Eight RCTs were included (3,599 participants, range 52 to 2,516). Six trials had a Jadad score of 3 and single trials scored 1 and 2 points.
No significant differences were observed between laparoscopy and laparotomy groups in overall, disease-free or cancer-related survival (three RCTs).
Laparoscopy compared to laparotomy was associated with more intraoperative complications (OR 1.33, 95% CI 1.03 to 1.73; seven RCTs), fewer postoperative complications (OR 0.59, 95% CI 0.46 to 0.75; eight RCTs), longer operative time (SMD 0.80, 95% CI 0.46 to 1.15; four RCTs), lower blood loss (SMD -2.29, 95% CI -3.67 to -0.91; three RCTs) and shorter hospital stay (SMD -2.60, 95% CI -3.47 to -1.72; three RCTs).
No significant differences were observed between groups in pelvic or para-aortic lymph node yield (two RCTs).
There was evidence of heterogeneity for perioperative time (p=0.012), blood loss (p=0.000), para-aortic lymphadenectomy (p=0.037) and length of hospital stay (p=0.001). No evidence of publication bias was found.