PARADISE-MI suggests a limited role of intensified neuro-hormonal inhibition in the management of acute myocardial infarction with reduced ejection fraction

Comment on ‘Angiotensin receptor-neprilysin inhibition in acute myocardial infarction’ published in the New England Journal of Medicine https://doi.org/10.1056/NEJMoa2104508. Key Points The industry-sponsored, international, multicenter, randomized, double-blind, active-comparator trial, PARADISE-MI1 (Prospective ARNI vs. ACE Inhibitor Trial to Determine Superiority in Reducing Heart Failure Events after Myocardial Infarction), evaluated the efficacy and safety of the angiotensin receptor–neprilysin inhibitor sacubitril–valsartan (S/V) (target dose, 97/103 mg twice-daily) vs. ramipril (target dose, 5 mg twice-daily) in 5661 patients with reduced left ventricular ejection fraction (LVEF ≤40%), pulmonary congestion requiring intravenous treatment, or both conditions following an acute myocardial infarction (AMI). Patients were eligible if they also had at least one of eight risk-augmenting factors [age ≥70 years, diabetes mellitus, previous MI, an estimated glomerular filtration rate of <60 mL/min per 1.73 m2 of body surface area, atrial fibrillation, LVEF <30%, Killip class III or IV, ST-segment elevation MI (STEMI) without reperfusion within 24 h after presentation]. It was also required that the systolic blood pressure be at least 110 mmHg in patients who had not taken an angiotensin-converting enzyme inhibitor (ACEi) or angiotensin receptor blocker (ARB) in the 24 h before randomization and at least 100 mmHg in patients who had. Patients with clinical haemodynamic instability were excluded. The trial was event driven and randomized 2830 patients to receive S/V and 2831 to ramipril. The patient population had a mean age of 64 ± 12 years, 24% were women, 76% presented with STEMI, PCI was performed in 87%, and the mean time to randomization was 4.3 ± 1.8 days after the index AMI. LVEF was 37 ± 9% and 57% were in Killip class ≥II. Dual antiplatelet therapy was prescribed in 92% of patients, 95% were treated with a statin, 85% received beta-blockers, and 42% a mineralcorticoid-receptor antagonist (MRA) at baseline. Either an ACEi or an ARB was used in 78% of patients and was discontinued at randomization, except for those assigned to S/V in whom the ACEi or ARBs was interrupted 36 hours before randomization. During a median follow-up of 22 months, 711 events occurred (708 primary-outcome events were required to provide the trial with 80% power to detect a hazard ratio of 0.81). No statistically significant difference in the primary composite outcome of cardiovascular (CV) death, heart failure (HF) hospitalization or outpatient development of HF, whichever occurred first, was found between the S/V and ramipril groups [11.9% vs. 13.2%; hazard ratio (HR) 0.90, 95% confidence interval (CI) 0.78–1.04; P = 0.17]. As component of the primary outcome, CV death occurred in 137 (4.8%) patients in the S/V arm vs. 136 (4.8%) in the ramipril arm; HF hospitalization or outpatient development of HF occurred in 201 (7.1%) patients vs. 237 (8.4%). Because the result for the primary outcome was not significant, subsequent analyses of the secondary outcomes were considered exploratory. All all-cause death occurred in 7.5% vs. 8.5% (HR 0.88, 95% CI 0.73–1.05). Hypotension was more frequent with S/V compared with ramipril (28.3% vs. 21.9% of patients, P < 0.001), while cough and hepatotoxicity were higher in the ramipril group. Permanent study discontinuation occurred similarly in both groups (17.8% vs. 18.4%), as well as discontinuation because of an adverse event (12.6% vs. 13.4%). Current management of AMI has seen multiple advances related to early primary PCI with drug-eluting stents and optimal pharmacological therapy. Despite the remarkable improvement in outcomes, AMI survivors with reduced LVEF, transient pulmonary congestion, or both remain at substantial risk for re-infarction, symptomatic chronic HF, or premature death. This leaves room for the intensification of therapeutic strategies and provides the rationale for testing potential benefits related to a broader neurohormonal inhibition using S/V. This hypothesis was based on the beneficial effect of this combination in reducing CV death, HF hospitalization, and all-cause mortality, in comparison with a standard dose of enalapril, in patients with chronic HF with reduced ejection fraction (HFrEF),2 and on the finding that S/V could be safely administered to patients admitted to hospital with worsening HF.3 Seminal clinical studies showed that very early use of ACEi improved the prognosis of patients with a recent MI complicated by left ventricular systolic dysfunction (LVSD) and HF.4–6 ARBs were subsequently tested in comparison to ACEi, in patients with recent AMI and additional risk factors. In the VALIANT7 trial, the use of valsartan was shown to be safe but not superior to captopril in reducing all-causes mortality. Also, MRAs were safe but showed little additive benefits on mortality or hospitalizations in patients with AMI and HF.8 The remarkable differences in patient management in the major randomized trials of ACEi and ARBs, vs. PARADISE-MI provide an appreciation of the impact of the evidence-based practice-changing paradigms.4–6,9 During the past 3 decades (since the introduction of ACEi, ARBs, and MRAs in the treatment of AMI with reduced LVEF) reperfusion strategies with PCI based on the use of more effective drug-eluting stents, and the earlier use of more intense platelet inhibition, high-dose statins and beta-blockers provided significant prognostic improvements.10 Indeed, in PARADISE-MI, there was a greater use of PCI, beta-blockers, statins, and dual antiplatelet therapy.1 These differences certainly had an impact on the event rates that was ∼2.5-fold lower in PARADISE-MI than in previous trials. For example, overall mortality was 8.5% at 2 years in the ramipril arm of PARADISE-MI,1 compared to nearly 20% in the VALIANT trial.7 Those observations raise some important issues. First, PARADISE-MI was adequately powered for its primary hypothesis of a 19% risk reduction in primary-outcome events,9 but largely underpowered to detect a modest 10% risk reduction. Second, the reasons for a smaller-than-expected effect size are most likely multifactorial, including the choice of the two components of the primary composite outcome (for only one of which there seemed to be a signal of responsiveness to S/V treatment, i.e. HF hospitalization or outpatient development of HF) as well as the optimal management of the enrolled patients, leaving little room for further improvement with a drug combination targeting neuro-hormonal activation. Finally, among the potential confounding factors that may deserve to be considered, the early (36 h) discontinuation of ACEi or ARB for safety reasons in patients assigned to S/V may have interfered differently with the early remodelling process in the two groups. This may be relevant in view of the early randomization of patients in the presence of variable exposure to neurohormonal inhibition (mean, 4.3 days). On the other hand, the study design based on the use of ramipril or S/V in the early stages of AMI without a run-in phase may support the relative safety of both treatments also in consideration of the fact that both therapies were associated with a limited discontinuation rate because of an adverse event (∼13%). At this stage, it is difficult to say whether or not there is further room for the benefits of neurohormonal inhibition or other pharmacological interventions after PARADISE-Mi, but we will probably learn more from ongoing and planned studies in this area. Conflict of interest: G.L. received grant support (to the Institution) for investigator-initiated research from American Heart Association, Italian National Health Service, and Italian Minister of Education, University and Research. She is currently involved in the Research Programs of the Italian Cardiovascular Network. She received personal fees from Astra Zeneca, Boehringer Ingelheim, Novo Nordisk, and Daiichi Sankyo.