Stroke mortality and its determinants in a resource-limited setting: A prospective cohort study in Yaounde, Cameroon

Background About three quarters of stroke deaths occur in developing countries including those in sub-Saharan African. Short and long-term stroke fatality data are needed for health service and policy formulation. Methods We prospectively followed up from stroke onset, 254 patients recruited from the largest reference hospitals in Yaounde (Cameroon). Mortality and determinants were investigated using the accelerated failure time regression analysis. Results Stroke mortality rates at one-, six- and 12 months were respectively 23.2% (Ischemic strokes: 20.4%, hemorrhagic strokes: 26.1%, and undetermined strokes: 34.8, p = 0.219), 31.5% (ischemic strokes: 31.5%, hemorrhagic strokes: 30.4%, and undetermined strokes: 34.8%, p = 0.927), and 32.7% (ischemic strokes: 32.1%, hemorrhagic strokes: 30.4%, undetermined strokes: 43.5%, p = 0.496). Fever, swallowing difficulties, and admission NIHSS independently predicted mortality at one month, six and 12 months. Elevated systolic blood pressure (BP) predicted mortality at one month. Elevated diastolic blood pressure was a predictor of mortality at one month in participants with hemorrhagic stroke. Low hemoglobin level on admission only predicted long term mortality. Conclusion In this resource-limited setting, post-stroke mortality was high with 1 out of 5 deaths occurring at one month and up to 30% deaths at six and twelve months after the index event. Fever, stroke severity, elevated BP and anemia increased the risk of death. Our findings add to the body of evidence for the poor outcome after stroke in resource limited environments.