Impaired Endothelium-Dependent Vasodilation of Large Epicardial and Resistance Coronary Arteries in Patients With Essential Hypertension

Patients

The study was done in seven patients with essential arterial hypertension and seven normotensive control subjects undergoing diagnostic cardiac catheterization for evaluation of chest pain. All patients and subjects had atypical chest pain, normal exercise stress test results, angiographically normal coronary arteries, and normal left ventricular function. To assess the isolated effects of hypertension on coronary endothelial function, we excluded patients with diabetes mellitus, hypercholesterolemia (total cholesterol level >6.2 mmol/L [240 mg/dL] at the time of study), a history of myocardial infarction, variant angina, valvular heart diseases, or heart failure. Individuals with evidence of left ventricular hypertrophy as assessed by electrocardiogram and echocardiogram were also excluded. A history of hypertension was defined as a blood pressure elevation greater than 160/95 mm Hg requiring the use of antihypertensive drugs. No patients or subjects were on cholesterol-lowering drugs before study. Hypertensive patients were being treated with various antihypertensive drugs.

The research protocol of this study was approved by the institutional committee for clinical research. Written informed consent was obtained from each participant.

Quantitative Coronary Angiography

Coronary cineangiograms were recorded with a cineangiographic system (Siemens Inc). An appropriate view that allowed the best visualization of the left anterior descending coronary artery (the study artery) was selected.

We determined changes in the luminal diameter of the left anterior descending coronary artery, a segment 2 to 3 mm distal to the tip of the Doppler catheter, as described previously.^{15 16 17} The end-diastolic frame was selected on a cine projector. The images of interest were taken into a videodensitometric analyzer (Kontron Instruments) and digitized. The diameter of the segment of interest was measured for later analysis. A Judkins catheter was used to calibrate the arterial diameter in millimeters. Arterial diameter measurements were done blindly without knowledge of the clinical characteristics of the patients or subjects.

Measurement of Coronary Blood Flow Velocity and Estimation of Coronary Blood Flow

An 8F angioplasty-guiding catheter was introduced into the left main coronary artery through a femoral approach. A 3F Doppler flow-velocity catheter (model DC-201, Millar Instruments) was introduced into the left anterior descending coronary artery. The Doppler catheter then was connected to a DC-101 Velocimeter (Millar Instruments) to obtain mean and phasic velocity signals. Increases in coronary blood flow were estimated from the product of the mean coronary blood flow velocity and the cross-sectional area of the arterial segment at the tip of the Doppler catheter and were expressed as percent increases from baseline as described previously.^{15 16 17}

Study Protocol

Cardiac catheterization was performed with patients and subjects in the fasting state after premedication with 5 mg oral diazepam. Control subjects were not receiving antianginal or antihypertensive drugs before the study. All hypertensive patients were receiving drugs such as nitrates, calcium channel blockers, β-blockers, and angiotensin-converting enzyme inhibitors. These drugs were discontinued 3 days before the study; during this period, patients were monitored for blood pressure elevation.

After completion of the diagnostic catheterization, the following interventions were performed: (1) a bolus injection of papaverine (10 mg/5 mL) through the guiding catheter; (2) infusion of saline (0.5 to 1.0 mL/min for 2 minutes) through the Doppler catheter; (3) infusions of acetylcholine (0.5 to 1.0 mL/min) at 1, 3, 10, and 30 μg/min (for 2 minutes at each dose) through the Doppler catheter; (4) infusions of substance P at 3, 10, and 30 ng/min through the Doppler catheter; and (5) bolus injection of 2 mg isosorbide dinitrate. After completion of the study with one drug, we waited for at least 5 minutes before beginning infusion of the next drug, by which time the coronary diameter and coronary blood flow velocity returned to baseline values. Coronary arteriography was performed before and 2 minutes after each drug. Mean and phasic coronary blood flow velocities, arterial pressure, heart rate, and standard 12-lead electrocardiograms were monitored continuously and recorded on a multichannel recorder (Nihon Kohden Polygraph System). Steady-state values were used for later analysis.

Statistical Analysis

Data are expressed as mean±SD or SEM. When serial changes in hemodynamic variables to the graded doses of drugs were compared, one-way ANOVA was used. When the responses were compared between study groups, two-way ANOVA for repeated measures followed by Bonferroni’s multiple comparison test was used. A value of P<.05 was considered statistically significant.

Clinical characteristics such as age, sex, arterial pressure, and serum cholesterol levels were compared between hypertensive patients and control subjects by Student’s t tests or χ² tests. The effects of these clinical factors on coronary vasomotion in response to acetylcholine were examined by multiple linear regression analysis.

Clinical Characteristics and Hemodynamic Data

The clinical characteristics of hypertensive patients and control subjects were comparable with regard to age, male sex, total serum cholesterol level, and history of cigarette smoking (Table⇓). Left ventricular diastolic and systolic dimensions and left ventricular posterior wall thickness as determined by echocardiogram did not differ between the two groups.

Systolic, diastolic, and mean arterial pressures at the time of study were significantly higher in hypertensive patients than control subjects. Heart rate and left ventricular end-diastolic pressure were similar between the two groups. The diameter of the proximal segments of the left anterior descending coronary artery at baseline was similar in the two groups. Baseline mean arterial pressure and heart rate before drugs did not significantly change during the study in either hypertensive patients or control subjects (data not shown). Mean arterial pressure and heart rate were not significantly altered by intracoronary administration of either acetylcholine, substance P, papaverine, or nitrate, as described previously.^{15 16 17 24 25}

Responses of the Large Epicardial Coronary Artery

Intracoronary infusion of saline did not change the diameter of the large epicardial coronary artery. In control subjects, acetylcholine at the low dose (3 μg/min) increased (P<.05 by one-way ANOVA and multiple comparison tests) and at the high dose (30 μg/min) decreased (P<.05) arterial diameter (Fig 1⇓). In hypertensive patients, acetylcholine significantly decreased (P<.01 by one-way ANOVA) arterial diameter in a dose-dependent manner. No significant vasodilation of the large epicardial coronary artery was noted in hypertensive patients. The vasoconstrictor responses of the large epicardial coronary artery to acetylcholine in hypertensive patients were significantly greater (P<.01 by two-way ANOVA and multiple comparison tests) than the responses in control subjects. The effects of clinical characteristics (Table⇑) on the vasoconstrictor response to 30 μg/min acetylcholine were assessed by multiple regression analysis; arterial pressure but not the other factors significantly (P<.01) correlated with the response to acetylcholine.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1.

Line graphs show effects of acetylcholine (left) and substance P (right) on vasomotor responses of the large epicardial coronary artery in hypertensive patients (•) and control subjects (○). Percent changes in large epicardial coronary artery diameter in response to drugs are presented. **P<.01 vs control by two-way ANOVA and multiple comparison tests. Values are mean±SEM.

Substance P significantly increased (P<.01 by one-way ANOVA) arterial diameter in hypertensive patients and control subjects. The vasodilator responses of the large epicardial coronary artery to substance P were comparable between the two groups (Fig 1⇑).

The vasodilator responses to papaverine (4±3% and 4±4% in hypertensive patients and control subjects, respectively) and nitrate (15±2% and 11±1%) were comparable between the two groups.

Responses of Coronary Blood Flow

Intracoronary infusion of saline did not change coronary blood flow. A real-time tracing of coronary blood flow velocity during drug infusion in a control subject is presented in Fig 2⇓. Both acetylcholine and substance P increased coronary blood flow in a dose-dependent manner in hypertensive patients and control subjects. The increases in coronary blood flow evoked with acetylcholine were markedly less (P<.01 by two-way ANOVA) in hypertensive patients than control subjects (Fig 2⇓). The percent increases in coronary blood flow with acetylcholine at 1, 3, 10, and 30 μg/min were 13±9%, 37±17%, 67±16%, and 184±76%, respectively, in hypertensive patients and 49±17%, 179±42%, 289±36%, and 372±23%, respectively, in control subjects. Multiple regression analysis showed that there was a significant negative correlation between arterial pressure and the percent increase in coronary blood flow evoked with 30 μg/min acetylcholine.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2.

Real-time tracings show electrocardiogram (ECG), arterial pressure (AoP), and coronary blood flow velocity (CBFV) during infusion of saline, papaverine, and acetylcholine.

The percent increases in coronary blood flow evoked with substance P were also significantly less (P<.01 by two-way ANOVA) in hypertensive patients than control subjects (Fig 3⇓). The percent increases in coronary blood flow evoked with substance P at 3, 10, and 30 ng/min were 10±5%, 20±6%, and 32±10% in hypertensive patients and 11±3%, 70±7%, and 143±28% in control subjects.

• Download figure
• Open in new tab
• Download powerpoint

Figure 3.

Line graphs show effects of acetylcholine (left) and substance P (right) on changes in coronary blood flow in hypertensive patients (•) and control subjects (○). Percent increases in coronary blood flow in response to acetylcholine and substance P are presented. **P<.01 vs control by two-way ANOVA and multiple comparison tests.

The percent increase in coronary blood flow in response to papaverine was similar between hypertensive patients (426±35%) and control subjects (394±34%, P=NS). The coronary blood flow response to nitrate was also similar between the groups (122±14% and 98±13%, hypertensive patients and control subjects, respectively, P=NS).

Endothelial Dysfunction in the Large Epicardial Coronary Artery in Hypertensive Patients

Our results that vasoconstrictor responses of the large epicardial coronary artery to acetylcholine in hypertensive patients were greater than those in control subjects are consistent with previous findings in animals and humans suggesting that hypertension is associated with diminished endothelium-dependent dilation of the large coronary artery.^{4 6 7 8} Previous studies have noted that atherosclerosis and various risk factors augment vasoconstrictor responses of angiographically normal-looking epicardial coronary arteries to acetylcholine.¹³ Our hypertensive patients had angiographically normal coronary arteries and no hypercholesterolemia or diabetes mellitus. Age, sex, and incidence of smoking were comparable between the two study groups. Furthermore, multiple linear regression analysis revealed that the levels of arterial pressure but not other clinical factors were significantly correlated with the responses of the large epicardial coronary artery and coronary blood flow to acetylcholine. Thus, as reported in previous studies,^{4 5 6 7 8} it is likely that augmented vasoconstrictor responses of the large epicardial coronary artery to acetylcholine were related to hypertension. However, we cannot exclude the possibility that the presence of angiographically undetected atherosclerosis was responsible for these findings.

It has been shown that acetylcholine and substance P dilate the large epicardial coronary artery by the release of endothelium-derived nitric oxide.^{26 27 28 29} However, acetylcholine and substance P act on different receptors to evoke the release of nitric oxide: muscarinic and tachykinin receptors, respectively.^{22 23} A new finding of the present study is that endothelium-dependent vasodilation of the large epicardial coronary artery evoked with substance P was preserved in hypertensive patients who had diminished endothelium-dependent vasodilation with acetylcholine. Endothelium-independent vasodilation with papaverine and nitrate was comparable in hypertensive patients and control subjects. These results suggest that diminished endothelium-dependent vasodilation of the large epicardial coronary artery with acetylcholine in hypertensive patients was related to a selective abnormality in the endothelial muscarinic receptor of the large epicardial coronary artery.

It has been shown that the increase in flow produces mechanical shear stress on the endothelial cell surface and thus causes endothelium-dependent vasodilation of large conduit vessels including the human epicardial coronary artery.^{30 31} Therefore, we do not exclude the possibility that diminished vasodilation of the large epicardial coronary artery in response to acetylcholine in hypertensive patients might have resulted from altered flow-induced endothelium-dependent vasodilation and not be related to an abnormality in the endothelial muscarinic receptor, since coronary blood flow could not be controlled in our patients. However, we consider it plausible that diminished endothelium-dependent vasodilation of the large epicardial coronary artery with acetylcholine in hypertensive patients may not totally be explained by attenuated flow-induced vasodilation. This hypothesis is based on the following findings. First, vasodilation of the large epicardial coronary artery with substance P in hypertensive patients was equivalent to that in control subjects, whereas the increases in coronary blood flow in response to substance P were markedly blunted in the hypertensive patients (Figs 1⇑ and 3⇑). Second, 1 μg/min acetylcholine, which did not significantly increase coronary blood flow in either group, produced a significant difference in the response of the large epicardial coronary artery between hypertensive patients and control subjects. Acetylcholine at high doses (10 to 30 μg/min), which markedly increased coronary blood flow, caused progressive vasoconstriction of the large epicardial coronary artery in both groups.

Endothelial Dysfunction in the Resistance Coronary Artery in Hypertensive Patients

It is reasonable to assume that the increases in coronary blood flow evoked with the drugs used in this study reflected vasodilation of the resistance coronary artery, because these drugs increased coronary blood flow without altering arterial pressure and heart rate. Although acetylcholine caused a modest vasoconstriction of the large epicardial coronary artery, we^{15 16 17} and other investigators³² have previously shown that this degree of vasoconstriction of the large epicardial coronary artery does not attenuate the coronary blood flow responses to vasodilator stimuli. Thus, it is unlikely that the acetylcholine-induced vasoconstriction of the large coronary artery might have limited the coronary blood flow responses to acetylcholine.

Previous studies have indicated that endothelium-dependent vasodilation of the resistance coronary artery evoked with acetylcholine is impaired in hypertensive patients,^{5 6} which agrees with the data presented in the present study. However, the coronary blood flow response to another endothelium-dependent vasodilator such as substance P has not been studied in hypertensive patients. We found in the present study that the coronary blood flow responses to both acetylcholine and substance P were attenuated in hypertensive patients, indicating that endothelium-dependent dilation of the resistance coronary artery in response to both drugs was impaired in our hypertensive patients. We also demonstrated that the endothelium-independent responses to papaverine and nitrate were similar between hypertensive patients and control subjects, indicating that altered endothelium-dependent responses in hypertensive patients did not result from a nonspecific abnormality in the response of vascular smooth muscle. These findings indicate that defective endothelium-dependent vasodilation of the resistance coronary artery in hypertensive patients is related to a more generalized abnormality of endothelial cells in response to diverse stimuli but not confined to a local endothelial abnormality at the muscarinic receptor level.

Recently, Panza et al,²¹ using a specific inhibitor of endothelium-derived nitric oxide (N^G-monomethyl-l-arginine), have suggested that impaired endothelium-dependent forearm vasodilation in hypertensive patients is related to a decreased release of nitric oxide. However, we could not determine the degree to which endothelium-derived nitric oxide might be involved in the mechanisms responsible for impaired endothelium-dependent vasodilation of the resistance coronary artery in hypertensive patients because responses of the resistance coronary artery to acetylcholine and substance P were not assessed before and after an inhibitor of endothelium-derived nitric oxide. Therefore, we do not know whether endothelial dysfunction in the resistance coronary artery in hypertensive patients is related to reduced release of endothelium-derived nitric oxide or hyperpolarizing factor,³³ inactivation of endothelium-derived relaxing factors,³⁴ or augmented release of endothelium-derived constricting factors.³⁵ Further studies are needed to determine the relative contributions of these possibilities in impaired endothelium-dependent coronary vasodilation in this condition.

Conclusion

The results of this study suggest that diminished endothelium-dependent dilation of the large epicardial coronary artery in hypertensive patients may be related to a selective abnormality in the muscarinic receptor of the endothelium and that blunted endothelium-dependent dilation of the resistance coronary artery in this condition may be related to a more generalized endothelial abnormality but not confined to the muscarinic receptor. These findings may contribute to our understanding of the pathophysiology of altered endothelial function of the coronary arteries in hypertensive patients.