Compliance matching between the host vessel and vascular grafts used for small-diameter arterial replacements is thought to be important for long-term patency. However, currently available grafts elicit fibroplastic reactions, resulting in decreasing compliance with time after implantation. Bioresorbable prostheses elicit ingrowth of myofibroblasts containing abundant contractile elements. This led us to investigate whether compliance of implanted bioresorbable prostheses decreased as a function of time and if the kinetics of change correlated with the progression of tissue ingrowth. Woven polyglactin 910 prostheses (10 mm x 4 mm i.d.) were implanted into adult NZW rabbit infrarenal aortas and replicates were harvested serially through 8 months. Control grafts were implanted, and immediately resected. Dynamic compliance was measured at 1-mm axial increments along each explant using a pulse duplicator apparatus which exposed the harvested samples to realistic pulsatile hemodynamics. Compliance was calculated for proximal, mid, and distal segments of each graft and averaged at each time point by grouping into control (zero time, n = 3), early (1-4 weeks, n = 13), and late (6-36 weeks, n = 9) explant periods. At late explant periods both proximal and distal compliance were significantly greater than mid graft compliance (p < .02 and p < .03, respectively). There was a significant increase in proximal compliance between early and late explant times (p < .01). Measured increases in mid and distal segment compliance over time did not reach statistical significance. Myofibroblast laden tissue ingrowth into the inner capsule followed macrophage phagocytosis and was nearly complete prior to the time that an increase in compliance was demonstrated. Thus since the major histologic episodes precede the change in compliance, these are not likely initiated by this biomechanical change. We hypothesize the graft resorption coupled with the ingrowth of more compliant tissue likely leads to the increased compliance of the graft material.

We evaluated the adherence of indium 111-radiolabeled endothelial cells to fibronectin-treated expanded polytetrafluoroethylene surfaces exposed to high (437 s-1) vs low (218 s-1) shear and the influence of shear on prostacyclin production. Canine jugular vein factor VIII-positive endothelial cells in passages 3 through 6 were incubated with 111Indium-oxine, and labeled cells were seeded onto fibronectin-treated expanded polytetrafluoroethylene patches. Patches with confluent cells were exposed to shear in a Weissenberg rheogoniometer for intervals ranging up to 60 minutes. Percent endothelial cell retention was determined by gamma counting of patches and media and by histologic evaluation. Prostacyclin production (tritiated radioimmunoassay of 6-keto-prostaglandin F1 alpha) was assayed on perfusing media. Results showed no differences in 6-keto-prostaglandin F1 alpha production between shear rates or time periods. Endothelial cell retention did not differ between the shear rates. Rotational shear caused persistent cell loss over time in either high- or low-shear conditions. This persistent cell loss in response to steady rotational shear differs from that in response to identical rates of pulsatile linear shear in our laboratory where cell loss approached zero after 15 minutes.