Stroke-heart syndrome: biomarkers and in-hospital outcomes—pilot data from the SHS-PARTICLES study

Background Stroke-Heart Syndrome (SHS) involves acute cardiac injury and hemodynamic stress following a cerebrovascular event. The differential magnitude of myocardial injury between hemorrhagic and non-hemorrhagic stroke and its impact on in-hospital outcomes, remains incompletely defined. Purpose To compare cardiac biomarker activation and short-term outcomes between hemorrhagic (HS) and ischaemic stroke (IS), and to identify independent predictors of length of stay (LOS) and in-hospital mortality. Methods We analyzed 396 consecutive stroke patients in 2025. We measured NT-proBNP and high-sensitivity troponin I (Hs-TnI) at admission and at peak. We assessed left ventricular ejection fraction (LVEF), LOS, and in-hospital death. Linear regression modeled LOS; logistic regression modeled in-hospital death (per 1 unit). Results The study comprised patients diagnosed with IS, n=370 (mean age 73.85±13.27, 53.24% female) and HS, n=26 (mean age 74.15±10.53, 38.46% female). Baseline comorbidities including heart failure (27.84%, n=103 vs 34.62%, n=9; p=0.46), diabetes mellitus (27.03%, n=100 vs 42.31%, n=11; p=0.09), hypertension (87.30%, n=323 vs 92.31%, n=24; p=0.45), hyperlipidemia (61.62%, n=228 vs 53.85%, n=14; p=0.43), chronic coronary syndrome (14.05%, n=52 vs 7.69%, n=2; p=0.36), past myocardial infarction (7.3%, n=27 vs 0.0%, n=0; p=0.15), past stroke (9.19%, n=34 vs 15.38%, n=4; p=0.3) did not differ significantly between groups. HS demonstrated markedly grater cardiac biomarker activation: peak NT-proBNP: 10645.6±12707.3 vs 2539.8±3974.0 pg/mL (p<0.001); peak Hs-TnI: 2124.3±9778.0 vs 346.5±2806.3 ng/L (p=0.02); The proportion of patients with elevated Hs-TnI was higher in HS (69.23% vs 41.08%, p=0.005) as was >10xULN Hs-TnI (26.92% vs 10.54%, p=0.012). HS was associated with longer hospitalization (17.58±9.42 vs 7.81±3.90 days, p<0.001) and higher in-hospital mortality (19.23%, n=5 vs 4.32%, n=16, p=0.001). In linear regression, age (β=0.042, 95%CI 0.003–0.081, p=0.03) and LVEF (β=−0.084, 95%CI −0.149 to −0.019, p=0.011) were independent predictors of LOS, while peak cardiac biomarkers were not. In logistic regression for in-hospital death, age (OR 1.047 per 1-year increase, 95%CI 0.034-1.207; p=0.030), peak NT-proBNP (OR+1000pg/mL 1.117 per 1,000 pg/mL increase, 95%CI 1.034-1.207, p=0.005), peak Hs-TnI (OR+1000pg/mL1.078 per 1,000 pg/mL increase, 95%CI 1.015-1.143, p=0.014) and LVEF+5% (OR 0.737, 95% CI 0.580-0.936, p=0.012) were all predictors. Conclusion(s) HS may be associated with substantially stronger SHS activation and higher in-hospital mortality than IS. While age and LVEF predict the duration of stay, cardiac biomarkers (NT-proBNP and Hs-TnI) serve as predictors of mortality. These findings support systematic cardiac screening across stroke subtypes to identify high-risk neurocardiac phenotypes.