Paneled Saphenous Vein Grafts Compared to Internal Jugular Vein Grafts in Venous Reconstruction after Pancreaticoduodenectomy

Abstract

Background: Venous resection during pancreaticoduodenectomy for the excision of pancreatic cancer allows for a more complete resection with negative margins, which increases survival. When the resected vein is greater than 3 cm, reconstruction with an interposition graft is recommended. However, consensus regarding the optimal venous conduit has not been reached. The objective of this study is to compare outcomes between the paneled saphenous vein graft (SVG) and internal jugular vein graft (IJVG) in portomesenteric venous reconstructions after pancreaticoduodenectomy.

Method: A retrospective review was performed of patients undergoing pancreaticoduodenectomy requiring an interposition graft for venous reconstruction between 2011 and 2019. Patients were stratified based on the type of conduit used (paneled SVG or IJVG). Preoperative patient characteristics, reconstruction details, and postoperative outcomes including graft patency were recorded.

Results: During the study period, 18 patients met inclusion criteria (10 female, mean age: 63 years, age range: 41-82 years). Thirteen patients underwent reconstruction with paneled SVG and five with IJVG. Comparing SVG and IJVG groups, there were no significant differences in venous resection length, venous diameters at the resection margins, or splenic vein ligation rate. For the paneled SVG, the average length of harvested vein was 168 mm which rendered 3-paneled grafts with an average diameter of 12 mm. The time to complete the venous reconstructions did not differ between the two groups (SVG: 263+/-204 min, IJVG: 216+/-77 min, P = 0.63). There were five graft thrombosis, three in the SVG group (mean follow-up time of 17 months) and two in the IJVG group (mean follow-up time of 8 months). All but one of the graft thromboses occurred during the index hospitalization. There was one donor site seroma and wound dehiscence in the SVG group and none in the IJVG group. Hospital length of stay was longer for the IJVG group (IJVG: 15.2 days, SVG: 10.2 days, P = 0.03). However, in-hospital and late mortality did not differ between the groups.

Conclusions: Paneled SVG and IJVG are both versatile and durable conduits for venous reconstruction after pancreaticoduodenectomy, able to accommodate a wide range of venous defects. In this small series, SVG has comparable outcomes to IJVG. Paneled SVG is a suitable alternative to IJVG for portomesenteric reconstruction.