A systematic review of diffusion-weighted magnetic resonance imaging in the assessment of postoperative cholesteatoma.

OBJECTIVE: A systematic review to determine whether the diffusion-weighted (DW) magnetic resonance imaging scan can reliably detect residual or recurrent cholesteatoma after mastoid surgery. DESIGN: A systematic review. DATA SOURCES: Databases including EMBASE, MEDLINE, CINAHL, Web of Science, and Cochrane Review were searched for studies published without language restriction from the start of the databases. Additional studies were identified from cited references. SELECTION CRITERIA: Initial search identified 402 publications, of which 16 studies met the inclusion criteria for the systematic review. The DW imaging (DWI) scan was used to detect residual or recurrent cholesteatoma and subsequent second-look surgery was performed to correlate the findings. REVIEW METHODS: Studies were assessed for their selection of patients for radiologic investigations, imaging parameters, and subsequent surgery. Outcome measures included sensitivity, specificity, positive and negative predictive values of the DWI, and the incidence and size of residual or recurrent cholesteatoma. RESULTS: Two different modalities of DWI sequences have been described. Eight studies with 225 patients analyzed echo-planar imaging (EPI) and 8 studies with 207 patients described the "non-EPI" scanning techniques. Non-EPI parameters are more reliable in identifying residual or recurrent cholesteatoma with sensitivity, specificity, and positive and negative predictive values of 91%, 96%, 97%, and 85%, respectively. CONCLUSION: The available evidence suggests that non-EPI such as half-Fourier acquisition single-shot turbo spin echo sequences are more reliable in identifying residual or recurrent cholesteatoma. This is a promising radiologic investigation; however, we think further studies are required with more patients and long-term results to establish its place as an alternative to a second-stage surgery after canal wall up surgery.