Epidemiological data further demonstrate that heart failure (HF) is a global disease with increasing prevalence and burden (Bozkurt et al., 2021). The most common cause of HF is coronary artery disease (CAD) (Conrad et al., 2018). Thus, it is extremely important for a physician to use important indexes to evaluate both the measure and efficacy of therapy outcome among CAD patients with HF.

Recently, a scientific panel has proposed a new working definition of HF with improved ejection fraction (HFimpEF) that includes a baseline left ventricular ejection fraction (LVEF) of ≤ 40%, a ≥ 10% increase from baseline LVEF, and a second measurement of LVEF of>40% (Wilcox et al., 2020; Bozkurt et al., 2021). The most recent consensus indicated the necessity to employ and use a universal definition and classification in research studies. In consequence, differences in the prognosis and clinical management of new patients with HFimpEF and those without improvement have attracted increasing attention, especially among patients with CAD.

However, current studies on the prognosis based on the new definition of HFimpEF are lacking. Thus, we sought to comprehensively investigate the clinical characteristics and outcomes of HFimpEF among patients with CAD.

This is the first study to investigate the characteristics and outcomes of a universal new definition of HFimpEF among patients with CAD. Our study showed that CAD patients with HFimpEF had fewer complications, and there were no statistically significant differences in terms of age and sex. Importantly, CAD patients with HFimpEF significantly achieved a 39% reduction in long-term mortality risk than patients with persistent HFrEF.

The burden of HF has increased to an estimated 23 million people globally, and approximately, 50% of cases are HFrEF, which results in millions of hospitalizations; hospitalized patients with HF continue to experience high post-discharge mortality and readmission rates (Ambrosy et al., 2014; Murphy et al., 2020). In addition, enhanced survival following AMI and the declining prevalence of hypertension and valvular heart disease are contributors to incident HF and have fueled the emergence of CAD as the primary risk factor for the development of HF (Lala and Desai, 2014). What complicates patient management is the difficulties clinicians face in terms of the uncertainties in the diagnosis of patients with CAD and HFrEF because of the non-specificity of the symptoms and signs of HF (Yancy et al., 2013). The evaluation of LVEF is among the most important and commonly used parameters in the diagnosis, characterization, prognosis, patient management, and treatment selection of HF (Lund et al., 2018). Improvement of left ventricular function is the one of the most important goals of HF therapy associated with reduced EF from any cause (Basuray et al., 2014).

It has been recently recognized that improvement of LVEF occurs in a proportion of HFrEF and is associated with a better prognosis. Multiple prior studies have documented outcomes with improved LVEF. In the study by Jørgensen et al. (2018) patients with improved LVEF (≥5% improvement) had a significantly lower risk of mortality and cardiovascular events compared with patients with persistently reduced LVEF. Perry et al. (2020) determined that an improvement in LVEF of ≥ 10% was independently associated with reduced mortality. Ghimire et al. (2019) also reported that HF patients with EF recovery (defined by EF absolute improvement≥10%) achieved a substantially better prognosis compared to patients with persistent HFrEF, even after multivariable adjustment. Taken together, these studies used different definitions for improvement or recovered LVEF.

The exact mechanisms underlying the association between CAD patients with HFimpEF and better prognosis have been extensively studied; hence, several hypotheses have been proposed. Howlett et al. (2020) indicated additional late improvement in LVEF and unchanged troponin levels for patients with HFimpEF, in contrast to those with persistent HFrEF and increase in serum troponin levels over time. These data provide evidence on the underlying mechanisms associated with late LV remodeling (Howlett et al., 2020). A large proportion of evidence supports reverse LV remodeling and recovery of LV function after implementation of evidence-based medical, device-based, and surgical interventions in patients with chronic HFrEF. Importantly, reverse LV remodeling is associated with improved myocyte contractility and improved LV chamber contractility (Wilcox et al., 2020). It is important to recognize that therapy-induced changes in LV remodeling are associated with fewer HF hospitalizations and decreased cardiovascular mortality and that there is a direct correlation between the extent of reverse LV remodeling and improvements in cardiac survival (Kramer et al., 2010). Importantly, in our study, we found that patients with CAD of younger age, with fewer comorbidities including PCI, history of PCI or AMI, and smaller left ventricular end diastolic diameter are more likely to advance to improved LVEF. Currently, there are no firm recommendations or guidelines on the role of PCI in managing HFrEF due to a lack of evidence from randomized controlled trials. The 2017 appropriateness criteria for revascularization do not provide a recommendation for PCI in patients with LVSD due to insufficient data (Parikh et al., 2021). At present, it is still controversial whether PCI is beneficial to left ventricular remodeling in patients with left ventricular dysfunction. In addition, the worse results in PCI patients could be explained by a more serious disease in patients with PCI. In addition, we noticed that neurohormonal medication that is proven to increase LVEF (especially beta-blockers) did not significantly alter our model. One reason was that our drug information was discharge medication. It cannot truly reflect whether patients take these drugs regularly and benefit from them. Further research is needed to validate these hypotheses and to explore the underlying mechanisms.

Risk stratification helps clinicians to screen patients with CAD whose EF may be difficult to improve in clinical practice and require investment of more medical resources. In fact, there is a substantial cross-over between treatment strategies for CAD and HFrEF, such as the use of β-blockers and an angiotensin receptor-neprilysin inhibitor, angiotensin-converting enzyme inhibitor, or angiotensin receptor blocker as foundational therapy, with addition of a mineralocorticoid receptor antagonist in patients with persistent symptoms (Chang et al., 2018). Ivabradine and hydralazine/isosorbide dinitrate also play a role in the care of certain patients with HFrEF (Murphy et al., 2020). More recently, sodium-glucose cotransporter 2 inhibitors have further improved disease outcomes and have reduced HF hospitalization in high-risk patients with HFrEF (Murphy et al., 2020). This may be an important treatment options for CAD patients with HFrEF in the future. In addition, EF is a poor index of myocardial efficacy and can be improved by the development of severe mitral regurgitation (MR) (Asgar et al., 2015; Goliasch et al., 2018). Thus, it is important to assess the severity of MR among CAD patients with HFrEF to guide and support clinical decisions and to establish treatment guidelines (Wilcox et al., 2020).

In addition, we found that patients with baseline LVEF and follow-up LVEF comprised a higher proportion of younger men, albeit there were no significant differences in terms of complications. Some drugs were used more frequently, and the mortality rates were lower, which was an interesting result. Since we recall patients for follow-up, younger men may pay closer attention to their health and may be suitable candidates for targeted drug therapy, thereby reducing mortality. However, this requires additional real-world data to explore the underlying relationship.

This study examined, for the first time, the characteristics and outcomes of the new definition of HFimpEF among patients with CAD. Although there are several limitations to be considered, the sizeable amount of data extracted from medical records allowed us to control a variety of confounders and selection bias in our analyses. First, the data were extracted from a single center, which hampered our control over a variety of confounders in analyses. Second, the population of our study may not represent all CAD patients with HFrEF because some patients without follow-up echocardiography were excluded. Third, we used echocardiography data from the 1-year follow-up to determine HFimpEF, even though we performed the sensitivity analysis and only included data from 6 months or more for the LVEF follow-up. Although the results did not change significantly, our sample size was too small to allow presenting them as main results. Fourth, we lack information about whether patients had accepted coronary artery bypass graft (CABG), which made it difficult to assess the role of CABG in our study. Finally, although we acquired information of LVEF of patients from very skilled cardiologists, LVEF data were determined without double-blinding, meaning that the two independent sonographers blinded from one another LVEF evaluation did not evaluate the LVEF of other. In addition, it is unknown how often biplane method of Simpson was used. This may lead to some errors in evaluating the improvement of ejection fraction. Finally, some potential confounding variables, such as drug compliance, valve disease, complete revascularization, and on the history of HF, were not considered in this study. However, our data analysis included the main clinical parameters.

Our study demonstrated that HFimpEF significantly reduced long-term mortality risk by 39% compared to patients with persistent HFrEF, which may serve as an effective end point evaluable in future studies. In addition, many independent important risk predictors such as younger age, fewer clinical comorbidities, and lower left ventricular end diastolic diameter are identified, and these predictors may be used to optimize patient care among individuals with persistent HFrEF. Importantly, more prospective studies, including randomized controlled trials, are needed to improve the recognition of the new definition of HF, which is more conducive to accurate diagnosis and treatment indications among patients with CAD.

The data analyzed in this study is subject to the following licenses/restrictions: Not applicable. Requests to access these datasets should be directed to YL, liuyong@gdph.org.cn.

The studies involving human participants were reviewed and approved by the Guangdong Provincial People’s Hospital Ethics Committee. Written informed consent for participation was not required for this study in accordance with the national legislation and the institutional requirements.

YL, NT, JC, YHu, and C-QO designed the study. SC, YHe, and BW collected and reviewed clinical and laboratory data. SC, HH, QL, and WL analyzed data. SC, HH, YHe, and BW performed the statistical analysis. HH, JL, ML, ZY, and KB performed drafting or revision of the manuscript. YL, NT, MA, and JC reviewed, interpreted, and checked clinical data. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.