[Efficacy of sildenafil citrate in men with erectile dysfunction following bilateral nerve-sparing radical prostatectomy: systematic review and meta-analysis].

OBJECTIVE: To systematically assess the efficacy and safety of oral sildenafil citrate for post bilateral nerve-sparig radical prostatectomy (post-BNSRP) erectile dysfunction (ED). METHODS: The following keywords: sildenafil, radical prostatectomy were used to search in Medline, Pubmed, Web of Science, Cochrane Library, CNKI, WanFang Database. The title, abstract and keywords of each article were independently screened by two reviewers. Randomized controlled trials (RCT) published between 1990 and 2014 were retrieved according to our inclusion and exclusion criteria. The efficacy and safety of oral sildenafil citrate for post-BNSRP ED were systematically assessed by meta-analysis. RESULTS: Four RCTs with 320 cases were included after literature retrieval and filtering. The potency rates were 32.1% (35/109) and 11.3% (7/62) between sildenafil and placebo groups after meta-analysis and showed statistically significant differences (OR = 4.66, 95% CI: 1.79-12.11). IIEF-5 score in sildenafil group was significant higher than that in the placebo group (WMD: 4.73, 95% CI: 3.26-6.19). In subgroup meta-analysis, the potency rates in high-dose, low-dose sildenafil groups and placebo groups were 30.4% (14/46), 25.0% (10/40) and 4.5% (2/44), respectively. There were statistically significant differences between the high-dose subgroup and placebo group (OR = 9.32, 95% CI: 1.96-44.23), and the low-dose subgroup and placebo group (OR = 6.99, 95% CI: 1.43-34.22). But there was no significant difference between high-dose and low-dose subgroups (OR = 1.31, 95% CI: 0.51-3.40). CONCLUSIONS: Compared with placebo, sildenafil has considerable efficacy for erectile function rehabilitationas as a primary treatment for post-BNSRP ED. There is no significant difference between high-dose and low-dose schedule for its efficacy. However, further studies are required to optimize treatment.