Comparing the operative, oncological, post-operative outcomes and complications of robotic and laparoscopic pancreaticoduodenectomy for the treatment of pancreatic and periampullary cancers: a systematic review and meta-analysis with subgroup analysis.

Abstract

The majority of previously published meta-analyses compare robotic and laparoscopic pancreaticoduodenectomy (RPD vs. LPD) across both benign and malignant lesions. This meta-analysis aims on focusing exclusively on malignant lesions, providing a detailed and targeted evaluation of operative, oncologic, and post-operative outcomes. This systematic review and meta-analysis adhered to PRISMA 2020 guidelines and the Cochrane Handbook and followed a pre-registered protocol on PROSPERO. A comprehensive search of PubMed, Scopus, and Cochrane Library databases was conducted up to August 1, 2024. Risk of bias was performed with the ROBINS-I tool. We calculated the odds ratios and the mean differences for the dichotomous and continuous outcomes, respectively. Sensitivity analysis was conducted to evaluate the robustness of findings. Subgroup analyses were performed for pancreatic cancer cases exclusively. In total, eight studies involving 6648 patients (1964 RPD and 4684 LPD) were included. Significant outcomes included reduced length of hospitalization for RPD (MD = -0.94, P = 0.005) and lower conversion rates (OR = 0.20, P < 0.00001). In addition, the number of harvested lymph nodes was significantly higher for RPD (MD = 1.02, P = 0.01). Overall morbidity was significantly lower for RPD (OR = 0.50, P = 0.05). Non-significant differences were observed for estimated blood loss (P = 0.72), operative duration (P = 0.28), blood transfusion rates (P = 0.12), R0 resection rates (P = 0.60), major complications (P = 0.54), pancreatic fistula rates (P = 0.06), delayed gastric emptying (P = 0.58), reoperation rates (P = 0.20), and 90-day mortality (P = 0.97). Sensitivity analysis reduced heterogeneity without altering significant results, with the exception of overall morbidity which became non-significant in some cases, highlighting study-specific influences. Subgroup analysis for pancreatic cancer showed consistent findings with the main analysis, except overall morbidity, which became non-significant, suggesting that periampullary cancers may have influenced the observed benefits of RPD. Further analysis was limited by data availability. While RPD offers potential benefits, including shorter hospitalization, lower conversion rates, higher number of harvested lymph nodes and lower morbidity, the limited number of high-quality studies, study heterogeneity, and conflicting evidence with prior meta-analyses underscore the need for further well-designed trials focusing on specific patient populations to guide surgical decision-making. PROSPERO registration CRD42025634636.