Predictors of mortality at one year after generalized convulsive status epilepticus

Status epilepticus (SE) is a life-threatening neurologic emergency, which requires prompt medical treatment. Little is known of the long-term survival of SE. The aim of this study was to investigate which factors influence 90 days and 1-year mortality after SE.

This retrospective study includes all consecutive adult (>16 years) patients (N = 70) diagnosed with generalized convulsive SE (GCSE) in Helsinki University Central Hospital (HUCH) emergency department (ED) over 2 years. We defined specific factors including patient demographics, GCSE characteristics, treatment, complications, delays in treatment, and outcome at hospital discharge and determined their relation to 90 days and 1-year mortality after GCSE by using logistic regression models. Survival analyses at 1 year after GCSE were performed with Cox proportional hazards regression analysis.

In-hospital mortality was 7.1%. Mortality rate was 14.3% at 90 days and 24.3% at 1 year after GCSE. In the univariate logistic regression analysis, Status Epilepticus Severity Score > 4 (STESS) (ODDS = 7.30, p = 0.012), worse-than-baseline condition at hospital discharge (ODDS = 3.5, p = 0.006), long delays in attaining seizure freedom (ODDS = 2.2, p = 0.041), and consciousness (ODDS = 3.4, p = 0.014) were risk factors for mortality at 90 days whereas epilepsy (ODDS = 0.2, p = 0.014) and Glasgow Outcome Scale (GOS) >3 at hospital discharge (ODDS = 0.05, p = 0.006) were protective factors. Risk factors for mortality at 1 year were STESS >4 (ODDS = 5.1, p = 0.028), use of vasopressors (ODDS = 8.2, p = 0.049), and worse-than-baseline condition at discharge (ODDS = 7.8, p = 0.010) while GOS >3 (ODDS = 0.2, p = 0.005) was protective. The univariate survival analysis at 1 year confirmed the significant findings regarding parameters STESS >4 (Hazard ratio (HR) = 4.1, p = 0.009), worse-than-baseline condition (HR = 6.2, p = 0.015), GOS >3 (HR = 0.2, p = 0.004) at hospital discharge and epilepsy (HR = 0.4, p = 0.044). Additionally, diagnostic delay over 6 h (HR = 3.8, p = 0.022) and Complication Burden Index (CBI) as an ordinal variable (0-2, 3-6, >6) (HR = 2.7, p = 0.027) were predictive for mortality. In the multivariate survival analysis, STESS > 4 (HR = 5.1, p = 0.007), CBI (HR = 3.2, p = 0.025, ordinal variable), diagnostic delay over 6 h (HR = 7.2, p = 0.003), and worse-than-baseline condition at hospital discharge (HR = 5.8, p = 0.027) were all independent risk factors for mortality at 1 year.

Severe form of SE, delayed recognition of GCSE, high number of complications during treatment period, and poor condition at hospital discharge are all independent predictors of long-term mortality. Most of these factors are also associated with mortality at 90 days, though at that point, delays in treatment seem to have a greater impact on prognosis than at 1 year. This article is part of the Special Issue "Proceedings of the 7th London-Innsbruck Colloquium on Status Epilepticus and Acute Seizures.