A retrospective study on aetiology based outcome of infantile spasms

PURPOSE The goal of this retrospective study is to review the causes of infantile spasms and to correlate aetiology with outcome. METHODS All children diagnosed with infantile spasms between 1990 and 2003 at our institution were included. Charts were reviewed for the presence or absence of a defined aetiology/association, response to treatment, long-term epileptic and cognitive outcome. RESULTS 80 out of 95 children are included in this review. 50 children (63%) had symptomatic spasms with disorders of cortical development being the most frequent cause followed by neonatal injury and tuberous sclerosis. Symptomatic children with developmental brain lesions responded at a rate of 54% to vigabatrin versus 62% for ACTH/prednisone, while other symptomatic aetiologies 83% responded to vigabatrin versus 63% for ACTH/prednisone. Cryptogenic spasms responded at a similar rate to both drugs. Other than children with cryptogenic spasms, very few went on to develop normally. Our results are however biased by on average more than 30 days of delay to diagnosis. None of our children developed Lennox-Gastaut syndrome but a number developed severe epilepsy with multifocal spikes. DISCUSSION The aetiology and prognosis of infantile spasms is evolving. To improve outcome, we need to reduce the delay to diagnosis and develop prospective double-blind randomized clinical trials looking at not only the epileptic outcome but also cognitive outcome of these children.