Comparative assessment of computed tomography and magnetic resonance imaging in the identification of the middle ear cholesteatoma

Abstract

Background: Chronic suppurative otitis media amounts up to 22.4% among all ear, nose and throat disorders. Its complication, a middle ear cholesteatoma, is one of the most frequent causes of patients' referral to an otologist. The literature on the preferential diagnostic method for the cholesteatoma is contradictory, despite that its main treatment approach is surgery. Therefore, it is important to identify the most valid diagnostic method, which would allow for the planning of the most effective surgical treatment. Aim: To compare sensitivity and specificity of computed tomography (CT) and magnetic resonance imaging (MRI) in the diagnosis of cholesteatoma, to assess the possibility of quantitative values of MRI diffusion limitation in cholesteatoma. Materials and methods: From 2015 to 2021, we examined 542 out- and in-patients (849 imagings of temporal bones) with chronic suppurative otitis media. The analysis of CT and MRI sensitivity and specificity included the data from 289 patient examinations, both with newly diagnosed cholesteatoma and having at least one past surgery and had their diagnosis verified intraoperatively and histologically. We analyzed the measured MRI diffusion coefficients from newly diagnosed and relapsed cholesteatoma. The MRI signal values were calculated for 266 masses from 238 patients. Results: MRI sensitivity for the diagnosis of cholesteatoma was 95.2%, specificity 81.2%. CT sensitivity and specificity for the diagnosis of cholesteatoma were 60.1 и 44.7%, respectively. There were no differences in the measured MRI diffusion coefficients between the relapsed and newly diagnosed cholesteatomas. The comparison of cholesteatoma signals with those from artifacts showed the overlapping in their mean values; therefore, it is impossible to rely only on the value of the diffusion coefficient. Conclusion: In the diagnosis of cholesteatomas, MRI is significantly more sensitive and specific than CT. No significant differences in MRI semiotics and the degree of MRI diffusion limitation at NON-EPI DWI (b1000) have been found for the cholesteatomas that had never been operated, and the relapsing cholesteatomas.