Abstracts

Rationale: Anakinra, an IL-1 modulator, is increasingly utilized as a therapy for febrile-infection related epilepsy syndrome (FIRES). However, the clinical course of children with FIRES on anakinra is not well understood. Here we describe the clinical course of 10 consecutive children with FIRES who were treated with anakinra.

Methods: We collected patient demographics, mechanical ventilation days, ICU and hospital length of stay (LOS), immune modulation, ketogenic diet, and anesthetic infusions. The anti-seizure medication (ASM) regimen and seizure frequency over a 24-h period for the following epochs were collected: 24 hrs prior to, 1 week and 3 weeks following anakinra therapy. We obtained serial serum c-reactive protein (CRP) and cytokine profiles. Pediatric cerebral performance category was used to assess functional outcome at 1 year following discharge, or the latest available information if < 1 year from hospital discharge. We reported patient demographic, clinical and laboratory data using descriptive statistics and expressed as median [interquartile range] or number (%). Pearson chi square test was used for categorical variables, and Wilcoxon rank sum test for continuous variables. We evaluated whether specific serum cytokine levels were associated with therapeutic response to anakinra using logistic regression analyses accounting for multiple observations by patients.

Results: Male children predominated (70%). The median age of presentation was 8.1 [6.6 - 14.9] yrs. The median time from seizure onset to anakinra initiation was 20.5 [10 - 28] d, with the median maximum dosage of 7.5 [4.2 - 7.8] mg/kg/d. Following 1 wk of anakinra therapy, 4 children had decreased seizures; anesthetic infusions were decreased in 5 children. Three children remained on pentobarbital infusions with 3 wks of anakinra therapy and were categorized as non-responders. There were no differences in the ASM regimen between the responders and non-responders. Serial CRP measurements revealed increased prevalence of abnormal CRP in the absence of an infection in 9 of 10 subjects. Serial serum cytokine profiles revealed that, controlling for the time to anakinra initiation, the responders had increased odds for elevated IL-6 levels (OR 4.4, 95% CI: 1.4 - 13.3, p < 0.01); and decreased odds for elevated IL-10 levels (OR 0.03, 95% CI: 0.002 - 0.39, p < 0.01). Compared with the non-responders, the responders had shorter mechanical ventilation days (55 [21 - 82] d vs. 199 [195 - 275] d, p < 0.05), ICU LOS (47 [29 - 64] d vs. 148 [137 - 275] d, p < 0.05), and hospital LOS (82 [42 - 112] d vs. 275 [215 - 283] d, p < 0.05). All children had refractory epilepsy. Of the responders, five children had normal/mild disability; and two had moderate disability. The three non-responders had severe disability/vegetative state at last follow-up.