Abstracts

Rationale: Systemic cytokines have been associated with outcome in New Onset Refractory Status Epilepticus (NORSE) and Febrile-Infection Related Epilepsy Syndrome (FIRES), suggesting that active systemic inflammation may contribute to the NORSE/FIRES pathology. Peripheral leukocyte subtypes are major contributors of cytokines, as well as the end-stage effector cells. Therefore, changes in leukocyte subtypes may provide insight into the systemic inflammatory state. We sought to evaluate whether peripheral leukocyte subtypes were associated with outcome in pediatric NORSE/FIRES.

Methods: Children with NORSE/FIRES admitted to the pediatric intensive care unit between 10/2012 and 9/2022 were included. Complete blood count (CBC) studies were included if they were obtained ≤90 d of seizure onset. CBC values were excluded if they were obtained during an active infection, defined as within ± 7 days of a positive infectious study. We collected demographic data, type and dose of immune modulators associated with each CBC, and neurological outcome at 1 year after seizure onset using pediatric cerebral performance category score. We quantified leukocyte subtypes as absolute number / uL. We constructed multivariable logistic regression model accounting for multiple observations by patient to evaluate the association of leukocyte subtypes with favorable outcome, defined as normal or mild disability.

Results: We included 16 patients in the study; 8 had favorable outcome. All received steroid; 15 received IVIG; 14 received anakinra. Patient with favorable outcome were older as compared with those with unfavorable outcome (12 [7-16] years vs. 5.5 [4-9.5] years, favorable vs. unfavorable, p=0.04). The median interval between seizure onset to CBC was 19.8 [IQR 8.2-34] days (n=91) in the favorable outcome group and 27.6 [IQR 14.7-41.7] days (n=147) in the unfavorable outcome group. Higher absolute eosinophil and basophil numbers were associated with favorable outcome in the univariate analyses. We identified the presence of anakinra being associated with increased eosinophil and basophil numbers. Multivariable logistic regression model containing age, eosinophil and basophil numbers, and anakinra demonstrated that age (OR: 1.37, 95% CI: 1.06-1.77, p=0.02) and eosinophil number (OR: 1.005, 95% CI: 1.002-1.008, p=0.001) remained to be independently associated with favorable outcome.

Conclusions: The association of eosinophilia with favorable outcome in pediatric NORSE/FIRES suggests a novel alteration of systemic immunity that has not been previously investigated. Changes in leukocyte subtypes may be a useful adjunct to assess the state of systemic inflammation in NORSE/FIRES.

Funding: None