Abstracts

Rationale: Infantile spasms (IS) is the most common infantile epilepsy and is associated with medically refractory seizures and poor neurodevelopmental outcomes. Use of standard medications for IS and lack of etiology are positive predictors of treatment response. Here, we compared clinical characteristics and treatment responses for IS between those with versus without seizures prior to IS onset. By understanding the neonatal and infantile characteristics of these patients, we aim to identify the impact of these antecedent neonatal and infantile seizures on IS characteristics. Methods: A retrospective review was conducted of patients with IS at Boston Children’s Hospital from 2012 to 2017. Time of IS onset, time to IS treatment, seizure history, treatment type and response, and etiology were collected. Standard treatment consisted of adrenocorticotropin, prednisolone, or vigabatrin with non-standard treatment being any other agent. Etiology was categorized as structural acquired (e.g. hypoxic ischemic encephalopathy, intraventricular hemorrhage), structural congenital (e.g., cortical dysplasia, lissencephaly), genetic, and unknown. We compared these variables between patients with or without prior seizures using chi-square and z-tests. Results: In 68 infants with IS, 57% were male. Sixteen patients (24%) who had seizures prior to onset belonged to one group while 52 patients without history of seizures formed the other group. Median age of IS onset did not differ between groups (5.8 months vs 6 months). The most common etiology was structural congenital or genetic in the group without prior seizures (37%, 19/52 for both etiologies) and structural acquired  (44%, 7/16) in the group with prior seizures. Standard therapy was used initially in 82% (56/68) while 18% (12/68) received non-standard therapy. The proportion of patients receiving standard therapy was similar between groups (83% (43/52) vs 81% (13/16)). 53% (36/68) of all patients received treatment within 2 weeks of IS onset. The response to standard therapy initially did not differ between groups (46% (6/12) with prior seizures vs. 23%(10/52) without, p=0.1). Notably, there was a higher rate of persistent spasms in the group without prior seizures (79% (41/52) vs 50% (8/16), p=0.02). Standard therapy was more likely to be used to treat persistent IS in the group without prior seizures versus with (78% (32/41) vs 38% (3/8), p=0.02). Despite this, there was no difference in responder rates between groups for standard (28% (9/32) vs 0% (0/3), p=0.28) or any treatment (22% (9/41) vs 0% (0/8), p=0.14) when given for persistent IS. Conclusions: The higher proportion of known etiology in the group with seizures prior to IS onset is not surprising. The similar response to initial therapy may be due to similar lead-time to treatment and use of standard therapy in both groups. A novel finding in our study was a higher rate of persistent spasms in infants without antecedent seizures, which warrants further analysis of specific etiology and treatment choice. Funding: None