Validation of the Status epilepticus severity score (STESS) at high-complexity hospitals in Medellín, Colombia.

Purpose: The Status Epilepticus Severity Score (STESS) is one of the most well-known clinical scoring systems to predict mortality in status epilepticus (SE). The objective of this study was to validate STESS in a Colombian population. Method: We evaluated historical data of adult patients (age =16 years) with a clinical or electroencephalographic diagnosis of SE admitted between 2014 and 2017. Prospectively, we included patients admitted from January to June of 2018. The primary outcome was in-hospital mortality. Receiver operating characteristic (ROC)-analysis, determination of best cutoff values, sensitivity, specificity, and positive and negative likelihood ratios were performed. Results: The sample was 395 patients, with in-hospital mortality of 16.8 %. The area under the ROC curve for STESS was 0.84. A cutoff point of >= 3 produced the highest sensitivity of 84.9 % (95 % CI 73.9 %-92.5 %) and a specificity of 65.7 % (95 % CI 60.2 %-70.8 %), with a positive likelihood ratio of 2.5 and a negative likelihood ratio of 0.2. Conclusions: STESS is a useful tool to predict mortality in patients with SE. In Medellin, Colombia, a STESS < 3 allows the identification of the patients who survive reliably. Those patients with a score <3 may have a better prognosis, and treatment with fewer side effects than anaesthetics could be suggested, always remembering the importance of the treating physician's clinical judgement.