Hypercholesterolemia in Minipigs Impairs Left Ventricular Response to Stress
Association With Decreased Coronary Flow Reserve and Reduced Capillary Density
Background— Hypercholesterolemia induces functional and structural changes of the microvasculature and reduces coronary flow reserve in humans and experimental animals. The effect of hypercholesterolemia on left ventricular (LV) function in the absence of coronary stenosis is, however, unknown. Our objective was therefore to assess the effect of hypercholesterolemia and cholesterol withdrawal on LV function in the presence of advanced coronary plaques that do not cause stenosis.
Methods and Results— Twenty-eight minipigs on cholesterol diet for 34 weeks and 16 control pigs were studied. Seven hypercholesterolemic pigs were withdrawn from the diet for 26 weeks. LV function was assessed with cine-MRI, myocardial blood flow with colored microspheres, and capillary density with immunohistochemistry, and microvascular endothelial cell apoptosis with terminal dUTP nick-end labeling staining. Hypercholesterolemia (17±8 versus 268±150 versus 12±10 mg/dL LDL cholesterol, control versus hypercholesterolemic versus cholesterol withdrawal; P<0.001) induced atherosclerosis but not stenosis in the left coronary artery. Baseline cardiac output, ejection fraction, and stroke volume were similar in control and hypercholesterolemic pigs. In dobutamine stress test, cardiac output (P<0.05) and stroke volume (P<0.01) were lower in hypercholesterolemic pigs compared with controls. The impaired response to dobutamine was reversible by dietary cholesterol withdrawal. Hypercholesterolemia reduced endomyocardial coronary flow reserve (P<0.01) and capillary density (P<0.05) and induced capillary endothelial cell apoptosis. Hypercholesterolemic pigs failed to reduce vascular resistance in response to increased LV workload and pharmacological vasodilation.
Conclusion— LDL hypercholesterolemia in minipigs impaired LV response to dobutamine stress in the absence of coronary stenosis.
Received April 25, 2002; accepted May 25, 2002.
Coronary artery disease and ischemic cardiomyopathy are major causes of morbidity and mortality worldwide. Hypercholesterolemia is a major risk factor for coronary artery disease. Deterioration of left ventricular (LV) function has mainly been attributed to coronary insufficiency secondary to formation of atherosclerotic lesions in epicardial coronary arteries that reduce blood flow at rest or in response to increased demands.1 Hypercholesterolemia also reduces coronary flow reserve and induces microvascular dysfunction2–4⇓⇓ that may contribute to myocardial ischemia and LV dysfunction. Patients with syndrome X have myocardial ischemia without stenosing coronary artery lesions on the basis of microvascular endothelial dysfunction.5,6⇓ Microvascular dysfunction is also present in patients with dilated cardiomyopathy.7
To date, there have been no reports on the effects of hypercholesterolemia-induced microvascular changes on LV function in the absence of epicardial coronary stenosis. We therefore examined the effect of hypercholesterolemia on LV performance in diet-induced hypercholesterolemia and after dietary cholesterol lowering in minipigs. We also studied the relation of LV function after dobutamine stress testing with endo- and epimyocardial coronary flow reserve, LV capillary density, and atherosclerosis in the left anterior descending coronary artery (LAD).
The local institutional review board approved all animal procedures. Minipigs8 were obtained by crossbreeding Gottinger and Yucatan minipigs (Charles River Laboratories, Lyon, France). Twenty-eight minipigs were placed on a 4%-cholesterol diet for 34±5.7 (mean±SD) weeks. After that period, 7 hypercholesterolemic pigs were continued on a regular low-cholesterol chow for 26±1 weeks to study the effect of dietary cholesterol lowering. Sixteen age-matched pigs were used as controls.
Peripheral venous blood was drawn from an ear vein. Total cholesterol, HDL cholesterol, and triglyceride levels were measured in the university hospital routine laboratory. LDL cholesterol levels were calculated with the Friedewald formula. Circulating oxidized LDL (ox-LDL) was measured as previously described.9 Blood viscosity was measured in 8 additional control and 8 hypercholesterolemic pigs with an LVDV-II+ Brookfield viscosimeter (Analis) with UL-adapter at 37°C.
MRI was performed on a 1.5-T imager (Siemens Magnetom Vision). According to stress level, cardiac MRI was acquired in 8 short-axis views and 2 horizontal long-axis views. A segmented k-space gradient-echo sequence (TR/TE/α 4.8/60 ms/20°; matrix, 110×256; field of view, 196×250 mm; slice thickness, 6 mm) was used. The amount of phases acquired per R-R interval was adapted to the effective heart rate of the animal, so that both systole and the largest part of diastole were covered. Cardiac MRI was obtained during a resting state and intravenous infusion of 5 μg · kg−1 · min−1 dobutamine and a dose of dobutamine that was targeted to increase heart rate by 20%.
Endo- and epimyocardial borders of the LV were semi-automatically traced using Argus software on the MRI software. A defined correction for through-plane long-axis shortening was performed. This delineation was used to obtain global functional parameters: end-diastolic (EDV) and end-systolic (ESV) volume, ejection fraction (EF), and stroke volume (SV).10
Coronary Artery Instrumentation
Coronary arteries were instrumented in deep propofol anesthesia as described before,11 and intravascular ultrasound (IVUS) loops of the proximal LAD were recorded to exclude significant epicardial coronary stenosis (IVUS Visions Five-64F/X, Endosonics). Lumen size was determined offline on calibrated IVUS-loops with the built-in image analysis tool of the IVUS device. Coronary angiograms were obtained from all hypercholesterolemic pigs.
Myocardial Blood Flow and Resistance
LV myocardial blood flow was determined by injecting 15-μm colored microspheres (CMs; Dye-Trak, Triton Technology) as described elsewhere.12 Hemodynamic parameters were allowed to stabilize for 10 minutes before 5 mL of thoroughly mixed CMs (18x106) were injected into the LV. Simultaneously arterial and ventricular pressures and heart rate (HR) were recorded. Dobutamine was started and titrated in 2.5 μg · kg−1 · min−1 steps to increase HR by 20% when a second dose of CMs was injected. Hemodynamics was recorded and dobutamine was discontinued; 0.8 mg/kg body weight of dipyridamole was injected into the LV, and a third dose of CMs was injected 5 minutes thereafter. The heart was removed, fixed, and divided into 5 slices of equal thickness. The central slice was used for spectrophotometric analysis of CM distribution. Coronary flow reserve was defined as the ratio of myocardial flow after dipyridamole to resting blood flow. Coronary vascular resistance was calculated based on flow and mean arterial pressure.
Histology and Immunohistochemistry
The LAD was excised after perfusion fixation with paraformaldehyde and divided into rings for histological processing. Blocks of myocardium from the anterior free wall were likewise excised from the hearts and cryoembedded for immunohistochemistry and terminal dUTP nick-end labeling (TUNEL)-staining. Coronary lesions were determined morphometrically by delineating the external elastic lamina, the internal elastic lamina, and the lumen. Transmural myocardial sections were stained with biotinylated Bandeiria simplicifolia isolectin B4 (Vector Laboratories) and developed with streptavidin-peroxidase conjugate to stain endothelial cells. Capillary density was assessed histomorphometrically on a Quantimet 600 (Leica) by manual delineation of capillaries in 8 high-power fields in the endomyocardial and 8 high-power fields in the epimyocardial layers of the free anterior LV wall, and a mean was established. Care was taken to only analyze areas with transverse capillary profiles.
Endothelial Cell Apoptosis
TUNEL assay was performed according to the manufacturer protocols (Intergen). Endothelial cells were labeled with a rabbit polyclonal anti-human von Willebrand Factor antibody (DAKO). Biotinylated goat-anti-rabbit antibody was used as secondary antibody. Omission of the primary antibody served as control. The peroxidase reaction was visualized with 3-amino-9-ethylcarbazole (Vector Laboratories) and 0.006% H2O2. The presence of apoptotic endothelial cells was semi-quantitatively scored on 5 randomly chosen high-magnification fields (200×) of 4 different sections of each animal. Apoptotic endothelial cells were considered present if ≥3 TUNEL-positive endothelial cells per high power field were counted.
Groups were compared by nonparametric ANOVA followed by Dunnett’s test (SPSS, version 10). Paired nonparametric testing compared hemodynamics and myocardial function at resting conditions and at stress testing. Correlations were determined by Spearman test. Multiple regression analysis was performed to determine the relation of cardiac output (CO) with LDL cholesterol, coronary flow reserve, capillary density, and atherosclerosis in LAD. Values of P<0.05 were considered statistically significant.
Lipids and Atherosclerosis
The cholesterol-rich diet induced LDL hypercholesterolemia (Table 1). Levels of oxidized LDL (ox-LDL) were higher in hypercholesterolemic pigs. Triglyceride levels were similar in control and hypercholesterolemic pigs (Table 1). Blood viscosity was 26% higher in hypercholesterolemic versus control pigs (5.4±0.63 versus 4.7±0.28 mPa×s; P<0.01). Cholesterol withdrawal resulted in a decrease of LDL cholesterol and ox-LDL (Table 1).
Mean intimal area in the proximal LAD was 0.81±0.17 mm2 in hypercholesterolemic compared with 0.15±0.03 mm2 in control pigs (n=13 and 16, P<0.01). Lumen area of the LAD, assessed by IVUS, was 9.2±0.6 mm2 in control and 9.6±0.4 mm2 in hypercholesterolemic pigs (P=NS). Coronary angiograms did not reveal significant stenosis. Cholesterol withdrawal did not significantly change the available lumen area of the LAD determined by IVUS (10.4±1.4 versus 10.1±1.1 mm2, hypercholesterolemic versus cholesterol withdrawal; n=6, P=NS). Mean intimal area in pigs after dietary cholesterol withdrawal was not different from that in hypercholesterolemic pigs (1.14±0.45 versus 0.81±0.17 mm2; n=7 and n=17).
Under resting conditions, HR, CO, EF, SV, EDV, and ESV were similar in control and hypercholesterolemic pigs (Table 2). To increase HR by 20%, higher dobutamine doses were required in the hypercholesterolemic and cholesterol withdrawal groups compared with controls (ANOVA P<0.05) (Table 2). Dobutamine stress test significantly increased CO in controls but not in hypercholesterolemic pigs (37±28% versus 3.3±27%, P<0.05) (Table 2). In hypercholesterolemic pigs, EF (P<0.05) and SV (P<0.01) decreased after dobutamine stress test (Table 2). There were no differences between hypercholesterolemic and cholesterol-lowering pigs before cholesterol withdrawal was initiated, both at rest and after dobutamine stress. In cholesterol withdrawal pigs, resting LV function was similar before and after cholesterol withdrawal (data not shown). After 6 months of cholesterol withdrawal, CO, EF, and SV in response to dobutamine were similar those of control pigs (Table 2).
After dobutamine stress test, SV was decreased in hypercholesterolemic pigs and increased in controls and pigs after cholesterol withdrawal (ANOVA P<0.05). The increase in CO was higher in controls than in hypercholesterolemic pigs. After dietary cholesterol lowering, increase in CO was not different from that in control pigs (ANOVA P<0.05) (Figure 1).
Blood pressure tended to be lower in hypercholesterolemic pigs at rest, after dobutamine and after dipyridamole (Table 3). Resting endo- and epimyocardial blood flow tended to be higher in hypercholesterolemic versus control pigs (Table 3). Endomyocardial flow was significantly higher than epimyocardial flow in control pigs (P<0.05) but not in hypercholesterolemic pigs (Table 3). The endomyocardial-to-epimyocardial flow ratio was significantly higher in control pigs than in hypercholesterolemic pigs (P<0.05) (Figure 2). Coronary flow reserve was reduced in the endomyocardium of hypercholesterolemic pigs (P<0.01) (Figure 2). Dobutamine increased endo- and epimyocardial flow to 175% and 168%, respectively, of flow under resting conditions (P<0.01) in control pigs compared with 121% and 118% (P=NS), respectively, in hypercholesterolemic pigs. Dipyridamole caused a redistribution of blood flow away from the subendocardium in hypercholesterolemic pigs but not in control pigs (endomyocardial/epimyocardial flow ratio, 0.73±0.086 and 1.08±0.055, respectively; P<0.05) (Figure 2).
Dobutamine, to increase HR by 20%, and dipyridamole decreased mean endomyocardial and epimyocardial resistance in control pigs (P<0.001 for both) but failed to do so in hypercholesterolemic pigs (Figure 3).
Capillary Density and Endothelial Cell Apoptosis
Endo- and epimyocardial capillary densities were not different. Mean LV capillary density was smaller in hypercholesterolemic pigs compared with control pigs (P<0.05) (Figure 4). Oil red O staining of adjacent sections revealed single small plaques in branching points of precapillary arterioles but no disseminated microvascular coronary artery disease or overt lipid deposition in the myocytes themselves. Hematoxylin/eosin staining demonstrated no gross abnormalities of the examined myocardium. TUNEL staining scored semiquantitatively revealed microvascular endothelial cell apoptosis in the myocardium of 6 out of 13 hypercholesterolemic pigs compared with only 1 out of 8 of control pigs.
CO, Myocardial Flow, and Capillary Density
Figure 5 demonstrates that the change in CO after dobutamine stress testing correlated with endomyocardial coronary flow reserve and with capillary density. The change in CO after dobutamine stress did not correlate with the mean intimal area in the proximal LAD (Figure 5). Multiple regression analysis—using a model containing the change in CO, LDL cholesterol, coronary flow reserve, capillary density, and plaque area in the LAD—showed that coronary flow reserve predicted the change in CO after stress test (R2=0.42; P=0.04).
The main findings of the present study are that LDL hypercholesterolemia in minipigs is associated with impaired LV performance in response to inotropic stress. Cholesterol withdrawal restored LV performance in dobutamine stress testing. The atherogenic diet induced atherosclerosis in the LAD as reported before8; however, coronary stenosis in the epicardial arteries was not detected, excluding major coronary stenosis as the cause for LV dysfunction. Microvascular dysfunction was evidenced by reduced endomyocardial coronary flow reserve, failure to decrease resistance in response to increased workload or pharmacological vasodilatation, disturbed transmural flow distribution, and reduced left capillary density. Thus hypercholesterolemia-induced microvascular changes were linked to LV dysfunction in the absence of stenosing coronary artery disease.
In hypercholesterolemic pigs, vascular dysfunction, as has been described in numerous clinical studies,13,14⇓ was demonstrated by reduced coronary flow reserve and disturbed flow regulation. Dipyridamole caused a distribution of transmural flow away from the endomyocardium. This intracoronary steal effect occurred even in the absence of epicardial stenosis and may have affected LV function, which is strongly dependent on endomyocardial flow.15 Vascular resistance did not decrease in response to increased oxygen demand owing to dobutamine in hypercholesterolemic pigs, suggesting that endothelial dysfunction was in part responsible for impaired LV function with increased demand. The response of the LV to dobutamine stress testing correlated with indices of vascular dysfunction such as coronary flow reserve in the endomyocardium and myocardial capillary density, but not with the degree of epicardial atherosclerosis. Reduced capillary density evidenced also structural alterations in the microvasculature. The reduced capillary density in hypercholesterolemic pigs may reflect an imbalance between capillary loss because of endothelial cell apoptosis and capillary replacement.3 We have demonstrated microvascular endothelial cell apoptosis, whereas others have demonstrated impaired angiogenesis.16,17⇓ Oxidized LDL induces endothelial cell apoptosis in vitro.18 In humans, coronary flow reserve inversely correlated with levels of autoantibody titer against oxidized LDL.19 Circulating levels of oxidized LDL were increased in hypercholesterolemic pigs and returned to baseline after lipid lowering. Hypercholesterolemia also reduces circulating endothelial cell progenitor cells,20 whereas lipid-lowering agents were found to restore the number of endothelial progenitor cells that could contribute to the neocapillarization of the myocardium after lipid lowering.21,22⇓ Apoptosis of myocytes, which may represent a further step in the progression to heart failure,23,24⇓ was not detected in the hearts of hypercholesterolemic pigs.
Hypercholesterolemia increases blood viscosity that may affect coronary flow reserve, as demonstrated by Rim and colleagues.25 In the latter study, triglyceride infusion acutely increased plasma triglyceride levels >100-fold and viscosity 6 to 7-fold. In our pigs, diet feeding resulted in an increase in LDL cholesterol but not triglycerides. Increased blood viscosity may have contributed to the reduced endomyocardial flow reserve, in addition to reduced capillary density and the shift of transmural flow distribution from endo- to epimyocardium. Lower arterial pressure after dipyridamole may also have contributed to lower coronary flow. To adapt for differences in arterial pressure, coronary resistance was calculated. It shows that control pigs but not hypercholesterolemic pigs are able to reduce coronary resistance after pharmacological vasodilatation.
Deterioration of LV function as a consequence of microvascular dysfunction and decreased coronary reserve in the absence of epicardial stenosis may be relevant to different clinical entities. Patients with idiopathic dilated cardiomyopathy have documented vitamin C-reversible microvascular dysfunction26,27⇓ and decreased capillary density.28 In patients with syndrome X, cardiovascular symptoms may result from microvascular dysfunction.5,6⇓ Despite the presence of normal coronary angiograms, a majority of patients with microvascular angina have epicardial plaques detectable with IVUS.29 Early atherosclerosis is associated with an impaired endothelium-dependent dilatation of the microvasculature and impaired coronary blood flow regulation.13 Abnormal myocardial blood flow distribution with shift of flow from endo- to epimyocardium was shown by myocardial contrast echocardiography in patients with angina but normal coronary angiogram.30 Hypercholesterolemia increases oxygen radical production. There is growing evidence that reactive oxygen species are involved in both myocardial and vascular dysfunction.2,31,32⇓⇓ They worsen myocardial function by inducing vascular dysfunction or by direct effects on myocardial cells.33 Vitamin C augmented the inotropic response in humans with normal LV function,34 and allopurinol enhanced the contractile response to dobutamine in dogs with pacing-induced heart failure.35
In conclusion, hypercholesterolemia impaired LV performance in the absence of coronary stenosis. Loss of contractility is linked to microvascular dysfunction and decreased capillary density. These features of microvascular disease are reversible on dietary normalization. Our study shows that hypercholesterolemia may induce myocardial dysfunction before epicardial stenosis occurs and may help explain the clinical observation that the degree of LV dysfunction is not entirely explained by the angiographic degree of coronary atherosclerosis.
This study was supported by the Interuniversitaire Attractiepolen Program (P5/01/02) of the Belgian government and by the Fonds voor Wetenschappelijk Onderzoek (FWO) grant G.0088.02. Dr Theilmeier was a recipient of a postdoctoral fellowship from the Deutsche Forschungsgemeinschaft and is currently a member of the Interdisciplinary Center for Clinical Research Münster, “IZKF-The chronic disease” University of Münster, Germany. Dr Verhamme is a doctoral fellow of the FWO, Vlaanderen. We thank M. Landeloos and V. Leunens for excellent technical assistance.
↵*Drs Theilmeier and Verhamme contributed equally to this work.
- ↵Rodriguez-Porcel M, Lerman A, Ritman EL, et al. Altered myocardial microvascular 3D architecture in experimental hypercholesterolemia. Circulation. 2000; 102: 2028–2030.
- ↵Holvoet P, Theilmeier G, Shivalkar B, et al. LDL hypercholesterolemia is associated with accumulation of oxidized LDL, atherosclerotic plaque growth, and compensatory vessel enlargement in coronary arteries of miniature pigs. Arterioscler Thromb Vasc Biol. 1998; 18: 415–422.
- ↵Holvoet P, Vanhaecke J, Janssens S, et al. Oxidized LDL and malondialdehyde-modified LDL in patients with acute coronary syndromes and stable coronary artery disease. Circulation. 1998; 98: 1487–1494.
- ↵Nagel E, Lehmkuhl HB, Bocksch W, et al. Noninvasive diagnosis of ischemia-induced wall motion abnormalities with the use of high-dose dobutamine stress MRI: comparison with dobutamine stress echocardiography. Circulation. 1999; 99: 763–770.
- ↵Varenne O, Pislaru S, Gillijns H, et al. Local adenovirus-mediated transfer of human endothelial nitric oxide synthase reduces luminal narrowing after coronary angioplasty in pigs. Circulation. 1998; 98: 919–926.
- ↵Kowallik P, Schulz R, Guth BD, et al. Measurement of regional myocardial blood flow with multiple colored microspheres. Circulation. 1991; 83: 974–982.
- ↵Jang JJ, Ho HK, Kwan HH, et al. Angiogenesis is impaired by hypercholesterolemia: role of asymmetric dimethylarginine. Circulation. 2000; 102: 1414–1419.
- ↵Dimmeler S, Haendeler J, Galle J, et al. Oxidized low-density lipoprotein induces apoptosis of human endothelial cells by activation of CPP32-like proteases: a mechanistic clue to the ‘response to injury‘ hypothesis. Circulation. 1997; 95: 1760–1763.
- ↵Scarabelli T, Stephanou A, Rayment N, et al. Apoptosis of endothelial cells precedes myocyte cell apoptosis in ischemia/reperfusion injury. Circulation. 2001; 104: 253–256.
- ↵Rim SJ, Leong-Poi H, Lindner JR, et al. Decrease in coronary blood flow reserve during hyperlipidemia is secondary to an increase in blood viscosity. Circulation. 2001; 104: 2704–2709.
- ↵Treasure CB, Vita JA, Cox DA, et al. Endothelium-dependent dilation of the coronary microvasculature is impaired in dilated cardiomyopathy. Circulation. 1990; 81: 772–779.
- ↵Abraham D, Hofbauer R, Schafer R, et al. Selective downregulation of VEGF-A165, VEGF-R1, and decreased capillary density in patients with dilative but mot ischemic cardiomyopathy. Circ Res. 2000; 87: 644–647.
- ↵Ide T, Tsutsui H, Kinugawa S, et al. Direct evidence for increased hydroxyl radicals originating from superoxide in the failing myocardium. Circ Res. 2000; 86: 152–157.
- ↵Cesselli D, Jakoniuk I, Barlucchi L, et al. Oxidative stress-mediated cardiac cell death is a major determinant of ventricular dysfunction and failure in dog dilated cardiomyopathy. Circ Res. 2001; 89: 279–286.
- ↵Mak S, Newton GE. Vitamin C augments the inotropic response to dobutamine in humans with normal left ventricular function. Circulation. 2001; 103: 826–830.
- ↵Ukai T, Cheng CP, Tachibana H, et al. Allopurinol enhances the contractile response to dobutamine and exercise in dogs with pacing-induced heart failure. Circulation. 2001; 103: 750–755.