Heart failure after myocardial infarction: Glass emptier than full

This article refers to ‘Declining risk of heart failure hospitalization following first acute myocardial infarction in Scotland between 1991–l2016’ by K.F. Docherty et al., published in this issue on pages 1213–1224.

Heart failure (HF) is an ominous and common complication following an acute myocardial infarction (MI). In fact, there is no single greater predictor of mortality following an MI than acute symptoms of HF.¹ Once the acute period surrounding an MI is past, the risk continues, and an estimated 12–15% of all patients experiencing an MI will develop HF requiring hospitalization within a year. This risk appears to be especially high for patients who are older, female, who have reduced ejection fraction or acute congestion at the time of MI, and/or in those with certain comorbidities such as type 2 diabetes or chronic kidney disease.¹ Once patients develop HF following an MI, their risk of death is markedly elevated. The risk of death associated with HF is substantially higher than the risk of death associated with a recurrent MI.² As a result, a number of drugs with demonstrated risk reduction in death and/or HF hospitalization are routinely used as guideline-directed medical therapy (GDMT) in high-risk patients following an MI, including beta-blockers, angiotensin-converting enzyme inhibitors or angiotensin receptor blockers, and mineralocorticoid receptor antagonists.³

It has been two decades since a post-MI drug trial identified a therapy that could reduce the risk for HF.³ Despite this, there are reasons to be hopeful that medical care for this population has improved, with regional efforts to improve acute coronary syndrome care management and activation of catheterization laboratories, new trials in antiplatelet therapies, and improved implementation of GDMT in this population, all of which may all contribute to improvements in outcomes for these patients.^{1, 4} Given the high morbidity and mortality associated with HF after an MI, it is important to understand the contemporary demographic and clinical profile of these high-risk patients in order to provide best care, and to inform future trials investigating potential therapies for this vulnerable population.

We therefore read with great interest the paper by Docherty et al.⁵ exploring trends in HF hospitalization after a first-time acute MI between 1991 and 2015 in Scotland. Patients with either a prior MI or a prior HF diagnosis were excluded from this analysis. The authors found that over a median follow-up of 6.7 years, 12.2% of patients required a HF hospitalization following an MI, with an overall decrease in the incidence of hospitalization for HF over time of approximately 50%. A similar decrease was seen in the risk of death following an MI. As in prior research, patients who were female, older, had type 2 diabetes or chronic kidney disease, or who did not receive revascularization were more likely to experience subsequent HF hospitalization after an MI. The authors also note that patients with socioeconomic deprivation, which was defined across seven measures including income, employment, health, education, skills and training, housing, geographic access, and crime, were at an elevated risk for future HF. The authors show that patients who develop HF requiring hospitalization after an MI were at a five-fold higher annualized risk for death compared to those who were never hospitalized for HF. Median survival after a hospitalization for HF was only 1.7–1.8 years and did not change significantly over the time span analysed, though in adjusted models there did appear to be a statistical decrease in mortality risk after HF hospitalization by about 30%.

This research is a valuable addition to the scientific literature and provides additional weight to prior observations that rates of HF and death after an MI have decreased over the past decades.¹ This analysis reinforces previous work that has identified age, sex, and certain comorbidities as critical risk factors for the development of HF. We do regret that the database used, while large, was unable to collect data on outpatient diagnoses/management of HF, or on other relevant clinical data such as ejection fraction or the presence of acute congestion or Killip class at the time of an MI, which are known to be major risk factors for development of future HF. However, we acknowledge that when utilizing large national datasets there are often inherent limitations to granularity.

This work also raises important questions about the evolution of MI diagnosis over the past several decades. The emergence of high-sensitivity troponin has resulted in potential reclassification of what was previously considered unstable angina to now be categorized as non-ST-elevation MIs. It is possible that many of these recategorized events, with relatively small (though dynamic) changes in troponin, which previously would have not been considered MI, are relatively unlikely to result in new-onset HF, given the minimal degree of cardiac damage that may have occurred.

Importantly, the results from Docherty et al.⁵ emphasize the high residual risk of de novo HF after a first time MI: 12.2% overall. The actual burden of HF is likely underestimated in this study, since patients with a previous history of MI are likely at even higher risk for future HF, and because some patients with HF will be diagnosed and treated in the outpatient setting. Despite this likely underestimation, the most contemporary rates of hospitalization for HF after an MI observed in this study were still 31.3 events per 1000 patient-years, making it likely that HF remains one of the largest clinical risks following an MI.¹ Furthermore, patients with HF post-MI were at a markedly higher risk of death than those without HF, emphasizing the critical need to better prevent and manage HF in these patients. It has been previously noted that the risk of death or hospitalization among patients with even stable HF is ∼10% per year and can exceed 30–40% among those with a recent HF hospitalization. In contrast, even for patients with a prior atherosclerotic cardiovascular disease event such as an MI, risk of death or hospitalization in a given year is unlikely to exceed ∼7–10%.² Though the current study did not examine risk of repeat hospitalizations among patients with and without a HF hospitalization, patients with a HF hospitalization had a fatality rate that was ∼5× higher than those who were not hospitalized for HF after their MI (Figure 1). Put another way, the 12.2% of patients who developed HF after MI carried ∼40% of the observed risk of death for all patients in the years following an index MI.

There remains, therefore, a critical unmet need to prevent the development of HF after an MI. To some extent, this effort likely lies in identifying the patients at highest risk, i.e. those who are elderly, female, with comorbidities, or with reduced ejection fraction and/or acute signs or symptoms of congestion following their MI, and ensuring the implementation of GDMT in these high-risk patient populations. We also note that the authors found that patients with socioeconomic deprivation were also at a higher risk for developing HF and worry that a part of this may be related to inadequate access to medical therapies that are known to help prevent and manage HF.

Beyond emphasizing aggressive implementation of current GDMT in high-risk populations, there is also a great need to identify new therapies to help decrease the risk of HF development. The recently completed Prospective ARNI versus ACE Inhibitor Trial to Determine Superiority in Reducing HF Events After MI (PARADISE-MI) trial did not find that the use of sacubitril-valsartan after an MI resulted in lower risk of cardiovascular death or incident HF when compared to ramipril.⁶ The ongoing Study to Test the Effect of Empagliflozin on Hospitalization for HF and Mortality in Patients With Acute MI (EMPACT-MI) trial will evaluate the impact of the sodium–glucose cotransporter 2 inhibitor (SGLT2i) empagliflozin on the risk of death or HF hospitalization after an MI,⁷ and the Dapagliflozin Effects on Cardiometabolic Outcomes in Patients with an Acute Heart Attack (DAPA-MI) trial will provide complementary information on the impact of dapagliflozin on a hierarchical endpoint of death, HF hospitalization, non-fatal MI, atrial fibrillation, new-onset diabetes, New York Heart Association class, and weight loss of ≥5%.⁸

In summary, the results of this well-performed study likely reflect successes of the healthcare system in taking care of patients with MI, including improved diagnosis and acute management, and the improved implementation of GDMT following an MI. However, these findings likely under report the true burden of HF post-ischaemic coronary events, given their exclusion of patients with prior history of MI, and of those who were diagnosed and treated exclusively in the outpatient setting. This study also underscores the high residual risk of developing HF after an MI, and the high mortality risk for those patients who do go onto develop HF after an MI. As a result, there remains a critical need to utilize all available therapies to reduce the risk of developing HF following an MI, and to identify new therapies to further reduce this risk. We therefore look forward to the results of the ongoing trials with SGLT2is.

Conflict of interest: J.H. receives salary support from T32 training grant T32HL069749. J.B. reports serving as a consultant to Abbott, Adrenomed, Amgen, Array, AstraZeneca, Bayer, Berlin Cures, Boehringer Ingelheim, Bristol-Myers Squib, CVRx, G3 Pharmaceutical, Innolife, Janssen, LivaNova, Luitpold, Medtronic, Merck, Novartis, Novo Nordisk, Occlutech, Relypsa, Roche, Sanofi, SC Pharma, V-Wave Limited, and Vifor.