Institutional review board approval and informed consent from all participants were obtained. We studied 30 patients with preoperatively suspected acquired cholesteatoma. Each patient underwent an MR examination including both SS-EP DWI and DSDE-TFE using a 3.0 T MR scanner. Images of the 30 patients (60 temporal bones including 30 with and 30 without cholesteatoma) were reviewed by two independent neuroradiologists. The confidence level for the presence of cholesteatoma was graded on a scale of 0-2 (0 = definite absence, 1 = equivocal, 2 = definite presence). Interobserver agreement as well as sensitivity, specificity, and accuracy for detection were assessed for the two reviewers.
Excellent interobserver agreement was shown for TFE-DSDE (¦Ê = 0.821) whereas fair agreement was obtained for SS-EP DWI (¦Ê = 0.416). TFE-DSDE was associated with significantly higher sensitivity (83.3 % ) and accuracy (90.0 % ) compared to SS-EP DWI (sensitivity = 35.0 % , accuracy = 66.7 % ; p < 0.05). No significant difference was found in specificity (96.7 % for TFE-DSDE, 98.3 % for SS-EP DWI)
With increased spatial resolution and reduced susceptibility artifacts, TFE-DSDE improves the accuracy in diagnosing acquired middle ear cholesteatomas compared to SS-EP DWI.