Determinants for bedside lateralization of benign paroxysmal positional vertigo involving the horizontal semicircular canal.

Abstract

This study aimed to define the factors affecting accuracy of bedside determination of the involved side in HC-BPPV. We developed 44 video clips including the nystagmus induced during supine head-roll test from patients with apogeotropic (n = 23) or geotropic (n = 21) HC-BPPV. The intensity of nystagmus was also quantified using video-oculography. Each video clip was presented twice to 25 participants, 14 medical students, and 11 medical personnel trained in neurology or neurotology, and the participants reported the lesion side using an evaluation sheet after each presentation. From the quantified video-oculographic data, absolute difference and asymmetry (absolute difference divided by the sum) of nystagmus intensity were calculated. The accuracy of bedside lateralization of HC-BPPV was 83.5% after the first presentation, and 86.0% after the second presentation, and was not different between the medical students and trained personnel after the second presentation. The accuracy was more closely correlated with the asymmetry than the absolute difference (Spearman's ρ = 0.627, p < 0.001). With a cut-off for the asymmetry at 30.8%, the accuracy was estimated at 93.8% with a sensitivity of 92.9%, and the area under the ROC curve of 0.779. The accuracy of bedside lateralization of the affected side is acceptable in HC-BPPV when the nystagmus asymmetry is more than 30%.