Abstract
OBJECTIVE: 1) Characterize corrective saccades (CS) in normal controls, and 2) examine the sensitivity of the video head impulse test (vHIT) for identifying vestibular loss using both gain and CS. STUDY DESIGN: Prospective combined with retrospective review. SETTING: Tertiary referral center. PATIENTS: Seventy subjects with normal vestibular function served as controls (mean age, 44.1 yr; range, 10-78) and data from 49 patients with unilateral and bilateral vestibular loss was retrospectively reviewed (mean age, 50; range, 7-81). INTERVENTION: vHIT; individual horizontal head impulses were then analyzed in MATLAB. MAIN OUTCOME MEASURES: Horizontal vHIT gain, CS peak velocity, frequency, and latency. RESULTS: There was not an age effect for CS velocity or latency, and only a weak relationship between CS frequency and age in the control group. Gain and CS latency were the only parameters affected by impulse side, demonstrating higher gain and longer latency on the right. The group with vestibular loss had significantly lower mean vHIT gain, higher mean CS frequency, higher mean CS velocity, earlier CS latency, and smaller mean CS standard deviations of the latency compared with the control group.When all factors were analyzed separately by logistic regression, vHIT gain provided the best classification (83.8%), closely followed by CS frequency (83.1%). Using a two variable approach (both gain and CS frequency) yielded the best diagnostic accuracy (overall classification = 84.6%). CONCLUSIONS: Along with gain, incorporating CS frequency in interpreting vHIT improves diagnostic accuracy. A repeatable CS (>81.89%) and/or low gain (<0.78) indicate vestibular loss.