Inhibition of neointimal hyperplasia in a sheep model of dialysis access failure with the bioabsorbable Vascular Wrap* paclitaxel-eluting mesh

Objective: This study evaluated the effect of a bioabsorbable mesh containing paclitaxel on neointimal hyperplasia in a sheep model of dialysis access failure. Methods. Forty neutered male sheep were randomized to one of five parallel groups: no mesh; or a 3-cm X 6-cm mesh with 0.0, 0.3, 0.7, or 1.2 mu g/mm(2) of paclitaxel for a total dose of 0.0, 0.6, 1.3, or 2.2 mg, respectively. Commercially available 6-mm internal diameter expanded polytetrafluoroethylene grafts were surgically placed between the left common carotid artery and the right external jugular vein. For those animals randomized to one of the mesh groups, the mesh was placed around the distal end of the graft and venous anastomosis. Patency was assessed at weekly intervals throughout the study. Animals were euthanized 8 weeks after implantation, and grafts and veins were harvested. After histologic processing, six cross sections were cut at the venous end of the graft and vessel. The primary and secondary efficacy outcome measures, respectively, were the area and capillary density of the neointima at the graft-vein anastomosis. Histologic analyses were also performed to investigate the effects of the paclitaxel-eluting mesh on the anastomotic site. Results. Graft's occluded before the scheduled sacrifice in five animals, and they were excluded from the study and not replaced. Control animals developed significant neointimal hyperplasia at the cross section taken perpendicular to the graft at its most distal end: the neointimal area measured 10.5 +/- 6.8 mm(2) in the no mesh group and 6.4 +/- 3.2 mm(2) in the zero-dose mesh group (P =.28). In contrast, neointimal area was significantly reduced in the paclitaxel mesh groups: 0.9 +/- 1.4 mm(2) in the 0.3 mu g/mm(2) group (P =.008 vs zero-dose mesh), 1.3 +/- 1.5 mm(2) in the 0.7 mu g/mm(2) group (P =.004 vs zero-dose mesh), and 1.2 +/- 1.4 mm(2) in the 1.2 mu g/mm(2) group (P =.008 vs zero-dose mesh). Capillary density in the neointima at the graft-vein anastomosis decreased with paclitaxel and was significantly reduced in the paclitaxel mesh groups with 0.3 and 1.2 mu g/mm(2) compared with the zero-dose mesh control (3.6 +/- 2.9 vs 8.9 +/- 5.6 per mm(2) [P=.022] and 1.1 +/- 1.7 vs 8.9 +/- 5.6 per mm(2) [P =.001] respectively). The paclitaxel mesh had no significant effect on healing of the anastomosis or on the thickness of the adjacent vein. Conclusions. In this model, the paclitaxel-eluting mesh significantly reduced neointimal hyperplasia and neointimal capillary density without apparent toxicity to the adjacent vein.