M yointimal thickening in experimental vein
grafts is dependent on wall tension
Lewis B. Schwartz, MD, Martin K. O'Donohoe, MB, FRCSI, Cemil M. Purut, MD,
Eileen M. Mikat, PhD, Per-Otto Hageri, PhD, and Richard L. McCann, MD,
Durham) N. C.
This study examines the relative contributions of intraluminal pressure, blood flow, wall
tension, and shear stress to the development ofmyointimal thickening in experimental vein
grafts. To study these different hemodynamic parameters, several experimental models
were created in 30 New Zealand White rabbits separated into six groups: common carotid
interposition vein grafts harvested at 4 weeks (VG-4) or 12 weeks (VG-12), common
carotid-linguofacial vein arteriovenous fistulas harvested at 4 weeks (AVF-4) or 12 weeks
(AVF-12), AVFs with partial outflow obstruction harvested at 4 weeks (AVFobs), and
combination VG-AVFs in series harvested at 4 weeks (VGAVF). Blood pressure and flow
in the graft or vein were measured by use of a transducer-tipped pressure catheter and
electromagnetic flow meter. At harvest, veins were perfusion-fixed and proximal, middle,
and distal sections were subjected to c.omputerized morphometric .analysis•. Vein grafts
were characterized by a high mean pressure (VG-4, 51 ± 4; VG-12, 62 ± 3 nim Hg), low
mean flow (VG-4, 17 ± 1; VG-12, 16 ± 4 ml/min), large luminal area (VG-4,
19.7 ± 2.4; VG-12, 19.3 ± 3.9 mm2 ), high wall tension (VG-4, 17.0 ± 1.5; VG-12,
19.5 ± 2.4 x 103 dyne/cm), low shear stress (VG-4, 0.75 ± 0.13; VG-12, 0.96 ± 0.38
dyne/cm2 ), and a high degree of myointimal thickening (VG-4, 5.89 ± 0.90; VG-12,
4.72 ±·0.83 mm2 ). Arteriovenous fistulas were characterized by a low mean pressure
(AVF-4,5 ± I,AVF-12,6 ± 2mmHg),elevatedbloodftow(AVF-4,82 ± 16;AVF-12,
82 ± 17 ml/min), small luminal area (AVF-4, 2.43 ± 0.58; AVF-12, 7.14 ± 2.68), low
wall tension (AVF-4, 0.62 ± 0.19; AVF-12, 0.89 ± 0.24 x 103 dyne/cm), elevated shear
stress (AVF-4, 108 ± 32; AVF-12, 71 ± 50 dyne/cm2 ), and decreased myointimal area
(AVF-4, 1.18 ± 0.26; AVF-12, 1.90 ± 0.55 mm2). The addition of outflow obstruction
toAVFs (AVFobs) resulted in elevated pressure (48 ± 2mmHg),decreasedftow(17 ± 4
ml/min), larger luminal area (8.71 ± 2.31 mm2), elevated wall tension (10.3 ± 1.7 x 103
dyne/cm), and a degree of myointimal thickening approaching that of vein grafts
(3.79 ± 0.66 mm2)~ The addition of fistulas to vein grafts (VGAVF) resulted in slightly
lower pressure (37 ± 2 mm Hg), elevated ftow (70 ± 11 ml/min), decreased luminal area
(14.7 ± 1.7mm2 ) decreased wall tension (10.7 ± 1.0 x 103 dyne/cm),andareductionin
myointimal area (2.78 ± 0.20 mm2 ;p < 0.05 compared to VG-4). Myointimal area most
strongly correlated with intraluminal area (r = 0.7210) and wall tension (r = 0.6846),
whereas weaker correlations were found with intraluminal pressure (r = 0.4419), blood
flow (r = 0.2454), and shear stress (r = 2887). It is concluded that the major stimulus
for vein graft myointimal thickening in this model is increased wall tension that causes
deformation of the vessel wall. Interventions that minimize vein graft wall tension may
serve to limit overall myointimal thickening, although luminal area can be expected to
decrease accordingly. (J VASe SUB..G 1992;15:176-86.)
From the Departments of Surgery (Drs. Schwartz, O'Donohoe,
Purut, Hagen, McCann), Pathology (Dr. Mikat), and Biochemistry (Dr. Hagen), Duke University Medical Center.
Supported by U. S. Public Health Service grants HLI5448,
HL32720, and HL08086.
Winner of the 1991 Liebig Foundation Award.
Presented at the Thirty-ninth Scientific Meeting of the International Society for Cardiovascular Surgery, North American
Chapter, Boston, Mass., June 3-4, 1991.
Reprint requests: Richard L. McCann, MD, Associate Professor,
Department of Surgery, Duke University Medical Center, Box
no. 2990, Durham, NC 27710.
24/6/33805
176
Myointimal thickening of vein grafts was first
noted by Carrel and Guthrie1 in their landmark study
of biterminal canine venous transplantation in 1906.
Since .that time myointimal thickening has been
documented in vein grafts in virtually every species
studied including rats, 2 rabbits,3-7 dogs,8-27 sheep,28
monkeys,29-31 and humans.9.10.32-42 It is now widely
accepted that vein transplantation into the arterial
circuit invariably stimulates smooth muscle cell
proliferation and some degree of intimal thickening.
Volume 15
Number 1
January
Myointimal thickening in vein grafts
1992
Despite wide acceptance .of this phe~omenon,
'derable controversy remaIns as to Its stImulus. A
conSl
·ety of inciting factors have been suggested,
.
varl
2039
. I ding harvestIng
techn°Ique,'
venous w all
~chCU 9-11 elevated wall tension with subsequent
15 emia,
.
dl al . .
mechanical wall deformatIon, * an or teratIons m
blood flow velocity or shear stress.14,16,17,21,22,24-2~~he
urpose of the pr~sent study wa~ t~ create distmct
modynamic environments for sllmlar segments of
;ew Zealand Whi~e ra?bit external jugu1~ veins to
assess relative contrIbutIons ofhemodynamIc parameters to myointimal thickening.
&
177
Table I. Experimental groups
Group
No.
SU"l!ical procedure
VG-4
VG-12
AVF-4
5
5
5
AVF-12
5
AVFobs
5
VGAVF
5
Carotid interposition vein graft
Carotid interposition vein graft
Carotid-linguofacial vein arteriovenous fistula
Carotid-linguofacial vein arteriovenous fistula
Arteriovenous fistula with outflow
obstmction
Vein graft-arteriovenous fistulas in
series
"Harvest
(weeks)
4
12
4
12
4
4
MATERIAL AND METHODS
Animal operations
Surgical procedures were performed in 30 2 to
2.5 kg New Zealand White rabbits separated into six
equal groups (Table I). Animal care complied with
"Principles of Laboratory Animal Care" and the
"Guide for the Care and Use ofLaboratory Animals"
(NIH Publication No. 80-23, revised 1985). Anesthesia was induced and maintained with subcutaneously injected ketamine hydrochloride (60 mg/kg,
Ketaset; Bristol Laboratories, Syracuse, N.Y.) and
xylazine (6 mg/kg, Rompun; Mobay Corp., Shawnee, Kan.). Neither mechanical ventilation nor intravenous fluid administration was required.
Carotid interposition vein grafts were performed
as previously described. 6 In brief, the vein was
harvested and placed nondistended and nonperfused
in lactated Ringers solution at room temperature.
Intravenous heparin was administered (200 U/kg;
Elkins-Sinn, Inc., Cherry Hill, N.J.), and the carotid
artery was dissected and clamped. The vein was
reversed, and anastomosis was performed end to side
by use of continuous 10-0 nylon suture (Ethicon,
Inc., Somerville, N.J.). The anastomoses were created so that the original in situ length of the vein
segment was preserved. The intervening carotid
segment was ligated and divided, and the clamps
were removed.
To create a model with high flow and low
pressure, arteriovenous fistulas were performed by
side-ta-side anastomosis of the linguofacial vein and
distal common carotid artery by use of an identical
anastomotic technique. Although the entire length of
the external jugular vein was not used during the
~peration, it was dissected free from the surrounding
tIssue to make the animal groups comparable.
To create a model with limited flow and high
p~essure, a partial outflow obstruction was placed
distal to surgically created arteriovenous fistulas
·References 4, 5, 12, 13, 15, 18, 19, 25-28, 32, 33, 36, 42, 43
(AVFob groups). Mer completion ofthe anastomosis, a stainless steel clip with a 1 mm opening was
placed around the external jugular vein near the
thoracic inlet. The clip was secured with a si~gle 4-0
nylon suture.
To examine a model in which grafts were exposed
to increased flow, a final group was enrolled consisting.ofvein grafts and arteriovenous fismlas in series
(VGAVF). This procedure necessitated excision of
the left external jugular vein for use as the graft
conduit since the right vein was needed for fisrula
outflow. The vein grafts were completed first, followed by the fistulas. Because of the length of the
procedure, an additional dose of heparin was .administered as well as infrequent additional doses of
anesthetic agents.
The characteristics ofthe outflow beds created by
these four surgical procedures in New Zealand White
rabbits was the subject of a previous communication
from this laboratory.44
Hemodynamic measurements
Diameters of the midportions of the veins were
measured with a digital caliper (Ultra-Cal IT; Fred V.
Fowler Co., Inc., Newton, Mass.) at the time of
surgery and at harvest. This was done to correct for
shrinkage in subsequent histologic measurements
(vide infra).
Measurement of intraluminal lateral blood pressure was performed by use of a 22-gauge tapered
pediatric intravenous cannula (Medicut; Sherwood
Medical, Tullamore, Ireland) secured with a 7-0
Prolene (Ethicon) purse-string suture. The fluidfilled cannula was connected to a stopcock assembly
housing a 5F transducer-tipped pressure catheter
(Micro-tip; Miliar, Inc., Houston, Texas). Lateral
blood pressure was measured in vein grafts via direct
insertion and in arteriovenous· fistulas via insertion
through a side branch. Blood flow was measured by
use of an electromagnetic flowmeter (SOlD; Caro-
Journal of
VASCULAR
SURGERY
178 Schwa-m et al.
lina Medical Electronics, Inc., King, N.C.) with
appropriately calibrated and sized probes. In all cases,
hematocrit was assumed to be 35%.
Pressure and flow waveforms were simultaneously recorded on strip chart paper (Model 7D
Polygraph; Grass Instrument Co., Quiney, Mass.)
and digitized at 200 Hz by use of a PC-based
acquisition system (Zenith Data Systems, Benton
Harbor, Mich.; Lab Master DMA; Scientific Solutions, Inc., Solon, Ohio). All mean values were
calculated from the digitized data. Measurements
were taken both at the time of operation and at
harvest; the average of the two measurements was
used in the final data analysis.
Harvest and histology
After the designated time interval (see Table I),
the animal was reanesthetized and the graft exposed.
After dissection and hemodynamic measurements,
the animal was heparinized and the incision was
extended inferiorly through the sternum. The bifur-.,
cation of the right common carotid and right
subclavian arteries was identified and cannulated.
After sacrifice, the carotid artery was perfused with
3.7% buffered formaldehyde, at the last recorded
blood pressure. Arteriotomy or fistulotomy distally
provided outflow for the fixative solution. Intraluminal blood pressures in the AVF-4 and AVF-12
groups were so low that these veins were immersion
fixed only.
After fixation, the enernal jugular vein (whether
graft, fistula, or combination) was excised. It was
placed in fixative and divided· into three equal
portions of approximately 1 cm length. All sections
were at least 0.5 cm from any anastomosis; anastomoses were not histologically examined in this study.
The portions were labelled as "proximal," "middle,"
and "distal" according to the direction ofblood flow,
not the original anatomic configuration. Note that
the study was designed so that for each experimental
group the direction of blood flow was opposite that
of the native vein.
The veins were stored at 4 0 C Wltil sectioning.
Sections measuring 6 f.Lm in thickness were cut and
stained with modified Masson's trichrome and VerhoefPs elastic tissue stain.45 Three consecutive sections 'were examined for each segment (270 total
sections). Total vessel area, myointimal area, and
luminal area were measured with the aid of a
computerized morphometric workstation (Videometric 150; American Innovision, San Diego, Cali£).
Total and luminal areas were quantified by use of a
digitizing pad, whereas myointimal area was quantified by use of thresholds of the characteristic red
color of smooth muscle cells stained with this
method. This was done in an effort to minimize
observer bias in the measurement ofmyointimal area.
All diameter measurements were derived from the
appropriate area measurements assuming a circular
lumen. To correct for tissue shrinkage during processing,46 a plot was made of in vivo measured
diameter versus histologically measured total diameter (calculated on the basis of total area; diameter,
2Varea/1r). This is shown in Fig. 1. Regression
analysis revealed that histologic diameter underestimated true diameter by approximately 10% (slope,
0.98; y-intercept, -0.48). Therefore all histologic
diameter measurements were corrected by 10% and
area measurements by 20%.
Calculations and statistical analysis
Measured variables included intraluminal blood
pressure (P), blood' flow (Q), total area (~»)
myointimal area (~), and luminal area (A;.). Assumed constants included blood viscosity (11 = 0.03
poise) .and specific gravity of blood (p = 1.056
g. cm~3). Derived dimensional variables included
luminal radius:
ri=~
and myointimal thickness:
hIII
=
~ - [.
~
'TT
1
Derived hemodynamic variables included wall
tension per unit length (Tw = P X ri ), shear stress
(rs = 4T)Q!1Tr/), and Reynold's number (2pQ!'TlT/ll)·
The dimensions ofthe middle segment were used for
hemodynamic variable analysis since pressure and
flow measurements were performed near the midpoint. Only open measurements were obtained so
wall tension calculations are appro~ations of the
true in situ values. For purposes of analysis, blood
flow was assumed to be laminar.
Statistical evaluation for the six groups was
performed by use of one·way analysis of variance
(ANOVA). For p values less than 0.05, differences
between groups were tested with the Bonferroni
correction. Linear regression was performed by the
method of least squares. For all tests of inference,
p values less than 0.05 were considered significant.
Volume 15
Number 1
January 1992
Myointimal thickening in vein grafts
179
0
m
n=30
y=O.98x-O.48
.s
co
p<O.0005
r 2=O.6982
"CD
,....
.......
e
....,
•E
.!!
Q
o
o
o
fQ
.2 It)
tta
.2
0
...,.
.....
Regression Line
:f
o
et')
o
N
3
2
5
4
6
7
Intra-operative Diameter
9
8
10
(mm)
Fig. 1. Comparison of measured intraoperative diameter and histologic diameter for 30
rabbits. Note that the histologic diameter underestimates intraoperative diameter byapproximately 10% (identity line drawn for comparison). Therefore all histologically derived diameter
measurements were corrected by 10% and all area measurements by 20% (see text) .
.Table Ill. Calculated hemodynamic
parameters *
Table 11. Hemodynamic measurements
Heart rate
Group
VG-4
VG-12
AVF-4
AVF-12
AVFobs
VGAVF
(bpm)
152
157
147
142
149
153
± 6
± 6
± 10
± 20
± 7
± 7
Intraluminal
pressure
(mmHg)
51
62
5
6
48
37
±
±
±
±
±
±
4
3
1*
2*
2
2*
Bloodjlqw
(ml/min)
17
16
82
82
17
70
±
±
±
±
±
±
1
4
16*
17*
4
11*
*p < 0.05 compared with VG-4 (ANOVA/Bonferroni).
Wall tension
Group
VG-4
VG-12
AVF-4
AVF-12
AVFobs
VGAVF
(xl0 3 dyne/cm)
17.0
19.5
0.00617
0.00891
10.4
10.7
±
±
±
±
±
±
1.5
2.4
0.00189t
0.00240t
1.8t
1.Ot
Shear stress
(dyne/cm2)
0.75
0.96
108
71
4.0
4.9
±
±
±
±
±
±
Reynold~s
number
0.13
0.38
32
50
1.5
1.2
22
21
310
220
36
110
*Computed by use of middle segment dimensions.
tp < 0.05 compared with VG-4 (ANOVA/Bonferroni).
RESULTS
Hemodynamics
Results of the measured and calculated hemodynamic parameters are shown in Tables 11 and Ill.
Intraluminal mean blood pressure in 4- and 12-week
vein grafts was 51 ± 4 mm Hg and 62 ± 3 mm Hg,
respectively. These were somewhat lower than values
previously measured for New Zealand White rabbits
likely because of the effect of long periods of
~esthesia.3 As expected, blood pressure was significantly reduced in arteriovenous fistulas (AVF-4
5 ± 1; AVF-12 6 ± 2 mm Hg). The addition of
outflow obstruction to the arteriovenous fistula
(AVFobs) resulted in an elevation in mean pressure
t~ 48 ± 2 mm Hg (p < 0.05). The addition of a
~~ fistula to the vein graft (VGAVF) resulted in a
significant decrease in mean blood pressure (37 ± 2
mm Hg; P < 0.05 compared to VG-4) because of
the close proximity of the fistula.
The mean blood flow of vein grafts in groups
VG-4 and VG-12 was 17 ± 1 and 16 ± 4 ml/min,
respectively. The creation of a fistula increased flow
approximately fourfold to 82 ± 16 and 82 ± 17 in
theAVF-4andAVF-12 groups"(p < 0.05). Outflow
obstruction (AVFobs) significantly reduced flow to
levels near that of the vein grafts. Addition of a distal
fistula to the vein graft (VGAVF) resulted in flow
nearly equal to that of the fistula alone.
.
Large differences were observed in calculated wall
tension bet\Veen the various groups (Table Ill). Vein
graft wall tensions were 17.0 ± 1.5 (VG-4) and
19.5 ± 2.4 x 103 dyne/cm (VG-12). Wall tension
in the control vein and arteriovenous fistula groups
was approximately 1 % to 5% of the vein graft level.
Journal of
VASCULAR
SURGERY
180 Schwartz et al.
cv
VG-4
VG-12
AVF-4
AVF-12
AVF
ob.
VGAVF
Fig. 2. Representative histologic cross sections_ . Vessels were perfusion-fixed at the last
recorded intragraft or intra-AVF mean blood pressure. (modified Masson's trichrome and
Verhoeff's elastic tissue stain; original magnification x 680.) As the samples are aligned with the
outer margin of smooth muscle cells at the bottom of the figure, the total thickness of the
myointimallayer is shown. Note the increased myointimal area ofvein grafts (VG-4 and VG-12)
compared with control ve~ (CV? and arteriovenous fistulas (A VF-4 and A VF-12) not operated
on. Intermediate degrees of myointimal thickening were observed in the A VFobs and VGA VF
groups.
Significant reductions were also noted in the AVFobs
and VGAVF groups.
There were large variations in shear stress across
the six groups but ANOVA did not identify intergroup differences because of the large standard
errors. Shear stresses in the vein graft groups were less
than 1 dyne/cm2 • Because shear stress varies direcdy
with flow and inversely with the third power of
radius, it is not surprising that shear stress is minimal
in the low flow but gready dilated grafts. In contrast,
shear stress approached 100 dyne/cm2 in the arteriovenous fistulas. Intermediate shear stress values were
calculated for the AVFobs and VGAVF groups.
Reynold's numbers ranged from about 10 to 400 and
were largest in the arteriovenous fistula groups.
Morphometric analysis
Representative histologic sections for each group
and for nonoperated immersion-fixed control veins
are shown in Fig. 2. Quantitative myointimal area
(.A.n) by group and location is given in Fig. 3.
Myointimal area increased more than 10 times in the
vein grafts as is characteristic of this model. No
significant difference was observed in myointimal
area· in 4-week and 12-week grafts. Myointimal area
was significandy reduced in the arteriovenous fistula
groups (p < 0.05). It appeared that the AVF-12
group developed more myointimal area than the
AVF-4 group, but this difference was not statistically
significant. The addition of outflow obstruction to
the fistula (AVFobs) resulted in increased myointi..
mal area that was not significandy different from
vein grafts. The addition of a distal fistula to the vein
grafts (VGAVF) resulted in decreased myointimal
area compared to VGs; this difference reached
statistical significance for middle and distal segments
only.
Regression analysis
With use of univariate linear regression, factors
associated with increased myointimal area were
identified. Positive correlations were found when
myointimal area was compared with luminal area
(r 2 = 0.7210; Fig. 4), wall tension (r 2 = 0.6846;
Fig. 5) or intraluminal pressure (r 2 = 0.4419; Fig.
6) as opposed to much weaker negative correlations
with flow (r2 = 0.2454; Fig. 7) and shear stress
(r 2 = 0.2887).
Volume 15
Number 1
January 1992
Myointimal thickening in vein grafts 181
o
Proximal Segments
~ Middle Segments
~ Distal Segments
o
VG-4
VG-12
AVF-4
AVF-12
VGAVF
AVF
obs
Fig. 3. Myointimal area (Am) by group and segment location. *p < 0.05 compared with VG-4
(ANOVA). The myointimal areas of the A VF-4, andAVF-12 groups were significantly smaller
than VG-4 for each of the proximal, middle, and distal segments. The myointimal area of the
VGA VF group was significantly smaller than VG-4 for middle and distal segments only.
0)
~
oS
a
n=30
r 2 =O.7210
D
CD
'"....
Q)
c:e
'ii
.~
.5
C")
0
::...
::E
0
0
5
10
15
20
25
Luminal Area (m",z)
Fig. 4. Least squares linear regression comparing myointimal area and luminal area.
DISCUSSION
Myointimal hyperplasia is the universal response
observed when veins are transplanted into the arterial
circulation. Despite the ubiquitous nature of the
proliferative response, many questions remain concerning the mechanism and extent of myointimal
thickening. The most pressing questions are, (1 )
What are the inciting stimuli that cause smooth
muscle cells to proliferate, and (2) why is the process
self-limited and insignificant in some cases while in
others it progresses to luminal obstruction and
compromise of graft function?
This study attempts to identify the initial stimulus
for vein graft myointimal thickening. Four distinct
hemodynamic environments were created for the
same segment of external jugular vein in New
Zealand White rabbits. The results suggest that the
primary stimulus for the development of myointimal
thickening is increased wall tension that causes
circumferential deformation (increased luminal
Journal of
VASCULAR
SURGERY
182 Schwartz et al.
0)
0
D
~
~
CO
tD
CD
....
oq:
to
.§
.S
...
D
(I')
D
0
~
0
0
5
10
20
15
25
Wall Tension (x1a' dyne/cm)
Fig. 5. Least squares linear regression comparing myointimal area and wall tension.
D
D
[]
D
D
D
D
D D
D
D
~
D
D
o
o
c
25
50
75
Pressure (mmHg)
Fig. 6. Least squares linear regression comparing myointimal area and intraluminal
pressure.
cross-sectional area). It is postulated that the increased wall tension stimulates smooth muscle cell
proliferation to such a degree that the final ratio of
myointimal are to luminal area is normalized.
Support for the hypothesis of tensiondeformation comes from several groups with use of
a variety of models. Both Vlodaver et al. 32 and Kern
et al. 33 studied human aortocoronary grafts in the
early 1970s and suggested that hemodynamic stress~
likely accounted for the observed histologic changes.
Animals studies were first performed by Brody
et al. II ,12 In 1972, in which reversed and in situ vein
and arterial interposition.grafts were created in dogs.
Medial fibrosis was seen in all dissected veins but
intimal proliferation was seen only in grafts subjected
to the arterial environment. In a similar study, Storm
et al. 13 studied the effects of mechanical dilation and
adventitial stripping but found that significant intimal hyperplasia was independent of preinsertion
manipulation. This suggested that the response
conformed to a set physiologic pattern and was due
to increased pressure.
Further support for the deformation and vein
remodeling theory can be derived from vein wrap-
Volume 15
Number I
January 1992
Myointimal thickening in vein grafts 183
D
D
~
D
~ co
...,
D
rP
CD
'-
D
[]
'lC(
D
"ii
.~
c:
·0
(I')
:::.....
~
o
o
50
100
150
Flow (ml/min)
Fig. 7. Least squares linear regression comparing myointirnal area and blood flow.
ping experiments. Karayannacos et al. 18 studied the
effects of wrapping canine vein bypass grafts with
Dacron mesh, which constricted the diameter of the
graft by 10%. The combined intimal~medial thickness was substantially less than in control vein grafts,
which the authors attributed to decreased dilation.
Similar experiments were performed by Barra et al. 28
in vein grafts in sheep. A 7 mm mesh tube was
applied to a jugular vein interposition graft, which
prevented distention by approximately 50%. Intimal
thickness was less in the wrapped grafts, and the
cellular pattern appeared significantly more regular.
They hypothesized that the reduction in medial
stretching was partially responsible for the observed
effect. More recently, wrapping experiments were
performed in the rabbit jugular vein graft model by
Kohler et al. 5 They showed that total cross-section
wall area, smooth muscle cell volume, and matrix
deposition were all decreased in segments tightly
wrapped (2.5 to 3 mm diameter polytetrafluoroethylene) compared to loosely wrapped or 'control
segments. Thus the effect ofvein graft wrapping with
its attendant decrease in deformation and wall
tension has been firmly established.
. More extensive experiments specifically addressIng the mechanism of myointimal thickening have
also been undertaken. Zwolak et al.,4 also using the
rabbit carotid interposition model, studied vein
grafts at intervals from 1 hour to 12 weeks. Morphometric measurements showed that the ratio of
luminal radius to wall thickness decreased steadily,
finally approaching carotid arterial values at 4 to 12
weeks. As in the present study, they noted no increase
in smooth muscle cell area after 4 weeks. It was
suggested that tangential stress on the smooth muscle
cells resulted in proliferation and protein synthesis in
an "attempt" to normalize radius to thickness ratio.
Final evidence for the importance of stress and
deformation comes from Dobrin et al. 25,26 in two
related studies of canine vein grafts. The authors
created several venous environments that differed
with respect to circumferential, longitudinal, and
radial stress, and blood flow velocity. They showed
that cuffed segments (preventing distention) developed less intimal and medial thickening than control
segments. They also showed that carotid interposi~
tion vein grafts in which the proximal carotid artery
was stenosed (resulting in a significant pressure drop
but little change in vein graft diameter) did not alter
intimal or medial characteristics. The authors concluded that the stimulus for medial hyperplasia was
mechanical deformation.
Indirect evidence for the modulating effects of
deformation and wall tension can also be found in cell
culture experiments and studies on the arterial wall.
In 1976 Leung et al. 47 stimulated rabbit aortic
smooth muscle cells with cyclic stretching in culture
and noted a twofold to fourfold increase in matrix
protein synthesis. Increases in actual endothelial cell
number as a result of experimentally created hypertension have also been reported. 48 Regarding the
composition of the arterial wall, studies have shown
that (1) the ratio of thoracic aortic radius to medial
lamellar unit (and radius to medial thickness) is nearly
a constant across mammalian species,49 (2) increased
stress in experimental hypertension produces medial
184 Schwartz et al.
accumulations of fibrous proteins,50 and (3) medial
cells adapt their synthetic response according to wall
tension during normal aortic and pulmonary trunk
growth. 51 ,52 In addition, atherosclerotic arteries have
larger diameters than their normal counterparts,
presumably in an attempt to increase luminal
area. 53,54 The finding, in this study and others,4 of
A;./~ "regulation" is particularly noteworthy since it
suggests that veins can "regulate" wall thickness at
nonphysiologic pressure and tension. We speculate
that attempts at minimizing wall tension in human
vein grafts either by constrictive wrapping28 or
adjunctive distal arteriovenous fistula 5S -S7 probably
result in less myointimal thickening. However, if the
A;./~ ratio is constant for a given vein, then less
luminal area will also be created, and this may be
detrimental to patency.
Although the results ofthese experiments suggest
that elevated wall tension with subsequent wall
deformation is the primary stimulus for myointimal
thickening, some caution should be exercised in their
interpretation. First, no attempt was made to separate intima from media in our studies since we feel
that this distinction was not possible for the arteriovenous fistula groups (Le., the internal elastic lamina
was impossible to identify with regularity). Therefore
our results address only total smooth muscle cell
cross-sectional area and cannot separate possible
differences in medial and intimal proliferation (if
such a true functional separation exists). Second, our
sttldies reveal that elevated wall tension with resultant
deformation is the most important s~ulus for
venous myointimal thickening but 'Ye do not suppose that it is the only variable that may modulate
cellular changes. It has been repeatedly theorized
that, when tension is held constant, lower Bow
velocities result in greater myointimal thickening.
Certain well-designed studies prove conclusively
that, as in arteries,58-65 Bow velocity and shear forces
play a role in vein graft remodeling. For example, in
the often quoted study by Berguer et al.,22 experiments were designed so that the intraluminal pressure
in two yein graft groups must have been equal
although .blood flow was markedly different. The
vein graft segment with lower Bow exhibited greater
intimal thickening that must be attributed to differences in Bow. Likewise, in the studies by Dobrin,
although pressure was not explicitly measured in the
grafts subjected to high Bow (common femoral vein
grafts with ligation of the native common femoral
artery designated as "Step 1" experiments), the only
reasonable conclusion is that greater velocity of Bow
resulted in less intimal hyperplasia?6 Finally, exper-
Journal of
VASCULAR
SURGERY
iments in dogs performed by Morinaga, et al. 24
showed that grafts with equivalent measured mean
blood pressures but lower flow rates and variations in
shear stress developed more profound intimal lesions,
and it is interesting to note that these lesions
regressed when flow conditions were normalized. In
agreement with these findings, flow and shear stress
in the present study were weakly negatively correlated
with myointimal thickening. The ''vein graft" configurations in this study, however, varied considerably with respect to intraluminal pressure and wall
tension. Therefore it is postulated that any modulatory effect of flow was overshadowed by the more
important factor of wall tension.
In summary, rabbit external jugular veins subjected to various hemodynamic environments remodel the vessel wall according to changes in wall
tension and mechanical deformation. The contribution of hemodynamic forces to the adaptive changes
in vein grafts will lead to a better understanding of
vein graft physiology and possibly to more welldesigned interventional strategies.
The.authors thank Lizzie Barber and Patsy Tidwell for
their technical assistance as well as Edna Ferrell for
manuscript preparation. In addition, we thank Ethicon,
Inc. for generously providing suture material and Edward
Week, Inc. (Research Triangle Park, N.C.) for financial
support.
REFERENCES
1'. Carrell A, Guthrie CC. Results of the biterminal transplantation of veins. Am J Med Sci 1906;132:415-22.
2. Saenz NC, Folkman J. T-cell mediated inhibition of smooth
muscle hyperplasia in venous interposition grafts. Surg Forum
1990;41:566-9.
3. Murday AJ, Gershlick AH, Syndercombe-Court YD, et al.
Intimal hyperplasia in arterial autogenous vein grafts: a new
animal model. Cardiovasc Res 1983;17:446-51.
4. Zwolak RM, Adams MC, Clowes AW. Kinetics of vein graft
hyperplasia: association with tangential stress. J VAse SURG
1987;5:126-36.
5. Kohler TR, Kirkman TR, Clowes AW. The effect of rigid
external support on vein graft adaptation to the arterial
circulation. J VAse SURG 1989;9:277-85.
6. El-Sanadiki MN, Cross KS, Murray 11, et al. Reduction of
intimal hyperplasia and enhanced reactivity of experimental
vein bypass grafts with verapamil treatment. Ann Surg
1990;212:87-96.
7. Kohler TR, Lirkman TR, Gordon D, Clowes AW. Mechanism of long-term degeneration of arterialized vein graft.
Am J Surg 1990;160:257-61.
8. Wyatt AP, Taylor GW. Vein grafts: changes in the endothe·
lium of autogenous free vein grafts used as arterial replacements. Br J Surg 1966;53:943-7.
9. McCabe M, Cunningham J, Wyatt AP, Rothnie NG, Taylor
GW. A histological and histochemical examination of autogenous vein grafts. Br J Surg 1967;54:147-55.
10. Imparato AM, Bracco A, Kim GE, Zeff R. Intimal and
Volume 15
Number 1
January 1992
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
neointimal fibrous proliferation causing failure of arterial
reconstructions. Surgery 1972;72: 1007-17.
Brody WR, Angell WW, Kosk JC. Histologic fate of the
venous coronary artery bypass in dogs. Am J Pathol 1972;
66: 111-30.
Brady WR, Kosek JC, Angell BE. Changes in vein grafts
following aorto-coronary bypass induced by pressure and
ischemia. J Thorac Cardiovasc Surg 1972;64:847-54.
Karayannacos PE, Geer J, GastM, Hodges R, Bond G, Vasko
JS. Wall strain in arterial vein grafts [Abstract]. Clin Res
1973;21 :813.
Faulkner SL, Fisher RD, ConIde DM, Page DL, Bender HW.
Effect of blood flow rate on subendothelial proliferation in
venous autografts used as arterial substitutes. Circulation
1975;51-52: 1163-72.
Storm FK, Gierson ED, Sparks FC, Barker WF. Autogenous
vein bypass grafts: biological effects of mechanical dilatation
and adventitial stripping in dogs. Surgery 1975;77:261-7.
Bond MG, Hostetler JR, Karayannacos PE, Geer JC, Vasko
JS. Intimal changes in arteriovenous bypass grafts. Effects of
varying the angle ofimplantation at the proximal anastomosis
and of producing stenosis in the distal runoff artery. J Thorac
Cardiovasc Surg 1976;71:907-16.
Rittgers SE, Karayannacos PE, Guy JF, et al. Velocity
distribution and intimal proliferation in autologous vein grafts
in dogs. Circ Res 1978;42:792-801.
Karayannacos PE, Hostetler JR, Bond MG, et al. Late failure
in vein grafts: mediating factors in subendothelial fibromuscular hyperplasia. Ann Surg 1978;187:183-8.
Karayannacos PE, Ringers SE, Kakos GS, Williams TB,
Meckstroth CV, Vasko JS. Potential role of velocity and wall
tension in vein graft failure. J Cardiovasc Surg 1980;21:
171-8.
Logerfo FW, Quist WC, Cantelmo NL, Haudenschild CC.
Integrity of vein grafts as a function of intimal and medial
preservation. Circulation 1983;68(suppl IT):11117-IT122.
Morinaga K, Okadome K, Kurkoi M, Miyazaki T, Muto Y,
Inokuchi K. Effect of wall shear stress on intimal thickening
of arterially transplanted autogenous veins in dogs. J VASC
SURG 1985;2:430-3.
Berguer R, Riggins RF, Reddy DJ. Intimal hyperplasia: an
experimental study. Arch Surg 1980;115:332-5.
Landymore RW, Kinley CE, Cameron CA. Intimal hyperplasia in autogenous vein grafts used for arterial bypass: a
canine model. Cardiovasc Res 1985;19:589-92.
Morinaga K, Eguchi H, Miyazaki T, Okadome K, Sugimachi
K. Development and regression of intimal thickening of
arterially transplanted autologous vein grafts in ~ogs. J VAse
SURG 1987;5:19-30.
Dobrin PB, Littooy FN, Golan J, Blakeman B, Fareed J.
Mechanical and histologic changes in canine vein grafts. J Surg
Res 1988;44:259-65.
Dobrin PB, Litooy FN, Endean ED. Mechanical factors
predisposing to intimal hyperplasia and medial thickening in
autogenous vein grafts. Surgery 1989;105:393-400.
Shin CS, Hatem IN, Abaci IF. Effects of diminished distal
blood flow on the morphologic changes in autogenous vein
grafts. Surg Gynecol 0 bstet 1978;147: 189-92.
Barra JA, Volant A, Leroy JP, et al. Constrictive perivenous
mesh prosthesis for preservation of vein integrity. J Thorac
Cardiovasc Surg 1986;92:330-6.
McCann RL, Larson RM, Mitchener JS, Fuchs JCA, Hagen
P-? Intimal thickening and hyperlipidemia in experimental
pnmate vascular autografts, Ann Surg 1979;189:62-7.
Myointimal thickening in vein grafts
185
30. McCann RL, Hagen P-O, Fuchs TCA. Aspirin and dipyridamole decrease intimal hyperplasia in experimental vein
grafts. Ann Surg 1980;191:238-43.
31. Boerboom LE, Bonchek LI, Kissibah AH, et al. Effect of
surgical trauma on tissue lipids in primate vein grafts: relation
to plasma lipids. Circulation 1980;62(suppl 1):1142-7.
32. Vlodaver Z, Edwards JE. Pathologic changes in aortocoronary arterial saphenous vein grafts. Circulation 1971;
719-28.
33. Kern WH, Dermer GB, Lindesmith GG. The intimal
proliferation in aortic-coronary saphenous vein grafts: light
and electron microscope studies. Am Heart J 1972;84:771-7.
34. Hamaker WR, Doyle WF, O'Connell TJ, Gomez AC.
Subintimal obliterative proliferation in saphenous vein grafts.
Ann Thorac Surg 1972;13:488-93.
35. Jones M, Conkle DM, Ferrans, et al. Lesions observed in
arterial autogenous vein grafts: light and electron microscope
evaluation. Circulation 1973;47-48(suppl III):IIII98-210.
36. Szilagyi DE, Elliott JP, Hageman]H, Smith RF, Dall'Olmo
CA. Biologic fate of autogenous vein implants as arteri~
substitutes: clinical, angiographic, and histopathologic observations in femoro-popliteal operations for atherosclerosis.
Ann Surg 1973;178:232-46.
37. Kennedy JH, Wieting DW, Hwang NHC, et al. Hydraulic
and morphologic study of fibrous intimal hyperplasia in
autogenous saphenous vein bypass grafts. J Thorac Cardiovasc
Surg 1974;67:805-13.
38. Dnni KK, Klottke BA, Titus lL, Frye RL, Wallace RB, Brown
AL. Pathologic changes in aortocdronary saphenous vein
grafts. Am J Cardiol 1974;34:526-32.
39. Ramos JR, Berger K, Mansfield PH, Sauvage LR. Histologic
fate and endothelial changes of distended and non-distended
vein grafts. Ann Surg 1976;183:205-28.
40. Lie IT, Lawrie GM, Morris GC. Aortocoronary bypass
saphenous vein graft atherosclerosis: anatomic study of 99
vein grafts from normal and hyperlipoproteinemic patients
up to 75 months postoperatively. Am J Cardiol 1977;40:
906-14.
41. Fuchs JCA, Mitchener JS, Hagen P-O. Postoperative changes
in autologous vein grafts. Ann Surg 1978;188:1-15.
42. Fonkalsrud EW, Sanchez M, Zerubavel R. Morphological
evaluation of canine autogenous vein grafts in the arterial
circulation. Surgery 1978;84:253-64.
43. Madras PN, Ward CA, Johnson WR, Singh PI. Anastomotic
hyperplasia. Surgery 1981;90:922-3.
44. Schwartz LB, Purut CM, O'Donohoe MK, Smith PK, Hagen,
P-O, McCann RL. Quantitation of vascular outflow by
measurement of impedance. J VASC SURG 1991;14:353-63.
45. Goodfellow BS, Mikat EM. Simultaneous demonstration of
collagen, muscle and elastic elements in manunalian tissue.
Lab Med 1988;19:243-4.
46. Zarins CK, Zatina MA, Glagov S. Correlation ofpostmortem
angiography with pathologic anatomy: quantitation of atherosclerotic lesions. In: Bond MG, Insu1l W, Glagov S,
Chandler AB, Cornmill F, eds. Clinical diagnosis of
atherosclerosis: quantitative methods of evaluation. New
York: Springer-Verlag, 1983 :283-306.
47. Leung DYM, Glagov S, Mathews MB. Cyclic stretching
stimulates synthesis of matrix components by arterial smooth
muscle cells in vitro. Science 1976;191:475-7.
48. Daniel RE, Boimott JK, Brown GD, Heptinstall RH.
Endothelial cell activity in experimental hypertension: effects
ofhemodynamic factors. Lab Invest 1983;48:690-7.
49. Wolinsky H, Glagov S. A lamellar unit of aortic medial
Journal of
VASCULAR
SURGERY
186 Schwartz et al.
50.
51.
52.
53.
54.
55.
56.
57.
structure and function in mammals. Circ Res 1967;20:99~
Ill.
Wolinsky H. Response of the rat aortic media to hyperten~
sion. Circ Res 1970;26:507-22.
Leung DYM, Glagov S, Mathews MB. Elastin and collagen
accumulation in rabbit ascending aorta and puhnonary trunk
growth during postnatal growth: correlation of cellular
synthetic response with medial tension. Circ Res 1977;41:
316-23.
Clark ]M, Glagov S. Transmural organization of the arterial
media: the lamellar unit revisited. Arteriosclerosis 1985;5: 1934.
Glagov S, Weisenberg E, Zarins CK. Compensatory enlarge~
ment of human atherosclerotic coronary arteries. N Engl J
Med 1987;316:1371~5.
Zarins CK, Weisenberg E, Kolettis G, et al. Differential
enlargement of artery segments in response to enlarging
atherosclerotic plaques. ] VASe SURG 1988;7:386~94.
Ibrahim IM, Sussman B, Dardik I, et al. Adjunctive arterio~
venous fistula with tibial and peroneal reconstruction for limb
salvage. Am J Surg 1980;140:246-51.
Dardik H, Sussman B, Ibrahim IM, et al. Distal arteriovenous
fistula as an adjunct to maintaining arterial and graft patency
for limb salvage. Surgery 1983;94:478-86.
Harris PL, Campbell H. Adjuvant distal arteriovenous shunt
with femorotibial bypass for critical ischaemia. Br ] Surg
to flow change in the canine carotid artery. Am ] Physiol
1980;239:HI4~21.
59. Zarins CK, Bomberger RA, Glagov S. Local effect of
stenoses: increased flow velocity inhibits atherogenesis. Cir~
culation 1981;64(suppl 11):11221-7.
60. Zarins CK, Giddens DP, Bharadvaj BK, Sottiurai VS, Mabon
RF, Glagov S. Carotid bifurcation atherosclerosis: Quantitative correlation of plaque localization -with flow ve.
locity profiles and wall shear stress. Circ Res 1983;53:
502-14.
61. Pomposelli F, Schoen F, Cohen R, O'Leary D, Johnson WR,
Madras PN. J VAse SURG 1984;1:525-535.
62. Langille BL, O'Donnell F. Reductions in arterial diameter
produced by chronic decreases in blood flow are endothelium~
dependent. Science 1986;231:405~7.
63. Levesque MJ, Liepsc D, Moravec S, Nerem RM. Correlation
ofendothelial cell shape and wall shear stress in a stenosed dog
aorta. Arteriosclerosis 1986;6:220-9.
64. Zarins CK, Zatina MA, Giddens DP-, Ku DN, Glagov S. Shear
stress regulation of artery lumen diameter in experimental
atherogenesis. J VAse SURG 1987;5:413-20.
65. Glagov S, Grade JP, Xu C, Giddens DP, Zarins CK. Limited
effects of hyperlipidemia on the arterial smooth muscle
response to mechanical stress. ] Gardiovasc Pharmacol 1989;
14(supp16):S90-7.
1983;70:377-80~
58. Kamiya A, Togawa T. Adaptive-regulation ofwall shear stress
Submitted June 10, 1991; accepted Sept. 18, 1991.