Abstract TP299: Early Mortality Prediction in Ischemic Stroke

    loading  Checking for direct PDF access through Ovid


Introduction: Ischemic stroke (IS) is a medical emergency with a high risk of morbidity, leading to disability, and mortality. IS’s impact on our patients’ lives and healthcare system make it essential to optimize its immediate management, triage requirements, and plans for transfer to comprehensive stroke centers. In this study, we hypothesized that it is possible to predict the mortality of IS patients within the first hour of hospital admission.Methods: With the approval of the Human Studies Committee (IRB #: 13.0396), we performed a retrospective analysis from the Stroke Quality Database. Patients who were admitted between 2007-2012 were included. In-hospital mortality was the main outcome of the study. Patients’ demographics, baseline illnesses, home medications, baseline neurological assessment scores, and basic laboratory values on admission were included in the analysis as potential factors. Multivariate analysis was performed to assess the independent contributing factors. Data is presented as odds ratio and 95% confidence intervals.Results: IS in-hospital mortality was about 10% in our study population. Multivariate analysis was performed in 2,131 patients. NIH Stroke Scale on admission, recurrent stroke, high WBC levels (>11K), elderly age, AHA Stage II hypertension, baseline use of Ca-channel blockers, and beta-blockers, admission glucose levels and albumin levels independently contributed to in-hospital mortality (Table).Conclusions: IS patients’ baseline characteristics, hospital admission assessment scales, and laboratory values on admission contributed to prediction of in-hospital mortality. Therefore, it is essential to rapidly initiate acute care for high-risk and complex stroke patients. Additionally, focusing on the modifiable factors such as blood pressure management, treating acute infections, and managing glucose levels may help to decrease acute IS mortality.

    loading  Loading Related Articles