Explore this issue:
August 2018Comment: The non-echo planar diffusion-weighted MRI (non-EP DWI) sequence is efficient in identifying the restricted diffusion in the substance of cholesteatoma, and, thanks to its sensitivity and precision, has changed the way otolaryngologists manage chronic ear disease. Even in diagnostic testing with high sensitivity and specificity, false positives and false negatives occur. In an attempt to decrease the incidence of erroneous results, we can examine the role of T1 imaging in cholesteatoma diagnosis. On T1 imaging, cholesteatomas are characteristically hypointense.
These authors examined 57 ears in 53 patients. A positive non-EP DWI alone identified 93% of the 46 ears with cholesteatoma (sensitivity), but a negative non-EP DWI only excluded cholesteatoma in 63% of patients who did not harbor cholesteatomas (specificity). When T1 imaging was added, 89% of cholesteatomas were correctly identified (sensitivity), but all (100%) 11 patients who did not have cholesteatoma were identified (specificity). Accuracy as defined by (true positives + true negatives)/all patients, improved from 87.7% for non-EP DWI alone to 91.2% with the addition of T1 imaging.
Because not every patient with nonspecific opacification of the mastoid on CT has cholesteatoma, and because non-cholesteatomatous ear disease may not require surgery, the results from the study could further help clinicians make decisions about treatment of the patient with chronic ear disease. —Samuel H. Selesnick, MD
What is the usefulness of T1-weighted (T1W) imaging on diagnostic magnetic resonance (MR) imaging for cholesteatoma?
Bottom line
T1W imaging may aid in the exclusion of false-positive cases on diagnostic non-EP DW MR imaging for cholesteatomas. A combination of non-EP DW and T1W imaging may improve the specificity and accuracy compared with non-EP DW imaging alone.
Study design: retrospective case review.
Setting: Tertiary referral center, Hokkaido University, Hokkaido, Japan
Synopsis: The authors identified 53 patients (57 ears) suspected to have cholesteatomas. Intervention included preoperative MR imaging, including non-echo planar (non-EP) diffusion-weighted (DW) and T1W imaging.
The primary outcome measures included the comparison between the diagnostic accuracy for the detection of cholesteatomas using non-EP DW imaging alone (criterion 1) and non-EP DW imaging along with T1W imaging (criterion 2). Diagnostic accuracy was evaluated in each case by comparing MR imaging with surgical findings. Secondary outcome measures included the comparison of the rates of cases showing a high T1W signal between cholesteatomas and noncholesteatomas, which showed a high non-EP DW signal.
The authors found that the sensitivity, specificity, and accuracy according to criterion 1 were 93.5, 63.6, and 87.7% and according to criterion 2 were 89.1, 100, and 91.2%, respectively. Of 43 cholesteatoma cases indicating a high non-EP DW signal, only two cases showed a high T1W signal (5%). On the other hand, all four noncholesteatoma cases indicating high non-EP DW signal showed a high T1W signal (100%), and these rates were significantly different (p < 0.001).