Angiotensin-converting enzyme inhibitors (ACEIs) are considered as primary drugs for the secondary prevention of myocardial infarction (MI), and angiotensin receptor blockers (ARBs) are used when ACEIs are not tolerated [1, 2]. Both these types of drugs are recognized as renin-angiotensin system inhibitors (RASI), but differences in their modes of action have been reported to be associated with clinically different outcomes.

The usefulness of ACEIs for secondary prevention of cardiovascular events in patients with acute myocardial infarction (AMI) and heart failure with reduced ejection fraction (HFrEF) has been established by many randomized controlled trials (RCTs) [3‐5]. However, the effectiveness of ACEIs and ARBs as first-line drugs in patients with AMI with preserved left ventricular (LV) function has not been established. While both classes are equally effective in patients who have vascular disease or high-risk diabetes but do not have heart failure, they may be differentially beneficial in MI patients with preserved LV systolic function [6]. Additionally, while many physicians use ACEIs as a first-line therapy, up to 20% of patients develop adverse reactions such as a cough and/or angio-edema, and change to ARBs [7], but no data is available indicating whether changing from ACEIs to ARBs in MI patients with preserved LV systolic function can affect long-term clinical outcomes including mortality.

We sought to investigate the effectiveness of treatment with ACEIs or ARBs on clinical outcomes over 3 years in AMI patients with preserved LV systolic function following percutaneous coronary intervention (PCI) from a nationwide large-scale registry of AMI patients in South Korea. We also investigated how many patients changed medication after 1 year and the impact of changing medication on long-term outcomes.

The main findings of our study are as follows. First, in AMI patients with preserved LV systolic function who underwent PCI, the risk of total death or recurrent AMI was lower in patients treated with ACEIs in the first year compared to patients treated with ARBs. Both groups showed lower risk compared to patients who did not receive any RASI. Second, for NSTEMI patients, treatment with ACEIs in the first year was associated with a greater reduction in risk compared to ARBs, but both were generally equivalent in risk reduction in patients with STEMI, hypertension, and diabetes mellitus. This suggests that NSTEMI patients should be treated with ACEIs in the first year. Third, AMI patients with preserved LV systolic function had a poorer prognosis if RASI were stopped after 1 year, suggesting that treatment should be continued beyond 1 year after AMI. Fourth, after 1 year, about a third of patients on ACEIs switched to ARBs, likely due to side effects, but this change in treatment did not alter their prognosis or risk of total death or recurrent AMI at least up to 3 years. This suggests that after the first year, switching the medication does not impact on long term major clinical outcomes. However, continuation of treatment is necessary.

Most randomized trials have demonstrated that ACEI therapy with captopril, enalapril, ramipril, trandolapril, or zofenopril started within 24 h to 16 days following an AMI improves the LV ejection fraction at one month to 1 year [3, 5, 17‐19]. In these studies, the great majority of patients had an STEMI; data are limited in patients with a NSTEMI [10]. In addition, most patients were treated with either fibrinolytic therapy or no reperfusion; data in patients who underwent PCI, particularly treated with DESs for AMI are limited. ACEIs have also been reported to be beneficial in low risk patients after AMI [20], although the role of routine long term treatment in low risk patients who have undergone revascularization after STEMI is less certain. Therefore, the guidelines recommend that ACE inhibitors should be considered in all patients after STEMI in the absence of contraindications. [21]

Yang et al. showed that ARB had beneficial effects in terms of cardiac death or MI, which were comparable with those achieved with ACEIs in patients with STEMI with preserved LV systolic function [22]. They studied 6698 patients with STEMI who underwent PCI and had a LV ejection fraction ≥ 40% among the same Korean registry used in this study, but with data from a different time period (Between November 2005 and September 2010). Yang’s study differs from our study in that they enrolled only STEMI patients with EF greater than 40%, and they used a different time period and different duration of progress observation (Yang followed patients for 1 year, whereas we followed patients for 3 years). Yang’s study examined patients from a 5-year period (2005–2010) with EF greater than 40%, whereas our study examined patients from a different 5-year period (2011–2016) with EF greater than 50%, which reflect on the clinical outcomes of current contemporary PCI with DESs. In other words, our study examined a different set of lower-risk patients and examined longer-term outcomes by following them for 3 years instead of 1 year. It is notable that despite of these differences, both studies found that STEMI patients have no difference in the incidence of total death or recurrent AMI in groups treated with either ACEI or ARB at their 1-year progress observations. However, in the NSTEMI group (which was not analyzed by Yang et al.), our study found that NSTEMI patients who were treated with ACEIs during the first year after MI was associated with better clinical outcomes than NSTEMI patients treated with ARBs, suggesting the possibility that ACEIs may be more effective in the NSTEMI group in the first year than ARBs, although this requires further research. There are few studies examining RASI in NSTEMI patients. Kim et al. demonstrated that ACEIs showed more prominent ability to decrease the occurrences of MACEs, any repeat revascularization, and TVR compared to the ARB group in NSTEMI patients during a 2-year follow-up period. Additionally Byun et al. found that compared to ARB, ACEIs did not impact on causes of death, cardiac death, or recurrence of MI, but ACEIs did reduce the incidence of revascularization and MACE in diabetic NSTEMI patients [1, 23]. However, their study’s population included all NSTEMI patients, including patients with reduced EF. Our study only looked at AMI patients with EF greater than 50%, and suggests that the ACEI group had better outcomes than the ARB group in the first year. However, in our study, 33% of patients switched from ACEIs to ARBs during the first year, and patients may have changed drugs as early as the three to six-month points if they experienced side effects of ACEI, so it is not possible to conclude with certainty that one is more effective than the other.

Although the participants in our study had AMI with preserved LV systolic function, they also had a risk of LV remodeling post MI. LV remodeling is one of the major determinants of survival after recovery from MI, and the mechanism of LV remodeling within a few weeks after myocardial infarct reperfusion is well known[24]. The three main biomechanisms that contribute to the increase in the LV chamber volume post MI are (1) expansion of the infarct in the subacute phase, (2) subsequent non-ischemic infarct extension into the adjacent non-infarcted region, and (3) hypertrophy and dilatation of the non-infarcted myocardium in the chronic phase. Previous studies have led to a general consensus that treatment with ACEIs and ARBs has an anti-fibrotic effect on collagen formation in infarcted and remote non-infarcted regions [25]. This is the rationale for implementing drug therapy as soon as possible after MI, and the current ACC/AHA guidelines recommend ACE inhibitors as Class I drugs for use within the first 24 h after MI. When considering the obvious clinical benefits of ACEIs and ARBs, this suggests that there may be excessive of fibrosis of the infarct in untreated patients, when a smaller, more compact scar is preferred. Although our study cannot determine the superiority of either ACEIs or ARBs, the clinical results of ACEIs tend to be superior to those of ARBs. A plausible hypothesis for this is that these two drugs (ACEIs and ARBs), which are in the same system in terms of RAS inhibition, have different mechanisms of action[26]. ACEIs inhibit the conversion of angiotensin I to angiotensin II, which reduces the level of serum angiotensin II and activates the bradykinin system. In contrast, ARBs may cause prolonged elevation of angiotensin II levels and the upregulation of angiotensin I. It is well known that elevated serum levels of angiotensin II play an important role in the pathogenesis of coronary artery disease [27]. In addition, elevated angiotensin II levels may direct an increase in serum cholesterol levels via the interaction with macrophage AT1 receptors, stimulating 3-hydroxy-3-methylglutaryl coenzyme A reductase gene expression, ultimately leading to cholesterol accumulation in macrophages and foam cell formation [28].

Some data showed that treatment with ACEIs or ARBs after AMI was associated with improved long-term survival compared to a group not receiving treatment, regardless of underlying renal function, and was accompanied by low rates of adverse renal events [29]. The findings of our study support earlier studies that showed patients with kidney disease receiving RASI treatment after MI have better outcomes than patients not receiving RASI treatment. Although, the statistical significance was not confirmed, but risk of death or MI were better in both the ACEI and ARB groups than in the No RASI group. In all cases, our results suggest that RASI are effective and should be used in patients following AMI with preserved LV systolic function.

There is much controversy over the MI paradox when using ARB. Angiotensin II receptor stimulation may be less beneficial than previously proposed and may even be harmful under certain circumstances through mediation of growth promotion, fibrosis, and hypertrophy, as well as proatherogenic and proinflammatory effects. However, in a total of 19 of 24 studies of data on AMI, representing a total of 31 569 patients, use of ARBs was not associated with an increased risk of MI compared with placebo (odds ratio [OR] 0.94, 95% CI 0.75 to 1.16) nor when compared with ACE inhibitors (OR 1.01, 95% CI 0.87 to 1.16). Our data showed that 33% of patients using ACEIs changed to ARBs at the end of a year, and these patients showed an equal or improved prognosis compared to those who remained on ACEIs. This suggests that ARBs are safe and effective in a population with AMI and preserved LV function, particularly AMI patients undergoing PCI with DESs.

First, our study did not have data on LV function at follow-up, meaning that our study cannot account for whether patient’s outcomes were impacted by changes in LV function or if they regained or lost LV function during the study period. Because it is possible that RASI have different effectiveness in populations with greater or reduced EF, it will be necessary to examine this more in the future to confirm these findings. Second, because the registry data is non-randomized, it is possible there was selection bias. Although we performed an IPWT matched analysis to account for these potential confounding factors, it was not possible to correct for unmeasured variables. For example, our database does not include information on why physicians switched patients from ACEIs to ARBs. In the future, it will be necessary to conduct randomized trials to confirm and expand the findings of this study. Third, adverse clinical events were not centrally adjudicated in our registry, and were instead identified by the patient’s physician and confirmed by the principal investigator of each hospital. As a result, it is possible that some adverse events were not captured in the database.