Bradykinin-Induced Vasodilation Is Impaired at the Atherosclerotic Site but Is Preserved at the Spastic Site of Human Coronary Arteries In Vivo
Background Bradykinin causes endothelium-dependent vasodilation of isolated human coronary arteries in vitro. However, the effect of bradykinin on vasomotion of human coronary arteries in vivo has not been studied. The aim of this study was to examine whether bradykinin-induced vasodilation is altered at the atherosclerotic or spastic site of human coronary arteries in vivo.
Methods and Results The effect of bradykinin on vasomotion of epicardial coronary arteries was evaluated in 8 patients with normal coronary arteries (control group), 14 patients with organic coronary stenosis (coronary artery disease [CAD] group), and 8 patients with vasospastic angina (VSA group). Changes in the diameter of epicardial coronary artery were assessed by quantitative coronary arteriography. Intracoronary administration of bradykinin at graded doses (60, 200, and 600 ng) dilated epicardial coronary arteries without altering arterial pressure or heart rate in all patients of either group. In the control group, vasomotor responses of the site where acetylcholine caused dilation were compared with the responses of the site where acetylcholine caused constriction. The magnitudes of bradykinin-induced dilation at the site with acetylcholine-induced dilation (mean±SD: 6±6%, 11±9%, and 15±9%) were comparable to that (3±6%, 8±8%, and 13±9%) at the site with acetylcholine-induced constriction. In the CAD group, vasomotor responses of the stenotic site (% diameter stenosis, 15% to 50%) and nonstenotic site were examined. The bradykinin-induced dilation at the stenotic site (0±4%, 3±8%, and 5±9%) was significantly less (P<.01) than at the nonstenotic site (3±4%, 8±6%, and 16±11%) and in the control group. Coronary vasodilation with nitrate at the stenotic site (20±11%) was comparable to that at the nonstenotic site (22±16%) and in the control group (21±10%). In the VSA group, vasomotor responses of the site with acetylcholine-induced spasm and the site without spasm were examined. The bradykinin-induced vasodilation at the spastic site (5±5%, 16±15%, and 33±17%) was comparable to that at the nonspastic site (4±8%, 12±14%, and 21±9%). Nitrate-induced dilation was comparable at the spastic site (51±19%) and the nonspastic site (32±13%). The ratio of bradykinin-induced vasodilation to nitrate-induced vasodilation at the spastic site was comparable to the control group.
Conclusions These results suggest that bradykinin causes vasodilation of human epicardial coronary arteries in vivo and that bradykinin-induced endothelium-dependent vasodilation is impaired at the stenotic site but is preserved at the angiographically normal site where endothelium-dependent vasodilation by acetylcholine is impaired and at the spastic site.
Vascular endothelial cells play an important role in the control of vasomotor tone by releasing vasoactive substances.1 2 3 4 Recent clinical studies have implied that coronary endothelial dysfunction may contribute to the pathogenesis of ischemic heart disease.5 Therefore, it is important to examine the alteration of endothelial function of human coronary arteries in various pathological states such as atherosclerosis or vasospasm. Endothelium-dependent vasodilation of human coronary arteries induced by acetylcholine is impaired at the atherosclerotic site6 7 8 and the angiographically normal site in patients with coronary artery disease9 10 11 or with coronary risk factors.8 12 In contrast, endothelium-dependent vasodilation by substance P is preserved at the angiographically normal site in patients with coronary artery disease13 and at the spastic site in patients with variant angina.7 14 These results are consistent with the hypothesis that impaired endothelium-dependent vasodilation of human coronary arteries during the atherosclerotic process may be selective to the muscarinic receptors that mediate the effects of acetylcholine.
It has been shown that bradykinin, an endogenous vasoactive substance, induces endothelium-dependent relaxation of isolated human coronary artery in vitro.15 16 17 18 However, the effects of bradykinin on vasomotion of epicardial coronary arteries in patients with coronary atherosclerosis or spasm have not been studied in vivo. The aim of this study was first, to determine whether bradykinin dilates epicardial human coronary arteries in vivo, and second, to examine whether endothelium-dependent vasodilation with bradykinin is altered at the atherosclerotic site in patients with coronary artery disease and at the spastic site in those with vasospastic angina.
Thirty patients were divided into three groups on the basis of their clinical findings, including coronary angiograms (Table 1⇓). The control group consisted of 8 patients with angiographically normal coronary arteries in all three coronary artery branches; 5 patients had atypical chest pain, 2 had valvular heart disease, and another had congenital heart disease (atrial septal defect). All patients in the control group showed a negative treadmill exercise test. The coronary artery disease (CAD) group consisted of 14 patients with organic coronary stenosis (>15%, <99%) in the left coronary arteries. The vasospastic angina (VSA) group consisted of 8 patients in whom vasospasm (% luminal reduction >75%) of the left coronary artery was provoked by intracoronary acetylcholine (100 μg). All patients had a typical history of anginal pain at rest. Three patients showed ST-segment elevation, and the other 5 patients showed ST-segment depression during the spasm. Four patients showed no organic stenosis in the three coronary artery branches, and the other 4 patients showed mild organic stenosis (<50%) in the left or right coronary arteries. The spastic site was angiographically normal after administration of isosorbide dinitrate (ISDN) in these 8 patients. There was no complex plaque or thrombus at any sites of the coronary arteries in any patient of either group.
The number of coronary risk factors (hyperlipidemia, diabetes mellitus, arterial hypertension, smoking habit, and family history of coronary artery disease) was significantly greater in the CAD group (1.9±0.9) than in the control group (0.9±0.6) (P<.05). That in the VSA group was 1.4±0.7.
Written informed consent was obtained from each patient after approval by the Institutional Review Board. Cardiac catheterization was performed in the fasting state after oral premedication of diazepam (5 mg). Antianginal and antihypertensive medications were discontinued at least 24 hours before the study (Table 1⇑).
After control arteriography of the left coronary artery was performed, the following studies were performed. First, bradykinin was administered into the left coronary artery by hand. Each dose of bradykinin was diluted with 3 mL of physiological saline, which was infused into the coronary artery through the Judkins catheter in 10 seconds. The catheter was then flushed by infusing 3 mL of saline in 20 seconds. Coronary arteriography was performed 1 minute after the end of the infusion. We evaluated the effects of three doses of bradykinin (60, 200, and 600 ng). We waited for 2 minutes before administration of the next dose of bradykinin. We found in the preliminary studies that bradykinin at doses >600 ng significantly decreased arterial pressure, which is consistent with the previous study by Bonner et al.19 Moreover, we found that bradykinin at 600 ng did not increase coronary blood flow velocity 1 minute after bradykinin infusion using a Doppler flow guide wire (Advanced Cardiovascular Systems Inc), which is consistent with the results by Pelc et al.20
Second, acetylcholine at 100 μg was administered into the left coronary artery. This dose of acetylcholine was diluted with 2 mL of physiological saline and was infused into the left coronary artery through the Judkins catheter in 10 seconds. The catheter was then flushed by infusing 3 mL of saline in 20 seconds. Coronary angiograms were recorded 1 minute after the beginning of the acetylcholine infusion to evaluate coronary spasm. The additional coronary angiograms were recorded 2 minutes after the beginning of the infusion to evaluate endothelium-dependent vasodilation by acetylcholine. In 5 patients of the CAD group, acetylcholine was not infused because they had severe organic stenosis.
Third and finally, ISDN (2 mg/4 mL) was administered into the left coronary artery, and coronary angiograms were recorded 2 minutes after the beginning of the ISDN infusion. Our previous study confirmed that intracoronary infusion of saline had no significant effects on the coronary diameter, arterial pressure, or heart rate.7
Phasic and mean aortic blood pressures, heart rate, and 12-lead ECGs were continuously monitored using a Nihon-Koden polygraph system and were recorded on a multichannel recorder.
Quantitative Coronary Arteriography and Its Analysis
Coronary cineangiograms were recorded using a Siemens cineangiographic system (Siemens Bicor & Hicor). Nonionic contrast media (Ioversol, Yamanouchi Pharmaceutical Co) was used. An appropriate view that permitted clear visualization of the coronary segment under study without overlapping branches was selected. An angle of the view, the distance from x-ray focus to the object, and that from the object to the image intensifier were carefully kept constant during the study. An end-diastolic frame of the angiogram was selected, and the luminal diameter was determined quantitatively by a cinevideodensitometric analysis system (Kontron Instruments).21 The readily identifiable branch points were used as a reference marker of the measurement to allow assessment serial changes in the diameter of the same site. The diameters were measured three times, and the mean value was used for the analysis. The size of the Judkins catheter was used for calibrating the arterial diameter in millimeters. The accuracy and precision of quantitative angiographic measurements were determined from the analysis of cinefilms of the phantom with the precision-drilled models of coronary arteries with diameters of 1.5, 2.0, and 3.0 mm (Kyoto Kagaku Hyouhon Co) filled with contrast medium and filmed under 5 cm of water.22 The accuracy was 0.7±0.2%, and the precision was 2.6±0.6% (mean±SD, n=165 measurements). Measurements with this system for interobserver and intraobserver reproducibility were high (r=.96 and r=.98, respectively). The changes in the coronary diameter in response to vasoactive drugs (bradykinin, acetylcholine, and ISDN) were expressed as percent changes from the baseline value. The degrees of organic stenosis were expressed as % diameter stenosis in angiograms after administration of ISDN.
In the control group, we measured changes in the luminal diameter at 25 sites where acetylcholine caused dilation (proximal [n=5], middle [n=8], and distal [n=12] segments of the left coronary artery) and those at 19 sites where acetylcholine caused vasoconstriction 2 minutes after acetylcholine. These measurements were designed because 6 of the 8 patients in the control group had at least one risk factor that has been shown to impair endothelium-dependent coronary dilation in response to acetylcholine.8 12 It is reported that responses of angiographically normal coronary artery segments to acetylcholine may be segmental rather than diffuse.23 Thus, we assumed that the angiographically normal segment with acetylcholine-induced dilation is functionally normal.
In the CAD group, we determined changes in the luminal diameter at 20 paired organic stenotic (15% to 50% stenosis) (Table 1⇑) and adjacent nonstenotic sites. The adjacent nonstenotic site was defined as the angiographically normal, smooth segment.
In the VSA group, we determined changes in the luminal diameter at eight paired spastic and nonspastic sites, none of which showed organic stenosis. The nonspastic site was defined as an adjacent segment proximal or distal to the spastic site when focal spasm was provoked or as a segment of another nonspastic vessel with the baseline diameter similar to that at the spastic site when diffuse coronary vasospasm was provoked.
Preparation of Bradykinin and Acetylcholine
Bradykinin (Sigma Chemical Co) was diluted with physiological saline at a concentration of 0.2 mg/mL and was sterilized at the Department of Pharmacy, Kyushu University Hospital. Acetylcholine chloride (Dai-ichi Pharmaceutical Co) was freeze-dried and stored at room temperature. Further dilution was made with physiological saline immediately before use.
Data are expressed as mean±SD. When serial changes in % increase in luminal diameter or hemodynamic variables in response to bradykinin were compared, ANOVA for repeated measures followed by Bonferroni’s multiple comparison test was used. The relation between nitrate-induced vasodilation and bradykinin-induced vasodilation was analyzed by a single linear regression. Probability of less than .05 was considered statistically significant.
Baseline mean aortic blood pressure and heart rate were comparable among the control group, the CAD group, and the VSA group (Table 2⇓). Intracoronary administration of bradykinin at graded doses did not alter arterial pressure or heart rate in the three groups (NS by one-way ANOVA) (Table 2⇓).
Bradykinin-Induced Coronary Vasodilation
In the control group, bradykinin dilated (P<.01 by one-way ANOVA) the site with acetylcholine-induced dilation (11±7%) in a dose-dependent manner (6±8%, 11±9%, and 15±9%) (Fig 1⇓). The bradykinin-induced vasodilation was comparable among the proximal, middle, and distal segments of the left coronary arteries (Table 3⇓). The bradykinin-induced vasodilation (3±6%, 7±8%, and 13±9%) at the site with acetylcholine-induced constriction (−15±12%) was comparable to that at the site with acetylcholine-induced dilation (Fig 1⇓).
In the CAD group, bradykinin at graded doses significantly increased the diameter at the nonstenotic site (P<.01 by one-way ANOVA) but not at the stenotic site (NS by one-way ANOVA) (Fig 2⇓). The % increases in the diameter evoked with bradykinin at the stenotic site (0±4%, 3±8%, and 5±9%) were significantly less (P<.01 by two-way ANOVA) than those at the nonstenotic site (3±4%, 8±6%, and 16±11%) and in the control group. There was no significant difference between bradykinin-induced coronary vasodilation at the nonstenotic site in the CAD group and in the control group (Fig 2⇓).
In the VSA group, coronary vasodilation evoked with bradykinin was comparable between the spastic site (5±5%, 16±15%, and 33±17%) and nonspastic site (4±8%, 12±14%, and 21±9%) (Fig 3⇓). Bradykinin-induced coronary vasodilation at the spastic site in the VSA group was significantly greater than in the control group (P<.01 by two-way ANOVA).
Nitrate-Induced Coronary Vasodilation
The % increases in the diameter evoked with nitrate were 21±10% in the control group, 20±11% and 22±16% (NS) at the stenotic and nonstenotic sites in the CAD group, and 51±19% and 32±13% (NS) at the spastic and nonspastic sites in the VSA group. Nitrate-induced vasodilation at the spastic site in the VSA group was significantly greater (P<.01) than in the control group. There was a significant positive correlation between nitrate-induced dilation and bradykinin-induced dilation in the control group (P<.01, r=.79). Thus, the ratio of bradykinin-induced vasodilation to nitrate-induced vasodilation was calculated. The ratio at the spastic site was comparable to that at the nonspastic site and in the control group (Fig 4B⇓). The ratio at the stenotic site in the CAD group was significantly less (P<.01 by two-way ANOVA) than at the nonstenotic site and in the control group (Fig 4A⇓).
We obtained four novel observations in the present study. First, intracoronary administration of bradykinin dilated human epicardial coronary arteries in vivo. Second, bradykinin-induced dilation was preserved at the angiographically normal site where acetylcholine-induced dilation was impaired in the control group. Third, coronary vasodilation evoked with bradykinin was impaired at the stenotic site in the CAD group. Fourth, bradykinin-induced dilation was preserved at the spastic site where coronary spasm was provoked by acetylcholine in the VSA group.
Bradykinin-Induced Vasodilation of Human Coronary Arteries In Vivo
In the present study, we administered bradykinin selectively at graded doses into the left coronary arteries to examine the local effect of bradykinin on vasomotion of human epicardial coronary arteries in vivo. Since preliminary studies had shown that bradykinin at doses exceeding 600 ng decreased systemic arterial pressure and that bradykinin at 600 ng did not change coronary blood flow velocity, we selected doses of bradykinin not exceeding 600 ng. The present study showed that bradykinin at these doses increased the diameters of epicardial coronary arteries without altering arterial pressure or heart rate. These results suggest that bradykinin-induced dilation resulted from the direct action of bradykinin on the epicardial coronary artery but not from the indirect action such as flow-mediated dilation. In vitro studies confirmed that bradykinin did not constrict isolated human coronary smooth muscle15 16 and that bradykinin-induced dilation of isolated human coronary artery was totally endothelium dependent.15 16 17 18 Thus, it appears reasonable to assume that bradykinin-induced dilation observed in the present study resulted from the release of endothelium-derived relaxing factor(s). It has been shown that bradykinin stimulates the release of endothelium-derived nitric oxide,15 16 17 hyperpolarizing factor,18 and prostacyclin24 ; however, we did not examine the relative contribution of these three factors in mediating bradykinin-induced coronary vasodilation in vivo.
Effect of Bradykinin at the Atherosclerotic Site
Previous studies have reported that endothelium-dependent vasodilation is impaired in atherosclerotic human epicardial coronary arteries.5 6 7 8 25 26 Endothelium-dependent coronary vasodilation evoked with acetylcholine is impaired not only at the atherosclerotic sites6 7 8 but also at the angiographically normal segments of coronary arteries in patients with coronary artery disease9 10 11 or with coronary risk factors,8 12 indicating that altered endothelium-dependent dilation with acetylcholine may precede the structural changes of arterial wall assessed by arteriography. On the other hand, substance P–induced vasodilation is not impaired at the angiographically normal segments in patients with coronary artery disease.13 These results suggest that endothelium-dependent vasodilation evoked with substance P may be preserved at the early stage of atherosclerotic process.5
In the present study, we demonstrated that bradykinin-induced coronary vasodilation at the stenotic site was significantly less than that at the nonstenotic site in the CAD group and in the control group. These results indicate that endothelium-dependent vasodilation evoked with bradykinin was impaired at the atherosclerotic site. We also demonstrated that bradykinin-induced vasodilation was preserved not only at the angiographically normal sites with acetylcholine-induced vasoconstriction in the control group but also at the nonstenotic site in the CAD group. The results of the present study and of the previous study suggest that endothelium-dependent vasodilation with bradykinin and substance P is preserved but that vasodilation with acetylcholine is impaired at the early stage of the atherosclerotic process.5 27
Effect of Bradykinin at the Spastic Site
Recent clinical studies have demonstrated that endothelium-dependent vasodilation with substance P,7 14 histamine,25 and the low dose of acetylcholine7 is preserved at the spastic site in patients with vasospastic angina. However, the effects of bradykinin on vasomotion at the spastic site have not been studied. The present study demonstrated that bradykinin-induced coronary vasodilation at the spastic site was comparable to that at the nonspastic site in the VSA group. There was no significant difference in nitrate-induced vasodilation between the spastic site and the nonspastic site. The ratio of bradykinin-induced vasodilation to nitrate-induced vasodilation in the VSA group was comparable to that in the control group. These results suggested that endothelium-dependent coronary vasodilation evoked with bradykinin was preserved at the spastic site and that augmented vasodilation evoked with bradykinin in the VSA group may be due to elevated basal coronary artery tone.28 Our results are in agreement with the suggestion that coronary spasm is caused by augmented reactivity of vascular smooth muscle but not by defective endothelial function alone.7 29
Limitations of the Study
The first limitation was that a small number of patients was studied. The second limitation was that most of our patients were taking various antianginal and antihypertensive drugs before the study (Table 1⇑). Since these drugs were discontinued 24 hours before the study, it is unlikely that the drugs have significant effects on coronary vasomotor tone.
The results of this study suggest that bradykinin causes vasodilation of human epicardial coronary arteries in vivo and that endothelium-dependent coronary vasodilation with bradykinin is impaired at the atherosclerotic site but is preserved at the angiographically normal site where endothelium-dependent coronary vasodilation by acetylcholine is impaired, at the angiographically normal site in patients with coronary atherosclerosis, and at the spastic site in patients with vasospastic angina. Our results may provide further information for understanding the pathophysiology of endothelial dysfunction in coronary atherosclerosis and vasospasm in humans.
This study was supported by Grants-in-Aid for JSPS Fellowships for Japanese Junior Scientists, by a Japan Cardiovascular Research Foundation Grant, Osaka, Japan, and by a Grant-in-Aid from the Ministry of Education, Science, and Culture, Tokyo, Japan. Dr T. Kuga is a Research Fellow of the Japan Society for the Promotion of Science. The authors are grateful to Tie Fukagawa for her technical assistance and to Dr Daisuke Teshima and Dr Osamu Fujishita, the Department of Pharmacy at Kyushu University Hospital, for preparing bradykinin for human use.
- Received November 14, 1994.
- Revision received January 3, 1995.
- Accepted January 10, 1995.
- Copyright © 1995 by American Heart Association
Furchgott RF. Role of endothelium in response of vascular smooth muscle. Circ Res. 1983;53:557-573.
Lucher TF, Richard V, Tschudi M, Yang Z, Boulanger C. Endothelial control of vascular tone in large and small coronary arteries. J Am Coll Cardiol. 1990;15:512-527.
Meredith IT, Yeung AC, Weidinger FF, Anderson TJ, Uehata A, Ryan TJ Jr, Selwyn AP, Gantz P. Role of impaired endothelium-dependent vasodilation in ischemic manifestations of coronary artery disease. Circulation. 1993;87(suppl V):V-56-V-66.
Egashira K, Inou T, Yamada A, Hirooka Y, Takeshita A. Preserved endothelium-dependent vasodilation at the vasospastic site in patients with variant angina. J Clin Invest. 1992;89:1047-1052.
Egashira K, Inou T, Hirooka Y, Yamada A, Maruoka Y, Kai H, Sugimachi M, Suzuki S, Takeshita A. Impaired coronary blood flow response to acetylcholine in patients with coronary risk factors and proximal atherosclerosis. J Clin Invest. 1993;91:29-37.
Werns SW, Walton JA, Hsia HH, Nabel EG, Sanz ML, Pitt B. Evidence of endothelial dysfunction in angiographically normal coronary arteries of patients with coronary artery disease. Circulation. 1989;79:287-291.
Hodgson JM, Marshall JJ. Direct vasoconstriction and endothelium-dependent vasodilation: mechanisms of acetylcholine effects on coronary flow and arterial diameter in patients with nonstenotic coronary arteries. Circulation. 1989;79:1043-1051.
Zeiher AM, Schachinger V, Hohnloser SH, Saurbier B, Just H. Coronary atherosclerotic wall thickening and vascular reactivity in humans: elevated high-density lipoprotein levels ameliorate abnormal vasoconstriction in early atherosclerosis. Circulation. 1994;89:2525-2532.
Vita JA, Treasure CB, Nabel EG, McLenachan JM, Fish RD, Yeung AC, Vekshtein VI, Selwyn AP, Ganz P. Coronary vasomotor response to acetylcholine relates to risk factors for coronary artery disease. Circulation. 1990;81:491-497.
Crossman DC, Larkin SW, Dashwook MR, Davies GJ, Yacoub M, Maseri A. Responses of atherosclerotic human coronary arteries in vivo to the endothelium-dependent vasodilator substance P. Circulation. 1991;84:2001-2010.
Forstermann U, Mugge A, Alheid U, Haverich A, Frolich JC. Selective attenuation of endothelium-mediated vasodilation in atherosclerotic human coronary arteries. Circ Res. 1988;62:185-190.
Toda N, Okamura T. Endothelium-dependent and -independent responses to vasoactive substances of isolated human coronary arteries. Am J Physiol. 1989;257:H988-H995.
Nakashima M, Mombouli J-V, Taylor AA, Vanhoutte PM. Endothelium-dependent hyperpolarization caused by bradykinin in human coronary arteries. J Clin Invest. 1993;92:2867-2871.
Bonner G, Schunk PU, Toussaint C, Kaufmann W. Hemodynamic effects of bradykinin on systemic and pulmonary circulation in healthy and hypertensive humans. J Cardiovasc Pharmacol. 1990;15(suppl 6):S46-S56.
Pelc LR, Gross GJ, Warltier DC. Mechanisms of coronary vasodilation produced by bradykinin. Circulation. 1991;83:2048-2056.
El-Tamimi H, Mansour M, Wargovich TJ, Hill JA, Kerensky RA, Conti R, Pepine CJ. Constrictor and dilator responses to intracoronary acetylcholine in adjacent segments of the same coronary artery in patients with coronary artery disease: endothelial function revisited. Circulation. 1994;89:45-51.
Zanzinger J, Zheng X, Bassenge E. Endothelium-dependent vasomotor responses to endogenous agonists are potentiated following ACE inhibition by a bradykinin dependent mechanism. Cardiovasc Res. 1994;28:209-214.
Golino P, Piscione F, Willerson JT, Cappelli-Bigazzi M, Focaccio A, Villari B, Indolfi C, Russolillo E, Condorelli M, Chiariello M. Divergent effects of serotonin on coronary artery dimensions and blood flow in patients with coronary atherosclerosis and control patients. N Engl J Med. 1991;324:641-648.
Shimokawa H, Flavahan NA, Vanhoutte PM. Loss of endothelial pertussis toxin–sensitive G protein function in atherosclerotic porcine coronary arteries. Circulation. 1991;83:652-660.
Fukai T, Egashira K, Hata H, Numaguchi K, Ohara Y, Takahashi T, Tomoike H, Takeshita A. Serotonin-induced coronary spasm in a swine model: a minor role of defective endothelium-derived relaxing factor. Circulation. 1993;88(pt 1):1922-1930.