Background
Approximately half of patients with heart failure (HF) have normal or only mildly impaired left ventricular ejection fractions (LVEFs) [
1,
2]. Patients with this profile, known as HF with preserved ejection fraction (HF-PEF), have signs, symptoms, quality of life (QoL), and prognoses similar to HF patients with a reduced ejection fraction (HF-REF) [
3,
4]. Furthermore, patients with acute myocardial infarction (MI) often have preserved ejection fraction (PEF) [
5]. Although many medical therapies benefit HF patients and post-MI patients with reduced LVEF [
6], effective, evidence-based pharmacologic treatments are not currently available for PEF patients [
7].
Aldeosterone-based activation of mineralocorticoid receptors has been demonstrated to contribute to the pathogenesis of HF and adverse cardiac remodeling after MI through multiple mechanisms, mainly including sympathetic activation, promotion of cardiac and vascular fibrosis, endothelial dysfunction, sodium retention, and potassium loss [
8,
9]. Mineralocorticoid receptor antagonists (MRAs) may inhibit these deleterious effects [
10] and may contribute to a beneficial therapeutic strategy for PEF patients. MRAs are effective for reducing total and cardiovascular mortality in patients with HF-REF (LVEF <35%) and post-MI patients with left ventricular dysfunction (LVEF <40%) [
11-
13]. However, whether they have a role in PEF remains to be clarified.
A recent series of studies assessed the efficacy of MRAs in HF-PEF patients and in patients with PEF after MI (MI-PEF) [
14-
19]. Although some studies failed to show a significant mortality benefit for MRA use [
14,
15], a number demonstrated a range of secondary benefits such as improved QoL, diastolic function, and cardiac remodeling, in response to MRA therapy [
16-
19]. As patients with PEF are usually older than HF-REF patients, a comprehensive evaluation may help provide support for therapy that improves symptoms and QoL, rather than mortality. In addition, since diastolic dysfunction and cardiac remodeling are considered the major underlying cardiac pathophysiology in HF-PEF and MI-PEF [
20], combining data regarding the impact of MRAs on these related parameters might elucidate some encouraging findings. However, data combining the experience from published randomized controlled trials to evaluate the effects of MRAs in PEF patients do not exist. Given the limited evidence concerning MRAs in PEF patients, this meta-analysis aimed to summarize the available data from randomized controlled trials (RCTs) to determine the efficacy and safety of MRAs in PEF (including both HF-PEF and MI-PEF) patients.
Methods
This meta-analysis was performed and reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines (Additional file
1) [
21].
Literature search
We searched the MEDLINE, EMBASE, Cochrane Library databases, and clinical trials databases (clinicaltrials.gov, controlled-trials.com, and clinicaltrialsregister.eu) for randomized controlled trials conducted between January 2000 and June 2014, using the following key words: i) mineralocorticoid receptor antagonists, aldosterone receptor antagonist, canrenoate, canrenoate potassium, canrenone, canrenoic acid, spironolactone, or eplerenone; ii) preserved left ventricular function, preserved ejection fraction, heart failure with normal ejection fraction, or diastolic heart failure; and iii) randomized controlled trials. Our literature search was limited to studies involving human subjects, reported in English. The list of full search strategies for EMBASE and MEDLINE is provided in Additional file
2. The search strategies for other databases are available on request.
Inclusion criteria
We included prospective, RCTs that: i) enrolled adult PEF patients with LVEFs ≥40% (including post-MI patients and those with symptomatic or asymptomatic HF), ii) assigned patients to MRA treatment versus placebo or control, iii) had at least one of the clinical outcomes of interest, and iv) had a study duration of at least 4 months.
Two independent reviewers screened all titles and the abstracts of all citations; potentially relevant articles were assessed according to the inclusion criteria. Disagreements were discussed until a consensus on inclusion/exclusion was reached. Information on patient characteristics, study design, quality, intervention strategies, and clinical outcomes was systematically extracted from each report using a standardized form. Data regarding safety and adverse events, including hyperkalemia and gynecomastia, were also noted. Hyperkalemia was defined as a potassium level >5.5 mmol/L. We used definitions of renal failure and gynecomastia as per the primary trial publication. The quality of the included RCTs was assessed using the Jadad quality scale [
22].
Outcome measures
The clinical outcomes for this meta-analysis were all-cause mortality and hospitalization due to HF. We also assessed echocardiographic parameters related to diastolic function, including E/e' (an estimate of filling pressure used to assess diastolic function), E/A (the ratio of early to late diastolic transmitral flow), E-wave deceleration time, and isovolumic relaxation time (IVRT) andvariables related to left ventricular structure and function, including LVEF, left ventricular end-diastolic volume index, left atrial volume index (LAVI), and left ventricular mass index (LVMI). More importantly, we also assessed relevant outcomes in terms of serum indicators and functional capacities: B-type natriuretic peptide, amino-terminal peptide of procollagen type-III (PIIINP), QoL, and 6-min walking distance.
Statistical analysis
For categorical variables, we calculated the relative risk (RR) and the absolute risk reduction (RD), as well as the corresponding 95% confidence intervals (CIs) for the outcome variables of interest, using the DerSimonian and Laird random effects model. Quantitative outcomes changing between baseline and follow-up were summarized and compared between the treatment and control groups using the weighted mean difference (WMD) and 95% CI, unless the outcomes used different scales, when the standardized mean difference (SMD) and 95% CI were used. The random-effects model using the DerSimonian and Laird method was used irrespective of heterogeneity because we anticipated heterogeneity between the trials [
23].
A priori, we defined significant heterogeneity between trials as an I
2 value of >50% [
24]. We assessed the evidence of publication bias using a funnel plot with an Eggers test [
25]. A two-sided
Pvalue <0.05 was considered statistically significant for all analyses.
Predefined subgroup analyses were conducted, a priori, according to the PEF subtypes (HF-PEF and MI-PEF), treatment durations (6 to 11 months, and ≥12 months), and MRA agent used (spironolactone, canrenoate, oreplerenone). If a given trial could be split into two or more separate studies, based on different treatment time points, the study with the longest follow-up was included in the meta-analyses. If a given trial could be split into two MRA groups with different doses, the group receiving the standard dose was included in the meta-analyses. Sensitivity analyses were conducted using sequential omission of a single study from the total studies and evaluating the influence of each study on the pooled effect estimates. All analyses were performed using Stata, version 11.2 (Stata Corp, College Station, TX, USA).
Discussion
In this meta-analysis of RCTs involving 6,248 patients, the effects of MRAs on patients with either MI-PEF or HF-PEF were evaluated. MRA treatment reduced the risk of hospitalization due to HF, improved QoL, reduced the E/e' or E/A ratio, increased LVEF, and reduced LVEDD and PIIINP levels in PEF patients. However, significant all-cause mortality benefits were not seen.
As patients with HF-PEF are usually older than HF-REF patients, hospitalization due to HF is increasing and represents a major burden in these patients [
1,
33], and emphasizes the growing need for effective, evidence-based therapies. However, previous pharmacological interventions, such as angiotensin-converting enzyme inhibitors [
34], angiotensin receptor blockers [
35], and beta-blockers [
36], have failed to show a significant reduction in hospitalizations due to HF. This meta-analysis provides important insights into the potential efficacy of MRA treatment for reducing the rate of hospitalizations due to HF in PEF patients, without increasing mortality. Reducing hospitalizations due to HF may help lower hospitalization costs and improve patient QoL. Additionally, significant MRA treatment benefits on composite outcome of death from cardiovascular causes, aborted cardiac arrests, or hospitalizations were observed after excluding patients recruited from Russia and the Republic of Georgia into the TOPCAT trial. HF-PEF patients from these jurisdictions, in that trial [
14], had extremely low placebo event rates, incompatible with those in prior HF-PEF studies [
35,
37]. The separate meta-analysis, excluding this population, might provide a more realistic insight into the effectiveness of MRAs in HF-PEF patients. Furthermore, we demonstrated that MRA treatment was associated with a significant improvement in QoL, measured by the KCCQ CCS. The KCCQ CCS has been reported to be a valid and reliable measure of health status and QoL in HF-PEF patients [
38]. Since the HF-PEF patients were elderly and typically demonstrated multiple comorbidities that might affect their mobility, it is not surprising that MRA treatment improved KCCQ CCS scores in these patients, but not exercise tolerance [
39]. Therefore, MRA treatment could be an option in PEF patients to improve their QoL.
Another encouraging finding from this meta-analysis was that MRAs improved both diastolic and systolic functions in PEF patients. Left ventricular diastolic dysfunction is the major underlying cardiac pathophysiology of PEF patients, and worse diastolic dysfunction has been associated with an increased risk of mortality [
40]. However, earlier pharmacotherapy did not achieve a significant improvement in diastolic function in HF-PEF patients [
7,
41,
42]. The present meta-analysis supports the potential clinical value of MRAs for improving diastolic function in PEF patients. We also found that MRA treatment significantly reduced the E/e' in HF-PEF patients and the E/A ratio in MI-PEF patients. Interestingly, MRAs were not associated with a significant reduction in the E/A ratio in HF-PEF patients. This may be because this ratio is rather complicated and cannot provide unequivocal evidence of diastolic dysfunction in HF-PEF [
43], whereas E/e' was demonstrated to be the most accurate, non-invasive index of diastolic function in HF-PEF [
43]. Our meta-analysis also found that MRA administration led to increased LVEF in PEF patients. A previous meta-analysis demonstrated that MRAs improved systolic function in HF-REF [
44]. Thus, the present meta-analysis shows that the beneficial effect of MRAs on systolic function also extends to PEF patients.
Although the favorable impact of MRAs on cardiac remodeling is well known in patients with HF and reduced LVEF [
44], the effect of MRAs in patients with preserved systolic function remained uncertain. This meta-analysis demonstrated that MRA administration could reverse cardiac remodeling in patients with preserved systolic function through a reduction of LVEDD and PIIINP levels. A subgroup analysis of LVEDD, based on treatment duration, found that the reduction became insignificant as the duration increased. This finding is consistent with a previous meta-analysis focusing on the effect of MRAs on cardiac structure in patients with left ventricular dysfunction [
44]. As PIIINP level has been proposed as an indicator of cardiac remodeling and poor clinical prognosis [
45], a reduction in serum PIIINP level might reflect the beneficial effects of MRAs on cardiac remodeling.
The benefits of MRAs on PEF patients are mainly attributed to the improvement of endothelial function and cardiac remodeling, as well as the decrease of myocardial fibrosis. Experimental evidence indicates that aldosterone-induced mineralocorticoid receptor activation provides an important unifying mechanism for many of the pathologic alterations of HF-PEF and MI-PEF [
46,
47]. The MRAs, through direct inhibition of aldosterone, were demonstrated to reduce myocardial fibrosis, improve vascular compliance and endothelial function, decrease inflammation and oxidative stress, and reduce the release of norepinephrine [
48]. These changes likely account for the diastolic function improvements seen on echocardiography and the reduction in collagen markers such as serum PIIINP level. MRA treatment was also associated with an increased risk of hyperkalemia and elevated serum creatinine levels. Our findings underscore the importance of monitoring electrolyte disorders and serum markers of kidney function during MRA treatment in clinical practice.
Several issues should be considered in the interpretation of our results. First, this meta-analysis was limited by the discrepancies in PEF diagnostic criteria employed in the clinical trials. The diagnosis of HF-PEF is still challenging because various criteria have been proposed to define patients with “diastolic HF” [
49]. The included RCTs had differing ejection fraction cut-off criteria (range, 40 to 50%) and challenges in diagnostic criteria for PEF, and may have resulted in a heterogeneous population. Nevertheless, patients with an ejection fraction of 40 to 50%, defined as borderline and intermediate PEF in the American College of Cardiology Foundation/American Heart Association guidelines [
49], were characteristically and prognostically similar to those with an ejection fraction ≥ 50% [
50]. Therefore, our meta-analysis does suggest a potential MRA treatment benefit for PEF patients. Second, some publication bias might exist in this meta-analysis, as we only included articles published in English. However, our statistical tests reported a low probability of publication bias in the pooling analysis. Third, different follow-up durations in the included trials might have produced heterogeneity, which limited the interpretation of pooled effect estimates. Finally, as the reported totals for all-cause mortality and hospitalizations due to HF were low, the assessment of the effect on clinical outcomes in PEF patients was of limited power. Despite the majority of evidence regarding clinical outcomes coming from the recently reported TOPCAT trial [
14], the findings of previous trials appear consistent.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made.
The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
YC and JB conceived and designed this meta-analysis. YC, HW, and YL performed the literature search and data extraction. YC, HW, XH, and JB conducted statistical analyses and data interpretations. YC and HW drafted the manuscript. YL and JB provided supervision. All authors critically revised the manuscript. All authors approved the final version of the manuscript submitted for publication and are guarantors of the study.