Comparisons with other studies
PPV of DW-MRI is around 96,3% in children and 88,5% in adults (1). Our results are consistent with the ones reported in the meta-analysis led by Van Egmond et al. demonstrating a better PPV in postoperative cases than in primary cases (2).
These PPV are a lot stronger than the ones calculated on CT scan images, which is around 67% (3). CT scan is usually systematic before the first surgery in order to study the petrous bone anatomy. The CT scan could be considered as enough before the first surgery, given that ossicular erosion corresponds to a cholesteatoma diagnosis in 90% of cases (4). However, for follow-up, MRI is preferred, due to the absence of ionizing radiation.
In our series, the use of DWI TSE sequences has led to a high positive predictive value of 89%. DW-MRI hypersignal may artefactual du to T2-shine through phenomenon and thus, ADC map has to be considered. DW-MRI sequence is based on the Brownian motion of water molecules (5). The differential diagnoses of cholesteatoma in DW-MRI are abscess, bone powder, cholesterol granuloma, fat tissue, silicone sheet, wax, proteinaceous fluid, cartilage, tympanosclerosis and squamous cell carcinoma (6). Moreover, other artefacts can also produce false positives: dental braces, eddy current, susceptibility, ghosting, chemical shift or motion (7).
The size of the DW-MRI hypersignal seems to be an important element in the diagnosis. In our study, false positives have a smaller average size than true positives. The inferior limit of detection of the cholesteatoma is between 3 and 5 mm (7,8). Although there is not a precise cut-off, we should be very careful in case of a small DW-MRI hypersignal in order to avoid an unnecessary surgery. One might to choose to perform a complementary MRI 6 months later to assess the changes and eventually propose a surgery if the mass has increased in size and the signal is in favor of a cholesteatoma.