Objective To investigate the relationship between early changes in left ventricular ejection fraction (LUEF) trajec- tories after acute myocardial infarction (AMI) and the long-term risk of heart failure (HF). Material and methods 299 AMI patients who underwent percutaneous coronary intervention (PCI) between January 2020 and December 2022 were retrospectively enrolled. Clinical and laboratory data were recorded and analyzed. The LUEF was measured within 24 hrs of admission (the baseline LUEF) and at one, three, and six months after discharge. Latent class growth mixture modeling (LCGMM) was used to identify post-discharge changes in LUEF trajectory patterns. During an additional 18 months follow-up, inci- dent HF and other outcome events were recorded. Cox proportional hazards regression models were applied to analyze the relationship between different LUEF trajectories and post-AMI HF risk. Results Four LUEF change trajectories were identified by LCGMM, and the patients were grouped accordingly: 1) Persistently Preserved LUEF Trajectory (98 patients, 32.8%). 2) Early LUEF Improvement Trajectory (89 patients, 29.8%). 3) Slow LUEF Improvement Trajectory (75 patients, 25.1%). 4) Persistently Reduced LUEF Trajectory (37 patients, 12.4%). Over a median additional follow-up of 18 mos, 62 patients (20.7%) developed incident HF. The HF incidence in the Persistently Reduced LUEF Trajectory patients was 62.2%, significantly higher than in the Slow LUEF Improvement Trajectory patients (30.7%), the Early LUEF Improvement Trajectory patients (12.4%), and the Persistently Preserved LUEF Trajectory patients (5.1%) (p<0.001 for all). Using the Persistently Preserved LUEF Trajectory patients as the reference, multivariable Cox regression analysis showed that the HF risk was increased in the Early LUEF Improvement Trajectory patients (p=0.024), further increased in the Slow LUEF Improvement Trajectory patients (p<0.001), and highest in the Persistently LUEF Reduced Trajectory patients (p<0.001). The area under the curve (AUC) for LUEF change trajectories in pre- dicting HF was 0.823, superior to the baseline LUEF alone (p=0.001). Diabetes mellitus, anterior wall myocardial infarction, Killip class II-III, and elevated peak N-terminal pro-B-type natriuretic pep- tide (NT-proBNP) were identified as significant, independent risk factors for HF, while angiotensin- converting enzyme inhibitor (ACEI)/angiotensin receptor blocker (ARB) use was a significant, inde- pendent protective factor. Conclusion Early changes in LUEF change trajectories after AMI are closely associated with the long-term risk of HF, with a persistently reduced LUEF trajectory conferring the highest HF risk. Dynamic monitoring of post-discharge LUEF changes provides superior predictive value compared with a single time-point measurement at admission. A combined prediction model integrating LUEF change trajectory, dia- betes mellitus, anterior wall MI, Killip classification, peak NT-proBNP, and ACEI /ARB use further enhanced predictive performance and facilitated early identification of patients at high risk of HF and individualized intervention.
