Pharmacologic and Nutritional Therapies in Idiopathic Male Infertility: A Systematic Review and Meta‐Analysis

ABSTRACT Background Pharmacologic and nutritional therapies are prescribed for men with idiopathic infertility, but their effectiveness by intervention class remains uncertain. Objectives To evaluate these interventions on semen parameters and clinical pregnancy in men with idiopathic infertility, stratified by mechanism of action. Materials and Methods We searched PubMed, Embase, Scopus, Cochrane CENTRAL, ClinicalTrials.gov, and Google Scholar (January 2010 to January 2026) for randomized controlled trials comparing pharmacologic or nutritional interventions with placebo or no treatment in men with idiopathic infertility. Semen parameters were the primary outcomes, analyzed by mechanism of action using three‐level random‐effects models with Knapp–Hartung adjustment. Certainty of evidence was assessed using GRADE. PROSPERO: CRD420261289976. Results Forty‐one RCTs (3615 men) were included. Antioxidants were associated with higher sperm concentration (SMD: 1.09, 95% CI: 0.10 to 2.08), progressive motility (SMD: 1.11, 95% CI: 0.32 to 1.90), total motility (SMD: 0.69, 95% CI: 0.26 to 1.13), and normal morphology (SMD: 0.43, 95% CI: 0.05 to 0.82); prediction intervals crossed the null for all. PDE inhibitors showed numerically higher point estimates for concentration (SMD: 1.96, 95% CI: −2.25 to 6.17) and morphology (SMD: 1.28, 95% CI: −0.47 to 3.02); methyl donors reduced DNA fragmentation (SMD: −1.17, 95% CI: −2.82 to 0.47). Confidence intervals for these included the null. Heterogeneity was substantial ( I 2 > 80%). Hormonal agents and clinical pregnancy were synthesized narratively. Discussion Pooled estimates represent averages of diverse effects; individual patient responses remain uncertain given wide prediction intervals. Conclusion Antioxidants were associated with higher values for several semen parameters, though certainty was very low and prediction intervals were wide. PDE inhibitors and methyl donors showed numerically positive point estimates but with confidence intervals that included the null. Larger trials reporting live births are needed.