The Robotic Trade-Off in Endometriosis Surgery: Robot-Assisted Versus Conventional Laparoscopy - A Systematic Review and Meta-Analysis.
Abstract
Objective
To compare intraoperative and postoperative outcomes between robotic-assisted and conventional laparoscopy for endometriosis and explore differences by disease phenotype. DATA SOURCES Systematic search of PubMed and EMBASE (2000-2026).
Methods
OF STUDY SELECTION RCTs, prospective and retrospective cohort studies comparing robotic-assisted with conventional laparoscopy in adult women with confirmed endometriosis were included. Two reviewers screened 1,067 records; 16 studies met eligibility. Data were pooled using random-effects models. TABULATION, INTEGRATION, AND
Results
Laparoscopy was associated with significantly shorter operative time (MD -26.47 minutes; 95% CI -39.41 to -13.54; p<0.0001; I²=89%; 14 studies; 2,450 patients). No significant differences were found for intraoperative bleeding, hemoglobin loss, conversion to laparotomy, hospital stay, intraoperative complications, fistula formation, or overall postoperative complications. Postoperative pain, voiding dysfunction, and the requirement for surgical drainage could not be pooled in a meta-analysis because of the high proportion of studies reporting zero events and were therefore reported descriptively; in the single study reporting postoperative pain, the incidence was lower in the robotic arm than in the laparoscopic arm (40.5% versus 76.7%). Subgroup analysis demonstrated laparoscopy was faster across all subtypes: ovarian (MD -32.09 min; 4 studies; 533 patients), unspecified disease location (MD -24.70 min; 8 studies; 1,605 patients), and bowel endometriosis (MD -22.24 min; 2 studies; 312 patients). Sensitivity analyses confirmed stable estimates; no publication bias was found (Egger's test: bleeding p=0.103, operative time p=0.704).
Conclusion
Conventional laparoscopy was significantly faster than robotic-assisted surgery across all endometriosis subtypes, while no significant differences were found for most other perioperative outcomes. Postoperative pain could not be pooled and was reported descriptively, with a lower incidence observed in the robotic arm in the single contributing study. The choice of surgical approach should be individualized based on disease location, surgical expertise, and disease complexity, rather than assuming superiority of one technique over the other.