Abstracts

Rationale: Infantile Spasms (IS) occur in approximately 40% of patients with Tuberous Sclerosis Complex (TSC), and are associated with poor neurocognitive outcomes. Vigabatrin (VGB) is a particularly effective treatment for IS in TSC, with response above 90%. Improved long-term cognition with early treatment has been reported in small retrospective series, and delayed treatment has been associated with worse neurological outcome. A small open label trial even reported possible neurocognitive benefits from pre-symptomatic treatment, in patients demonstrating EEG abnormalities but not spasms yet. . Given the clear preference for VGB as a first line agent, and the apparent benefits on neurological outcome of early therapeutic intervention, we studied the effect of time from IS onset to VGB therapy on neurological outcome (resolution of spasms, later refractory epilepsy, autism, and cognition).Methods: A retrospective chart review of all patients with definite Tuberous Sclerosis Complex (TSC) followed at Boston Children's Hospital Multidisciplinary TSC Clinic was performed [n=182][Figure 2]. Patients with a history of infantile spasms and sufficient follow up data were identified [n=59, 32.4%]. Demographic and genetic information was collected. Charts were reviewed for time of IS onset, time of VGB therapy initiation, use of other medications prior to VGB, and resolution and recurrence of IS. Long term outcome variables included presence of autism spectrum disorder, and cognitive outcome (categorized as normal or near-normal for age, mild to moderate delays, and severe delays).Results: Kaplan-Meier time to VGB survival curves of normal cognition, mild to moderate delays, and severe delays are displayed in Figure 2. The differences were significant (p <0.5, log-rank test). Time to VGB was found to be significant for different cognition groups (Log-Rank p=0.036). Hazard ratios (HR) for the subjects with mild to moderate delay in comparison to the normal cognition was 0.42 (0.18-0.99). Similarly, HR for the subjects with severe delay in comparison to the normal cognition was 0.29 (0.11-0.80). Time to VGB was highly significant for different number of pre-drug groups (Log-Rank p=0.023). Hazard ratios (HR) for the subjects with 1 in comparison to the 0 pre-drugs was 0.43 (0.22-0.83) whereas for the subjects with >1 vs 0 pre-drugs was 0.16 (0.03-0.55). The time to VGB was not related to presence of autism. Similarly, there was no relationship between time to VGB and time to IS resolution.Conclusions: The time of initiation of VGB therapy for IS in TSC is a determinant of neurological outcome. In this retrospective series, the earlier VGB was started, the more likely a good cognitive outcome was achieved. Similarly, delays in the time of VGB initiation caused by trials of other medications, were associated with a more poor outcome.