Phosphodiesterase type 5 inhibitors and premature ejaculation: an overview of systematic reviews/meta-analyses using the AMSTAR 2, ROBIS, and GRADE tools

Abstract

Introduction The place of phosphodiesterase type 5 inhibitors (PDE5-Is) in treating premature ejaculation (PE) remains a therapeutic challenge. Objectives (1) To summarize the evidence of the efficacy and safety of PDE5-Is from published systematic reviews/meta-analyses (SRs/MAs). (2) To evaluate the reporting, methodological quality, and evidence quality of SRs/MAs concerning PE. Methods Nine databases were searched to retrieve SRs/MAs on using PDE5-Is for PE from inception to July 2022. Methodological quality and risk of bias were assessed with the AMSTAR 2 (A Measurement Tool to Assess Systematic Reviews) and ROBIS (Risk of Bias in Systematic Reviews). GRADE criteria (Grading of Recommendations, Assessment, Development, and Evaluations) were used to assess the evidence quality. Results The literature search revealed 15 relevant SRs/MAs covering 28 primary studies (9 pairwise MAs and 6 network MAs) rated as low or very low quality and high risk of bias except 1 review that was rated as moderate quality and low risk of bias. Among the 27 outcome measures related to efficacy and safety, the quality of evidence according to GRADE criteria was low in 4 and critically low in 23. Oral PDE5-Is have demonstrated a possible benefit over placebo in lifelong PE and mixed PE. The results of pairwise and network MAs advocated that the combined use of PDE5-Is and SSRIs is of possible benefit as compared with either SSRIs or PDE5-Is alone. The total adverse effects were more frequent with PDE5-Is than placebo. Conclusion PDE5-Is are of a possible benefit than placebo in lifelong PE and mixed PE. The results favor coadministration of PDE5-Is plus SSRIs over SSRIs alone or PDE5-I monotherapy. These conclusions should be interpreted cautiously due to the low methodological quality and low quality of evidence of most available reviews. Additional higher-quality randomized controlled trials, SRs, and MAs are warranted to provide a better estimate of any effect size.