Key result
Novel five-factor model predicts acute heart failure after STEMI PCI with ~0.88 AUC.
Why the study?
To establish and evaluate the clinical performance of a predictive model for acute heart failure occurrence in STEMI patients following PCI.
Can a predictive model using clinical and laboratory parameters accurately predict acute heart failure after PCI in STEMI patients?
Cohort (n=419)
No
Can a predictive model using clinical and laboratory parameters accurately predict acute heart failure after PCI in STEMI patients?
Effect estimate: AUC 0.883 (95% CI 0.845-0.920)
p-value: p=<0.0001
A novel predictive model incorporating systolic blood pressure, neutrophils, total bilirubin, urea nitrogen, and LVEF accurately predicts acute heart failure following PCI in STEMI patients.
May aid AHF risk stratification post-STEMI PCI; leaves open prospective validation before clinical adoption.
Objective To establish a predictive model for acute heart failure (AHF) occurrence in patients with ST-segment elevation myocardial infarction (STEMI) following percutaneous coronary intervention (PCI) and evaluate its clinical performance. Methods A retrospective analysis was conducted on 419 STEMI patients treated at the Cardiology Department of Maanshan People's Hospital from January 2018 to December 2024. Patients were divided into AHF group ( n = 100) and non-AHF group ( n = 319) based on AHF occurrence. Logistic regression analysis identified independent risk factors for model construction. Model performance was assessed using receiver operating characteristic (ROC) curves with optimal threshold determination via maximum Youden index. Statistical validation included Omnibus and Hosmer-Lemeshow tests. Results The AHF prediction model was: Logit(P) = 3.084 − 0.026 × systolic blood pressure + 0 .083 × neutrophil count − 0.041 × total bilirubin + 0 .238 × urea nitrogen − 0.045 × left ventricular ejection fraction (LVEF). AHF was predicted when logit(P) > 0.231. Statistical validation showed Omnibus test χ 2 = 7.112, P = 0.008, and Hosmer-Lemeshow test χ 2 = 6.551, P = 0.586, indicating good model fit. The model achieved 74% sensitivity, 86.8% specificity, and 82.4% diagnostic accuracy. Comparative ROC analysis demonstrated superior performance vs. established scores: predictive model + Grace Score 0.902 vs. baseline model 0.715 (Delong test: Z = 0.235, P < 0.0001), with continuous net reclassification index (NRI) 0.7472 ( P < 0.0001) and integrated discrimination index (IDI) 0.2455 ( P < 0.0001); predictive model + CAMI -STEMI Score 0.909 vs. baseline model 0.717 (Delong test: Z = 0.4930, P < 0.0001), with continuous NRI 0.7245 ( P < 0.0001) and IDI 0.2101 ( P < 0.0001). Clinical decision curve analysis demonstrated net clinical benefit when threshold probability ranged from 0.1 to 0.99. Conclusion Systolic blood pressure, total bilirubin, LVEF, neutrophils, and urea nitrogen are independent factors affecting AHF occurrence after PCI in STEMI patients. This exploratory predictive model demonstrates potential for assessing AHF risk in STEMI patients following PCI and may provide valuable clinical decision support for risk stratification and patient management. However, external validation in independent, multicenter populations is essential before clinical implementation.
No takes yet. Share an insight, caveat, or question.
Li et al. (2026) conducted a cohort in ST-segment elevation myocardial infarction (STEMI) post-PCI (n=419). Predictive model (systolic blood pressure, neutrophils, total bilirubin, urea nitrogen, LVEF) vs. Grace Score was evaluated on Prediction of acute heart failure (AUC 0.883, 95% CI 0.845-0.920, p=<0.0001). A novel predictive model incorporating systolic blood pressure, total bilirubin, LVEF, neutrophils, and urea nitrogen accurately predicted acute heart failure after PCI in STEMI patients with an AUC of 0.883.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: