Comparison of surgical versus endovascular occlusion models in pig femoral arteries.

Abstract

PURPOSE To compare an endovascular technique with a well established surgical approach to achieve long-term occlusions of large porcine arteries while preserving the integrity of periarterial tissue. METHODS The femoral arteries in 11 pigs were occluded using surgical techniques on one side and blinded stent-grafts in the contralateral vessel. Feasibility, safety, primary and long-term success, and the extent of vascularization were determined over a 3-month period by conventional angiography and histological analysis. A subgroup of animals (n=5) was treated with a locally administered plasmid coding for vascular endothelial growth factor (pVEGF165) to compare both occlusion techniques under conditions of collateral growth induction. RESULTS The primary and long-term success rates for both occlusion models were 100%. Surgical occlusion of arteries resulted in a significant amount of scar dehiscence and local groin infection compared to the endograft-occluded side. There was no significant difference in capillary densities and collateralization of periarterial areas in a comparison of the occlusion technique: the cross-sectional area of the superficial femoral artery (SFA) was 300 +/- 24 mm2 for endovascular occlusion versus 320 +/- 23 mm2 for surgical occlusion (p=0.559). In the profunda femoris artery, respective values were 418 +/- 35 and 448 +/- 18 mm2 (p=0.474). The local delivery of pVEGF165 resulted in a significant increase in collateral growth in both occlusion models with comparable neovascularization: cross-sectional SFA area increased from 310 +/- 16 to 428 +/- 13 mm2 (p<0.0001); in the PFA, the area increased from 422 +/- 19 to 658 +/- 49 mm2 (p<0.0001). CONCLUSIONS Endovascular arterial occlusions using blinded stent-grafts allow easy and safe creation of long-term occlusions. Previously described collateralization following surgical occlusions was not observed, indicating that those collaterals may be associated with wound healing rather than ischemia. The occlusion of arteries using blinded stent-grafts in pigs may therefore be an appropriate model for assessing the effects of angiogenic factors in vivo.