Nicorandil (N-[2-hydroxyethyl] nicotinamide nitrate) has been shown to be useful for the treatment of various types of ischemic heart disease (1,2). In basic pharmacologic studies, it has been demonstrated that its mechanism of action has two components: a cyclic guanosine monophosphate (GMP)-dependent component as a nitrate and an adenosine triphosphate (ATP)-sensitive K+ (KATP) channel-dependent component as a KATP channel opener (3). The increases in the duration of exercise and the time to the onset of ischemic ST-segment depression after oral nicorandil (20 mg) were almost equivalent to those after sublingual nitroglycerin (0.3 mg) in patients with stable exercise-induced angina pectoris (4). In many clinical cases, the antiischemic effect of nicorandil is almost equivalent to that of nitrates (1). However, nicorandil alleviates some cases of angina in which nitrates are not fully effective (1,2,5,6). These observations suggest the participation of the KATP channel-opening component. Recent clinical studies have suggested that nicorandil has cardioprotective effects because of the opening of KATP channels in patients with angina pectoris (7,8).
Previous basic studies have demonstrated that the two mechanistic components of nicorandil are quantitatively separable and act independently in the relaxation of isolated arteries (9,10). However, it is unclear which component is predominant when nicorandil concentrations are varied (11). Previous in vitro studies using isolated bovine (9) or rabbit (10) arteries have indicated that the effect of nicorandil as a KATP channel opener occurs at lower concentrations, whereas its cyclic GMP activity increases at higher concentrations. In contrast, an in vitro study using isolated dog coronary arteries indicated that the nitrate action of nicorandil was predominant at lower concentrations, whereas the KATP channel-opening action was observed at higher concentrations (12).
The aims of this study were to examine the effects of the intravenous infusion of nicorandil on coronary circulation as well as systemic hemodynamics in humans, and to discuss the relation between nicorandil concentration and the participation of the cyclic GMP- and KATP channel-dependent components. Unfortunately, it is difficult to apply KATP channel blockers such as glibenclamide (13,14) to humans. However, it has been demonstrated that nicorandil dilates the epicardial large coronary arteries in a manner similar to nitrates and relaxes the small coronary resistance arteries by opening KATP channels (11,15). Thus we assessed the changes in coronary artery diameter by angiography as an indicator of nitrate action, and evaluated the changes in coronary vascular resistance using Doppler flow as an indicator of KATP channel-opening action.
We also compared coronary hemodynamic changes produced by nicorandil with those produced by intracoronary nitroglycerin or papaverine. Nitroglycerin dilates epicardial coronary arteries although it has a minimal effect on coronary resistance arteries, and reduces preload more strongly than afterload (16). Papaverine is known to produce maximal dilation of coronary arteries (17,18).
Fourteen patients (13 men and one woman; mean age, 58 years; range, 37-74 years) undergoing cardiac catheterization for investigation of chest pain or ischemic changes on electrocardiograms (ECGs) participated in this study. All had normal sinus rhythm and an angiographically normal left anterior descending coronary artery. The study protocol was approved by the Ethical Committee of Shiga Medical University, and all patients gave their written informed consent before the study. None had left ventricular (LV) dysfunction, vasospastic angina, myocardial infarction, severe valvular heart disease, or cardiomyopathy. Nine of the 14 patients had an atypical history of anginal pain and showed a negative treadmill exercise test. They showed no organic stenosis in the right or left coronary arteries. Five of the 14 patients showed mild organic stenosis (<50%) only in the left circumflex artery or right coronary artery. None of the patients had diabetes mellitus, and thus none received glibenclamide before catheterization. All vasoactive medication was withdrawn ≥72 h before the study.
After the completion of diagnostic catheterization, an additional 3,000 U of heparin was given, and a 7F guide catheter was introduced into the proximal segment of the left coronary artery. A 3F infusion catheter was advanced through the guide catheter into the middle segment of the artery. Coronary blood flow velocity was measured with a 0.014-in. Doppler flow guidewire (Flowire; Cardiometrics, Mountain View, CA, U.S.A.). The guidewire was advanced through the guide catheter into the middle segment of the artery. This guidewire system has a miniature Doppler ultrasound crystal that transmits signals at a carrier frequency of 15 MHz and receives pulsedwave ultrasound signals, sampled at a distance of 5 mm from the guidewire tip (19-21). The tip of the guidewire was positioned distal to the end of the infusion catheter. The position was carefully selected in a segment of the vessel that was straight and free of any major branches, and that could be imaged without overlap from other vessels, thus allowing quantitative measurements of the coronary diameter. The Doppler signals were analyzed with a FloMap instrument (Cardiometrics). A quantitative estimate of coronary flow was calculated from the arterial cross-sectional area at the site of the Doppler sample volume and the time-averaged peak velocity. During catheterization, atrial pacing was performed at a pulse rate of 90 beats/min, which was selected to be in the physiologic range but greater than the patient's own heart rate. A Swan-Ganz catheter was inserted via the femoral vein to determine hemodynamic parameters, and cardiac output was measured by the thermodilution technique. The hemodynamic parameters were recorded using a multichannel cardiac monitor/recorder (RMC-1100; Nihon Kohden, Tokyo, Japan).
As shown in Fig. 1, nicorandil was continuously infused into the left brachial vein at a total dose of 12 mg over a period of 45 min. Before and 15, 30, and 45 min after the initiation of nicorandil infusion, blood samples were drawn from the right femoral artery and vein. Blood samples were centrifuged, and the plasma was separated and frozen. The plasma samples were assayed for nicorandil by high-performance liquid chromatography (22). At the same time points during nicorandil infusion, peak coronary flow velocity was measured, and coronary angiography was performed. Systemic blood pressure, pulmonary artery pressure, pulmonary capillary wedge pressure, right atrial pressure, and cardiac output were measured. After measuring the hemodynamic parameters with nicorandil, patients were given an intracoronary injection of 250 μg of nitroglycerin. Coronary flow velocity was measured, and coronary angiography was performed. Patients were then given an intracoronary injection of 12 mg of papaverine, which produces maximal dilation of coronary arteries in humans (17,18). Peak flow velocity measurement and coronary angiography were repeated. In addition, we examined the effects of intracoronary nitroglycerin at 10, 50, and 250 μg in preliminary experiments. The dilatory effect of 250 μg nitroglycerin on epicardial coronary arteries was almost equivalent to that of 50 μg nitroglycerin, suggesting that 250 μg nitroglycerin produces maximal coronary vasodilation. Thus this dose of nitroglycerin was chosen in our study.
Coronary angiograms were recorded using a cineangiographic system (Philips, Eindhoven, Netherlands), and quantitatively analyzed with a QCA-CMS (Medis, Leiden, Netherlands). The hemodynamic variables were calculated as follows. EQUATIONS (1)-(8)
Data are expressed as mean ± SEM. The alterations in hemodynamic parameters during nicorandil infusion are expressed as percentages of the baseline values before treatment with nicorandil. Single comparisons were performed with Student's paired t test. Multiple comparisons were performed by an analysis of variance (ANOVA) followed by Scheffé's post hoc test. Probability values of <0.05 were considered statistically significant.
Plasma level of nicorandil during the continuous intravenous infusion of nicorandil
Figure 2 shows plasma concentrations of nicorandil at 15, 30, and 45 min during the continuous infusion of nicorandil at a total dose of 12 mg for 45 min of infusion. Arterial plasma concentrations of nicorandil were higher than those in venous plasma. At 15, 30, and 45 min during nicorandil infusion, arterial plasma concentrations of nicorandil were 174.1 ± 11.3, 244.2 ± 17.2, and 315.2 ± 22.8 ng/ml, respectively, and venous plasma concentrations were 75.9 ± 4.1, 144.5 ± 5.3, and 203.3 ± 9.7 ng/ml, respectively (n = 14).
Response of systemic hemodynamics to nicorandil
As shown in Fig. 3, nicorandil significantly reduced systolic aortic pressure (−5.6 ± 1.0%, p < 0.05, n = 14), starting at a plasma level of ≥174.1 ng/ml, and caused a concentration-dependent decrease in systolic aortic pressure. The plasma levels are shown as mean arterial plasma concentrations of nicorandil at 15, 30, and 45 min during nicorandil infusion. Nicorandil produced significant decreases in pulmonary artery pressure at plasma levels of ≥244.2 ng/ml (−37.2 ± 8.4% at 244.2 ng/ml, −43.7 ± 7.3% at 315.2 ng/ml). As shown in Fig. 4, nicorandil also produced significant decreases in pulmonary capillary wedge pressure and the cardiac index at a plasma level of ≥244.2 ng/ml (pulmonary capillary wedge pressure, −82.5 ± 34.0% at 244.2 ng/ml; cardiac index, −10.5 ± 3.0% at 244.2 ng/ml). In the plasma concentration range of 174.1-315.2 ng/ml, however, nicorandil did not significantly change either the systemic or pulmonary vascular resistance index (Fig. 5).
Response of coronary circulation to nicorandil
Table 1 shows the coronary hemodynamic values at baseline, during the intravenous infusion of nicorandil, after the bolus injection of intracoronary nitroglycerin (250 μg), and after the bolus injection of intracoronary papaverine (12 mg), and Fig. 6 shows the changes in the coronary artery diameter and coronary vascular resistance index versus plasma levels of nicorandil. Nicorandil produced significant dilation of the epicardial coronary artery (+14.0 ± 3.3%; p < 0.05; n = 14) at a plasma level as low as 174.1 ng/ml. Although the continuous infusion of nicorandil caused a concentration-dependent increase in the epicardial coronary artery diameter (21.3 ± 3.6% at 244.2 ng/ml, 23.4 ± 3.8% at 315.2 ng/ml), neither nitroglycerin nor papaverine caused further increases in the coronary artery diameter (Table 1), suggesting that nicorandil at a plasma level of 315.2 ng/ml induced maximal vasodilation of the epicardial coronary artery. Conversely, nicorandil produced significant decreases in the coronary vascular resistance index at a plasma level of ≥244.2 ng/ml (−13.2 ± 3.6% at 244.2 ng/ml, −13.8 ± 3.9% at 315.2 ng/ml). Although nitroglycerin caused no further decrease in coronary vascular resistance, papaverine induced a 75% decrease in coronary vascular resistance (Table 1), suggesting that even at a maximal level of 315.2 ng/ml, nicorandil did not induce maximal dilation of coronary resistance arteries.
We examined the effects of nicorandil on systemic and coronary hemodynamics in humans. A 45-min intravenous infusion of nicorandil increased epicardial coronary artery diameter and decreased coronary vascular resistance, aortic pressure, pulmonary artery pressure, pulmonary capillary wedge pressure, and cardiac index. However, nicorandil had no significant effect on systemic or pulmonary vascular resistance. Nitroglycerin and papaverine were subsequently administered. Nitroglycerin caused no further changes in coronary artery diameter or coronary vascular resistance, whereas papaverine caused no further increase in coronary artery diameter, but markedly decreased coronary vascular resistance.
Nicorandil's mechanism of action has two components: a KATP channel-dependent component as a KATP channel opener and a cyclic GMP-dependent component as a nitrate (3). However, it remains unclear which component acts predominantly when nicorandil concentrations are varied (11). Kukovetz et al. (9) examined two mechanisms of relaxation by nicorandil in isolated bovine coronary arteries, and reported that at low concentrations, nicorandil predominantly exerts a relaxing effect by opening K+ channels, whereas at higher concentrations, the effect of cyclic GMP becomes greater. Meisheri et al. (10) also examined the quantitative separation of these two components in isolated rabbit mesenteric arteries, and concluded that the effect of nicorandil as a K+ channel opener occurred at lower concentrations, whereas its nitrate action was observed at higher concentrations. In contrast, our previous study using isolated dog coronary arteries indicated that the cyclic GMP-dependent component was predominant at low concentrations, whereas the KATP channel-dependent component was prevalent at higher concentrations (12). These differences in the vascular response to nicorandil may be explained by species-specific differences.
In this study, we assessed the changes in epicardial coronary artery diameter and coronary vascular resistance as indicators of the nitrate and KATP channel-opening actions of nicorandil, respectively. Nitrates preferentially dilate large coronary arteries (23), whereas KATP channel openers predominantly dilate small coronary arteries (11,24-26). Previous animal studies have indicated that the vasodilatory action of nicorandil on epicardial coronary arteries is not attenuated by the KATP channel blocker glibenclamide, suggesting that its action is related to its function as a nitrate (11,15). In contrast, its dilatory action on coronary resistance arteries was inhibited by KATP channel blockers, suggesting that its effect is related to its KATP channel-opening action. In the present study, nicorandil increased epicardial coronary artery diameter at lower concentrations (arterial plasma concentration <200 ng/ml). At higher concentrations (arterial plasma concentration >200 ng/ml), nicorandil significantly decreased coronary vascular resistance. Neither nitroglycerin nor papaverine caused further dilation of the epicardial coronary artery. The coronary vascular resistance after nicorandil treatment was markedly decreased by papaverine. Papaverine (12 mg) induces maximal dilation of both large and small coronary arteries. Nitroglycerin (250 μg) produces maximal dilation of large coronary arteries, but has only a minimal effect on small coronary arteries. Our results suggested that nicorandil at the higher concentration range used in this study (arterial plasma concentration of ∼300 ng/ml) caused maximal dilation of the epicardial coronary artery, but, even at the highest concentration, did not induce maximal dilation of coronary resistance arteries. Taken with the results of our previous study (22), our observation suggests that the participation of the KATP channel-dependent component increases as the nicorandil concentration increases (arterial plasma concentration >200 ng/ml), but, even at the maximal clinical level, the nitrate action is still predominant.
Previously we examined the in vivo coronary effects of the continuous intravenous infusion of nicorandil (0.5-100 μg/kg/min) in conscious dogs instrumented with ultrasonic crystals and an electromagnetic flow meter in the circumflex coronary artery (22). The epicardial coronary artery responded to nicorandil at a lower concentration (an arterial plasma concentration of ∼100 ng/ml), and reached almost maximal dilation at a higher concentration of >200 ng/ml. Dilation of the coronary resistance artery (i.e., decreased coronary vascular resistance) took place only at high concentrations (arterial plasma concentrations of >200 ng/ml), and maximal dilation did not occur even at a much higher concentration of >1,000 ng/ml. Thus our results concerning the relation between the plasma concentration of nicorandil and coronary hemodynamic changes in humans are compatible with those obtained in our previous studies using dogs.
In this study, nicorandil decreased pulmonary capillary wedge pressure, which reflects LV preload, but did not significantly alter systemic vascular resistance as an indicator of LV afterload. The vasodilation caused by nitrates is essentially predominant in the venous circulation (27), and thereby nitrates have a stronger effect on LV preload than on LV afterload (16). In contrast, the KATP channel-opening action is predominant for arterial dilation with a resultant reduction of afterload. Our results suggest that nicorandil decreases cardiac preload mainly through its nitrate action. Thus in systemic hemodynamics as well as coronary hemodynamics, the nitrate action of nicorandil is predominant. In addition, nicorandil reduced coronary vascular resistance but did not alter systemic or pulmonary vascular resistance in our study. These findings suggest that coronary resistance vessels are more sensitive to the KATP channel-opening action of nicorandil than are systemic or pulmonary resistance vessels. Nicorandil decreased the cardiac index in our study. It has been reported that nicorandil decreases systemic vascular resistance and increases the cardiac index in patients with ischemic heart disease (28) or congestive heart failure (29). This difference is probably because we studied patients with almost normal cardiac function, and heart rate was kept constant by cardiac pacing.
We previously reported that the venous plasma concentrations of nicorandil averaged 78 ± 83 (SD) ng/ml and 313 ± 142 ng/ml ∼30 min after the oral administration of 10 or 30 mg, respectively, in humans (30). The plasma concentration of nicorandil after the oral administration of 10 mg was equivalent to that at 15 min after intravenous infusion in this study, and the plasma concentration of nicorandil after the oral administration of 30 mg was somewhat higher than that at 45 min after intravenous infusion in this study. After each administration of high- or low-dose nicorandil, the treadmill exercise test was performed. Patients who had plasma levels >100 ng/ml with 10 mg of nicorandil showed significantly improved exercise capacity, but did not show any further increase with a dose of 30 mg. Our results indicate that in this concentration range, nicorandil acts on systemic and coronary hemodynamics largely through its nitrate action. Thus our previous and present studies suggest that, as a vasodilator, the clinical effect of nicorandil is largely attributable to its action as a nitrate. However, it has been shown that nicorandil alleviates angina in some patients in whom nitrates are not fully effective (1,2,5,6). This effect may be related to its action on myocardial KATP channels to cause pharmacologic preconditioning. Recent animal and clinical studies (7,8) have shown that KATP channel openers afford cardioprotection against ischemia through their pharmacologic preconditioning effects, which may be mediated through mitochondrial KATP channels in the myocardium (31). It has recently been reported that nicorandil can specifically activate the mitochondrial KATP channels in rabbit ventricular myocytes (32). Thus nicorandil may exert clinical benefits mainly by two mechanisms of action: vasodilatory action on coronary and systemic hemodynamics as a nitrate and cardioprotective action through myocardial KATP channels to cause pharmacologic preconditioning as a KATP channel opener. However, further studies are required to determine whether nicorandil at clinical doses activates mitochondrial KATP channels in human myocardium.
In summary, as assessed by its vasodilatory action on coronary and systemic hemodynamics, the nitrate action of nicorandil was predominant at clinical doses. The contribution of the KATP channel-opening action of nicorandil was observed at high concentrations. However, even at the clinically maximal level, the nitrate action was predominant over its KATP channel-opening action.
Acknowledgment: We thank Ms. Ikuko Sakaguchi for her technical assistance.
1. Kinoshita M, Sakai K. Pharmacology and therapeutic effects of nicorandil. Cardiovasc Drugs Ther
2. Frampton J, Buckley MM, Fitton A. Nicorandil: a review of its pharmacology and therapeutic efficacy in angina pectoris. Drugs
3. Taira N. Nicorandil as a hybrid between nitrates and potassium channel activators. Am J Cardiol
4. Hayata N, Araki H, Nakamura M. Effects of nicorandil on exercise tolerance in patients with stable effort angina: a double-blind study. Am Heart J
5. Kato K, Iimura H, Hirosawa K, et al. Clinical evaluation of intravenous SG-75 in treatment of unstable angina: a multicenter double-blind comparative study with intravenous ISDN. Rinsho Iyaku
6. Liu Q, Nakae I, Takahashi M, et al. Comparative efficacy of nicorandil and nitroglycerin in patients with chronic stable angina pectoris. Int Med J
7. Saito S, Mizumura T, Takayama T, et al. Antiischemic effects of nicorandil during coronary angioplasty in humans. Cardiovasc Drugs Ther
8. Patel DJ, Purcell HJ, Fox KM on behalf of the CESAR 2 investigation. Cardioprotection by opening of the KATP
channel in unstable angina: Is this a clinical manifestation of myocardial preconditioning? Results of a randomized study with nicorandil. Eur Heart J
9. Kukovetz WR, Holzmann S, Braida C, Poch G. Dual mechanism of the relaxing effect of nicorandil by stimulation of cyclic GMP formation and hyperpolarization. J Cardiovasc Pharmacol
10. Meisheri KD, Cipkus-Dubray LA, Hosner JM, Khan SA. Nicorandil-induced vasodilation: functional evidence for K+
channel-dependent and cyclic GMP-dependent components in a single vascular preparation. J Cardiovasc Pharmacol
11. Nakae I, Liu Q, Hashimoto K, Sugimoto Y, Tsutamoto T, Kinoshita M. Mechanism of the vasodilatory action of nicorandil on coronary circulation in dogs. Cardiovasc Drugs Ther
12. Matsumoto T, Takahashi M, Omura T, et al. Heterogeneity of the vasorelaxing effect of nicorandil on dog coronary arteries: comparison with other NO donors. J Cardiovasc Pharmacol
13. Schmid-Automarchi H, De Weille J, Fosset M, Lazdunsky M. The receptor for antidiabetic sulfonylureas controls the activity of the ATP-modulated K+
channel in insulin-secreting cells. J Biol Chem
14. Eltze M. Glibenclamide is a competitive antagonist of cromakalim, pinacidil and RP 49356 in guinea-pig pulmonary artery. Eur J Pharmacol
15. Imagawa J, Akima M, Nabata H, Taira N. Spasmolytic action of nicorandil in canine conductive coronary arteries in vivo is not modified by glibenclamide. J Cardiovasc Pharmacol
16. Hashimoto K, Ohbayashi Y, Kinoshita M. Left ventricular effects of nicorandil in comparison with nitroglycerin in chronic conscious dogs. Cardiovasc Drugs Ther
17. Wilson RF, White CW. Intracoronary papaverine: an ideal coronary vasodilator for studies of the coronary circulation in conscious humans. Circulation
18. Hongo M, Goto T, Watanabe N, et al. Relation of phasic coronary flow velocity profile to clinical and hemodynamic characteristics of patients with aortic valve disease. Circulation
19. Anderson HV, Kirkeeide RL, Willerson JT, Smalling RW, Schroth G. Coronary artery flow monitoring: the value of intravascular Doppler for detection of complications after interventions. Eur Heart J
20. Sudhir K, Chou TM, Hutchison SJ, Chatterjee K. Coronary vasodilation induced by angiotensin-converting enzyme inhibition in vivo: differential contribution of nitric oxide and bradykinin in conductive and resistance arteries. Circulation
21. Shiode N, Morishima N, Nakayama K, Yamagata T, Matsuura H, Kajiyama G. Flow-mediated vasodilation of human epicardial coronary arteries: effect of inhibition of nitric oxide synthesis. J Am Coll Cardiol
22. Hashimoto K, Kinoshita M, Ohbayashi Y. Coronary effects of nicorandil in comparison with nitroglycerin in chronic conscious dogs. Cardiovasc Drugs Ther
23. Chu A, Murray JJ, Lin C-C, Russell M, Hagen P-O, Cobb FR. Preferential proximal coronary dilation by activators of guanylate cyclase in awake dogs. Am J Physiol
24. Giudicelli J-F, Drieu La Rochelle C, Berdeaux A. Effects of cromakalim and pinacidil on large epicardial and small coronary arteries in conscious dogs. J Pharmacol Exp Ther
25. Yoneyama F, Satoh K, Taira N. Nicorandil increases coronary blood flow predominantly by K-channel opening mechanism. Cardiovasc Drugs Ther
26. Arai K, Wang Y, Sato K, et al. Vasodilatory effect of nicorandil on coronary arterial microvessels: its dependency on vessel size and the involvement of the ATP-sensitive potassium channels. J Cardiovasc Pharmacol
27. Mackenzie JE, Parratt JR. Comparative effects of glyceryl trinitrate on venous and arterial smooth muscle in vitro; relevance to antianginal activity. Br J Pharmacol
28. Aizawa T, Ogasawara K, Kato K. Effects of nicorandil on coronary circulation in patients with ischemic heart disease: comparison with nitroglycerin. J Cardiovasc Pharmacol
29. Tsutamoto T, Kinoshita M, Nakae I, et al. Absence of hemodynamic tolerance to nicorandil in patients with severe congestive heart failure. Am Heart J
30. Kinoshita M, Nishikawa S, Sawamura M, et al. Comparative efficacy of high-dose versus low-dose nicorandil therapy for chronic stable angina pectoris. Am J Cardiol
31. Liu Y, Sato T, O'Rourke B, Marban E. Mitochondrial ATP-dependent potassium channels: novel effectors of cardioprotection? Circulation
32. Sato T, Sasaki N, O'Rourke B, Marbán E. Nicorandil, a potent cardioprotective agent, acts by opening mitochondrial ATP-dependent potassium channels. J Am Coll Cardiol