Predictive model for acute heart failure in STEMI patients after PCI shows high accuracy

This retrospective cohort study involved 419 STEMI patients who underwent percutaneous coronary intervention (PCI) at the Cardiology Department of Maanshan People's Hospital. The research focused on establishing and evaluating a predictive model for acute heart failure occurrence, comparing it to the baseline Grace Score. The primary outcome was model performance, with secondary metrics including sensitivity, specificity, and diagnostic accuracy.

Main results showed the predictive model achieved an area under the curve (AUC) of 0.902, indicating high discriminatory ability, compared to the Grace Score's AUC of 0.715, though the P-value for the Delong test was not reported. The model demonstrated a sensitivity of 74%, specificity of 86.8%, and diagnostic accuracy of 82.4%. Additional statistical tests supported model validity: the Hosmer-Lemeshow test indicated good fit (χ² = 6.551, P = 0.586), and the Omnibus test was significant (χ² = 7.112, P = 0.008).

Safety and tolerability data were not reported in the study. Key limitations include the retrospective design, which may introduce bias, and the single-center setting, limiting generalizability. The sample size of 419 is moderate but may not capture all patient variability. Funding and conflicts of interest details were not provided.

In practice, this model shows promise for predicting acute heart failure in STEMI patients post-PCI, with high accuracy metrics. However, clinicians should interpret these results cautiously due to the observational nature and lack of external validation. Further prospective studies are needed to confirm its clinical applicability and impact on patient outcomes.