Abstracts

Vigabatrin (VGB), an antiepileptic drug and irreversible inhibitor of GABA-transaminase, has not received FDA approval because of retinal toxicity. The safety of VGB was initially challenged by animal studies showing intramyelinic edema and T2 hyperintensity on MRI. Multiple case reports have shown little compelling evidence of treatment-associated MRI abnormalities. Three adult cases showed increased T2 signal in the corpus callosum during VGB treatment; however it was difficult to prove a strong association because no patient had a series of pre-, intra-, and post-treatment MRI scans., A retrospective chart review of patients who had cranial MRI during VGB therapy was conducted, following detection of new basal ganglia and thalamic hyperintensities in two treated infants. Patients were assessed for age at time of MRI scans, MRI findings from before, during, and after VGB therapy when available, and underlying diagnosis., 15 patients were identified as having MRI scans during the course of VGB therapy (age range 5m-20y; mean 6.1y). Underlying diagnoses were tuberous sclerosis complex (TSC) in 5 (33%), cortical dysplasia in 5 (33%), cryptogenic infantile spasms in 2, and 1 each of: Down syndrome, chromosome 18p-, MTS. Three patients (20%) had new-onset VGB-related T2 hyperintensities within the basal ganglia, thalami, anterior commissure, corpus callosum, or midbrain. These resolved following discontinuation of therapy in 2; 1 patient is undergoing VGB taper. DWI was positive with ADC coefficient maps demonstrating intracellular cytotoxic edema. The basal ganglia were affected in all 3 patients. Each was 12-13 months of age; none of these patients had TSC. Of the remaining 12 patients (age range 5m-20y; mean 7.2y; median 4.6y), none showed evidence for new T2-weighted signal abnormalities during VGB therapy. Ten had abnormalities which were also present prior to initiating VGB therapy. Two additional patients without pre-VGB MRI had abnormalities that were consistent with pre-VGB disease states: TSC and moderate bifrontal hygromas., MRI abnormalities attributable to VGB therapy occur in 20% of patients, and unexplained hyperintense MRI signal may be due to adverse effects of VGB on neuronal metabolism or white matter integrity. Young infants may represent a high risk group. We recommend MRI surveillance of children on VGB including a pre-treatment scan, given the high rate of baseline structural abnormalities and changes attributable to VGB therapy., (Supported by Delman Family Fund for Pediatric Neurology Research.)