Quantitative Video-oculography to Differentiate Stroke from Vestibular Neuritis in Acute Vertigo―Vestibulo-ocular Reflex Gain Distributions and Optimal Test Thresholds for Diagnosis (P6.289)

OBJECTIVE: Describe the range and distribution of vestibulo-ocular reflex (VOR) gain measurements in patients with acute, continuous vertigo/dizziness to identify optimal thresholds for discriminating central from peripheral causes. BACKGROUND: The acute vestibular syndrome (AVS) is usually caused by vestibular neuritis or stroke. Clinical presentations are similar, but strokes can be differentiated from neuritis based on eye movements. The horizontal head impulse test (h-HIT) of VOR function is the single best bedside predictor. The h-HIT can be measured by quantitative video-oculography, enabling more consistent diagnostic interpretation. DESIGN/METHODS: Prospective, observational study at two academic medical centers (08/2011-06/2012). We recruited adult ED patients with AVS defined as new, persistent vertigo/dizziness, nystagmus, and (1) nausea/vomiting, (2) head motion intolerance, or (3) new gait unsteadiness. We recorded the h-HIT VOR using a portable video-oculography device. Gold standard diagnoses were determined by MRI with diffusion-weighted imaging and follow-up. The number of h-HIT trials needed was determined using general linear mixed model-based simulations. RESULTS: We enrolled 26 patients (16 vestibular neuritis, 10 strokes). Strokes were located in the posterior inferior cerebellar artery (PICA) territory (n=7) and the anterior inferior cerebellar artery (AICA) territory (n=3). Vestibular neuritis patients had ipsilesional VOR deficits, PICA stroke patients had no VOR deficits, and AICA stroke patients had variable VOR gain patterns (no deficit, unilateral deficit, bilateral deficit). Mean gains (ipsilesional, contralesional with standard errors [SE]) were as follows: vestibular neuritis (0.52 [SE 0.04], 0.87 [SE 0.04]), PICA strokes (0.94 [SE 0.04], 0.93 [SE 0.04]), AICA strokes (0.84 [SE 0.10], 0.74 [SE 0.10]). A mean gain threshold of 0.70 (below=peripheral; above=central) was optimal (90% sensitivity, 88% specificity for stroke). Ten h-HIT trials per ear was sufficient to achieve consistent classification. CONCLUSIONS: Peripheral and central causes of AVS can be differentiated by quantitative, portable video-oculography. Some AICA strokes mimic vestibular neuritis closely. Disclosure: Dr. Saber Tehrani has nothing to disclose. Dr. Mantokoudis has nothing to disclose. Dr. Wozniak has nothing to disclose. Dr. Eibenberger has nothing to disclose. Dr. Kattah has received personal compensation for activities with Pfizer, Inc. as a consultant. Dr. Guede has nothing to disclose. Dr. Zee has received personal compensation for activities with Sun Pharma and Abbott. Dr. Zee has received royalty payments from Oxford University Press. Dr. Newman-Toker has received personal compensation for activities with Janssen Pharmaceuticals. Dr. Newman-Toker has received compensation for serving on the board of Society to Improve Diagnosis in Medicine. Dr. Newman-Toker has received research support from GN Otometrics.