| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Regular Articles |
From the Department of Pathology,*
University of
Massachusetts Medical School, Worcester, Massachusetts; and the
Department of Mechanical Engineering,
Worcester Polytechnic Institute, Worcester, Massachusetts
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Twelve male Wistar rats (Charles River Breeding Laboratories, Wilmington, MA) 4–6 weeks of age and weighing 225–250 g were maintained in accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals (DHHS, NIH Publication No. 85–23, 1985) and guidelines of the Animal Care Advisory Committee of the University of Massachusetts Medical School.
Stenosis
Mushroom-shaped Plexiglas plugs were prepared by the
University of Massachusetts Medical School machine shop using a 10-inch
Hardinge lathe. The nearly hemispherical cap measured approximately
0.9 x 1.5 mm and the stem 3 x 1 mm (Figure 1)
. These dimensions were chosen within
the limits imposed by the size of the aortic system in the rat for the
purpose of creating an approximately 50% occlusion of the aortic
lumen.
|
Under Nembutal anesthesia administered intraperitoneally,
the rat was laid supine and a 5-cm incision was made along the linea
alba. The left renal artery and vein were cleared of overlying
peritoneum and fat, and the aorta was isolated 5 mm proximal and distal
to the left renal vessels. The left renal artery was ligated with 4–0
silk 4–5 mm from its aortic origin. The left renal vein was
double-ligated 2 mm from its entrance into the inferior vena cava and
sectioned between ligatures. The abdominal aorta was then cross-clamped
with atraumatic aneurysm clips 1 mm below the right renal artery and 5
mm below the left renal artery. When needed, an additional clamp was
placed across adjacent spermatic branches. The left renal artery was
sectioned 2–3 mm from the aorta, dilated and held open with no. 5
jeweler's forceps. The plug was inserted through the renal artery
stump into the lumen of the aorta using a microneedle holder. This
required gentle force because the head of the plug was wider than the
natural lumen of the renal artery. Two 8–0 nylon ligatures were then
placed around the renal artery, which cuffed the stem of the plug. The
aortic clamps were then removed, the left kidney and adrenal gland were
excised, and minor bleeding vessels were cauterized. The abdominal wall
was closed with 4–0 silk running sutures; the skin was closed with
stainless steel clips. The entire surgical procedure (Figure 2)
lasted approximately 30 minutes, after
which the animal was returned to its cage and fed a normal diet.
|
Twelve rats were used. Stenosis was produced in six; six were kept as unstenosed controls. The rats with stenosis were fed a normal diet for 1 month to allow recovery and healing from surgery; thereafter, they were placed on a hypercholesterolemic diet for 3 months. The controls were also fed a hypercholesterolemic diet for 3 months after 1 to 4 weeks on a normal diet.
The purpose of the experiment was to compare the pattern of lipid deposition in unstenosed rats with that in rats with the plug stenosis. Surgical sham controls were not performed; ligating the renal artery would have created a new geometry of the aorta not directly relevant to the experimental model.
Diets
Hypercholesterolemia was induced by feeding the rats ad libitum a diet supplemented with 4% cholesterol, 1% cholic acid, and 0.5% 2-thiouracil (CCT Diet, Bioserve, Frenchtown, NJ).28 Normal diet was standard Purina Lab Chow No. 5008 (Ralston Purina, St. Louis, MO).
Plasma Cholesterol
Before sacrifice, 1 ml of blood was obtained for measuring plasma cholesterol using Sigma kit 351-20 (Sigma Chemical, St. Louis, MO).
Perfusion Fixation and Intimal Staining
Under deep ether anesthesia, the rats were sacrificed by perfusion-fixation at 110 mm Hg.28 Perfusion staining of the intercellular junctions with silver nitrate and of lipid deposits with oil red O was performed as previously described.14
Light Microscopy
One-centimeter segments of the fixed and oil red O-stained
abdominal aortas, containing the plug site in stenosed animals (Figure 3)
and the corresponding aortic segment
in controls, were cut open longitudinally along the right lateral wall,
pinned flat on paraffin, immersed in water, and viewed en
face through a surgical microscope. The plugged aorta of one
hypercholesterolemic rat was set aside for embedding in Epon as
previously described.14
Semithin 1-µm longitudinal
sections through the stenosis site were cut and stained with toluidine
blue.
|
The aortic segment containing the plug site was opened, pinned out, and photographed. A Kodachrome slide was projected onto paper and the perimeter of the aortic specimen and outlines of all fat deposits were traced. The same procedure was carried out with the controls.
Biophysical Models: Hydrodynamic Studies
Dye streamline analysis was used to study the flow field in an enlarged 34:1 scale Plexiglas model of the in vivo hemispherical plug stenosis. In the model, the radius of the aorta and hemispherical plug were equal (25.4 mm). To ensure fully developed flow, a straight entrance region 5 m long preceded the transparent test section.
In unsteady flow, the Strouhal number29 is a dimensionless quantity that compares the relative importance of the local time-varying component of acceleration due to pulsatile flow to the convective component of acceleration due to changes in vessel geometry. A small Strouhal number indicates that the acceleration terms caused by changes in vessel geometry predominate. Based on the low Strouhal number (0.04) found in rats,13 the flow in the region of the stenosis may be considered quasi-steady.
Hence, in a first study, a steady flow with mean (systolic) Reynolds
number of 250 was analyzed. Water was maintained at a constant flow
rate in the 5.1-cm diameter Plexiglas tube. Dye was injected at 40
different locations and the pathlines were recorded by videotaping at a
frame rate of 60 frames/second. Mirrors allowed the simultaneous
viewing of the velocity components in both the frontal and sagittal
planes. A total of 1100 velocities were measured and en face
wall shear stresses computed from the resulting velocity profiles. The
wall shear stresses measured along the center line of the plug were
normalized using the wall shear stress 
o found
in an unobstructed vessel with the same flow rate. This quantity,
/o, was then plotted as a function of axial
position.
In a second study, a sinusoidal flow pulse equal to 20% of the total flow was created using a rotating ball valve and superimposed on the steady flow component. The fluid was a 65% glycerol/35% water mixture. The mean Reynolds number was 250 and the Strouhal number was 0.04 (Womersley number of 1.6).13,29 As with the steady flow experiments, dye visualization studies were used to determine the time-dependent extent of the low-velocity/recirculating regions near the plug.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Hypercholesterolemic Controls without Plug Stenosis
Plasma cholesterol at time of sacrifice was greater than 500
mg/dl. The aortic intima viewed en face showed extensive,
spotty, disseminated lipid deposits; their distribution appeared to be
random except for some selectivity for arterial ostia, especially that
of the left renal artery (Figure 4A)
.
|
Plasma cholesterol at time of sacrifice was greater than
500 mg/dl. A narrow band of oil red O-positive lipid was seen all
around the base of the plug; heavier, crescent-shaped deposits were
present proximally and distally. On average, the oil red O-positive
proximal region extended upstream 0.5 diameters and the distal region
extended downstream 1.5 diameters. The aortic intima opposite the plug
contained almost no lipid. The surface of the plug had little or no
lipid near the apex and small droplets were scattered over the rest of
the hemisphere. As in the control arteries from hypercholesterolemic
rats, lipid deposits were also seen at the ostia of arteries branching
from the aorta (Figure 4B)
. In most animals, the plug surface was seen
grossly to be completely covered by a new intima (in two animals the
surface was only partially covered); the endothelial cell junctions
were clearly outlined by silver and the cells contained sparse lipid
(Figure 5)
. Epon sections showed that the
new intimal layer covering the plug was continuous with the aortic
intimal lining (Figure 6)
and contained
smooth muscle cells (Figure 7, A and B)
.
Intracellular lipid droplets were present in the intimal thickening at
the base of the plug and at the ostium between the aorta and renal
artery.
|
|
|
Steady Flow
The hydrodynamic steady flow studies showed that very low
shear stresses occurred upstream and downstream of the plug along the
center line of the model aorta. High shear stresses up to 13 times
baseline occurred on the surface 45° anterior and 30° posterior
from the top of the plug. Very low fluid shear stresses were measured
in a region surrounding the base of the plug, extending approximately
0.5 diameters upstream and downstream of the stenosis (Figure 8A)
. Along the wall opposite the plug,
shear stresses reached 2–3 times baseline value (Figure 8B)
.
|
. One stagnation point, approximately
0.5 diameters upstream, was associated with a bound vortex. A second
region of stagnation and recirculation extended for about 1 diameter
downstream. The flow separated just beyond the top of the bulge. The
inner stagnation/separation regions formed triangular shapes (Figure 9)
.
|
In the unsteady flow studies, the pattern was similar to that of steady flow. The triangular region downstream of the stenosis was marked by low velocities and recirculation; a smaller, similar pattern was found upstream. The upstream region varied in size from 0.25 diameters at minimum flow to 0.5 diameters at peak flow, whereas the downstream region extended from 0.75 to 1.25 diameters.
Due to the technique of plug insertion, the exact position and orientation of the plug within the aorta in vivo cannot be precisely determined. To determine the sensitivity of the flow field to the position of the plug, unsteady model studies were conducted with the plug tilted upstream at approximately 45° to the vessel axis. This increased the upstream stagnation region, which varied from 0.5 to 0.75 diameters, and the downstream region, which varied from 1 to 2 diameters. Within this downstream region a horseshoe vortex scoured the aortic wall with an aperiodic fluctuating velocity.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
. Tissue Response to the Plug
Locally, the aorta seemed to tolerate the plug well. The
luminal surface was partly or completely overgrown by endothelium
accompanied by a subendothelial layer of smooth muscle cells. This
layer was thicker along the proximal and especially along the distal
edge of the plug, and therefore acquired a somewhat elongated shape
(Figure 4B)
. This subendothelial thickening occurred in areas that were
also subject to lipid deposition; it could therefore be a response to
the lipid deposit, special local flow conditions (see discussion
below), or both. Whatever its mechanism, this phenomenon suggests that
raised atherosclerotic plaques should also tend to grow preferentially
in the direction of flow.
Correlation of Biological and Biophysical Findings
In control rats (hypercholesterolemia without plug stenosis), the lipid deposits in the abdominal aorta were diffusely distributed with some accentuation around the ostia, as expected.31 The effect of the plug was to alter the lipid distribution in a consistent manner. There were no lipid deposits in the intimal area opposite to the plug, whereas the base of the plug was surrounded by a heavy deposit of oil red O-positive lipid, advancing upstream 0.5 diameters on average and trailing downstream 1.5 diameters. The physical models of the plug stenosis indicate that the lipid-free area opposite the plug corresponds to increased fluid shear stress, whereas the regions proximal and distal to the base of the plug are areas of low shear stress, flow separation, and recirculation.
Fukushima and Azuma32 conducted a detailed study of the steady flow field resulting from a hemispherical protuberance in a cylindrical vessel. They described a portion of the resulting flow field as a horseshoe vortex. As expected, the flow field in our models corresponds to their description. In unsteady flow, a horseshoe vortex was still observed but the extent of the separated regions varied over time. In models, the regions proximal and distal to the plug are characterized by flow separation, flow reversal, and low shear. These regions are strongly correlated with the crescent-shaped areas of lipid deposition in vivo. Similar findings have been inferred by others31,33,34 studying the focal nature of atherosclerotic lesions. The plug stenosis further confirms these findings in a controlled in vivo experiment.
The sensitivity of the flow field to positioning of the plug within the rat aorta was investigated in model studies under conditions of unsteady flow. The general characteristics of the flow were unaltered even when the plug axis was tilted upstream at an angle of 45° rather than orthogonal to the flow. The size of the upstream and downstream separation regions increased as the plug axis was tilted; however, the flow field still contained a horseshoe vortex. These studies indicate that the in vivo flow field caused by the plug stenosis is relatively insensitive to experimental variations in plug deployment and renal branch angle.
In our prior studies of experimental aortic stenosis an external clip was used. The clip has a shielding effect that limits wall motion14 in the throat of the stenosis. In such a system, fluid shear stress is very high beneath the clip, whereas the intramural stress and strain components are low. Under these conditions, the throat was spared from lipid deposition. These findings suggested that high fluid shear stress inhibited lipid deposition; however, as noted previously, the possible effect of low intramural stress could not be completely discounted. The hemispherical plug stenosis demonstrates that in the presence of normal intramural strains, increased shear correlates with decreased lipid deposition. Thus, the present study adds strong evidence to support the contention that increased shear inhibits lipid deposition. Our studies do not address the mechanism whereby low shear stress favors lipid deposition; extensive work is being performed in this area, focusing especially on endothelial mechanotransduction35,36 and flow-mediated regulation of endothelial gene expression.37-39
Finally, the ring of intima in contact with the base of the plug is also an area of lipid deposition. However, the intima in very close proximity to the base of the plug is likely to experience altered stresses and strains. Hence, this ring of lipid deposition cannot be directly and exclusively attributed to disturbed flow.
In summary, the hemispherical plug stenosis minimizes the potentially confounding effects of intramural parameters on the data related to shear stresses. In vivo studies using the plug stenosis can focus more sharply on the relationship between lipid deposition and hydrodynamic disturbances. Our results show that elevated shear stress reduces lipid deposition, whereas low shear and recirculation promote lipid deposition.
|
Acknowledgements |
|---|
|
Footnotes |
|---|
Supported in part by grants HL25973 and HL33529 from the National Heart, Lung and Blood Institute. Contents are solely the responsibility of the authors and do not necessarily represent the official views of the National Institutes of Health.
Accepted for publication April 15, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  S. Noria, F. Xu, S. McCue, M. Jones, A. I. Gotlieb, and B. L. Langille Assembly and Reorientation of Stress Fibers Drives Morphological Changes to Endothelial Cells Exposed to Shear Stress Am. J. Pathol., April 1, 2004; 164(4): 1211 - 1223. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. M Wasserman and J. N Topper Adaptation of the endothelium to fluid flow: in vitro analyses of gene expression and in vivo implications Vascular Medicine, February 1, 2004; 9(1): 35 - 45. [Abstract] [PDF]
|
|
|
|
 |
|
 B. A. Wasserman, W. I. Smith, H. H. Trout III, R. O. Cannon III, R. S. Balaban, and A. E. Arai Carotid Artery Atherosclerosis: In Vivo Morphologic Characterization with Gadolinium-enhanced Double-oblique MR Imaging—Initial Results Radiology, May 1, 2002; 223(2): 566 - 573. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
