Controlled Nicorandil Administration for Myocardial Protection During Coronary Artery Bypass Grafting Under Cardiopulmonary Bypass

Abstract

 

Nicorandil is a hybrid potassium channel opener, and recent experimental studies have demonstrated its efficacy in myocardial protection against ischemia-reperfusion. This clinical study was designed to examine the myocardial protective effect of nicorandil administered during cardiopulmonary bypass. Seventy adult patients, 53 men and 17 women, undergoing elective coronary artery bypass grafting were randomly assigned to two groups, one receiving nicorandil during cardiopulmonary bypass (group N, n = 35) and the other receiving no nicorandil for control (group C, n = 35). Nicorandil was administered at each dose of 0.1 mg/kg into the cardiopulmonary bypass circuit according to the following schedule: (¹) bolus injection at the initiation of cardiopulmonary bypass, (²) continuous infusion for 5 min before aortic cross-clamping, (³) bolus administration at 5 min before reperfusion, and (⁴) continuous infusion for 5 min before reperfusion. The time required for achieving cardiac arrest after the initial cardioplegia was significantly reduced in group N in comparison with that in group C. After aortic unclamping, the number of patients showing a significant ST segment change on the electrocardiogram was significantly fewer in group N, whereas the number of patients showing spontaneous recovery of heart beat was significantly greater. As for the myocardial protective effect, group N showed lower plasma levels of malondialdehyde, human-heart fatty acid-binding protein, and peak creatine kinase-MB, and required lower doses of catecholamine. Our results suggest that nicorandil administration during cardiopulmonary bypass provides enhanced myocardial protective effects against ischemia-reperfusion in patients undergoing coronary artery bypass grafting.

Nicorandil, an adenosine triphosphate-sensitive potassium (KATP) channel opener (¹), has beneficial pharmacologic effects that are clinically applicable for hemodynamic management after open heart surgery. Hyperpolarization induced by potassium channel opening secondarily reduces calcium influx, leading to relaxation of arterial smooth muscle and dilatation of peripheral and small coronary arterioles (²). Thus, nicorandil has a relatively selective vasodilating effect on coronary vessels in comparison with other vasodilating agents such as nifedipine, verapamil, and diltiazem (^3,4). In patients with coronary artery disease, spastic contraction of the coronary vessels is easily induced (^5,6). During coronary artery bypass grafting (CABG), mechanical stimulation of the vascular wall induced by cardiopulmonary bypass can trigger perioperative coronary artery spasms (^7,8), which sometimes cause perioperative myocardial infarction associated with increased risk of postoperative mortality (⁸). Furthermore, the occurrence of circulatory collapse after coronary revascularization often makes it difficult to use other vasodilating agents possibly affecting systemic circulation. Thus, nicorandil is gaining clinical popularity for the prevention of coronary spasms, perioperatively in particular (^9–11).

Previous experimental studies demonstrated the myocardial protective effect of potassium channel openers for ischemic or pharmacologic preconditioning (^11–13) and as an adjunct to cardioplegia (^14–18). Ischemic and pharmacologic preconditioning, mediated by protein kinase C, are thought to be associated with potassium channel opening (¹⁹). Recent experimental studies have suggested that the mitochondrial K_ATP channel is both a trigger and distal effector of ischemic and pharmacologic preconditioning (^20,21). As for cardioplegic myocardial protection, cardioplegia solution containing potassium channel opener is reported to induce hyperpolarized cardioplegic arrest without a high concentration of potassium (¹⁸). These pharmacologic effects of potassium channel openers seem to be useful in improving perioperative myocardial protection. Namely, pretreatment with a potassium channel opener before cardioplegic arrest provides myocardial tolerance to ischemia-reperfusion to the same degree that ischemic preconditioning does (^11–13). Rapid cardioplegic arrest can be achieved by a combination of potassium channel opener and high-concentration potassium, which attenuates myocardial injury caused by the introduction of ischemia (^14–18). Conversely, these experimental demonstrations have been performed in a variety of species in both in vivo and in vitro studies, which suggests that there may be a difference among various experimental models, and thus, clinical evaluation should be needed. These beneficial effects of potassium channel openers have not been applied in clinical settings, however. We designed a study to apply nicorandil to myocardial protection during open heart surgery and to further elucidate the myocardial protective effects of potassium channel openers that have been demonstrated in previous experimental studies.

METHODS

Study population

Seventy patients who underwent elective CABG under tepid-thermic cardiopulmonary bypass in our institution between 1997–1999 were enrolled in this study. Fifty-three were men and 17 were women, and their ages at the time of surgery ranged from 43–82 years, with a mean of 61.2 ± 9.5 years. None received either nitroglycerin or diltiazem hydrochloride from initial hospitalization until myocardial reperfusion during CABG. All patients gave informed consent to participate in this study, and we followed the guidelines of our internal review board.

This study was designed to elucidate the pharmacologic effects of nicorandil that had already been demonstrated in the experimental situations. Nicorandil administration in this study aimed at the following three points: (¹) from cardiopulmonary bypass initiation until aortic cross-clamping, (²) during the initial cardioplegia infusion, and (³) just after myocardial reperfusion. Patients were randomly divided into two groups. One group (group N, n = 35) received nicorandil during cardiopulmonary bypass; the other group (group C, n = 35) served as a control and did not receive nicorandil during cardiopulmonary bypass. Nicorandil (Sigmart, molecular weight = 211.18, Chugai Pharmaceutical Co., Tokyo, Japan) was administered at each dose of 0.1 mg/kg into the cardiopulmonary bypass circuit according to the following schedule: (¹) bolus injection at initiation of cardiopulmonary bypass, (²) continuous infusion for 5 min before aortic cross-clamping, (³) bolus administration at 5 min before reperfusion, and (⁴) continuous infusion for 5 min before reperfusion (Fig. 1). Patient characteristics are shown in Table 1 according to their distribution in the two groups. Both preoperative and perioperative variables are included.

FIG. 1.: The study protocol. Nicorandil was administered as follows: (¹) bolus injection at cardiopulmonary bypass (CPB) initiation; (²) continuous infusion for 5 min before aortic cross-clamping (ACC); (³) bolus administration at 5 min before reperfusion; and (⁴) continuous infusion for 5 min before reperfusion.

TABLE 1: Cardiopulmonary bypass data per study group

Operative management

The cardiopulmonary bypass circuit comprised a centrifugal pump, a membrane oxygenator, an arterial filter, a venous reservoir, and tubing lines, which were primed without blood components. The total volume of priming solution was approximately 1,000 ml. Components of the cardiopulmonary bypass circuit were not heparin coated. Heparin at a dose of 3 mg/kg was infused, and cardiopulmonary bypass was instituted in a routine fashion. A venous cannula was placed directly into the superior vena cava and another into the inferior vena cava, and venous drainage was carried out with a vacuum system. An arterial cannula was positioned directly into the ascending aorta.

Anesthesia was induced and maintained with diazepam (0.4 mg/kg), fentanyl (50 μg/kg), and inhaled isoflurane via endotracheal intubation. To reduce cardiopulmonary bypass-induced inflammatory response, nafamostat mesilate (FUT-175, Torii Pharmaceutical Co., Tokyo, Japan), which is a serine protease inhibitor, was added at a dose of 1 mg/kg into the venous reservoir at the initiation of cardiopulmonary bypass and was administered continuously during cardiopulmonary bypass at a dose of 0.5 mg/kg/h (²²).

Cardiopulmonary bypass was controlled on a basis of α-stat management with blood-flow rates of 2.2–2.6 l/min/m² to maintain mean arterial pressure between 60–80 mm Hg, using vasoactive agents such as chlorpromazine hydrochloride and norepinephrine if necessary. We measured the blood temperature in the arterial line just after it passed through the heat exchanger and controlled it at 34°C.

Oxygenated blood was obtained from the oxygenator reservoir and mixed with cold potassium crystalloid solution (500 ml of lactate Ringer's solution, 40 mEq/l of potassium chloride, and 5 ml of 2.2% sodium citrate buffer solution) [Na⁺, 30 mEq/l; K⁺, 40 mEq/l; Ca⁺⁺, 3 mEq/l] (²³). Cardiac arrest was achieved by aortic cross-clamping and an initial administration of 2:1 bloodcrystalloid cardioplegia solution containing a final concentration of 16 mEq/l potassium and 0.8–1.0 mEq/l calcium. The initial dose of cardioplegia solution was 10 ml/kg body weight. Cardioplegia solution was infused into the aortic root until cardiac arrest was confirmed on the electrocardiogram, at which point the remaining dose was infused into the coronary sinus. A 4:1 bloodcrystalloid cardioplegia solution at a dose of 5 ml/kg body weight was infused every 20 min through the coronary sinus for myocardial protection during ischemia. The temperature of the cardioplegia solution was measured in the cardioplegia line just after it passed through the heat exchanger and was maintained at 15–20°C. Myocardial temperature at the ventricular septum was monitored and maintained below 20°C by additional cardioplegia infusion. Terminal warm blood cardioplegia, with or without depletion of leukocytes, was not used for any study patient (²⁴).

The patients were rewarmed after aortic unclamping, and cardiopulmonary bypass was terminated when rectal temperature reached 35°C. Catecholamines were administered as needed on the basis of the patient's hemodynamic condition just after the termination of cardiopulmonary bypass. For patients showing marked ST segment changes on the electrocardiogram after myocardial reperfusion, 0.3 μg/kg per min of nitroglycerin and 1 μg/kg per min of diltiazem hydrochloride were administered immediately. Patients not showing marked ST changes were given nitroglycerin and diltiazem hydrochloride in these same doses after they entered the intensive care unit.

Evaluations

Significant ST changes on the electrocardiogram, previously described by Krumenacker and Roland (²⁵), were considered possible perioperative coronary spasms during cardiopulmonary bypass (at cardiopulmonary bypass initiation and after aortic unclamping). The time required to achieve cardioplegic arrest was used to assess the effect of the potassium channel opener nicorandil on hyperpolarization. Spontaneous recovery of heart beat after aortic unclamping was considered sufficient myocardial protection indicating no effect on the specialized conduction system.

Immediately after myocardial reperfusion, arterial blood and coronary sinus effluent were obtained to measure the difference in plasma levels of malondialdehyde as an index of myocardial lipid peroxidation (²⁶). Fifty minutes after reperfusion, arterial blood was obtained for measurement of plasma human-heart fatty acid-binding protein (HH-FABP), an early indicator of myocardial damage after cardiac surgical procedures (²⁷). Postoperatively, plasma creatine kinase-MB was measured every 6 h after reperfusion, the maximum value during the first 24 h postoperatively being considered the extent of myocardial damage, a generally used index (²⁸). The maximum dose of catecholamine (dopamine plus dobutamine; DOA+DOB) required at the time of weaning from cardiopulmonary bypass and during the postoperative course was also used as an index of myocardial damage. These values were compared between groups to evaluate the myocardial protective effect of nicorandil administered during cardiopulmonary bypass.

Statistical analysis

All data are expressed as mean ± SD. χ² test for independence and unpaired Student's t test were used to compare values between groups. Analyses were performed with the Statview v5.0 statistical package (Abacus Concepts Inc., Berkeley, CA, U.S.A.). A p value of < 0.05 was considered statistically significant.

RESULTS

All study patients tolerated the surgical procedures, and no mechanical circulatory assist devices such as intra-aortic balloon pumping or percutaneous cardiopulmonary support were needed for weaning from cardiopulmonary bypass or postoperatively. No complications related to nicorandil administration were observed, and all patients were discharged.

The time required for achieving cardiac arrest after the initial cardioplegia was significantly less in group N (42.3 ± 15.9 s) than in group C (72.7 ± 20.6 s, p < 0.01) (Fig. 2). No patients in either group showed a significant ST segment change on the electrocardiogram from cardiopulmonary bypass initiation to initial cardioplegia infusion. However, the number of patients showing significant ST changes after aortic unclamping was significantly smaller in group N (n = 3) than in group C (n = 12, p < 0.05) (Fig. 3A). The number of patients showing spontaneous recovery of heart beat after reperfusion was significantly greater in group N (n = 28) than in group C (n = 17, p < 0.05) (Fig. 3B).

FIG. 2.: The time required for achieving cardiac arrest after initial cardioplegia infusion.

FIG. 3. A: . Number of patients showing significant ST segment change on the electrocardiogram after aortic unclamping. B. Number of patients showing spontaneous recovery of heart beat after myocardial reperfusion.

The difference in malondialdehyde concentration between coronary sinus effluent and arterial blood immediately after reperfusion was significantly less in group N than in group C. HH-FABP 50 min after myocardial reperfusion, postoperative peak creatine kinase-MB, and maximum DOA + DOB dose were also significantly less in group N than in group C, as shown in Figure 4.

FIG. 4. A: . Coronary sinus effluent (CS)-arterial blood (Ao) difference in the plasma levels of malondialdehyde (MDA) immediately after myocardial reperfusion. Group N: 1.31 ± 0.50 μM; group C: 1.79 ± 0.72 μM; p <0.01. B. Plasma levels of human-heart fatty acid-binding protein (HH-FABP) 50 min after myocardial reperfusion. Group N: 87.2 ± 43.0 IU/L; group C: 133.7 ± 63.4 IU/L; p <0.01. C. Peak levels of plasma creatine kinase-MB (CK-MB) concentrations during the first 24 h postoperatively. Group N: 12.8 ± 7.2 IU/L; group C: 20.5 ± 12.1 IU/L; p < 0.01. D. Maximum doses of catecholamine (dopamine plus dobutamine) required at the time of weaning from cardiopulmonary bypass (CPB) and during postoperative course. Group N: 3.76 ± 1.73 μg/kg/min; group C: 5.44 ± 2.77 μg/kg/min; p <0.01. Data are expressed as mean ± SD. (^*p < 0.05 vs. group C, ^**p < 0.01 vs. group C).

DISCUSSION

Cardioplegic myocardial protection under cardiopulmonary bypass has been a standard method during open heart surgery. Several experimental studies have demonstrated that adjuvants to cardioplegia enhance myocardial protective effects (^29–32). The combination of potassium channel opener and high-concentration potassium cardioplegia may enhance myocardial protective effects in terms of pharmacologic preconditioning, rapid cardioplegic arrest, prevention of coronary spasms, and so on. We examined the myocardial protective effect of nicorandil based on clinically appropriate protocol. In comparison with the conventional methods of perioperative myocardial protection, nicorandil administration during CABG in this study (¹) shortened the time required to achieve cardioplegic arrest; (²) decreased the chance of arrhythmic events such as ventricular fibrillation after myocardial reperfusion; (³) prevented ST segment changes possibly induced by coronary spasms after reperfusion; (⁴) attenuated usual increases in plasma levels of malondialdehyde, HH-FABP, and creatine kinase- MB; and (⁵) reduced the catecholamine requirements at the weaning from cardiopulmonary bypass and during the postoperative course.

No patient in either group showed significant ST segment changes on the electrocardiogram at cardiopulmonary bypass initiation. During the preoperative period, no patients were given a vasodilating agent for this study, and it is possible that there were no patients with spastic coronary artery disease severe enough to produce coronary spasms by cardiopulmonary bypass initiation alone. Some patients, however, did show significant ST changes after aortic unclamping. Some chemotactic mediators cause vasoconstriction as a result of cardiopulmonary bypass-induced inflammatory response (³³) and are thought to affect coronary vascular tone after myocardial reperfusion. Our findings suggest that nicorandil was effective against spastic contraction enhanced by cardiopulmonary bypass-induced inflammatory response.

Several experimental studies have demonstrated that potassium channel opening induces hyperpolarization (2, 17,18). It follows, then, and has been demonstrated that nicorandil can be used as a substitute for high-concentration potassium. In the current study, nicorandil combined with hyperkalemic cardioplegia provided for rapid cardioplegic arrest in comparison to that achieved by conventional cardioplegia alone. When cardioplegia solution is infused, the time required for obtaining complete hyperpolarization is considered a crucial ischemic period influencing perioperative myocardial injury. From this standpoint, rapid cardioplegic arrest achieved by nicorandil administration may have an additional benefit for attenuating myocardial injury during CABG.

The relatively selective vasodilating effect of nicorandil likely contributes to increased coronary blood flow after aortic unclamping as well as to prevention of coronary spasms. Feldman et al. (³⁴) demonstrated that nitroglycerin alone did not significantly increase coronary blood flow, despite reduced systemic vascular resistance. Grover et al. (³⁵) demonstrated, however, that the myocardial protective effect of nicorandil was not attributable to an increase in coronary blood flow. We did not examine the relation between coronary blood flow and myocardial protective effects; further studies are needed to elucidate this issue.

The effective concentration of nicorandil remains controversial; the various experimental study protocols have differed in terms of the effective dose of nicorandil. According to the study protocol, the concentration of nicorandil was defined by circulatory blood volume including cardiopulmonary bypass priming solution. As a result, about 100 μM of nicorandil was included in cardiopulmonary bypass circulation volume, about 33% of which was included in the initial cardioplegia solution. This concentration is nearly the same as that in some previous studies. Although plasma concentrations of nicorandil were not measured in this study and the effect of subsequent nicorandil administrations on the change in plasma levels is unknown, the present nicorandil administration protocol appears effective in enhancing conventional cardioplegic myocardial protection.

Nicorandil has various pharmacologic benefits demonstrated in experimental models. We first demonstrated a clinically myocardial protective effect against myocardial injury during open-heart surgery, which does not elucidate the mechanism of nicorandil in the clinical settings because of the limitation of the current study design. During the CABG procedure, the pattern of coronary blood flow changes five times as follows: (¹) at the initiation of cardiopulmonary bypass; (²) during aortic cross-clamping and initial cardioplegia administration; (³) during intermittent cardioplegia administration during ischemia; (⁴) at aortic unclamping and myocardial reperfusion; and (⁵) after the termination of cardiopulmonary bypass. The nicorandil administration protocol of this study was mainly aimed at the following three periods: before ischemia under cardiopulmonary bypass, during initial cardioplegia until achieving cardiac arrest, and after reperfusion prior to weaning from cardiopulmonary bypass. Therefore, the greatest effective period of nicorandil administration has not been clarified. In particular, Imagawa et al. (¹³) reported in a rabbit infarct model that nicorandil administration after reperfusion did not demonstrate a significantly satisfactory protective effect against myocardial ischemia-reperfusion injury compared with its administration before the ischemic period. Further clinical studies into the most effective and simple means of nicorandil administration are essential.

In summary, this is the first clinical study demonstrating that administration of nicorandil during cardiopulmonary bypass provides enhanced myocardial protective effects in comparison with conventional cardioplegic myocardial protection. The study suggests that nicorandil can be used as an effective adjunct to current perioperative myocardial protection strategies. Although further clinical studies into the mechanism of the myocardial protection by nicorandil are needed, the results of this study suggest the potential benefits of nicorandil for improving myocardial tolerance to surgically induced cardioplegic ischemic arrest.

Acknowledgment: We gratefully thank Chugai Pharmaceutical Co. for their critical suggestions on this manuscript.