Construction and Validation of a Prediction Model for Killip Classes II-IV During Hospitalisation in Patients With...

To perform a comprehensive assessment of the predictive value of soluble growth stimulator gene 2 protein (sST2) in predicting in-hospital Killip classes II-IV among patients with acute ST-segment elevation myocardial infarction (STEMI). This study aimed to provide more precise prognostic insights for informed clinical decision-making.

A retrospective cohort study was performed. The clinical records of STEMI patients admitted to Tianjin TEDA International Cardiovascular Hospital and who received primary percutaneous coronary intervention (PPCI) within 24 hours of symptom onset from July 2021 to March 2023 were analyzed. Statistical methodologies, including univariate and multivariate analyses, were applied to identify potential risk factors associated with the development of in-hospital Killip classes II-IV and to construct a reliable prediction model.

Among a total of 232 enrolled STEMI patients, 50 experienced Killip classes II-IV during their hospitalisation. Compared to those with Killip class I, the Killip class II-IV patients presented with significantly elevated sST2 concentrations and a higher heart rate (HR) at the first visit. In contrast, the left ventricular ejection fraction (LVEF) and estimated glomerular filtration rate (eGFR) values in these patients were significantly lower. Multivariate logistic regression analysis revealed that an sST2 level >77.3 ng/mL (odds ratio (OR) = 2.813, 95% confidence interval (CI): 1.201-6.586, p = 0.017), a first-visit HR >94 bpm (OR = 7.286, 95% CI: 2.778-19.106, p < 0.001), an LVEF <50% (OR = 3.336, 95% CI: 1.458-7.631, p = 0.004), and an eGFR <84 mL/(min·1.73 m²) (OR = 3.807, 95% CI: 1.556-9.316, p = 0.003) were independent risk factors for the occurrence of in-hospital Killip classes II-IV in STEMI patients treated with PPCI. Receiver operating characteristic (ROC) curve analysis, along with decision curve analysis (DCA), indicated that the combined predictive model integrating sST2, first-visit HR, LVEF, and eGFR exhibited a significantly stronger predictive ability compared to any single parameter.

In STEMI patients undergoing PPCI, the combination of sST2, first-visit HR, LVEF, and eGFR can effectively predict patients with Killip classes II-IV during hospitalisation, which may contribute to early intervention and improved patient outcomes.