[Meta-analysis of safety and efficacy on paraaortic lymphadenectomy in gastric cancer]

To evaluate the safety and efficacy of paraaortic lymphadenectomy in gastric cancer with meta-analysis.

Randomized controlled trial comparing D(2)/D(3) and D(2)/D(3) plus paraaortic lymphadenectomy in gastric cancer were identified by two investigators after search strategy was established. Seven randomized controlled trials with 1446 cases were included in this meta-analysis. Fixed effect model or random model was separately used to evaluate the safety and efficacy of paraaortic lymphadenectomy in gastric cancer.

Total complication rate in the D(2)/D(3) plus paraaortic lymphadenectomy group was significantly higher than that in the D(2)/D(3) group (RR 1.34, 95% CI 1.03-1.75, P=0.03), while surgery associated complication between the two groups was almost the same. Lymphorrhea and severe diarrhea was more common in the D(2)/D(3) plus paraaortic lymphadenectomy (RR 16.23 and 9.56, 95% CI 1.99-131.98 and 1.70-53.67, respectively). Anastomotic leak, pancreatic fistula, abdominal abscess, ileus, postoperative bleeding and wound infection were similar. There were no advantages in operating time, amount of blood transfusion, in-hospital stay, reoperation, recurrence and 5-year survival favoring the paraaortic lymphadenectomy group except intra-operative blood loss and rate of blood transfusion (WMD 202.95, 95% CI 116.80-289.10; RR 1.93, 95% CI 1.20-3.10).

Paraaortic lymphadenectomy is associated with increased total complication without altering postoperative recurrence and survival. Strict large-scale multicenter controlled trials with long-term follow-up are required.