F23. Predicting mortality and functional outcome at 3–6 months after status epilepticus requiring critical care

Abstract

Introduction Status epilepticus (SE) often requires intensive care and is associated with significant mortality and morbidity. Mortality in the general critical care population may be predicted by the Acute Physiology and Chronic Health Evaluation (APACHE) score; in patients with SE, the Status Epilepticus Severity Score (STESS) has been developed to predict death or return to baseline at hospital discharge. It is unclear how accurate APACHE models mortality after SE. Functional outcome at 3–6 months has been understudied. We sought to determine if one of these tools is superior for predicting mortality and functional outcome for critically ill patients that present with SE. Methods This retrospective observational cohort study included all patients admitted to the Neuroscience Intensive Care Unit (ICU) from 3/2016 to 3/2017 with a primary diagnosis of SE. We excluded patients based upon the following criteria: secondary seizures related to co-occurring brain injury (such as hypoxic-ischemic injury), no definite witnessed clinical or electrographic seizure, ICU length of stay less than 24 h, or care directed by non-ICU clinical team. STESS and APACHEII scores, age, gender and modified Rankin Scale (mRS) were retrieved or calculated from electronic medical record. All patients received telephone follow-up between 3 and 6 months from admission; our primary outcome was 3-month mRS. Poor outcome was defined as mRS > 2. Neurodeterioration was defined as change from baseline mRS > 1 point. We constructed generalized linear models using stepwise regression with pre-defined significant level of p Results We identified n = 82 patients with a primary diagnosis of SE admitted to the ICU for >24 h. The median baseline mRS prior to admission was 3 (interquartile range 1–3). The median STESS score was 3 (IQR 3–5) and the mean APACHEII score was 20.1 ± 5.6. At follow-up, 16/82 (19.5%) had died, and 20/48 (41.7%) contacted survivors had experienced neurodeterioration. Follow-up mRS was independently associated with STESS (OR 1.8 95 CI 1.0–3.3; p = 0.04) plus baseline mRS (OR 2.7 95 CI 1.4–5.2; p = Conclusion We found that a combination of STESS score and baseline level of functioning achieved the best area-under-the-curve for absolute functional outcome at 3–6 months following SE. APACHEII did not predict mortality or functional outcome in this patient population, suggesting that disease-specific factors play a key role in mediating outcome.