Normalization Reduces Intersubject Variability in Cervical Vestibular Evoked Myogenic Potentials

Abstract

Cervical vestibular evoked myogenic potentials are used to assess saccular and inferior vestibular nerve function. Normalization of the VEMP waveform has been proposed to reduce the variability in vestibular evoked myogenic potentials by correcting for muscle activation. In this study, we test the hypothesis that normalization of the raw cervical VEMP waveform causes a significant decrease in the intersubject variability. Prospective cohort study. Large specialty hospital, department of otolaryngology. Twenty healthy subjects were used in this study. All subjects underwent cervical vestibular evoked myogenic potential testing using short tone bursts at 250, 500, 750, and 1,000 Hz. Both intersubject and intrasubject variability was assessed. Variability between raw and normalized peak-to-peak amplitudes was compared using the coefficient of variation. Intrasubject variability was assessed using the intraclass correlation coefficient and interaural asymmetry ratio. cVEMPs were present in most ears. Highest peak-to-peak amplitudes were recorded at 750 Hz. Normalization did not alter cVEMP tuning characteristics. Normalization of the cVEMP response caused a significant reduction in intersubject variability of the peak-to-peak amplitude. No significant change was seen in the intrasubject variability. Normalization significantly reduces cVEMP intersubject variability in healthy subjects without altering cVEMP characteristics. By reducing cVEMP amplitude variation due to nonsaccular, muscle-related factors, cVEMP normalization is expected to improve the ability to distinguish between healthy and pathologic responses in the clinical application of cVEMP testing.