HSC Section 8_April 2017

134 / 264

Non-EPI DW MRI in Planning the Surgical Approach
to Primary and Recurrent Cholesteatoma
*Lela Migirov, *Michael Wolf, † Gahl Greenberg, and †Ana Eyal
*Department of Otolaryngology and Head and Neck Surgery; and ÞDepartment of Diagnostic Imaging, Sheba
Medical Center, Tel Hashomer and Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel
Objective: To investigate a correlation between preoperative
non-echo planar diffusion-weighted magnetic resonance imag-
ing (non-EPI DW MRI) with surgical findings of localization
and extension of cholesteatoma and to develop criteria for sur-
gical planning.
Patients: Preoperative non-EPI DWMRI was available and
positive for cholesteatoma in 27 patients with primary and 23
with residual/recurrent lesions.
Interventions: Patients with cholesteatoma limited to the mid-
dle ear and its extensions were managed with a transcanal en-
doscopic approach. Patients with extension of the cholesteatoma
posteriorly to the lateral semicircular canal underwent retroauricular
mastoidectomy combined with an endoscopic approach.
Main Outcome Measure: Comparison of preoperative radio-
logic to surgical findings.
Results: DWI showed isolated tympanic and attic extension in
33 cases and attico-antral and mastoid extension in 17 cases.
MRI findings correlated with surgical findings in all patients
with primary cholesteatoma, 19 of whom were managed with a
transcanal endoscopic approach and 8 with endoscope-assisted
ear surgery. The transcanal endoscopic approach was applied in
14 of the patients with residual/recurrent cholesteatoma, and the
other 9 residual/recurrent lesions were eradicated using endoscope-
assisted mastoidectomy. DWI overestimated cholesteatoma sites in
1 patient with residual lesion. The smallest cholesteatoma detected
on DWI was a 3-mm lesion in the middle ear over the facial nerve.
Conclusion: Primary and residual/recurrent cholesteatoma was
accurately detected on non-EPI DWI with 98% clinical and
radiologic concordance. Lesions less than 8 mm confined to the
middle ear and its extensions can be eradicated with a minimally
invasive endoscopic transcanal technique, whereas endoscope-
assisted retroauricular mastoidectomy is preferred for larger
lesions. Key words: CholesteatomaVImagingVSurgery.
Otol Neurotol 35:121 Y125, 2014.
Non-echo planar (non-EPI) diffusion-weighted (DW)
magnetic resonance imaging (MRI) has emerged as the
optimal imaging technique for diagnosing the presence
and extent of cholesteatoma. Recent studies have already
shown a high correlation between preoperative non-EPI
DWI and findings at surgery, demonstrating that DWMRI
can accurately predict the presence of cholesteatoma in
both primary and residual cases. The application of non-
EPI DWI with a detection limit for a cholesteatoma as low
as 2 mm is rapidly becoming a widely accepted practice
in the postoperative follow-up of these patients (1Y14).
The surgical management of cholesteatoma tends to
use the least invasive surgical techniques (15 Y18). The
choice of surgical approach depends on the extension of
the disease and on the preoperative otoscopic and radio-
logic findings. Cholesteatoma is usually endoscopically
accessible when the lesion does not involve the mastoid
beyond the level of the lateral semicircular canal (15),
whereas mastoid obliteration techniques can be used in
more extended cases (19). The growing utilization of
endoscopic procedures in the eradication of choles-
teatoma requires precise preoperative imaging data for
assistance in optimal planning of endoscopic ear surgery
(EES) or endoscope-assisted ear surgery (EAES).
The objective of the present work was to investigate
a correlation between preoperative non-EPI DWI and
surgical findings in terms of localization and extension
of primary and residual/recurrent cholesteatoma and to
develop criteria for surgical technique planning. This is
the first study to present the results of endoscopic or
endoscope-assisted ear surgeries that were planned ac-
cording to the preoperative non-EPI DW MRI findings.
METHODS
Only the surgeries performed by the same surgeon (L. M.)
were analyzed in the current study. Between July 2008 and June
2013, an endoscopic approach was applied in 185 surgeries, of
which, 120 were performed for primary (n = 87) or residual/
Address correspondence and reprint requests to: Lela Migirov M.D.,
Department of Otolaryngology and Head and Neck Surgery, Sheba Medical
Center, Tel Hashomer, 5262l, Israel; E-mail:
The authors disclose no conflicts of interest.
Otology & Neurotology
35:121Y125 2013, Otology & Neurotology, Inc.
Reprinted by permission of Otol Neurotol. 2014; 35(1):121-125.
113