Abstract

Background: Recurrence is the main risk that may occur during the follow-up of operated middle ear cholesteatoma. Imaging plays an important role in its diagnosis, leading to avoid surgical second look when it is not mandatory. The aim of our study was to evaluate postoperative CT and MRI in patients who had undergone middle ear cholesteatoma surgery. Methods: Retrospective study from June 2010 to June 2015 including operated patients for middle ear cholesteatoma whom follow-up was made in the ENT department of Rabta hospital and who had postoperative CT and/or MRI in the imaging department. Comparison of radiological and second look surgical findings was made with analysis of sensitivity, specificity, PNV, PPV for each type of imaging exam. Results: Forty ears included (36 patients, median age=38. 5, sex-ratio=1.1). Thirty four ears had CT showing well aerated middle ear cleft (n=1), total opacification (n=7), partial soft-tissue opacity with convex margins (n=11), pearl-shaped lesion (n=7) and concave margins opacity (n=8). CT was not able to further characterize these opacities (specificity 20%) but it was efficient in the evaluation of ossicular and bony walls lysis. Twenty five ears had MRI showing recurrent cholesteatoma (n=15), scar tissue (n=8) and aerated postoperative cavity with alteration of the labyrinth T2 signal (n=2). MRI specificity was about 25%. 100% PNV allowed excluding recurrence when MRI was showing no soft tissue mass. PPV of diffusion weighted imaging (DWI) and delayed post contrast T1 weighted imaging was respectively 83.3% and 71.4%. A hypersignal on DWI and no contrast uptake were highly in favor of cholesteatoma. Conclusion: CT is insufficient for the diagnosis of recurrent cholesteatoma. MRI contribution is hindered by false negatives due to too small lesions to be detected.

Keywords: CT scan; MRI; Cholesteatoma; Middle ear; Tympanoplasty