European Journal of Cardio-thoracic Surgery 22 (2002) 898–903
www.elsevier.com/locate/ejcts
A new outside stent – does it prevent vein graft intimal proliferation? q
Michal Krejca a,*, Janusz Skarysz a, Przemyslaw Szmagala a, Danuta Plewka b,
Grazyna Nowaczyk b, Andrzej Plewka b, Andrzej Bochenek a
a
1st Cardiac Surgery Department, Silesian University of Medicine, SPSK nr 7, 40-635 Katowice, ul. Ziolowa 45-47, Poland
Department of Histology and Embryology, Silesian University of Medicine, 40-752 Katowice, ul. Medykow 18-20, Poland
b
Received 28 June 2002; received in revised form 5 September 2002; accepted 6 September 2002
Abstract
Objective: The saphenous vein subjected to arterial pressure stretches to its elastic limits and constitutes intimal hyperplasia. Sheathing of
the vein graft with pressure-resistant tubing might prolong vein graft patency. Methods: Twenty-one sheep received radial vein grafts or
hybrid grafts composed of radial vein, collagen fibrin glue and highly flexible torlen/dacron mesh tubing transplanted into the carotid artery
position. Veins were examined with the use of light and electron microscopy. Proliferating cell antigen (Ki-67) stains served as markers of
proliferation. Results: The mean wall thickness of both intimal and medial layers was evaluated. The mean intimal wall thickness was
19 ^ 11 mm in hybrid grafts vs. 24 ^ 7 mm in unsheathed grafts (P , 0:001); 22 ^ 6 vs. 26 ^ 10 mm (P , 0:001); 23 ^ 8 vs. 52 ^ 15 mm
(P , 0:001); 37 ^ 21 vs. 90 ^ 31 mm (P , 0:001); 57 ^ 31 vs. 104 ^ 28 mm (P , 0:001); 58 ^ 21 vs. 133 ^ 32 mm (P , 0:001); and
72 ^ 22 vs. 244 ^ 100 mm (P , 0:001) after respectively 5 days, 9 days, 4 weeks, 6 weeks, 8 weeks, 10 weeks and 12 weeks from
implantation. Electronic microscope examination of hybrid grafts revealed a smooth endothelial layer with intact nuclei and an intima
composed of layers of collagen and muscle fibers. In unsheathed grafts endothelial edema and nuclear destruction were observed. Conclusions: The external vein graft support with mesh tubing reduces intimal and medial layer thickening and cell proliferation in composite vein
grafts transplanted in the arterial position. q 2002 Elsevier Science B.V. All rights reserved.
Keywords: Coronary heart disease; Coronary vein graft; Hybrid graft; Remodeling; Atherosclerosis
1. Introduction
Since the beginning of the nineties the tendency towards
surgical treatment of coronary artery disease (CAD), by
means of coronary artery bypass grafting (CABG) with
the use of artery grafts, has become truly significant. The
use of radial and gastroepiploic arteries is becoming more
and more popular, while internal thoracic artery (ITA) graft
to LAD is a universally accepted standard [1]. However,
despite proven longer patency of the arterial grafts, those
made out of the autogenous saphenous vein are most
commonly used [2].
The occlusion of the venous grafts is a result of the slow
process of atherosclerosis acting over the years and progressive thickening of the intima and media acting over the first
months [3].
q
Presented at CTT 2002: Current Trends in Thoracic Surgery VIII, Fort
Lauderdale, FL, USA, January 23–26, 2002 and the 20th International
Cardiovascular Surgical Symposium, Zurs, Austria, March 2–9, 2002.
* Corresponding author. Tel.: 148-32-202-40-25, ext. 1640; fax: 14832-252-70-66.
E-mail address: [email protected] (M. Krejca).
The venous graft transplanted into the arterial system
adapts to a higher pressure acting on it by thickening of
its wall [4]. The external elastic membrane of arterial
vessels probably allows them to withstand much higher
pressures. The changes in each element of the graft wall
may affect its permeability and accelerate formation of the
atherosclerotic lesion [5,6]. The remodeling of a venous
graft is modulated by factors similar to those affecting the
vascular system, such as atherosclerosis, hypertension, and
mechanical trauma following the angioplasty.
According to the Laplace law, the pressure under the
spherical surface (single mesh in tubing net) with radius r
is described by the following formula:
p¼
2g
r
and the pressure under the cylinder surface (vein without
mesh tubing) with radius R is described by the following
formula:
p¼
1010-7940/02/$ - see front matter q 2002 Elsevier Science B.V. All rights reserved.
PII: S10 10- 7940(02)0058 7-0
2g
R
M. Krejca et al. / European Journal of Cardio-thoracic Surgery 22 (2002) 898–903
where g ¼ s £ d, s is the stress inside the wall, and d is the
wall thickness.
When considering the model of a vein without mesh
tubing one can assume that the relation between the pressure
inside the vein and stress in its wall is linear. In the model of
the vein with mesh tubing, with increasing pressure inside
the vein, the vein walls in the mesh form a spherical surface.
Its radius decreases with increasing pressure and reaches a
minimal value which is equal to the mesh radius.
As a result of that, the stress in the vein wall is increasing
slower than linear compared to the increase in the pressure.
At identical pressure inside the vein with and without the
mesh tubing and assuming that the mesh radius is equal to 1/
10 of the vein radius, the stress exerted in the meshprotected vein wall could be even 20 times lower than the
stress in the non-protected vein wall.
The goal of this research was to evaluate the efficacy of
the extravascular stent, made of a dacron mesh that we
developed, in prevention of venous grafts degeneration.
The dacron mesh placed around the graft should serve as
a support to its wall making it more resistant to high pressure, decrease tangential stress and retard graft degeneration
processes.
2. Material and methods
The research was performed in 2000 and 2001 at the
Central Animal Laboratory of the Silesian University of
Medicine, Katowice, Poland. The approval of the Local
Ethical Committee on Animal Research was obtained.
Animals received human care in compliance with the
European Convention on Animal Care.
Twenty-one male sheep of the same breed at ages
between 8 and 10 months and weighing 35–40 kg were
used. Genetically, the animals were very close.
In collaboration with Tricomed S.A. Lodz, Poland – a
company that has years of experience in vascular prostheses
production – we constructed a 4 mm in diameter mesh made
out of polyester fiber (torlen/dacron) (see Fig. 1). This stent
consists of empty rhomboidal spaces created by single interlacing torlen threads and plays a role of external support of
the native vein. This extravascular stent (patent pending) is
very resistant to bending and has an ability to change its
diameter depending on forces acting along its long axis. The
mesh constructed for further human use is 6 mm in diameter
and easily changes its diameter from 10 to 2 mm.
899
vein was harvested. All side branches were obliterated with
the 4-0 silk ligature. The vein with a diameter of 3–4 mm
formed the biological component of the hybrid graft. It was
then wrapped with the mesh and glued to it with the use of
the tissue fibrin glue Beriplast (Centeon, Mannheim,
Germany).
The coronary forceps were introduced inside the mesh
while the mesh was gently pushed to minimize its long
diameter. The 10–12 cm coronary forceps can be fully introduced inside 25 cm long mesh. The distal part of the vein
was then grasped by the forceps and the mesh was pulled on
the vein. The mesh was ligated to the needle placed in the
proximal part of the vein. The distal part of the vein was
clamped together with surrounding mesh and the vein was
filled gently with saline. With this simple maneuver the
mesh nearly automatically closely covers the vein. The
composite graft was dried with the sponge, and finally the
fibrin part of the glue and then thrombin was applied. The
construction of hybrid graft together with glue application
lasts about 3–4 min. The total length of the hybrid graft was
about 10 cm. The remaining portion of the vein was later
implanted as a classical graft.
A 10–15 cm long fragment of a jugular artery was
bypassed with the hybrid graft and then cut and ligated for
blood to flow exclusively through the graft. The end to side
anastomoses were completed with the use of Prolene 7-0
(Ethicon, Somerville, USA). The same was done on the
opposite side of the neck, where just an autologous vein
graft covered only with tissue glue (without the polyester
mesh) was used. Animals were extubated on the operating
table and allowed to recover.
2.2. Histology and immunohistochemistry
Samples of tissues were examined after 5 days, 9 days
and 4, 6, 8, 10, and 12 weeks, each time from three animals.
After premedication animals were anesthetized with Pentobarbital/Morbital 20–30 ml i.v. Samples of vessels were
2.1. Surgical technique
After premedication (Atropin 1 mg/20 kg body weight
s.c., Xylazine 2% 0.8 ml i.m.), animals were given intravenous anesthesia (Pentobarbital/Vetbutal 0.25 ml/kg b.w.
i.v., Biovetalgin/Metamizol 5–10 ml i.v.). The animals
were then ventilated (FiO2 of 0.21–0.30) and a gastric
tube was introduced to prevent flatulence.
Heparin was administered i.v. (5000 units) and the radial
Fig. 1. The polyester mesh used to prepare a hybrid graft (THG, trico hybrid
graft).
900
M. Krejca et al. / European Journal of Cardio-thoracic Surgery 22 (2002) 898–903
divided into proximal, medial and distal parts and at least
three samples from each part were obtained for particular
examination. The material to be examined using an optical
microscope was flushed with 0.9% NaCl and preserved for
20 h in 4% formaldehyde with phosphate buffer. The
sections for histology and immunocytochemistry were
dehydrated and embedded in paraffin. Sections for histology
were stained with eosin, hematoxylin and van Gieson to
make the elastic fibers visible.
Portions of vessels to be examined under electron microscopy were fixed in 2.5% glutaraldehyde with cacodyl
buffer, and osmium tetroxide, dehydrated and then
embedded in Epon fixative. Once obtained, they were
contrasted with uranyl acetate and lead nitrate for evaluation
in a transmission electron microscope.
The pictures obtained from the optical microscope were
scanned which allowed computer image analysis and
measurements of media and intima of the venous graft
(group I – radial vein RA) and of the hybrid graft (group
II – trico hybrid graft, THG). The database consists of many
hundreds of measurements at each time point to eliminate as
much as possible the potential influence of variability of
histological changes along the length of the grafts on final
study results. Intima was measured from the subendothelial
basal lamina to the internal elastic lamina, from where
media extends to the adventitial layer.
Proliferation of the medial and intimal cells was assessed
with the use of immunohistochemical reaction with the Ki67 monoclonal antigen [7,8]. The specimens were fixed
overnight in 10% neutral-buffered formalin (phosphate
buffer). The samples were subsequently passed through
graded alcohol solutions, processed three times in xylene
and finally embedded in paraffin blocks. Slices of 5 mm
thickness were placed on Apes-coated slides, deparaffinized
and rehydrated. To unmask the antigen, sections were boiled
in 0.01 M citrate buffer (pH 6.0) in a microwave oven for 10
min at 800 W. For quenching of endogenous peroxidase
activity, tissue sections were blocked with normal rabbit
serum for 10 min and incubated with primary antibodies
for 60 min at 25 8C (Ki-67 Antigen Kit, Novocastra,
Newcastle upon Tyne, UK). Next, biotinylated secondary
antibodies and ABC reagent were added (30 min, 25 8C
each). DAB and H2O2 were used as peroxidase substrates.
Finally, tissues were stained with hematoxylin, dehydrated
and coverslipped.
Under a 250 £ magnification field an assessment of a
percentage of stained, proliferating cells’ nuclei was
performed. Usually approximately 100 nuclei were found
in one field of vision. Counting was terminated once a
number of 1000 was reached or when all cells in a particular
section were evaluated.
Student’s t-test or the Mann–Whitney U-test were used.
Differences were considered significant when the P value
was less than 0.05. Results are presented as the mean ^
standard deviation.
3. Results
Thickening of the graft wall noted in both examined
groups was largely due to the overgrowth of the intimal
layer (neointima). The mean difference in the total wall
thickness between the groups was only 10%. However,
the neointima of the venous grafts was a few times thicker,
compared to the hybrid grafts, and this discrepancy
increased over time (Fig. 2).
The continuous degenerative changes in the venous graft
wall seen in the optical microscope were significantly more
pronounced in group I (venous graft). Flattened endothelial
cells, with focal loss of continuity, proliferation of fibroblasts and myofibroblasts, increasing density of collagen
fibers and spots of hyalinization in media and intima were
noted in samples collected from this group. In addition an
irregularity of collagen fibers in the surrounding tissues was
observed. Vasa vasorum were singular and visible.
In group II endothelial cells with intact continuity and
proliferation of collagen fibers on the boundary of media
and adventitia were found. The vasa vasorum were numerous, some with thickened walls. Focally, infiltration with
2.3. Statistical analysis
Normal distribution was confirmed in both groups using
the Kolmogorov–Smirnov test. For any further analysis,
Fig. 2. Changes of the thickness (in micrometers) of the intima (A), media
(B) and intima and media (C) over time in both examined groups. Group I,
RA (radial vein); group II, THG (trico hybrid graft).
M. Krejca et al. / European Journal of Cardio-thoracic Surgery 22 (2002) 898–903
901
Table 2
Changes in histological grading over time (note that only the time periods
with significant differences are shown)
Observation
Group
Results
5 days
Group I (RA)
Group II (THG)
Group I (RA)
Group II (THG)
Group I (RA)
Group II (THG)
Group I (RA)
Group II (THG)
Group I (RA)
Group II (THG)
G1
G0
G1/G2
G1
G2
G1
G2
G1/G2
G3
G1/G2
9 days
6 weeks
10 weeks
12 weeks
preserved structure of the nucleus were observed. Intima
was made of singular layers of dense collagen fibers.
The ratio of proliferating cells was assessed in the aforementioned time points with the use of immunocytochemical
reaction of monoclonal Ki-67 antibody with nuclear antigen. The proliferation ratio after 4 weeks from surgery was
similar in both groups and equaled about 8%. However,
after 6 and 8 weeks the appropriate values were 7% and
7.5% (group I) vs. 5% and 5.5% (group II). In group I the
number of proliferating cells continues to decrease after 10
through 12 weeks (6.5% and 5.5%) while in group II the
process is already nearly terminated (1.5%) (Fig. 4).
Fig. 3. Cross-section of (A) the radial vein with no stent (group I) and (B)
the radial vein with a stent (group II) assessed after 12 weeks from grafting.
Optical microscope, magnification 150 £ . Eosin, hematoxylin, van
Gieson’s staining. INT, intima between the arrows; MED, media between
the circles; ADV, adventitia.
inflammatory cells could be noted in the proximity of the
stent fibers (see Fig. 3).
For the purpose of this study the grading system of vascular changes shown in Table 1 has been applied. Table 2
shows the changes in histological grading over time (note
that only the time periods with significant differences are
shown).
The electron microscope examination of the sample slices
obtained from group I revealed a degradation of the smooth
muscle cells, often attributable to apoptosis. The cells were
shrunk and nuclei presented with numerous invaginations or
were occasionally fragmented. Often the endothelium was
covering the interrupted intimal layer. In group II endothelial cells with a smoothened cell membrane with a well
4. Discussion
A graft patency depends on its biological characteristics,
i.e. anatomy, size and physiological properties of the vessel
used for grafting [9]. Typical atherosclerotic lesions are
extremely rarely encountered in human veins, though they
develop quickly once a vein is transplanted into the arterial
system [10]. Remodeling is a homeostatic response of the
normal arteries to the changes of the blood flow characteristics and stretching of their walls. This allows for shear
stress and wall tension to be brought back to normal levels.
Table 1
Histological grading system of vascular changes
G0
G1
G2
G3
No change
Increased number of fibroblasts within intima and media
Increased intimal and medial thickness, increased number of
fibers and focal hyalinization
Markedly increased intimal and medial thickness, indistinct
borders between layers, diffuse hyaline changes
Fig. 4. Changes in the number of proliferating cells in both examined
groups over time. Group I, RA (radial vein); group II, THG (trico hybrid
graft).
902
M. Krejca et al. / European Journal of Cardio-thoracic Surgery 22 (2002) 898–903
Increased wall tension and endothelial disruption permits
blood cellular elements to penetrate the vessel wall, a
process that induces intensive proliferation. Injured
endothelium is unable to produce and secrete adequate
amounts of the antiplatelet, vasodilatating and antimitotic
factors. As a consequence, growth factors, tromboxan,
angiotensin and endothelin are released [11]. Such an abnormal wall remodeling is characterized by migration and
proliferation of smooth muscle cells from intima to media
and formation of neointima [12]. The pathological mechanisms described have certain distinct features, being associated with a specific vein physiology and histological
structure.
Unlike in the saphenous vein, the synthesis of nitric oxide
(NO) by the ITA endothelium is stimulated by plateletderived products, such as ADP and ATP. ITA endothelium
has an ability to promote vasodilation, exert antiplatelet
effects and help to break forming platelet cloths, features
that saphenous vein endothelium does not have. Hormones
like bradykinin, histamine, and substance P stimulate NO
production to a much greater degree in ITA then in saphenous vein [13]. Also prostacycline (PGI2) is secreted in
much lesser quantities in saphenous vein than in ITA [14].
With time venous grafts become more reactive to the
vasorelaxing effects of acetylcholine, ADP and thrombin,
but never to the same extent as arterial grafts [15].
Several research groups tried to modify venous grafts
with external stents in order to make them more pressureresistant [6,16–20]. In cooperation with Tricomed S.A.
(Lodz, Poland), we were able to create a mesh stent which
is highly resistant to forced bending. Several of its advantages need to be pointed out, i.e. preparing the hybrid graft
in the conditions of the operating room is simple, attaching
an external surface of the vein with the use of fibrin glue is
easy, and the torlen/dacron net remains adherent to the vein
after it is trimmed with scissors before performing an
anastomosis.
Most of the hybrid grafts described in the literature were
unique and hand-made from vascular prostheses of a different degree of porousness [6,16,19]. In 1996 Zurbrugg et al.
[17] described the use of an external stent, made of steel
mesh, and reported its protective effects on intima and
media. In our study we showed that overgrowth of both
intima and media in a vein wrapped with a stent is retarded,
and that the effect is especially pronounced in intima. We
observed the media overgrowth due to infiltration with acute
inflammatory cells and transformation of medial myocytes
into fibrocytes in both groups, however the difference is
pronounced strongly after 12 weeks.
The data presented here seem to confirm that the use of an
extravascular mesh stent significantly decreases overgrowth
of the intima that is observed after 4 weeks by the other
methods. The present study shows that the effect is already
evident in the early period of observation (5–9 days; 19–22
vs. 24–26 mm), intensifies with time (8–10 weeks; 57–58 vs.
107–133 mm), and the difference exceeds nearly three times
the initial level after 3 months from grafting (72 vs. 245
mm).
Most probably the formation of the neointima plays the
key role in a total change in intima and media thickness.
Changes occurring over time in a venous graft wall are
simply a reflection of the underlying proliferative processes.
The endothelial damage caused by surgical trauma, thrombosis and elevated transmural pressure leads to intimal
fibrous proliferation. We can speculate that the mesh use
should decrease the transmural pressure. The histological
changes related to increased transmural pressure are probably becoming more and more evident within weeks from
grafting. The early changes are mostly due to endothelial
damage caused by surgical trauma.
Angellini et al. [20] noted a significantly lower ratio of
proliferating cells in a wall of a vein supported with an
extravascular stent. Rodriguez et al. [21] described proliferation of non-smooth muscle cells (NSMC) in the early
period following grafting. Both authors used the PC-10 antigen (anti-proliferating cell nuclear antigen: PCNA). PCNA
is a nuclear antigen, present at low, but detectable concentrations in nonproliferating cells. Therefore, it makes a
useful tool for qualitative assessment, although its ability
to determine changes that occur later over time is limited.
Antigen reacting with Ki-67 antibody undergoes a very fast
catabolism at the end of phase M of cell proliferation [7,8].
This helps to determine changes in proliferative processes
with greater accuracy. The differences become significant as
early as after 6 weeks: the proliferation rate in stent-free
veins (group I) decreases slightly at that time, while it is
nearly abolished within hybrid grafts (group II). A lower
proliferation rate on day 5 in group II can be explained by
a delayed triggering of mechanisms that stimulate that
process. At those very early stages it is mostly due to
mechanical trauma to the endothelium. It is worth remembering though, that the hybrid graft was constructed from
the same portion of the vein as was the standard graft, not to
mention that it underwent manipulations aiming at placing
the vein inside the dacron net and gluing with elevated
intravenous pressure that could further damage the endothelium. Stooker et al. [22] showed, in an in vitro model, the
protective effect of the extravascular stent on the endothelium in human saphenous vein perfused for 1 h with blood.
We can therefore assume that early endothelium reparative
processes (re-endothelialization) occur at faster rates in
veins reinforced with extravascular stents, and that the
degenerative processes causing higher permeability of the
vein are activated later. The fibrin glue, commonly used for
hybrid graft construction, probably has no influence on the
vein wall structure and function [23]. The fibrin sealant
allowed a vein wall and a mesh to adhere and avoid formation of inner lumen foldings [17].
The presence of a stent made a vein lumen remain widely
open even in an empty state and suturing an anastomosis
incorporating dacron fibers was easy. Fibrin sealant exerts
some positive influences on wound healing: it promotes
M. Krejca et al. / European Journal of Cardio-thoracic Surgery 22 (2002) 898–903
granulation, early fibroblast proliferation, accelerated reepithelization and revascularization as well as inhibiting
leukocyte infiltration [24]. Because of its biocompatibility
the fibrin glue has also been advocated as a substrate for
endothelial cell seeding in vascular prostheses [25].
The extravascular dacron mesh stent wrapped around a
vein that is to be implanted into the arterial system prevents
the hypertrophy of the graft’s wall, impedes the overgrowth
of the intima and decreases the proliferation rate of venous
graft cellular elements.
Acknowledgements
The project was supported by a grant from Silesian
University of Medicine, Katowice, Poland.
References
[1] Lytle BW, Loop FD, Cosgrove DM, Ratliff NB, Easley K, Taylor PC.
Long term (5–12 years) serial studies of internal mammary artery and
saphenous vein coronary bypass grafts. J Thorac Cardiovasc Surg
1985;89:248–258.
[2] Lytle BW, Loop FD, Thurer RL, Groves LK, Taylor PC, Cosgrove
DM. Isolated left anterior descending coronary arteriosclerosis: longterm comparison of internal mammary artery and venous autografts.
Circulation 1980;61:869–874.
[3] Dobrin PB, Littooy FN, Endean ED. Mechanical factors predisposing
to intimal hyperplasia and medial thickening in autogenous vein
grafts. Surgery 1989;105:393–400.
[4] Glagow S, Zarins CK. Is intimal hyperplasia an adaptative response or
a pathologic process? Observation on the nature of nonatherosclerotic
intimal thickening. J Vasc Surg 1989;10:571–573.
[5] Berceli SA, Borovetz HS, Sheppeck RA, Moosa HH, Warty VS,
Armany MA, Herman IM. Mechanisms of vein graft atherosclerosis:
LDL metabolism and endothelial reorganisation. J Vasc Surg
1991;13:336–347.
[6] 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. J Thorac Cardiovasc Surg 1986;92:330–336.
[7] Gerdes J, Li L, Schlueter C, Schlueter C, Duchrow M, Wohlenberg C,
Gerlach C, Stahmer I, Kloth S, Brandt E, Flad HD. Immunobiochemical and molecular biologic characterization of the cell proliferationassociated nuclear antigen that is defined by monoclonal antibody Ki67. Am J Pathol 1991;138:867–873.
[8] McCormick D, Yu C, Hobbs C, Hall PA. The relevance of antibody
concentration to the immunohistological quantification of cell proliferation-associated antigens. Histopathology 1993;22:543–547.
903
[9] Luscher TF. Vascular biology of coronary bypass grafts. Curr Opin
Cardiol 1991;6:868–876.
[10] Campeau LC, Enjalbert M, Lasperance J, Bourassa MG. The relation
of risk factors to the development of atherosclerosis in saphenousvein grafts and the progression of disease in native circulation. N Engl
J Med 1984;311:1329–1332.
[11] Walton KW, Slaney G, Ashton F. Atherosclerosis in vascular grafts
for peripheral vascular disease. Atherosclerosis 1985;54:49–64.
[12] Andersen HR, Maeng M, Thorwest M, Falk E. Remodeling rather
than neointimal formation explains luminal narrowing after deep
vessel wall injury; insights from a porcine coronary (re)stenosis
model. Circulation 1996;93:1717–1721.
[13] De Ruiter MC, Pocimann RE, vanMunsteren JC. Embryonic endothelial cells transdifferentiate into mesenchymal cells expressing smooth
muscle actins in vivo and in vitro. Circ Res 1997;80:444–449.
[14] Schwartz SM. Perspective series: cell adhesion in vascular biology.
Smooth muscle migration in atherosclerosis and restenosis. J Clin
Invest 1997;99:2814–2820.
[15] Nabel EG, Yang Z, Plautz G. Recombinant fibroblast growth factor-1
promotes intimal hyperplasia and angiogenesis in arteries in vivo.
Nature 1993;362:844–851.
[16] Parsonnet V, Attai Lari A, Shah IH. New stent for support of veins in
arterial grafts. Arch Surg 1963;87:696–702.
[17] Zurbrugg HR, Wied M, Angelini GD, Hetzer R. Reduction of intimal
and medial thickening in sheathed vein grafts. Ann Thorac Surg
1999;68:79–83.
[18] Kohler TR, Kirkman TR, Clowes AW. The effect of rigid external
support on vein graft adaptation to the arterial circulation. J Vasc Surg
1989;2:277–285.
[19] Violaris AG, Newby AC, Angelini GD. Effects of external stenting on
wall thickening in arteriovenous bypass grafts. Ann Thorac Surg
1993;55:667–671.
[20] Angelini GD, Izzat MB, Bryan AJ, Newby AC. External stenting
reduces early medial and neointimal thickening in a pig model of
arteriovenous bypass grafting. J Thorac Cardiovasc Surg
1996;1:79–84.
[21] Rodriguez E, Lambert EH, Magno MG, Mannion JD. Contractile
smooth muscle cell apoptosis early after saphenous vein grafting.
Ann Thorac Surg 2000;70:1145–1153.
[22] Stooker W, Niessen HW, Baidoshvili A, van Hinsbergh VW, Wildevuur CR, Eijsman L. Perivenous support reduces early changes in
human vein grafts: studies in whole blood perfused human vein
segments. J Thorac Cardiovasc Surg 2001;2:290–297.
[23] Rousou J, Levitsky S, Gonzales-Lavin L. Randomized clinical trial of
fibrin sealant in patients undergoing resternotomy or reoperation after
cardiac operations. J Thorac Cardiovasc Surg 1989;97(2):1194–1203.
[24] Jackson MR, Alving BM. Fibrin sealant in preclinical and clinical
studies. Curr Opin Hematol 1999;6:415–419.
[25] Zitta P, Deutsch M, Fischlein T, Minar E. Endothelial cell seeding of
polytetrafluoroethylene vascular grafts in humans. A preliminary
report. J Vasc Surg 1987;6:535–541.