Background Angular vestibular ocular reflex (VOR) can be non-invasively and easily assessed with the video head-impulse test (vHIT), allowing the quantification of VOR dynamic parameters (latency and gain) at a physiological frequency domain, as well as the characterisation of quick eye movements (QEM) triggered with head impulses. Here we characterise these findings in HD patients.

Methods We explored 18 genetically confirmed HD patients (44.7 ± 8.1 years; male=9), classified as Shoulson and Fahn severity stages 1 (n = 5; pre-symptomatic), 2 (n = 4), 3 (n = 8) and 4 (n = 1) based on their Total Functional Capacity (TFC) scores, and 40 healthy controls (39.9 ± 16.5 years; male=20). We calculated the VOR latency and gain and determined the latency, peak-velocity and occurrence rate of the QEM triggered during and after head impulses. All patients had an oculomotor evaluation (spontaneous nystagmus, smooth pursuit, gaze fixation and saccades) before vHIT testing.

Results VOR latency and VOR gain were not different from controls (p > 0.29 for both comparisons). No differences were found between presymptomatic and symptomatic patients. QEM were present in 11 patients and 18 controls, always after the head impulse and compensatory (overt saccades). QEM latency (174.5 ± 41.2 ms), peak velocity (71.58 ± 34.48 º/s) and occurrence rate (0.45 ± 0.29) were not different from controls (p > 0.11 for all comparisons). Though not realising it, 5 of the symptomatic patients majorly failed to direct gaze in darkness, both horizontal as vertically, this correlating with TFC scores (Spearman r = 0.65, p = 0.005).

Conclusions VOR seems to be preserved at physiological frequency domains in HD patients, even in more advanced stages of the disease. Although both voluntary saccades and quick phases of nystagmus are known to be slower in HD, quick eye movements triggered with impulses showed no differences in comparison to controls. Gaze failure in darkness may prove beneficial as a biological marker for HD.