Aims: We assessed the clinical characteristics and prognosis of chronic heart failure patients with improved ejection fraction (HFIEF) compared with persistently reduced ejection fraction (HFpREF) after evidence-based therapy. Methods and results: We performed a meta-analysis including 24 eligible observational studies comparing 2663 HFIEF (≥5% left ventricular ejection fraction (LVEF) improvement) versus 8355 HFpREF patients who received recommended drug therapy, cardiac resynchronization therapy and/or intracardiac defibrillator. LVEF was assessed at baseline and reassessed after 19 ± 19 months. The primary endpoints were all-cause mortality and appropriate shocks. The mean duration of follow-up was 39 ± 12 months. Among HFIEF patients, LVEF improved 16.3 percentage points (95% confidence interval 15.9–16.6, p < 0.0001). Compared with HFpREF patients, HFIEF patients had a comparable mean age (60.9 years vs. 62.4 years, p = 0.11), were more often women (33% vs. 25%), had a higher prevalence of non-ischaemic heart failure (58% vs. 53%), less diabetes (27% vs. 28%), higher systolic blood pressure (127.5 ± 9 vs. 122 ± 12 mmHg) and lower left ventricle end-diastolic diameter (64.1 ± 3.7 vs. 67.4 ± 4.9 mmHg), all p-values < 0.05. Absolute risk of all-cause mortality was lower in HFIEF (5.8%) compared with HFpREF (17.5%) with a risk ratio of 0.34 (95% confidence interval 0.28-0.41), p < 0.001. Risk of appropriate shocks was significantly lower in HFIEF versus HFpREF (risk ratio 0.58 (95% confidence interval 0.46–0.74), p < 0.001). Conclusion: In heart failure patients, we identified several baseline characteristics in favour of an improved LVEF, in response to evidence based therapy. Patients with improved LVEF had significantly lower risks of mortality and appropriate shocks compared with patients with persistently reduced LVEF.