RESUMO
OBJECTIVES: The purpose of this study is to evaluate the reliability of temporal bone density measurements for diagnosing otosclerosis. MATERIALS AND METHODS: A retrospective case-control study is presented. Bone density was measured in Hounsfield units (HUs) by using high-resolution computed tomography in eight regions of interest (ROI) where otosclerotic foci are usually localized. The density of 113 otosclerotic ears was compared with that of 33 nonotosclerotic ears to determine sensitivity and specificity. Furthermore, the binormal receiver operating characteristic curve of each ROI's density was calculated to estimate the diagnostic value for osteosclerosis. In addition, the radiological density of seven cases-where radiological visual examination exhibited no findings but surgery confirmed stapes fixation-was compared with nonotosclerotic controls. RESULTS: ROI densities were significantly lower in otosclerotic patients compared with nonotosclerotic controls. The area under the curve of the fissula ante fenestram (FAF) presented the highest diagnostic performance: 1,871 HU cut-off value (area under the curve = 0.986), 96.64% sensitivity, and 100% specificity. Significantly lower densities in the FAF area were observed in the seven cases with negative radiology but intraoperatively confirmed otosclerosis. CONCLUSION: The high-resolution computed tomography density of the FAF is a reliable measurement for diagnosing otosclerosis. A value less than 1,871 HU exhibited the highest sensitivity and specificity in a European Caucasian population.