Does PGA external stenting reduce compliance mismatch in venous grafts?
- Zhong-zhao Teng†1Email author,
- Guang-yu Ji†2,
- Hong-jun Chu2,
- Zhi-Yong Li3,
- Liang-jian Zou2,
- Zhi-yun Xu2 and
- Sheng-dong Huang2
© Teng et al; licensee BioMed Central Ltd. 2007
Received: 31 August 2006
Accepted: 16 April 2007
Published: 16 April 2007
Autogenous vein grafting is widely used in regular bypassing procedures. Due to its mismatch with the host artery in both mechanical property and geometry, the graft often over expands under high arterial blood pressure and forms a step-depth where eddy flow develops, thus causing restenosis, fibrous graft wall, etc. External stents, such as sheaths being used to cuff the graft, have been introduced to eliminate these mismatches and increase the patency. Although histological and immunochemical studies have shown some positive effects of the external stent, the mechanical mismatch under the protection of an external stent remains poorly analyzed.
In this study, the jugular veins taken from hypercholesterolemic rabbits were transplanted into the carotid arteries, and non-woven polyglycolic acid (PGA) fabric was used to fabricate the external stents to study the effect of the biodegradable external stent. Eight weeks after the operation, the grafts were harvested to perform mechanical tests and histological examinations. An arc tangent function was suggested to describe the relationship between pressure and cross-sectional area to analyse the compliance of the graft.
The results from the mechanical tests indicated that grafts either with or without external stents displayed large compliance in the low-pressure range and were almost inextensible in the high-pressure range. This was very different from the behavior of the arteries or veins in vivo. The data from histological tests showed that, with external stents, collagen fibers were more compact, whilst those in the graft without protection were looser and thicker. No elastic fiber was found in either kind of grafts. Furthermore, grafts without protection were over-expanded which resulted in much bigger cross-sectional areas.
The PGA external extent contributes little to the reduction of the mechanical mismatch between the graft and its host artery while remodeling develops. For the geometric mismatch, it reduces the cross-section area, therefore matching with the host artery much better. Although there are some positive effects, conclusively the PGA is not an ideal material for external stent.
Autogenous vein grafting is widely used in regular bypassing procedures in the treatment of ischemia due to occlusive vascular lesions, such as atherosclerosis. However its patency is limited by progressive intima hyperplasia, which causes serious clinical problems, needing repeated angioplasty and graft surgery. Extensive studies have been carried out for capturing the mechanisms of restenosis. Currently the mechanical property/compliance and geometric mismatches are regarded as the main factors contributing to graft failure. Due to the poorly developed venous middle layer, the graft is over-expanded in an arterial environment where the blood pressure is much higher than that in the vein. The over-expansion induces extremely high stresses in the venous wall, approximately 5 times those in the artery wall and 140 times those in the normal vein , which definitely damages the living components in the venous wall [2–4]. The distension also leads to a geometrical mismatch around the anastomosis, where the eddy flow forms [5–7] hence causing disordered shear stress distribution and long particle residence time.
Limited by these mismatches, vein patency is generally less good than those of selected arterial grafts, such as mammary artery , the right gastro-epiploic artery , and the radial artery , whose mechanical property meets that of their hosts much better. However, most arteries are necessary, so available arterial conduits are very limited. Therefore this generates an interest in looking for approaches to increase the patency of venous grafts. One of the most efficient is called external stenting (ES), which was first introduced by Parsonnet et al.  by using a sheath to cuff the vein section to reduce the mismatch between graft and host. It has been shown that this protection could preserve the endothelium cells (ECs), smooth muscle cells (SMCs) and elastic fibers and increase vasa vasorum [12, 13]. It also indicates that the ES can decrease wall thickness and matrix deposition significantly [13–15], prevent atherosclerosis around the anastomosis [16–20] even in hypercholesterolemic animals , and reduce cell apoptosis [3, 7, 21, 22]. Moreover, ES can not only improve the patency of the normal vein graft, but also means that varicose and dilated veins can be used [23, 24].
Up to now, investigations on ES have been largely focused on morphological, histological and immunochemical aspects, whilst quantitative results of the mechanical properties of the graft with ES are still little known, particularly, whether or not the compliance mismatch can be eliminated, with the degrading of ES remaining unknown. In this study, the compliance and distensibility of venous grafts with and without ES were investigated. Studies were performed on hypercholesterolemic rabbits. The jugular vein was harvested and implanted into the carotid artery transpositionally to form an end-to-end graft and externally stented with a non-woven polyglycolic acid (PGA) fabric. After 8 weeks the animal was sacrificed. The graft was harvested for mechanical tests as well as morphological and histological examinations.
Twenty-five adult male rabbits, with a weight ranging from 2.0 to 2.5 kg, were divided into 5 groups at random (5 per group). One group was given a normal diet, named as NDG, and the other four groups, named as HSM, HM, HSH and HH, respectively, were given a hypercholesterolemic diet (HD) by the additional of 1 g of cholesterin powder per day. The grafts from groups HSM and HM were used for mechanical study, and the grafts from groups HSH and HH were used for morphological and histological examinations (Please refer to the Abbreviations for details of each group). At the beginning of week 0,2,4,8 and12, blood samples were taken from all the rabbits from groups NDG and HH to perform total cholesterol (TC), triglyceride (TG), low density lipoprotein (LDL) and high density lipoprotein (HDL) examinations (BIO-RAD3550-UV, Baxter, USA). Rabbits were fasted 12 hours before blood extraction via ear edge vein. At the end of the 12th week, the grafting operations were carried out on all hypercholesterolemic animals and external stents were used for groups HSM and HSH. The diet remained unchanged after the operation.
Main procedures of the grafting operation
(1). The feeding was stopped 12 hours before the surgery and the water supply was stopped 4 hours before. The animal was anaesthetised with 1% nembutal solution (35–40 mg/kg) and heparinized with a dose of 1 mg/kg. An additional nembutal solution (5–10 mg/kg) dose was injected if needed during the operation.
(3). For the groups (HM & HH) without ES, the serrefine close to the heart and the one at the distal end were released to reopen the blood flow. The operating area was then closed in layers with 4# silk thread.
(4). For the groups (HSM & HSH) with ES, when the target vein was exposed, its outer diameter was measured at both ends and the middle point to calculate the width of the ES. The length of the ES had to be sufficent to cover both cuff tubes. Therefore a rectangular piece of PGA non-woven fabric (provided by GUNZE LTD, Japan) was made to wrap the graft and sutured with 6-0 proline line (Johnson & Johnson Medical LTD., USA). Then the operating area and wound were closed.
(5). A dose of 0.8-million units of penicillin was administered as soon as the operation was completed and 3 days later. Aspirin was administered at 12.5 mg/kg/day in the post-operative 4 weeks. The hyper-cholesterolemic diet was continued on the following day.
The animals in groups HSH and HH were used for histological examinations. 8 weeks after the operation, the graft, around 3 cm in length, was harvested. Snipping off 0.5 cm- long sections at both ends, the remaining part was cut into 4 rings and 3 were selected arbitrarily for morphological and histological examinations: the thickness of the intima and media was measured by the Leica Qwin image analysis system (U.K.); victoria blue-ponceaux (VB) was used to stain elastic fiber and collagen fiber.
The grafts and the joined arteries (0.5–1 cm) from the animals in groups HSM and HM were harvested and submerge in ringer solution containing narceine (1 mg/ml) immediately and preserved at 4°C for mechanical tests. The non-touched artery and vein (around 3.5 cm) opposite to the graft were also harvested (the arteries were grouped as HA, and the veins were HV). All vessels were mounted in an apparatus built in our laboratory (I3A, Spain) to perform expansion tests in 6 hours after cutting. A detailed description of the experimental apparatus has been well presented elsewhere [27, 28]. The sensors used in the apparatus are the following: 1. LVDT (SX 20 ME × 200, Sensorex, France) for measuring the volume of the liquid pumped into the specimen; 2. Microliter syringe (Hamilton, Switzerland); 3. Pressure sensor (TP3, AEP, Italy) for measuring the pressure; 4. A/D card (NI PCI 6014, National Instrument, USA) for recording the digital signals.
It has to be pointed out that the graft with protection usually had many vasa vasorums, so some of the bigger ones were ligated carefully while those smaller ones were blocked by dried blood particles from the blood of the NDG group. The procedures of the expansion test are presented below.
Main procedures of the mechanical test
(1). The graft was cannulated at both ends until the tips touched the cuff tubes and then tied with 4# silk thread twice. The samples were submerged in sodium lactate ringer's injection (Guangdong Otsuka Pharmaceutical Co., Ltd.) and the temperature was controlled at 23°C by a dipped heater.
(2). Once mounted on the holders, the ringer solution containing blood particles was injected into the graft through a triple valve until the internal pressure could maintain at 50 mmHg in 1 minute without obvious decrease.
Maximum pressure under different axial stretch ratios in different group (mmHg)
group mp (mmHg) λ z
(4). After the mechanical test, the arteries were cut into rings (around 1 mm in length) and the geometric information was recorded to calculate the cross-section area (Anon-load) in load free state. The compliance of the system was measured by replacing the specimen with a rigid glass tube and it was taken into account in the experimental data analysis.
Data fitting and statistics
in which A p eq denotes the value of A eq when pressure equals p. From (Eq.(5)), reflects the relative expansion rate when pressure changes from p 1 to p 2.
The data in this study are presented as Mean ± SD. The comparison between two groups was tested using a Student-T test. p > 0.05 was used to indicate there is no statistical difference; 0.01 < p < 0.05 denotes significant statistical difference (marked with * or #) and p < 0.01 means a highly significant statistical difference (marked with ** or ##).
Index of blood-fat
The blood-fat level (mmol/L) (in each group n = 5)
1.67 ± 0.43
1.72 ± 0.45
1.72 ± 0.25
1.98 ± 0.55
2.02 ± 0.58
1.80 ± 0.20
1.88 ± 0.22
2.59 ± 0.76*#
18.57 ± 1.25**##
19.08 ± 1.31**##
1.06 ± 0.31
1.09 ± 0.45
1.11 ± 0.38
1.28 ± 0.35
1.36 ± 0.55
1.09 ± 0.15
1.12 ± 0.11
2.09 ± 0.87*#
3.83 ± 1.75**##
4.07 ± 1.54**##
0.61 ± 0.20
0.69 ± 0.26
0.65 ± 0.31
0.58 ± 0.17
0.76 ± 0.27
0.53 ± 0.07
0.55 ± 0.09
1.21 ± 0.35*#
2.04 ± 1.11**##
2.12 ± 1.23**##
0.93 ± 0.21
0.95 ± 0.33
1.01 ± 0.49
0.89 ± 0.27
0.98 ± 0.25
0.94 ± 0.26
1.06 ± 0.24
1.84 ± 0.36*#
14.68 ± 1.42**##
15.13 ± 1.64**##
Expansion rate and compliance
Figure 2 indicates that the expansibility of the graft is much smaller than that of the vein. When pressure increases from 0 to 20 mmHg, A eq of the vein could increase more than 10 times (Figure 2A); however, that of the grafts only increased by about 50–80% (Figure 2B). Although the expansibility of the graft with ES was better than the one without protection, it was still much worse than that of the artery. Figure 2 also shows that for both vein and graft without ES, Δ increases only a little when the pressure is more than 40 mmHg.
The expansion rate in different pressure ranges (λ z = 1.4)
η 0–20 (%)
20.20 ± 8.93
56.40 ± 7.34
82.43 ± 3.73
η 80–120 (%)
16.55 ± 3.86
2.99 ± 0.92
0.45 ± 0.10
According to the conclusion from Table 3 and the values in Figure 3, the compliance of the grafts in the low pressure range (p < 20 mmHg) is much higher than that in the high pressure range (p > 20 mmHg). With a given λ z , when the pressure is less than 20 mmHg, the compliance of HM (without ES) was higher than that of HSM (with ES). Particularly, in the cases of p ranging from 0 to 5 mmHg, significant differences were found between HM and HSM. However, the decrease of compliance of HM with pressure was much faster than that of HSM. Therefore, when pressure exceeded 30 mmHg, the compliance of HSM was higher than that of HM, especially when λ z equaled 1.3 and 1.4.
Under the same pressure, the compliance of the vein was usually 10 times that of the graft (HSM or HM). This implies that the venous section becomes much stiffer in order to tolerate much higher loading in the arterial environment. With the protection of the ES, the mechanical property of HSM was more similar to that of the artery when this was compared with HM, even though there were still important differences in compliance (see Figure 3) and expansibility (see Table 3). Therefore at least 8 weeks after the grafting operation, the PGA ES does not have a significant impact on improving venous mechanical tone to make it similar to its host artery.
In fact, thickening is not detrimental to the graft, which decreases the high stress induced by the arterial blood pressure. However, ideal remodeling would be able to change the microstructure of the wall and finally arterize the venous graft. Now it is clear that remodeling leads grafts without protection to having a fibrotic wall, filled with foam cells, inflammation cells and, more likely to fail. The purpose of biodegradable ESs is to make remodeling develop well and finally eliminate the geometrical and mechanical mismatches between the vein and the host artery. However, the results in this study indicate that PGA ES does not guarantee an ideal remodeling process.
The disappearance of elastic fiber may be the reason for the convex shapes of p-Δ curve of the grafts (Figure 2(B)), the high expansibility (Table 3) and the high compliance in the low pressure range (Figure 3). It is well known that in the low pressure range, the load is mainly undertaken by elastic fiber, and when pressure exceeds a certain value, (for artery it is around 50 mmHg,) collagen fibers begin to tighten to provide strength for the wall, which is the reason of the 'S' shape of the arterial p-Δ curve (Figure 2(A)). Grafts in this study lacking elastic fiber could be compared with a tuck net. During the early period of liquid being pumped in, pressure increases little, but as long as the volume reaches a certain value, tighting the collagen fibers, the pressure increases quickly. Thus the explanation for the mechanical behaviour of the grafts is the collagen fiberising of the wall.
From the foregoing characterization of the compliance and the micro-component of the wall, we have come to the conclusion that the PGA ES can improve the venous graft tone, giving rise to a smaller expansion rate in the low pressure range and a higher value in the physiological range (Table 3), a compliance closer to the artery and more compact collagen in the wall (Figures 6). However, the compliance of the venous graft with ES is still quite different from that of the artery. At least 8 weeks after the operation, the biodegradable PGA ES could not eliminate the mechanical mismatch between the venous graft and its host artery. Although we could not extend the validities of this conclusion for a longer period, (because after 8 weeks the PGA ES had not degraded thoroughly and the residual part could still be seen), we do not expect the improvement of graft tone to become similar to the artery after a longer period of remodeling.
The weakening of ES will allow the venous wall to undertake more and more loading and this causes tissue remodeling. Therefore the degrading property of ES should be important. According to technological reports from the provider, the PGA non-woven fabric used in our experiments degrades totally in 15 weeks in the saline solution in vitro. Actually, we found that after 12 weeks there was no perceptible remnant in vivo (the data is not shown in this paper). Despite its importance, the mechanical property evolution of ES in vivo is extremely poorly understood. Except for the weakening process, the degradation product is another important factor. The products of PGA are CO2 and H2O, which could decrease the local pH and affect the local tissue, possibly promoting apoptosis.
Cross section areas in load free state (mm2)
1.28 ± 0.11
2.36 ± 0.64
3.87 ± 0.17
18.15 ± 7.91
From the reported results, ESs have a quite similar protection effect, such as preventing hyperplasia in the inner layer, preserving EC and SMC, regulating the extracellular matrix, reducing the lipid deposition, stopping the granulocyte and macrophage invasion and the formation of foam cells, and finally improving the bypass patency significantly. However Es currently has not been applied clinically. This is mainly because of the following remaining questions: (1) how the mechanical property of the graft varies during the remodeling process when different kinds of ESs are used, (2) what is the long-term outcome, (3) what is the standard operation process of cuffing, and (4) what is the ideal material for ES. Further studies to address these issues would be very useful.
Grafts both with and without ES displayed a large compliance and high expansibility in the low pressure range and were almost inextensible in the high pressure range, and such behavior was more prominent in those without ESs. The collagenous fiberized graft wall and the disappearance of elastic fiber were the key factors leading to this mechanical behavior. Although the PGA ES reduced the compliance mismatch between graft and host artery, the gap remained very large after 8 weeks. Therefore from the mechanical point of view, the PGA non-woven fabric is not an ideal material for ES. However, its effect is obvious in reducing the geometrical mismatch, which is important for the shear stress pattern in the graft.
Non-touched artery from hypercholesterolemic animal
High density lipoprotein
For the hypercholesterolemic animal, whose graft is without external stent and not for mechanical test
For the hypercholesterolemic animal, whose graft is without external stent and for mechanical test
For the hypercholesterolemic animal, whose graft is with external stent and not for mechanical test
For the hypercholesterolemic animal, whose graft is with external stent and for mechanical test
Non-touched vein from the hypercholesterolemic animal
Low density lipoprotein
Control group with normal diet
Smooth muscle cell
The authors gratefully acknowledge the supports from Shanghai Board of Health, China through the grant of 034119 and the Juan de la Cierva program. We thank Mr Tim Baynes from University of Cambridge for checking the English.
- Liu SQ, Fung YC: Changes in the organization of the smooth muscle cells in rat vein grafts. Ann Biomed Eng 1998, 26: 86–95. 10.1114/1.52View ArticleGoogle Scholar
- Cheng W, Li B, Kajstura J: Stretch-induced programmed myocyte cell death[J]. J Clin Invest 1995, 96: 2247–2259.View ArticleGoogle Scholar
- Moore MM, Goldman J, Patel AR, Chien S, Liu SQ: Role of tensile stress and stain in the induction of cell death in experimental vein graft. Journal of Biomechanics 2001, 34: 298–297. 10.1016/S0021-9290(00)00217-7View ArticleGoogle Scholar
- Wolfgang H Goldmann: Mechanical aspects of cell shape regulation and signalling. Cell Biology International 2002, 4: 313–317.Google Scholar
- Hughes PE, How TV: Effects of geometry and flow division on flow structure in models of the distal end-to-side anastomosis. J Biomechanics 1996, 29: 855–872. 10.1016/0021-9290(95)00168-9View ArticleGoogle Scholar
- Hughes PE, Shortland AP, How TV: Visualization of vortexshedding at the proximal side-to-end artery-graft anastomosis. Biorheology 1996, 33: 305–317. 10.1016/0006-355X(96)00024-8View ArticleGoogle Scholar
- Liu SQ, Moore MM, Yap C: Prevention of mechanical stretch-induced endothelial and smooth muscle cell injury in experimental vein grafts. J Biomech Eng 2000, 122: 31–38. 10.1115/1.429625View ArticleGoogle Scholar
- Peck AW, Orszulak TA, Anderson BJ, Schaff HV: Single versus bilateral internal mammary artery graft: 10-year outcome analysis. Ann Thorac Surg 1997, 64: 599–605. 10.1016/S0003-4975(97)00620-6View ArticleGoogle Scholar
- Grandjean JG, Boonstra PW, den Heyer P, Ebels T: Arterial revascularization with the right gastroepiploic artery and internal mammary arteries in 300 patients. J Thorac Cardiovasc Surg 1994, 107: 1309–1316.Google Scholar
- Chen AH, Nakao T, Brodman RF, Greenberg M, Charney R, Menegus M, Johnson M, Grose R, Hu EC, Choi H-K, Safyer S: Early postoperative angiographic assessment of radial artery grafts used for coronary artery bypass grafting. J Thorac Cardiovasc Surg 1996, 111: 1208–1212. 10.1016/S0022-5223(96)70223-4View ArticleGoogle Scholar
- Parsonnet V, Lari AA, Shah IH: New stent for support of veins in arterial grafts. Arch Surg 1963, 87: 696–702.View ArticleGoogle Scholar
- Barra JA, Volant A, Leroy JP, Braesco J, Airiau J, Boschat J, Blanc JJ, Penther P: Constrictive perivenous mesh prosthesis for preservation of vein integrity. Experimental results and application for coronary bypass grafting. J Thorac Cardiovasc Surg 1986, 92: 330–336.Google Scholar
- Bambang LS, Moczar M, Lecerf L, Loisance D: External biodegradable supporting conduit protects endothelium in vein graft in arterial interposition. Int J Artif Organs 1997, 20: 397–406.Google Scholar
- Kohler TR, Kirkman TR, Clowes AW: The effect of rigid external support on vein graft adaptation to the arterial circulation. J Vasc Surg 1989, 9: 277–285. 10.1067/mva.1989.vs0090277View ArticleGoogle Scholar
- Batellier J, Wassef M, Merval R, Duriez M, Tedgui A: Protection from atherosclerosis in vein grafts by a rigid external support. Arterioscler Thromb 1993, 13: 379–384.View ArticleGoogle Scholar
- Sottiurai VS, Yao JS, Batson RC, Sue SL, Jones R, Nakamura YA: Distal anastomotic intimal hyperplasia: histopathologic character and biogenesis. Ann Vasc Surg 1989, 3: 26–33.View ArticleGoogle Scholar
- Angelini GD, Bryan AJ, Williams HM, Morgan R, Newby AC: Distention promotes platelet and leukocyte adhesion and reduces short-term patency in pig arteriovenous bypass grafts. J Thorac Cardiovasc Surg 1990, 99: 433–439.Google Scholar
- Bassiouny HS, White S, Glagov S, Choi E, Giddens DP, Zarins CK: Anastomotic intimal hyperplasia: mechanical injury or flow induced. J Vasc Surg 1992, 15: 708–717. 10.1067/mva.1992.33849View ArticleGoogle Scholar
- White SS, Zarins CK, Giddens DP, Bassiouny H, Loth F, Jones SA: Hemodynamic patterns in two models of end-to-side vascular graft anastomoses: effects of pulsatility, flow division, Reynolds number, and hood length. J Biomech Eng 1993, 115: 104–111.View ArticleGoogle Scholar
- Liu SQ, Moore MM, Glucksberg MR, Mockros LF, Grotberg JB, Mok AP: Partial prevention of monocyte and granulocyte activation in experimental vein grafts by using a biomechanical engineering approach. J Biomechanics 1999, 32: 1165–1175. 10.1016/S0021-9290(99)00117-7View ArticleGoogle Scholar
- O'Brien JamesE Jr, Ormont MichaelL, Shi Yi, Wang Dian, Zalewski Andrew, Mannion JohnD: Early injury to the medial after saphenous vein grafting. Ann Thorac Surg 1998, 65: 1273–1278. 10.1016/S0003-4975(98)00175-1View ArticleGoogle Scholar
- Mayr M, Li C, Zou Y, Huemer U, Hu Y, Xu Q: Biomechanical stress-induced apoptosis in vein graft involves p38 mitogen-activated protein kinases. FASEB Journal 2000, 15: 261–270.Google Scholar
- Moritz A, Raderer F, Magometschnigg H, Trubel W, Ullrich R, Laufer G, Staudacher M: The use of mesh-tube-constricted dilated or varicose veins as arterial bypass conduit. Thorac Cardiovasc Surg 1992, 40: 356–360.View ArticleGoogle Scholar
- Zurbrugg HR, Hetzer R: The use of biocompound-grafts together with varicose veins. First clinical experience. J Cardiovasc Surg 1996, 37: 143–146.Google Scholar
- Heren I: Kidney transplantation in the rabbit: A new method. Acta Pathol Microbil Scand 1970, 78: 90–95.Google Scholar
- Kamada N, Calne RY: A surgical experience with five hundred thirty liver transplants in the rat. Surgery 1983, 93: 64–69.Google Scholar
- Gunst SJ, Stropp JQ: Pressure-volume and length-stress relationship in canine bronchi in vitro. Journal of Applied Physiology 1988, 64: 2522–2531.Google Scholar
- Tiddens HAWM, Hofhuis W, Bogaard JM, Hop WCJ, de Bruin H, Willems LNA, de Jongste JC: Compliance, histeresis and collapsibility of human small airways. American Journal of Respiratory and Critical Care Medicine 1999, 160: 1110–1118.View ArticleGoogle Scholar
- Teng ZZ, Liu ZR, Wang YQ, Tang WC, Liu BY: p-V Relationship and Limit Flow for Collapsible Trachea. Chinese Quarterly Mechanics 2001,22(1):40–46.Google Scholar
- Zweep HP, Satoh S, van der Lei B, Hinrichs WL, Dijk F, Feijen J, Wildevuur CR: Autologous vein supported with a biodegradable prosthesis for arterial grafting. Ann Thorac Surg 1993, 55: 427–433.View ArticleGoogle Scholar
- Hinrichs WL, Zweep HP, Satoh S, Feijen J, Wildevuur CR: Supporting, microporous, elastomeric, degradable prostheses to improve the arterialization of autologous vein grafts. Biomaterials 1994, 15: 83–91. 10.1016/0142-9612(94)90255-0View ArticleGoogle Scholar
- Liu SQ: Prevention of focal intimal hyperplasia in rat vein grafts by using a tissue engineering approach. Atherosclerosis 1998, 140: 365–377. 10.1016/S0021-9150(98)00143-9View ArticleGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.