Can demographic and admission laboratory variables be useful to identify long-stay patients with acute ischemic stroke? A hospital-based cohort study in Singapore.

The demographic and laboratory predictors of long-stay patients with ischemic stroke were sought in this retrospective hospital-based study. In the univariate and multivariate analysis, advanced age, male gender, leukocytosis, elevated creatinine, low-serum albumin, elevated alkaline transaminases, and lactate dehydrogenase were identified as independent predictors of "long" stayers. At an optimal probability cut-offs, the receiver operating curve incorporating these variables was 0.70, sensitivity 68%, specificity 80%, positive-predictive value 39% and negative-predictive value 95%. Application of this information may assist physicians to triage patients at risk of severe stroke for early therapy and care.