Test–Retest Reliability of Video Head Impulse Test in Healthy Individuals and Individuals with Dizziness

Abstract

AbstractVideo head impulse test (vHIT) is widely accepted as a test for the assessment of functional integrity of semicircular canals (SCCs). It allows for the evaluation of the functioning of all six SCCs independent of each other. It works on the principle of the vestibulo-ocular reflex (VOR). In individuals with vestibular pathologies, the VOR is impaired, and hence, the use of vHIT may provide vital information about the functional status of SCCs and the VOR pathway originating from them.In the recent past, studies reported excellent test–retest reliability of vHIT in healthy individuals. However, these studies used analysis of variance or the nonparametric counterpart Wilcoxon signed-rank test, which are insufficient statistical methods for conclusions about test–retest reliability. Further, because vHIT assesses VOR function in individuals with vestibular pathologies, it is important to assess test–retest reliability in the pathological group as well. Therefore, the present study aimed to evaluate test–retest reliability of vHIT in healthy individuals and individuals with vestibular pathology.Repeated measures.Twenty healthy individuals with no history of vestibular pathology and 20 individuals with known vestibular pathology were included.Each participant underwent vHIT testing for all three SCCs of both sides on four different occasions. VOR gain and the presence of pathological saccades were noted and analyzed for each recording.Intraclass correlation coefficient (ICC) revealed excellent test–retest reliability for VOR gain in both groups (ICC ≥ 0.76). Kappa coefficient analysis for the presence of refixation saccades demonstrated moderate to excellent agreement between test sessions (K ≥ 0.63) for the lateral canal. For the anterior and posterior SCC, there was large variability between sessions for refixation saccades.This study provides evidence about test–retest reliability of VOR gain and refixation saccades assessed using vHIT in healthy individuals and individuals with vestibulopathies. These findings suggest that both measures are highly reliable and replicable across test sessions, except refixation saccades in vertical canals which varied between sessions in some individuals.