The American Journal of Otology 21:774781 2000, The American Journal of Otology, Inc.

Extensive Intratemporal Cholesteatoma: Surgical Strategy

Alexis Bozorg Grayeli, Isabelle Mosnier, Hani El Garem, Didier Bouccara, and Olivier Sterkers
Department of OtolaryngologyHead Neck Surgery, Hopital Beaujon, AP-HP, Universite Paris 7, Clichy, France

Objective: To evaluate the decisional elements in the surgical strategy for extensive intratemporal cholesteatomas. Study Design: A retrospective review of cases followed up between 1985 and 1996. Setting: Tertiary referral center. Patients: Nineteen patients with temporal bone cholesteatoma extending beyond the middle ear limits and surgically treated were included. Preoperative imaging distinguished apical (8), infralabyrinthine (3), supralabyrinthine (3), retrolabyrinthine (1), and translabyrinthine (4) cholesteatomas. Intervention: Apical and supralabyrinthine lesions were treated through a middle fossa approach. Infralabyrinthine and translabyrinthine locations were exposed through a subtotal petrosectomy or a transotic route, depending on the preoperative audiovestibular status and labyrinthine destruction on computed tomography. The retrolabyrinthine lesion was approached through a retrolabyrinthine route. Main Outcome Measures: Patients were assessed for postoperative audiologic and facial functions and for recurrence of tumor.

Results: The facial nerve was neither rerouted nor interrupted during surgery. Among the 12 patients with preoperative facial palsy (FP), 5 cases of improvement (42%), 6 cases of stable function (50%), and 1 case of mild deterioration (8%) were observed postoperatively. In patients without preoperative FP, facial function remained unchanged postoperatively. The labyrinth could be preserved in three patients (16%), with postoperative stable hearing function in two (11%), and a 40-dB mean auditory deterioration in one (5%). Complete macroscopic resection was obtained in all patients. Two cases (11%) of postoperative recurrence were observed. Conclusion: The surgical strategy, principally based on cholesteatoma location and preoperative auditory function, yielded a high rate of local disease control and facial function preservation. Key Words: Intratemporal cholesteatomaFacial nerveFacial palsy. Am J Otol 21:774781, 2000.

Although chronic otitis media has dramatically decreased in developed countries because of a better understanding of its pathophysiology and its management in early stages, the incidence of extensive cholesteatomas seems constant (1). These lesions remain a great surgical challenge with two principal goals: maximum functional preservation and minimum risk of recurrence. Many approaches to cranial base lesions have been described, whose aim is to limit the risk to the adjacent neurovascular structures (25). Nevertheless, each approach carries inherent anatomic limitations and should be tailored to individual cases. Consequently, a precise preoperative examination determines the surgical strategy and the postoperative functional results. Because these cranial base lesions vary so greatly, a

simple and exhaustive strategic scheme cannot be proposed. However, major decisional factors can be highlighted, and principal surgical attitudes can be deduced for each type of lesion and location. The aim of this retrospective study was to assess preoperative decisional elements and to analyze surgical strategy in the light of the postoperative results and the pathophysiology of extensive cholesteatoma. MATERIALS AND METHODS
A retrospective study of 19 patients undergoing surgery for cholesteatoma extending beyond the middle ear limits between 1985 and 1996 in our department was undertaken. Data from medical files were collected concerning medical history, preoperative symptoms, neurologic and audiovestibular clinical status, audiometric and vestibular test results, facial electromyography (EMG) and blink reflex test results in cases of clinical abnormality or encasement of the facial nerve in

Address correspondence and reprint requests to Pr. Olivier Sterkers, Service dOto-Rhino-Laryngologie, Hopital Beaujon, 100 Boulevard General Leclerc, F-92118, Clichy Cedex, France.

774

EXTENSIVE INTRATEMPORAL CHOLESTEATOMA

775

FIG. 1. Intratemporal cholesteatoma extension types according to Fisch et al.(4): supralabyrinthine (A), apical (B), infralabyrinthine (C), retrolabyrinthine (D), and translabyrinthine (E).

cholesteatoma, and radiologic explorations, including computed tomography (CT) and magnetic resonance imaging (MRI). The type of approach, observations on cholesteatoma extent during surgery, the surgical sacrifice of neurovascular structures, and the quality of macroscopic resection were also noted. Postoperative data were obtained regarding complications, neurologic deficits, audiovestibular tests, and radiologic appearance. Cholesteatoma locations and extensions assessed on preoperative CT were classified as apical, supralabyrinthine, infralabyrinthine, retrolabyrinthine, and translabyrinthine according to Fisch and Mattox (4). This classification is summarized in Figure 1. Facial function was clinically assessed according to House and Brackmann (6). Mean hearing thresholds were calculated by air conduction audiometry at 500, 1,000, and 2,000 Hz. These were considered to be within normal limits when <25 dB, mild hearing loss when 25 and <45 dB, moderate hearing loss when 45 and <65 dB, severe hearing loss when 65 and <85 dB, and profound hearing loss when 85 dB. Values are shown as mean standard deviation. The statistical test used in this study was an unpaired t test. Significant difference was considered to be p < 0.05.

of a history of otitis media, temporal bone surgery, or trauma. The population was composed of 12 male and 7 female patients (sex ratio 1.7). The mean age was 41 years (range 1660). The mean follow-up period was 27 months (range 294). Twelve patients with cholesteatoma were operated on for the first time (63%), and 7 were surgically treated for recurrence (37%). Presenting symptoms The presenting symptoms are summarized in Table 2. Facial paresis was present in apical cholesteatomas in six of eight patients (86%). Apical lesions were observed in three of four patients with intense headaches (75%). Neither audiovestibular symptoms nor any other group of symptoms was associated with a particular cholesteatoma location. Three cases were revealed by systematic CT performed for recurrent bacterial meningitis (16%). In these three patients, CT showed an intracranial extension of the cholesteatoma in the supralabyrinthine, retrolabyrinthine, and apical regions. Preoperative audiovestibular and neurologic status Otoscopy confirmed the diagnosis by visualization of the tumor in 12 patients (63%). The remaining patients had a normal tympanic membrane (5 patients, 26%), a tympanic perforation (1 patient, 5%) or a radical mastoidectomy cavity without evidence of cholesteatoma recurrence (1 patient, 5%). Preoperative hearing test results were available in 18 patients (95%). These are summarized in Table 3. Ten patients (53%) experienced preoperative profound hearing loss. In this group, six had previously undergone surgery on the same ear. The profound hearing loss was associated with destruction of labyrinthine bone on preoperative CT in seven patients (70%) and with a temporal bone fracture involving the cochlea in three patients (30%). Among the nine remaining patients with preoperative serviceable hearing, two (22%) had a large cochlear fistula on the basal turn on preoperative imaging (Fig. 2 A,C), and one (5%) showed a lateral semicircular canal fistula. Five patients (26%) showed clinical vestibular signs preoperatively. Rombergs sign was observed in all these
The American Journal of Otology, Vol. 21, No. 6, 2000

RESULTS Population The population comprised 17 patients with acquired cholesteatoma (89%) defined by an abnormal tympanic membrane on clinical examination and/or medical history of otitis media, middle ear surgery, or middle ear trauma (Table 1). Two congenital cholesteatomas were observed (11%). These were defined by a normal tympanic membrane on clinical examination and the absence
TABLE 1. Medical history in acquired cholesteatomas
Medical history Middle ear cholesteatoma, previously operated on Chronic otitis media, not operated on Acute mastoiditis Temporal bone fracture None Total No. of patients (%) 9 (47) 4 (21) 1 (5) 3 (16) 2 (11) 19

776

A. B. GRAYELI ET AL.
TABLE 2. Presenting clinical symptoms in different cholesteatoma extension types

Extension type Apical (8) Infralabyrinthine (3) Supralabyrinthine (3) Translabyrinthine (4) Retrolabyrinthine (1) Total (19)

Otalgia 1 1 (5%)

Intense headaches 3 1 4 (21%)

Hypoacusis 2 1 1 2 1 7 (37%)

Tinnitus 2 1 3 (16%)

Vertigo 3 2 1 1 7 (37%)

Facial paresis 6 1 1 3 11 (58%)

Miscellaneous Mental disorders (1) Seizure (1) Depression (1) Cutaneous fistula (1) Intracranial hypertension (1) Bacterial meningitis (1) Depression (1) Intracranial hypertension (1) Bacterial meningitis (1) Otorrhea (1) Bacterial meningitis (1) 8 (42%)

patients. A spontaneous nystagmus was associated with Rombergs sign in two patients (10%). Preoperative vestibular caloric test results were available in 12 patients (63%), of whom 5 showed clinical signs (Table 3). They revealed vestibular hyporeflexia in four patients (33%) and areflexia in eight patients (66%) on the affected side. Areflexia was associated with a labyrinthine fistula in one patient (5%). Preoperative facial palsy was clinically observed in 12 patients (63%). Facial function assessments are summarized in Figure 3. The facial nerve had been interrupted during previous surgery in one patient. Facial electromyography or a blink reflex test was performed in nine patients (47%). Among those, three (16%) had a delayed R1 response on the blink reflex associated with a normal facial function (grade I). Mild degenerative EMG signs were present in two patients with grade I (11%) and 2 with grade II facial function (11%). Severe degenerative EMG signs were detected in a grade II (5%) and a grade IV (5%) facial paresis. Other preoperative neurologic symptoms included trigeminal deficit in one patient (5%) and intracranial hypertension syndrome in two patients (10%), associated with abducens nerve palsy in one patient (5%). Cholesteatoma location and extension on preoperative imaging The apical location was most frequently observed (8 patients, 42%). In this group, cochlear fistula was present in four patients (50%), and extension to the cerebel-

lopontine angle in one patient (5%). The greatest cholesteatoma diameter on CT axial views in this group was 35 10.9 mm. Infralabyrinthine lesions were observed in three patients (16%). This location was associated with apical extension in two patients and with labyrinthine involvement in two patients. The mean greatest diameter on axial views in this group was 38 3.5 mm. Supralabyrinthine lesions were observed in three patients (16%). Preoperative imaging showed lysis of the tegmen tympani associated with cholesteatoma extension toward the middle fossa in all patients and a superior semicircular canal fistula in two patients. The mean greatest diameter on axial views in this group was 25 5.0 mm. The cholesteatoma diameter was smaller in supralabyrinthine than in apical or translabyrinthine lesions (p < 0.05). Four cases (21%) of translabyrinthine extension with large labyrinthine destruction were observed. The mean greatest diameter on axial views was 36 4.8 mm. One case of retrolabyrinthine location without destruction of the otic capsule was observed (5%). The lesion measured 25 mm in its greatest diameter. The cholesteatoma extension followed retrolabyrinthine cells directly to the posterior fossa dura in the presinusal region. No intradural extension was observed. Surgical approaches The surgical approaches used in this series were middle cranial fossa (MCF) with or without mastoidec-

TABLE 3. Preoperative audiovestibular function in different cholesteatoma extension types

Preoperative hearing loss* (n Extension type Apical Infralabyrinthine Supralabyrinthine Translabyrinthine Retrolabyrinthine Total Mild ( 25 to <45 dB) 2 2 (11%) Moderate ( 45 to <65 dB) 2 1 3 (17%) 18) Profound ( 85 dB) 3 2 4 1 10 (55%) Preoperative vestibular tests (n Areflexia 2 1 2 3 1 8 (66%) 12) Hyporeflexia 3 1 4 (33%)

Severe ( 65 to <85 dB) 2 1 3 (17%)

*Mean hearing loss assessed by tonal audiometric tests and calculated on 500-, 1,000-, and 2,000-Hz air conduction thresholds. Vestibular function assessed by caloric tests during electronystagmography or videonystagmography. The American Journal of Otology, Vol. 21, No. 6, 2000

EXTENSIVE INTRATEMPORAL CHOLESTEATOMA

777

FIG. 2. Intratemporal cholesteatoma extension on preoperative imaging in four representative cases. A: Supralabyrinthine extension with cochlear destruction on frontal computed tomography (CT) view. B: Apical extension with carotid canal involvement on axial CT view. C: Infralabyrinthine extension with cochlear destruction on frontal CT view. D: Translabyrinthine extension with cerebellopontine angle involvement on axial magnetic resonance imaging in T1 sequence

tomy, extended middle cranial fossa (EMCF) (4), retrolabyrinthine (5), subtotal petrosectomy with otic capsule preservation, and transotic with otic capsule resection according to Fisch and Mattox (4) (Table 4). Endoscopic assessment of the surgical resection was performed in cases of bilobulated lesions and intracranial extension. Factors influencing surgical strategy Cholesteatoma location and extension The transotic approach was used in apical cholesteatomas in two lesions involving the entire otic capsule and encasing the intratemporal carotid canal. These lesions extended to the cerebellopontine angle in one patient and the internal auditory canal in the second. The EMCF or the MCF approach was used in five patients with apical lesions and was combined with a radical mastoidectomy

in two patients. The EMCF or MCF approach was used for lesions extending upward through the supralabyrinthine cell tract and the geniculate ganglion region with limited or no involvement of the cerebellopontine angle. Destruction of the cochlea by the lesion was encountered in two patients of this group. Similarly, three cases of supralabyrinthine lesions with neither apical nor internal extensions were operated on by the MF and EMF approaches. Infralabyrinthine cholesteatomas were approached by a transotic route in two cases of anterior and internal extension through the infralabyrinthine cell tract associated with cochlear destruction. In the third case, an anterior extension of cholesteatoma limited to the posterior aspect of the vertical portion of the carotid canal was approached by a subtotal petrosectomy route. All translabyrinthine lesions were approached by a
The American Journal of Otology, Vol. 21, No. 6, 2000

778

A. B. GRAYELI ET AL.

FIG. 3. Preoperative audiometric results; data were available in 18 patients. Mean air conduction hearing thresholds on 500, 1,000, and 2,000 Hz for each subgroup (profound hearing loss, 85 dB; severe hearing loss, 65 to <85 dB; moderate hearing loss, 45 to <65 dB, mild hearing loss, 25 to <45 dB; normal, <25 dB) are represented on abscissa. Open bars = total number of patients. Hatched bars = number in each subgroup requiring surgical labyrinthine sacrifice.

transotic route because of the labyrinthine destruction and the large cholesteatoma diameters. Preoperative auditory function The otic capsule was destroyed during surgery in 16 patients (84%), principally in those with preoperative severe or profound hearing loss (Fig. 3). Moreover, five patients in this group (31%) had ipsilateral vestibular areflexia. The labyrinth was preserved in two patients (11%) with supralabyrinthine cholesteatoma 20- and 30dB preoperative hearing loss and in one patient (5%) with an infralabyrinthine tumor and preoperative mixed severe hearing loss. Involvement of the cranial nerves by cholesteatoma The facial nerve was encased in the lesion in four apical (21%), one supralabyrinthine (5%), and one infralabyrinthine (5%) extensions. The nerve was involved in its tympanic segment in five apical and supralabyrinthine extensions (26%) and in its mastoid segment in the infralabyrinthine lesion. The surgical approach did not necessitate facial nerve rerouting, and the nerve was decompressed without interruption in all these cases. The greater superficial petrosal nerve was encased in an apical lesion and was consequently interrupted (5%). The trigeminal nerve was involved in one case of apical cho-

lesteatoma (5%) but could be preserved during the surgical resection by an EMCF approach. No caudal cranial nerve involvement was observed in our series. Postoperative results The otic capsule was surgically preserved in three patients (16%). Two of these patients had a 20- to 30-dB conductive preoperative hearing loss associated with a supralabyrinthine cholesteatoma. Both underwent surgical resection by the MCF approach. Postoperatively, hearing thresholds remained stable in one patient, and a mean deterioration of 40 dB with mixed hearing loss was noted in the second. The third patient had an 80-dB mixed hearing loss associated with an infralabyrinthine lesion preoperatively. A subtotal petrosectomy was performed. Hearing thresholds remained stable postoperatively. The otic capsule was surgically removed in 10 patients (53%) with profound hearing loss with large labyrinthine destruction on preoperative CT, in 2 patients (11%) with large cochlear fistula with mild and severe hearing loss, and in 3 patients (16%) with extensive lesions associated with moderate and mild hearing loss without evidence of radiologic labyrinthine destruction. Preoperative and postoperative facial functions are summarized in Figure 4. In seven patients, preoperative normal facial function remained unchanged after sur-

TABLE 4. Surgical approaches used in different cholesteatoma extension types

Cholesteatoma location Approach Middle fossa Extended middle fossa Middle fossa + mastoid Retrolabyrinthine Subtotal petrosectomy Transotic Total Apex 3 2 3 8 (42%) Infralabyrinthine 1 2 3 (16%) Supralabyrinthine 2 1 3 (16%) Translabyrinthine 4 4 (21%) Retrolabyrinthine 1 1 (5%) Total (%) 2 (11) 4 (21) 2 (11) 1 (5) 1 (5) 9 (47) 19

The American Journal of Otology, Vol. 21, No. 6, 2000

EXTENSIVE INTRATEMPORAL CHOLESTEATOMA

779

FIG. 4. Facial function grading at preoperative examination (open bars) and at last follow-up assessment (hatched bars) according to House and Brackmann (6).

gery. In 12 patients with preoperative facial palsy, improvement after nerve decompression was observed in 5 patients (42%) (4 grade II to grade I, and 1 grade IV to III). A deterioration of facial function from grade II to III was noted (5%). In patients with severe or total facial paralysis (grade V and VI), the last follow-up assessment remained unchanged. Other postoperative deficits were one case of ipsilateral ninth cranial nerve paresis, which regressed at the last follow-up examination, and one case of preoperative cerebellar syndrome, which persisted at the last examination. Postoperative complications were limited to two cases of cerebrospinal fluid (CSF) leak through the incision, which was cured under conservative treatment. Surgical resection was macroscopically complete in all cases. No residual mass was detected on postoperative imaging. Two cases of recurrence (11%) were observed on systematic follow-up radiologic examination 2 years after the surgery. These concerned two apical lesions, accompanied by severe hearing loss, excised by the MCF and EMCF routes and associated with mastoidectomy. These patients underwent surgical resection of their recurrent lesions. Information concerning postoperative professional activity was obtained in 15 patients. Among those patients, 14 (93%) had resumed their previous activities at the last follow-up examination. The decisional elements determining the type of surgical approach are summarized in Figure 5. DISCUSSION Cholesteatomas extending beyond the middle ear limits are not frequent: 15% of middle ear cholesteatomas involve the inner ear structures and are referred to as extensive in large series (7). Nevertheless, these lesions raise individual diagnostic and therapeutic challenges in relation to different types of extension. The presenting signs of an extensive intratemporal

cholesteatoma may be misleading. Intact tympanic membrane was observed in 10% of our patients and in a variable proportion in other series, ranging from 0% (8) to 12% (9). Although some authors classify all these cases as primary lesions (8), others have a more restrictive definition of a primary cholesteatoma and add the absence of otitis media, paracentesis, and temporal bone trauma in the history as supplementary conditions (10). Intense headaches were experienced by four patients in our series, associated with an apical lesion in three patients (75%). This symptom has been reported as an early sign of apical extension in other series (11,12). The headache may be explained by the meningeal traction and irritation in contact with the lesion (12). Bacterial meningitis was the presenting symptom in three of our patients (16%), all associated with superior extension of the lesion and disruption of the tegmen tympani. By comparison, the intracranial complication rate was evaluated as 1.2% in a series of 1,907 cholesteatomas (13). In the latter series, bacterial meningitis was the most frequent intracranial complication, and it was related to a tegmen tympani disruption in 30% of patients (13). Sensorineural hearing loss, vertigo, and facial paresis are the clinical signs indicating an extensive lesion in the presence of a cholesteatoma. Their frequency in different series varies from 42% (14) to 100% (15) for sensorineural hearing loss, from 7% (9) to 61% (16) for vestibular signs, and from 26% (9) to 83% (17) for facial dysfunction. As observed in our series and reported by other authors, facial paresis is frequently associated with apical extension of the lesion (12,17). This symptom is often related to the involvement of the geniculate ganglion and the tympanic segment of the nerve following the extension of the cholesteatoma to the supratubal recess from the attic compartment (18). The decision process in the treatment of intratemporal cholesteatomas generally favors complete resection by a wide exposure and preservation of facial function by avoidance of nerve rerouting when possible (8,19,20). Because of the variety of cholesteatoma locations, extension types, and involved structures, different approaches may be considered in each case. In our series, the selection of the approach type was mainly based on the radiologic examination and the audiovestibular functional test results (Fig. 5). The preoperative radiologic examination in extensive cholesteatomas is based on high-definition CT (20,21). Magnetic resonance imaging provides important information in case of intracranial extension and on the probable histologic type of the lesion (20,22). In addition to confirmation of the diagnosis, the radiologic assessment of cholesteatomas distinguishes several routes of extension (4,14,15). These routes mainly follow petrous bone cell tracts, which offer low resistance to the cholesteatoma outgrowth. The relative frequency of different extension types varies in different series. Sanna et al. (14) reported a predominance of the perilabyrinthine extension type (43%) in their population. Charachon et al. (9) observed that the supralabyrinthine extension was the most frequently encounThe American Journal of Otology, Vol. 21, No. 6, 2000

780

A. B. GRAYELI ET AL.

FIG. 5. Summary of surgical strategies in this series. +, preserved auditory function; , destroyed auditory function; EMCF, extended middle fossa approach; STP, subtotal petrosectomy; MF, middle cranial fossa approach; TO, transotic approach; RL, retrolabyrinthine approach.

tered (47%) and that in this group, 80% of tumors involved the apex. Similarly, in our series, the apical extension was the most frequently encountered. Reported differences in extension type frequency may be due in part to differences in recognizing extensions, because many tumors have complex extension routes and may be classified in two different categories (9,15). In the apical extension, the lesion follows the supralabyrinthine cell tract from the attic and involves the greater petrosal nerve canal, or it invades the inferior perilabyrinthine cell tract from the mastoid cavity and progresses toward the apex along the carotid canal (18,23). The latter type of extension is the basis of another category, defined as infralabyrinthine-apical by Fisch and Mattox (4). The cochlear basilar turn may be involved in this type of cholesteatoma progression (18), as was observed in three of our eight apical lesions. The supralabyrinthine type of extension eroding the tegmen tympani concerned the smallest lesions in our series. The MCF approach gave the best access to the lesion and permitted the preservation of the auditory function in the absence of labyrinthine fistula in two of three patients. The translabyrinthine type of extension concerned patients with large tumors and ipsilateral proThe American Journal of Otology, Vol. 21, No. 6, 2000

found hearing loss in our series. The translabyrinthine lesion was approached through a transotic route in all patients. This approach carried a low risk for the facial nerve. Moreover, it permitted surgical resection through a large access route and reduced the risk of recurrence in these patients. Independently from the approach width, temporal bone endoscopy offers enhanced control during surgical resection, may avoid rerouting of the facial nerve, and diminishes the risk of CSF leak (24,25). In addition to radiologic examination, audiometric tests helped determine the selection of the approach, because the labyrinthine sacrifice concerned patients with profound hearing loss in 82% of patients (Fig. 3). Vestibular function deficit represented an additional indication of inner ear involvement in patients with hearing loss, but it was not a determinant in the selection of surgical approach. Similarly, other cranial nerve deficits did not influence the type of approach in our series. The frequency of postoperative recurrence is variable and cannot be directly compared between the different reports because of the heterogeneity of cholesteatoma sizes, the types of extension, and the different approaches used (8,9,12). It varies from 0% (12) to 14% (9). In our series, recurrences concerned two apical le-

EXTENSIVE INTRATEMPORAL CHOLESTEATOMA sions (11%) approached by the EMCF and MCF with mastoidectomy routes. The poor auditory prognosis in our series was caused by frequent cochlear destruction by a large fistula. Although limited vestibular fistulas can be repaired with satisfactory auditory results (27,28), involvement of the cochlea by the lesion has a poor prognosis and occurred in the majority of our patients. Our postoperative complications were limited to two postoperative regressive CSF leaks. Complications reported in the literature mainly include CSF leak (7%) (9)and bacterial meningitis (8%) (16). Mortality rates ranging from 0.2% (3) to 5% (9) are reported in large series, underlining the importance of the preoperative examination and the surgical decision to the final outcome in patients with extensive intratemporal cholesteatomas. In conclusion, the choice of surgical approach was principally based on the cholesteatoma location, as assessed by CT scan and MRI, and the auditory function. The strategy favored complete macroscopic resection and preservation of the facial nerve function. The poor auditory prognosis was related to involvement of the labyrinthine structures by the cholesteatoma in the majority of our patients. REFERENCES
1. House WF, Hitselberger WE. The transcochlear approach to the skull base. Arch Otolaryngol Head Neck Surg 1976;102:33442. 2. Fisch U, Fagan P, Valvanis A. The infratemporal approach for the skull base. Otolaryngol Clin North Am 1984;17:51352. 3. Wigand ME. The enlarged middle fossa approach to the cerebellopontine angle: technique and indications. Rev Laryngol Otol Rhinol 1998;119:15962. 4. Fisch U, Mattox D. Microsurgery of the skull base. New York: Theime, 1988:12740. 5. Darrouzet V, Guerin J, Aouad N, et al. The widened retrolabyrinthine approach: a new concept in acoustic neuroma surgery. J Neurosurg 1997;86:81221. 6. House JW, Brackmann DE. Facial nerve grading system. Otolaryngol Head Neck Surg 1985;3:18493. 7. Sheehy JL, Brackmann DE. Cholesteatoma surgery: management of the labyrinthine fistula, a report of 97 cases. Laryngoscope 1979;89:7887. 8. Atlas MD, Moffat DA, Hardy DG. Petrous apex cholesteatoma: diagnostic and treatment dilemmas. Laryngoscope 1992;102: 13638.

781

9. Charachon R, Martin C, Gratacap B, et al. Temporal bone cholesteatoma. In: Tos M, Thomsen J, Peitersen E, eds. Proceedings of the Third International Conference on Cholesteatoma and Mastoid Surgery. Amsterdam: Kugler & Ghedini, 1989:10818. 10. Zappia JJ, Wiet RJ. Congenital cholesteatoma. Arch Otolaryngol Head Neck Surg 1995;121:1922. 11. Flood LM, Kemink JL. Surgery in lesions of the petrous apex. Otolaryngol Clin North Am 1984;17:56575. 12. Glasscock ME, Woods CI, Poe DS, et al. Petrous apex cholesteatoma. Otolaryngol Clin North Am 1989;22:9811003. 13. Fisch U: Intracranial complications of cholesteatoma. In: Sade J, ed. Proceedings of the Second International Conference on Cholesteatoma and Mastoid Surgery. Amsterdam: Kugler, 1982:369 79. 14. Sanna M, Zini C, Mazzoni A, et al. Cholesteatoma of the temporal bone. In: Tos M, Thomsen J, Peitersen E, eds. Proceedings of the Third International Conference on Cholesteatoma and Mastoid Surgery. Amsterdam: Kugler & Ghedini, 1989:107780. 15. Martin C, Prades JM, Bertholon P, et al. Intrapetrous cholesteatoma. Rev Laryngol Otol Rhinol 1996;117:15763. 16. Burggraaff B, Luxford WM, Doyle KJ. Neurotologic treatment of acquired cholesteatoma. Am J Otol 1995;16:4805. 17. Yanagihara N, Matsumoto Y. Cholesteatoma in the petrous apex. Laryngoscope 1981;91:2728. 18. Bartels LJ. Facial nerve and medially invasive petrous bone cholesteatomas. Ann Otol Rhinol Laryngol 1991;100:30816. 19. Parisier SC. Management of cholesteatoma. Otolaryngol Clin North Am 1989;22:92740. 20. Sterkers JM, Sterkers O. Deep-seated destruction of the inner ear by cholesteatoma. In: Sade J, ed. Proceedings of the Second In ternational Conference on Cholesteatoma and Mastoid Surgery. Amsterdam: Kugler, 1982:3638. 21. Phelps PD, Wright A. Imaging cholesteatoma. Clin Radiol 1990; 41:15662. 22. Mafee MF. MRI and CT in the evaluation of acquired and congenital cholesteatomas of the temporal bone. J Otolaryngol 1993; 22:23948. 23. Robert Y, Dubreuile F, Carcasset, S et al. Petrous bone extension of middle-ear acquired cholesteatoma. Acta Radiol 1996;37:166 70. 24. Rosenberg SI. Endoscopic otologic surgery. Otolaryngol Clin North Am 1996;29:291300. 25. Axon PR, Fergie N, Saeed SR, et al. Petrosal cholesteatoma: management considerations for minimizing morbidity. Am J Otol 1999; 20:50510. 26. Proctor B. Surgical anatomy of the ear and the temporal bone. New York: Thieme, 1989:858. 27. Kobayashi T, Sakurai T, Okitsu T, et al. Labyrinthine fistulae caused by cholesteatoma: improved bone conduction by treatment. Am J Otol 1989;10:510. 28. Ars B, Claes J, Casselman J, Ars-Piet N. Preservation of cochlear function after extensive labyrinthine destruction. Am J Otol 1996; 17:405.

The American Journal of Otology, Vol. 21, No. 6, 2000