The value of non-echo planar HASTE diffusion-weighted MR imaging in the detection, localisation and prediction of extent of postoperative cholesteatoma.

OBJECTIVE

To evaluate the diagnostic performance of half-Fourier-acquisition single-shot turbo-spin-echo (HASTE) diffusion-weighted magnetic resonance imaging in the detection, localisation and prediction of extent of cholesteatoma following canal wall up mastoid surgery.

STUDY DESIGN

Prospective blinded observational study.

SETTING

University affiliated teaching hospital.

PATIENTS

Forty-eight patients undergoing second-look surgery after previous canal wall up mastoid surgery for primary acquired cholesteatoma.

INTERVENTIONS

All patients underwent non-echo planar HASTE diffusion-weighted imaging prior to being offered 'second-look' surgery.

MAIN OUTCOME MEASURES

Radiological findings were correlated with second-look intra-operative findings in 38 cases with regard to presence, location and maximum dimensions of cholesteatoma.

RESULTS

Half-Fourier-acquisition single-shot turbo-spin-echo diffusion-weighted imaging accurately predicted the presence of cholesteatoma in 23 of 28 cases, and it correctly excluded in nine of 10 cases. Five false negatives were caused by keratin pearls of <2 mm and in one case 5 mm. Overall sensitivity and specificity for detection of cholesteatoma were 82% (95% confidence interval [CI] 62-94%) and 90% (CI 55-100%), respectively. Positive predictive value and negative predictive value were 96% (CI 79-100%) and 64% (CI 35-87%), respectively. Overall accuracy for detection of cholesteatoma was 84% (CI 69-94%). Half-Fourier-acquisition single-shot turbo-spin-echo diffusion-weighted imaging has good performance in localising cholesteatoma to a number of anatomical sub-sites within the middle ear and mastoid (sensitivity ranging from 75% to 88% and specificity ranging from 94% to 100%). There was no statistically significant difference in the size of cholesteatoma detected radiologically and that found during surgery (paired t-test, P = 0.16). However, analysis of size agreement suggests possible radiological underestimation of size when using HASTE diffusion-weighted imaging (mean difference -0.6 mm, CI -5.3 to 4.6 mm).

CONCLUSION

Half-Fourier-acquisition single-shot turbo-spin-echo diffusion-weighted imaging performs reasonably well in predicting the presence and location of postoperative cholesteatoma but may miss small foci of disease and may underestimate the true size of cholesteatoma.