Differential effects of the phosphodiesterase inhibition in chronic heart failure depending on the echocardiographic phenotype (HFREF or HFpEF): a meta-analysis.

INTRODUCTION

According to 2015 ESC/ERS guidelines for the diagnosis and treatment of pulmonary hypertension (PH), no specific drug is currently indicated for PH related to left heart disease (PH-LDH), i.e., the one secondary to left chronic heart failure (CHF), which coincides with the group 2 of the PH classification endorsed by the above-mentioned guidelines. Indeed, adoption of therapies that specifically apply for so-called pulmonary arterial hypertension (group 1 of the PH classification) has been regarded as substantially contraindicated in patients with PH-LHD, according to current ESC/ERS guidelines. Nevertheless, based on some previous studies, phosphodiesterase 5 inhibitors (PDE5i) would seem to exert a beneficial effect in CHF patients, although the comparison between these studies shows quite inconsistent or heterogeneous findings. Thus, in order to better evaluate the effect of PDE5i therapy in CHF patients, we performed a meta-analysis of randomized controlled trials (RCTs).

EVIDENCE ACQUISITION

PubMed and EMBASE databases were searched for RCTs that compared PDE5i with placebo in CHF with reduced (HFREF) or preserved (HFpEF) left ventricular ejection fraction. The endpoints of interest were: a composite of all-cause death or hospitalization, adverse events, peak VO2, Six-Minute Walk Test (6MWT), left ventricular ejection fraction (LVEF), E/e' ratio, mean pulmonary arterial pressure (mPAP), pulmonary arterial systolic pressure (PASP) and pulmonary vascular resistance (PVR).

EVIDENCE SYNTHESIS

Fourteen studies, enrolling a total of 928 patients, were comprised in the meta-analysis. Among these, 13 were RCTs and one was a subgroup analysis. Among patients with HFREF (N.=555), a significant benefit was conferred by PDE5i against the risk of the composite endpoint of death and hospitalization (OR=0.28; 95% CI: 0.10 to 0.74; P=0.03). Furthermore, among HFREF patients, therapy with PDE5i improved peak VO2 (difference in means [MD] 3.76 mL/min/kg; 95% CI: 3.27 to 4.25), as well as the 6MWT (MD=22.7 m; 95% CI: 8.19 to 37.21) and LVEF (MD=4.30%; 95% CI: 2.18% to 6.42%). For patients with HFREF, PDE5i therapy yielded a nonsignificant decrease in mPAP, while PASP was significantly reduced (MD=-11.52 mmHg; 95% CI: -15.56 to -7.49; P<0.001). By contrast, in the RCTs of patients with HFpEF (N.=373), no benefit ensued from PDE5i use regarding all of the investigated clinical, ergospirometric or hemodynamic endpoints.

CONCLUSIONS

Therapy with PDE5i caused a statistically significant improvement of clinical outcomes, exercise capacity and pulmonary hemodynamics in patients with HFREF, but not in HFpEF. However, in view of the relatively small sample size of the HFpEF population recruited so far in the RCTs that investigated the effects of PDE5i treatment, further research in this field is required to clarify whether PDE5i are beneficial even in this subset.