Stratification of early mortality risk in patients with stroke-heart syndrome: insights from a cluster analysis of pooled individual patient data analysis in the VISTA database

Background Stroke-heart syndrome (SHS), characterised by cardiac complications during the acute phase of acute ischaemic stroke, is associated with poor prognosis. We employed hierarchical cluster analysis to identify phenotypic profiles associated with early mortality in patients with SHS. Methods We analysed data of patients with acute ischaemic stroke in the Virtual International Stroke Trials Archive (VISTA). We included patients with SHS who developed cardiac complications within 30 days of stroke onset, such as acute myocardial infarction, heart failure, arrhythmias, and cardiorespiratory arrest. Using hierarchical cluster analysis, we identified distinct phenotypic risk profiles and compared 90-day mortality incidence and risk across these profiles. Results We included 1,787 patients (mean age 73±11 years; 52% male), and identified there phenotypic profiles: Profile 1 ("atherosclerotic risk factors"), Profile 2 ("cardiac co-morbidities"), and Profile 3 ("elderly and female"). Within 90 days post-stroke, 620 patients (35%) died. Profile 3 has the highest risk of 90-day mortality (hazard ratio [95% confidence interval]: 1.44 [1.15-1.81] vs. Profile 1), followed by Profile 2 (1.19 [0.95-1.48]). The incidence of cardiorespiratory arrest increased progressively across profiles, with Profile 3 showing a significantly higher risk than Profile 1 (2.18 [1.24-3.82], p=0.006). Conclusions Hierarchical cluster analysis identified distinct SHS phenotypic profiles associated with early mortality and cardiorespiratory arrest risk.