AUTHOR=Tran Emma D. , Swanson Austin , Sharon Jeffrey D. , Vaisbuch Yona , Blevins Nikolas H. , Fitzgerald Matthew B. , Steenerson Kristen K. 

TITLE=Ocular Vestibular-Evoked Myogenic Potential Amplitudes Elicited at 4 kHz Optimize Detection of Superior Semicircular Canal Dehiscence

JOURNAL=Frontiers in Neurology

VOLUME=Volume 11 - 2020

YEAR=2020

URL=https://www.frontiersin.org/journals/neurology/articles/10.3389/fneur.2020.00879

DOI=10.3389/fneur.2020.00879

ISSN=1664-2295

ABSTRACT=Introduction: High-resolution temporal bone computed tomography (TBCT) is considered the gold standard for diagnosing superior semicircular canal dehiscence (SCD). Ocular vestibular-evoked myogenic potentials (oVEMPs)—most commonly conducted at 500 Hz stimulation—are increasingly used to detect SCD. Previous research demonstrated that oVEMP n10 response elicited at 4 kHz can have near-perfect sensitivity and specificity in detecting radiographic SCD. With a larger cohort, we seek to determine whether assessing the 4 kHz oVEMP n10-p15 amplitude rather than the binary n10 response alone would optimize the detection of clinically significant SCD, as well as subgroups of radiographic, symptomatic, and surgical SCD. 

Methods: A cross-sectional study of patients who have undergone oVEMP testing at 4kHz. Using the diagnostic criteria proposed by Ward et al., patients were determined to have SCD if dehiscence was confirmed on temporal bone CT by two reviewers, patient-reported characteristic symptoms, and if they had at least one positive physiologic test suggestive of SCDS. Receiver operating characteristic (ROC) analysis was conducted to identify the optimal 4 kHz oVEMP amplitude cut-off. Comparison of 4 kHz oVEMP amplitude across radiographic, symptomatic, and surgical SCD subgroups was conducted using the Mann-Whitney U test.

Results: 902 patients underwent 4 kHz oVEMP testing. After evaluating 150 TBCTs, we identified 49 patients (n, ears = 61) who had radiographic SCD. Of those, 33 patients (n, ears = 36) were determined to have clinically significant SCD. 4 kHz oVEMP n10 responses had a sensitivity of 86.5% and a specificity of 87.8% for detecting SCD. ROC analysis demonstrated that accounting for the inter-amplitude of 4kHz oVEMP was more accurate in detecting SCD than the presence of n10 response alone (AUC 91% vs 87%). Additionally, using an amplitude cut-off of 15uV improves specificity to 96.8%. Surgical and symptomatic SCD cases had significantly higher 4 kHz oVEMP amplitudes, while radiographic SCD cases without characteristic symptoms had similar amplitudes, compared to cases without no evidence of SCD.

Conclusion: Our results suggest that accounting for 4 kHz oVEMP amplitude can improve detection of SCD compared to the binary presence of n10 response. The optimal 4kHz oVEMP amplitude cut-off for our cohort is 15 uV.