Abstract 350: Optimizing Stroke Assessment of Dizziness: The Role and Impact of Video Head Impulse Testing

Introduction Dizziness is a frequent and challenging symptom in patients with posterior ischemic stroke, often making it difficult to distinguish from peripheral vestibular disorders. The video head impulse test (vHIT) is a non‐invasive tool that measures the vestibulo‐ocular reflex (VOR) to assess semicircular canal function. This study evaluates the effectiveness of vHIT in the emergency department (ED) for differentiating between central (stroke‐related) and peripheral causes of dizziness. The primary objective is to confirm that vHIT is an effective tool for enhancing diagnostic confidence in ruling out stroke and avoiding unnecessary admissions, thereby optimizing resource allocation for other stroke patients. Methods In this prospective study, 30 patients who presented to the ED of the stroke center with dizziness and were subsequently admitted under standard stroke protocol were analyzed. Each patient underwent vHIT in addition to the standard stroke evaluation, which included MRI. The vHIT assessed the VOR in the lateral semicircular canal to detect vestibular hypofunction through measured gain reduction and saccadic eye movements. Two specialists independently analyzed the vHIT results, both blinded to the MRI findings. Results Of the 30 patients included in the study, MRI was performed on 27, serving as the gold standard for stroke detection. MRI confirmed the presence of a stroke in 10 cases (37%). Among these patients, the vHIT indicated a possible central etiology in 9 patients, with a sensitivity of 100% (95% CI: 66.37% to 100%). One patient's vHIT recordings were of poor quality, making the results uninterpretable. In the subgroup of 17 patients with negative MRI findings, vHIT suggested a peripheral cause of dizziness in 2 patients. The specificity of vHIT for central etiology was 15.38% (95% CI: 1.92% to 45.45%) with a negative predictive value of 100% (95% CI: 15.81% to 100.00%) and a positive predictive value of 40.97% (95% CI: 35.50% to 46.67%). One patient with a negative MRI but positive vHIT for peripheral etiology was unnecessarily admitted for 3 days. Conclusion Our findings suggest that incorporating vHIT into the ED workflow could be an effective strategy for ruling out stroke before proceeding with extensive stroke admissions and workups. vHIT appears to be a valuable tool in the initial assessment of patients with dizziness, facilitating the differentiation between central and peripheral causes. This approach has the potential to reduce unnecessary hospital admissions for patients without central pathology.