Skip to main content
ESC Heart Failure logoLink to ESC Heart Failure
. 2022 Aug 31;9(6):4088–4099. doi: 10.1002/ehf2.14125

Long‐term mortality in heart failure with mid‐range ejection fraction: systematic review and meta‐analysis

Deep Chandh Raja 1,2,3,4, Indira Samarawickrema 4,2, Souvik Das 3,4, Abhinav Mehta 1, Lukah Tuan 1,4, Sanjiv Jain 2, Sanjay Dixit 5, Frank Marchlinski 5, Walter P Abhayaratna 1,3, Prashanthan Sanders 6, Rajeev K Pathak 1,2,3,4,
PMCID: PMC9773707  PMID: 36045010

Abstract

Aims

Heart failure patients with mid‐range ejection fraction (HFmrEF) have overlapping clinical features, compared with patients with reduced ejection fraction (HFrEF) and preserved ejection fraction (HFpEF). We aim to perform a meta‐analysis of studies reporting long‐term outcomes in HFmrEF compared with HFrEF and HFpEF.

Methods and results

Data from 18 eligible large‐scale studies including 126 239 patients were pooled. Patients with HFmrEF had a lower risk of all‐cause death than those with HFrEF [risk ratio (RR) = 0.92; 95% CI = 0.85–0.98; P < 0.001]. This significant difference was seen in the follow‐up at 1, 2, and 3 years. Patients with HFmrEF had significantly lower risk of cardiovascular (CV) deaths than HFrEF (RR = 0.77; 95% CI = 0.65–0.92; P < 0.001). Subgroup analysis showed that studies recruiting >50% of males had higher risk of deaths with HFrEF (RR = 1.15; 95% CI = 1.04–1.26; P = 0.006). When compared with HFpEF, patients with HFmrEF had comparable risk of all‐cause death (RR = 1.02; 95% CI = 0.96–1.09; P = 0.53). Similarly, there were no differences in the 1, 2, and 3 year deaths; CV and non‐CV deaths were insignificant between HFmrEF and HFpEF.

Conclusions

The results of the study support that HFmrEF has better prognosis than HFrEF but similar prognosis when compared with HFpEF. Gender disparity between studies seems to influence the results between HFmrEF and HFrEF. Transition in left ventricular ejection fraction (LVEF), which could not be addressed in the study, may play a decisive role in determining outcomes. PROSPERO review registration number CRD42021277107.

Keywords: Systematic review, Meta‐analysis, Heart failure, Mid‐range ejection fraction, Mortality, Gender differences

1. Introduction

Mortality in patients with symptomatic chronic heart failure varies from 25% to 80% at 5 years, depending on the stage of heart failure. 1 Left ventricular ejection fraction (LVEF) is an important prognostic indicator for the risk of all‐cause mortality as well as sudden cardiac death in chronic heart failure patients. 2 Since the introduction of categories of chronic heart failure based on LVEF, heart failure with mid‐range (borderline) ejection fraction (HFmrEF) is recognized as an entity with overlapping clinical, treatment, and outcome characteristics compared with heart failure with reduced ejection fraction (HFrEF) and heart failure with preserved ejection fraction (HFpEF). 3

Large‐scale registries, multi‐centre, and single‐centre studies have reported their observations on prognosis in each of these categories of heart failure. 4 , 5 , 6 , 7 , 8 , 9 , 10 , 11 , 12 , 13 , 14 , 15 , 16 , 17 , 18 , 19 , 20 , 21 However, only a few meta‐analyses have been carried out comparing the prognosis of HFmrEF with HFrEF and HFpEF. 22 , 23 , 24 Moreover, these meta‐analyses did not then have studies adequately powered to examine the annual mortality risk in the follow‐up between these groups. Additionally, none of the meta‐analyses have explored the difference in outcomes according to certain important subgroups, which could be important moderators influencing the results. To explore these gaps in knowledge, we proposed a systematic review and meta‐analysis to determine the long‐term cumulative and annual mortality risk, cardiovascular and non‐cardiovascular deaths in patients with HFmrEF compared with those with HFrEF and HFpEF.

2. Methods

2.1. Objectives

The primary objective was to determine the effect size of long‐term cumulative mortality risk and all‐cause mortality risk at 1, 2, 3, 4, 5, and 10 years; in persons with HFmrEF compared with HFrEF and HFpEF. Our secondary objectives were to (1) explore the long‐term cause‐specific mortality risk for deaths from cardiovascular, non‐cardiovascular, heart failure, and sudden cardiac deaths in persons with HFmrEF compared with HFrEF and HFpEF and (2) identify clinical variables influencing the long‐term mortality risk estimates.

2.2. Search strategy and eligibility criteria

We followed the Preferred Reporting Items for Systematic review and Meta‐Analyses (PRISMA) guidelines in conducting this systematic review and the meta‐analysis. 25 We searched MEDLINE, PubMed, EMBASE, Web of Science, and Cochrane Library from inception to 30 September 2021, for studies reporting mortality outcomes between persons with heart failure with mid‐range/borderline ejection fraction (40%–49%, HFmrEF) compared with reduced (<40%, HFrEF) or preserved ejection fraction (≥50%, HFpEF). The key search terms used were ‘ejection‐fraction’, ‘heart‐failure’, ‘death*’, ‘mortality*’, and ‘hazard‐ratio*’. The detailed search strategy is in Supporting Information, Table S1 . To identify additional studies, we manually searched the reference lists of the included studies. This study is registered at the PROSPERO review database (CRD42021277107).

The inclusion criteria were full‐text peer reviewed papers in English, randomized controlled trials, cohort studies, observational studies with consecutive recruitment of patients with heart failure; standard case definition of heart failure, studies defining HFmrEF or borderline heart failure ejection fraction as 40%–49%, HFrEF as <40%, and HFpEF as ≥50% in adults aged 18 years and above; reporting cumulative all‐cause deaths in HFmrEF and HFrEF/HFpEF/both along with annual all‐cause, cardiovascular, non‐cardiovascular, heart failure deaths or sudden cardiac deaths; follow‐up ≥1 year; the sample size ≥100 with HFmrEF and the total sample size of heart failure patients ≥1000 when the power of the study was not listed. When there were ≥2 articles on the same group of patients, we selected the newest study best representing the cohort and the inclusion criteria.

The exclusion criteria were study protocols, brief reports, case studies, abstracts, theses, reviews, duplicate studies, incomplete and unpublished studies, studies of pregnant women, participants with a specific disease or a health condition in addition to heart failure and duplicate studies reporting outcomes from the same study population.

2.3. Data extraction

Retrieved studies were exported to CovidenceR. I. S. (Indira Samarawickrema) and D. R. (Deep Raja) screened the titles and the abstracts independently in CovidenceR for eligibility. R. P. (Rajeev Pathak) was the third independent reviewer. I. S. and D. R. independently extracted data from the eligible full‐text studies into the custom designed data extraction tables in CovidenceR. Corresponding authors of studies without data on mortality outcomes were contacted. We analysed for consensus to ensure reliability of the data. Inter‐rater agreement in data extraction was calculated. We had meetings to come to a consensus to resolve any discrepancies.

The data collated comprised study design, dataset, study period, study population, definitions of mid‐range or borderline, reduced and preserved ejection fraction in heart failure, mean follow‐up and standard deviation, study power, sample size, mean age and standard deviation, proportion of males, number of patients at baseline, number of cumulative, cardiovascular, non‐cardiovascular and sudden cardiac deaths and all‐cause deaths at 1 year intervals up to 10 years.

2.4. Quality appraisal

I. S. and D. R. independently assessed the quality of eligible used Newcastle‐Ottawa quality assessment scale (NOS) for cohort studies and Cochrane Risk of Bias tool for randomized clinical trials (RoB 2). 25 We calculated inter‐rater agreement for quality assessment and consensus reached following discussions for the disagreements. The third independent reviewer was R. P. The objective of the quality appraisal was to exclude low quality studies, that is, NOS score <7 and RoB 2 score <2. This study did not have patient or public involvement.

2.5. Statistical analysis

Pooled data of mortality and survival data for HFmrEF were compared with HFrEF and HFpEF for cumulative, cardiovascular, non‐cardiovascular, heart failure and sudden cardiac deaths; and cumulative deaths at 1, 2, 3, 4, 5, and 10 years; to determine the effect size [risk ratio (RR)] for the risk of death in the meta‐analysis. To minimize potential of biases, each meta‐analysis required ≥2 studies. Heterogeneity was calculated as I 2, and the cut‐off for low heterogeneity was I 2 < 30%. We used random effects maximum likelihood model for I 2 ≥ 30% and fixed effect model with inverse‐variance for I 2 < 30%. We reported exponentiated effect sizes from the log risk ratios. The effect size for death was determined as a risk ratio with 95% confidence intervals in the meta‐analysis.

We explored heterogeneity between studies due to publication bias with visual inspection of funnel plots and Egger's linear regression test for the influence from small studies. Sensitivity analysis excluding one study at a time was also performed to examine the influence of individual study on the effect size. We examined potential factors for between‐study heterogeneity with clinically and epidemiologically relevant covariates (moderators) in the subgroup analysis and in the random‐effects meta‐regression. The covariates included were the proportion of men below or above 50%, mean age of the participants below or above 50 years, studies recruiting outpatients versus hospitalized patients, studies recruiting de novo versus chronic heart failure patients, single versus multi‐centre studies, and follow‐up period ≤2 versus > 2 years. We conducted statistical analyses with Stata 17.0 (STATA Corporation, Texas, USA).

3. Results

The search process is outlined in Figure 1 , and the search strategy is listed in Supporting Information, Table S1 . Eighteen eligible studies were included in the meta‐analysis of the mortality outcomes between HFmrEF and HFrEF, and 17 eligible studies were included in meta‐analysis of the mortality outcomes between HFmrEF and HFpEF. All studies were follow‐up of patient cohorts from observational studies and there were no eligible randomized controlled trials. Inter‐rater reliability in data extraction was 0.91, and Cohen's kappa was 0.39. The quality of evidence of each selected study is listed in Supporting Information, Table S2 . Inter‐rater reliability in quality appraisal was 0.84, and Cohen's kappa was 0.67.

Figure 1.

Figure 1

Flow chart of search process and results.

3.1. Study and patient characteristics

There were 14 prospective and 4 retrospective cohort studies; 13 including hospitalized patients; 3 including outpatients; 14 multi‐centre and 4 single‐centre studies; 13 studies with enrolments before 2010; and 5 studies with enrolments during or after 2010 (Table  1 ). The participant follow‐up ranged from 1 to 10 years. The noteworthy large‐scale studies were the 1 and 5 year data from the Get With The Guidelines–Heart failure (GWTG‐HF) registry, 3 year data from the Swedish Heart Failure Registry, and 1 year data from the ESC Heart Failure long‐term registry. 5 , 8 , 15 , 19 Choi et al. reported outcomes for both de novo heart failure and acute decompensated chronic heart failure patients and was hence considered as two separate studies. 9

Table 1.

Characteristics of the included studies

Source a Study design Country Study period Follow‐up in years Age b (years) (mean ± SD/median ± IQR) Total sample size Number of patients
HFrEF (<40%) HFmrEF (40%–49%) HFpEF (≥50%)
No. Male % No. Male % No. Male %
Bonsu et al., 2017 Retrospective single hospital registry Ghana January 2009 to December 2013 5 60.8 ± 14.6 1488 345 48.1 265 50.2 878 43.3
Cheng et al., 2014 Prospective multi‐hospital registry USA January 2005 to December 2011 1 80 ± 74–86 40 239 15 716 60 5626 49.5 18 897 32.7
Chioncel et al., 2017 Prospective multi‐hospital registry Europe and Mediterranean countries May 2011 to April 2013 1 68.6 ± 13.7 9134 5460 78.4 2212 68.5 1462 52.1
Choi et al., 2018 de novo Prospective multi‐hospital registry South Korea March 2011 to February 2014 4 71.7 ± 14.5 2867 1631 61.8 492 50 744 42.3
Choi et al., 2018 ADHF Prospective multi‐hospital registry South Korea March 2011 to February 2014 4 72.6 ± 12.0 2547 1551 60.7 383 43.9 613 33.8
Coles et al., 2014+ Prospective multi‐hospital registry USA 1995, 2000, 2002, 2004 4 76.5 ± 11.9 3604 1479 56.5 346 45.4 1779 33.4
Farre et al., 2017 Prospective multi‐hospital registry Spain Aug 2001 to Jun 2015 10 73.5 ± 11.4 3580 2232 75.7 504 66.9 844 44
Ganapathi et al., 2020 Retrospective single hospital registry India January 2001 to December 2010 10 48.8 ± 14.7 1502 404 78.2 231 74.5 867 51.7
Gomez‐Otero et al., 2017 Prospective multi‐hospital registry Spain October 2013 to December 2014 1 72.5 ± 11.1 1420 583 76.7 227 67 610 46.7
Guisado‐Espartero et al., 2018 Prospective multi‐hospital registry Spain February 2008 to February 2009 1 81 ± 76–86 2753 808 62.5 281 58 1664 37.4
Hamatani et al., 2018 Prospective multi‐hospital registry Japan June 2005 to April 2016 5 77 ± 11 1792 860 71.2 318 64.5 614 47.4
Koh et al., 2017 Retrospective multi‐hospital registry Sweden January 2000 to December 2012 3 77 ± 11 42 061 23 402 71.2 9019 60.8 9640 45.4
Lam et al., 2018 Prospective multi‐hospital registry New Zealand, Singapore March 2010 to August 2014 2 71.5 ± 11.8 2039 1209 83.5 256 69.9 574 52.3
Liu et al., 2021+ Retrospective single hospital registry Germany July 2009 to December 2017 8 69 ± 13 2018 1067 76.3 951 74.2 ‐‐ ‐‐
Pascual‐Figal et al., 2017 Prospective multi‐hospital registry Spain April 2003 to January 2011 4 72.1 ± 12.2 3446 2351 76.8 460 73 635 42.8
Shah et al., 2017 Prospective multi‐hospital registry USA January 2005 to 30 December 2009 5 82 ± 75–87 39 982 18 398 59 3285 48.5 18 299 32.4
Shiga et al., 2019 Retrospective multi‐hospital registry Japan April 2013 to March 2014 2 81 ± 72–87 1156 412 68.2 248 53.2 496 39.5
Vergaro et al., 2019+ Prospective single hospital registry Italy January 2000 to December 2016 5 71 ± 12 2791 1539 76.3 623 71.6 629 4.5

Abbreviations: ADHF, acute decompensated heart failure patients; HFrEF, heart failure with reduced ejection fraction; HFmrEF, heart failure with mid‐range ejection fraction; HFpEF, heart failure with preserved ejection fraction.

a

Studies have been mentioned in order of first author followed by year of publication; additionally, these studies had defined HFmrEF as 41%–49%, remainder of the studies defined HFmrEF as 40%–49%.

b

Average age is presented as mean and 95% standard deviation or median with 25th and 75th interquartile ranges.

The included studies had 126 239 heart failure patients inclusive of; 64 779 patients with HFrEF (51.3%); 20 301 patients with HFmrEF (16.1%); and 41 159 patients with HFpEF (32.6%). The average age of the patients was 71.9 years. The proportion of men in the overall cohort was 60.6% with 60.5% in HFrEF, 69% in HFmrEF and 40.4% in HFpEF. The HFmrEF group had significantly reduced proportion of patients with hypertension, diabetes mellitus, kidney disease, atrial fibrillation, and coronary artery disease (9.7%, 5.2%, 3.2%, 7%, 8.1% respectively) than both HFrEF (29%, 17%, 11%, 19%, 27% respectively) and HFpEF groups (23%, 11%, 8.5%, 14.6%, 12.2% respectively) (Supporting Information, Table S3 ).

3.2. Risk of all‐cause mortality

Patients with HFmrEF had a significantly lower risk of all‐cause death than those with HFrEF (Risk Ratio (RR) = 0.92; 95% CI = 0.85–0.98; P < 0.001), using the random‐effects model (Figure  2 ). Subgroup analysis (Figure  3 ) revealed that studies recruiting >50% of men had a lower risk ratio for HFmrEF versus HFrEF (RR = 0.87; 95% CI = 0.79–0.96; P = 0.006). This translated into a higher risk of deaths with HFrEF in studies recruiting >50% of males (RR = 1.15; 95% CI = 1.04–1.26; P = 0.006). However, studies ≤50% of men had insignificant differences in the risks of deaths (HFmrEF vs. HFrEF: RR = 1.00; 95% CI = 0.99–1.03; P = 0.63). Subgroup analysis also revealed that studies including only outpatients had lesser risk of death with HFmrEF (RR = 0.78; 95% CI = 0.65–0.93; P = 0.006) than studies including only hospitalized patients (RR = 0.97; 95% CI = 0.94–1.01; P = 0.13).

Figure 2.

Figure 2

Forest plots demonstrating all‐cause deaths in (A) HFmrEF and HFrEF, and (B) HFmrEF and HFpEF. HFmrEF, heart failure with mid‐range (borderline) ejection fraction; HFrEF, heart failure with reduced ejection fraction; HFpEF, heart failure with preserved ejection fraction.

Figure 3.

Figure 3

Forest plots of sub‐group analysis of long‐term all‐cause mortality in (A) HFmrEF and HFrEF, and (B) HFmrEF and HFpEF. HFmrEF, heart failure with mid‐range (borderline) ejection fraction; HFrEF, heart failure with reduced ejection fraction; HFpEF, heart failure with preserved ejection fraction.

Patients with HFmrEF had comparable risk of all‐cause death with HFpEF (pooled RR = 1.02; 95% CI = 0.96–1.09; P = 0.53) using the random‐effects model (Figure  2 ). Subgroup analysis (Figure  3 ) revealed higher risk of deaths in HFmrEF, than HFpEF, in single‐centre studies (RR = 1.13; 95% CI = 1.03–1.24; P = 0.008) compared with multi‐centre studies (RR = 1.00; 95% CI = 0.92–1.07; P = 0.91).

There was high level of heterogeneity between the included studies for HFmrEF versus HFrEF (I 2 = 87%) as well as for HFmrEF versus HFpEF (I 2 = 80.2%). To assess publication bias, a sensitivity analysis was conducted by exclusion of individual studies and the smaller studies, and this was found to not affect the results. Meta‐regression analysis for between‐study variance in overall mortality risk ratio between HFmrEF and HFrEF identified that studies on hospitalized patients and multi‐centre studies contributed to 98.6% variance (P = 0.0001). Meta‐regression analysis for between‐study variance in overall mortality risk ratio between HFmrEF and HFpEF identified that multi‐centre studies attribute to 53.7% of variance (P = 0.005).

3.3. Annual all‐cause mortality in the follow‐up

Analysis of studies with annual all‐cause deaths showed that patients with HFmrEF had a significantly lower risk of all‐cause death than those with HFrEF at 1 year (7 studies; pooled RR = 0.84; 95% CI = 0.74–0.95; P = 0.03; I 2 = 54%), 2 years (4 studies; pooled RR = 0.86; 95% CI = 0.75–0.98; I 2 = 34%), and 3 years (2 studies; pooled RR = 0.38; 95% CI = 0.32–0.44; I 2 = 88%) (Figure  3 ). The differences were insignificant at 5 years (4 studies) and 10 years (2 studies). In contrast, patients with HFmrEF had comparable risk of all‐cause death with HFpEF at all the specified years (Figure  4 ).

Figure 4.

Figure 4

Forest plots demonstrating annual all‐cause deaths in (A) HFmrEF and HFrEF, and (B) HFmrEF and HFpEF. HFmrEF, heart failure with mid‐range (borderline) ejection fraction; HFrEF, heart failure with reduced ejection fraction; HFpEF, heart failure with preserved ejection fraction.

3.4. Cardiovascular, non‐cardiovascular deaths, heart failure, and sudden cardiac deaths

Patients with HFmrEF had significantly lower risk of cardiovascular deaths than HFrEF (pooled RR = 0.77; 95% CI = 0.65–0.92; P < 0.001; I 2 = 76%) (Figure 4 , Supporting Information, Figures S1 S3 ). Patients with HFmrEF had significantly lower risk of SCD than in HFrEF (pooled RR = 0.59; 95% CI = 0.41–0.85; P < 0.001; I 2 = 0). Patients with HFmrEF had trends of lower risk of heart failure related deaths than in HFrEF (pooled RR = 0.72; 95% CI = 0.43–1.21; P = 0.21; I 2 = 86%). However, the risk of non‐cardiovascular deaths was comparable (pooled RR = 1.20; 95% CI = 1.05–1.37; P = 0.15; I 2 = 12.7%) between HFmrEF and HFrEF groups.

In contrast, patients with HFmrEF and HFpEF had comparable risks of cardiovascular deaths (pooled RR = 1.10; 95% CI = 0.99–1.21; P = 0.38; I 2 = 6%), heart failure deaths (pooled RR = 0.93; 95% CI = 0.70–1.24; P = 0.62; I 2 = 44%), SCD (pooled RR = 1.33; 95% CI = 0.85–2.09; P = 0.27; I 2 = 23%), and non‐cardiovascular deaths (pooled RR = 1.05; 95% CI = 0.77–1.44; P = 0.76; I 2 = 77%) (Figure 5 ).

Figure 5.

Figure 5

Forest plots demonstrating cardiovascular deaths in (A) HFmrEF and HFrEF, and (B) HFmrEF and HFpEF. HFmrEF, heart failure with mid‐range (borderline) ejection fraction; HFrEF, heart failure with reduced ejection fraction; HFpEF, heart failure with preserved ejection fraction.

4. Discussion

The salient features of our meta‐analysis are (1) The risk of long‐term all‐cause mortality is reduced in patients with HFmrEF compared with HFrEF at 1, 2, and 3 years; beyond which the differences in mortality risk are insignificant; (2) the risk ratios for all‐cause deaths between HFmrEF and HFrEF are different when accounting for gender disparity, while studies recruiting > 50% men show a higher risk of deaths in the HFrEF category compared with HFmrEF, the risk ratios are indifferent in studies recruiting higher proportion of women; (3) the risk of long‐term all‐cause mortality, cardiovascular and non‐cardiovascular deaths is similar between patients with HFmrEF and HFpEF.

Our study is the largest meta‐analysis of large‐scale studies reporting long‐term mortality outcomes in HFmrEF compared with HFrEF and HFpEF. The last meta‐analysis on this topic included studies till end of April 2019 and had evaluated only the adjusted mortality ratios between these groups. 23 While two of the previous meta‐analyses reported both short‐term and long‐term mortality ratios between the groups, only one meta‐analysis reported the 1 year mortality ratios. 22 , 24 The results of our study on the cumulative all‐cause mortality outcomes between HFmrEF and HFrEF as well as HFpEF groups are consistent with the earlier meta‐analyses. 22 , 26 While HFmrEF bears a better long‐term prognosis compared with HFrEF, the outcomes seem to be similar compared with HFpEF.

Our study, in addition, explores the annual mortality outcomes in the follow‐up. Beyond 3 years, the risk differences seem to level out between HFmrEF and HFrEF. This observation could be due to either transition in LVEF from one group to the other in the follow‐up, or inadequate power of the very long‐term studies to detect differences in the deaths between the two groups. Transition of patients from one group to the other has not been reported uniformly in the registries. In a seminal study involving 4942 patients, amongst those patients with HFmrEF, 37% patients had worsened to HFrEF and 25% patients improved to HFpEF category. Also, 21% patients of HFpEF had progressive failing of LV function and shifted to HFmrEF category and 16% patients of HFrEF improved to the HFmrEF category. 27 This shows that HFmrEF is likely a heterogenous entity comprising patients with improved ejection fraction, stable left ventricular functions, and progressive LV dysfunction. Though LVEF helps to categorize these patients, the dynamic change in LVEF is an important consideration to keep in mind while interpreting the long‐term outcomes.

Gender disparity between studies and its influence on the outcomes is revealed in our analysis. Studies with more proportions of women had comparable risk of deaths between HFmrEF and HFrEF, while those with more proportion of men had higher risk of deaths in HFrEF. Studies on gender differences in incidence and prognosis of heart failure show higher risk of HFrEF, incident heart failure and deaths in men compared with higher incidence of HFpEF amongst women. 28 Hospitalization is an important event in the natural history of HF that portends poor prognosis. 29 In our analysis, the studies including only outpatients had reported lesser all‐cause mortality rates ranging from 7.6% to 28% in HFmrEF group compared with mortality rates ranging from 18% to 75% in the studies including only hospitalized patients. Predictably, our subgroup analysis showed a significantly reduced deaths in the HFmrEF compared with HFrEF only amongst the studies reporting outpatients.

An individual patient data meta‐analysis showed no differences in the risks of all‐cause mortality as well as cardiovascular deaths at LVEF ≥40%. 26 Our observation also shows no differences in absolute mortality, cardiovascular, non‐cardiovascular and sudden cardiac deaths, between the HFmrEF and HFpEF groups. Thus, this group is a large cluster of heart failure patients, wherein LVEF fails to dictate prognosis. This underlines the need to look for additional non‐cardiac and cardiac prognostic indicators like associated co‐morbidities, cardiac MRI, and arrhythmia burden. Cardiac specific investigations need to be considered beyond the conventional echocardiogram in this subset. Recent large‐scale studies have shown that cardiac MRI could identify subtle myocardial abnormalities in this subset that may portend a poor prognosis both in terms of worsening myocardial dysfunction and increase in risk of arrhythmias. 30

4.1. Limitations

(1) In spite of application of strict eligibility criteria, our meta‐analysis does bear a high level of heterogeneity between the included studies. This heterogeneity was adequately addressed statistically with appropriate sensitivity, subgroup and meta‐regression analysis. The results seem to be unaffected by any of the included studies. (2) The lack of a standard reporting protocol of long‐term outcomes in registries was evident during our systematic review, such as variable period of follow‐up; gender disparity; variable age groups; variable outcomes in single and multi‐centre studies; inclusion of both hospitalized and outpatients in a few studies; inclusion of both de novo and chronic heart failure patients in a few registers; and representation of more valvular heart diseases and myocardial infarction in a few registries. We have attempted to address this heterogeneity arising due to clinical variables in our subgroup analysis. (3) Medical therapy has improved enormously with regards to heart failure. However, due to considerable heterogeneity in the reporting of medical therapy in the included studies, we could not address this in the subgroup analysis. (4) It is possible that, the recruited studies did not have adequate power to detect significant differences in risk‐ratios in heart failure deaths and very long‐term (beyond 3 years) deaths between HFmrEF and HFrEF groups. Similarly, the single‐centre studies may not have sufficient power to detect enough deaths in the HFpEF group.

4.2. Future directives

(1) Adoption of a standard reporting guideline on long‐term outcomes with respect to the clinically relevant confounders could reduce the level of heterogeneity between studies; (2) transition in LVEF needs to be considered while analysing the outcomes in chronic heart failure; (3) additional risk‐stratification tools should be aggressively sought to understand better the similarities and differences between HFmrEF and HFpEF.

5. Conclusions

Our meta‐analysis concludes that risk of long‐term all‐cause mortality is less in patients with HFmrEF compared with HFrEF till 3 years; beyond which the differences in mortality risk were insignificant, for which transition in LVEF could play an important role. The insignificant differences in the risk‐ratio between HFmrEF and HFrEF while considering studies representing greater proportion of women, suggests that gender disparity may play a divisive role in determining outcomes. The differences in risk of long‐term all‐cause mortality were comparable between patients with HFmrEF compared with HFpEF, thus suggesting the need to explore the mortality risks with tools other than LVEF.

Conflicts of interest

Dr Sanders reports having served on the advisory board of Medtronic, Abbott Medical, Boston Scientific, CathRx and PaceMate. Dr Sanders reports that the University of Adelaide has received on his behalf lecture and/or consulting fees from Medtronic, Abbott Medical, and Boston Scientific. Dr Sanders reports that the University of Adelaide has received on his behalf research funding from Medtronic, Abbott Medical, Boston Scientific, BD, and Microport. All other authors have no disclosures. Dr Pathak reports having served on the advisory board of Medtronic, Abbott Medical, Boston Scientific. Dr Pathak reports that Canberra Heart Rhythm Foundation has received on his behalf lecture and/or consulting fees from Medtronic, Abbott Medical, Boston Scientific and Biotronik. Dr Pathak reports that Canberra Heart Rhythm Foundation has received on his behalf research funding from Medtronic, Abbott Medical, Boston Scientific, and Biotronik.

Supporting information

Figure S1: Forest plots demonstrating non‐cardiovascular deaths in (A) HFmrEF and HFrEF, and (B) HFmrEF and HFpEF; HFmrEF‐ heart failure with mid‐range (borderline) ejection fraction, HFrEF‐ heart failure with reduced ejection fraction, HFpEF‐ heart failure with preserved ejection fraction.

Figure S2: Forest plots demonstrating heart failure deaths in (A) HFmrEF and HFrEF, and (B) HFmrEF and HFpEF; HFmrEF‐ heart failure with mid‐range (borderline) ejection fraction, HFrEF‐ heart failure with reduced ejection fraction, HFpEF‐ heart failure with preserved ejection fraction.

Figure S3: Forest plots demonstrating sudden cardiac deaths in (A) HFmrEF and HFrEF, and (B) HFmrEF and HFpEF; HFmrEF‐ heart failure with mid‐range (borderline) ejection fraction, HFrEF‐ heart failure with reduced ejection fraction, HFpEF‐ heart failure with preserved ejection fraction.

Table S1: Enumeration of the detailed search strategy.

Table S2: Newcastle‐Ottawa quality assessment scale (NOS) for cohort studies.

Table S3: The distribution of patient characteristics between the three groups.

Acknowledgements

We acknowledge the support extended by Murray Turner, liaison librarian for Faculty of Health, Library, University of Canberra. Dr Sanders is supported by a Practitioner Fellowship from the National Health and Medical Research Council of Australia and by the National Heart Foundation of Australia.

Raja, D. C. , Samarawickrema, I. , Das, S. , Mehta, A. , Tuan, L. , Jain, S. , Dixit, S. , Marchlinski, F. , Abhayaratna, W. P. , Sanders, P. , and Pathak, R. K. (2022) Long‐term mortality in heart failure with mid‐range ejection fraction: systematic review and meta‐analysis. ESC Heart Failure, 9: 4088–4099. 10.1002/ehf2.14125.

Funding information: The authors declare there is no source of funding.

[Correction added on 06 September 2022, after first online publication: The first author's name has been corrected in this version.]

References

  • 1. Ammar KA, Jacobsen SJ, Mahoney DW, Kors JA, Redfield MM, Burnett JC Jr, Rodeheffer RJ. Prevalence and prognostic significance of heart failure stages: Application of the American College of Cardiology/American Heart Association heart failure staging criteria in the community. Circulation. 2007; 115: 1563–1570. [DOI] [PubMed] [Google Scholar]
  • 2. Solomon SD, Anavekar N, Skali H, McMurray JJ, Swedberg K, Yusuf S, Granger CB, Michelson EL, Wang D, Pocock S, Pfeffer MA. Candesartan in heart failure reduction in mortality I. influence of ejection fraction on cardiovascular outcomes in a broad spectrum of heart failure patients. Circulation. 2005; 112: 3738–3744. [DOI] [PubMed] [Google Scholar]
  • 3. Srivastava PK, Hsu JJ, Ziaeian B, Fonarow GC. Heart failure with mid‐range ejection fraction. Curr Heart Fail Rep. 2020; 17: 1–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4. Bonsu KO, Owusu IK, Buabeng KO, Reidpath DD, Kadirvelu A. Clinical characteristics and prognosis of patients admitted for heart failure: A 5‐year retrospective study of African patients. Int J Cardiol. 2017; 238: 128–135. [DOI] [PubMed] [Google Scholar]
  • 5. Cheng RK, Cox M, Neely ML, Heidenreich PA, Bhatt DL, Eapen ZJ, Hernandez AF, Butler J, Yancy CW, Fonarow GC. Outcomes in patients with heart failure with preserved, borderline, and reduced ejection fraction in the Medicare population. Am Heart J. 2014; 168: 721–730. [DOI] [PubMed] [Google Scholar]
  • 6. Coles AH, Fisher K, Darling C, Yarzebski J, McManus DD, Gore JM, Lessard D, Goldberg RJ. Long‐term survival for patients with acute decompensated heart failure according to ejection fraction findings. Am J Cardiol. 2014; 114: 862–868. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7. Bhambhani V, Kizer JR, Lima JAC, van der Harst P, Bahrami H, Nayor M, de Filippi CR, Enserro D, Blaha MJ, Cushman M, Wang TJ, Gansevoort RT, Fox CS, Gaggin HK, Kop WJ, Liu K, Vasan RS, Psaty BM, Lee DS, Brouwers FP, Hillege HL, Bartz TM, Benjamin EJ, Chan C, Allison M, Gardin JM, Januzzi JL Jr, Levy D, Herrington DM, van Gilst WH, Bertoni AG, Larson MG, de Boer RA, Gottdiener JS, Shah SJ, Ho JE. Predictors and outcomes of heart failure with mid‐range ejection fraction. Eur J Heart Fail. 2018; 20: 651–659. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8. Chioncel O, Lainscak M, Seferovic PM, Anker SD, Crespo‐Leiro MG, Harjola VP, Parissis J, Laroche C, Piepoli MF, Fonseca C, Mebazaa A, Lund L, Ambrosio GA, Coats AJ, Ferrari R, Ruschitzka F, Maggioni AP, Filippatos G. Epidemiology and one‐year outcomes in patients with chronic heart failure and preserved, mid‐range and reduced ejection fraction: An analysis of the ESC heart failure long‐term registry. Eur J Heart Fail. 2017; 19: 1574–1585. [DOI] [PubMed] [Google Scholar]
  • 9. Choi KH, Lee GY, Choi JO, Jeon ES, Lee HY, Cho HJ, Lee SE, Kim MS, Kim JJ, Hwang KK, Chae SC, Baek SH, Kang SM, Choi DJ, Yoo BS, Kim KH, Park HY, Cho MC, Oh BH. Outcomes of de novo and acute decompensated heart failure patients according to ejection fraction. Heart. 2018; 104: 525–532. [DOI] [PubMed] [Google Scholar]
  • 10. Farre N, Lupon J, Roig E, Gonzalez‐Costello J, Vila J, Perez S, de Antonio M, Sole‐Gonzalez E, Sanchez‐Enrique C, Moliner P, Ruiz S, Enjuanes C, Mirabet S, Bayes‐Genis A, Comin‐Colet J, Investigators G . Clinical characteristics, one‐year change in ejection fraction and long‐term outcomes in patients with heart failure with mid‐range ejection fraction: A multicentre prospective observational study in Catalonia (Spain). BMJ Open. 2017; 7: e018719. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11. Ganapathi S, Jeemon P, Krishnasankar R, Kochumoni R, Vineeth P, Mohanan Nair KK, Valaparambil AK, Harikrishnan S. Early and long‐term outcomes of decompensated heart failure patients in a tertiary‐care Centre in India. ESC Heart Fail. 2020; 7: 467–473. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12. Gomez‐Otero I, Ferrero‐Gregori A, Varela Roman A, Seijas Amigo J, Pascual‐Figal DA, Delgado Jimenez J, Alvarez‐Garcia J, Fernandez‐Aviles F, Worner Diz F, Alonso‐Pulpon L, Cinca J, Gonzalez‐Juanatey JR, Red Espanola de Insuficiencia Cardiaca r . Mid‐range ejection fraction does not permit risk stratification among patients hospitalized for heart failure. Rev Esp Cardiol (Engl Ed). 2017; 70: 338–346. [DOI] [PubMed] [Google Scholar]
  • 13. Guisado‐Espartero ME, Salamanca‐Bautista P, Aramburu‐Bodas O, Conde‐Martel A, Arias‐Jimenez JL, Llacer‐Iborra P, Davila‐Ramos MF, Cabanes‐Hernandez Y, Manzano L, Montero‐Perez‐Barquero M, group Ri. Heart failure with mid‐range ejection fraction in patients admitted to internal medicine departments: Findings from the RICA registry. Int J Cardiol. 2018; 255: 124–128. [DOI] [PubMed] [Google Scholar]
  • 14. Hamatani Y, Nagai T, Shiraishi Y, Kohsaka S, Nakai M, Nishimura K, Kohno T, Nagatomo Y, Asaumi Y, Goda A, Mizuno A, Yasuda S, Ogawa H, Yoshikawa T, Anzai T. Investigators for the WETNCP. Long‐term prognostic significance of plasma B‐type natriuretic peptide level in patients with acute heart failure with reduced, mid‐range, and preserved ejection fractions. Am J Cardiol. 2018; 121: 731–738. [DOI] [PubMed] [Google Scholar]
  • 15. Koh AS, Tay WT, Teng THK, Vedin O, Benson L, Dahlstrom U, Savarese G, Lam CSP, Lund LH. A comprehensive population‐based characterization of heart failure with mid‐range ejection fraction. Eur J Heart Fail. 2017; 19: 1624–1634. [DOI] [PubMed] [Google Scholar]
  • 16. Lam CSP, Gamble GD, Ling LH, Sim D, Leong KTG, Yeo PSD, Ong HY, Jaufeerally F, Ng TP, Cameron VA, Poppe K, Lund M, Devlin G, Troughton R, Richards AM, Doughty RN. Mortality associated with heart failure with preserved vs. reduced ejection fraction in a prospective international multi‐ethnic cohort study. Eur Heart J. 2018; 39: 1770–1780. [DOI] [PubMed] [Google Scholar]
  • 17. Liu D, Hu K, Lau K, Kiwitz T, Robitzkat K, Hammel C, Lengenfelder BD, Ertl G, Frantz S, Nordbeck P. Impact of diastolic dysfunction on outcome in heart failure patients with mid‐range or reduced ejection fraction. ESC Heart Fail. 2021; 8: 2802–2815. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18. Pascual‐Figal DA, Ferrero‐Gregori A, Gomez‐Otero I, Vazquez R, Delgado‐Jimenez J, Alvarez‐Garcia J, Gimeno‐Blanes JR, Worner‐Diz F, Bardaji A, Alonso‐Pulpon L, Gonzalez‐Juanatey JR, Cinca J. Music, groups RIr. Mid‐range left ventricular ejection fraction: Clinical profile and cause of death in ambulatory patients with chronic heart failure. Int J Cardiol. 2017; 240: 265–270. [DOI] [PubMed] [Google Scholar]
  • 19. Shah KS, Xu H, Matsouaka RA, Bhatt DL, Heidenreich PA, Hernandez AF, Devore AD, Yancy CW, Fonarow GC. Heart failure with preserved, borderline, and reduced ejection fraction: 5‐year outcomes. J Am Coll Cardiol. 2017; 70: 2476–2486. [DOI] [PubMed] [Google Scholar]
  • 20. Shiga T, Suzuki A, Haruta S, Mori F, Ota Y, Yagi M, Oka T, Tanaka H, Murasaki S, Yamauchi T, Katoh J, Hattori H, Kikuchi N, Watanabe E, Yamada Y, Haruki S, Kogure T, Suzuki T, Uetsuka Y, Hagiwara N, Investigators H‐HI. Clinical characteristics of hospitalized heart failure patients with preserved, mid‐range, and reduced ejection fractions in Japan. ESC Heart Fail. 2019; 6: 475–486. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21. Vergaro G, Ghionzoli N, Innocenti L, Taddei C, Giannoni A, Valleggi A, Borrelli C, Senni M, Passino C, Emdin M. Noncardiac versus cardiac mortality in heart failure with preserved, midrange, and reduced ejection fraction. J Am Heart Assoc. 2019; 8: e013441. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22. Altaie S, Khalife W. The prognosis of mid‐range ejection fraction heart failure: A systematic review and meta‐analysis. ESC Heart Fail. 2018; 5: 1008–1016. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23. Guo P, Dai JF, Feng C, Chen ST, Feng JP. Special prognostic phenomenon for patients with mid‐range ejection fraction heart failure: A systematic review and meta‐analysis. Chin Med J (Engl). 2020; 133: 452–461. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24. Lauritsen J, Gustafsson F, Abdulla J. Characteristics and long‐term prognosis of patients with heart failure and mid‐range ejection fraction compared with reduced and preserved ejection fraction: A systematic review and meta‐analysis. ESC Heart Fail. 2018; 5: 685–694. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25. Moher D, Shamseer L, Clarke M, Ghersi D, Liberati A, Petticrew M, Shekelle P, Stewart LA, Group P‐P . Preferred reporting items for systematic review and meta‐analysis protocols (PRISMA‐P) 2015 statement. Syst Rev. 2015; 4: 1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26. Meta‐analysis Global Group in Chronic Heart F . The survival of patients with heart failure with preserved or reduced left ventricular ejection fraction: An individual patient data meta‐analysis. Eur Heart J. 2012; 33: 1750–1757. [DOI] [PubMed] [Google Scholar]
  • 27. Savarese G, Vedin O, D'Amario D, Uijl A, Dahlstrom U, Rosano G, Lam CSP, Lund LH. Prevalence and prognostic implications of longitudinal ejection fraction change in heart failure. JACC Heart Fail. 2019; 7: 306–317. [DOI] [PubMed] [Google Scholar]
  • 28. Taylor CJ, Ordonez‐Mena JM, Jones NR, Roalfe AK, Lay‐Flurrie S, Marshall T, Hobbs FDR. National trends in heart failure mortality in men and women, United Kingdom, 2000‐2017. Eur J Heart Fail. 2021; 23: 3–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29. Loehr LR, Rosamond WD, Chang PP, Folsom AR, Chambless LE. Heart failure incidence and survival (from the atherosclerosis risk in communities study). Am J Cardiol. 2008; 101: 1016–1022. [DOI] [PubMed] [Google Scholar]
  • 30. Di Marco A, Brown PF, Bradley J, Nucifora G, Claver E, de Frutos F, Dallaglio PD, Comin‐Colet J, Anguera I, Miller CA, Schmitt M. Improved risk stratification for ventricular arrhythmias and sudden death in patients with nonischemic dilated cardiomyopathy. J Am Coll Cardiol. 2021; 77: 2890–2905. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Figure S1: Forest plots demonstrating non‐cardiovascular deaths in (A) HFmrEF and HFrEF, and (B) HFmrEF and HFpEF; HFmrEF‐ heart failure with mid‐range (borderline) ejection fraction, HFrEF‐ heart failure with reduced ejection fraction, HFpEF‐ heart failure with preserved ejection fraction.

Figure S2: Forest plots demonstrating heart failure deaths in (A) HFmrEF and HFrEF, and (B) HFmrEF and HFpEF; HFmrEF‐ heart failure with mid‐range (borderline) ejection fraction, HFrEF‐ heart failure with reduced ejection fraction, HFpEF‐ heart failure with preserved ejection fraction.

Figure S3: Forest plots demonstrating sudden cardiac deaths in (A) HFmrEF and HFrEF, and (B) HFmrEF and HFpEF; HFmrEF‐ heart failure with mid‐range (borderline) ejection fraction, HFrEF‐ heart failure with reduced ejection fraction, HFpEF‐ heart failure with preserved ejection fraction.

Table S1: Enumeration of the detailed search strategy.

Table S2: Newcastle‐Ottawa quality assessment scale (NOS) for cohort studies.

Table S3: The distribution of patient characteristics between the three groups.


Articles from ESC Heart Failure are provided here courtesy of Wiley

RESOURCES