Utilization of chemoradiation therapy provides strongest protective effect for avoidance of postoperative pancreatic fistula following pancreaticoduodenectomy: A NSQIP analysis

Background The utilization of neoadjuvant therapy (NAT) before performing pancreaticoduodenectomy for malignancy has been well established as a protective factor for the prevention of postoperative pancreatic fistula (POPF). However, there is a paucity of published data evaluating the specific NAT regimen that is the most protective against POPF development. We evaluated the differences between neoadjuvant chemotherapy (CT) and chemoradiation therapy (CRT) with regard to the effect on POPF rates. Methods The main and targeted pancreatectomy American College of Surgeons National Surgical Quality Improvement Program registries for 2014-2016 were retrospectively reviewed. A total of 10,665 pancreaticoduodenectomy cases were present. The primary outcome was POPF development. The factors that have previously been shown to be associated with or suspected to be associated with POPF were evaluated. The factors included NAT, sex, age, body mass index (BMI), diabetes, smoking, steroid therapy, preoperative weight loss, preoperative albumin level, perioperative blood transfusions, wound classification, American Society of Anesthesiologists classification, duct size (<3 mm, 3-6 mm, and >6 mm), gland texture (soft, intermediate, and hard), and anastomotic technique. The factors identified to be statistically significant were then used for propensity score matching to compare POPF development between the cases utilizing CT versus CRT. Results A total of 10,117 cases met the inclusion criteria. The development of POPF was significantly associated, on multivariate analysis, with a lack of NAT, male sex, higher BMI, nondiabetic status, nonsmoker status, decreased weight loss, preoperative albumin level, decreased duct size, and soft gland texture. NAT, duct size, and gland texture had the strongest associations with the development of POPF (p < .0001). The overall 1765 cases (17.45%) received NAT and the POPF rate for cases with NAT was 10.20% versus 20.10% for cases without NAT (p < .0001). A total of 1031 cases underwent CT and 734 cases underwent CRT, respectively. A total of 708 paired cases were selected for analysis based on propensity score matching. The POPF rates were 11.20% versus 3.50% for CT and CRT, respectively (p < .0001). There was no difference in the frequencies of specific POPF grades. The decreased POPF rate with CRT correlated with firmer gland texture rates. Conclusions To our knowledge, this is the largest analysis of specific NAT regimens with regard to the development of POPF following pancreaticoduodenectomy. CRT provided the strongest protective effect. That protective effect is most likely due to increased fibrosis in the pancreatic parenchyma from radiation therapy. These findings provide additional support to consider CRT over CT alone in the treatment of pancreatic cancer when NAT will be utilized.