Stroke Prediction Based on the Spontaneous Nystagmus Suppression Test in Dizzy Patients: A Diagnostic Accuracy Study

Abstract

Objective: Failure of fixation suppression of spontaneous nystagmus is sometimes seen in patients with vestibular strokes involving the cerebellum or brainstem; however, the accuracy of this test for the discrimination between peripheral and central causes in patients with an acute vestibular syndrome (AVS) is unknown.

Methods: Patients with AVS were screened and recruited (convenience sample) as part of a prospective cross-sectional study in the emergency department between 2015 and 2020. All patients received neuroimaging, which served as a reference standard. We recorded fixation suppression with video-oculography (VOG) for forward, right, and left gaze. The ocular fixation index (OFI) and the spontaneous nystagmus slow velocity reduction was calculated.

Results: We screened 1,646 patients reporting dizziness in the emergency department and tested for spontaneous nystagmus in 148 patients with AVS. We analyzed 56 patients with a diagnosed acute unilateral vestibulopathy (vestibular neuritis) and 28 patients with a confirmed stroke. There was a complete nystagmus fixation suppression in 49.5% of patients with AVS, in 40% of patients with vestibular neuritis, and in 62.5% of patients with vestibular strokes. OFI scores had no predictive value for detecting strokes; however, a nystagmus reduction of less than 2 °/s showed a high accuracy of 76.9% (confidence interval 0.59-0.89) with a sensitivity of 62.2% and specificity of 84.8% in detecting strokes.

Conclusions: The presence of fixation suppression does not rule out a central lesion. The magnitude of suppression was lower compared to patients with vestibular neuritis. The nystagmus suppression test predicts vestibular strokes accurately provided that eye movements are recorded with VOG.

Classification of evidence: This study provides Class II evidence that in patients with an AVS, decreased fixation suppression recorded with VOG occurred more often in stroke (76.9%) than in vestibular neuritis (37.8%).

Figure 1. Flow Diagram Shows the Process of Screening, Inclusions, Exclusions, and Enrollment

*Patients could have had one or several reasons for exclusion. AVS = acute vestibular syndrome; BPPV = benign paroxysmal positional vertigo; ED = emergency department; EVS = episodic vestibular syndrome.

Figure 2. Box Plots Demonstrate Lower Quartile, Median, and Upper Quartile of Nystagmus Slow Phase Velocity (SPV) for Patients With Vestibular Neuritis and Stroke

(A) Vestibular neuritis. (B) Stroke. Whiskers represent 1.5 times the interquartile range (o = outliers <1.5 box size distance, * = outliers >3 box size distance). Data are grouped by gaze direction. 0° = gaze straight ahead. The direction of the nystagmus beats corresponds to the fast phase direction. Gray boxes indicate measurements with visual fixation, blue boxes without visual fixation. (C) Median reduction of nystagmus velocity (ΔEyeVel), with visual fixation. Green boxes represent patients with vestibular neuritis; red boxes, patients with stroke.

Figure 3. Receiver Operating Characteristic Curve for the Nystagmus Suppression Test in Different Gaze Positions for Differentiating Dizzy Patients With and Without Stroke

ΔEyeVel represents any eye velocity (slow phase velocity) difference between testing in darkness and in light. The diagonal reference line represents a likelihood ratio of 1 at all thresholds, indicating a useless test. The best discrimination cutoff (best sensitivity and specificity) for stroke is indicated for each test condition.