Perioperative outcomes of robotic vs laparoscopic pancreatoduodenectomy: a meta-analysis and trial sequential analysis.

BACKGROUND

Minimally invasive pancreatoduodenectomy has gained widespread acceptance among hepatopancreatobiliary surgeons due to its demonstrated advantages in perioperative outcomes compared to the conventional open approach. This meta-analysis, along with trial sequential analysis, aimed to compare the outcomes of robotic pancreatoduodenectomy and laparoscopic pancreatoduodenectomy based on the current available evidence.

METHODS

A systematic search of PubMed, Cochrane, Scopus, and Web of Science was conducted from inception to July 2024. We assessed intraoperative and postoperative outcomes, including open conversion rates, readmission, reoperation, and mortality. Additional factors, such as blood loss and the number of lymph nodes resected, were also evaluated. Quality assessment was performed using Cochrane's risk of bias tools. Subgroup analysis was conducted based on propensity score matching status, and trial sequential analysis was applied to statistically significant results on first analysis.

RESULTS

We meta-analyzed seventeen studies, with a total of 5,483 patients. The meta-analysis reported lower rates of open conversion (OR 0.40; 95% CI 0.26-0.61; p = < 0.001) and a higher number of lymph nodes resected (MD 3.5; 95% CI 1.45-5.55; p = 0.0008) in the RPD group. These results remained consistent after subgroup analysis. Trial sequential analysis confirmed true positive results, indicating that further studies are unnecessary to detect a significant difference. However, other perioperative outcomes did not show statistical significance and failed to reach the required information size for definitive conclusions.

CONCLUSION

This meta-analysis of non-randomized cohorts found lower rates of open conversion and a higher number of lymph nodes resected in patients undergoing robotic pancreatoduodenectomy. However, the initial analysis did not reveal statistically significant differences in transfusion rates, multivisceral and vascular resections, hemorrhage, delayed gastric emptying, mortality, or readmission rates. Future studies should focus on randomized designs, target more specific populations, and include long-term outcomes for a more comprehensive analysis.