Abstract Introduction Many erectile dysfunction (ED) patients respond inadequately to phosphodiesterase type 5 inhibitors (PDE5is). Non-pharmacological interventions (NPIs) have emerged as potential alternative or adjunct treatments for improving erectile function. Objectives This study aimed to evaluate and compare the efficacy and safety of various NPIs for ED through a systematic review and network meta-analysis. Methods Registered in PROSPERO (CRD420251083286), this NMA followed PRISMA guidelines. Eight databases were searched for RCTs through December 2025. Evaluated NPIs included Electrical Stimulation (ES), Exercise, Low-intensity Extracorporeal Shockwave Therapy (Li-ESWT), Acupuncture-Based Therapies (ABT), and Hyperbaric Oxygen Therapy (HOT). Outcomes were the International Index of Erectile Function-5 (IIEF-5), Erection Hardness Score (EHS), Peak Systolic Velocity (PSV) and Adverse Events (AEs). Risk of bias (RoB 2.0) and evidence certainty (CINeMA) were independently assessed. Results 31 RCTs (1874 participants) were included. For IIEF-5, ES + Exercise indicated the most notable improvement versus Sham/Placebo (Mean Difference MD = 6.81, 95% CI 3.50–10.12; Surface Under the Cumulative Ranking Curve SUCRA 98.1%; high-certainty evidence).ES alone also showed significant gains (MD = 2.86, 95% CI 0.86–4.85; SUCRA 82.3%; high-certainty evidence). Li-ESWT offered modest benefits (MD = 0.65, 95% CI 0.00–1.29), while HOT and other modalities ranked lower in efficacy. Regarding secondary outcomes, ES was most associated with EHS improvement (MD = 3.38), and Li-ESWT+Exercise with PSV increase (MD = 7.33). No serious AEs occurred; minor events included transient muscle strain and headache. Evidence certainty ranged from low to high. Conclusion This network meta-analysis suggest that NPIs, especially ES-based therapies, can significantly improve ED. ES combined with Exercise showed the most consistent and clinically meaningful benefits. However, the overall quality of evidence remains low due to intervention variability and small sample sizes. Further large-scale, high-quality trials are necessary to confirm the long-term efficacy and safety of these interventions across diverse patient populations.
He et al. (2026) studied this question.