The head impulse, nystagmus type, test of skew (HINTS) protocol set a new paradigm to differentiate peripheral vestibular disease from stroke in patients with acute vestibular syndrome (AVS). The relationship between degree of truncal ataxia and stroke has not been systematically studied in patients with AVS. We studied a group of 114 patients who were admitted to a General Hospital due to AVS, 72 of them with vestibular neuritis (based on positive head impulse, abnormal caloric tests, and negative MRI) and the rest with stroke: 32 in the posterior inferior cerebellar artery (PICA) territory (positive HINTS findings, positive MRI) and 10 in the anterior inferior cerebellar artery (AICA) territory (variable findings and grade 3 ataxia, positive MRI). Truncal ataxia was measured by independent observers as grade 1, mild to moderate imbalance with walking independently; grade 2, severe imbalance with standing, but cannot walk without support; and grade 3, falling at upright posture. When we applied the HINTS protocol to our sample, we obtained 100% sensitivity and 94.4% specificity, similar to previously published findings. Only those patients with stroke presented with grade 3 ataxia. Of those with grade 2 ataxia (n = 38), 11 had cerebellar stroke and 28 had vestibular neuritis, not related to the patient’s age. Grade 2–3 ataxia was 92.9% sensitive and 61.1% specific to detect AICA/PICA stroke in patients with AVS, with 100% sensitivity to detect AICA stroke. In turn, two signs (nystagmus of central origin and grade 2–3 Ataxia) had 100% sensitivity and 61.1% specificity. Ataxia is less sensitive than HINTS but much easier to evaluate.

vHIT provides phenotypic information that differentiates these autosomal ataxias and can serve as a strategy to orient genetic diagnosis. A correlation between VOR and SARA raises the possibility of using VOR gain as a neurophysiologic biomarker for disease severity.

Differentiating central from peripheral origins of vestibulo-ocular reflex (VOR) lesions can be challenging. A 36-year old man presented with a 1-year history of progressive unsteadiness. The video-Head Impulse Test revealed a significantly reduced VOR gain in both horizontal and posterior canals (0.49 ± 0.05 and 0.38 ± 0.06) but normal VOR responses in both anterior canals (0.89 ± 0.08 and 1.04 ± 0.15). No plausible combination of end-organ lesion should be responsible for these observations. A brain magnetic resonance imaging disclosed a left inferior cerebellar peduncle lesion suggestive of a glioma.