Risk factors for ASD among children with IESS are poorly characterized.

In this retrospective cohort study we identified subjects with history of IESS who were at least 2 years of age at their most recent follow-up. Data were abstracted from the electronic medical record. Presence or absence of ‘likely ASD’ was determined by impressions of neurologists and developmental pediatricians, explicit formal testing  (ADOS, ADI-R, BOSA), and supplementary testing (VABS, SRS-2).

185 children (27% female) with IESS were identified, with median (IQR) age of onset of 6.0 (4.0 – 9.0) months. Etiology was known among 69% and development was normal at IESS onset among 49%. Response to IESS treatment and subsequent relapse were observed among 69% and 16%, respectively. With median (IQR) follow-up of 69 (40 – 98) months, 49 (26%) children were classified as exhibiting likely ASD. Subjects with likely ASD were more likely to be male (P = 0.09) and less likely to have ever responded to treatment for IESS (P = 0.07). Clinical attributes, including age of IESS onset, development prior to IESS onset, response to IESS treatment, epileptic spasms relapse, and etiology, were not associated with likely ASD. Among the 128 subjects who responded to IESS treatment, likely ASD was neither associated with duration of IESS prior to response, nor any clinical EEG feature at the time of response, including interictal slowing and the abundance and distribution of epileptiform discharges.

Our estimate of ASD prevalence among children with IESS is high and concordant with prior studies. We did not observe a “dose-effect;” likely ASD was not associated with IESS refractoriness, duration from IESS onset to response, epileptic spasms relapse, or diagnosis of LGS.