Prognosis after discontinuing renin angiotensin aldosterone system inhibitor for heart failure with restored ejection fraction after acute myocardial infarction

Abstract

Prognostic effect of discontinuing renin-angiotensin-aldosterone-system-inhibitor (RAASi) for patients with heart failure (HF) after acute myocardial infarction (AMI) whose left ventricular (LV) systolic function was restored during follow-up is unknown. To investigate the outcome after discontinuing RAASi in post-AMI HF patients with restored LV ejection fraction (EF). Of 13,104 consecutive patients from the nationwide, multicenter, and prospective Korea Acute Myocardial Infarction-National Institutes of Health (KAMIR-NIH) registry, HF patients with baseline LVEF < 50% that was restored to ≥ 50% at 12-month follow-up were selected. Primary outcome was a composite of all-cause death, spontaneous MI, or rehospitalization for HF at 36-month after index procedure. Of 726 post-AMI HF patients with restored LVEF, 544 maintained RAASi (Maintain-RAASi) beyond 12-month, 108 stopped RAASi (Stop-RAASi), and 74 did not use RAASi (RAASi-Not-Used) at baseline and follow-up. Systemic hemodynamics and cardiac workloads were similar among groups at baseline and during follow-up. Stop-RAASi group showed elevated NT-proBNP than Maintain-RAASi group at 36-month. Stop-RAASi group showed significantly higher risk of primary outcome than Maintain-RAASi group (11.4% vs. 5.4%; adjusted hazard ratio [HR_adjust] 2.20, 95% confidence interval [CI] 1.09-4.46, P = 0.028), mainly driven by increased risk of all-cause death. The rate of primary outcome was similar between Stop-RAASi and RAASi-Not-Used group (11.4% vs. 12.1%; HR_adjust 1.18 [0.47-2.99], P = 0.725). In post-AMI HF patients with restored LV systolic function, RAASi discontinuation was associated with significantly increased risk of all-cause death, MI, or rehospitalization for HF. Maintaining RAASi will be necessary for post-AMI HF patients, even after LVEF is restored.

Figure 1

Study flow. Study flow of the present study is shown. From the KAMIR-NIH registry, a total of 726 HFresEF patients were selected for analysis and follow-up during 36-month after index procedure. Abbreviations: AMI, acute myocardial infarction; FU, follow-up; HFresEF, heart failure with restored ejection fraction; KAMIR-NIH, Korea Acute Myocardial Infarction-National Institutes of Health; LVEF, left ventricular ejection fraction; RAASi, renin–angiotensin–aldosterone-system-inhibitor.

Figure 2

Changes of systemic hemodynamics according to treatment strategy of RAAS inhibitor after post-AMI 12-month. Serial change of systemic hemodynamic parameters, (A) SBP, (B) DBP, (C) heart rate, and (D) cardiac workload is presented in 3 groups; Maintain-RAASi, Stop-RAASi, and RAASi-Not-Used groups. Abbreviations: b.p.m., beats per minute; DBP, diastolic blood pressure; SBP, systolic blood pressure; otherwise as in Fig. 1.

Figure 3

Changes of LV systolic function and levels of NT-proBNP according to Treatment Strategy of RAAS Inhibitor after Post-AMI 12-Month. Serial change of (A) LV ejection fraction, and (B) concentration of NT-proBNP is presented in 3 groups; Maintain-RAASi, Stop-RAASi, and RAASi-Not-Used groups. Abbreviations: LV, left ventricular; NT-proBNP, N-terminal pro-B-type natriuretic peptide; otherwise as in Fig. 1.

Figure 4

Comparison of composite of all-cause death or spontaneous MI after post-AMI 12-month according to treatment strategy of RAAS inhibitor. Comparison of cumulative incidence and Kaplan–Meier curves of (A) a composite of all-cause death, spontaneous MI, or rehospitalization for HF, and (B) all-cause death or spontaneous MI among the 3 groups; Maintain-RAASi, Stop-RAASi, and RAASi-Not-Used groups, are presented. Abbreviations: HF, heart failure, otherwise as in Fig. 1.