As a common consequential condition of a wide variety of cardiac diseases, heart failure (HF) is a frequent burden of global medical resources. Left ventricular ejection fraction (LVEF) draws additional attention as an essential reference for the detailed diagnosis and treatment strategy of HF since it characterized the state of systolic and diastolic function to a heart; however, relationship between LVEF and prognosis of HF remains conflicting. On the basis of LVEF, previous studies have historically established two distinct entities including HF with reduced ejection fraction (HFrEF, LVEF <40%) and HF with preserved ejection fraction (HFpEF, LVEF ≥50%). While in 2013 the American College of Cardiology Foundation/American Heart Association HF guidelines introduced HF with borderline preserved ejection fraction (HFbEF, LVEF 41%–49%) as a new sub-group. Then in 2016 the conception of HF with mid-range ejection fraction (HFmrEF, LVEF 41%–49%) was proposed into a three-category classification of HF by the European Society of Cardiology HF guidelines along with HFrEF and HFpEF. Data showed that HFmrEF accounted for a proportion of 11.9% to 24.0%[4–6] of HF and was characterized by mild systolic dysfunction and diastolic dysfunction. As for biological markers of HFmrEF, Tromp's study revealed that markers related to both inflammatory responses and cardiac stretch had been frequently involved. Clinical studies focusing on HFmrEF have not yet met in agreement prognostically and therapeutically.
Therefore, we investigated studies with HFpEF, HFmrEF, and HFrEF patients to quantitatively analyze and compare the baseline characteristics and prognosis among them by systematic review and meta-analysis.
This study was designed in adherence to preferred reporting items for systematic reviews and meta-analyses requirements and registered in the PROSPERO international prospective register of systematic reviews (CRD42019133109). PubMed, Embase, and Web of Science databases were searched for studies concerning outcomes of HF patients from the inception up to 23 April 2019 without language restriction. The search terms used were as follows: Heart Failure [MeSH Terms] or Heart Diseases or Cardio-Renal Syndrome or Dyspnea, Paroxysmal or Edema, Cardiac or Heart Failure, Diastolic or Heart Failure, Systolic or HFmrEF or mid-range ejection or borderline ejection fraction or HFbEF or intermediate ejection fraction or heart failure rehospitalization or cardiovascular death or mortality. References of targeted studies and systematic reviews, meta-analyses were hand searched for relevant studies. Abstracts, meeting proceedings, and letters were excluded from this study. Initial search and assessments of titles and abstracts for considered citations were conducted by two independent reviewers. Full texts of potentially eligible studies were reviewed and discrepancies were settled by further discussion with a third reviewer when needed. Authors were contacted to provide any clarification of missing information. If a same population cohort was repeatedly reported, we retained data of study with the largest sample.
Inclusion criteria were as follows: (1) studies focusing on human; (2) observational studies (prospective or retrospective cohort studies); (3) studies reported at least one endpoint among all-cause mortality, HF rehospitalization and cardiovascular death; (4) hazard ratios (HRs) and confidence intervals (CIs) of the above endpoints were available or could be estimated.
Exclusion criteria were as follows: (1) duplicate publication data; (2) HRs have not been adjusted by multiple factors; (3) population sample size <100.
Data extraction was carried out by two independent investigators according to a pre-designed form. Terms of publication information (including first author, country of the author's affiliate, publication year, study design, sample size), demographic characteristics, clinical items (including type of HF, follow-up period, interventions, complications, endpoints, and adjusted HRs with 95% CIs) were extracted and then pooled together into the three-category groups.
Endpoints targeted for synthesis include long-term all-cause mortality (LAM), short-term all-cause mortality (SAM), long-term cardiovascular death (LCD) and long-term HF rehospitalization (LHR). Long-term endpoints were followed-up at least one year while short-term endpoint less than one year. The primary outcome was LAM. Secondary outcomes were SAM, LCD, and LHR.
Quality assessments of literature were conducted independently by two reviewers with discrepancies properly resolved. Evaluation of risk of bias for included studies was performed using Newcastle-Ottawa scale (NOS) with results displayed in Supplementary Table 1, http://links.lww.com/CM9/A166. In our research, studies that achieved five or more stars on the modified NOS were considered high quality.
Baseline data were pooled and expressed as either mean ± standard deviation (SD) for continuous variables or simple summation and proportion for categorical variables. The weighted mean difference method was applied to obtain means and their SDs. Baseline data were tested by Student's test or Chi-squared test. HRs and their 95% CIs were used to evaluate risks of different endpoints between HF sub-groups. Transformation of HR was achieved by Hamling conversion formula (http://www.pnlee.co.uk/software.htm) when the reference group of HR was different among studies. Random-effects model was applied for all meta-analyses. Heterogeneity across studies was examined by the Cochran Q test and I2 statistic value. Sources of heterogeneity were explored by subgroup analysis and sensitivity analysis When I2 > 50% with over eight studies. Sub-group analyses were established and interaction for each subgroup was evaluated by random-effects analysis. In sensitivity analysis, the influence of every single study on overall estimates was assessed. Publication bias and selective reporting were investigated firstly by funnel plot and then Egger's and Begg's tests to detect statistical significance.
Statistical analyses were performed with either Stata (version 15.0, StataCorp, College Station, TX, USA), Reviewer Manager (RevMan, Version 5.3, The Cochrane Collaboration, Copenhagen, Denmark), and Excel (version 2016). A two-tailed P value <0.05 was considered statistically significant.
Literature search strategy
The initial search identified 1085 records, of which 193 duplicates and 831 irrelevant papers based on titles and abstracts were excluded. Detailed retrieving of the remained 61 full-text articles eventually yielded 19 studies eligible for this study [Figure 1].[4,6,11–27] Information of the selected studies was listed in Table 1. All included studies were of high quality, as indicated by individual NOS scores ranging from 5 to 8.
A total of 164,678 patients were enrolled in this study, including 63,998 HFpEF patients, 26,614 HFmrEF patients, and 74,066 HFrEF patients. Baseline information including demographic and clinical features and endpoints is shown in Table 2. Follow-up time was 3.6 ± 2.5 years for long-term endpoints while 30 days for short-term endpoints. Rates of the four endpoints named LAM, SAM, LCD, and LHR in HFpEF patients were higher than HFmrEF group but lower than HFrEF by the end of follow-up [Figure 2].
Fifteen studies with 149,659 patients involving reported HRs of LAM among different HF groups. All studies but four,[14,16,22,25] applied transformation of HR. Pooled data significantly indicated that the types of HF were independently associated with LAM. The risk of HFmrEF patients was increased compared with HFpEF patients (reference) with HR: 1.07, 95% CI: 1.00 to 1.15, I2 = 63%, P = 0.0005 [Figure 3A], while reduced compared with HFrEF patients (reference) with HR: 0.80, 95% CI: 0.73 to 0.88, I2 = 70%, P < 0.0001 [Figure 3B]. Subsequent analyses were conducted since notable heterogeneity appeared among different studies.
Sub-group analyses based on sample size, scale, follow-up time, publication year, and fund support were performed to explore impact on heterogeneity [results are shown in Table 3 and Supplementary Figures 1–6, http://links.lww.com/CM9/A166]. Scale was outlined to be a source of heterogeneity in comparison of HFrEF vs. HFmrEF (reference): in multi-center studies HR: 0.83, 95% CI: 0.76 to 0.91, I2 = 71%, P < 0.0001; in single-center studies HR: 0.60, 95% CI: 0.46 to 0.79, I2 = 11%, P = 0.33. It was statistically significant in interaction test (P = 0.03). Varied conditions among countries, ethnicities, environment, and medical conditions in multi-center studies may be a conspicuous source of heterogeneity. The remaining sub-groups were proofed innocent in heterogeneity.
Further sensitivity analyses [Supplementary Tables 2–3, http://links.lww.com/CM9/A166 and Supplementary Figure 7, http://links.lww.com/CM9/A166] demonstrated that Siontis's study had driven high heterogeneity in the comparison of HFpEF vs. HFmrEF. The heterogeneity was significantly reduced after removing this study (I2 = 44%, P = 0.04). In Siontis's article, the disparity of ages and proportions of diabetes between patients with HFmrEF and HFpEF was significantly narrowed compared with that in our meta-analysis, thus the risk of LAM in HFmrEF patients was relatively high, making it a possible source of heterogeneity. However, meta-analyses in HFmrEF vs. HFpEF group had not changed much before (HR: 1.07, 95% CI: 1.00–1.15) and after the rejection (HR: 1.04, 95% CI: 0.98–1.10), which manifested the stability of the original synthesis. No source of heterogeneity was confirmed in HFrEF vs. HFmrEF group. Considering that HR transformation may cause minor errors, we applied sensitivity analysis by excluding studies using this method [Supplementary Figure 8, http://links.lww.com/CM9/A166]. Results showed only a little difference of the HRs and I2 after removing the studies using HR transformation but still in accordance with our primary outcome.
Egger's test and Begg's test [Supplementary Table 4, http://links.lww.com/CM9/A166] confirmed no statistically significant publication bias existed in the analyses performed although Funnel plot asymmetry were detected visually [Figure 4].
Data of LHR from five studies (one did not apply HR transformation) were pooled together [Supplementary Figure 9, http://links.lww.com/CM9/A166]. HFmrEF patients had a 7% higher rate of LHR (HR: 1.07, 95% CI: 0.95–1.21) compared with HFpEF patients, but a 19% rate lower than HFrEF patients (HR: 0.81, 95% CI: 0.73–0.91). Four studies of which two[14,22] did not use HR transformation reported the risk of LCD with results of meta-analyses showed (HR: 1.39, 95% CI: 1.08–1.79) for HFmrEF vs. HFpEF while (HR: 0.71, 95% CI: 0.54–0.94) for HFmrEF vs. HFrEF [Supplementary Figure 10, http://links.lww.com/CM9/A166]. Test of heterogeneity and publication bias were not performed because of limitations of study numbers, although considerable heterogeneity had presented in the mentioned meta-analyses above. Pooling analyses were implemented for SAM using available information from seven studies (HR transformation were not applied in three studies[17,20,22]) one of which only provided data of HFmrEF vs. HFpEF group [Supplementary Figure 11, http://links.lww.com/CM9/A166]. HFmrEF patients still run a higher risk of SAM than HFpEF patients (HR: 1.13, 95% CI: 0.92–1.38) but a lower rate than HFrEF patients (HR: 0.74, 95% CI: 0.62–0.88). Satisfying syntheses were achieved owing to a low heterogeneity (I2). The number of involving studies was too small to detect publication bias.
This systematic review and meta-analysis investigated risks of endpoints including mortality and re-admission among patients with HFmrEF, HFmrEF, and HFmrEF using adjusted HRs as indicators. Baseline information of this study showed that HFmrEF was a unique subtype distinct from HFpEF and HFrEF since meta-analysis confirmed its distinctive characteristics including the lowest rate of New York Heart Association class (NYHA) III–IV, the lest use of digoxin, and the highest application of percutaneous coronary intervention (PCI). The rates of endpoint events were lowest in HFmrEF patients, followed by HFpEF patients, and highest in HFrEF patients, however, HRs of poor prognosis after multivariable analysis increased successively by HFpEF, HFmrEF, and then HFrEF.
Evidences varied in prognosis of HFmrEF patients. Two meta-analyses, which were quite different from our findings,[28,29] summarized that HFmrEF patients had the lowest relative risk (RR) of all-cause mortality and cardiac deaths. Besides, the indicator RR could not assess and prove the impact of time and other confounding factors on the results. Therefore, we applied multivariable adjusted HR as indicator in this study and obtained results that the risks of LAM, SAM, LCD, and LHR in HFmrEF patients were higher than that of HFpEF but lower than HFrEF, while rates of the mentioned endpoints in HFmrEF patients were the lowest. We named the inconsistency between the risks and rates of the endpoints as “separation phenomenon,” which may partly because of the complexity of patient population and the diversity of complication in HFpEF group. Confounding factors that increase risks of endpoints including advanced age, renal insufficiency, and female sex, were all calibrated by the COX regression model, then the risk of HFmrEF highlighted. We also detected detailed source of heterogeneity of LAM. Results showed that study scale might be a potential source since multi-center studies involve more different countries, ethnicities, environment, and medical conditions than the singles; then Siontis study should be mentioned as well, because the disparity of ages and proportions of diabetes between patients with HFmrEF and HFpEF was significantly narrowed compared with that in our meta-analysis. The “separation phenomenon” unveiled the significance of HFmrEF and promoted the individual management for different types of HF in clinical work. For patients with HFmrEF, more aggressive cardiovascular-related treatments should be taken to improve their prognosis. While for HFpEF patients, treatment of complications and other chronic diseases should never be ignored, patients may benefit more from comprehensive treatment.
LVEF is a dynamic indicator intensively associated with cardiac function and risks of adverse outcomes. HFmrEF is an independent but unstable subtype with a changeful LVEF, it resembles to the other two types in some features and could easily convert to them. HFmrEF in female patients or in those without ischemic heart disease (IHD) were more likely to convert to HFpEF, while to HFrEF in patients with IHD. Some studies even stated that HFmrEF was the early stage of HFrEF patients in those with IHD. We also found proportion of IHD in patients with HFmrEF was similar to that of HFrEF but significantly higher than HFpEF patients. Conversion of HFmrEF to HFpEF was reported more common than to HFrEF; however, the latter would gradually increase with the growing of IHD as data from studies of Yamamoto et al, Gwag et al, Tsuji et al, and Vedin et al exemplified. This interrelated incremental relationship was more thoroughly revealed by Vedin and colleagues’ study in which the proportion of IHD in the HFmrEF cohort was as high as 60.7%, therefore, the conversion to HFrEF was higher than to HFpEF (36.5% vs. 23.6%) in his study. Accordingly, we consider that IHD plays a vital role in the conversion of HFmrEF, and will eventually affect prognosis of HF patients. As bewritten by Savarese et al, that conversion from HFmrEF to HFpEF might reflect recovery after myocardial infarction, while downward conversion to HFrEF might indicate progressive HF or a new ischemic event. Therefore, additional attention should be paid to the history and recurrence of IHD in HFmrEF patients, and relatively aggressive treatments were recommended to prevent conversion to HFrEF if IHD was involved. Further studies are urgently required since the improvement, maintenance and deterioration of LVEF in HFmrEF patients remain inconclusive.
Patients with NYHA III–IV in HFmrEF group were lower than that in HFpEF group as depicted in this study (35.68% vs. 36.70%, P = 0.043), suggesting heavier symptoms of HFpEF patients in spite of preserved ejection fractions. Yet heavier symptoms may also due to an older population, various comorbidities especially the highest proportion of COPD, of which clinical manifestations may interfere with the judgment of NYHA. Given the inconsistency between LVEF, symptoms and other HF assessment scales, the value of LVEF on evaluating cardiac function should be taken with caution. We also found that HFmrEF group used less digoxin than HFpEF group, although the latter had higher LVEF. The high proportion of atrial fibrillation or flutter in HFpEF group may explain this phenomenon because digoxin is also a kind of arrhythmia drugs controlling ventricular rate in patients with atrial fibrillation or flutter. Besides, the implementation of PCI was significantly more in HFmrEF group than in HFrEF group, although the latter companied with a higher IHD proportion. This may because of the favorable applying of CABG therapy or conservative treatment in HFrEF patients as many of them were too ill to accept PCI whereas most HFmrEF patients with IHD could withstand it.
Limitations should not be neglected alongside the results presented in this review. First, all studies included were of an observational nature which was highly subject to selection bias. Second, follow-up spans varied from 1 year to 5 years across studies reporting LAM, which may generate inconsistency of results. Furthermore, heterogeneity of LCD and LHR were high yet hard to discern due to the relatively small number of studies. And some HRs in this study were obtained by conversion which may cause minor errors.
In conclusion, distinctive characteristics especially “separation phenomenon” highlighted the remarkable significance of the new classification of HFmrEF. Further exploration is eagerly expected both in clinical management and prognosis of HFmrEF.
This study was supported by grants from the China Cardiovascular Association-Cardiac Rehabilitation and Metabolic Therapy Research Fund (No. CCA-CRMT-1805) and the Science and Technology Funding of Tianjin Chest Hospital (No. 2018XKZ17).
Conflicts of interest
1. Potter E, Marwick TH. Assessment of left ventricular function by echocardiography: the case for routinely adding global longitudinal strain to ejection fraction
. JACC Cardiovasc Imaging
2018; 11:260–274. doi: 10.1016/j.jcmg.2017.11.017.
2. Yancy CW, Jessup M, Bozkurt B, Butler J, Casey DE Jr, Drazner MH, et al. 2013 ACCF/AHA guideline for the management of heart failure
: executive summary: a report of the American College of Cardiology Foundation/American Heart Association Task Force on practice guidelines. Circulation
2013; 128:1810–1852. doi: 10.1161/CIR.0b013e31829e8807.
3. Ponikowski P, Voors AA, Anker SD, Bueno H, Cleland JG, Coats AJ, et al. 2016 ESC guidelines for the diagnosis and treatment of acute and chronic heart failure
: The Task Force for the diagnosis and treatment of acute and chronic heart failure
of the European Society of Cardiology (ESC). Developed with the special contribution of the Heart Failure
Association (HFA) of the ESC. Eur J Heart Fail
2016; 18:891–975. doi: 10.1002/ejhf.592.
4. Lam C, Gamble GD, Ling LH, Sim D, Leong K, Yeo P, et al. 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: 10.1093/eurheartj/ehy005.
5. Vedin O, Lam C, Koh AS, Benson L, Teng T, Tay WT, et al. Significance of ischemic heart disease in patients with heart failure
and preserved, midrange, and reduced ejection fraction
: a nationwide cohort study. Circ Heart Fail
2017; 10:e003875doi: 10.1161/CIRCHEARTFAILURE.117.003875.
6. Toma M, Ezekowitz JA, Bakal JA, O’Connor CM, Hernandez AF, Sardar MR, et al. The relationship between left ventricular ejection fraction
and mortality in patients with acute heart failure
: insights from the ASCEND-HF trial. Eur J Heart Fail
2014; 16:334–341. doi: 10.1002/ejhf.19.
7. Tromp J, Khan M, Mentz RJ, O’Connor CM, Metra M, Dittrich HC, et al. Biomarker profiles of acute heart failure
patients with a mid-range ejection fraction
. JACC Heart Fail
2017; 5:507–517. doi: 10.1016/j.jchf.2017.04.007.
8. Moher D, Liberati A, Tetzlaff J, Altman DG. PRISMA Group. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. Int J Surg
2010; 8:336–341. doi: 10.1016/j.ijsu.2010.02.007.
9. Hamling J, Lee P, Weitkunat R, Ambühl M. Facilitating meta-analyses by deriving relative effect and precision estimates for alternative comparisons from a set of estimates presented by exposure level or disease category. Stat Med
2008; 27:954–970. doi: 10.1002/sim.3013.
10. Badhiwala JH, Nassiri F, Alhazzani W, Selim MH, Farrokhyar F, Spears J, et al. Endovascular thrombectomy for acute ischemic stroke: a meta-analysis
2015; 314:1832–1843. doi: 10.1001/jama.2015.13767.
11. Borovac JA, Novak K, Bozic J, Glavas D. The midrange left ventricular ejection fraction
(LVEF) is associated with higher all-cause mortality during the 1-year follow-up compared to preserved LVEF among real-world patients with acute heart failure
: a single-center propensity score-matched analysis. Heart Vessels
2019; 34:268–278. doi: 10.1007/s00380-018-1249-7.
12. Margolis G, Khoury S, Ben-Shoshan J, Letourneau-Shesaf S, Flint N, Keren G, et al. Prognostic implications of mid-range
left ventricular ejection fraction
on patients presenting with ST-segment elevation myocardial infarction. Am J Cardiol
2017; 120:186–190. doi: 10.1016/j.amjcard.2017.04.005.
13. Choi KH, Lee GY, Choi JO, Jeon ES, Lee HY, Cho HJ, et al. Outcomes of de novo and acute decompensated heart failure
patients according to ejection fraction
2018; 104:525–532. doi: 10.1136/heartjnl-2017-311813.
14. Pascual-Figal DA, Ferrero-Gregori A, Gomez-Otero I, Vazquez R, Delgado-Jimenez J, Alvarez-Garcia J, et al. 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: 10.1016/j.ijcard.2017.03.032.
15. Guisado-Espartero ME, Salamanca-Bautista P, Aramburu-Bodas Ó, Conde-Martel A, Arias-Jiménez JL, Llàcer-Iborra P, et al. 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: 10.1016/j.ijcard.2017.07.101.
16. Shah KS, Xu H, Matsouaka RA, Bhatt DL, Heidenreich PA, Hernandez AF, et al. Heart failure
with preserved, borderline, and reduced ejection fraction
: 5-year outcomes. J Am Coll Cardiol
2017; 70:2476–2486. doi: 10.1016/j.jacc.2017.08.074.
17. Miró Ò, Javaloyes P, Gil V, Martín-Sánchez FJ, Jacob J, Herrero P, et al. Comparative analysis of short-term outcomes of patients with heart failure
with a mid-range ejection fraction
after acute decompensation. Am J Cardiol
2019; 123:84–92. doi: 10.1016/j.amjcard.2018.09.021.
18. Farré N, Lupon J, Roig E, Gonzalez-Costello J, Vila J, Perez S, et al. 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:e018719doi: 10.1136/bmjopen-2017-018719.
19. Shiga T, Suzuki A, Haruta S, Mori F, Ota Y, Yagi M, et al. 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: 10.1002/ehf2.12418.
20. Farmakis D, Simitsis P, Bistola V, Triposkiadis F, Ikonomidis I, Katsanos S, et al. Acute heart failure
left ventricular ejection fraction
: clinical profile, in-hospital management, and short-term outcome. Clin Res Cardiol
2017; 106:359–368. doi: 10.1007/s00392-016-1063-0.
21. Koh AS, Tay WT, Teng T, Vedin O, Benson L, Dahlstrom U, et al. A comprehensive population-based characterization of heart failure
with mid-range ejection fraction
. Eur J Heart Fail
2017; 19:1624–1634. doi: 10.1002/ejhf.945.
22. Siontis GC, Branca M, Serruys P, Silber S, Räber L, Pilgrim T, et al. Impact of left ventricular function on clinical outcomes among patients with coronary artery disease. Eur J Prev Cardiol
2019; 26:1273–1284. doi: 10.1177/2047487319841939.
23. Lund LH, Claggett B, Liu J, Lam CS, Jhund PS, Rosano GM, et al. Heart failure
with mid-range ejection fraction
in CHARM: characteristics, outcomes and effect of candesartan across the entire ejection fraction
spectrum. Eur J Heart Fail
2018; 20:1230–1239. doi: 10.1002/ejhf.1149.
24. Avula HR, Leong TK, Lee KK, Sung SH, Go AS. Long-term outcomes of adults with heart failure
by left ventricular systolic function status. Am J Cardiol
2018; 122:1008–1016. doi: 10.1016/j.amjcard.2018.05.036.
25. Miro O, Javaloyes P, Gil V, Jacob J, Herrero-Puente P, Martin-Sanchez FJ, et al. Mortality after an episode of acute heart failure
in a cohort of patients with intermediate ventricular function: global analysis and relationship with admission department. Med Clin (Barc)
2018; 151:223–230. doi: 10.1016/j.medcli.2017.11.025.
26. Wang K, Li HL, Bei WJ, Guo XS, Chen SQ, Islam S, et al. Association of left ventricular ejection fraction
with contrast-induced nephropathy and mortality following coronary angiography or intervention in patients with heart failure
. Ther Clin Risk Manag
2017; 13:887–895. doi: 10.2147/TCRM.S137654.
27. Delepaul B, Robin G, Delmas C, Moine T, Blanc A, Fournier P, et al. Who are patients classified within the new terminology of heart failure
from the 2016 ESC guidelines. ESC Heart Fail
2017; 4:99–104. doi: 10.1002/ehf2.12131.
28. 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: 10.1002/ehf2.12353.
29. 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: 10.1002/ehf2.12283.
30. Savarese G, Vedin O, D’Amario D, Uijl A, Dahlström U, Rosano G, et al. Prevalence and prognostic implications of longitudinal ejection fraction
change in heart failure
. JACC Heart Fail
2019; 7:306–317. doi: 10.1016/j.jchf.2018.11.019.
31. Lupón J, Díez-López C, de Antonio M, Domingo M, Zamora E, Moliner P, et al. Recovered heart failure
with reduced ejection fraction
and outcomes: a prospective study. Eur J Heart Fail
2017; 19:1615–1623. doi: 10.1002/ejhf.824.
32. Hwang SJ, Melenovsky V, Borlaug BA. Implications of coronary artery disease in heart failure
with preserved ejection fraction
. J Am Coll Cardiol
2014; 63:2817–2827. doi: 10.1016/j.jacc.2014.03.034.
33. Yamamoto M, Seo Y, Ishizu T, Nishi I, Hamada-Harimura Y, Machino-Ohtsuka T, et al. Different impact of changes in left ventricular ejection fraction
between heart failure
classifications in patients with acute decompensated heart failure
. Circ J
2019; 83:584–594. doi: 10.1253/circj.CJ-18-0881.
34. Gwag HB, Lee GY, Choi JO, Lee HY, Kim JJ, Hwang KK, et al. Fate of acute heart failure
patients with mid-range ejection fraction
. Circ J
2018; 82:2071–2078. doi: 10.1253/circj.CJ-17-1389.
35. Tsuji K, Sakata Y, Nochioka K, Miura M, Yamauchi T, Onose T, et al. Characterization of heart failure
patients with mid-range
left ventricular ejection fraction
-a report from the CHART-2 study. Eur J Heart Fail
2017; 19:1258–1269. doi: 10.1002/ejhf.807.
36. Triposkiadis F, Butler J, Abboud FM, Armstrong PW, Adamopoulos S, Atherton JJ, et al. The continuous heart failure
spectrum: moving beyond an ejection fraction
classification. Eur Heart J
2019; 40:2155–2163. doi: 10.1093/eurheartj/ehz158.