Clinical profile and comparison of scoring tools for the prediction of in-hospital mortality in convulsive status epilepticus in elderly.

ABSTRACT

PURPOSE: The present study was aimed to study the clinical profile, etiologies, and outcome of convulsive status epilepticus (CSE) in elderly patients and also to compare the predictive accuracy of the Status Epilepticus Severity Score (STESS) and Epidemiology-based Mortality Score in Status Epilepticus - etiology, age, level of consciousness (EMSE-EAL) score for in-hospital mortality. METHODS: Eighty-five elderly patients (≥60 years of age) with a diagnosis of CSE were consecutively enrolled. The distinction between the score performances was determined by comparing the means area under the receiver operating characteristic curve (AUC). RESULTS: The mean age of respondents was 66.3 ±â€¯7.4 years; the most common etiology of CSE was stroke (acute and remote symptomatic) in 48.2% of cases. In-hospital mortality was 16.5% in our series, and on multivariate analysis, variables significantly related with mortality were lack of response to first-line drugs (odds ratio (OR) = 43.05, 95% confidence interval (CI) = 4.7-386.8; p = .001) and higher EMSE-EAL score (OR = 0.08, 95% CI = 0.015-0.47; p = .005). On comparison, STESS with the cutoff value of ≥3 has AUC of 0.678 (95%CI = 0.54-0.81), whereas ESME-EAL with the cutoff value of ≥40 showed AUC of 0.901 (95% CI = 0.83-0.97). CONCLUSIONS: Most frequent cause of CSE in elderly in our series was stroke and was also associated with high mortality. For the prediction of in-hospital mortality in elderly, EMSE-EAL-40 score is superior to STESS-3, which can be easily applied in resource-poor sectors with limited diagnostic facilities especially where continuous video-electroencephalogram (EEG) monitoring is unavailable.