This study aimed to compare video head impulse test (vHIT) metrics in relapsing–remitting multiple sclerosis (RRMS) versus controls, test differences by cerebellar MRI status within RRMS, and explore clinical/MRI correlates of corrective saccades.

Single-center prospective study including 35 adults with RRMS and 35 age-/sex-matched controls. Disability was quantified by EDSS; a blinded neuroradiologist classified cerebellar involvement. The MS cohort was categorized into two subgroups according to the occurrence of cerebellar lesions. Each group participated in a vHIT to assess VOR gain.

RRMS participants had low disability (mean EDSS 0.95). Compared with controls, RRMS showed lower VOR gain in the right anterior/posterior, left posterior, and left lateral canals and higher asymmetry across RALP, LARP, and lateral planes (p=0.012, 0.024, <0.001). Corrective saccades differed between groups only in the left anterior canal (p=0.025). Within RRMS, cerebellar-positive versus cerebellar-negative subgroups did not differ for gain or asymmetry (all p>0.05); corrective-saccade prevalence was similar (33.3% vs 35.0%; chi-square p=0.918). In regression, cerebellar involvement, relapse count, EDSS, and disease duration were not associated with recurrent corrective saccades.

vHIT reveals canal-selective gain reductions and elevated asymmetry in patients with mildly disabled RRMS, but these abnormalities do not differ by cerebellar MRI status.