Purpose: The neurodevelopmental outcome of infants with early-onset epilepsies varies widely, ranging from typical development to global developmental delay. Visual skills, which emerge during infancy, are crucial for the development of cognitive functions. The aim of this observational cohort study was to explore gaze behavior in infants with early-onset epilepsy and evaluate if eye tracking could support prognostication of their neurodevelopment.
Methods: Fifty-one infants (22 females, mean seizure onset-age 5, SD ± 2, months) from a prospective epilepsy cohort underwent repeated eye tracking and Hammersmith Infantile/Neonatal Neurological examination (HINE/HNNE). Neurodevelopment at age two was categorized as typical development (mean Bayley [BSID-III] cognitive and language or Griffiths [GMDS-III] scales score ≥ 85) and developmental delay. At initial (age 3-10 months) and 12-month visit, we compared reliability of fixation, probability of gaze shifts and saccadic reaction times (SRTs) in a non-competitive SRT-task between developmental groups. Gaze behavior was also compared across etiologies, syndrome groups and between those with optimal versus suboptimal first HINE/HNNE.
Results: Infants with typical developmental outcome (n = 23) had higher reliability of fixation (p = 0.007) and higher probability of gaze shifts (p = 0.012) at initial eye tracking than those with delay (n = 28). SRTs became faster during the follow-up but did not differ significantly between the developmental groups. Gaze behavior associated with epilepsy syndrome, etiology, and initial HINE/HNNE result.
Conclusions: Ability to fixate reliably and shift gaze soon after the epilepsy diagnosis is associated with developmental outcome in infants with early-onset epilepsy, suggesting that eye tracking could be useful as an additional prognostic tool.
Keywords: Early-onset epilepsy; Eye tracking; Gaze behavior; Neurodevelopment; Structured neurological examination.
Copyright © 2025 The Author(s). Published by Elsevier Inc. All rights reserved.