery. This is
CWU technique in pediatric cholesteatoma (3Y5). due to the high percentage of cases in which the stapes

Otology & Neurotology, Vol. 29, No. 7, 2008

Copyright @ 2008 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.
 BONY OBLITERATION IN PEDIATRIC CHOLESTEATOMA 959

superstructure was absent (61.5%), to the fixed footplate 2) The cost of the expensive gadolinium is avoided. 3)
in 3 cases, and to the fact that, in cholesteatoma ears, an The examination becomes much less burdening for the
essential precondition for functional success, that is, nor- patient on the emotional level. In conclusion, the finan-
mal middle ear aeration, is often lacking. cial and emotional cost of follow-up screening for resi-
dual disease of cholesteatoma becomes negligible in
Hygienic Status of the Ear comparison with the cost of exploratory surgery in
At latest follow-up, a safe, dry, and trouble-free graft cases without residual disease.
was observed in 46 children (88.5%). All patients are still
followed up by yearly otoscopic control. With the excep-
tion of the perforated case, all ears are dry, self-cleaning,
and water-resistant to date. The combined presence of a CONCLUSION
normal-sized external meatus, achieved in 52% of the
Our results indicate that the BOT, performed in chil-
cases by means of an M-meatoplasty, of a normal-sized
dren with primary/recurrent cholesteatoma or unstable
ear canal protected by a solid bony wall, and of a TM
cavities reconstruction of cavities, is a very useful tech-
well placed in its normal position seems to provide the
nique to deal with the higher rates of residual and recur-
ideal basis for a stable hygienic condition of the ear.
rent cholesteatoma in the pediatric population. The
results show that the incidence of recurrent cholestea-
Long-Term Safety Issues
toma has dramatically declined. Acceptable functional
The long-term safety of the ear is a primary concern in
results have been achieved. The problem of residual cho-
cholesteatoma surgery, especially in children. Therefore,
lesteatoma can be safely dealt with by staging and/or
surgical staging to detect residual cholesteatoma has been
postoperative imaging, using a combination of CT and
advocated by most authors (1Y5,9,10). Adequate imaging
new MRI techniques including nonYEPI DWI MRI. We
follow-up of obliterated mastoids is necessary to prevent
are using this surgical technique in the majority of cho-
late complications when residual cholesteatoma remains
lesteatoma cases.
buried in the bony obliterated mastoid. HRCT has been
shown to be very effective in excluding and detecting
small pearls in obliterated mastoids, presenting as
punched-out lesions in the dens bone (24,25). However, REFERENCES
in the case of associated opacified lesions, further char-
1. Shelton C, Sheehy JL. Tympanoplasty: review of 400 staged cases.
acterization of the lesion by MRI will be necessary. Until Laryngoscope 1990;100:679Y81.
recently, small residual pearls (G5 mm) could not be 2. Shohet JA, de Jong AL. The management of pediatric cholestea-
detected using DWI (24,26). toma. Otolaryngol Clin North Am 2002;35:841Y51.
The recent availability of the nonYEPI DWI MRI 3. Darrouzet V, Duclos JY, Portmann D, et al. Preference for the
closed technique in the management of cholesteatoma of the mid-
sequence offers a safe, noninvasive, selective, sensitive, dle ear in children: a retrospective study of 215 consecutive patients
and comparatively cheap alternative to exploratory treated over 10 years. Am J Otol 2000;21:474Y81.
staged surgery. It allows for the detection of cholestea- 4. Sanna M, Zini C, Gamoletti R, et al. The surgical management of
toma pearls down to a size of 2 mm (27Y29). Because we childhood cholesteatoma. J Laryngol Otol 1987;101:1121Y6.
allow for the possibility that it takes more than 1 year for 5. Schuring AG, Lippy WH, Rizer FM, et al. Staging for cholestea-
toma in the child, adolescent and adult. Ann Otol Rhinol Laryngol
a residual pearl to develop into a detectable lesion, we 1990;99:256Y60.
now routinely image all our cholesteatoma cases at 1 and 6. Vartiainen E, Nuutinen J. Long term results of childhood choles-
5 years postoperatively. MRI thus effectively replaces teatoma. Int J Pediatr Otorhinolaryngol 1992;24:201Y8.
second-stage surgery in our department for 2 years. Off- 7. Brown JS. A ten-year statistical follow-up of 1142 consecutive
cases of cholesteatoma: the closed vs. the open technique. Laryn-
setting the cost of surgical staging to the cost of twice- goscope 1982;92:390Y6.
repeated MRI, this protocol is not only less burdening to 8. Nyrop M, Bonding P. Extensive cholesteatoma: long-term results
the patient on a practical and emotional level but also of three surgical techniques. J Laryngol Otol 1997;111:521Y6.
compares most favorably on the budgetary level for both 9. Mercke U. The cholesteatomatous ear one year after surgery with
obliteration technique. Am J Otol 1987;8:534Y6.
the patient and the public health system. Our analysis
10. Gantz BJ, Wilkinson EP, Hansen MR. Canal wall reconstruction
showed that the full HRCT and MRI protocol, repeated tympanomastoidectomy with mastoid obliteration. Laryngoscope
twice, is six times cheaper than exploratory surgery 2005;115:1734Y40.
within the Belgian context. We recently completed the 11. Mirck PG. The M-meatoplasty of the external auditory canal.
validation process of the nonYEPI DWI MRI sequence Laryngoscope 1996;106:367Y9.
12. Marquet J. Twelve years experience with homograft tympano-
(29). Therefore, we now use only this sequence in the plasty. Otolaryngol Clin North Am 1977;10:581Y93.
follow-up screening for residual cholesteatoma. For 13. Martinez-Lage J, Poza M, Sola J, et al. Accidental transmission of
screening purposes, we deem the use of HRCT and the Creutzfeldt-Jacob disease by dural cadaveric grafts. J Neurol Neu-
use of the MRI sequence 45 minutes after injection with rosurg Psychiatry 1994;57:1091Y4.
14. Heckmann JG, Lang CJ, Petruch F, et al. Transmission of
gadolinium no longer necessary. The advantages are evi- Creutzfeldt-Jakob disease via a corneal transplant. J Neurol Neu-
dent. 1) The nonYEPI DWI sequence by itself takes less rosurg Psychiatry 1997;63:388Y90.
than 5 minutes to execute. On the logistic level, this op- 15. Janssens de Varebeke S, Goubau P, et al. Tympano-ossicular allo-
timizes patient put-through in the radiology department. grafts and HIV transmission. Am J Otol 1998;19:704Y8.

Otology & Neurotology, Vol. 29, No. 7, 2008

Copyright @ 2008 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.
960 J.-P. VERCRUYSSE ET AL.

16. Schilder AG, Govaerts PJ, Somers T, et al. Tympano-ossicular resolution computed tomography and magnetic resonance imaging
allografts for cholesteatoma in children. Int J Pediatr Otolaryngol in the detection of residual cholesteatomas in primary bony obliter-
1997;18:31Y40. ated mastoids. Am J Otolaryngol 2007;28:230Y4.
17. Black B. Mastoidectomy elimination: obliterate, reconstruct, or 25. Yung MM, Karia KR. Mastoid obliteration with hydroxyapatiteV
ablate? Am J Otol 1998;19:551Y7. the value of high, resolution CT scanning in detecting recurrent
18. Yung MW. The use of hydroxyapatite granules in mastoid oblit- cholesteatoma. Clin Otolaryngol 1997;22:553Y7.
eration. Clin Otol 1996;21:480Y4. 26. Vercruysse JP, De Foer B, Pouillon M, et al. The value of diffusion-
19. Mahendran S, Yung MW. Mastoid obliteration with hydroxyapatite weighted MR imaging in the diagnosis of primary acquired and
cement: the Ipswich experience. Otol Neurotol 2004;25:19Y21. residual cholesteatoma: a surgical verified study of 100 patients.
20. Leatherman BD, Dornhoffer JL. The use of demineralised bone Eur Radiol 2006;16:1461Y7.
matrix for mastoid cavity obliteration. Otol Neurotol 2004; 27. De Foer B, Vercruysse JP, Vertriest R, et al. Echo planar diffusion-
25:22Y5. weighted MR imaging versus spin echo planar diffusion-weighted
21. Charachon R, Gratacap B, Tixier C. Closed versus obliteration MR imaging in the detection of acquired middle ear cholesteatoma:
technique in cholesteatoma surgery. Am J Otol 1988;9:286Y92. case report. AJNR Am J Neuroradiol 2006;27:1480Y2.
22. Robertson JB Jr, Mason TP, Stidham KR. Mastoid obliteration: 28. De Foer B, Vercruysse JP, Bernaerts A, et al. The value of single-
autogenous cranial bone pâté reconstruction. Otol Neurotol 2003; shot turbo spin-echo diffusion-weighted MR imaging in the detec-
24:132Y40. tion of middle ear cholesteatoma. Neuroradiology 2007;49:841Y8.
23. Lee WS, Choi JY, Song MH, et al. Mastoid and epitympanic 29. De Foer B, Vercruysse JP, Bernaerts A, et al. Detection of post-
obliteration in canal wall up mastoidectomy for prevention of operative residual cholesteatoma with non echo-planar diffusion-
retraction pocket. Otol Neurotol 2005;26:1107Y11. weighted magnetic resonance imaging. Otol Neurotol 2008;
24. De Foer B, Vercruysse JP, Pouillon M, et al. Value of high- 29:513Y7.

Otology & Neurotology, Vol. 29, No. 7, 2008

Copyright @ 2008 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.