Author + information
- Received June 1, 2000
- Revision received September 5, 2000
- Accepted October 12, 2000
- Published online February 1, 2001.
- Martin Rodriguez-Porcel, MD∗,
- Amir Lerman, MD∗,
- Patricia J.M Best, MD∗,
- James D Krier, MS†,
- Claudio Napoli, MD, PhD‡,§ and
- Lilach O Lerman, MD, PhD†,* ()
- ↵*Reprint requests and correspondence: Dr. Lilach O. Lerman, Division of Hypertension, Mayo Clinic, 200 First Street SW, Rochester, Minnesota 55905
We intended to study the effect of hypercholesterolemia (HC) on myocardial perfusion and permeability response to increased cardiac demand.
Hypercholesterolemia is associated with increased incidence of cardiac events and characterized by impaired coronary vascular function, possibly mediated partly through increased pro-oxidative conditions in plasma and tissue. However, it is yet unclear whether HC is also associated with impaired myocardial perfusion and vascular permeability responses in vivo.
For 12 weeks pigs were fed a normal, HC or HC diet supplemented daily with antioxidants (HC + AO, 100 IU/kg vitamin E and 1 g vitamin C). Myocardial perfusion and vascular permeability were measured in vivo using electron beam computed tomography before and after cardiac challenge with intravenous adenosine. Plasma and tissue oxidative status was determined ex vivo.
Plasma cholesterol increased in all cholesterol-fed pigs but was associated with increased markers of oxidative stress only in HC pigs. Myocardial perfusion increased in response to adenosine in normal and HC + AO (+37 ± 13% and +58 ± 22%, respectively, p < 0.05 vs. baseline) but not in HC, whereas vascular permeability index increased only in HC pigs (+ 92 ± 25%, p = 0.002). In HC animals, tissue endogenous oxygen radical scavengers and antioxidant vitamins were depleted and LDL oxidizability enhanced, but both were normalized in HC + AO pigs. Myocardial perfusion response was directly, and permeability inversely, associated with plasma and tissue vitamin concentrations.
This study demonstrates that experimental HC is associated with blunted myocardial perfusion and increased vascular permeability responses in vivo to increased cardiac demand, which may be partly mediated by a shift in oxidative status.
Hypercholesterolemia (HC) is considered a major risk factor for progression of coronary atherosclerosis and is associated with an increase in the incidence of myocardial ischemia and cardiac events (1,2). Hypercholesterolemia is characterized by coronary endothelial dysfunction, the hallmark of which is an altered vasodilation to endothelial-dependent vasodilators (3,4). Because many regulatory mechanisms contribute to the determination of basal vascular tone, it appears to be balanced under resting conditions, and abnormalities are often apparent only in stimulated situations (5)such as exercise or increased cardiac demand. Alterations in microvascular permeability (6)have also been postulated to be a measure of coronary endothelial dysfunction in HC (7,8)and may indicate functional or structural loss of coronary vascular integrity. Nevertheless, the functional significance of these impairments in coronary vascular function remains unclear, because it is yet to be elucidated whether HC, as a sole risk factor, is associated with significant abnormalities of myocardial perfusion in vivo.
One mechanism that may underlie the abnormal coronary vascular function in HC is an alteration in the oxidative status (9)accompanied by increased production of several oxidants such as peroxynitrites, as well as peroxidation compounds and oxidative end-products such as PGF2-α isoprostanes (9–11). Furthermore, a shift in oxidative status with a decrease in nitric oxide (NO) and an increase in oxygen radicals may have a deleterious effect on vascular permeability (12–14). Thus, these mechanisms may play a pivotal role in inducing abnormalities of both myocardial perfusion and vascular permeability associated with HC.
Quantification of the myocardial perfusion and permeability abnormalities that exist in HC in vivo may have important implications for understanding and management of HC as a risk factor for coronary heart disease. However, few methods are capable of accurately and noninvasively measuring myocardial perfusion and permeability in vivo. Electron beam computed tomography (EBCT), a fast computed tomography (CT) scanner, provides a unique tool to accurately (15), reproducibly (16)and noninvasively study in vivo myocardial perfusion (15,17)and microvascular permeability (18–20). Electron beam computed tomography measurements of myocardial perfusion correlate well with concurrent radioactive microspheres measurements (15)and show little short-term variability (16). This method uses fast-acquired cross-sectional tomographic images during a single intravenous bolus injection of contrast media and permits repeated study of myocardial perfusion in animals (17)and humans (15)within the same myocardial region, thus minimizing effects of spatial heterogeneity or imaging artifacts on the calculated perfusion or microvascular permeability. Hence, EBCT provides a unique tool to noninvasively calculate myocardial vascular function in different pathophysiological states in vivo.
The present study was designed to test the hypothesis that experimental HC is associated with abnormal myocardial perfusion and microvascular permeability response to increased myocardial demand. Furthermore, this study demonstrates that these abnormalities are due partly to a shift in oxidative status.
All procedures were designed in accordance with the National Institutes of Health Guidelines and were reviewed and approved by the Institutional Animal Care and Use Committee. Domestic crossbred pigs (23 to 35 kg) were divided in three groups (subsequently treated for 12 weeks): group 1 (n = 8) received a normal diet. Pigs in groups 2 (HC, n = 8) and 3 (n = 6) were placed on an atherogenic diet of 2% cholesterol and 15% lard by weight (TD 93296, Harlan Teklad, Madison, Wisconsin). In addition, animals from group 3 (HC + AO) received daily supplementation of an antioxidant combination consisting of vitamin E (100 IU/kg) and vitamin C (1,000 mg). These doses of vitamins are higher than those used clinically but are commonly used experimentally to explore their effects and participation in disease mechanisms (21–24)
In vivo studies, plasma lipid profiles (Roche, Nutley, New Jersey), total isoprostanes (25), low density lipoprotein (LDL) oxidizability, tissue oxygen radical scavengers and circulating levels of vitamin E and C were determined after 12 weeks of diet in all three groups.
Assessment of myocardial perfusion
Animals were anesthetized with a mixture of ketamine and xylazine, intubated and mechanically ventilated with room air. Anesthesia was maintained with a constant infusion of ketamine (17.5 mg/kg/h) and xylazine (2.3 mg/kg/h) in saline. The carotid artery and jugular vein were cannulated, and under fluoroscopic guidance an 8F guide catheter was positioned in the descending aorta for on-line measurement of arterial pressure. A 5-F pigtail catheter was inserted in the right atrium for subsequent injection of contrast media.
Animals were then placed supine in the EBCT (Imatron C-150, Imatron Inc., South San Francisco, California) gantry. Using localization scans, images containing the mid-left ventricle (mid-LV) were identified in the neutral axis of the heart. After allowing 30 min of stabilization and saline infusion (5 ml/min), a baseline perfusion study was performed. Two tomographic levels were selected and 40 consecutive end-diastolic scans were obtained (17)at intervals of one to three heartbeats after a bolus injection of nonionic contrast iopamidol (Isovue-370, Squibb Diagnostics, Princeton, New Jersey, 0.3 cc/kg over 2 s). Ten minutes later an IV infusion of adenosine (approximately 400 μg/kg/min) was initiated (simulating a cardiac stress test). After stabilization of blood pressure, hemodynamic measurements were obtained and the perfusion study was repeated. After completion of the studies, animals were euthanized by intravenous administration of 30 mg/kg pentobarbital sodium (Sleepaway, Fort Dodge Laboratories, Iowa). Images were then transferred and stored on a Sun workstation
EBCT data analysis
Regions of interest were traced in the anterior wall and chamber of the LV and time-density curves (17,26)were obtained (Fig. 1A). To define the transmural distribution of blood flow, the myocardial region of interest was further subdivided into two equidistant subepicardial and subendocardial zones. Time-density curves were then generated and analyzed with commercially available computer software (KaleidaGraph, Synergy Software, Reading, Pennsylvania) containing custom-designed algorithms, which allow modeling of intravascular and extravascular transit of contrast media (17)(Fig. 1B).
Calculation of myocardial perfusion used the intravascular curve, which portrays transit of contrast media within the intramyocardial vasculature. Intramyocardial vascular blood volume (BV) (ml/cc tissue) (17,27–29)and mean transit time (MTT) (s) (30)were calculated as previously described.
Myocardial perfusion (ml/g/min) was calculated (15,17)as 60 × (BV/MTT)/[1.05 × (1−BV)]. Perfusion of the subendocardial and subepicardial regions was similarly obtained and their ratio (endocardial/epicardial) was calculated.
For calculation of myocardial vascular permeability index, the rate of vascular leakage was estimated from the extravascular accumulation of contrast media, as previously described in the heart (18,19)and the kidney (31). Contrast extraction rate (analogous to permeability-surface area product) was then calculated from the curve depicting the extravascular permeation of contrast media (17)as: where slope is the maximal slope of the ascending arm of the extravascular curve, MTT estimates the duration of leakage and 1.05 g/cc is the specific density of the myocardium. Blood volume was subsequently used as a surrogate for vascular surface area (19)and permeability index (arbitrary units; AU) was calculated as extraction rate/BV (19).
LDL isolation and oxidation
Plasma was obtained from each experimental group and LDL was rapidly isolated by two consecutive steps of discontinuous density ultracentrifugation in a KBr gradient as previously described (32). A Sephacryl S-300 column was used to desalt and remove low-molecular-weight components. Fresh LDL preparations were immediately used to minimize spontaneous peroxidation (32). Protein content was determined by the Lowry assay (33). Low density lipoprotein (300 μg/ml) was incubated for 12 h at 37° C with 1 μm copper sulfate and lag time was measured spectrophotometrically as previously detailed (32,34). Malondialdehyde content was assayed by thiobarbituric acid and relative electrophoretic mobility (REM) of lipoproteins on agarose gel (0.8%) was assessed (32,34). Vitamin E and C concentration in plasma and tissue was determined by high-performance liquid chromatography (35).
Determinations of tissue antioxidant enzymes
Tissue concentrations of oxygen-radical scavengers were determined in left coronary artery segments and in the myocardium. Homogenates in potassium phosphate buffer, pH 7.4, containing 10 mol/l deferoxamine, 0.03% butylated hydroxytoluene and 2% ethanol, equilibrated with nitrogen (to reduce auto-oxidation) were centrifuged at 1,000 gfor 15 min at 4 °C to remove nuclei and tissue debris. The supernatant was centrifuged again at 30,000 gfor 35 min at 4°C. Glutathione peroxidase, catalase, copper zinc-superoxide dismutase (SOD) and manganese-SOD tissue activities were determined spectrophotometrically as previously described (36). All enzyme activities were normalized for protein content (33).
Data are expressed as mean ± SEM. Comparisons between the groups were performed using analysis of variance (ANOVA) followed by the Bonferroni corrected ttest or unpaired Student ttest, as applicable. Comparisons within each group were done using paired Student ttest. Regressions were calculated by the least-squares method. Statistical significance was accepted for p ≤ 0.05.
Basal mean arterial pressure and heart rate were similar among the three experimental groups (Table 1). In response to IV adenosine, mean arterial pressure decreased significantly and similarly in all groups (Table 1).
Serum cholesterol and 8-iso-PGF2alpha isoprostane levels
Total and LDL cholesterol levels were significantly increased in both groups of HC pigs compared with normal pigs (Table 1). There was no difference in cholesterol levels between HC and HC + AO pigs (Table 1), indicating that antioxidant supplementation had no effect on cholesterol levels.
The increase in serum cholesterol was associated with a significant increase in plasma isoprostanes in HC animals compared with normal pigs (p = 0.005, Table 1), representing increased oxidative stress (37). The increase in plasma isoprostanes observed in HC was attenuated, although not completely normalized, in HC animals receiving long-term vitamin supplementation, because their levels were similar to those in both groups.
Anterior wall myocardial perfusion was similar under basal conditions among the three experimental groups (normal: 0.95 ± 0.1, HC: 0.89 ± 0.1, and HC + AO: 0.93 ± 0.1 ml/min/g myocardial tissue, ANOVA, p = 0.48). Adenosine infusion led to a significant increase in myocardial perfusion in normal animals (to 1.25 ± 0.2 ml/min/g tissue, p = 0.02, Fig. 2, top panel) but not in HC pigs (to 0.93 ± 0.1 ml/min/g tissue, p = 0.28). On the other hand, in HC + AO pigs, the response of myocardial perfusion to adenosine was normalized (to 1.67 ± 0.2 ml/min/g tissue, p = 0.03, Fig. 2, top panel).
The basal ratio of endocardial/epicardial perfusion was similar in the normal, HC and HC + AO groups (1.26 ± 0.27, 1.33 ± 0.13, and 1.21 ± 0.14, respectively, p = 0.914). Intravenous infusion of adenosine was not associated with significant intramyocardial redistribution of perfusion in any group.
Myocardial microvascular permeability
The 12-week diet led to similar basal myocardial permeability index in normal, HC and HC + AO pigs (1.61 ± 0.2, 1.15 ± 0.2, and 1.30 ± 0.2 AU, respectively, p = 0.34). In response to adenosine, there was no change in permeability index in normal pigs (to 1.76 ± 0.3 AU, p = 0.56, Fig. 2, bottom panel). However, in HC pigs adenosine infusion led to a significant increase in the permeability index (to 2.08 ± 0.3 AU, p = 0.002, Fig. 2). ⇓Similar to normal pigs, in HC + AO pigs the permeability index response to adenosine was preserved (1.02 ± 0.46 AU, p = 0.21, Fig. 2, bottom panel).
Status of antioxidant defenses and LDL oxidizability
Plasma vitamins E and C were significantly decreased in HC pigs compared with normal (p = 0.02, Table 2), although tissue levels were similar. In contrast, tissue scavenging enzyme activities were all significantly lower in HC pigs compared with both normal and HC + AO pigs (Table 2). The susceptibility of LDL for oxidation was also markedly increased in the HC pigs, as evidenced by an increase in LDL-malondialdehyde and LDL-REM and a shortening of LDL-lag time (Table 2).
Twelve weeks of daily supplementation resulted in a significant increase in plasma levels of both vitamins E and C in the HC + AO pigs compared with both normal and HC pigs (both p < 0.00001, Table 2). More importantly, in the HC + AO pigs antioxidant supplementation completely normalized tissue levels of scavenging enzymes, all of which were also significantly higher than in the HC pigs, indicating restoration of endogenous antioxidant and scavenging defenses. Tissue levels of vitamin E were significantly higher than normal (p = 0.0002), whereas LDL-REM was significantly lower (p = 0.015) and LDL-lag time significantly longer (p = 0.001) than in normal pigs as well as in HC pigs (p < 0.00001 for all measurements).
The change in myocardial perfusion in response to adenosine in HC and HC + AO pigs correlated directly with plasma concentrations of vitamins E and C (r = 0.68; p = 0.01 and r = 0.48; p = 0.03, respectively, Fig. 3, top left panel), LDL lag time (r = 0.68, p = 0.0006), and tissue levels of copper zinc-SOD (r = 065, p = 0.04). Moreover, the change tended to correlate with tissue levels of vitamins E and C (r = 0.53, p = 0.09 and r = 0.53, p = 0.09, respectively, Fig. 3, top right panel). Changes in microvascular permeability correlated inversely with plasma concentrations of vitamins E and C (r = −0.58; p = 0.007 and r = −0.59; p = 0.006, respectively, Fig. 3, bottom left panel), as well as with tissue levels of vitamin E (Fig. 3, bottom right panel) and manganese-SOD (r = −0.68; p = 0.02 each). In normal pigs there were no significant correlations; myocardial perfusion responses to adenosine did not correlate significantly in any group with either total or LDL cholesterol or with 8-iso-PGF2alpha isoprostanes.
This study demonstrates that experimental HC, a pro-oxidant state (9–11), is associated with an attenuated myocardial perfusion and increased microvascular permeability in response to increased cardiac demand. Moreover, these abnormalities may be partly mediated by an increase in oxidative stress.
Hypercholesterolemia and myocardial perfusion
Hypercholesterolemia is a major risk factor for development of coronary atherosclerosis, and several clinical studies have shown that its presence is associated with an increased incidence of cardiac events (1,2). Early HC is characterized by coronary endothelial dysfunction. Because the increase in coronary flow and myocardial perfusion is in part mediated by the endothelium, endothelial dysfunction may conceivably contribute to myocardial ischemia. Indeed, studies from our laboratory and others have demonstrated that in humans coronary endothelial dysfunction per se may lead to myocardial perfusion defects (38,39)and increased prevalence of myocardial events (40). Hence, it was postulated that early HC might impose a deleterious effect on myocardial perfusion response to increased cardiac demand. The current study supports this hypothesis in demonstrating that experimental HC is characterized by an attenuated myocardial perfusion response to increased cardiac demand. Nevertheless, in addition to increased cardiac demand, myocardial perfusion response in this study was likely also elicited by both endothelium-dependent and independent mechanisms triggered by modest doses of IV adenosine. The transmural distribution of perfusion was unaffected by HC, suggesting homogeneity of this impairment throughout the anterior cardiac wall.
Hypercholesterolemia and oxidative stress
One of the mechanisms that may underlie endothelial dysfunction is an increase in pro-oxidative conditions (41). Hypercholesterolemia is associated with increased production of oxygen radicals and increased oxidation of LDL cholesterol (9–11). Oxidation of LDL is implicated in the pathogenesis of early atherogenesis in human fetuses (42)and children (43). Oxidized LDL may directly induce endothelial dysfunction that can be prevented by vascular incorporation of vitamin E (44). Accordingly, in the present study LDL oxidizability was increased in HC pigs and significantly reduced by the antioxidant vitamins E and C. Moreover, production of oxygen radicals may further decrease the bioavailability of NO, thus leading to inadequate endothelium-dependent responses (41). This study shows that an increase in oxidative stress may contribute to the impaired myocardial perfusion response in HC, as evidenced by the preservation of the myocardial perfusion response achieved when HC pigs received long-term combination antioxidant supplementation.
Hypercholesterolemia and microvascular permeability
In addition to providing the myocardium with adequate blood supply, intramyocardial circulation has other momentous roles. For example, the integrity of the endothelium serves as a barrier between the circulation and the vascular wall and may thus be regarded as a parameter of endothelial function. Furthermore, in pathophysiological states such as HC, abnormalities in myocardial vascular permeability (7)have been suggested as a parameter of endothelial dysfunction. Although our model of early atherosclerosis does not show development of early atheromatous plaques or coronary luminal obstruction (45), it is characterized by increased fenestration of the coronary internal elastic lamina (46). This study supports the notion that such changes may have functional significance upon increased demand. It also demonstrates for the first time that experimental HC is associated with an increase in microvascular permeability in response to cardiac challenge in vivo and that this impairment may be mediated by an increase in oxidative status.
Vitamin E and vitamin C both are potent antioxidants, which retard oxidative modification of LDL (47)and endothelial dysfunction (44). Such effects may have clinical benefits (48)that are likely to be dependent, among the rest, on the duration, dose, timing and combination of dietary supplementation, as recently reviewed in detail (24,49). For example, antioxidant effects may be partly synergistic, because vitamin C contributes to regeneration of vitamin E. In this study we investigated another important effect of antioxidants and showed that daily dietary supplementation of a high-dose combination of both vitamins E and C was effective in elevating their circulating levels and tissue concentrations and in normalizing oxidative status, particularly myocardial perfusion and permeability response to challenge. A direct vasodilatory effect of vitamin E at the relatively high doses utilized in this study cannot be excluded either. The preservation of endogenous scavenging activities in the heart of the vitamin-treated group can also improve cardiac function.
In summary, this study demonstrates for the first time that experimental HC is associated with blunted myocardial perfusion and increased microvascular permeability in vivo in response to cardiac challenge. This study also indicates that these abnormalities are associated with an increase in oxidative stress in HC.
The authors are grateful to the staff of the EBCT for their technical assistance with the performance of experiments and to Y. Lee and F. de Nigris for their skillful technical assistance.
☆ This study was supported by grant numbers HL-03621 and HL-63282 from the National Institutes of Health, by the Bruce and Ruth Rappaport Program in Vascular Biology grant, by the Mayo Foundation, and by grant ISNIH 99.56980 (C. Napoli)-32767.
- analysis of variance
- arbitrary units
- blood volume
- computed tomography
- electron beam computed tomography
- HC + AO
- HC plus long-term vitamin supplementation
- low density lipoprotein
- left ventricle
- mean transit time
- nitric oxide
- relative electrophoretic mobility
- superoxide dismutase
- Received June 1, 2000.
- Revision received September 5, 2000.
- Accepted October 12, 2000.
- American College of Cardiology
- Heistad D.D,
- Lopez J.A,
- Baumbach G.L
- Wu C.C,
- Chang S.W,
- Chen M.S,
- Lee Y.T
- Piatti P,
- Fragasso G,
- Monti L.D,
- et al.
- Davi G,
- Alessandrini P,
- Mezzetti A,
- et al.
- Dauber I.M,
- Lesnefsky E.J,
- VanBenthuysen K.M,
- Weil J.V,
- Horwitz L.D
- Bell M.R,
- Lerman L.O,
- Rumberger J.A
- ↵Mohlemkamp S, Behrenbeck TR, Lerman LO, et al. Quantitation of long-term changes in coronary microvascular functional reserve using electron beam tomography—preliminary results in a porcine model. Radiology 2000. In Press.
- Canty J.M Jr.,
- Judd R.M,
- Brody A.S,
- Klocke F.J
- Reckelhoff J.F,
- Kanji V,
- Racusen L.C,
- et al.
- Gilligan D.M,
- Sack M.N,
- Guetta V,
- et al.
- Rumberger J,
- Bell M
- Liu Y.H,
- Bahn R.C,
- Ritman E.L
- Thompson H.K,
- Starmer F,
- Whalen R.E,
- McIntosh H.D
- Lerman L.O,
- Ritman E.L,
- Pelaez L.I,
- Sheedy P.F.I,
- Krier J.D
- Napoli C,
- Mancini F.P,
- Corso G,
- et al.
- Lowry O.H,
- Roseburgh H.J,
- Farr A.L,
- Randall R.J
- Napoli C,
- Witztum J.L,
- de Nigris F,
- et al.
- Delanty N,
- Reilly M.P,
- Pratico D,
- et al.
- Hasdai D,
- Gibbons R.J,
- Holmes D.R Jr.,
- Higano S.T,
- Lerman A
- Zeiher A.M,
- Krause T,
- Schachinger V,
- Minners J,
- Moser E
- Al Suwaidi J,
- Hamasaki S,
- Higano S.T,
- et al.
- Ignarro L.J,
- Cirino G,
- Napoli C
- Napoli C,
- D’Armiento F.P,
- Mancini F.P,
- et al.
- Jialal I,
- Fuller C.J,
- Huet B.A