Predictors of refractoriness in a Turkish status epilepticus data bank

Refractory status epilepticus (RSE) is known to constitute approximately 10-50% of all cases of status epilepticus (SE) and is associated with significant morbidity and mortality. In the present study, data from a prospectively collected SE database were analyzed. Patients with RSE (defined as a SE episode requiring a second line of intravenous treatment following intravenous phenytoin) were compared with patients with nonrefractory SE (NRSE); 290 episodes of SE were identified, of which 108 (38%) were defined as RSE. Univariate analysis revealed that age, female gender, SE type, SE duration, and acute etiology were associated with refractoriness, whereas electroencephalographic patterns were not. Nonconvulsive SE, which is probably associated with delays in treatment initiation, was a predictor of RSE, although it was not retained as a predictor in multivariate analysis. In the latter analysis, female gender (odds ratio: 1.815, 95% CI: 1.053-3.126) and acute etiology (odds ratio: 0.619, 95% CI: 0.429-0.894) were shown to be the only significant independent predictors of refractoriness.