Secondary Logo

Journal Logo

Institutional members access full text with Ovid®

Negative Predictive Value of Non-Echo-Planar Diffusion Weighted MR Imaging for the Detection of Residual Cholesteatoma Done at 9 Months After Primary Surgery Is not High Enough to Omit Second Look Surgery

Horn, Roelof J.*; Gratama, Jan Willem C.*; van der Zaag-Loonen, Hester J.; Droogh-de Greve, Kitty E.*; van Benthem, Peter-Paul G.

doi: 10.1097/MAO.0000000000002270

Objectives: To evaluate non echo-planar diffusion weighted magnetic resonance imaging (non-EP DW MRI) at 9 months after primary surgery to rule out residual cholesteatoma in patients scheduled before second-look-surgical exploration.

Study Design: Prospective observational study.

Setting: Secondary teaching hospital.

Patients/Interventions: Patients who were scheduled for second-look-surgery after primary canal wall up repair of cholesteatoma underwent 1.5 T MRI including non-EP DWI and high-resolution coronal T1 and T2-FS SE sequences.

Main Outcome Measures: Imaging studies were evaluated for the presence of cholesteatoma by three independent observers. Intraoperative observations were regarded the standard of reference. Ear, nose, throat (ENT) surgeons were blinded for imaging findings. The primary outcome was the negative predictive value (NPV) of MR imaging, secondary outcomes were sensitivity, specificity, and positive predictive value.

Results: Thirty-three patients underwent both MRI and surgery, among whom 22 had a cholesteatoma. Mean time between primary surgery and MRI was 259 days (standard deviation [SD] 108). NPV of non-EP DW MRI in detecting recurrent cholesteatoma was 53% (95% CI: 32–73%). Sensitivity and specificity were 59% (39–77%) and 91% (62–98%), respectively. The positive predictive value was 93% (69–99%). In five out of nine false-negative cases, recurrent cholesteatoma measured 3 mm or less. Using a 3 mm detection threshold, NPV increased to 79%.

Conclusion: Non-EP DW MRI cannot replace second look surgery in ruling-out residual cholesteatoma at 9 months after primary surgery. It could be used in a follow-up strategy in low risk patients. Further research is needed which types of residual cholesteatoma are not revealed by MRI.

*Department of Radiology

Department of Clinical Epidemiology

Department of Ear Nose and Throat, Gelre Ziekenhuizen Apeldoorn, The Netherlands

Address correspondence and reprint requests to Roelof J. Horn, M.D., Department of Radiology, St Anna Ziekenhuis, PO BOX 90, 5660 AB Geldrop, The Netherlands; E-mail:

Authors’ contributions: All authors agree to be accountable for all aspects related to accuracy or integrity of this work. Study concepts/design: R.J.H., J.W.C.G., P.P.G.B. Data acquisition and analysis: R.J.H., J.W.C.G., K.E.D. Literature research: R.J.H.

Statistical analysis: H.J.Z. Manuscript drafting, editing and final approval: all authors.

§Present address of author Roelof J. Horn: Department of Radiology, St Anna Ziekenhuis Geldrop, The Netherlands.

Present address of author Peter-Paul G. van Benthem: Department of ORL-HNS, Leiden University Medical Centre, Leiden, The Netherlands.

The authors disclose no conflicts of interest.

Copyright © 2019 by Otology & Neurotology, Inc. Image copyright © 2010 Wolters Kluwer Health/Anatomical Chart Company