BackgroundPrevious studies suggested that phosphodiesterase 5 inhibitors (PDE5i) have a beneficial effect in patients with heart failure (HF), although the results were inconsistent. We performed a meta-analysis to evaluate the effect of PDE5i in HF patients, and investigated the relationship between PDE5i effects and pulmonary hemodynamics.MethodWe searched PubMed, EMBASE and the Cochrane Library for randomized controlled trials (RCTs) that compared PDE5i with placebo in HF with reduced ejection fraction (HFrEF) or HF with preserved EF (HFpEF). PDE5i effects were interpolated according to baseline pulmonary arterial pressure (PAP) or according to changes in PAP after PDE5i treatment.ResultsThirteen RCTs enrolling 898 HF patients, and two sub-analysis studies with different study outcomes, were included in the meta-analysis. Among patients with HFrEF, PDE5i improved peak VO2 (mean difference [MD], 3.76 mL/min/kg; 95% confidence interval [CI], 3.27 to 4.25; P < 0.00001), VE/VCO2 slope (MD, −6.04; 95% CI, −7.45 to −4.64; P < 0.00001), LVEF (MD, 4.30%; 95% CI, 2.18 to 6.42; P < 0.0001), and pulmonary vascular resistance (MD, −80.74 dyn·sec/cm5; 95% CI, −110.69 to −50.79; P < 0.00001). The effects of PDE5i in patients with HFpEF were heterogeneous. Meta-regression analyses indicated that the beneficial effect of PDE5i was related to the baseline PAP as well as the extent of PDE5i-mediated PAP decrease.ConclusionPDE5i improved pulmonary hemodynamics and exercise capacity in patients with HFrEF, but not in HFpEF. The relationship between the benefits by PDE5i with the baseline PAP and the changes in PAP indicates the therapeutic potential of PDE5i in HF according to pulmonary hemodynamics.Electronic supplementary materialThe online version of this article (doi:10.1186/s12872-017-0576-4) contains supplementary material, which is available to authorized users.