• Yi Wei Wu, Amit Karandikar, Julian Pn Goh, and Tiong Yong Tan.
• Department of Diagnostic Radiology, Tan Tock Seng Hospital, Singapore.
• Ann Acad Med Singap. 2020 Feb 1; 49 (2): 65-71.

IntroductionThis study aimed to identify imaging features on single-sequence noncontrast magnetic resonance imaging (MRI) that differentiate the vestibular ganglion from small intracanalicular schwannomas.Materials And MethodsNinety patients (42 men and 48 women; age: 24‒87 years old) with 102 internal auditory canal (IAC) nodules (59 vestibular ganglia and 43 intracanalicular schwannoma) who underwent both single-sequence T2-weighted (T2W) non-contrast enhanced MRI studies and contrast-enhanced T1-weighted (T1W) MRI studies between May 2012 and April 2017 were evaluated. The length, width, distance to the IAC fundus and length/width ratios for all lesions were obtained and compared among groups. Diagnostic performance and cutoff values of the parameters were evaluated with receiver operating characteristics curve analysis. Area under the curve (AUC) value was calculated.ResultsVestibular ganglia have significantly smaller lengths and widths compared to intracanalicular vestibular schwannomas (1.7 ± 0.4 mm and 1.0 ± 0.2 mm versus 5.6 ± 3.0 mm and 3.7 ± 1.5 mm). They are more fusiform in shape compared to vestibular schwannomas (length/width ratio: 1.8 ± 0.4 versus 1.5 ± 0.4). The lesion width demonstrated the highest diagnostic performance (AUC: 0.998). Using a cutoff width of <1.3 mm, the sensitivity, specificity and overall accuracy for diagnosing vestibular ganglia were 97% (57/59), 100% (43/43) and 98% (100/102), respectively.ConclusionVestibular ganglia may mimic intracanalicular vestibular schwannomas on a single-sequence T2W MRI. However, a fusiform shape and width <1.3 mm increases confidence in the diagnosis of ganglia. Identifying the vestibular ganglion on single-sequence T2W MRI studies may obviate the need for a contrast-enhanced MRI, reducing the risks of contrast administration, additional scanning time and cost.

22 keywords...

hide…