Front Neurol. 2026 May 20;17:1789811. doi: 10.3389/fneur.2026.1789811. eCollection 2026.
ABSTRACT
OBJECTIVE: To evaluate oculomotor, vestibular, reaction time, and cognitive (OVRT-C) function in patients with vestibular migraine (VM) using objective eye-tracking-based metrics and to identify patterns of dysfunction relative to healthy controls.
BACKGROUND: Vestibular migraine is a common yet underdiagnosed cause of vertigo. Diagnosis remains primarily clinical, because objective neurologic and vestibular findings are often absent or inconsistent. Quantitative methods capable of capturing the multisystem manifestations of VM may improve objective characterization of the disorder.
METHODS: Participants with a clinical diagnosis of vestibular migraine were recruited from a tertiary headache center and assessed using a battery of OVRT-C tests administered with a portable eye-tracking system (Spryson Dx-100; n = 52). The test battery assessed gaze stability, saccades, antisaccades, smooth pursuit, vergence, optokinetic responses, and visual and auditory reaction times. Participants also underwent bedside neurologic and vestibular examinations and completed the Dizziness Handicap Inventory (DHI). OVRT-C metrics were compared with normative data from a database of 300 healthy adults. Univariate and stepwise multivariate logistic regression models were used to identify metrics that differentiated VM patients from controls.
RESULTS: A substantial proportion of VM patients aged 18 to 45 demonstrated abnormal OVRT-C performance compared with normative data, most prominently in horizontal and vertical saccades (54.3 and 51.4%, p < 0.0001), vertical smooth pursuit (62.9%, p < 0.0001), optokinetic responses (43.8%, p < 0.0001), and gaze stability (65.7%). Several OVRT-C metrics showed strong discriminative ability in logistic regression analyses when considering a single OVRT-C metric is considered (adjusted for participant age and gender). A multiple logistic regression model identified six OVRT-C metrics as significant indicators of VM and demonstrated excellent classification performance (AUC = 0.996), with estimated specificity of 95.9% and sensitivity of 99.7% for probability cutoff of 0.5. Moderate but statistically significant correlations were observed between OVRT-C metrics, bedside neurologic and vestibular findings, and DHI domain scores.
CONCLUSION: Objective OVRT-C testing reveals quantifiable abnormalities in oculomotor, vestibular, and cognitive dysfunction in patients with vestibular migraine. These findings support the feasibility of eye-tracking-based multimodal assessments as complementary tools for characterizing vestibular migraine and warrant further validation in larger and longitudinal cohorts.
PMID:42246042 | PMC:PMC13229772 | DOI:10.3389/fneur.2026.1789811
