Differential Prognostic Impact of Risk-Prediction Models for Heart Failure in Acute Myocardial Infarction: The Original and Revised Heart Failure Time-Points.

Abstract

Background: We previously developed a risk-scoring system for heart failure (HF) in patients with acute myocardial infarction (MI), namely "HF time-points (HFTPs)". In the original HFTPs, the presence of HF on admission, during hospitalization, and at short-term follow-up was individually scored. This study examined whether the revised HFTPs, with additional scoring of previous HF, provide better predictivity. Methods: This multicenter registry included a total of 1331 patients with acute MI undergoing percutaneous coronary intervention. HF was evaluated at four time-points before and after acute MI onset: (1) a history of HF; (2) elevated natriuretic peptide levels on admission; (3) in-hospital HF events; and (4) elevated natriuretic peptide levels at a median of 31 days after the onset. When HF was present at each time-point, one point was assigned to a risk scoring system, namely the original and revised HFTPs, ranging from 0 to 3 and from 0 to 4. The primary endpoint was a composite of cardiovascular death and HF rehospitalization after discharge. Results: Of the 1331 patients, 65 (4.9%) had the primary outcome events during a median follow-up period of 507 (interquartile range, 335-1106) days. The increase in both original and revised HFTPs was associated with an increased risk of the primary outcomes in a stepwise fashion with similar diagnostic ability. Conclusions: The original and revised HFTPs were both predictive of long-term HF-related outcomes in patients with acute MI undergoing percutaneous coronary intervention. Yet, the original HFTPs may be sufficient to estimate HF risks after MI.