Objective To evaluate the role of systematic lymphadenectomy (SL) vs. unsystematic lymphadenectomy (USL) for improving overall survival (OS) in epithelial ovarian cancer (EOC). Methods The databases such as PubMed, EMbase, The Cochrane Library, Evidence-Based Medicine Reviews (EBMR), CBM, CNKI and VIP were searched between January 1, 1995 and December 31, 2010, the randomized controlled trials (RCTs) and observational studies on SL vs. USL in treating EOC were included. Based on Cochrane handbook, the data were extracted, the methodological quality was assessed, and then meta-analyses were conducted by using RevMan 5.0 software. Results The total 13 studies involving 22 796 patients were included, including 5 420 patients in the SL group, and the other 17 376 patients in the USL group. Two of the 13 studies were RCTs, and the other 11 were observational studies (including 2 studies retrieved from SEER data). The analyses on 2 RCTs showed that compared with USL, a) SL could not improve 5-PFS (OR=0.70, 95%CI 0.40 to 1.22, P=0.21) in early-stage EOC (FIGO I to II), but it did improve 5-PFS (OR=0.62, 95%CI 0.40 to 0.96, P=0.03) in advanced-stage EOC (FIGO III to IV); b) SL could not improve 5-OS in both early-stage EOC (OR=0.84; 95%CI 0.44 to1.58, P=0.58) and advanced-stage EOC (OR=0.93, 95%CI 0.64 to 1.37, P=0.73); and c) SL could not improve 5-OS in both early-stage (OR=0.84, 95%CI 0.44 to 1.58, P=0.58) and advanced-stage (OR=0.93, 95%CI 0.64 to 1.37, P=0.73) of EOC patients who had optimal tumor dubulking surgery. The analyses on observational studies showed that compared with USL, a) SL could not improve 5-PFS in both early-stage EOC (OR=0.38, 95%CI 0.08 to 1.74, P=0.21) and advanced-stage (OR=2.88, 95%CI 0.95 to 8.72, P=0.06) EOC; b) Whether SEER impacts were excluded or not, SL did improve 5-OS in both early-stage EOC (OR=0.54, 95%CI 0.46 to 0.63, Plt;0.000 01) and advanced-stage (OR=0.47, 95%CI 0.43 to 0.52, Plt;0.000 01) EOC; and c) For EOC patients who had optimal tumor dubulking surgery, SL could not improve 5-OS in early-stage (OR=0.32, 95% CI 0.02 to 6.19, P=0.45), but it did improve 5-OS in advanced-stage (OR=0.53, 95%CI 0.32 to 0.88, P=0.01). Conclusion These findings suggest that maybe SL can improve 5-PFS and 5-OS in EOC. However, the efficacy of SL on 5-PFS and 5-OS is still undetermined, so more relevant studies are required for further investigating the role of SL in EOC.