Portal vein reconstruction using primary anastomosis or venous interposition allograft in pancreatic surgery.

OBJECTIVE: Superior mesenteric vein/portal vein (SMV/PV) resection and reconstruction during pancreatic surgery are increasingly common. Several reconstruction techniques exist. The aim of this study was to evaluate characteristics of patients and clinical outcomes for SMV/PV reconstruction using interposed cold-stored cadaveric venous allograft (AG+) or primary end-to-end anastomosis (AG-) after segmental vein resections during pancreatic surgery.

METHODS: All patients undergoing pancreatic surgery with SMV/PV resection and reconstruction from 2006 to 2015 were identified. Clinical and histopathologic outcomes as well as preoperative and postoperative radiologic findings were assessed.

RESULTS: A total of 171 patients were identified. The study included 42 and 71 patients reconstructed with AG+ and AG-, respectively. Patients in the AG+ group had longer mean operative time (506 minutes [standard deviation, 83 minutes] for AG+ vs 420 minutes [standard deviation, 91 minutes] for AG-; P < .01) and more intraoperative bleeding (median, 1000 mL [interquartile range (IQR), 650-2200 mL] for AG+ vs 600 mL [IQR, 300-1000 mL] for AG-; P < .01). Neoadjuvant therapy was administered more frequently for patients in the AG+ group (23.8% vs 8.5%; P = .02). Patients with AG+ had a longer length of tumor-vein involvement (median, 2.4 cm [IQR, 1.6-3.0 cm] for AG+ vs 1.8 cm [IQR, 1.2-2.4 cm] for AG-; P = .01), and a higher number of patients had a tumor-vein interface >180 degrees (35.7% for AG+ vs 21.1% for AG-; P = .02). There was no difference in number of patients with major complications (42.9% for AG+ vs 36.6% for AG-; P = .51) or early failure at the reconstruction site (9.5% for AG+ vs 8.5% for AG-; P = 1). A subgroup analysis of 10 patients in the AG+ group revealed the presence of donor-specific antibodies in all patients.

CONCLUSIONS: The short-term outcome of SMV/PV reconstruction with interposed cold-stored cadaveric venous allografts is comparable to that of reconstruction with primary end-to-end anastomosis. Graft rejection could be a contributing factor to severe stenosis in patients reconstructed with allograft.