Abstract
OBJECTIVE: To determine the prevalence of retinal defect in children with infantile spasms (IS) unrelated to treatment with vigabatrin and clarify if specific primary etiologies for IS are associated with retinal defect more than others. METHODS: This was an observational cohort study including 312 patients (176 male, 136 female) with IS who were vigabatrin-naive. Participants ranged from 1.7 to 34.7 months of age (mean 8.8 months). Electroretinograms (ERGs) were performed according to the International Society for Clinical Electrophysiology of Vision. Retinal defect was identified as abnormal if the 30-Hz flicker ERG amplitude was lower than the age-corrected normal 95% prediction interval. The primary etiology for IS, as determined by the treating pediatric neurologist(s), was obtained from patient health records and classified into 1 of 9 etiologic subgroups: (1) genetic disorders alone, (2) genetic-structural disorders, (3) structural-congenital, (4) structural-acquired (perinatal), (5) structural-acquired (postnatal), (6) metabolic disorders, (7) immunologic disorders, (8) infectious, and (9) unknown causes. RESULTS: Fifty-nine of the 312 vigabatrin-naive children (18.9%) showed retinal defect and the prevalence of retinal defect was highest (24.4%) in the structural-acquired (perinatal) subgroup, which included hypoxic-ischemic defect. Retinal function compared across subgroups showed no significant difference. CONCLUSIONS: Care is required in diagnosing retinal toxicity, which would be enhanced by baseline flicker ERG in children with IS prior to starting vigabatrin.