Prophylaxis of Atrial Fibrillation With Magnesium Sulfate After Coronary Artery Bypass Grafting William J. Fanning, MD, Clarence S. Thomas, Jr, MD, Albert Roach, PharmD, Richard Tomichek, MD, William C. Alford, MD, and William S. Stoney, Jr, MD Department of Cardiac and Thoracic Surgery, Vanderbilt University Medical Center and St. Thomas Hospital, Nashville, Tennessee

Ninety-nine consecutive consenting patients were prospectively entered into a randomized, double-blind, placebo-controlled trial to determine the efficacy of postoperative magnesium therapy on the incidence of cardiac arrhythmias after elective coronary artery bypass grafting. No patient had documented or suspected arrhythmias preoperatively. Forty-nine patients received 178 mEq of magnesium given over the first 4 postoperative days, and 50 patients received only placebo. The clinical characteristics of both groups were similar. The preoperative mean serum magnesium concentration was similar in both study (1.90 mEq/L) and placebo (1.90 mEq/L) groups. The mean postoperative serum magnesium concentration in study patients was significantly elevated over postoperative days 1 through 4 when compared with preoperative levels ( p < 0.001). The postoperative mean

S

upraventricular tachyarrhythmias occur commonly after coronary artery bypass grafting. A variety of pharmocologic agents have been used to prevent the occurrence of supraventricular tachyarrhythmias including propranolol [l],timolol [2], digoxin [3], and verapamil [4]. Magnesium administration has proved beneficial in preventing ventricular and supraventricular tachyarrhythmias after acute myocardial infarction [5-71, and although hypomagnesemia has been reported after cardiopulmonary bypass [%lo], there is little information regarding the use of magnesium to prevent supraventricular tachyarrhythmias after coronary artery bypass grafting. This study was undertaken to determine whether prophylactic postoperative magnesium administration can reduce the incidence of supraventricular tachyarrhythmias after coronary artery bypass grafting.

Material a n d Methods This study was approved by the Human Subjects Committee of St. Thomas Hospital, Nashville, Tennessee. After giving informed consent, 99 consecutive consenting patients undergoing elective first-time coronary artery bypass grafting were prospectively randomized in a double-blind fashion to receive postoperative magnesium (49 patients) or placebo (50 patients). Excluded from the Accepted for publication May 14, 1991.

Address reprint requests to Dr Fanning, 300 E Town St, 12th Floor, Columbus, OH 43215.

0 1991 by The Society of Thoracic Surgeons

serum magnesium concentration in control patients declined and remained significantly depressed through postoperative day 3 ( p < 0.001),but increased to preoperative levels by postoperative day 4. The mean serum magnesium concentration was significantly greater in the study patients as compared with the control patients over postoperative days 1 through 4 ( p < 0.001). Although there was no significant difference between groups with respect to episodes of ventricular arrhythmias, there was a significant decrease in the number of episodes of atrial fibrillation in the group receiving magnesium therapy ( p < 0.02). There were no recognized adverse effects of magnesium therapy. Prophylactic magnesium administration seems to lessen the incidence and severity of atrial fibrillation after coronary artery bypass grafting. (Ann Thoruc Surg 1991;52:529-33)

study were patients with a serum creatinine level greater than 25 mg/L (2.5 mgidL), those with a history of secondor third-degree heart block, patients with a permanent pacemaker, and those refusing to participate. Patients with any documented or suspected supraventricular or ventricular arrhythmias, including isolated atrial or ventricular premature depolarizations noted on preoperative surface electrocardiography, were excluded as well. All patients underwent left heart catheterization and selective coronary arteriography because of progressive angina pectoris and were referred for surgical myocardial revascularization. Patients requiring additional procedures, such as valve replacement or left ventricular aneurysmectomy, were excluded. Patients were randomly assigned to receive in a doubleblind fashion either (1)placebo group maintenance fluids (5% dextrose in water with 20 mEq of potassium chloride per liter) or (2) study group maintenance fluids (5% dextrose in water with 20 mEq of potassium chloride per liter plus 40 mEq of magnesium, as magnesium sulfate, per liter). These intravenous fluids were given as a continuous infusion at 100 mL/h over the first 24 hours postoperatively followed by 25 mL/h over postoperative hours 25 through 96. The study group, therefore, received 96 mEq of magnesium over the first 24 hours postoperatively (4 mEq/h) followed by 72 mEq over postoperative hours 25 though 96 (1 mEq/h). All patients had determination of serum magnesium and potassium levels preoperatively, immediately postop0003-4975/91/$3.50

530

FANNING ET AL MAGNESIUM AND CORONARY BYPASS

eratively, and on postoperative days 1 through 4. Supplemental potassium was given as needed to maintain a serum potassium level greater than 4 mEq/L. For the purposes of defining end points in this study, postoperative cardiac arrhythmias were defined as: (1) ventricular fibrillation, (2) ventricular tachycardia (greater than 3 premature ventricular contractions in a row at a rate greater than 100 beatslmin), (3) premature ventricular contractions greater than 5 per minute, multiformed, couplets, or R on T phenomenon, and (4) atrial fibrillation or flutter and supraventricular tachycardia at rates greater than 110 beats/min lasting greater than 30 seconds. Frequent premature atrial contractions were not treated unless deemed hemodynamically significant. Using these definitions, cardiac arrhythmias were treated under the direction of the attending surgeon. All patients continued the study protocol regardless of the arrhythmia noted or the treatment received. Although the patient's physicians were not blinded to the results of the serum magnesium testing, no patient received additional magnesium therapy. Each episode of arrhythmia was recorded, and for patients with a persistent arrhythmia lasting more than 24 hours, each new 24-hour period was recorded as an additional episode. All operations were performed using hypothermic cardiopulmonary bypass with a crystalloid, non-magnesium-containing prime and a membrane oxygenator. Multidose blood cardioplegia with topical hypothermia was used in 82 patients. The cardioplegic component contained no additional magnesium. In these 82 patients (39 study, 43 control), standard surgical techniques were used with all distal coronary anastomoses performed during a single period of aortic cross-clamping. Proximal vein graft anastomoses were performed with partial aortic occlusion during rewarming. The left internal thoracic artery was used as a bypass conduit in the majority of patients. Seventeen patients (10 study, 7 control) underwent myocardial revascularization using hypothermic intermittent ischemic arrest as described by Olinger [ll]. All patients underwent careful intraoperative and postoperative hemodynamic monitoring with arterial, central venous, and pulmonary artery pressure monitoring with thermodilution cardiac output determination. Atrial, atrioventricular sequential, or ventricular pacing in the early postoperative period was used as required for bradycardia. Patients receiving P-blockers had these medications continued until the day of operation. No patient received P-blockers, calcium channel blockers, or digitalis postoperatively unless a specific arrhythmia dictated their use. Lidocaine was not routinely used postoperatively unless specifically indicated for the treatment of a postoperative ventricular arrhythmia. Postoperative hypertension was managed with intravenous nitroglycerin or nitroprusside, or both. Inotropic or pressor support was used to maintain a mean arterial pressure greater than 70 mm Hg and a cardiac index greater than 2 L * min-'

. m-2

Apart from the study protocol involving the magnesium infusion or placebo, each patient's postoperative management was directed by a single group of cardiovas-

Ann Thorac Surg 1991;52:529-33

Table 1. Patient Characteristics" Variable

N Sex (M/F) Age (Y) History of MI p-Blockers Ca2+ Blockers Diuretics Digitalis LVEF XC time (min) No. of grafts a

Study Group

Control Group

49

50 39111

35114 59 (4S75)

62 (42-79)

0 0.50 (0.3M.70) 66 (2&167)

21 13 22 10 7 0.49 (0.25-0.70) 66 (1&135)

4.0 (2-7)

3.6 (1-7)

17 13 24 5

Values in parentheses are ranges

LVEF = left ventricular ejection fraction; XC = aortic cross-clamp.

MI

=

myocardial infarction;

cular surgeons and anesthesiologists at this institution. All patients underwent continuous electrocardiographic monitoring through postoperative day 4 using the Mennen Horizon 2000 Bedside Arrhythmia Detection System (Mennen Medical Co, Clarence, NY). At the time of termination of the study, the code was broken and the patient data were assembled into placebo and magnesium groups. The incidence of postoperative arrhythmias was analyzed for group differences by using a two-sided x2 analysis. The mean serum magnesium levels (both within and between the control and study groups) and the mean number of episodes of arrhythmias between groups were analyzed using a paired or two-sample t test where appropriate.

Results Patient demographics are assembled in Table 1. The two groups were well matched with respect to age, sex, and history of prior myocardial infarction. There was no significant difference between groups with respect to P-blocking or calcium channel blocking medication, although preoperative digoxin and diuretic use was more common in the control group. Left ventricular systolic function was similar between groups as were operative variables, including ischemic time, number of grafts, and the method of myocardial protection. There was one death in the study period occurring in a 79-year-old man 4 days after two-vessel coronary bypass grafting due to recurrent ventricular tachycardia degenerating to asystole from which he could not be resuscitated. This patient was in the control group. Serum magnesium level in this patient on the day of death was 2.0 mEqIL. Permission for autopsy was denied. There were four postoperative myocardial infarctions as determined by new Q waves, compatible clinical course, and increase in creatine kinase-MB level, with two in each group. Seventeen of 49 study and 27 of 50 control patients received

531

FANNING ET AL MAGNESIUM AND CORONARY BYPASS

Ann Thorac Surg 1991;52:529-33

Table 2. Mean Serum Magnesium Levelf Postoperative Day Group

Preop

0

1

2

3

4

Study

1.90 2 0.03 1.90 2 0.04

1.27 2 0.03 1.17 2 0.03

2.44 2 0.05 1.25 -+ 0.03

2.30 2 0.05 1.44 2 0.04

2.20 C 0.04 1.75 2 0.04

2.08 2 0.05 1.85 2 0.04

Control a

Values are expressed in milliequivalents per liter

?

standard error of the mean

inotropic or pressor support postoperatively to maintain a mean arterial pressure greater than 70 mm Hg and a cardiac index greater than 2 L min-' m-' during the first 18 postoperative hours. Forty study and 31 control patients received nitroprusside for hypertension. Intravenous nitroglycerin was used routinely in both groups. The mean serum magnesium concentrations for the two groups are shown in Table 2. A comparison of the changes in serum magnesium concentration over time for the two groups is shown in Figure 1. There was no difference between groups in the preoperative mean serum magnesium concentration. In both the study and control group, there was a significant decline in the mean serum magnesium concentration immediately after coronary artery bypass grafting ( p < 0.001). Compared with preoperative levels, the mean serum magnesium concentration was significantly elevated in study patients on postoperative days 1 through 4 ( p < 0.001). In control patients, the mean serum magnesium concentration remained depressed as compared with preoperative levels on postoperative days 1 through 3 ( p < 0.001) but returned to preoperative levels by postoperative day 4. The mean serum magnesium concentration in study patients was significantly greater than that of controls on postoperative days 1 through 4 ( p < 0.001). The episodes of the various dysrhythmias are shown in Figure 2. There were no significant differences between groups for the various ventricular arrhythmias. The difference between groups with respect to the numbers of patients with atrial fibrillation did not reach statistical significance; however, significantly fewer episodes of atrial fibrillation occurred in the group receiving supplemental magnesium. Seven patients had 12 episodes of atrial fibrillation in the study group as compared with 14

control patients experiencing 42 episodes ( p < 0.02). Nine of these 14 control patients experienced persistent or recurrent atrial fibrillation requiring multiple pharmacologic interventions, and 2 of these 9 required synchronized direct-current cardioversion because of hemodynamic embarrassment associated with the atrial fibrillation. All 9 were eventually discharged receiving oral digoxin. Additionally, 7 required oral verapamil and 3 patients required oral quinidine for control of the atrial fibrillation. By comparison, only 2 patients in the study group experienced more than one episode of atrial fibrillation. No patient in the study group required cardioversion. Two of 17 patients undergoing coronary revascularization using intermittent ischemic arrest experienced atrial fibrillation (both in the control group). Four of 7 control patients receiving preoperative digoxin experienced atrial fibrillation. The number of patients receiving preoperative p-blockers in whom postoperative atrial fibrillation developed was similar in both groups (4 study, 3 control). No side effects could be attributed to the magnesium therapy. As noted previously, atrial and ventricular pacing was used as required and no patient in this study had artificial pacing requirements beyond 36 hours. There was no difference in pacing requirements between the two groups. The study code was not broken throughout the study.

Comment Magnesium is the second most abundant intracellular cation and is an essential cofactor of the myocardial cell

Mg++Concentration

3

10

dlop

0

0.5

PAOP

I

I

I

I

1

2

3

4

Days

Fig 1 . Magnesium concentration (mEqIL). (Solid line = study patients; dotted line = control patients )

V Fib

V Tach PVC s Atrial Dysrhythmia Study Group 0Control Group

Fig 2 Episodes of dysrhythmia after coronary artery bypass grafting (Atrial = atrial fibrillation, PVCs = premature ventricular contractions; V Fib = ventricular fibrillation, V Tach = ventricular tachycardia.)

532

FANNING ET AL MAGNESIUM AND CORONARY BYPASS

membrane associated enzyme sodium-potassium adenosinetriphosphatase, which maintains normal transmembrane sodium and potassium gradients. Magnesium deficiency may adversely affect this ionic pump, impairing the ability of the myocardial cell to maintain normal intracellular potassium concentrations, leading to alterations in membrane potential, potassium conductance, and repolarization, and may thereby be potentially arrhythmogenic [12]. Hypomagnesemia is seen in a variety of clinical states and is commonly noted after cardiopulmonary bypass [%lo]. Although in vitro data exist to support a potential arrhythmogenic influence of hypomagnesemia [13], in vivo evidence is inconclusive and based on testimonials to arrhythmia suppression by the administration of magnesium salts [14]. This arrhythmia suppression with magnesium is, however, nonspecific and does not prove that hypomagnesemia induced the arrhythmia. Nonetheless, numerous uncontrolled observations do support the contention that exogenous magnesium suppresses a variety of tachyarrhythmias [15-191. In humans it has been difficult to separate the effects of magnesium deficiency from other associated electrolyte abnormalities, especially potassium deficiency. As a result, much of our understanding of the cardiovascular effects of magnesium are derived from in vitro observations and the in vivo effects of exogenous magnesium administration. Detailed electrophysiologic studies in humans [20-231 have examined the effects of magnesium on cardiac conduction. In these studies, patients have differed in baseline conduction abnormalities, presence of discernible heart disease, and degree of induced hypermagnesemia. Magnesium therapy did, however, induce consistent prolongation of atrioventricular conduction with PR interval prolongation seen on the surface electrocardiogram and AH interval prolongation detected by intracardiac electrograms. In general, there was little effect on ventricular conduction or refractoriness. These studies provide some rationale for the potential effectiveness of magnesium in the treatment of some cardiac arrhythmias, particularly supraventricular tachyarrhythmias. Most reports attesting to the effectiveness of magnesium therapy in the treatment of cardiac arrhythmias are case reports or collections of case reports. Recently, however, a few controlled studies have shown a decrease in tachyarrhythmias after acute myocardial infarction in patients treated with magnesium as compared with controls ~ 7 1 . These observations in the reported literature and the frequent development of supraventricular tachyarrhythmias and hypomagnesemia after coronary artery bypass prompted this study. The results of this study confirm previous observations of significant hypomagnesemia after cardiopulmonary bypass [%lo]. All patients experienced a significant decrease in serum magnesium level, which is probably explained by hemodilution and possibly a P-receptor-mediated phenomenon secondary to elevated epinephrine levels commonly noted after cardiopulmonary bypass [13,24]. In the control group, the mean serum magnesium decreased to its nadir immediately

Ann Thorac Surg 1991;52:529-33

postoperatively but then gradually increased to preoperative levels by the fourth postoperative day. By this protocol, the mean serum magnesium concentration in the study group was significantly elevated over the first 4 postoperative days. No patient had a serum magnesium concentration greater than 3.8 mEq/L, and there were no recognized adverse effects to the drug. Although there was no significant difference between groups with respect to the numbers of patients with atrial or ventricular tachyarrhythmias, there was a significant decrease in the number of episodes of atrial fibrillation in the study group. In addition, these dysrhythmias, when they occurred in the study group, were well tolerated and more easily treated as reflected by the more frequent requirement for cardioversion and multiple drug interventions in the control group. There was, however, no demonstrable difference in the duration of hospital stay between groups reflecting the usually benign nature of supraventricular arrhythmias after coronary artery bypass grafting [25]. Interestingly, the 7 control patients receiving preoperative digoxin were afforded no protection from postoperative supraventricular arrhythmias, as 4 of 7 patients experienced 18 episodes of atrial fibrillation. Abrupt discontinuation of P-blockers in the postoperative period did not seem to influence these results. Four of 13 study patients and 3 of 13 control patients receiving preoperative P-blockers had development of atrial fibrillation in the postoperative period (seven episodes in each group). Likewise, the method of myocardial protection did not seem to influence these results. Although conceivably atrial anoxia during ischemic arrest periods could potentiate postoperative supraventricular arrhythmias, this was not observed. Only 2 of 17 patients (both control) had postoperative atrial fibrillation. Others have noted that the incidence of atrial arrhythmias is similar whether the method of myocardial protection is cold potassium cardioplegia or intermittent aortic occlusion [26]. Although the monitoring system used was not a Holter tape recording system, it automatically records bradyarrhythmias or tachyarrhythmias, allowing analysis of the arrhythmia. It is unlikely that any arrhythmias as defined by the end points in this study were missed. When this study was initiated there were no reports on the use of magnesium in patients after cardiopulmonary bypass. Extrapolating from the existing literature, this study was based on the protocol used by Rasmussen and associates [7], which demonstrated a reduced incidence of supraventricular arrhythmias in patients with acute myocardial infarction. Admittedly this protocol is somewhat cumbersome, requiring a 4-day continuous infusion, and clearly further studies should be done both to confirm these results and to simplify the method of magnesium administration. The observation of increased requirements for inotropes and vasopressors in control patients cannot be easily explained. The patient groups were similar with respect to measurable variables, the study was blinded, and postoperative hemodynamic management was directed by a single group of cardiovascular surgeons and

Ann Thorac Surg 1991;52:529-33

anesthesiologists with a consistent philosophy regarding the use of inotropes and vasopressors. Conceivably the decreased incidence of postoperative supraventricular arrhythmias attributed to magnesium replacement could be related to increased catecholamine use in the control group. Alternatively, the decreased requirements for inotropes in the study group could be related by some undefined mechanism to magnesium administration. These two alternative explanations are speculative and should be addressed by further study. It should be stressed that the observations in this study should not be overinterpreted. They do not prove that hypomagnesemia is the cause of supraventricular arrhythmias after coronary artery bypass grafting, but only that supplemental magnesium seems to lessen their incidence and severity. Additional studies are needed to confirm these observations and, more importantly, to establish the mechanism of the antiarrhythmic properties of magnesium.

References 1. Hammon JW Jr, Wood AJJ, Prager RL, Wood M, Muirhead J, Bender HW Jr. Perioperative beta blockade with propranolol: reduction in myocardial oxygen demands and incidence of atrial and ventricular arrhythmias. Ann Thorac Surg 1984;38: 363-7. 2. White HD, Antman EM, Glynn MA, et al. Efficacy and safety of timolol for prevention of supraventricular tachyarrhythmias after coronary artery bypass surgery. Circulation 1984; 70:479-84. 3. Csicsko JF, Schatzlein MH, King RD. Immediate postoperative digitalization in the prophylaxis of supraventricular arrhythmias following coronary artery bypass. J Thorac Cardiovasc Surg 1981;81:419-22. 4. Davison R, Hartz R, Kaplan K, Parker M, Feiereisel P, Michaelis L. Prophylaxis of supraventricular tachyarrhythmia after coronary artery bypass surgery with oral verapamil: a randomized, double-blind trial. Ann Thorac Surg 1985;39: 36&9. 5. Abraham AS, Rosenmann D, Kramer M, et al. Magnesium in the prevention of lethal arrhythmias in acute myocardial infarction. Arch Intern Med 1987;147:75>5. 6. Morton BC, Nair RC, Smith FM, et al. Magnesium therapy in acute myocardial infarction: a double blind study. Magnesium 1984;3:345-52. 7. Rasmussen MS, Suenson M, McNair P, Norregard P, Balslev S. Magnesium infusion reduces the incidence of arrhythmias in acute myocardial infarction. A double blind placebo controlled study. Clin Cardiol 1987;10:351-6. 8. Scheinman MM, Sullivan RW, Hyatt KN. Magnesium metabolism in patients undergoing cardiopulmonary bypass. Circulation 1969;39(Suppl 1):23541.

FANNING ET AL MAGNESIUM AND CORONARY BYPASS

533

9. Holden MP, Ionescu MI, Wooler GN. Magnesium in patients undergoing open heart surgery. Thorax 1972;27:212-8. 10. Khan RMA, Hodge JS, Bassett NF. Magnesium in open heart surgery. J Thorac Cardiovasc Surg 1973;66:185-91. 11. Olinger G. Hypothermic intermittent ischemic arrest in myocardial preservation: clinical application. Cardiac Surg State Art Rev 1988;2:155-66. 12. Eisenberg MJ. Magesium deficiency and cardiac arrhythmias. N Y State J Med 1986;86:1334. 13. Zwerling HK. Does exogenous magnesium suppress myocardial irritability and tachyarrhythmias in the nondigitalized patient? Am Heart J 1987;113:104&53. 14. Surawicz B. Is hypomagnesia or magnesium deficiency arrhythmogenic? J Am Coll Cardiol 1989;14:109M. 15. Boyd LJ, Scherf D. Magnesium sulfate in paroxysmal tachycardia. Am J Med Sci 1943;206:4>8. 16. Chadda KD, Lichstein E, Gupta P. Hypomagnesia and refractory cardiac arrhythmia in a nondigitalized patient. Am J Cardiol 1973;31:98-100. 17. Iseri LT, Chung P, Tobias J. Magnesium therapy for intractable ventricular tachyarrhythmias in normomagnesemic patients. West J Med 1983;138:823-8. 18. Tzivoni D, Banai S, Schuger C, et al. Treatment of torsade de points with magnesium sulfate. Circulation 1988;77:392-7. 19. Allen BJ, Brodsky MA, Capparelli EV, Luckett CR, Iseri LT. Magnesium sulfate therapy for sustained monomorphic ventricular tachycardia. Am J Cardiol 1989;64:1202-4. 20. Rasmussen HS, Thomsen PEB. The electrophysiological effects of intravenous magnesium on human sinus node, atrioventricular node, atrium and ventricle. Clin Cardiol 1989;12:85-90. 21. Dicarlo LA, Morady F, deBuitleir M, Krol RB, Schurig Annesley TM. Effects of magnesium sulfate on cardiac conduction and refractoriness in humans. J Am Coll Cardiol 1986;7: 135U2. 22. Sager PT, Widerhorn J, Petersen R, et al. Prospective evaluation of parenteral magnesium sulfate in the treatment of patients with reentrant AV supraventricular tachycardia. Am Heart J 1990;119:308-16. 23. Kulich DL, Hong R, Ryzen E, et al. Electrophysiologic effects of intravenous magnesium in patients with normal conduction systems and no clinical evidence of significant cardiac disease. Am Heart J 1988;115:367-73. 24. Reves IF, Karp RB, Buttner EE, et al. Neuronal and adrenomedullary catecholamine release in response to cardiopulmonary bypass in man. Circulation 1982;66:49-55. 25. Rubin DA, Nieminski KE, Reed GE, Herman MV. Predictors, prevention and long term prognosis of atrial fibrillation after coronary artery bypass graft operations. J Thorac Cardiovasc Surg 1987;94:331-5. 26. Pattison CW, Dimitri WR, Williams BT. Dysrhythmias following coronary artery surgery. A comparison between cold cardioplegic and intermittent ischemic arrest (32°C) with the effect of right coronary endarterectomy. J Cardiovasc Surg 1988;29:601-5.

Prophylaxis of atrial fibrillation with magnesium sulfate after coronary artery bypass grafting.

Ninety-nine consecutive consenting patients were prospectively entered into a randomized, double-blind, placebo-controlled trial to determine the effi...
924KB Sizes 0 Downloads 0 Views