Background: We performed a meta-analysis to evaluate the efficacy and safety of short-term (<= 6 months) and long-term (>6 months), regular (OaD) and on-demand (PRN) regimens of phosphodiesterase type 5 inhibitors (PDE5-Is) in treating erectile dysfunction (ED) after nerve-sparing radical prostatectomy (NSRP). Methods: We conducted a literature search in August 2016. Sources included PubMed, EMBASE, and MEDLINE databases. The main outcome was International Index of Erectile Function-Erectile Function (IIEF-EF) domain score, and the secondary outcome was treatment-emergent adverse events (TEAEs). Results: Eight articles involving 13 randomized controlled trials (RCTs) were used in this analysis: they suggested that PDE5-Is can improve the IIEF-EF distinctly in comparison with placebo in short and long term (mean difference [MD]: 2.26, 95% confidence interval [CI]: 1.45-3.08, P < 0.00001, and MD: 4.5, 95% CI: 3.6-5.4, P < 0.00001), and long-term use of PDE5-Is (>6 months) can improve the IIEF-EF distinctly in comparison with short-term use of PDE5-Is (<= 6 months) (MD: 3.9, 95% CI: 3.01-4.8, P < 0.00001). OaD of PDE5-Is significantly improved the IIEF-EF compared to placebo in short and long term (MD: 4.08, 95% CI: 3.2-4.97, P < 0.00001, and MD: 4.74, 95% CI: 3.79-5.69, P < 0.00001). No significant differences were found in IIEF-EF changes between PRN and placebo (<= 6 months) (MD: 2.64, 95% CI: -0.87 to 6.14, P=0.14), and between PRN and OaD group (<= 6 months) (MD: -0.58, 95% CI: -9.86 to 8.74, P=0.91). There were more TEAEs in PDE5-Is group in comparison with placebo (odds ratio [OR]: 1.55, 95% CI: 1.26-1.91, P < 0.0001), and TEAEs in OaD group were not significantly different from those seen in PRN group (OR: 1.05, 95% CI: 0.78-1.4, P= 0.77). Conclusion: Our meta-analysis suggests that PDE5-Is are efficient and safe for treatment of ED after NSRP, and we should choose the regular regimen for short term and regular or ondemand regimen for long term. Further high-quality RCTs are needed to validate this result.