The Biologic Behavior of Balloon Hyperinflation–Induced Arterial Lesions in Hypercholesterolemic Pigs Depends on the Presence of Foam Cells
Abstract Lack of a large-animal model of accelerated atherosclerosis has limited study of the biologic behavior of atherosclerotic lesions. We hypothesized that mechanical vascular trauma combined with diet-induced hypercholesterolemia would result in rapid development of complex atherosclerosis-like lesions. Accordingly, we induced deep injury to a carotid artery by repetitive balloon hyperinflations in minipigs that were fed either an atherogenic diet (n=30) or a standard diet (controls, n=4) and examined the resultant lesions 1 month later. The neointimal lesions that evolved in 23 patent vessels from cholesterol-fed animals were complex, exhibiting infiltration of smooth muscle and foam cells and evidence of organized thrombus, recent thrombus, hemorrhage, and calcification. Lesions were separable histologically into two groups: foam-cell rich (n=12), with 33±10 foam cells per high-power field, and foam-cell poor (n=11), with 4±1 foam cells per high-power field. Minipigs with foam cell–rich lesions had higher serum cholesterol levels than those with foam cell–poor lesions (712±178 vs 468±240 mg/dL, P<.02). The incidence of intralesional thrombus was also significantly greater in foam cell–rich than in foam cell–poor lesions (50% vs 9%, P<.04). In addition, the degree of luminal stenosis was greater in the presence of lesions containing thrombus compared with those without thrombus (60±38% vs 30±29%, P=.05). Lesions in the control animals were fibrocellular and lacked foam cells and thrombus. Thus, hypercholesterolemia appeared to affect lesion composition and behavior. Lesions with an abundance of foam cells were more likely to show evidence of intralesional thrombosis, which was associated with increased luminal stenosis. Our findings suggest that foam cells may predispose to lesion instability and thrombosis, leading to even more severe luminal obstruction.
Reprint requests to Samuel A. Wickline, MD, Jewish Hospital at the Washington University School of Medicine, Division of Cardiology, 216 S Kings Highway, St Louis, MO 63178.
- Received August 2, 1994.
- Accepted April 21, 1995.
Results from clinical studies have suggested that the severity of luminal stenosis caused by an atherosclerotic plaque may not be a reliable predictor of future ischemic events.1 2 Patients with mild to moderately severe obstructive lesions in the coronary or cerebrovascular circulation frequently develop unstable ischemic syndromes in an accelerated fashion, apparently as a consequence of plaque rupture with subsequent thrombus formation.3 4 5 6 7 The mechanisms responsible for plaque weakening and rupture have not been clearly identified, although lesions that contain numerous macrophages and soft, lipid-enriched cores appear to be more unstable and prone to rupture with rapid disease progression.8 Several mechanisms have been suggested for this behavior, including hemodynamic stresses,9 fluctuations in pressure,10 coronary vasospasm,4 11 12 variations in the tensile strength of the lesion cap,13 14 and an acute inflammatory process.15 An animal model exhibiting similar lesion morphology and behavior would be useful for studying the mechanisms of plaque rupture and potential interventions that could facilitate plaque stability.
Lack of a large-animal model of accelerated atherosclerosis has been a limitation in atherosclerosis research.16 Previous studies have employed balloon hyperinflation–induced arterial injury in pigs with normal serum lipid levels, resulting in proliferative intimal lesions that lack lipid-laden macrophages (foam cells) but contain abundant smooth muscle cells and extracellular matrix,16 17 which are more characteristic of restenosis following angioplasty than of atherosclerosis. We hypothesized that rapid development of lesions more like the complex atherosclerotic plaques that are observed in humans could be accomplished in minipigs with use of a combination of balloon-induced vascular trauma and concomitant hypercholesterolemia. This study was designed to characterize the morphology of lesions induced with this combined approach and to examine the relationship between lesion morphology and biologic behavior.
Thirty-four male Yucatan minipigs (Charles River Laboratories) weighing 20.4±3.3 kg were used in the study. A baseline blood sample was obtained for measurement of serum cholesterol. Thirty animals were fed a previously described18 atherogenic diet containing 4% cholesterol and 30% fat (Purina Test Diet No. 5748 M-A). The remaining four animals were fed a standard diet without added cholesterol and served as the control group. After 14 days, the serum cholesterol level was measured and the left carotid artery injured as described below. Forty-eight hours after injury, patency of the injured vessel was assessed by duplex ultrasound (Hewlett-Packard Sonos 1500 scanner with a 7.5-MHz linear-array transducer). If the vessel was patent, no further intervention was performed. If the vessel was occluded, the injury procedure was repeated within 1 week on the right carotid artery. Animals were continued on the high-cholesterol or standard diet after the injury procedure. One month after carotid artery injury, vessel patency was reassessed by ultrasound, carotid artery angiography was performed, and the arteries were excised for histological analysis.
The pigs were sedated with an intramuscular injection of ketamine (22 mg/kg body wt), acepromazine (1.1 mg/kg body wt), and atropine (0.05 mg/kg body wt). An ear vein was cannulated and maintenance fluids were administered. The trachea was intubated and surgical anesthesia was induced and maintained with repeated doses of intravenous pentobarbital (20 mg/kg body wt).
An 8F catheter sheath was inserted into the exposed right femoral artery, and a bolus of intravenous heparin (200 U/kg body wt) was administered. The animals were kept anticoagulated for 3 hours after the procedure to maintain the activated clotting time at >400 seconds. No antiplatelet agents were administered.
Baseline carotid artery angiography was performed and a balloon catheter (Proflex 5, 8 mm×2 cm, Mallinckrodt) was then advanced into the left carotid artery to the level of the second or third cervical vertebra. The balloon was inflated five times to a distending pressure of 8 atm for 30 seconds with 30 seconds between inflations. Measurements of the vessel and inflated balloon on angiograms revealed luminal diameters (mean±SD) of 4.36±0.18 and 4.94±0.20 mm, respectively. This value represents an increase in vessel diameter of 13.4±3.3% during balloon inflation, which others have reported is associated with consistent disruption of the internal elastic lamina.19
Four weeks after carotid artery injury and after repeated angiography, the tip of the angiographic catheter was withdrawn to the level of the brachiocephalic artery. Through a median sternotomy, a ligature was placed around the proximal brachiocephalic artery containing the catheter to obstruct blood flow, and the carotid arteries were perfused antegradely with 500 mL of 0.9% NaCl followed by 500 mL of 4% paraformaldehyde at a pressure of 120 mm Hg. Both the injured and uninjured (when available) carotid arteries were excised via a midline neck incision and cut into 1-cm segments for histological analysis. The segment from the injured artery with the smallest luminal diameter was embedded in paraffin and cut through its entirety at a thickness of 5 μm, and sets of sections collected every 100 μm were stained with hematoxylin and eosin, Masson’s trichrome for collagen, Verhoeff–van Gieson’s stain for elastic tissue, and Martius scarlet blue (MSB) to distinguish recently formed fibrin, organized thrombus, and red blood cells.20
The extent of luminal stenosis in sections from the vessel segment with the smallest luminal diameter was measured by digitizing the images of the sections stained for elastic tissue (to facilitate identification of the internal elastic lamina) with the use of a Nikon Optiphot-2 microscope with a CCD camera attached to a Macintosh IIci computer outfitted with a NuVista frame-grabber board. A reference dimensional scale was also digitized to allow calibration of the image-analysis software (Image 1.44, National Institutes of Health). The cross-sectional areas of the lumen and intima were measured by planimetry by tracing the margin of the lumen and internal elastic lamina, respectively. The percent luminal obstruction was then calculated from these measurements. Five different sections from each artery were analyzed and the results averaged.
Foam cell content in neointimal lesions was quantified by counting the number of foam cells per high-power field (×400) in 25 randomly selected fields. The presence of thrombus was defined by the identification of fibrin within the lesion and was graded as either present or absent.
Statistical analysis was performed with commercially available computer software (Statview 4.0, Abacus Concepts, Inc). Data are expressed as mean±SD unless otherwise indicated. Continuous variables were analyzed with either Student’s t test for unpaired data (two groups) or ANOVA (more than two groups). Nominal data were analyzed by χ2 testing. Significance was assigned at P<.05.
All 34 animals (30 on the high-cholesterol diet) underwent carotid artery injury without complications. None exhibited acute occlusion of the injured artery. Four animals died within 4 days: two of shock due to hypothermia, one of infection at the femoral cutdown site, and another of a tracheal abrasion. Two additional animals did not complete the protocol because of technical difficulty in achieving adequate balloon hyperinflations. A total of 28 animals completed the protocol. Four animals exhibited occlusion of the injured vessel, as identified by ultrasound after 48 hours, and underwent a second procedure on the contralateral vessel. Two additional animals exhibited vessel occlusion after 4 weeks, resulting in a total of six of 32 (19%) injured vessels that became occluded. Twenty-six vessels (81%) remained patent and were used for analysis. Of these, 23 were from animals fed cholesterol and three were from animals fed a standard diet.
Serum Cholesterol Levels
Serum cholesterol levels increased markedly after 2 weeks in the minipigs that were fed the atherogenic diet: from 101±22 mg/dL at baseline to 587±235 mg/dL (P<.0001). There was no significant difference in cholesterol level 4 weeks later (540±260 mg/dL, P=.73). Cholesterol levels also did not change throughout the study period in animals that were fed the standard diet (92±37 mg/dL at baseline and 79±26 mg/dL at injury, P=.95).
Severe mural damage was observed in all balloon-injured vessels, as manifested by disruption and fragmentation of the internal elastic lamina and replacement of medial smooth muscle cells with collagen, a finding consistent with necrosis (Fig 1⇓). Neointimal lesions in cholesterol-fed animals were heterogeneous and composed of foam cells, smooth muscle cells, and areas of thrombus. Some lesions appeared to be organized in tiers, with a fibrous cap, foam cells at the base, and smooth muscle cells and matrix between the cap and the matrix. Lesions appeared as two distinct types on the basis of their overall content of foam cells, ie, foam-cell rich and relatively foam-cell poor (Fig 2⇓). Foam cell–rich lesions (>10 foam cells per high-power field) exhibited foam cells predominantly at their base near the internal elastic lamina and at lesion shoulders (Fig 1⇓). Foam cell–poor lesions (<10 foam cells per high-power field) generally contained abundant smooth muscle cells (Fig 1⇓). The characteristics of foam cell–rich and foam cell–poor lesions are shown in the Table⇓. Neointimal lesions in control animals fed the standard diet were essentially devoid of foam cells and appeared similar to those described previously induced by the same injury technique but without cholesterol feeding.16 19 No intimal thickening or foam cells were observed in the contralateral uninjured carotid arteries from animals fed cholesterol for 6 weeks.
Thrombus was a common feature of foam cell–rich lesions (the Table⇑ and Fig 3⇓). MSB staining revealed the presence of both recent and prior fibrin together with extravasated red blood cells, suggesting that repetitive episodes of thrombosis had occurred. Neovascularization, hemosiderin-laden macrophages, and calcification were observed in many lesions, indicating organization of the thrombus (Figs 3⇓ and 4⇓). Evidence of discrete, adjacent areas at different stages of organization was present in some lesions, with the appearance of focal mononuclear cell infiltration around the fibrous cap, a finding consistent with the possibility of lesion growth and remodeling (Fig 4⇓). Thrombus was not observed in lesions from animals fed the standard diet.
Relationships Between Lesion Composition and Biologic Behavior
An association between serum cholesterol levels and the number of foam cells in lesions was apparent (Fig 5⇓). Furthermore, foam cells appeared to be associated with intralesional thrombus. Six of 12 (50%) foam cell–rich lesions contained thrombus, whereas only one of 11 (9%) foam cell–poor lesions showed evidence of thrombus (the Table⇑).
The degree of luminal narrowing was less severe in foam cell–poor than in foam cell–rich lesions, but the difference was only marginally significant (the Table⇑). However, thrombus-containing lesions manifested significantly greater luminal stenosis than did those without thrombus (Fig 6⇓). Thus, although the extent of luminal stenosis at 4 weeks was not correlated with the level of serum cholesterol (r=.01), high serum cholesterol levels did predispose to the formation of foam cell–rich lesions, which were significantly associated with thrombosis and increased lesion severity.
One aim of our study was to rapidly induce complex, atherosclerosis-like lesions in the carotid arteries of normal minipigs by using a combination of balloon-induced injury and cholesterol feeding. Using this approach, we observed heterogeneous lesions that contained infiltrations of smooth muscle cells, foam cells, and organized thrombus after only 4 weeks (Fig 1⇑). Lesions also showed evidence of instability and remodeling, as indicated by the presence of recent fibrin deposition, hemorrhage, and mononuclear cell infiltration (Figs 3⇑ and 4⇑). In contrast, animals that were fed a standard diet and underwent only balloon-induced injury developed less severe lesions that lacked foam cell infiltration and hemorrhage, as previously reported.16 19
Swine spontaneously develop atherosclerotic lesions within 2 years, which exhibit histological features similar to those of human atheroma.21 22 23 Cholesterol feeding accelerates this process.24 25 26 27 Cholesterol feeding with superimposed balloon denudation of the endothelium accelerates atherogenesis even more18 28 29 30 but produces lesions that are composed primarily of foam cells and frequently lack evidence of plaque rupture and thrombosis, as observed clinically and in our study. Thus, the combination of balloon hyperinflation–induced arterial injury and atherogenic diet appears to not only accelerate atherosclerosis but also to induce vascular lesions with characteristics more comparable to human atheroma.
The severity of hypercholesterolemia appeared to be an important determinant of lesion composition and behavior. Animals with higher levels of serum cholesterol in general exhibited more abundant foam cells in the induced lesions (the Table⇑ and Fig 5⇑). Moreover, lesions rich in foam cells contained more thrombus, which in turn was associated with increased luminal stenosis (the Table⇑ and Fig 6⇑). Lesions that were foam-cell poor exhibited less thrombus and less marked luminal stenosis. Thus, the presence of foam cells in lesions appeared to be associated with lesion instability and thrombosis, leading to even more severe luminal obstruction.
A few animals exhibited cholesterol levels >500 mg/dL but few foam cells in their vascular lesions. Similar findings have been reported in rabbits fed high-cholesterol diets.31 The reasons for this are unclear but may be related to differences in cholesterol handling or plasma lipid profiles that could affect lipid deposition and the recruitment of foam cells to vascular lesions. Alternatively, recruitment of foam cells may also depend on the extent of initial thrombosis at the time of vessel injury, a hypothesis that has been recently proposed by Schwartz et al.32 Although thrombus may contribute to foam cell deposition, our data, which show preferential accumulation of foam cells at the base and shoulders of lesions as well as evidence of recurrent fibrin deposition, argue against a major role for the initial thrombus in foam cell enrichment of lesions.
Previous studies have implicated macrophages in the behavior of atherosclerotic plaque. Lendon et al33 observed that an increased density of macrophages within plaques was associated with weakening of the fibrous cap. Foam cells have frequently been observed at the site of plaque rupture and thrombosis.8 These cells secrete a number of potent proteolytic enzymes, including stromelysins34 and collagenase,35 that can digest extracellular matrix molecules. In addition, Rouis et al36 have shown that cholesterol is a potent stimulator of the secretion of elastolytic enzymes from macrophages. These studies suggest that the presence of macrophages within the atherosclerotic plaque contributes to lesion instability, a notion consistent with our data.
Macrophages and lipid in atherosclerotic plaque may also exacerbate thrombus formation after plaque fissuring and rupture, thereby contributing to an unstable ischemic syndrome.4 Wilcox et al37 have shown that macrophages in atherosclerotic plaques express tissue factor, which activates coagulation. Lesnik et al38 have recently shown that expression of tissue factor by macrophages in cell culture is enhanced by cholesterol loading. In addition, acetylated LDL has been shown to stimulate macrophages to activate plasminogen as well as to degrade the extracellular matrix.39 Thus, the combination of lipid and macrophages within an atherosclerotic lesion may greatly contribute to its instability and subsequent thrombogenicity.
One limitation of our study is that we did not fractionate serum cholesterol into LDL and HDL components or measure Lp(a), all of which have been shown to be important in atherosclerosis and thrombus formation40 and might have provided further insight into the cause of variations in lesion behavior. However, Moreland27 showed that the alterations in serum lipid profiles observed in swine fed an atherogenic diet were similar to those seen in humans. These authors also found no evidence of lipid deposition in the liver or other organs that might indicate a lipid-storage disease. Furthermore, atherosclerotic lesions that are induced by dietary manipulation alone as well as those that are produced in an accelerated fashion by dietary manipulation combined with vessel injury16 30 41 are morphologically similar to those occurring spontaneously in older swine fed normal diets,26 suggesting that our results are not explained by pathological lipid storage.
Another limitation of this study was exclusion of 19% of the animals with vascular injury due to thrombotic occlusion. These animals were excluded from further analysis because the time of occlusion was unknown, and it would presumably have altered the course of subsequent neointimal formation. The incidence of thrombotic occlusion that we observed is similar to that reported previously with the same balloon hyperinflation technique.19
In conclusion, our findings suggest that hypercholesterolemia in the setting of arterial injury is an important determinant of lesion composition and stability. Lesions in animals with higher levels of serum cholesterol were more likely to be rich in foam cells, which in turn were more frequently associated with lesion instability, as manifested by intralesional thrombus. In turn, thrombus was associated with greater luminal obstruction. Accordingly, control of serum cholesterol may reduce acute vascular events in growing lesions by modifying lesion composition and the propensity for recurrent thrombosis.
This study was supported in part by grant HL-42950 from the National Institutes of Health, Bethesda, Md (Dr Wickline); an Established Investigator Award from the American Heart Association (Dr Wickline); and an Affiliate Fellowship Award from the Missouri Affiliate of the American Heart Association (Dr Recchia). We thank John Engelbach, Delbert McGraw, and Daud Ashai, MD, for technical assistance and Ava Ysaguirre and Barbara Donnelly for assistance with the manuscript.
- ↵Little WC, Constantinescu M, Applegate RJ, Kutcher MA, Burrows MT, Kahl FR, Santamore WP. Can coronary angiography predict the site of a subsequent myocardial infarction in patients with mild-to-moderate coronary artery disease? Circulation. 1988;78:1157-1166.
- ↵Davies MJ, Thomas AC. Plaque fissuring—the cause of acute myocardial infarction, sudden ischemic death, and crescendo angina. Br Heart J. 1985;53:363-373.
- ↵Loree HM, Kamm RD, Atkinson CM, Lee RT. Turbulent pressure fluctuations on the surface of model vascular stenosis. Am J Physiol. 1991;261:H644-H650.
- ↵Lin CS, Penha PD, Zak FG, Lin JC. Morphodynamics of acute coronary thrombosis, with special reference to volcano-like eruption of atheromatous plaque caused by coronary artery spasm. Angiology. 1988;39:535-547.
- ↵Kuga T, Tagawa H, Tomoike H, Mitsuoka W, Egashira S, Ohara Y, Takeshita A, Nakamura M. Role of coronary artery spasm in progression of organic coronary stenosis and acute myocardial infarction in a swine model: importance of mode of onset and duration of coronary artery spasm. Circulation. 1993;87:573-582.
- ↵Falk E. Why do plaques rupture? Circulation. 1992;86(suppl III):III-30-III-42.
- ↵Cheng GC, Loree HM, Kamm RD, Fishbein MC, Lee RT. Distribution of circumferential stress in ruptured and stable atherosclerotic lesions: a structural analysis with histopathological correlation. Circulation. 1993;87:1179-1187.
- ↵van der Wal A, Becker AE, van der Loos CM, Das PK. Site of intimal rupture or erosion of thrombosed coronary atherosclerotic plaques is characterized by an inflammatory process irrespective of dominant plaque morphology. Circulation. 1993;89:36-44.
- ↵White CJ, Ramee SR. Swine models of atherosclerosis. In: White RA, ed. Atherosclerosis and Arteriosclerosis. Boca Raton, Fla: CRC Press; 1989:207-234.
- ↵Weiner BH, Ockene IS, Jarmolych J. Comparison of pathologic and angiographic findings in a porcine preparation of coronary atherosclerosis. Circulation. 1985;72:1081-1086.
- ↵Steele PM, Chesebro JH, Stanson AW, Holmes DR, Dewanjee MK, Badimon L, Fuster V. Balloon angioplasty: natural history of the pathophysiological response to injury in a pig model. Circ Res. 1985;57:105-112.
- ↵Lendrum AC, Fraser DS, Slidders W, Henderson R. Studies on the character and staining of fibrin. J Clin Pathol. 1962;15:401-413.
- ↵French JE, Jennings MA, Florey HW. Morphologic studies on atherosclerosis in swine. Ann N Y Acad Sci. 1965;12:780-794.
- ↵Moreland AF. Experimental atherosclerosis in swine. In: Roberts JC, Strauss R, eds. Comparative Atherosclerosis: The Morphology of Spontaneous and Induced Atherosclerotic Lesions in Animals and its Relation to Human Disease. New York, NY: Harper and Row; 1965:21-24.
- ↵Clopath P. Rapid production of advanced atherosclerosis in swine. Artery. 1977;3:429-438.
- ↵Shepard RK, Miller JG, Wickline SA. Quantification of atherosclerotic plaque composition in cholesterol-fed rabbits with 50-MHz acoustic microscopy. Arterioscler Thromb. 1992;12:1227-1234.
- ↵Henney AM, Wakeley PR, Davies MJ, Foster K, Hembry R, Murphy G, Humphries S. Localization of stromelysin gene expression in atherosclerotic plaques by in situ hybridization. Proc Natl Acad Sci U S A. 1991;88:8154-8158.
- ↵Shah PK, Falk E, Badimon JJ, Levy G, Ortiz A, Fallon J, Fuster V. Human monocyte-derived macrophages express collagenase and induce collagen breakdown in atherosclerotic fibrous caps: implications for plaque rupture. Circulation. 1993;88(suppl I):I-254. Abstract.
- ↵Rouis M, Nigon F, Lafuma C, Hornebeck W, Chapman MJ. Expression of elastase activity by human monocyte-macrophages is modulated by cellular cholesterol content, inflammatory mediators, and phorbol myristate acetate. Arteriosclerosis. 1990;10:246-255.
- ↵Wilcox JN, Smith KM, Schwartz SM, Godron D. Localization of tissue factor in the normal vessel wall and in the atherosclerotic plaque. Proc Natl Acad Sci U S A. 1989;86:2839-2843.
- ↵Lesnik P, Rouis M, Skarlatos S, Kruth HS, Chapman MJ. Uptake of exogenous free cholesterol induces upregulation of tissue factor expression in human monocyte-derived macrophages. Proc Natl Acad Sci U S A. 1992;89:10370-10374.
- ↵Rabbani LE, Loscalzo J. The relationship between thrombosis and atherosclerosis. In: Loscalzo J, Schafer AI, eds. Thrombosis and Hemorrhage. Boston, Mass: Blackwell Scientific Publications; 1994:769-789.