Asymmetric Hearing Loss Prompting MRI Referral in a Military Population: Redefining Audiometric Criteria

Abstract

Objective To reevaluate asymmetric sensorineural hearing loss (ASNHL) criteria used to justify magnetic resonance imaging (MRI) in the evaluation of retrocochlear tumors in a military population. Study Design Retrospective case-control study. Setting Tertiary care military medical center. Subjects and Methods Patients with military service and a history of ASNHL prompting referral for MRI, with or without retrocochlear tumors, were compared between 2005 and 2016. Predictor variables included pure tone ASNHL, speech audiometry, and a history of noise exposure. Logistic regression models for hearing asymmetries were performed, and receiver operator curves were used to calculate sensitivity and specificity. Results Thirty-eight retrocochlear tumors were identified. The MRI diagnosis rate for patients with ASHNL was 0.85%. Patients with tumors were slightly older (42 vs 37 years, P = .021) and had less noise exposure (47% vs 85%, P < .001). A sensitivity of 0.83 and a specificity of 0.58 were calculated for asymmetries ≥10 dB at 2000 Hz without adjusting for noise exposure. Instituting this imaging threshold would have reduced the number of MRI scans by half while missing 16% of tumors. Conclusion The tumor diagnosis rate among those undergoing MRI for ASNHL is low in the military population, likely because service-related noise exposure commonly causes ASNHL. Optimal MRI referral criteria should conserve resources while balancing the risks of over- and underdiagnosis. For those with a history of military service, an asymmetry ≥10 dB at 2000 Hz among patients meeting current ANSHL referral criteria is most predictive of a retrocochlear tumor.