Clinical testing of otolith function: perceptual thresholds and myogenic potentials

Abstract

Cervical and ocular vestibular-evoked myogenic potential (cVEMP/oVEMP) tests are widely used clinical tests of otolith function. However, VEMP testing may not be the ideal measure of otolith function given the significant inter-individual variability in responses and given that the stimuli used to elicit VEMPs are not physiological. We therefore evaluated linear motion perceptual threshold testing compared with cVEMP and oVEMP testing as measures of saccular and utricular function, respectively. A multi-axis motion platform was used to measure horizontal (along the inter-aural and naso-occipital axes) and vertical motion perceptual thresholds. These findings were compared with the vibration-evoked oVEMP as a measure of utricular function and sound-evoked cVEMP as a measure of saccular function. We also considered how perceptual threshold and cVEMP/oVEMP testing are each associated with Dizziness Handicap Inventory (DHI) scores. We enrolled 33 patients with bilateral vestibulopathy of different severities and 42 controls to have sufficient variability in otolith function. Subjects with abnormal oVEMP amplitudes had significantly higher (poorer) perceptual thresholds in the inter-aural and naso-occipital axes in age-adjusted analyses; no significant associations were observed for vertical perceptual thresholds and cVEMP amplitudes. Both oVEMP amplitudes and naso-occipital axis perceptual thresholds were significantly associated with DHI scores. These data suggest that horizontal perceptual thresholds and oVEMPs may estimate the same underlying physiological construct: utricular function.

FIG. 1.

Individual staircase history and psychometric fits for two different subjects in two different conditions. A Staircase history for 2D1U staircase terminated after 50 trials. Filled and unfilled points show correct and incorrect responses, respectively. B Cumulative Gaussian fit to data from (A). Proportion of correct responses is plotted as a function of the log of the stimulus magnitude. Threshold is the stimulus value corresponding to 84 % correct shown by red dashed lines in (A) and (B). C Staircase history same as in (A) except a 3D1U stepping rule was used and the staircase terminated after ten total reversals. D Cumulative Gaussian fit to data from (C).

FIG. 2.

Sample right (A) and left (B) ocular VEMP traces from a control subject. N1 and P1 peaks are marked. Sample right (C) and left (D) cervical VEMP traces from a control subject; two separate trials are depicted for each side. P1 (P13) and N1 (N23) peaks are marked. N1 amplitudes were used in oVEMP analyses, and peak-to-peak amplitudes were used in cVEMP analyses, as specified in the text.

FIG. 3.

Spearman's rank correlations between linear motion perceptual thresholds and VEMP amplitudes and age in controls. A Correlation between IA axis threshold and age (r = 0.22, p = 0.1532). B Correlation between HV axis threshold and age (r = 0.31, p = 0.0458). C Correlation between corrected cVEMP peak-to-peak amplitude and age (r = −0.53, p = 0.0006). D Correlation between NO axis threshold and age (r = 0.44, p = 0.0032). E Correlation between oVEMP n1 amplitude and age (r = −0.08, p = 0.5771).