|
|
| Journal of Vascular Access 2003; 4: 111 - 117 |
| A porcine model of intimal-medial hyperplasia in polytetrafluoroethylene arteriovenous grafts |
K. Baig1, R.C. Fields 1, J. Gaca 1, S. Hanish 1, L.G. Milton 1, W.J. Koch 1, J.H. Lawson2
1Department of Surgery, Duke University Medical Center, Durham USA
2Department Pathology, Duke University Medical Center, Durham USA
|
|
|
|
Search Medline for articles by:
 K. Baig
R.C. Fields
J. Gaca
S. Hanish
L.G. Milton
W.J. Koch
J.H. Lawson
|
|
 Printable Format (PDF)
|
|
|
|
|
|
|
ABSTRACT
PURPOSE: Vascular access polytetrafluoroethylene (PTFE) graft failure is a major cause of morbidity in the hemodialysis population. The most common cause of graft failure is thrombosis secondary to stenosis at the venous outflow tract. Venous outflow stenosis is characterized by intimal-medial hyperplasia. We have developed a porcine arteriovenous (AV) graft model that may be used to investigate this proliferative response and aid in the development of new therapies to prevent intimal-medial hyperplasia and improve graft patency.
METHODS: Left carotid to right external jugular vein PTFE (6 mm) grafts were implanted in the necks of swine. Immediately following anatomosis, flow rates were recorded. In one group of animals (n = 4) the venous outflow tract was harvested after 7 days and morphometric analysis of intimal and medial area was performed. In a second group (n = 8) the graft patency was monitored until 28 days.
RESULTS: All porcine PTFE fistula grafts were patent at 7 days and 100% patency was maintained until 14 days. After 28 days, 75% of the grafts failed due to thrombosis. The venous outflow tract developed a significant proliferative response. After 7 days the intimal and medial areas were 469 ± 9 µm2 and 875 ± 26 µm2 respectively. At 28 days the intimal and medial areas were 913 ± 55 µm2 and 1437 ± 182 µm2 respectively. Luminal flow rate of the venous outflow tract was reduced significantly (344 ± 11 ml/min at Day 0 to 129 ± 14 ml/min at Day 7, p < 0.05).
CONCLUSIONS: This porcine model rapidly, reliably and robustly reproduces the flow reducing stenosis and intimal-medial hyperplasia at the venous outflow tract of PTFE arteriovenous fistula. It represents a promising tool for investigating the mechanisms of intimal-medial hyperplasia, evaluating therapeutic interventions and new graft materials. (The Journal of Vascular Access 2003; 4: 111-7)
Key Words. AV Fistula, Dialysis access, Intimal hyperplasia
INTRODUCTION
The introduction of the endogenous arteriovenous fistula by Brescia and colleagues in 1960 allowed for the development of chronic hemodialysis therapy for patients with end stage renal disease (1). Since then there has been an exponential increase with more than one million AV graft procedures performed for hemodialysis access every year in the United States with polytetrafluoroethylene (PTFE) graft being the most widely used conduit (2, 3). However, the one year patency rate for PTFE dialysis grafts is only 40 to 50% with a two-year patency rate of approximately 25% (4). These graft failure result in significant morbidity and hospitalization in the hemodialysis population in the United States at a cost exceeding $1 billion per year (5-7). The most common cause of vascular access graft failure is thrombosis secondary to stenosis at the venous anastomosis or outflow tract. The underlying pathological process is thought to be smooth muscle cell proliferation in the vessel wall, termed intimal-medial hyperplasia which narrows the lumen of the outflow vein (8-10). The principle stimuli for this process are endothelial injury, inflammation and increased vessel wall stress but knowledge of the cellular and molecular mechanisms of this process is limited (11). There are currently no effective treatments for the prevention or reduction of the proliferative response in the venous outflow tract of PTFE arteriovenous fistulas (12, 13).
In order to further understand the pathophysiology of intimal-medial hyperplasia in arteriovenous fistulas and study effective therapies, several animal models have been developed. The canine model of hemodialysis access has been most widely studied but been limited by the dissimilarity between canine and human vein responses to presence of arteriovenous fistula in terms of degree and rate of intimal hyperplasia lesions (14, 15). Lemson et al described a goat model of carotid artery to ipsilateral jugular vein (16) and Kohler et al developed a sheep model carotid artery to contralateral external jugular vein (17). The pig has been widely used in cardiovascular research and become accepted as the non-primate model of choice for many preclinical studies. There are many similarities between porcine and human anatomy and physiology as well as the nature of intimal hyperplasia response (18-20). We have developed at our institution based on extensive experience of porcine cardiovascular surgery a unique model of carotid artery to contralateral external jugular vein PTFE arteriovenous fistula. This model rapidly, robustly and reliably reproduces the intimal-medial hyperplasia response, is easy to use and accessible and allows potential for therapeutic interventions.
METHODS
Animals
All activities were pre-approved by the institutional animal review committee. 70 lb. Yorkshire cross-bred swine (Walnut Hill Farms, NC) were housed at the Duke University Vivarium. Animals were pre-treated with 650 mg aspirin PO for two days before surgery. Animals were not fed after midnight on the day of surgery.
Surgical protocol
Animals were tranquilized with a ketamine-acepromazine-glycopyrrolate solution and sedated with 2.5% thiopental, intubated with a size 6 endotracheal tube, and maintained on isofluorane for the duration of the procedure. Prior to skin incision, animals were given 1 g cefazolin i.v. The animal was placed in the supine position on the operating table and prepped in a sterile fashion. The animal was draped and a 15 cm midline longitudinal neck incision was made. The left common carotid artery was isolated first, followed by the right external jugular vein. An 8 cm segment of vein was freed from surrounding tissues and all tributaries off of the vein were ligated with 3-0 silk suture (Ethicon Inc, Sommerville, NJ). The animal was then treated with 100 U/kg of heparin i.v. followed by 1,000 U/hr for the duration of the procedure.
The left carotid artery was clamped and a 7 mm arteriotomy performed. An oblique end-to-side anastomosis was performed between the artery and a 6 mm internal diameter PTFE graft (Atrium, Hudson, NH) using a running 6-0 prolene suture (Ethicon Inc, Sommerville, NJ). Once fashioned, the arterial clamp was removed and the graft flushed with a heparin-saline solution. Flow was documented through the artery and into the graft. The graft was then tunneled beneath the sterno-clidomastoid muscles and brought into the proximity of the right external jugular vein.
A 7 mm venotomy was performed directly in the external jugular vein. The arteriovenous fistula was then completed with an oblique end-to-side anastomosis between the PTFE graft and the right external jugular vein, again using a running 6-0 prolene suture (Fig. 1a).
|
|
View larger version |
Fig. 1
a) Schematic diagram of graft postioning in the neck. Arrows indicate direction of blood flow. b) Diagram of PTFE to EJV anastomosis, indicating venous outflow tract hyperplasia and location of histological sections. |
|
|
|
All clamps were removed and good flow was observed through the graft. The left carotid artery distal to the PTFE anastomosis was then doubly tied off with 3-0 silk. Hemostasis was achieved. The neck wound was then filled with sterile saline and flow probe analysis was performed at the distal venous outflow tract using a 6 mm Transonic flow probe (Transonics Systems, Ithaca, NY) The muscular and subcutaneous layers were closed in one layer with running 2-0 vicryl suture (Ethicon Inc, Sommerville, NJ) and the skin closed. Animals were maintained at the Duke University Vivarium during the post-operative period following approved IACUC protocols. Animals were treated with 325 mg of aspirin PO QD post-operatively.
One group of animals were harvested at 7 days (n = 4) and another group of animals (n = 8) allowed to survive upto 28 days to assess long term patency. The patency of the graft was checked tri-weekly by means of palpation of a thrill or if not detected via a handheld ultrasound probe (Hewlett-Packard, Palo Alto, CA). Graft failure was defined as absence of palpable thrill and absence of flow by ultrasound analysis. If graft failure was detected then the animal was sacrificed at that time point. At the time of harvest, animals were tranquilized and sedated as described above. The old surgical incision was re-opened and the 8 cm segment of the right external jugular vein was isolated and freed from surrounding tissues. The neck wound was filled with sterile saline and flow probe analysis was performed at the distal outflow tract using a 6 mm Transonic flow probe (Transonic Systems Inc Ithaca, NY).
Histological staining and stenosis measurements
The 8 cm anastomosis segments of right external jugular vein were harvested at either 7 or 28 days post-operatively and perfusion-fixed with formalin. Venous segments were embedded in paraffin and cut in cross-section for histological staining and measurements. Five mm cross-sections were taken every 100 mm and stained with Masson trichrome (Fig. 1b). At least 50 sections were obtained from each venous anastomosis, and the 5 sections with maximal hyperplasia were identified and measured. Digital images were taken of these sections and measured with StainPoint 1.14 software (Lynx Graphics) for cross-sectional area comparison.
Modifications
Subcutaneous tunneling
A subcutaneous tunneled AV fistula may be created by starting with two lateral incisions, 4 cm from the midline, and exposing the left carotid artery and right external jugular vein. This modification facilitates graft palpation and enables percutaneous puncture procedures to be performed in this model.
Statistical analysis
Data are presented as mean ± SE. In vivo histological findings of hyperplasia were analyzed by ANOVA.
RESULTS
Surgical results
Arteriovenous PTFE fistulas were created between the common carotid and external jugular vein in 12 animals. There was no perioperative or post operative mortality. There were no strokes in spite of the ligation of the distal common carotid artery indicating an ample residual circulation through the Circle of Willis and collaterals via the vertebral vessels.
Patency
After 7 days all 4 grafts were patent and 100% patency was maintained for 14 days following implantation (Fig. 2).
|
|
View larger version |
Fig. 2
Patency rate of porcine PTFE AV fistula grafts. Graph demonstrates the percent of vascular conduits patent, as assessed clinically and by ultrasound analysis at various timepoints. |
|
|
|
However, by 28 days only 25% of grafts remained patent with thrombosis being the primary cause of graft failure.
Venous outflow tract hemodynamics
The flow recorded in the venous outflow tract immediately following completion of the anastomosis was 344 ± 11 ml/min. After 7 days, the flow through the venous outflow tract was significantly reduced to 129 ± 14 ml/min (p < 0.05 vs day 0) (Fig. 3).
|
|
View larger version |
Fig. 3
Hemodynamic flow through venous outflow tract of PTFE AV fistula grafts, immediately following fashioning of anastomosis and 7 days after surgery. Data represent mean ± SEM. * p
< 0.05 for flow rate at time of surgery vs 7 days.
|
|
|
|
Histology and intimal-medial hyperplasia
All the harvested grafts demonstrated significant proliferative response at the venous outflow tract (Figs. 4, 5).
|
|
View larger version |
Fig. 4
Intimal-medial surface areas of venous outflow tracts of PTFE AV fistula grafts. Histograms represent measurements of areas of the intimal and medial layers of venous outflow tract 7 and 28 days following implantation. Data represent the mean ± SEM. |
|
|
|
After 7 days the intimal area was 469 ± 9 µm2 and the medial area 875 ± 26 µm2. The proliferative trend progressed until 28 days with an intimal area of 913 ± 55 µm2 and medial area of 1437 ± 182 µm2. The intima was composed of smooth muscle cells as demonstrated by histologic analysis (Fig. 5d).
|
|
View larger version |
Fig. 5
Representative histologic sections through venous outflow tract of porcine AV fistula grafts, stained with Masson Trichome. a) Control external jugular vein (day 0) (4x). b) 7 day explanted vein (4x). c) 28 day explanted vein (4x). d) High power of neointima formation at the venous outflow tract showing smooth muscle cell proliferation (40x). |
|
|
|
DISCUSSION
In the present study we demonstrated that pigs develop significant proliferation of the venous outflow tract of PTFE arteriovenous fistulas. The hyperplastic response increases with time and is so robust that stenosis at the venous anastomosis in conjunction with thrombus formation leads to graft occlusion in 75% of the 28 day animals. Stenosis due to intimal-medial thickening increases resistance in these normally high flow grafts and reduced flow results in thrombotic occlusion. Our data at 7 days establishes the presence of significant hyperplasia at the venous outflow tract in 4 patent thrombus free grafts indicating that intimal hyperplasia precedes the thrombotic occlusion seen in the 28 day group. These failure rates are much higher than what is seen in the human patient population, but it is the rate of intimal-medial hyperplasia which makes this an excellent model. In short, we have found that this 28 day model can mimic the human proliferative response that may take months.
The term intimal hyperplasia is used extensively in the literature in association with the underlying pathological process and lesions that cause vein graft and AV fistula PTFE graft failure. However, we are of the opinion that the term intimal-medial, or as some have suggested fibromuscular hyperplasia is more appropriate since the proliferative response is not confined to the intima (8). Several recent studies have highlighted that the smooth muscle cells that constitute the neointima are derived from the media and migrate following stimulation by injury, inflammation or wall stress. In fact, medial hyperplasia appears to play a far greater role in reducing luminal area and thus creating a stenosis. Therefore, therapies directed towards improving the patency of AV fistulas must also target the medial proliferative response.
Several animal models have been developed to study the process of intimal-medial hyperplasia. The rabbit and rat have been extensively studied with significant intimal hyperplastic responses being documented in vein and PTFE grafts (21). However, the disparity in size of vessels in relation to humans makes such animals unsuitable for studying larger caliber grafts. Dogs have traditionally been the most studied large animal model with femoral and carotid arteries being of comparable caliber to humans. However, intimal hyperplasia in canines occurs at a far slower rate than in humans, with the maximal response being situated within the graft itself rather than compromising the lumen of the outflow tract (14, 15). Lemson et al heralded a new goat model of intimal hyperplasia but the rate of progression of intimal hyperplasia with time was so slow that there was no significant difference between 10 days and 7 weeks (16). Kohler et al described a sheep model in which PTFE grafts were also placed between the carotid artery and jugular vein (17). They reported a rapid intimal hyperplastic response at 4 weeks with thrombus formation predominating. However, they failed to demonstrate whether intimal hyperplasia preceded thrombus formation indicating that thrombosis is the causative factor in intimal lesion formation, unlike in humans.
Pigs have been used since 1972 when Soyer et al implanted a PTFE prosthesis into the venous circulation and demonstrated clearly neointima formation throughout the length of the graft after 2 weeks (24). Pigs have been favored due to their anatomic and physiologic similarity to humans. Johnson et al recently proposed a porcine hemodialysis access model in which they constructed AV fistulae between the external iliac artery and external iliac vein using PTFE (25). In a similar model, Kelly et al recently demonstrated that the nature of the venous neointimal hyperplasia in a pig model of arteriovenous PTFE placed between the femoral artery and vein closely resembled lesions analyzed taken from tissue samples from hemodialysis patients with stenosed fistulas (26). Although a significant degree of intimal hyperplasia and stenosis at the venous outflow tract was reproduced in these models they are severely limited by the fact that the short length of conduit (only 2 cm and 7 cm respectively) and the small PTFE graft diameter of 4 mm employed.
The model we propose has several advantages over previous ones discussed above. We prefer the neck vessels since they are of similar caliber to human brachial artery (common carotid artery 3-5 mm) and basilic vein (external jugular vein 4-8 mm) that are routinely used for hemodialsis access grafts. The courses of the internal carotid artery and contralateral external jugular vein are easily accessible via a single midline incision and they run almost parallel to each other, avoiding potential hazards of crossing joints. We were able to use 6 mm PTFE grafts, identical to those routinely implanted in humans and lengths of graft were typically around 20 cm.
The pigs in our study reliably and robustly produced intimal hyperplasia and stenotic lesions in the venous outflow tract, which are histologically similar to those that occur in humans. The rate of development of these lesions is more rapid than described in dogs, sheep and goats which allows for effective earlier evaluation of therapeutic interventions against intimal hyperplasia. Even by the 7 day time point the degree of intimal-medial hyperplasia is very impressive and the resulting flow reducing stenosis is an important determinant of subsequent thrombosis that reduced patency as shown in the animals studied over 28 days.
The proposed model may be used effectively for improving the properties of PTFE graft design and function such as modifications aimed at reducing thrombosis and stimulating transmural tissue and capillary ingrowth. The presence of a bridge of skin across the midline allows the potential for subcutaneous tunneling facilitating subsequent puncture of the graft. The effects of repeated puncturing on the patency of graft may also be investigated in this way.
Therapeutic interventions such as local/systemic anti-proliferative/anti-thrombotic agents, local radiation treatment and seeding of the graft with endothelial cells may be examined with this model. A modification of the model that allows specific treatment of the venous outflow tract with anti-proliferative gene therapy has also been conducted. Further, with the current interest surrounding drug eluting stents, this model may represent a useful platform for evaluating the effectiveness of such therapies in preventing or even reversing proliferation due to intimal-medial hyperplasia.
This model represents a relatively inexpensive, readily available, easy to use and adaptable pre-clinical tool to investigate the mechanisms of intimal hyperplasia, apply therapeutic interventions and evaluate the performance of new graft materials.
Acknowledgements
This work was supported in part by grants HL56205, HL65360, and HL59533 from the National Institutes of Health (W.J.K) and the Department of Surgery, Duke University.
W.J.K. receives sponsored research support from Genzyme (Framingham, MA). J.H.L. is a Clinician Scientist Awardee from the American Heart Association and Genentech (San Francisco, CA).
Address for correspondence:
Jeffrey H. Lawson, M.D., Ph.D.
Departments of Surgery & Pathology
Duke University Medical Center
DUMC Box 2622
Room 481 MSRB, Research Dr.
Durham, NC 27710
lawso006@mc.duke.edu
REFERENCES
1. Brescia MJ, Cimino JE, Appel K., Hurwich BJ. Chronic hemodialysis using venipuncture and a surgically created arteriovenous fistula. N Eng J Med 1966; 275: 1089-92.
2. Kapoian T, Sherman RA. A brief history of vascular access for hemodialysis: An unfinished story. Semin Nephrol 1997; 17: 239-45.
3. Stanley JC, Barnes RW, Ernst CB, Hertzer NR, Mannick JA, Moore WS. Vascular Surgery in the United States: Workforce issues. Report of the Society for Vascular Surgery and the International Society for Cardiovascular Surgery, North American Chapter, Committee on Workforce Issues. J Vasc Surg 1996; 23: 172-81.
4. Schwab SJ, Harrington JT, Singh A. Vascular access for hemodialysis. Kidney Int 1999; 55: 2078-90.
5. System USRD. USRDS 1997 annual data report. Ann Arbor: University of Michigan.
6. Feldman, HI, Held PJ, Hutchinson JT, Stoiber E, Hartigan MF, Berlin JA. Hemodialysis vascular access morbidity in the United States. Kidney Int 1993; 43: 1091-6.
7. Hakim R, Himmelfarb J. Hemodialysis access failure: A call to action Kidney Int 1998; 54: 1029-40.
8. Swedberg SH, Brown BG, Sigley R, Wight TN, Gordon D, Nicholls SC. Intimal fibromuscular hyperplasia at the venous anastomosis of PTFE grafts in hemodialysis patients. Clinical, immunocytochemical, light and electron microscope assessment. Circulation 1989; 80: 1726-36.
9. Roy-Chaudhury P, Kelly B, Miller MA, et al. Venous neointimal hyperplasia in polytetrafluoethylene dialysis grafts. Kidney Int 2001; 59: 2325-34.
10. Rekhter MD, Nicholls SC, Ferguson M, Gordon D, Cell proliferation in human arteriovenous fistulas used for hemodialysis. Arterioscler Thromb 1993; 13: 609-17.
11. Newby AC, Zaltsman AB. Molecular mechanisms in intimal hyperplasia. J Pathol 2000; 190: 300-9.
12. Kovalik EC, Schwab SJ. Can venous stenosis be prevented? Semin Dial 1999; 12: 145.
13. Sukhatme V. Vascular access stenosis: Prospects for prevention and therapy. Kidney Int 1996; 49; 1161-74.
14. Fillinger MF, Reinitz ER, Schwartz RA, Resetarits DE, Paskanik AM, Bredenberg CE. Beneficial effects of banding on venous intimal-medial hyperplasia in arteriovenous loop grafts. Am J Surg 1989; 158; 87.
15. Fillinger MF, Reinitz ER, Schwartz RA, et al. Graft geometry and venous intimal-medial hyperplasia in arteriovenous loop grafts. J Vasc Surg 1990; 11: 556.
16. Lemson MS, Daemen MJAP, Kitslaar PJEHM, Tordoir JHM. A new animal model to study intimal hyperplasia in arteriovenous fistulas. J Surg Res 1999; 85: 51-8.
17. Kohler T, Kirkman TR. Dialysis access failure: A sheep model of rapid stenosis. J Vasc Surg 1999; 30: 744-51.
18. Ferrel M, Fuster V, Gold HK, Chesebro JH. A dilemma for the 1990s: choosing appropriate experimental model for the prevention of restenosis. Circulation 1992; 85: 1630-1.
19. Schwartz RS, Edwards WD, Bailey KR, Camrud AR, Jorgenson MA, Holmes DR. Jr. Differential neointimal response to coronary artery injury in pigs and dogs: implications for restenosis models. Arterioscler Thromb 1994; 14: 395-400.
20. Sims FH. A comparison of the structural features of the walls of coronary arteries from 10 different species. Pathology 1989; 21: 115-24.
21. Friedman EW, Hamilton AJ. Polytetrafluorethylene grafts in the peripheral circulation of rabbits. Am J Surg 1983; 146: 355.
22. Kohler TR, Kirkman TR, Krais LW, Zierler BK, Clowes AW. Increased blood flow inhibits neointimal hyperplasia in endothelialised vascular grafts. Circ Res 1991; 69: 1557.
23. Geary RL, Kohler TR, Vergel S, Kirkman TR, Clowes AW. Time course of flow-induced smooth muscle cell proliferation and intimal thickening of endothelialised baboon vascular grafts. Circ Res 1994; 74: 14-23.
24. Soyer T, Lempinen M, Cooper P, Norton L, Elseman BA. A new venous prosthesis. Surgery 1972; 72: 864-72.
25. Johnson MS, McLennan G, Lalka SG, Whitfield RM, Dreesen RG. The porcine hemodialysis access model. J Vasc Interv Radiol 2001; 12: 969-77.
26. Kelly BS, Narayana A, Heffelfinger SC, et al. External Beam Radiation attenuates venous neointimal hyperplasia in a pig model of arteriovenous polytetrafluoroethylene (PTFE) graft stenosis. Int J Radiation Oncology Biol Phys 2002; 54: 263-9.
|
|
