Exercise training in heart failure with preserved and reduced ejection fraction: A systematic review and meta-analysis

Background<br />While exercise training (ET) is an established tool in heart failure (HF), no research to date has analysed the efficacy of ET in both preserved (HFpEF) and reduced (HFrEF) ejection fraction phenotypes across the same clinically important parameters.<br /><br />Methods<br />A comprehensive systematic search was performed to identify trials published between 1990 and May 2021. Controlled trials of adults reporting pre- and post-ET peak VO2, 6-min walk distance (6MWD), Minnesota Living with Heart Failure Questionnaire (MLHFQ), Kansas City Cardiomyopathy Questionnaire (KCCQ) and left ventricular ejection fraction (LVEF) were considered. Parameters of cardiac diastolic function, brain natriuretic peptides (BNP)/N-terminal prohormone of BNP (NTproBNP) and follow-up hospitalisation and mortality data were also analysed.<br /><br />Results<br />Ninety-three studies (11 HFpEF and 82 HFrEF) were included in the final analysis, with a pooled sample size of 11,081 participants. HFpEF analysis demonstrated significant improvements in peak VO2 (weighted mean difference: 2.333 ml·min-1·kg-1, Pfixed < 0.001), 6MWD (WMD: 35.396 m, Pfixed < 0.001), MLHFQ (WMD: − 10.932, Prandom < 0.001), KCCQ (WMD: 3.709, Pfixed = 0.037) and E/e′ (WMD: − 1.709, [95% CI] = − 2.91–0.51, Prandom = 0.005). HFrEF analysis demonstrated significant improvements in peak VO2 (WMD: 3.050 ml·min-1·kg-1, Prandom < 0.001), 6MWD (WMD: 37.299 m, Prandom < 0.001), MLHFQ (WMD: − 10.932, Prandom < 0.001), LVEF (WMD: 2.677%, Prandom = 0.002) and BNP/NTproBNP (SMD: − 1.349, Prandom < 0.001). Outcome analysis was only performed in HFrEF, which found no significant changes in hospitalisation, all-cause mortality or composite end-points.<br /><br />Conclusion<br />ET significantly improves exercise capacity and quality of life in both HFpEF and HFrEF patients. In HFpEF patients, ET significantly improved an important index of diastolic function, with significant improvements in LVEF and NTproBNP/BNP seen in HFrEF patients only. Such benefits did not translate into significantly reduced hospitalisation or mortality after short-term follow-up.