Heterotopic ossification of an expanded polytetrafluoroethylene vascular graft

Dialysis access failure: A sheep model of rapid stenosis

PURPOSE

Intimal hyperplasia at the venous anastomosis of dialysis access grafts causes early failure, although increased flow inhibits intimal hyperplasia in arterial grafts and after vessel injury. We designed a sheep model to study this process.

METHODS

Polytetrafluoroethylene (PTFE) grafts were placed in the necks of sheep from the carotid artery to the external jugular vein. Grafts were harvested after perfusion fixation at 4, 8, and 12 weeks and submitted for histologic and immunohistochemical examination, including morphometry of neointimal lesions.

RESULTS

The venous anastomoses developed thick neointima within the PTFE graft by 4 weeks. Lesions at the venous end were significantly thicker than those at the arterial end by 8 weeks (1.2 +/- 0.1 vs 0.38 +/- 0.05 mm, P <.02) and had greater cross-sectional area at both 4 (0.32 +/- 0.21 vs 3.6 +/- 0.8 mm(2), n = 7, P <.02) and 8 weeks (9.8 +/- 1.9 vs 1.1 +/- 0.7 mm(2), n = 7, P <.02). Only one of the four grafts (25%) in the 12-week group remained patent. Lesions were composed of smooth muscle cells, matrix, and thrombus of various ages. Cellular proliferation was prominent in neointima adjacent to thrombus and in granulation tissue surrounding the graft. Organizing thrombus contributed significantly to luminal narrowing.

CONCLUSION

The sheep model of dialysis access reliably produces venous stenosis within 4 weeks. Lesions develop in the absence of uremia, graft puncture, or dialysis, suggesting that these factors are not necessary for graft failure. The continued presence of thrombus and high rates of cellular proliferation suggest ongoing injury is an important cause of lesion formation. This model allows study of the cellular mechanisms of dialysis failure.

Hemodynamic and mechanical performance of arterial grafts assessed by numerical simulation: a design oriented study

The hemodynamic and mechanical characteristics of an end-to-end implanted prosthesis for a small artery were theoretically investigated. The changes in the main physical and geometrical properties of the prosthesis were simulated by means of a numerical model of arterial hemodynamics. Variation in the pressure-radius curve due to changes in lumen size, wall thickness, elasticity, and tapering were considered. The effects of such changes on pressure, flow, and wall stresses during the cardiac cycle were evaluated. To avoid superimposing the effects, only 1 of the graft properties was varied with respect to a reference condition in each simulation. A prosthesis with a reduced lumen size (20%) was subjected to higher shear stress, which was dangerously doubled. Wall thickening (200%) primarily determined decreased circumferential stress (300%) and increased wall shear stress (48%) because it caused a reduction of the graft lumen size. The time averages of flow and pressure over the cardiac cycle were not significantly influenced by the simulated changes, being imposed primarily by the proximal and distal circulations.

Bone morphogenetic protein induced repair of compartmentalized segmental diaphyseal defects

In adult rabbits, mid-diaphyseal segments of the radius or ulna were excised to produce defects greater than the critical size for spontaneous bone repair. The defects were enveloped in sleeves composed of nonbiodegradable expanded polyfluoroethylene (ePTFE), pore size 30, 60, 90 microns, and compared with sleeves of three biodegradable materials. Bone morphogenetic protein and associated noncollagenous bone matrix protein (BMP/NCP) or recombinant human morphogenetic protein (rhBMP-2) were implanted inside the sleeves. Albumin was implanted for a control system. Without intracompartmental BMP, only about 10%-15% of the defect was repaired by bone growth extending from the bone ends into the sleeves composed of ePTFE, pore size 30 microns. With sleeves with pore size 60 or 90 microns and intracompartmental BMP/NCP, 54%-96% regeneration occurred within 8 weeks after the operation. Sleeves of biodegradable nonimmunogenic materials such as polyorthoester (POE) and polylactic-polyglycollic acids (PLA/PGA) permitted 86%-98% restoration of bone continuity, but only when BMP was present in the lumen. With puncture holes (0.5 mm in diameter), implants of BMP/NCP in the 30-micron PTFE sleeve produced transmembrane external callus formation and bone regeneration to 147%. Sleeves composed of aorta first calcified, then induced complete intracompartmental bone regeneration. Atelocollagen sleeves incited a low-grade inflammatory cell reaction and did not promote complete regeneration. Under conditions presently undisclosed segments of the ulna bridged with ePTFE, were incompletely paired, even with intracompartmental BMP/NCP. Puncture holes of 0.5 mm admitted ingrowth of capillaries and introduced local conditions favorable for the response to BMP/NCP. BMP/NCP may promote proliferation of nutrient vessels and differentiation of bone marrow stroma cells between the open bone ends. For further investigation, the hypothesis to be examined is that the optimum response to BMP/NCP and rhBMP-2 would emerge in compartments containing first a high concentration gradient and second proliferating perivascular cells.

Expanded polytetrafluoroethylene arterial prostheses in humans: histopathological study of 298 surgically excised grafts

The expanded polytetrafluoroethylene vascular prosthesis is considered to be the best synthetic alternative for peripheral arterial reconstruction. Most studies on the healing characteristics of expanded polytetrafluoroethylene prostheses have been carried out on animals, and very few data are available on prosthesis implanted in humans long term. We implanted 298 expanded polytetrafluoroethylene grafts as arterial substitutes in humans. The mean duration of implantation was 523 d and the grafts were implanted mainly for infrainguinal or axillofemoral bypass. The cellular and collagen infiltration of the microporous expanded polytetrafluoroethylene structure was generally poor. Infiltration occurred mainly in the external region of the prosthetic wall and increased with the duration of implantation. The external reinforcement was not a major factor in limiting tissue infiltration. The luminal surfaces were covered with a thin, irregular layer of organized fibrin, interspersed with exposed expanded polytetrafluoroethylene areas. Mineral deposits were observed in five cases. Despite poor healing, the clinical performance of expanded polytetrafluoroethylene vascular prostheses is relatively good. Since the chief advantage of this material is good mechanical stability in vivo, any modifications of the graft to improve healing characteristics or thrombogenic properties should not be made at the expense of stability in vivo.

Expanded PTFE prostheses as arterial substitutes in humans: late pathological findings in 73 excised grafts

Through collaboration of surgeons, pathologists and bioengineers at five centers in Canada and France, this study analyzed the late pathology and structural changes in 73 expanded PTFE arterial prostheses harvested from patients at autopsies and reoperations. The degree of tissue encapsulation increased with the duration of implantation but was reduced by the presence of infection. In several cases, the fibrous tissue penetrated the wall of the prosthesis and partitioned off the thin outer layer, thus disrupting the delicate microporous structure of the wall. The presence of aneurysms was observed in models that had no external reinforcing layer and among grafts that apparently suffered from surgical trauma. Wrinkling of grafts was noted at areas of flexion and was often associated with thickening of the external capsule and reduced luminal diameters. Endothelialization was found within only a few millimeters of the anastomoses. The luminal surfaces were generally not well healed. The PTFE structure was usually readily visible under a thin covering of loosely adhering thrombotic deposits. Bacteria were observed in 46% of the cases, even though only 29% were considered clinically infected. The incidence of lipid or cholesterol deposits was high. Avoiding iatrogenic trauma to the external wall of the prosthesis during implantation is important. Those features where design improvements are required to provide longer term structural integrity and dimensional stability in future models of expanded PTFE prostheses should be identified.