Prognostic Models of Mortality Following First-Ever Acute Ischemic Stroke: A Population-Based Retrospective Cohort Study

Abstract

Background and Aims

There is a lack of population-based studies focusing on guideline-based prognostic models for stroke. This study aimed to develop and validate a prognostic model that predicts mortality following a first-ever acute ischemic stroke.

Methods

The study included 899 adult patients ( ≥ 18 years) with confirmed diagnosis of first-ever acute ischemic stroke enrolled in the Malaysian National Stroke Registry (NSR) from January 2009 to December 2019. The primary outcome was mortality within 90 days post-stroke (266 events [29.6%]). The prognostic model was developed using logistic regression (75%, n = 674) and internally validated (25%, n = 225). Model performance was assessed using discrimination (area under the curve (AUC]) and calibration (Hosmer-Lemeshow test [HL]).

Results

The final model includes factors associated with increased risk of mortality, such as age (adjusted odds ratio, aOR 1.06 [95% confidence interval, CI 1.03, 1.10; p < 0.001]), National Institutes of Health Stroke Scale (NIHSS) score aOR 1.08 (95% CI 1.08, 1.13; p = 0.004), and diabetes aOR 2.29 (95% CI 1.20, 4.37; p = 0.012). The protective factors were antiplatelet within 48 h. aOR 0.40 (95% CI 0.19, 0.81; p = 0.01), dysphagia screening aOR 0.30 (95% CI 0.15, 0.61; p = 0.001), antiplatelets upon discharge aOR 0.17 (95% CI 0.08, 0.35; p < 0.001), lipid-lowering therapy aOR 0.37 (95% CI 0.17, 0.82; p = 0.01), stroke education aOR 0.02 (95% CI 0.01, 0.05; p < 0.001) and rehabilitation aOR 0.08 (95% CI 0.04, 0.16; p < 0.001). The model demonstrated excellent performance (discrimination [AUC = 0.94] and calibration [HL, X² p = 0.63]).

Conclusion

The study developed a validated prognostic model that excellently predicts mortality after a first-ever acute ischemic stroke with potential clinical utility in acute stroke care decision-making. The predictors could be valuable for creating risk calculators and aiding healthcare providers and patients in making well-informed clinical decisions during the stroke care process.