Effects of Pancreatic Fistula After Minimally Invasive and Open Pancreatoduodenectomy

Abstract

Importance: Postoperative pancreatic fistulas (POPF) are the biggest contributor to surgical morbidity and mortality after pancreatoduodenectomy. The impact of POPF could be influenced by the surgical approach.

Objective: To assess the clinical impact of POPF in patients undergoing minimally invasive pancreatoduodenectomy (MIPD) and open pancreatoduodenectomy (OPD).

Design, setting, and participants: This cohort study was conducted from 2007 to 2020 in 36 referral centers in Europe, South America, and Asia. Participants were patients with POPF (grade B/C as defined by the International Study Group of Pancreatic Surgery [ISGPS]) after MIPD and OPD (MIPD-POPF, OPD-POPF). Propensity score matching was performed in a 1:1 ratio based on the variables age (continuous), sex, body mass index (continuous), American Society of Anesthesiologists score (dichotomous), vascular involvement, neoadjuvant therapy, tumor size, malignancy, and POPF grade C. Data analysis was performed from July to October 2023.

Exposure: MIPD and OPD.

Main outcomes and measures: The primary outcome was the presence of a second clinically relevant (ISGPS grade B/C) complication (postpancreatic hemorrhage [PPH], delayed gastric emptying [DGE], bile leak, and chyle leak) besides POPF.

Results: Overall, 1130 patients with POPF were included (558 MIPD and 572 OPD), of whom 336 patients after MIPD were matched to 336 patients after OPD. The median (IQR) age was 65 (58-73) years; there were 703 males (62.2%) and 427 females (37.8%). Among patients who had MIPD-POPF, 129 patients (55%) experienced a second complication compared with 95 patients (36%) with OPD-POPF (P < .001). The rate of PPH was higher with MIPD-POPF (71 patients [21%] vs 22 patients [8.0%]; P < .001), without significant differences for DGE (65 patients [19%] vs 45 patients [16%]; P = .40), bile leak (43 patients [13%] vs 52 patients [19%]; P = .06), and chyle leak (1 patient [0.5%] vs 5 patients [1.9%]; P = .39). MIPD-POPF was associated with a longer hospital stay (median [IQR], 27 [18-38] days vs 22 [15-30] days; P < .001) and more reoperations (67 patients [21%] vs 21 patients [7%]; P < .001) but comparable in-hospital/30-day mortality (25 patients [7%] vs 7 patients [5%]; P = .31) with OPD-POPF, respectively.

Conclusions and relevance: This study found that for patients after MIPD, the presence of POPF is more frequently associated with other clinically relevant complications compared with OPD. This underscores the importance of perioperative mitigation strategies for POPF and the resulting PPH in high-risk patients.