Can the adverse effects of antiepileptic drugs be detected in saccadic eye movements

Abstract Purpose The objective of this study was to determine whether the adverse effects of antiepileptic-drugs could be assessed by the eye movements of epilepsy patients. Methods This study was performed prospectively in a single tertiary hospital. The inclusion criteria for this study were as follows: (1) consecutive patients with epilepsy taking antiepileptic-drugs regularly for at least 1 year, (2) the absence of structural lesions on MRI, (3) an age ≥16 years old, (4) not using medications that could influence eye movement, and (5) a normal neurological examination. The latency, peak velocity and accuracy of the saccades and the gain of the pursuits were recorded by video-based electro-oculography. We analyzed the differences in the parameters of the eye movements for 75 patients with epilepsy and 20 normal controls matched for age and sex. Results The total latency (1017.7±148.9ms vs. 1150.7±106.6ms, p =0.0003) and accuracy [370.7% (95% CI 364.1–376.4%, range 306–408.2%), 92.7% as total accuracy normalized value vs. 383.6% (95% CI 378.8–398%, range 322.9–417.4%), 95.9% as total accuracy normalized value, p =0.0005] were significantly different between the patients with epilepsy and normal controls. For the detection of nystagmus with video-based electro-oculography, the clear cutoff values of total accuracy (≤388.7%, 97.2% as total accuracy normalized value) revealed 93.4% sensitivity and 28.6% specificity, and the clear cutoff values of total latency (≤1005.5ms) showed 49.2% sensitivity and 78.6% specificity. Conclusions The total latency and accuracy of video-based electro-oculography may be screened to identify patients with a high risk of adverse effects with antiepileptic-drugs.