Impact of a robotic approach on hypoattenuated area formation leading to postoperative pancreatic fistula in patients after pancreatoduodenectomy.

Abstract

Background: Hypoattenuated area (HA) formation at the pancreatojejunostomy (PJ) site on contrast-enhanced computed tomography (CE-CT) is significantly associated with clinically relevant postoperative pancreatic fistula (CR-POPF) after open pancreaticoduodenectomy (PD) (O-PD). Here, we evaluated the impact of HA formation in robotic PD (R-PD) and surgical factors predictive of HA formation.

Methods: The study retrospectively analyzed 66 patients who underwent either O-PD or R-PD and exhibited a drain amylase level exceeding three times the upper limit of normal range, with CE-CT assessment performed on postoperative days 3-14. Patients were divided into two groups, with evident HA (≥ 5 mm) (E-HA) and subtle HA (< 5 mm) (S-HA), and their data were analyzed by multivariate and propensity-score matching analyses.

Results: Among the patients, 24 (36.3%) exhibited E-HA and 42 (63.7%) S-HA. The percentages of R-PD and CR-POPF in E-HA group were significantly lower and higher, respectively, than S-HA group (R-PD: 29.2% vs 54.8%, p = 0.0446; CR-POPF: 70.8% vs 4.8%, p < 0.0001). Multivariate analysis revealed the surgical approach as a significant factor associated with E-HA formation (odds ratio: 0.26; p = 0.0223). Propensity-score matching analysis revealed significantly fewer patients with E-HA formation and CR-POPF in R-PD group than O-PD group (E-HA: 14.3% vs 64.3%, p = 0.0068; CR-POPF: 14.3% vs 57.1%, p = 0.0180).

Conclusion: The impact of HA formation in predicting CR-POPF was confirmed in the patients undergoing PD, including O-PD and R-PD. Furthermore, the data suggest that R-PD, compared with O-PD, significantly decreased the incidence of E-HA formation, indicating an advantage of R-PD over O-PD in reducing CR-POPF via HA formation.