Automated alternate cover test for ‘HINTS’ assessment: a validation study

Abstract

Abstract Objective The alternate cover test (ACT) in patients with acute vestibular syndrome is part of the ‘HINTS’ battery test. Although quantitative, the ACT is highly dependent on the examiner’s experience and could theoretically vary greatly between examiners. In this study, we sought to validate an automated video-oculography (VOG) system based on eye tracking and dedicated glasses. Methods We artificially induced a vertical strabismus to simulate a skew deviation on ten healthy subjects, aged from 26 to 66, using different press-on Fresnel prisms on one eye while recording eye position with VOG of the contralateral eye. We then compared the system’s performance to that of a blinded trained orthoptist using conventional, semi-quantitative method of skew measurement known as the alternate prism cover test (APCT) as a gold standard. Results We found a significant correlation between the reference APCT and the Skew VOG (Pearson’s R2 = 0.606, p < 0.05). There was a good agreement between the two tests (intraclass correlation coefficient 0.852, 95 CI 0.728–0.917, p < 0.001). The overall accuracy of the VOG was estimated at 80.53% with an error rate of 19.46%. There was no significant difference in VOG skew estimations compared with the gold standard except for very small skews. Conclusions VOG offers an objective and quantitative skew measurement and proved to be accurate in measuring vertical eye misalignment compared to the ACT with prisms. Precision was moderate, which mandates a sufficient number of tests per subject.