Slower Onset of Low Shear Stress Leads to Less Neointimal Thickening in Experimental Vein Grafts

Vein grafts respond to low flow and shear stress (τw) by generating thicker walls and smaller lumens through the processes of neointimal hyperplasia and remodeling. Clinically, however, vein grafts with obviously low τw, such as those distal to high-grade proximal obstructions, are not infrequently found to be widely patent and pliable. One possible explanation for this phenomenon may be that vein grafts remodel more favorably in response to changes in shear that occur gradually over time compared to abruptly. This hypothesis was tested in an experimental animal model in this report. Two separate models of experimental vein graft failure were created, causing either immediate exposure to ultralow τw (<1 dyne/cm2) or delayed exposure to ultralow τw. Under general anesthesia and using a sterile technique, the right external jugular (EJ) veins of 28 New Zealand white rabbits were surgically exposed and isolated. An end-to-side distal EJ/common carotid artery anastomosis was created, resulting in a widely patent arteriovenous fistula. For the immediate exposure group (n = 5), the EJ was suture-ligated just proximal to the thoracic inlet, distal to a small 10-50 μm venous tributary. This created a reversed vein segment immediately and abruptly exposed to high wall tension (2.0 ± 0.3 × 104 dyne/cm) and ultralow τw (0.15 ± 0.08 dyne/cm2). For the delayed exposure group (n = 22), the EJ was ligated over a 0.035 guidewire, leaving a small aperture to sustain some measure of blood flow and τw. This predictably resulted in slightly less wall tension (1.4 ± 0.2 × 104 dyne/cm) and higher τw (0.68 ± 0.21 dyne/cm2) than the immediate exposure group. During the first week, the small outflow aperture in the delayed exposure grafts thrombosed, eventually exposing them to the same low level of τw as the immediate exposure grafts. Thus, the only difference in the two models was that delayed exposure grafts enjoyed a slower decline in τw than immediate exposure grafts. Fourteen rabbits in the delayed exposure group were harvested over the first 7 days to define the patency curve of the restricted outflow channel. As expected, the small aperture had thrombosed in all animals by 7 days. The remaining 14 grafts were harvested after 4 weeks, and 13/14 remained patent. Examination of the hemodynamic parameters at the time of death confirmed that wall tension and τw had equalized (wall tension 0.9 ± 0.1 vs. 1.1 ± 0.1 × 104 dyne/cm, τw 0.45 ± 0.12 vs. 0.30 ± 0.08 dyne/cm2). Histological examination revealed less neointimal hyperplasia in the delayed exposure group compared to the immediate exposure group (wall thickness 266 ± 16 vs. 180 ± 24 μm, p = 0.025) as well as a slightly greater luminal diameter (0.30 ± 0.02 vs. 0.40 ± 0.02 cm, p = 0.038). The results of this experiment suggest that slow exposure to reduced τw results in more favorable remodeling (less thickening) than abrupt exposure. This finding may explain the occasional clinical observation of a widely patent vein graft even in the face of proximal arterial obstruction and very low flow; the change in τw presumably occurred slowly mitigating the remodeling response.