Robotic pancreatoduodenectomy attenuates perianastomotic exudative changes and reduces pancreatic fistula in patients with low-grade malignancies: a retrospective cohort study

To compare the short-term outcomes of robotic pancreatoduodenectomy (RPD) and open pancreatoduodenectomy (OPD) in patients with low-grade malignancies, focusing on perianastomotic CT findings and pancreatic fistula. Ninety-four patients who underwent PD with pancreatojejunostomy reconstruction for low-grade malignancies at our institution between February 2013 and October 2025 were retrospectively analyzed (RPD n = 14, OPD n = 80). The perianastomotic Low-Density Area (LDA) on postoperative day 7 CT was the primary surrogate marker. Multivariable logistic and linear regression, propensity score matching, and inverse probability of treatment weighting were used to adjust for baseline confounders. The clinically relevant POPF (CR-POPF) rate was significantly lower in the RPD group (14.3% vs 57.5%, P = 0.003), as was major morbidity (14.3% vs 48.8%, P = 0.020). The LDA was significantly smaller in the RPD group (median 3.8 vs 10.1 mm, P < 0.001). After full multivariable adjustment for MPD diameter, pancreatic thickness, and soft pancreatic texture, RPD remained independently associated with reduced CR-POPF (adjusted odds ratio 0.089, 95% CI 0.017–0.475, P = 0.005) and reduced LDA ( β = -5.03 mm, P < 0.001). Findings were preserved in sensitivity analyses. In this retrospective cohort, RPD for low-grade malignancies was associated with attenuated perianastomotic exudative changes and a lower rate of CR-POPF compared with OPD, even after multivariable adjustment. These findings support further evaluation of the robotic approach in larger prospective studies.