Background and objectives: Although etiology is considered central to outcomes in status epilepticus (SE), previous studies often lacked standardized classification and adjustment for confounders, particularly withdrawal of life-sustaining treatment (WLST). This study examined the association between SE etiology, mortality, and neurologic recovery using the International League Against Epilepsy (ILAE) classification while accounting for confounders and WLST.
Methods: This 2-center observational study included adults (≥18 years) with SE treated at the University Hospitals of Basel and Geneva from 2015 to 2023. Etiologies were classified as acute symptomatic, remote symptomatic-unprovoked, progressive CNS disorders, epilepsy without additional triggers, or cryptogenic. Demographics, SE type, SE severity score, Charlson Comorbidity Index, treatment data, complications, and WLST were assessed. The primary outcome was in-hospital mortality; secondary outcomes were 30-day mortality and recovery to premorbid neurologic function at discharge. Associations were assessed using Poisson regression with robust error variance, adjusted for age, nonconvulsive SE (NCSE) with coma, comorbidity, and center.
Results: Among 967 patients (median age 67 years, interquartile range 54-78; 46.5% female), SE was terminated in 95%, with 48.5% of patients recovering to premorbid function. Acute symptomatic SE accounted for 34.2%, remote symptomatic SE for 27.6%, SE due to progressive CNS disorders for 14.4%, epilepsy without additional triggers for 16.7%, and cryptogenic SE for 7.1%. In-hospital and 30-day mortality were 7.9% and 13.9%, respectively, while 48.5% recovered to premorbid function. Etiology was associated with neurologic recovery, with intracranial hemorrhage (relative risk [RR] 0.49, 95% CI 0.35-0.67) and acute symptomatic SE (RR 0.71, 95% CI 0.60-0.83) being associated with reduced likelihood of recovery, whereas known epilepsy was associated with increased likelihood of recovery (RR 1.40, 95% CI 1.23-1.60). NCSE with coma (11.9%) was independently associated with higher in-hospital and 30-day mortality and reduced recovery across all ILAE etiology groups. WLST did not significantly alter these associations.
Discussion: Etiology was associated with neurologic recovery but not with short-term mortality after adjustment for confounders and WLST. By contrast, NCSE with coma showed the strongest association with adverse outcomes. This suggests that while etiology informs prognosis for recovery, SE type, particularly NCSE with coma, is the more critical determinant of survival.