Glutaraldehyde-treated homologous vein graft as a vein substitute: experimental study in rabbits

Vein reconstruction using grafts may prevent sequelae of venous interruption or lesion. Autologous vein is sometimes unsuitable or absent for a vascular restoration. The aim of this study was to study glutaraldehyde-treated homologous vein graft as vein substitute and compare it with autologous vein as a substitute for a vena cava segment in rabbits.

Sixty rabbits were allocated into two groups: autologous vein graft (AG), and glutaraldehyde-treated homologous vein graft (HG). Each group was subdivided into three subgroups (N.=10) to be studied at: 24 hours, 14 days, and 28 days. The veins were treated in 0.19% glutaraldehyde, pH=7.4, for 1 hour and kept at 4 degrees C in saline with added gentamicin and amphotericin B. The animals received benzanthine penicillin on the day of graft implantation and heparin only during surgery. The grafts were implanted into the vena cava. Anastomosis was performed with interrupted sutures. Cavography was performed, after surgery, and at the time the animals were killed. Evaluation of the veins was made macroscopically and by light and scanning electron microscopy.

Fibrosis was seen around the grafts at 14 and 28 days, with no difference in intensity between the groups. Cavography performed before euthanasia of the animals showed 4 partial thrombi in AG (2 at 24 hours and 2 at 14 days), 3 in HG (2 at 24 hours and 1 on day 14), and 4 occlusive thombi in HG (3 at 14 days and 1 at 28 days). Macroscopic examination did not show any thrombus in AG. In HG, two partial thrombi were confirmed at 24 hours and three occlusive thrombi at 14 days. There was no statistical difference in relation to patency between the two groups. At 14 and 28 days, the histological sections showed intimal hyperplasia of similar intensity and variable distribution in both groups. Evaluation by electron microscopy showed at 24 hours lesion areas characterized by absence of the endothelium on the graft surface, presence of inflammatory cells, and, at some sites, presence of mural thrombi in AG and HG. Both groups at 14 and 28 days showed endothelial cells covering the lesion area on the graft surface, this covering being larger in AG than in HG.

In the studied model, both grafts behaved similarly in relation to patency and morphological characteristics. This suggests that the glutaraldehyde-treated graft can be a promising alternative for vein reconstruction, justifying further animal studies with the aim of using it in human surgery.