Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Otology & Neurotology 30:1152Y1159 Ó 2009, Otology & Neurotology, Inc. Reliability of High-Resolution CT Scan in Diagnosis of Otosclerosis *Sebastien Lagleyre, *Tommaso Sorrentino, *Marie-Noelle Calmels, *Young-Je Shin, †Bernard Escudé, *Olivier Deguine, and *Bernard Fraysse *Department of Otology-Neurotology and Skull Base Surgery, Purpan University Hospital; and ÞDepartment of Neuroradiology, Clinique Pasteur, Toulouse, France Objective: To assess the reliability of high-resolution computed tomographic scan (HRCT scan) for the diagnosis of otosclerosis and to determine its usefulness to predict hearing deterioration and surgical difficulties. Study Design: Prospective study. Setting: Tertiary reference center. Patients: Two hundred nine ears (200 patients) presenting progressive conductive hearing loss with normal tympanic membrane, abnormal stapedial reflex, and scheduled for stapes surgery. The mean age was 47.3 years. Intervention: All patients underwent HRCT scan before surgery (slice thickness of 0.6 to 1 mm). Stapedotomy was performed in 99% of cases. Main Outcome Measures: High-resolution computed tomographic scan results were categorized as positive, doubtful, or negative. We classified a CT scan as positive for otosclerosis when a hypodense focus was seen around the otic capsule. Preoperative and postoperative air- and bone-conduction thresholds were collected. Results: Of 209 HRCT scans, 84.2% were classified positive, 8.6% doubtful, and 7.2% negative. In all patients with positive CT scan, otosclerosis was confirmed in surgery. Among 15 negative cases, we found 4 minor malformations and 1 fracture of the stapes. Footplate incidents (mobilized, floating, or fractured footplate; 5.3%) occurred significantly more frequently when an HRCT scan was negative or doubtful ( p = 0.05). Mean preoperative air-bone gap was 27.7 dB (standard deviation, 10). Mean postoperative air-bone gap was within 10 dB in 65% and within 20 dB in 92% of cases. Greater than 10 dB deterioration of bone-conduction thresholds occurred in 2% of cases. Mean preoperative and postoperative bone-conduction thresholds were significantly lower in cases of round window obliteration, pericochlear, or internal auditory canal hypodensities ( p G 0.005 and p G 0.0001, respectively). Conclusion: In our series, the sensitivity of HRCT scan to otosclerosis was 95.1%. Hypodense otosclerotic foci were mostly localized at the anterior part of footplate. Negative or doubtful cases were associated with the highest incidence of stapes footplate complications. Foci involving otic capsule, internal auditory canal, or round window led to a significantly higher risk of sensorineural hearing loss. Key Words: CT scanVFloating stapes footplateVMobile footplateV OtosclerosisVStapedotomy. Otol Neurotol 30:1152Y1159, 2009. Stapes surgery remains the established treatment for otosclerosis. Stapedotomy is usually a safe procedure, performed under local anesthesia with minimal complications. Developments in otologic surgery are aimed to minimize the risk of failure. Since its introduction into otology, the computed tomographic (CT) scan has been shown to be an useful instrument for diagnosis of otosclerosis (1,2). Many studies found a reliability of more than 90% in the diagnosis of otosclerosis (1,3). The reliability was only limited in cases of isolated otosclerotic foci (4Y6) or by superficial foci too small to be visualized on the CT scan (1,3). The main goal of this study was to specify the reliability of high-resolution CT scans (HRCT scans) in the diagnosis of otosclerosis based on comparing radiologic and surgical findings. Results are compared with those of previous studies (3,7) conducted in the same department with lower-resolution CT scans. In addition, we discuss the role of preoperative CT scans in preventing intraoperative incidents and their importance in localizing hypodensities for hearing prognosis. Address correspondence and reprint requests to Sebastien Lagleyre, M.D., Service d’ORL et Otoneurologie, CHU Purpan, Place du Docteur Baylac, 31059 Toulouse Cedex, France; E-mail: [email protected] Additional Information: This work was presented in part at the 2nd International Symposium of the Politzer Society on Otolosclerosis and Stapes Surgery, May 8Y10, 2008, Biarritz, France. MATERIALS AND METHODS The only way to be certain of the diagnosis of otosclerosis is to find specific features of otosclerosis during surgery, as 1152 Copyright @ 2009 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited. RELIABILITY OF HRCT SCAN IN OTOSCLEROSIS FIG. 1. Computed tomography, axial cut, showing fenestral otosclerosis characterized by a hypodensity immediately adjacent to the anterior margin of the oval window, including the anterior branch of the stapes. This CT scan was considered positive. defined as the presence of macroscopic otosclerotic foci on the oval window with stapedial fixation. Therefore, we conducted an open prospective study, including patients with Btypical[ and Bnontypical[ clinical forms of otosclerosis who were scheduled for surgery. A Btypical[ clinical form of otosclerosis was defined by the presence of a bilateral conductive hearing loss, a normal tympanic membrane, and the absence of stapedial reflex and a family history of otosclerosis. We considered patients had bilateral conductive hearing loss when air-bone gap (ABG) was more than 10 dB in both ears and evidence of family history when a previous surgery for otosclerosis had occurred in the family. Patients with Bnontypical[ clinical form of otosclerosis presented a conductive hearing loss with normal FIG. 2. Computed tomography, axial cut, showing absence of hypodensity or thickening of the stapes footplate. This CT scan was considered negative. 1153 tympanic membrane and abnormal (biphasic or absent) stapedial reflex in at least 1 ear. Between September 2004 and September 2006, we collected data on 209 ears (200 patients), including demographic, radiologic, audiologic, and surgical data. Before surgery, all patients underwent HRCT scan with helical acquisition in axial and coronal planes. High-resolution CT scans were performed in 25 different radiology departments. Reconstructed slices were of thickness 0.6 mm from 19 departments to 1 mm from 6 departments. One hundred twenty-one examinations (57.9 %) were obtained with a slice thickness of 0.6 mm. For each case, the result of the CT scan was confirmed by an experienced neuroradiologist. Computed tomographic scan results with respect to the presence of otosclerosis were classified as positive, negative, or doubtful: positive when hypodense foci were seen around the otic capsule (Fig. 1); negative in the absence of any hypodensity or other temporal bone finding, including tympanosclerosis, ossicular malformation or fixation, modiulus anomalies, and superior semicircular canal dehiscence (Fig. 2); and doubtful in cases of minimal hypodense foci such as an isolated thickening of the anterior part of the footplate or a triangular widening of the anterior branch of the stapes (Fig. 3). Different sites of otosclerosis were classified according to Veillon et al. (2): anterior fenestral hypodensity (AFH; Ib, II), isolated thickening of the footplate hypodensity (Ia), hypodensity extended to the endosteum (III) or to the vestibule (IVb), pericochlear hypodensity (PH; IVa), internal auditory canal hypodensity (IACH; IVa), and ossification of the oval window or of the round window. The decision to perform a surgical middle ear exploration was based upon the history of the pathology, clinical reports, and the needs of the patient in terms of hearing. During surgery, performed under local anesthesia by a classical tympanomeatal approach, the mobility of the ossicular chain was tested, and the presence of an otosclerosis focus was determined by observation. Positive surgical diagnosis of otosclerosis was confirmed FIG. 3. Computed tomography, axial cut, showing an isolated thickening of the anterior part of the footplate or a triangular widening of the anterior branch of the stapes. This CT scan was considered doubtful. Otology & Neurotology, Vol. 30, No. 8, 2009 Copyright @ 2009 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited. 1154 S. LAGLEYRE ET AL. TABLE 1. Details of locations of otosclerotic hypodensities across all CT scans (several locations could be encountered in the same ear) Location of otosclerotic hypodensities (classification of Veillon) n (% of all CT scans) Isolated or associated AFH (Ib, II) Isolated AFH Isolated thickening of the footplate hypodensity (Ia) Hypodensity extended to the endosteum (III) Isolated or associated PH (IVa) Isolated PH IACH (IVa) Ossification of the round window (IVa) Stage A (normal window) Stage B (edge hypodensity with normal round window membrane) Stage C (round window obliteration) Stage D (basal turn ossification of the cochlea) Hypodensity extended to the vestibule (IVb) No hypodensities 185 (88.5) 143 (68.4) 4 (1.9) 5 (2.4) 20 (9.6) 2 (1) 17 (8.1) 6 (2.9) V 4 (1.9) 2 (1) 0 (0) 0 15 (7.2) AFH indicates anterior fenestral hypodensity; PH, pericochlear hypodensity; IACH, internal auditory canal hypodensity. by the presence of macroscopic otosclerotic foci on the oval window with stapedial fixation. In the case of an absence of these signs, other diagnoses were considered. In positively diagnosed cases of otosclerosis, stapedotomy was performed using an argon laser, followed by a skeeter microdrill. Anatomic difficulties or complications encountered during surgery were noted. Pure-tone air-conduction (AC) and bone-conduction (BC) thresholds were measured in all patients the day before surgery and 2 months later in the same double-walled soundproof room. Hearing tests were realized according to the Committee on Hearing and Equilibrium (8). The preoperative and postoperative ABG and ABG closure (ABGC) were calculated from airand bone-conduction thresholds. The sensorineural hearing level (SNHL) was estimated by averaging the bone-conduction levels for frequencies 1, 2, and 4 kHz. The preoperative to postoperative change in SNHL was the measure of surgically induced sensorineural hearing loss. Positive values reflected improvements in bone-conduction levels, referred to as overclosure, whereas negative values indicated surgically induced high-frequency sensorineural hearing loss. The change in SNHL was stratified as worse, 910 dB; unchanged, T10 dB; and improved, 910 dB. Statistical analyses were performed using the W2 test and Fischer’s exact test. The significance threshold was 0.05. surgery. Therefore, the Btypical[ form of otosclerosis was rare in our sample. Radiologic Results: Operated and Opposite Ears Operated Ear Of 209 CT scans, 176 were positive (84.2%), 18 were doubtful (8.6%), and 15 were negative (7.2%). Across all CT scans (Table 1), otosclerotic hypodensities were classified as anterior fenestral hypodensities in 88.5%, pericochlear in 9.6%, and internal auditory canal in 8.1% of the cases (Fig. 4). Hypodensities involved the round window in 6 ears (2.9%), and for 2 of these, it was obliterated. For positive CT scans, anterior fenestral (isolated or associated) was the most frequent location (96.6% of cases). Other positions of hypodensities for positive CT scans were pericochlear in 12.5%, internal auditory canal in 9.6%, and involving the round window in 3.4%. Among doubtful cases, a triangular widening of the anterior branch of the stapes was found in 83.3% of ears. Other associated diagnoses were suspected in 3 patients (2 cases of ossicular chain malformation and 1 case of malleus fixation). Computed tomographic scans results were positive bilaterally for 135 cases (64.6%). Of the 200 patients included, only 9 patients (4.5%) were operated bilaterally during the period of study (mean preoperative ABG: first side, 31.4 dB [SD, 13.9]; second side, 26.5 dB [SD, 7.3]; mean delay, 12 mo). Here, of the 126 unoperated opposite ears, ABG was less than 10 dB in 50% of cases. Opposite Ear Concerning the opposite ear, we defined 3 groups from pure-tone audiogram: normal hearing, pure sensorineural hearing loss, and conductive hearing loss. RESULTS For the 209 ears (200 patients), the mean age was 47.3 (range, 17Y77) years, with a male-female ratio of 2.21. Mean preoperative ABG was 27.7 dB (standard deviation [SD], 10). The stapedial reflex was absent in 47.5% of cases, and we found a biphasic reflex in 51% of cases. Of the 200 patients, 120 (60%) presented a unilateral conductive hearing loss. Among the 80 patients with bilateral hearing loss, only 15 (7.5%) presented the Btypical[ clinical form of otosclerosis. The remaining 65 patients (32.5%) with bilateral conductive hearing loss did not exhibit all the typical features of otosclerosis such as absent stapedial reflex and/or family history of FIG. 4. Computed tomography, axial cut, showing an internal auditory canal hypodensity on the internal auditory meatus associated with a large anterior fenestral hypodensity. Otology & Neurotology, Vol. 30, No. 8, 2009 Copyright @ 2009 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited. RELIABILITY OF HRCT SCAN IN OTOSCLEROSIS 1155 TABLE 3. Details of intraoperative variations and incidents (several problems could be encountered in the same ear) Intraoperative problems Incidents Stapes footplate incidents: mobilized, floating, and fractured footplates Vertigo Variations Narrow oval window niche, including overhanging facial nerve Facial nerve dehiscence Incus malformation Obliterated footplate Persistent stapedial artery Other FIG. 5. Comparison of surgical and radiological findings. In the opposite ears of 200 patients, we found 120 (60%) with no conductive component, defined as ABG less than or equal to 10 dB (mean AC, 23.6 dB [SD, 16.8]; mean BC, 16 dB [SD, 12.1]). Among these 120 ears, 69 ears presented normal hearing (34.5%), defined as ABG less than or equal to 10dB and mean AC less than 20 dB (mean AC, 18.7 dB [SD, 18.6]; mean BC, 8.4 dB [SD, 4.1]). The remaining 51 ears presented a pure sensorineural hearing loss (25.5%; mean AC, 30.2 dB [SD, 11.2]; mean BC, 26.3 dB [SD, 11.8]). In the 69 opposite ears with normal hearing, we found 32 positive (46.5%), 27 negative (39%), and 10 doubtful CT scans (14.5%). The position of the hypodensities in these normal-hearing opposite ears on positive CT scans was anterior fenestral in 87%, pericochlear in 12%, around the internal auditory canal in 15%, and involving the round window in 6%. In the 51 opposite ears with pure sensorineural hearing loss, we found 35 positive (68%), 11 negative (22%), and TABLE 2. n % of surgeries 11 5.3 1 0.5 7 3.3 6 3 2 1 6 2.9 1.4 0.9 0.5 3.4 5 doubtful CT scans (10%). The position of the hypodensities in these opposite ears on positive CT scans was anterior fenestral in 94%, pericochlear in 20%, around the internal auditory canal in 20%, and involving the round window in 6%. Finally, in the 80 opposite ears with conductive hearing loss (mean AC, 34.4 dB [SD, 16.9]; mean BC, 21.8 dB [SD, 16]), CT scan was positive in 81%, negative in 11%, and doubtful in 8%. The position of the hypodensities in these opposite ears on positive CT scans was anterior fenestral in 100%, pericochlear in 17%, around the internal auditory canal in 15%, and involving the round window in 3%. In conclusion, for unoperated normal-hearing opposite ears, the CT scan indicated the presence of otosclerotic hypodensities in 46% of cases, with a greater proportion of hypodensities with endosteal extension than in operated ears. Opposite ears with pure sensorineural hearing loss had the greatest proportion of hypodensities with endosteal extension compared with normal-hearing opposite ears, bilateral conductive hearing loss ears, and operated ears. Comparison of Surgical and Radiologic Findings The results of comparing surgical and radiologic findings are presented in Figure 5. All ears with positive CT scan results had a positive diagnosis of otosclerosis during surgery. Among 18 ears with doubtful hypodensity, during surgery, we found 17 with otosclerotic foci (94.4%) and in 1 a minor malformation of the anterior branch of the stapes. Comparison of surgical and radiological findings Positive and doubtful scans Negative scans Total TABLE 4. Positive surgical otosclerosis Other diagnosis Total 193 10 203 1 5 6 194 15 209 Ten surgical otosclerotic foci were missed on the CT scan (i.e., sensitivity of 95.1%), and 1 minor malformation was confused with a radiologic hypodensity (i.e., specificity of 99.5%). Stapes footplate complications Positive scan (176) Doubtful scan (18) Negative scan (15) Total (209) Stapes footplate complications and radiologic diagnosis Mobilized and floating footplate 4 3 0 7 Fracture Incidence, % 2 0 2 4 Z 3.4 16.7^ p = 0.05 13.3 5.3 Otology & Neurotology, Vol. 30, No. 8, 2009 Copyright @ 2009 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited. S. LAGLEYRE ET AL. 21.2 34 31.3 27.2 20.5 23 (10.9) (21.2) (19.4) (12.6) (12.3) (13.5) +2.8 +1.7 +0.1 +1.4 +3 +2.6 dBHL dBHL dBHL dBHL dBHL dBHL (9) (4.9) (9.2) (5.2) (4.8) (8.6) 27.5 (10) 32 (10.7) 32 (10.8) 31 (8.8) 23.2 (6.5) 27.7 (10) 9.6 9.2 9.3 12.7 12 10 (7.1) (6) (8) (16.1) (6.5) (7.4) 17.9 22.8 22.7 18.3 11.3 17.4 (10.9) (10.2) (10.1) (12.4) (7.4) (10.6) Among 15 negative CT scans, we found 4 minor malformations of the ossicular chain and 1 fracture of the stapes. The remaining 10 cases (66.6%) were positively diagnosed as otosclerosis during surgery and were defined as Binfraradiologic[ forms (i.e., a patient with negative CT scan presenting a positive surgical otosclerosis). In summary, the preoperative sensitivity of HRCT scan in the diagnosis of otosclerosis was 95.1%, and its specificity was 99.5% (Table 2). During surgery, we found 51 (17.7%) variations or incidents (Table 3). Anatomic variations were narrow oval window niche and overhanging facial nerve (3.3%), facial nerve dehiscence (2.9%), incus malformation (1.4%), obliterated footplate (0.9%), and persistent stapedial artery (0.5%). No associated malleus fixations were found. The main surgical incident concerned the stapes footplate (5.3%). The risk of a mobilized, floating, or fractured footplate was significantly more frequent ( p =0.05) in cases of doubtful (16.7%) or negative (13.3%) CT scans than in positive ones (3.4%; Table 4). It was possible that in doubtful or negative CT scans, the superficial otosclerotic foci made the stapes less immobile, and that this increased the risk of footplate fragility. (15) (24.2) (22.8) (20) (8.1) (16.6) 20.3 25.4 23.6 19.8 14 20.1 dBHL dBHL dBHL dBHL dBHL dBHL (14.1) (11.3) (13.5) (16.2) (7.9) (13.5) 24 35.7 31.4 28.6 22.7 25.5 (11.8) (20.9) (19.1) (11.8) (14) (13.7) Hearing Results and the Position of Hypodensities For the 209 ears, mean postoperative ABG was 10 dB (SD, 7.4). Mean postoperative ABG was within 10 dB in 65% and within 20 dB in 92% of cases. Sensorineural hearing level improved by more than 10 dB in 13.4% of cases and was greater than 10 dB worse in 1.9%. No patient had a dead ear postoperatively. Hearing results before and 2 months after stapedotomy are classified in Table 5 according to hypodensity location. For the 36 ears with endosteal involvement (pericochlear, internal auditory canal, or round window hypodensities), mean preoperative and postoperative BC thresholds were significantly lower than the 143 ears Hearing results between patients with and without endosteal involvement 29 40.6 37.7 36.7 31.2 31.2 TABLE 6. n 143 22 17 6 15 209 Hypodensity location Isolated anterior fenestral Pericochlear Internal auditory canal Round window No location All dBHL indicates decibel hearing level. dBHL 49.4 (14.9) 66 (25.7) 61.5 (25.2) 56.4 (14.4) 44.2 (12.6) 51.3 (17) Mean postoperative BC (SD) Mean preoperative BC (SD) Mean AC improvement (SD) Mean postoperative AC (SD) Mean preoperative AC (SD) TABLE 5. Hearing results versus hypodensity location Mean change SNHL (SD) Mean preoperative ABG (SD) Mean postoperative ABG (SD) Mean ABGC (SD) 1156 Mean preoperative AC (SD) Mean postoperative AC (SD) Mean AC improvement (SD) Mean preoperative BC (SD) Mean postoperative BC (SD) Mean change SNHL (SD) Mean preoperative ABG (SD) Mean postoperative ABG (SD) Mean ABGC (SD) SNHL percentage of ears Improved 910 dB No change (T10 dB) Worse 910 dB Group 1 (n=36) Group 2 (n=143) Statistical significance 61.2 (23.4) 39.2 (23.7) 22.1 (12.9) 31.4 (19) 30.9 (19.5) +0.5 (6.9) 31.7 (10.5) 10.9 (8.8) 20.8 (10.3) 49.4 (14.9) 29 (15) 20.3 (14.1) 24 (11.8) 21.2 (10.9) +2.8 (9) 27.5 (10) 9.6 (7.1) 17.9 (10.9) p G 0.0001 p G 0.005 ns p G 0.005 p G 0.0001 p = 0.1 p G 0.005 NS p G 0.05 5.6 88.8 5.6 14.7 83.9 1.4 Z NS Group 1, patients with endosteal involvement (pericochlear, internal auditory canal, and round window hypodensities); Group 2, patients without endosteal involvement (isolated anterior fenestral hypodensity). dBHL indicates decibel hearing level; ABG, Air-bone gap; ABGC, Air-bone gap closure; SNHL, sensorineural hearing level. Otology & Neurotology, Vol. 30, No. 8, 2009 Copyright @ 2009 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited. RELIABILITY OF HRCT SCAN IN OTOSCLEROSIS without endosteal involvement (isolated AFH; p G 0.005 and p G 0.0001, respectively; Table 6). Comparing the group with endosteal extension (pericochlear, internal auditory canal, and round window hypodensities) and the group without endosteal involvement (isolated anterior fenestral hypodensities), a significant difference was found in the overclosure; +0.5 versus +2.8 dB. The percentages for SNHL worse than 10 dB were 5.6% and 1.4%, respectively. In cases of a completely obliterating focus in the round window (2 ears), we found a worsening of hearing results compared with the group without endosteal involvement, with a mean change SNHL of -4.2 versus +2.8 dB and with a mean ABGC of 7.5 versus 17.9 dB, respectively. Therefore, we think that there is a risk of a worsening of hearing in patients with endosteal involvement of otosclerosis. DISCUSSION Sensitivity of the CT Scan in Diagnosis of Otosclerosis In 1993, Valvassori (1) reported that otoslerotic hypodensities must be 1 mm or more in diameter to be visible 1157 in CT slices. Therefore, infraradiologic forms of otosclerosis were thought to be due to superficial otosclerotic foci, too small to be seen on CT scan. According to Thiers et al. (4), developments in CT imaging will produce images of higher resolution. Better collimation and decreased pixel size will improve the sensitivity of CT scans in the diagnosis of otosclerosis. In previous retrospective studies (3,7), performed between 1996 and 1999 on 437 cases, CT scans had a sensitivity of 91.3%. Slice thickness of the machine was 1 mm. In our present series with better resolution, sensitivity was improved to 95.1%. The sensitivity of HRCT scan in otosclerosis diagnoses followed an improvement in resolution. Future reductions in slice thickness should further reduce the frequency of infraradiologic forms. Naumann et al. (9) reported that the detection rate of hypodense foci increased by 11% (from 74 to 85%) when computer-aided analysis was used instead of simple manual reading. Despite this improvement, 2 of their patients without any radiologic findings showed at surgery otosclerotic foci around the anterior oval window. Infraradiologic forms may be due to inactive lesions of otosclerosis. The same density of these foci to the surrounding bone left them undetected (4,6). This FIG. 6. A, Computed tomography, axial cut, showing a normal round window. This image is classified as Stage A. B, Computed tomography, axial cut, showing an edge hypodensity on the round window niche with normal round window membrane. This image is classified as Stage B. C, Computed tomography, axial cut, showing a round window obliteration. This image is classified as Stage C. D, Computed tomography, axial cut, showing basal turn ossification of the cochlea with pericochlear and round window hypodensities. This image is classified as Stage D. Otology & Neurotology, Vol. 30, No. 8, 2009 Copyright @ 2009 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited. 1158 S. LAGLEYRE ET AL. hypothesis can only be confirmed with a histopathologic sample from these patients. However, inactive isolated forms of otosclerosis were rare in histologic reports. Shucknecht (10) considered it common to find inactive and active regions in otosclerotic foci. For Mafee et al. (11), the CT scan is likely normal in cases of otosclerotic foci less than 2 to 3 mm or in the case of a simple fibrous fixation of the footplate. Finally, infraradiologic forms could be considered like isolated and superficial otosclerotic lesions in the antefenestram fissula region. Schuknecht and Barber (12) localized histologic foci at the fissula antefenestram in 95.9% of cases. Among positive CT scans of our series, AFH was the main localization in 96.6% of cases. Thus, the anterior part of the oval window is the most important site to investigate in radiologic diagnosis of otosclerosis especially in doubtful and normal CT scans. The Use of the CT Scan in Foreseeing the Risk of Stapes Footplate Problems After a positive result from CT scan, it would be useful to foresee possible problems that may be encountered at surgery. Each surgeon should be aware of the risk of stapes footplate problems before surgery and should inform the patient. However, we think that in light of this information at the preoperative stage, one may reconsider surgery as a function of the surgeon’s experience. In our series, we encountered footplate incidents in 5.3% of cases. Ayache et al. (13) observed an incidence of 5.8% of floating footplate, associated with narrow oval window niche in half of cases. Lippy et al. (14) reported an incidence of 1.2%, but the authors analyzed only nonremoving floating footplate. In our series, this situation occurred in only 1 case (0.5%). In our study, incidents involving the stapes footplate were significantly more frequent when the CT scan was doubtful or negative. In a previous study, Shin et al. (3) found the same results with a significant risk ( p = 0.04) of complication of the stapes footplate in cases of infraradiologic forms. Infraradiologic forms correspond to superficial histopathologic otosclerotic foci usually associated with reduced stapes fixation and a higher risk of fragile footplate. Thus, in cases of negative or doubtful HRCT scan otosclerosis, could not be excluded, but the surgeon is alerted to the higher risk of encountering a stapes footplate problem. Impact of the CT Scan on Audiometric Results For Veillon et al. (2), pericochlear hypodensities may correspond to an extension of the anterior foci of otosclerosis through the otic capsule or an independent focus of the fissula antefenestram developed from the cartilagenous residue. Generally, pericochlear hypodensities seem bigger than anterior fenestral lesions and are easily detected on CT scans (15), contrary to internal auditory canal hypodensities. Shin et al. (7) found preoperative and postoperative bone conduction thresholds significantly lower ( p G 0.05) in patients with pericochlear foci on the CT scan. This occurred especially when the endosteum was involved. In our study, we found similar results with mean preoperative and postoperative bone conduction significantly lower ( p G 0.005 and p G 0.0001, respectively) in ears with endosteal involvement (pericochlear, internal auditory canal, or round window hypodensities). Otosclerosis may involve the round window niche. Thiers et al. (4) recommended inspecting this area in primary stapes surgery. The radiologic and anatomic presence of round window foci varied between 3.2 (7) and 36.1% (16). For several authors, it was the second most frequent anatomic site for otosclerosis (12,17). Round window hypodensities were detected in 2.9% of cases in our series. They were associated with pericochlear and internal auditory canal hypodensities in 50% of cases. Round window obliteration is classically considered as a possible cause of surgical failure (18Y22). The incidence of round window obliteration is generally less than 1% (23Y25). According to Shin et al. (7), the risk of sensorineural hearing loss was significantly higher ( p G 0.05) in cases of round window obliteration. However, in their study, this obliteration was associated with a pericochlear focus in 8 of 14 cases. Therefore, the degree of sensorineural hearing loss could not be interpreted as solely due to the effect of round window obstruction. Wiet et al. (26) considered that round window obliteration could induce a hearing loss only if the round window niche is totally obliterated. To clarify otosclerotic involvement of the round window, we suggest adopting a new classification in 4 progressive stages (Fig. 6AYD). Stage A is defined as a normal window, Stage B when hypodensity involves the edge of the niche but the round window membrane is normal, Stage C in case of round window obliteration, and in Stage D, the basal turn of the cochlea is ossified. In our series, we observed 2 cases of obliterated round window niche (Stage C; 1%), both associated with postoperative high-frequency sensorineural hearing loss (mean change, SNHL -4.2 dB). Following the suggestions of several authors (25,27) we did not drill the round window area to avoid the increased risk of sensorineural hearing loss. From these results, we suggest that for round window lesions classified as Stages C and D from preoperative HRCT scans, patients should be informed of the possible risk of postoperative sensorineural hearing loss. CONCLUSION The current study showed that for patients presenting with conductive hearing loss, normal tympanic membrane, and abnormal stapedial reflex, the sensitivity of preoperative HRCT scan in the diagnosis of otosclerosis was 95.1%. We found that the preoperative HRCT scan allowed us to allow the positive diagnosis of otosclerosis in cases of Bnontypical[ clinical form (Btypical[ defined as a bilateral conductive hearing loss associated with normal tympanic membrane, absent stapedial reflex, and a familial Otology & Neurotology, Vol. 30, No. 8, 2009 Copyright @ 2009 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited. RELIABILITY OF HRCT SCAN IN OTOSCLEROSIS history of otosclerosis), foresee anatomic difficulties (i.e, narrow oval window, facial dehiscence) and look for other associated diagnoses (i.e, malleus fixation, ossicular malformation), identify a risk of mobilized stapes footplate in cases of negative or doubtful images, and inform patients of the possible risk of postoperative sensorineural hearing loss in cases of pericochlear, internal auditory canal, or round window involvement. The systematic use of HRCT scan before stapes surgery should be based on the experience of the surgeon and on the socioeconomic context of each country, taking into account the great variability of CT scan costs. Acknowledgments: The authors thank Vanina Bongard, M.D. (Department of Epidemiology, Purpan Medicine Faculty, Paul-Sabatier University, Toulouse, France) for statistical support and Chris James, M.D., for help. 10. 11. 12. 13. 14. 15. 16. 17. REFERENCES 1. Valvassori GE. Imaging of otosclerosis. Otolaryngol Clin North Am 1993;26:359Y71. 2. Veillon F, Stierle JL, Dussaix J, et al. [Otosclerosis imaging: matching clinical and imaging data]. J Radiol 2006;87:1756Y64. 3. Shin YJ, Deguine O, Cognard C, et al. Reliability of CT scan in the diagnosis of conductive hearing loss with normal tympanic membrane. Rev Laryngol Otol Rhinol (Bord) 2001;122:81Y4. 4. Thiers FA, Valvassori GE, Nadol JB Jr. Pathology case of the month: otosclerosis of the cochlear capsule: correlation of computerized tomography and histopathology. Am J Otol 1999;20:93Y5. 5. Bonafe A. Imaging of conductive hearing loss. J Radiol 1999;80: 1772Y9. 6. Guneri EA, Ada E, Ceryan K, et al. High-resolution computed tomographic evaluation of the cochlear capsule in otosclerosis: relationship between densitometry and sensorineural hearing loss. Ann Otol Rhinol Laryngol 1996;105:659Y64. 7. Shin YJ, Fraysse B, Deguine O, et al. Sensorineural hearing loss and otosclerosis: a clinical and radiologic survey of 437 cases. Acta Otolaryngol 2001;121:200Y4. 8. Committee on Hearing and Equilibrium guidelines for the evaluation of results of treatment of conductive hearing loss. American Academy of OtolaryngologyYHead and Neck Surgery Foundation, Inc. Otolaryngol Head Neck Surg 1995;113:186Y7. 9. Naumann IC, Porcellini B, Fisch U. Otosclerosis: incidence of 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 1159 positive findings on high-resolution computed tomography and their correlation to audiological test data. Ann Otol Rhinol Laryngol 2005;114:709Y16. Schuknecht HF. Disorders of Bone. Philadelphia, PA: Lea and Febiger, 1993:365Y79. Mafee MF, Henrikson GC, Deitch RL, et al. Use of CT in stapedial otosclerosis. Radiology 1985;156:709Y14. Schuknecht HF, Barber W. Histologic variants in otosclerosis. Laryngoscope 1985;95:1307Y17. Ayache D, Sleiman J, Tchuente AN, et al. Variations and incidents encountered during stapes surgery for otosclerosis. Ann Otolaryngol Chir Cervicofac 1999;116:8Y14. Lippy WH, Fucci MJ, Schuring AG, et al. Prosthesis on a mobilized stapes footplate. Am J Otol 1996;17:713Y6. Saunders JE, Derebery MJ, Lo WW. Magnetic resonance imaging of cochlear otosclerosis. Ann Otol Rhinol Laryngol 1995;104: 826Y9. Hueb MM, Goycoolea MV, Paparella MM, et al. Otosclerosis: the University of Minnesota temporal bone collection. Otolaryngol Head Neck Surg 1991;105:396Y405. Guild SR. Histologic otosclerosis. Ann Otol Rhinol Laryngol 1944;53:246Y67. Farrior J, Sutherland A. Revision stapes surgery. Laryngoscope 1991;101:1155Y61. Cokkeser Y, Naguib M, Aristegui M, et al. Revision stapes surgery: a critical evaluation. Otolaryngol Head Neck Surg 1994;111: 473Y7. Hough JV, Dyer RK Jr. Stapedectomy. Causes of failure and revision surgery in otosclerosis. Otolaryngol Clin North Am 1993;26: 453Y70. Somers T, Govaerts P, Marquet T, et al. Statistical analysis of otosclerosis surgery performed by Jean Marquet. Ann Otol Rhinol Laryngol 1994;103:945Y51. Battista RA, Wiet RJ, Joy J. Revision stapedectomy. Otolaryngol Clin North Am 2006;39:677Y97, v-vi. Ginsberg IA, Hoffman SR, Stinziano GD, et al. StapedectomyVin depth analysis of 2405 cases. Laryngoscope 1978;88:1999Y2016. Gristwood RE, Venables WN. Otosclerotic obliteration of oval window niche: an analysis of the results of surgery. J Laryngol Otol 1975;89:1185Y217. Shea JJ, Farrior JB. Stapedectomy and round window closure. Laryngoscope 1987;97:10Y2. Wiet RJ, Harvey SA, Bauer GP. Complications in stapes surgery. Options for prevention and management. Otolaryngol Clin North Am 1993;26:471Y90. Causse JB, Causse JR, Wiet RJ, et al. Complications of stapedectomies. Am J Otol 1983;4:275Y80. Otology & Neurotology, Vol. 30, No. 8, 2009 Copyright @ 2009 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.