Electroretinogram Indicators of Toxicity in Children Taking the Anti Epiletic Medication Vigabatrin

Abstract

Abstract: : Purpose:The antiepileptic drug vigabatrin results in retinal toxicity in about 30% of adult patient. Adults can be monitored with visual field testing which is not possible in young children. At Sick Kids we use ERGs to monitor children with infantile spasms. ERG changes occur as a result of the drug. It is important to understand which changes occur as a result of non–toxic drug effect and which result from toxicity. ERG change with daily drug dosage is likely reversible, whereas change associated with cumulative dosage are more likely the result of irreversible damage. Our purpose is to determine which ERG types change with daily drug dosage, cumulative dosage and retinal toxicity. Methods: Longitudinal, prospective, observational study including 40 infants on vigabatrin therapy (1–25 months of age, mean 9.8 months). Infants were tested before vigabatrin therapy (n=31) or within one week of first taking vigabatrin (n=9). The ERG was recorded at 6 month intervals, according to ISCEV standards, with the addition of a higher intensity (4 cd.s/m²) flash for the cone response. Photopic oscillatory potentials (OPs) were analyzed according to early (first 2 OPs) vs. late OP. Repeated measures ANOVA identified the effect of daily drug dosage, cumulative dosage and retinal toxicity on ERG result. Results: Evidence of toxicity was observed in 12 patients (30%). All data are described in terms of log difference from normal control data. The following ERG types were linearly associated with change in daily drug dosage: amplitude of early OPs (p<0.0001) and flicker response implicit time (p=0.0053). OP implicit time was also linearly related (p=0.05) but this result has to be treated with caution due to multiple testing. The amplitude of early OPs was quadratically associated with cumulative drug dosage (p=0.0028). Although flicker amplitude varied linearly (p=0.045) and cone b–wave amplitude varied quadratically(p=0.023) with cumulative dosage these results are treated with caution due to multiple testing. In addition to reduction associated with cumulative drug the presence of retinal toxicity was associated with further reduction in cone b–wave amplitude, amplitude of early OPs and flicker amplitude. Conclusions: Reduction in amplitude of early OPs, cone b–wave amplitude, and amplitude of the flicker response should be considered an important indicator of viagabatrin attributed retinal toxicity.

Keywords: electroretinography: clinical • drug toxicity/drug effects • infant vision