Reduced Left Jugular Venous Remodeling and Graft Patency in a Preclinical Model

Abstract

Venous remodeling, the adaptive structural and functional reorganization of the venous wall after intervention, is still not well understood. Up to 60% of arteriovenous fistulas fail to mature adequately to sustain hemodialysis as the vein fails to adequately thicken and dilate in response to arterial flow. To examine venous remodeling, we used a pig arteriovenous graft (AVG) model to expose veins to arterial flow without changing their geometry. Because humans have smaller diameter and cross-sectional area of the left internal jugular vein (IJV) compared with the right IJV, we hypothesized that left-sided AVGs may have different remodeling and patency compared with right-sided AVGs in a preclinical model. Ten Yorkshire male pigs (mean weight, 48 kg; age, 3.4 months) underwent ipsilateral or bilateral placement of AVGs from the proximal common carotid artery to the distal IJV using polytetrafluoroethylene (6 mm in diameter, 6-7 cm in length). Pigs were observed for 1 week, 2 weeks, or 3 weeks. Select pigs underwent ultrasound measurements of flow and ultrasound and caliper measurements of vessel diameters before graft placement. Vessels were harvested from two additional pigs that had undergone procedures unrelated to the IJV for control measurements. Grafts and vessels were excised and analyzed with histologic evaluation. At baseline, there was no significant difference in peak systolic or end-diastolic velocities between the left and right IJVs. The left and right IJVs also did not demonstrate any difference in caliper-measured diameters or ultrasound-measured luminal diameters. Histologic analysis of preoperative bilateral IJVs showed no difference in wall thickness or intima-media surface area. There were 10 left-sided and 8 right-sided polytetrafluoroethylene grafts placed; 4 of 10 (40%) were patent on the left and 7 of 8 (88%) were patent on the right (P = .03996, χ2 test). Postoperatively, the right IJV showed an increase in wall thickness (0.15 to 0.35 mm; P < .0001) and intima-media surface area (0.17 to 0.35 mm2; P = .0024) compared with baseline, but significant thickening was not seen on the left side. Left-sided grafts had increased luminal macrophages at the arterial anastomosis compared with right-sided grafts, but there was no significant difference in macrophage number at the venous anastomosis. Left-sided jugular veins do not thicken to the degree that right-sided veins thicken on exposure to the AVG environment. Lack of left jugular venous remodeling was associated with reduced graft patency in this preclinical model. These data suggest anatomic differences in different venous beds, emphasizing the need to understand the biology of venous remodeling to optimize graft patency.