EXPERIMENTAL STUDY
Multiple endothelial injury in epicardial coronary artery induces downstream microvascular spasm as well as remodeling partly via thromboxane A2
Shu-ichi Saitoha,
Futoshi Onogia,
Kazuhiko Aikawaa,
Mitsuru Mutoa,
Tomiyoshi Saitoa,
Kazuhira Maeharaa and
Yukio Maruyamaa
a Fukushima Medical University, Fukushima, Japan
Manuscript received May 2, 2000;
revised manuscript received August 15, 2000,
accepted September 28, 2000.
Reprint requests and correspondence: Dr. Yukio Maruyama, First Department of Internal Medicine, Fukushima Medical University, 1-Hikari-ga-oka, Fukushima, 960-1295, Japan
maruyama{at}fmu.ac.jp
 |
Abstract
|
|---|
OBJECTIVES
The study was undertaken to develop a coronary microvascular spasm model in pigs by repeated epicardial coronary artery endothelial injury.
BACKGROUND
The pathophysiologic mechanisms responsible for coronary microvascular spasm remain unclear, in large part because a suitable animal model has yet to be found.
METHODS
Balloon endothelial denudation was done just distal to the site of an implanted Doppler flowmeter in the left anterior descending coronary artery (LAD) every two weeks for a total of four times. Changes in LAD blood flow by intracoronary administration of vasoactive agents were assessed before each denudation.
RESULTS
In the epicardial LAD endothelial denudation pigs, decreases in LAD blood flow caused by acetylcholine were augmented. Before denudation, it was –15 ± 4%, and at week 8 (i.e., two weeks after the fourth denudation) it was –100% (i.e., zero flow [p < 0.01]). The LAD flow changes in response to 5-hydroxytryptamine (5-HT) changed from an increase to a decrease, accompanied by medial thickening of microvessels in the LAD perfusion area. These flow responses were observed without significant changes in LAD diameter. In contrast, the LAD blood flow responses to acetylcholine and 5-HT did not change throughout the experiment in pigs given aspirin and a thromboxane A2 (TXA2) synthase inhibitor orally.
CONCLUSIONS
This microvascular spasm model indicates that hypersensitivity to vasoactive substances in the microvascular beds as well as microvascular remodeling are brought about partly through TXA2. This model should be useful for examining the pathophysiology and treatment of microvascular angina.
|
Abbreviations and Acronyms
| | ADP | = adenosine diphosphate | | ECG | = electrocardiogram | | ED group | = repeated LAD endothelial denudation group | | 5-HT | = 5-hydroxytryptamine | | ISDN | = isosorbide dinitrate | | LAD | = left anterior descending coronary artery | | LCA | = left coronary artery | | L-NAME | = N -nitro-L-arginine methyl ester | | PRP | = platelet-rich plasma | | TXA2 | = thromboxane A2 | | TXB2 | = thromboxane B2 |
|
Despite many studies, there is still no consensus on the pathophysiology of coronary microvascular angina or spasm (1–4). Therefore, not only the pathogenesis of coronary microvascular angina but also its prevention and treatment remain important clinical problems. One reason for the lack of progress is that a microvascular angina model is not available.
It has been reported that hypertension (5), hypercholesterolemia (6) and diabetes mellitus (7) induce epicardial coronary artery endothelial damage, which is likely to occur repeatedly over the course of a lifetime. To mimic this process, we performed repeated epicardial coronary artery endothelial injury every two weeks, leading to the remodeling of the injured epicardial vessel as well as changes in its vascular reactivity (8). In this procedure, platelet activation may occur occasionally at the damaged endothelial site (9), which may produce potent vasoconstrictive substances such as thromboxane A2 (TXA2) and 5-hydroxytryptamine (5-HT) (10). It is possible that these substances might induce downstream microvascular spasm and remodeling. In the present study, we tried to develop a new coronary microvascular spasm model in the pig by producing endothelial injury repeatedly in the upstream epicardial coronary artery.
|
Methods
|
|---|
After general anesthesia with sodium thiopental (2 mg/kg, IV) and 1% halothane, under sterile conditions and with artificial ventilation, pigs (22 ± 0.8 kg, n = 36) underwent a left thoracotomy at the fourth intercostal space. We implanted an ultrasonic transit-time flowmeter (Transonic T201D, Transonic System, New York) around the epicardial left anterior descending coronary artery (LAD). The chest was closed in layers, and analgesia was provided after surgery with 0.2 mg IM bupurenorphine (Otsuka, Osaka, Japan). Prophylactic 1 g IM flomoxef sodium (Shionogi, Osaka, Japan) and 600 mg IM clindamycin phosphate (Nihon-Upjhon, Tokyo, Japan) were administered for five days after surgery.
Experiments were initiated two weeks after surgery. At first, we made control recordings of the LAD diameter using a coronary angiogram, the LAD blood flow with a flowmeter, electrocardiograms (ECGs) from limb (III, aVF) and precordial lead (V5), and aortic blood pressure via the 8F sheath introducer (Termo, Tokyo, Japan) inserted from the carotid artery using a pressure transducer (AP641G, Nihon Koden). We calculated the mean aortic pressure-rate product (mm Hg/min) and LAD vascular resistance (mean aortic pressure/mean LAD blood flow; mm Hg/ml/min). The ECG ST-segment shifts were determined at a point 80 ms after the QRS onset. Next, these measurements were repeated after administration of vasoactive substances into the left coronary artery (LCA) at doses similar to those reported previously (11,12) [acetylcholine 0.05 µg/kg (Ovisot, Daiichi, Tokyo, Japan), 5-HT 1 µg/kg (Wako Pure Chemical, Tokyo, Japan), isosorbide dinitrate (ISDN) 10 µg/kg (Eisai, Tokyo, Japan), and adenosine 10 µg/kg (Sigma Chemical, St. Louis, Missouri)].
We also tested different doses of acetylcholine (0.01 and 0.5 µg/kg) and 5-HT (0.1 and 10 µg/kg). When larger doses of these drugs were administered, the epicardial conduit artery showed coronary vasoconstriction [% diameter reduction after 0.5 µg/kg acetylcholine (n = 5) from 1 ± 1% (week 0) to 12 ± 5% with severe delay of contrast medium (week 8, p < 0.05); after 10 µg/kg 5-HT (n = 5) from 7 ± 4% (week 0) to 92 ± 8% (week 8, p < 0.01)], and a greater flow reduction occurred earlier [% flow reduction after 0.5 µg/kg acetylcholine (n = 5) from –32 ± 5% (week 0) to –100% (week 4, p < 0.01); after 10 µg/kg 5-HT (n = 5) from –3 ± 5% (week 0) to –72 ± 8% (week 8, p < 0.01)].
In contrast, smaller doses produced minor effects on the epicardial coronary artery [% diameter reduction after 0.01 µg/kg acetylcholine (n = 5) from 0 ± 1% (week 0) to 2 ± 2% (week 8); after 0.1 µg/kg 5-HT (n = 5) from 0 ± 2% (week 0) to 3 ± 3% (week 8)], as well as on LAD blood flow [% flow changes after 0.01 µg/kg acetylcholine (n = 5) from 14 ± 3% (week 0) to –68 ± 5% (week 8, p < 0.01); after 0.1 µg/kg 5-HT (n = 5) from 6 ± 3% (week 0) to –10 ± 4% (week 8)]. An acetylcholine dose of 0.05 µg/kg and a 5-HT dose of 1 µg/kg were defined as suitable for observing vascular responses in the microvascular beds throughout the time of the experiment. All drugs were diluted with a 0.9% NaCl solution to a volume of 1 ml, and the same amount of 0.9% NaCl was used to flush residual drug from the handmade Judkins-type catheter, which was inserted from the sheath introducer. As for the change in coronary artery diameter after treatment with vasoactive substances, each drug was again infused into the LCA after observing the preinjection hemodynamic state, and then coronary angiography was performed at the time when initial changes in coronary blood flow in response to the vasoactive substances were maximum.
We also investigated whether an abnormality in nitric oxide release occurs with repeated epicardial LAD endothelial denudation. After 10 mg/kg N-nitro-L-arginine methyl ester [L-NAME; Sigma Chemical] in 100 ml of 0.9% NaCl was administered intravenously over a period of 30 min, the same dose of acetylcholine and 5-HT was administered into the LCA and the variables described above were measured.
Next, balloon endothelial denudation was done just distal to the site of the flowmeter in the LAD using a 2F Fogarty catheter, as previously described (8). Briefly, the catheter balloon was inflated with 0.05 ml of contrast medium and 0.05 ml of 0.9% NaCl. After balloon inflation, three denudation procedures were performed over a length of 4 cm in the same portion of the artery. Two weeks after denudation, we performed the same protocol as for the examination of coronary and systemic hemodynamics, and then the denudation protocol was done again. The hemodynamic measurements and denudation procedure were repeated every two weeks from week 0 until week 6, but the LAD denudation was not performed in week 8. Thus, a total of four denudations (weeks 0, 2, 4, and 6) and five hemodynamic measurements (week 0, 2, 4, 6 and 8) were done throughout the experimental period.
Coronary flow response to each vasoactive substance was expressed as % change of the initial peak flow response from the preinjection baseline value, which was nearly constant throughout the experiment, since later on coronary flow followed by severe flow reduction was affected by reactive hyperemia.
Before the first epicardial LAD endothelial denudation and at week 8, venous blood samples were taken from the coronary sinus and the peripheral artery for measuring platelet aggregation rate and plasma concentrations of thromboxane B2 (TXB2) and 6-keto-PGF1 . Platelet-rich plasma (PRP) was prepared, platelet aggregation was induced with adenosine diphosphate (ADP), and collagen was measured using the modified method of Born and Cross (13). TXB2 and 6-keto-PGF1 concentrations in plasma were measured with the RIA-PEG method.
At the end of the experiment, pigs were euthanized with a lethal dose of sodium pentbarbital. For fixation of maximum vasodilated coronary arteries, the coronary arteries were flushed with 10 ml saline containing 10 µg/kg nitroglycerin and 10 µg/kg adenosine after the hearts were excised. The hearts were then perfused with 10% formaldehyde at a pressure of 100 mm Hg and fixed in 10% formaldehyde for a few days. Thereafter, the hearts were cut transversely from base to apex serially at 1-cm intervals. The tissue was embedded in paraffin, sectioned (5 µm thick), and stained with hematoxylin-eosin and elastica van Gieson stains. Pictures of intramyocardial microvessels (10 to 200 µm) in the LAD perfusion areas and of the LAD endothelial denuded portions were taken at 40x to 200x magnification, and the parameters described later were measured with a photomicroscope analysis system. In each heart, the lumen diameter from 10 µm to 200 µm and the corresponding external elastic membrane cross-sectional area to lumen cross-sectional area, which is the index of medial thickening in various sizes of microvessels, were measured from 6 to 30 microvessels. These values were then averaged in every 10-µm range of lumen diameters and were plotted (14,15). This analysis enabled us to evaluate the vascular remodeling in different sizes of microvessels. For the LAD endothelial denuded portion, the ratio of intima area (mm2)-to-media area (mm2), indicating intimal thickening, was also calculated to clarify the remodeling in the repeatedly denuded portion. Histopathology and morphometry were performed by investigators who were blind to the treatment performed.
There were three groups of animals in the experiment: a control group with flowmeter implantation without LAD endothelial denudation (Control group, n = 12); a group with repeated LAD endothelial denudation every two weeks (ED group, n = 12); and a group with repeated LAD denudations every two weeks, which was given 1 g/day TXA2 synthetase inhibitor (Ozagrel hydrochloride, Ono Ltd., Osaka, Japan) (16) and 100 mg/day aspirin (Sigma Chemical) orally immediately after the first denudation (Drug group, n = 12).
Data are expressed as means ± SEM. Serial changes of hemodynamic variables, coronary vascular responses and hemostatical variables prior to and following administration of drugs were carried out using the two-way analysis of variance (ANOVA) followed by the Scheffé post hoc test. The Student t-test was used in comparing paired or unpaired data. Differences in intimal thickening in LAD and medial thickening corresponding to the difference in lumen diameters in microvessels were evaluated by one-way ANOVA followed by the Fisher exact test for multiple comparisons. The Spearman correlation was used to investigate the relation between the vessel diameter and medial thickening in microvessels. A level of p < 0.05 was considered significant.
All experiments using animals were performed according to the guidelines of Fukushima Medical University and the Japanese Government Notification on Feeding and Safekeeping of Animals (No. 6).
|
Results
|
|---|
Effects of repeated endothelial denudation of the epicardial LAD on histopathological changes in downstream microvessels.
In the epicardial LAD denuded portion, intimal thickening was observed in the ED group compared to the Drug and Control groups. The intima to media ratio at eight weeks in the ED group was 0.30 ± 0.06 (sample n = 8, p < 0.01 vs. Drug and Control groups); in the Drug group, 0.08 ± 0.04 (sample n = 8, p < 0.05 vs. Control group); and in the Control group, 0.04 ± 0.02 (sample n = 8) (Fig. 1).
View larger version (121K):
[in this window]
[in a new window]
|
Figure 1 Photomicrographs of week 8 coronary arteries stained with hematoxylin-eosin. Large epicardial coronary arteries (top) and downstream microvascular arterioles (middle; 50 µm  lumen diameter <150 µm; bottom; lumen diameter <50 µm). Left panel is from the Control group, center panel from the ED group and right panel from the Drug group. In the ED group, thickening of the media and increased perivascular fibrosis are shown. Epicardial coronary arteries in the center and right upper panels show the denuded portions. Scale bars indicate 100 µm.
|
|
The mean lumen diameters of microvessels examined were 100.1 ± 5.16 µm (range 10 to 185 µm, averaged plotted sample n ± 59, measured sample n = 173), 98.1 ± 4.56 µm (range 15 to 200 µm, averaged plotted sample n = 69, measured sample n = 199) and 99.3 ± 5.17 µm (range 18 to 182 µm, averaged plotted sample n = 54, measured sample n = 212) in the Control, ED and Drug groups, respectively. Medial thickening in the downstream microvessels of the LAD perfusion beds in vessels with diameters from 50 µm up to <150 µm was greater in the ED group (2.62 ± 0.11) than in the Control (1.82 ± 0.06, p < 0.05) and Drug groups (2.11 ± 0.43, p < 0.05). However, in microvessels with diameters of <50 µm and  150 µm, medial thickening was not different among the three groups (Figs. 1 and 2).
View larger version (14K):
[in this window]
[in a new window]
|
Figure 2 Scatterplots showing the relation between the averaged values of external elastic membrane (EEM) cross-sectional area (CSA)-to-lumen CSA ratios, and the averaged values of vessel diameter in the Control group (a), the ED group (b) and the Drug group (c). Each point in the plot represents the average of vessels in a 10 µm range of diameters in an individual pig. In the Control and Drug groups, the ratios are relatively constant in coronary arteries of various diameters from 10 to 200 µm, and there was no significant correlation between the averaged values of EEM CSA-to-lumen CSA ratios and the vessel diameters. In the ED group, a significant correlation existed between the EEM CSA-to-lumen CSA ratios and the vessel diameters (y = –0.0001x2 + 0.0230x + 1.6921, r2 = 0.208, p = 0.006).
|
|
Baseline pressure-rate product, coronary vascular resistance, and coronary diameter of the repeated denuded portions.
The baseline mean aortic pressure-rate product and LAD vascular resistance did not change in any of the groups (n = 12 each) throughout the experimental time period [pressure-rate product at week 8 (mm Hg/min): Control group, 8164 ± 425; ED group, 7992 ± 438; Drug group, 7558 ± 486; vascular resistance at week 8 (mm Hg/ml/min): Control group, 3.1 ± 1.2; ED group, 2.8 ± 1.1; Drug group, 2.7 ± 0.9]. The baseline coronary artery diameters before the first denudation were 2.1 ± 0.2, 2.2 ± 0.3 and 2.1 ± 0.3 mm and at week 8 were 2.5 ± 0.3, 2.4 ± 0.4 and 2.5 ± 0.4 mm in the Control, ED and Drug groups, respectively.
Systemic and coronary hemodynamic responses to acetylcholine, 5-HT, ISDN and adenosine.
The mean aortic pressure and heart rate after intracoronary administration of acetylcholine, 5-HT, ISDN and adenosine did not change in any of the groups throughout the experimental period (data not shown).
In week 0, before epicardial LAD endothelial denudation, acetylcholine initially decreased LAD blood flow (Control group, –14 ± 5%; ED group, –15 ± 4%; Drug group, –18 ± 5%), which was followed by a small increase in flow (Control group, 21 ± 6%; ED group, 22 ± 5%; Drug group, 24 ± 4%) without any changes in epicardial coronary artery diameters (Figs. 3a and 4a). Conversely, in week 0 coronary blood flow was increased by 5-HT (Control group, 23 ± 5%; ED group, 25 ± 5%; Drug group, 28 ± 6%) (Fig. 4b), ISDN (Control group, 94 ± 12%; ED group, 92 ± 10%; Drug group, 94 ± 12%) (Fig. 4c) and adenosine (Control group, 202 ± 12%; ED group, 206 ± 10%; Drug group, 204 ± 12%) (Fig. 4d).
View larger version (56K):
[in this window]
[in a new window]
|
Figure 3 Representative tracings and coronary angiograms from a pig in the ED group at (a) week 0 (before endothelial denudation) and (b) week 8 (2 weeks after four denudations). Coronary blood flow at week 8 initially decreased to zero flow, and subsequent reactive hyperemia appeared with ECG ST depression in limb lead III after intracoronary administration of 0.05 µg/kg acetylcholine (ACh) without showing epicardial coronary artery vasoconstriction. BP = aortic blood pressure; CBF = coronary blood flow in LAD; ECG = electrocardiogram.
|
|
The LAD flow responses to each drug did not change in the Control group throughout the experimental period (Fig. 4). Decreased coronary blood flow response to acetylcholine in the denuded portion in the ED group was enhanced by repeated LAD endothelial denudation, and by week 8 acetylcholine reduced coronary blood flow to zero (p < 0.01 vs. Control group) with ECG ST-segment depression (limb lead III, 0.24 ± 0.08 mV; aVF, 0.22 ± 0.06 mV; precordial lead V5, 0.26 ± 0.08 mV, p < 0.01 vs. each preinjection state) (Figs. 3b and 4a). When LAD blood flow was reduced to zero, a severe delay in contrast medium washout without epicardial spasm in the repeated denuded portion was observed in angiography (% diameter reduction, 8 ± 3%) (Fig. 3b). The duration of zero coronary flow was 46 ± 2.4 (38 to 57) s with the dose of acetylcholine used in this study. This cessation of flow was followed by reactive hyperemia (% maximal increase of flow, 214 ± 28%). Epicardial spasm by acetylcholine in the denuded portion did not appear throughout the experiment (% diameter reduction, 1 ± 2%, 3 ± 3%, 4 ± 3% and 5 ± 3% in weeks 0, 2, 4 and 6, respectively).
Repeated endothelial denudation also changed the effect of 5-HT on blood flow from an increase (+25 ± 5%) to a decrease with time (week 8, –20 ± 4%, p < 0.05 vs. Control group) (Fig. 4b); however, the diameter reduction in the repeatedly denuded portion induced by 5-HT was negligibly small in week 8 (7 ± 2%) in the ED group. In contrast, increases in coronary blood flow in response to ISDN (week 8, 96 ± 8%) and adenosine (week 8, 216 ± 14%) also did not change in the ED group (Fig. 4c, d).
Decreases in coronary blood flow in response to acetylcholine (week 8, ED group [–100%] vs. Control group [–21 ± 5%], p < 0.01) and 5-HT (week 8, ED group [–20 ± 4%] vs. Control group [20 ± 6%], p < 0.05) in the ED group were prevented in the Drug group (Fig. 4a, b), whereas coronary flow responses to ISDN and adenosine did not differ from those of the other two groups (Fig. 4c, d).
Effects of L-NAME on acetylcholine and 5-HT responses.
In all of the groups (n = 12 each), IV administration of 10 mg/kg L-NAME tended to increase aortic pressure and decrease heart rate, but not significantly (data not shown).
Pretreatment with IV L-NAME did not alter the LAD flow response to acetylcholine in the ED group (week 0, –25 ± 7% vs. –15 ± 4% without L-NAME [NS]; week 6, –100% vs. –91 ± 5% without L-NAME [NS]; week 8, –100% vs. –100% without L-NAME [NS]). From week 2 to week 4, there also was no significant difference, although coronary flow reduction tended to be greater in the ED group receiving L-NAME.
Additionally, L-NAME always decreased the LAD blood flow response to 5-HT in the ED group (week 0, 40 ± 6% vs. 25 ± 5% without L-NAME [p < 0.01]; week 8, –78 ± 6% vs. –20 ± 4% without L-NAME [p < 0.01]). However, as mentioned above, the coronary flow-reducing effect of L-NAME on 5-HT responses did not change throughout the experimental period (week 0,  65 ± 7% [from 25 ± 5% to –40 ± 6%]; week 8, 58 ± 8% [from –20 ± 4% to –78 ± 6%]; NS).
The effects of L-NAME on acetylcholine and 5-HT-induced LAD flow responses in the Control group and the Drug group were similar to those of the ED group, and consequently there was no significant effect of L-NAME on the coronary flow responses to acetylcholine or 5-HT in any of the three experimental groups. The differences between groups in response to acetylcholine and 5-HT were preserved, regardless of treatment with L-NAME.
Hemostasis analysis.
In the Control group (n = 12), plasma concentrations of 6-keto-PGF1, and TXB2 in coronary sinus blood did not change throughout the experimental period (6-keto-PGF1, 66.8 ± 5.6 pg/ml [week 0] to 72.2 ± 6.8 pg/ml [week 8]; TXB2, 118 ± 13.4 pg/ml [week 0] to 142 ± 10.6 pg/ml [week 8]). However, the plasma concentration of 6-keto-PGF1 in coronary sinus blood decreased in the repeated LAD endothelial denudation group irrespective of drug administration (ED group [n = 12]: from 64.8 ± 4.9 pg/ml [week 0] to 40.4 ± 5.2 pg/ml [week 8], p < 0.05 vs. Control group; Drug group [n = 12]: from 69.5 ± 5.2 pg/ml to 42.6 ± 5.8 pg/ml, p < 0.05 vs. Control group). Moreover, the plasma concentration of TXB2 in coronary sinus blood increased in the ED group (from 108 ± 12.8 pg/ml [week 0] to 309 ± 23.4 pg/ml [week 8], p < 0.01 vs. Control group) but decreased in the Drug group (from 115 ± 11.2 pg/ml [week 0] to 22 ± 4.2 pg/ml [week 8], p < 0.01 vs. Control and ED groups).
Adenosine diphosphate and collagen-induced PRP aggregation rates in the Control group (n = 12) did not change in coronary sinus blood or in peripheral arterial blood (coronary sinus blood; ADP, 36 ± 5% to 42 ± 5%; collagen, 28 ± 6% to 32 ± 6%; peripheral arterial blood; ADP; 35 ± 4% to 40 ± 6%; collagen, 26 ± 4% to 30 ± 5%, week 0 to 8, respectively). In the ED group (n = 12), ADP and collagen-induced PRP aggregation rates were increased in coronary sinus blood (ADP, from 36 ± 4% [week 0] to 88 ± 6% [week 8], p < 0.01 vs. Control group; collagen, from 25 ± 5% [week 0] to 72 ± 8% [week 8], p < 0.05 vs. Control group]; however, they did not change in peripheral arterial blood (ADP, 35 ± 5% [week 0] to 48 ± 7% [week 8]; collagen, 26 ± 6% [week 0] to 39 ± 5% [week 8]).
In the Drug group (n = 12), ADP and collagen-induced PRP aggregation rates were decreased in coronary sinus blood (ADP, from 38 ± 6% [week 0] to 6 ± 2% [week 8]; collagen, from 29 ± 5% [week 0] to 4 ± 2% [week 8]; p < 0.05 vs. Control and p < 0.01 vs. ED group) and in peripheral arterial blood (ADP, from 36 ± 6% [week 0] to 4 ± 3% [week 8]; collagen; from 27 ± 6% [week 0] to 3 ± 3% [week 8], p < 0.05 vs. Control and ED groups).
|
Discussion
|
|---|
Epicardial coronary artery endothelial injury and microvascular spasm with remodeling.
Recently, according to Wiedermann et al. (17), small-vessel disease is frequently observed by intravascular ultrasound in patients showing abnormal epicardial vessels (plaque or intimal thickening) who have entirely normal coronary angiograms. However, a causal relationship between epicardial coronary artery and downstream microvascular abnormalities has not been established. Moreover, despite clinical observation (9) it remains unclear whether small but additional narrowing of the epicardial stenosed coronary artery plays a role in microvascular spasm with ECG ST change and lactate production in coronary sinus blood in the human, and it also remains unclear how coronary flow actually decreases in a microvessel anginal attack.
In the present study we showed that, in a pig model of coronary microvascular spasm showing transient coronary zero flow by administration of acetylcholine, the abnormal microvascular flow response is accompanied by microvascular remodeling similar to that in the human (2,4,18). As far as we know, this is the first demonstration that coronary flow in a downstream perfusion bed perfused by a repeatedly denuded epicardial coronary artery stops briefly after acetylcholine injection and is accompanied by ischemic ECG ST changes but not epicardial coronary spasm. Coronary microvascular remodeling in this model, which was induced by upstream repeated coronary artery endothelial injury, occurred especially in smaller microvessels (50 to 150 µm). The reason for this is unclear, but it may have been related to altered expression of platelet-derived growth factors in the coronary microcirculation.
Therefore, to better understand the mechanisms of the microvascular remodeling process or the different responses (depending on the diameter of small vessels) to repeated epicardial coronary artery endothelial denudation, the transmyocardial production of platelet-derived growth factors should be measured. Another important issue to be clarified is where the spastic site resides in microvessels in the zero flow state. In particular, no information is presently available as to how microvessel morphological changes relate to functional flow alterations in response to drugs.
Thromboxane A2 and coronary microvascular remodeling.
Epicardial coronary artery endothelial injury is likely to occur repeatedly during a lifetime as a result of local shear forces and/or various risk factors. An injured surface of endothelium activates platelets (9). Activated platelets release several vasoconstrictive substances, such as 5-HT, TXA2, ADP and thrombin, that are also major contributors to vascular smooth muscle cell migration (10,19). In addition, endothelial injury mimics inflammatory changes, in which infiltration of neutrophils and macrophages into the vessel is likely to occur and in which TXA2 is considered to be released from endothelial cells through the action of Cox-2 (20). Thromboxane A2 markedly enhances thrombin-induced proliferation of vascular smooth muscle cells (21).
Moreover, according to our results, TXA2 also seems to be an important factor in microvascular remodeling because a TXA2 synthetase inhibitor and aspirin prevented microvascular remodeling. However, we did not examine the effect of TXA2 synthetase inhibitor alone. Additionally, it is plausible that a decrease in the production of a prostacyclin (6-keto-PGF1) associated with repeated epicardial coronary artery denudation might be a cofactor for promoting abnormal structural and functional changes in the microvasculature. Thus, further study is needed to clarify how the combined effects of TXA2 synthetase inhibitor and aspirin differ from treatment with each drug alone on the microvascular remodeling process and/or spasm induction.
Nitric oxide and microvascular spasm.
Although pretreatment with IV L-NAME affected baseline coronary flow rate in each group, it did not affect the LAD flow response to acetylcholine and 5-HT throughout the experimental period. Therefore, it seems unlikely that an abnormality in nitric oxide release from the microvascular endothelium per se contributed to the prominent flow reduction following each drug administration at the later stages of the repeated denuding procedures.
Study limitations.
First, platelet deposition at the site of endothelial injury was not consecutively determined after endothelial injury; however, in our model, regenerated endothelial cells covered the injured surface two weeks after injury without the formation of large thrombi (8). According to Yan et al. (9), de-endothelialization resulted in the immediate adhesion of platelets to the injured vessel wall; but large thrombi were not formed, and the platelets disappeared from the injured surface within a few days.
Second, although our animal microvessel disease model is similar to morphological changes in microvessel or the vasoconstrictive responses including ECG ST changes to various cardiovascular drugs in humans (22), spontaneous microvascular spasm attacks were not observed. Third, we did not quantitatively estimate regional wall-motion abnormalities during attacks of microvascular spasm. It has been reported that long-lasting microvascular abnormalities, including spasm, may lead to myocardial ischemia in human diabetes mellitus and in an animal cardiomyopathy model (23,24). However, there are little data showing wall-motion abnormality during an attack of coronary microvascular spasm, although an ischemic cause of syndrome X is a hypothesis that has been abandoned by some of its initial proponents (25). With our model in only a few cases examined during the coronary flow cessation state with echocardiography, it was evident that the altered microvascular flow responses resulted in significant regional myocardial dysfunction of even a transient nature. However, the data are sparse and premature and further study is needed to clarify this issue.
|
Footnotes
|
|---|
This work was supported by the Fukushima Society for the Promotion of Medicine.
|
References
|
|---|
- Cannon RO, Epstein SE. "Microvascular angina" as a cause of chest pain with angiographically normal coronary artery. Am J Cardiol. 1988;61:1339–1343
- Mosseri M, Yarom R, Gostmann MS, Hasin Y. Histologic evidence for small-vessel coronary artery disease in patients with angina pectoris and patent large coronary arteries. Circulation. 1986;74:964–972[Abstract/Free Full Text]
- Opherk D. Reduced coronary dilatory capacity and ultrastructural changes of the myocardium in patients with angina pectoris but normal coronary arteriograms. Circulation. 1981;63:817–825[Abstract/Free Full Text]
- Mori M, Koyanagi M, Egashira K, et al. Angina pectoris caused by microvascular spasm. Lancet. 1998;351:1165–1169[CrossRef][Medline]
- Toda N, Okamura T. Endothelium-dependent and -independent response to vasoactive substances of isolated human coronary arteries. Am J Physiol. 1989;257:H988–H995
- Treasure CB, Klein JL, Weintraub WS. Beneficial effects of cholesterol-lowering therapy on the coronary endothelium in patients with coronary artery disease. N Engl J Med. 1985;332:481–488
- Nitenberg A, Valensi P, Sachs R, Dali M, Aptecar E, Attali JR. Impairment of coronary vascular reserve and Ach-induced coronary vasodilation in diabetic patients with angiographically normal coronary arteries and normal left ventricular systolic function. Diabetes. 1993;42:1017–1025[Abstract]
- Saitoh S, Saito T, Maruyama Y, et al. Morphological and functional changes in coronary vessel evoked by repeated endothelial injury in pigs. Cardiovasc Res. 1998;38:772–781[Abstract/Free Full Text]
- Yan Z, Yokota T, Zhang W, Hansson GK. Expression of inducible nitric oxide synthase inhibits platelet adhesion and restores blood flow in injured artery. Circulation. 1996;79:38–44
- Hourani SMO, Cusack NJ. Pharmacological receptors on blood platelets. Am Soc Pharmacol Exper Ther. 1991;43:243–298
- Hata H, Egashira K, Fukai T, et al. The role of endothelium-derived nitric oxide in acetylcholine-induced coronary vasoconstriction in closed-chest pigs. Coron Artery Dis. 1993;4:891–898[Medline]
- Kadokami T, Egashira K, Kuwata K, et al. Altered serotonin receptor subtypes mediated coronary microvascular hyperreactivity in pigs with chronic inhibition of nitric oxide synthesis. Circulation. 1996;94:182–189[Abstract/Free Full Text]
- Born GVR, Cross MJ. The aggregation of blood platelets. J Physiol. 1963;168:178–195[Free Full Text]
- Heagerty AM, Aalkjaer C, Bund SJ, Korsgaard N, Mulvany MJ. Small artery structure in hypertension. Hypertension. 1993;21:391–397[Free Full Text]
- Numaguchi K, Egashira K, Takemoto M, et al. Chronic inhibition of nitric oxide synthesis causes coronary microvascular remodeling in rats. Hypertension. 1995;26:957–962[Abstract/Free Full Text]
- Taniguchi K, Wakitani K, Omawari N, et al. Pharmacological studies on the TXA2 synthetase inhibitor (E)-3-[p-(1H-Imidazol-1-Yl methyl) Phenyl]-2-Propenoic Acid (OKY-046). Jpn J Pharmacol. 1986;41:393–401[Medline]
- Wiedermann JG, Schwartz A, Apfelbaum M. Anatomic and physiologic heterogeneity in patients with syndrome X: an intravascular ultrasound study. J Am Coll Cardiol. 1995;25:1310–1317[Abstract]
- Suzuki H, Takeyama Y, Koba S, Suwa Y, Katagiri T. Small vessel pathology and coronary hemodynamics in patients with microvascular angina. Int J Cardiol. 1994;43:139–150[CrossRef][Medline]
- Tamura K, Kanazaki T, Saito Y, Otabe M, Morisaki N. Serotonin enhances migration of rat aortic smooth muscle cells through 5-HT2 receptors. Atherosclerosis. 1997;132:139–143[CrossRef][Medline]
- Bustos M, Coffman TM, Saadi S. Modulation of eicosanoid metabolism in endothelial cells in a xenograft model. J Clin Invest. 1997;100:1150–1158[Medline]
- Zucker T, Bonisch D, Muck S, et al. Thrombin-induced mitogenesis in coronary artery smooth muscle cells is potentiated by thromboxane A2 and involves upregulation of thromboxane receptor mRNA. Circulation. 1998;97:589–595[Abstract/Free Full Text]
- Serota H, Kerm MJ, Deligonul U, Aguirre F, Caralis DG. Ergonovine-induced myocardial ischemia without epicardial coronary vasospasm: evidence for ischemia produced by small-vessel vasoconstriction. Am Heart J. 1991;121:1807–1809[CrossRef][Medline]
- Philip JN, Ronald EB, Heigi O, Michael DW, James DR. Maximal coronary flow reserve and metabolic coronary vasodilation in patients with diabetes mellitus. Circulation. 1995;91:635–640[Abstract/Free Full Text]
- Stephen MF, Minase T, Sangho C, Renee D. Microvascular spasm in the cardiomyopathic syrian hamster. Circulation. 1982;66:342–354[Abstract/Free Full Text]
- Kaski JC. Commentary. Chest pain and normal coronary arteriograms: role of "microvascular spasm". Lancet. 1998;351:1144–1145[CrossRef][Medline]
This article has been cited by other articles:
|
|
|
|
 
M. Feletou and P. M. Vanhoutte
Endothelial dysfunction: a multifaceted disorder (The Wiggers Award Lecture)
Am J Physiol Heart Circ Physiol,
September 1, 2006;
291(3):
H985 - H1002.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
O. Sorop, E. N. T. P. Bakker, A. Pistea, J. A. E. Spaan, and E. VanBavel
Calcium channel blockade prevents pressure-dependent inward remodeling in isolated subendocardial resistance vessels
Am J Physiol Heart Circ Physiol,
September 1, 2006;
291(3):
H1236 - H1245.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
E. N.T.P. Bakker, C. L. Buus, J. A.E. Spaan, J. Perree, A. Ganga, T. M. Rolf, O. Sorop, L. H. Bramsen, M. J. Mulvany, and E. VanBavel
Small Artery Remodeling Depends on Tissue-Type Transglutaminase
Circ. Res.,
January 7, 2005;
96(1):
119 - 126.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
H. Kunichika, O. Ben-Yehuda, S. Lafitte, N. Kunichika, B. Peters, and A. N. DeMaria
Effects of glycoprotein iib/iiia inhibition on microvascular flow after coronary reperfusion: A quantitative myocardial contrast echocardiography study
J. Am. Coll. Cardiol.,
January 21, 2004;
43(2):
276 - 283.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
J. A. Barrabes, D. Garcia-Dorado, M. Mirabet, R.-M. Lidon, B. Soriano, M. Ruiz-Meana, P. Pizcueta, J. Blanco, Y. Puigfel, and J. Soler-Soler
Lack of effect of glycoprotein IIb/IIIa blockade on myocardial platelet or polymorphonuclear leukocyte accumulation and on infarct size after transient coronary occlusion in pigs
J. Am. Coll. Cardiol.,
January 2, 2002;
39(1):
157 - 165.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
Top
Abstract
Methods
p < 0.01 vs. 0 week at the same portion in the same condition. 
p < 0.01 vs. two weeks at the same portion in the same condition. @p < 0.01 vs. the Control group in the same experimental period. ¶p < 0.01 vs. the Drug group in the same experimental period.