Seizure recurrences in generalized convulsive status epilepticus under sedation: What are its predictors and its impact on outcome?

Abstract

BackgroundManagement of status epilepticus (SE) is focused on the early seizure termination. Refractory SE is an indication for sedation in patients with SE, but up to 75% of patients may be ventilated due to a neurological or respiratory failure. In patients requiring sedation, the clinical assessment is not sufficient to assess seizure control. Identifying those at risk of recurrent seizures could be useful to adapt their management. On the other hand, patients with low risk could benefit from an early withdrawal of sedation to avoid the impact of inappropriate sedation on outcome. ObjectiveTo determine the prevalence and the predictors of uncontrolled SE and its impact on outcome in patients with generalized convulsive SE (GCSE) requiring mechanical ventilation (MV). MethodsWe retrospectively included patients admitted to the intensive care unit with GCSE requiring MV. Uncontrolled SE was defined as persistent or recurrent seizures during sedation or within 24hours following withdrawal. A multivariable logistic regression model was used to assess the associated factors. ResultsUncontrolled SE occurred in 37 out of 220 patients (17%). Persistent seizures at admission, higher SAPS II and central nervous system infection were associated with a higher risk of uncontrolled SE. Acute toxic or metabolic etiologies were associated with a decreased risk of uncontrolled SE. In a supplementary analysis, decrease of albumin blood levels was associated with uncontrolled SE. Uncontrolled SE was associated with a poor functional outcome and mortality at 90 days. ConclusionsSeventeen percent of patients with a GCSE requiring MV suffered from uncontrolled SE. Etiology and persistent seizures at admission were the main predictors of uncontrolled SE. Patients with uncontrolled SE had a longer duration of sedation and MV, a poor functional outcome and a higher mortality. Further studies are required to determine the impact of continuous electroencephalogram monitoring on the clinical course.