Abstracts

Rationale: The psychomotor arrest and regression associated with infantile spasms renders it one of the main causes of cognitive deterioration during infancy. The long-term implications of this encephalopathic epilepsy are significant, with frequent occurrence of intellectual impairment and persistent seizures in survivors, and increased incidence of early mortality. The aim of this study was to identify patient and treatment factors that were associated with neurodevelopmental and epilepsy outcomes one or more years after onset of infantile spasms. Methods: Using a retrospective chart review, we identified 109 patients who had adequate, long-term follow-up data at our institution. Baseline variables included gender, age at diagnosis, etiology (cryptogenic versus symptomatic), and preexisting medical conditions. The treatment variables we recorded included age at initiation of treatment, treatment lag, pre- and post-treatment EEG patterns, and initial and subsequent treatments used (primarily adrenocorticotropic hormone, vigabatrin, or conventional anticonvulsants). The specific outcomes included seizure control, cognitive status, and motor functioning as documented during the final follow-up visit. Results: The mean period of follow-up for qualifying patients was 5.4 years, with a broad range of 1 to 16.2 years. Twenty-eight patients (26%) had cryptogenic infantile spasms, while 81 patients (74%) fulfilled criteria for a symptomatic cause of their seizures. The most common symptomatic etiologies included congenital brain malformation, perinatal encephalopathy from various causes, and tuberous sclerosis. There were no differences between cryptogenic and symptomatic patients in gender, age at onset of spasms, or age at start of treatment. Cognitive outcome was normal in 21% of cryptogenic versus 8% of symptomatic patients (P-value= 0.047). Similarly, motor outcome was good or mildly impaired in 57% of cryptogenic subjects versus 33% of symptomatic patients (P-value = 0.022). However, intractable seizures were equally frequent in symptomatic and cryptogenic groups. Lag time to initiation of treatment was not predictive of any developmental outcome, nor for need to use a second agent to resolve spasms, even when controlling for etiology. However, patients who responded to the first medication with resolution of spasms and hypsarrythmia had significantly better long term seizure control, and better cognitive outcome. The impact of individual medications on developmental outcome could not be analyzed because almost half of all patients were exposed to multiple agents. Conclusions: While etiology of infantile spasms and response to the first medication predict better outcomes, the majority of patients with infantile spasms continue to have epilepsy and some degree of motor and cognitive impairment long-term. Only through coordination of efforts involving multiple centers could adequate prospective evidence accrue to impact decisions regarding the short and long-term treatment of infantile spasms.