Abstracts

Rationale: The etiology of IS is unknown in one third of patients despite investigation for structural, metabolic and chromosomal or other known genetic etiologies. Patients with IS of unknown cause (UCIS) are typically thought to have a better prognosis than those with an identified cause, but limited clinical data exists to support this. We report the clinical features and predictors of developmental outcome of the largest known cohort of UCIS patients.Methods: Participants were enrolled in the Epilepsy Phenome/Genome Project (EPGP) Infantile Spasms Project, a multi-center NINDS-sponsored study. Patients with a history of UCIS and EEG showing hypsarrhythmia or a hypsarrhythmia variant were included. Initial exclusion criteria included developmental delay of greater than 50% prior to onset of IS, prematurity (<32 weeks gestation) and brain malformation other than focal cortical dysplasia on imaging. For this study, medical records were reviewed by two independent reviewers. EEG and MRI brain imaging were reviewed. Patients were then excluded for this analysis if an etiology (including focal cortical dysplasia) was identified.Results: Of 189 UCIS patients, there was no significant difference in gender (M:F; 0.86:1). IS onset occurred at a mean age of 5.8 months. Treatment was initiated within one month in 57.5% patients and corticosteroids were first line therapy in 60.1% patients. 30.3% had seizure freedom for at least six months, whereas 55.8% of patients had seizures subsequent to their spasms. Developmental delay occurred prior to IS in just 17.0%, however at last assessment, only 21.3% had normal development. Normal development was seen in 31.3% (OR 2.31) of patients treated within one month and 26.3% (OR 2.11) of patients who were treated with corticosteroids first. 54.5% of patients with resolution of their spasms went on to develop normally while 79.5% (OR 0.06) of those with seizures subsequent to IS had abnormal developmental outcomes. Conclusions: This large cohort of UCIS patients shows significant heterogeneity of developmental and seizure outcomes, but highlights a predominantly unfavorable developmental outcome. Factors previously identified to predict a good prognosis including prompt initiation of treatment and first line treatment with steroids did show a trend toward better outcomes but were not statistically significant. The variable clinical outcomes suggest that the underlying causes are likely heterogenous. The overall poor developmental outcomes within this group outline the importance of further understanding the underlying etiology. Under the auspices of the Epi4K Project, the exomes of these patients are being sequenced. Phenome-genome correlations will provide valuable prognostic, diagnostic, and therapeutic options in the future; ultimately leading to better outcomes.