- Email: [email protected]
Postoperative atrial tachyarrhythmias in patients undergoing coronary artery bypass graft surgery without cardiopulmonary bypass: A role for intraoperative magnesium supplementation
Postoperative Atrial Tachyarrhythmias in Patients Undergoing Coronary Artery Bypass Graft Surgery Without Cardiopulmonary Bypass: A Role for Intraoperative Magnesium Supplementation Andrew D. Maslow, MD, Meredith M. Regan, ScD, Stephanie Heindle, MD, Peter Panzica, MD, William E. Cohn, MD, and Robert G. Johnson, MD Objective: To determine if intraoperative magnesium supplementation would be associated with a reduction in postoperative atrial tachyarrhythmias (POAT) in patients undergoing coronary artery bypass grafting (CABG) surgery without cardiopulmonary bypass (off-pump CABG surgery). Design: Retrospective study. Setting: University Medical Center. Participants: Patients who had undergone off-pump CABG surgery (n ⴝ 124). Interventions: None. Measurements and Main Results: The charts of 124 patients who had undergone off-pump CABG surgery (64 by anterior thoracotomy and 60 by median sternotomy) were retrospectively reviewed. Demographic data and perioperative care were recorded and compared among patients who did and did not experience POAT and among patients who did and did not receive intraoperative magnesium supplementation. Logistic regression analysis was used to assess the association between magnesium supplementation and incidence of POAT, controlling for other covariables. Of the 124 patients, 16 had a prior history of atrial or ventricular arrhythmias and/or were receiving antiarrhythmic medications. Medical records of the remaining 108 patients were
reviewed. Twenty-four patients (22%) had POAT. Forty-two patients (39%) received intraoperative magnesium. In patients receiving intraoperative magnesium, the incidence of POAT was significantly decreased (12% v 29%; p ⴝ 0.03). In these patients, initial postoperative serum magnesium was significantly higher (2.37 mEq/L v 1.86 mEq/L; p F 0.01). In patients not receiving intraoperative magnesium, 35% had hypomagnesemia (serum magnesium F 1.8 mEq/L) compared with 9% of patients receiving magnesium ( p F 0.01). Patients who received intraoperative magnesium and -adrenergic blockers had a lower incidence of POAT (5%) than patients who received only one (19%) or neither (33%) (p F 0.05). Conclusions: Intraoperative magnesium supplementation is associated with a decrease in POAT after off-pump CABG surgery. The combination of a -blocker and magnesium may reduce POAT further. It is recommended that intraoperative magnesium supplementation be part of the care of patients undergoing off-pump CABG surgery. Copyright r 2000 by W.B. Saunders Company
P
complications of CPB. Several surgical approaches have been described. One approach accomplishes CPB surgery through a small anterior thoracotomy (minimally invasive direct [MID] CABG surgery). A second surgical approach is by a median sternotomy incision (off-pump [OP] CABG sternotomy). Although there are a variety of approaches to OP revascularization, they all have in common the avoidance of cardiac cannulation, CPB, and cardioplegic arrest. Despite the avoidance of CPB, the incidence of POAT in OP CABG surgery, regardless of surgical approach, remains substantial and not significantly different from conventional CABG, when patient populations are matched for a variety of predictors of POAT, including age.13-16 The benefits of magnesium administration to reduce POAT in OP CABG surgery patients is unexplored. To examine whether prophylactic magnesium might reduce the incidence of POAT after OP CABG surgery, the authors retrospectively examined the incidence of early POAT (within 7 days of surgery) in patients after MID CABG surgery or OP CABG sternotomy.
OSTOPERATIVE ATRIAL TACHYARRHYTHMIAS (POAT) (atrial fibrillation, atrial flutter, supraventricular tachycardia) occur in 20% to 30% of patients undergoing conventional coronary artery bypass graft surgery (CABG) (cardiopulmonary bypass [CPB] with cardioplegic arrest).1-4 POAT is a cause of and contributes to significant morbidity, including neurologic injury and hypotension.2-4 POAT is associated with increased intensive care unit (ICU) and hospital stays and increased health care costs in this population.5 Several investigations have shown a benefit when magnesium sulfate is used for prophylaxis against POAT.6-11 Studies have shown that perioperative serum magnesium levels may be reduced in 70% of patients undergoing conventional CABG surgery using hypothermic CPB.12 It is postulated that decreased magnesium stores may be associated with increased incidence of atrial and ventricular tachycardias12 and that prophylactic administration of magnesium may be effective in reducing the incidence of tachyarrhythmias after cardiac surgery.6-11 CABG surgery without the use of CPB and cardioplegic arrest has become an increasingly popular way to avoid
From the Department of Anesthesia, Rhode Island Hospital, Providence, RI; and Biometrics Center, and Departments of Anesthesia and Critical Care and Surgery, Beth Israel Deaconess Medical Center, Boston, MA. Address reprint requests to Andrew D. Maslow, MD, Department of Anesthesia, Rhode Island Hospital, Davol Building, Room 128, 593 Eddy St, Providence, RI 02903. Copyright r 2000 by W.B. Saunders Company 1053-0770/00/1405-0006$10.00/0 doi:10.1053/jcan.2000.9485 524
KEY WORDS: arrhythmias, magnesium, cardiac surgery, coronary artery surgery (off-pump)
METHODS The authors sought to determine whether prophylactic magnesium might reduce the incidence of POAT after OP CABG surgery by median sternotomy or MID CABG surgery. According to the literature, the incidence of POAT does not appear to be different for the 2 surgical approaches.13-16 Preliminary analysis in this study showed the incidence to be similar. Both groups were combined to form one group labeled OP CABG (off-pump CABG surgery). The hospital course of 124 consecutive patients who had undergone nonemergent OP CABG surgery at this institution between 1997 and 1998 was retrospectively examined; this includes the first 64 patients who underwent MID CABG surgery and first 60 patients who underwent OP CABG sternotomy. Data were collected from medical record
Journal of Cardiothoracic and Vascular Anesthesia, Vol 14, No 5 (October), 2000: pp 524-530
MAGNESIUM IN OFF-PUMP CABG
review. Exclusion criteria for analysis included a history of atrial tachyarrhythmias and all preoperative antiarrhythmic medication for ventricular tachycardias (eg, amiodarone, procainamide, disopyramide) with the exception of -adrenergic antagonists and calcium channel blockers taken for control of angina or hypertension. Because digoxin has not been shown to affect the incidence of POAT after cardiac surgery, the use of digoxin in patients with reduced systolic function was not considered an exclusionary criterion as long as there was no history of atrial tachyarrhythmia. A total of 108 patients met the criteria for evaluation. Demographic variables collected included age, sex, ejection fraction, history of myocardial infarction, distribution of coronary artery disease, coexisting diseases, preoperative cardiac medications, and previous revascularization procedure. Intraoperative data collected included invasive monitoring, anesthetic technique, use and dosage of magnesium, use of -adrenergic blockers, use of calcium channel blockers, and the surgical procedure. Additional perioperative data included preoperative hematocrit, hematocrit on admission to the ICU, need for perioperative blood transfusion, serum magnesium and potassium levels on admission to the ICU, serum magnesium level at the time POAT was diagnosed, time of extubation, duration of ICU, and duration of hospital stay. In particular, it was recorded whether an early (within 7 days of surgery) POAT occurred. The timing of POAT and the magnesium level before treatment were recorded. POAT was defined as a documented nonsinus rhythm originating from the atrial tissue that lasted ⬎10 minutes or required pharmacologic or electric (cardioversion) therapy. Patient characteristics for the entire cohort of 108 patients are summarized as mean ⫾ standard deviation or number (and percent) of patients. A preliminary analysis examined whether the data analysis should be conducted separately by type of surgery or could be combined across the 2 types of surgery. Important variables, such as incidence of POAT, number of patients receiving magnesium supplementation, and median dose, were comparable between surgeries. Patient characteristics were summarized and compared between patients who did and did not experience POAT. Categoric variables were compared using chi-square or Fisher’s exact test, and continuous variables were compared using a two-sample t-test or Wilcoxon rank sum test, as appropriate. Similarly, characteristics of patients who did and did not receive intraoperative magnesium were compared. A p value ⱕ0.05 was considered statistically significant. The study examined whether the incidence of POAT was decreased in patients who received intraoperative magnesium, while controlling for demographic and intraoperative covariables, including type of surgery, using multiple logistic regression. Demographic and intraoperative variables that tended to be related (ie, p ⬍ 0.25) to the incidence of POAT or use of magnesium in univariate analysis were considered as potential covariates. A forward stepwise selection technique was used, and variables that contributed significantly to the model ( p ⱕ 0.05) were kept. RESULTS
Between 1997 and 1998, 124 patients had one of the CABG procedures; 64 underwent OP CABG surgery by an anterior thoracotomy incision (MID CABG), and 60 underwent OP CABG surgery by median sternotomy (OP CABG sternotomy). Sixteen patients were excluded from the final analysis because of a history of atrial tachyarrhythmia or ventricular arrthythmia requiring treatment with amiodarone, procainamide, or disopyramide. Six patients had been receiving digoxin before surgery because of a history of congestive heart failure. Of these 6, 3 underwent MID CABG surgery (left ventricular ejection fraction, 15%, 50%, and 55%) and 3 underwent OP CABG
525
sternotomy (left ventricular ejection fraction, 25%, 30%, and 40%). Two patients received magnesium (4 g each). None of the 6 patients had POAT. These patients were included in the data analysis because there was no preoperative history of arrhythmias. The anesthetic technique consisted of an induction agent (thiopental, propofol, or etomidate), maintenance with a fentanyl analog (fentanyl or sufentanil), isoflurane, and muscle relaxation (vecuronium or pancuronium). Of patients, 81% were extubated on the same day of surgery. The mean hospital stay was 5 days. These data are summarized in Table 1. Of 108 patients remaining for this study, 24 (22%) experienced POAT. Forty-two patients (39%) received intraoperative magnesium, with a median dose of 5 g of intravenous magnesium. Magnesium was administered before manipulation of the native coronary arteries in all patients. Forty patients (37%) received intraoperative -adrenergic blockers. Of these, 16 received esmolol, 22 received metoprolol, and 2 received labetalol. -Adrenergic blockers were administered for elevated heart rates (⬎95 beats/min) or if coronary ischemia was suspected. When the data from the 2 surgical techniques (MID CABG v OP CABG) were compared, there were no significant differences seen with respect to postoperative serum magnesium levels (2.13 mEq/L v 2.02 mEq/L), administration of intraoperative magnesium (42% v 36%), timing of POAT (2.1 v 2.1 days), or incidence of POAT, whether supplementation was administered or not (MID CABG, 13/55 [24%] v OP CABG 11/53 [21%]). Although there were differences seen in left ventricular ejection fraction (54% v 49%) and number of vessels bypassed (1.0 v 2.4), these differences were not found to be relevant to the occurrence of POAT. The data for the 2 surgical types are combined and shown in Table 1. For the study population, the preoperative and postoperative hematocrit values were 39% (4.9%) and 33% (4.4%). Approximately 30% of patients received packed red blood cells during the first 3 postoperative days. Patients undergoing OP CABG received an average 640 mL (⫾380 mL) of cell-saved blood. Table 2 displays data comparing patients who experienced POAT with patients who did not. Preoperative demographics, anesthetic technique, and surgical procedure were not different between the 2 groups. POAT occurred on postoperative day 2 in 62.5% of patients and on postoperative days 1 and 3 in 16.7% each. Of 24 patients experiencing POAT, 5 received magnesium (21%) compared with 37 of 84 whose hospital course was not complicated by POAT (44%). Of 24 patients (21%) experiencing POAT, 5 received intraoperative -blockers compared with 35 of 84 whose postoperative course was not complicated by POAT (41%; p ⫽ 0.09). Patients who experienced POAT tended to be older (68.0 v 62.5 years; p ⫽ 0.06). There was no statistically significant difference in initial postoperative magnesium or potassium levels between patients with and patients without POAT. Patients with POAT had significantly longer hospital stays (7.5 v 4.3 days; p ⫽ 0.0004). Preoperative and postoperative hematocrit values, transfusion requirements, and cell-saved volume were similar between the 2 groups. Table 3 compares data between patients who did and patients who did not receive intraoperative magnesium. Preoperative demographics, anesthetic technique, and surgical procedure
526
MASLOW ET AL
Table 1. Summary of Patient Data Preoperative Variable
Mean age (y) ⫾ SD Sex Male Female Ejection fraction (%) (⫾SD) 40% Previous procedure PTCA CABG (conventional) Coronary lesions (stenosis ⱖ70%) Left anterior descending artery Left circumflex artery Right coronary artery Posterior descending artery Preoperative medicines -blocker Calcium channel blocker Digoxin Diuretics ACE inhibitors Coexisting disease Myocardial infarction Congestive heart failure Hypertension Tobacco/COPD Renal insufficiency (Cr ⱖ 2 mg/dL) Surgical procedure MID CABG OP CABG sternotomy No. of bypasses (mean grafts ⫾ SD) ⱖ3 bypasses Other intraoperative medicines -blocker Calcium channel blocker Intraoperative magnesium Yes No Dose of magnesium (g) (SD) Median Mean Time of extubation Operating room ICU (same day) Postoperative day 1 Other Mean ICU stay (d) (SD) % ⬎1 day Mean hospital stay (d) (SD) Atrial fibrillation
63.7 (⫾12.7) 73 (67.6%) 35 (32.4%) 51.4 (⫾11.6) 21 (19.4%) 36 (34.3%) 10 (9.3%) 103 (95.4%) 60 (55.6%) 22 (20.4%) 40 (37.0%) 97 (89.8%) 43 (39.8%) 6 (5.6%) 22 (20.4%) 21 (19.4%) 41 (38.0%) 12 (11.1%) 54 (50.0%) 45 (41.7%) 6 (5.5%) 55 (50.9%) 53 (49.1%) 1.9 (1.2) 34 (31.5%) 40 (37.0%) 2 (1.9%) 42 (38.9%) 66 (61.1%) 5 6.5 (2.3) 39 (36.1%) 49 (45.4%) 19 (17.6%) 1 (0.9%) 1.3 (⫾1.4) 10 (9.3%) 5.0 (⫾4.1) 24 (22.2%)
NOTE. All data are presented as absolute number of patients and percent of group (n ⫽ 108) unless otherwise stated. Abbreviations: SD, standard deviation; PTCA, percutaneous transluminal coronary angioplasty; CABG, coronary artery bypass graft; ACE, angiotensin-converting enzyme; COPD, chronic obstructive pulmonary disease; Cr, creatinine; MID, minimally invasive direct; OP, off-pump; ICU, intensive care unit.
were not different between the 2 groups. Of the 108 patients, 42 (39%) received intraoperative magnesium. The initial postoperative serum magnesium level was greater in patients receiving magnesium (2.4 mEq/L v 1.9 mEq/L; p ⬍ 0.05). In patients who did not receive intraoperative magnesium, the incidence of
postoperative hypomagnesemia (⬍1.8 mEq/L) was 35%, compared with 9% in patients who received intraoperative magnesium. The timing of POAT was similar whether patients received intraoperative magnesium or not (2.0 v 2.1 days). Table 2. Data Comparing Patients Experiencing Atrial Fibrillation With Patients Who Did Not Preoperative Variable
Mean age (y) (⫾SD) Sex: male Mean LVEF (%) (⫾SD) ⱕ40% Previous procedure PTCA CABG Coronary lesions Left anterior descending artery Left circumflex artery Right coronary artery/patent ductus arteriosus Preoperative medicines -blocker Calcium channel blocker Digoxin Diuretics ACE inhibitors Coexisting disease Myocardial infarction Congestive heart failure Hypertension Tobacco/COPD Renal disease Surgical procedure MID CABG OP CABG sternotomy No. bypasses performed Intraoperative medicines -blockers Calcium channel blockers Intraoperative magnesium Electrolyte levels (mEq/L) (SD) Postoperative magnesium Postoperative potassium Magnesium at POAT Time of extubation Operating room Same day ICU Postoperative day 1 Other Mean ICU stay (d) (SD) Mean hospital stay (d) (SD)
POAT (n ⫽ 24)
68.0 (14.4) 15 (62.5%) 52.7 (8.9%)
No POAT (n ⫽ 84)
62.5 (12.0)* 58 (69.0%) 51.0 (12.3%) 15 (17.8%)
3 (12.5%) 7 (29.2%)
29 (34.5%) 9 (10.7%)
1 (4.2%) 22 (91.7%)
81 (96.4%) 45 (53.6%)
15 (62.5%)
45 (53.6%)
13 (54.2%) 22 (91.7%) 12 (50.0%) 0 4 (16.7%)
75 (89.3%) 31 (36.9%) 6 (7.1%) 18 (21.4%) 18 (21.4%)
3 (12.5%) 9 (37.5%) 2 (8.4%) 11 (45.8%) 11 (45.8%)
32 (38.1%) 10 (11.9%) 43 (51.2%) 34 (40.5%) 3 (3.6%)
13 (54.2%) 11 (45.8%) 1.8 (1.1)
42 (50%) 42 (50%) 1.9 (1.2)
5 (20.8%) 0 5 (20.8%)
35 (41.7%)† 2 (2.4%) 37 (44.0%)‡
1.87 (0.30) 4.22 (0.44) 1.76 (0.17)
2.15 (0.57) 4.19 (0.44)
12 (50.0%) 6 (25.0%) 6 (25.0%) 0 1.08 (0.65) 7.5 (7.0)
37 (44.0%) 33 (39.4%) 13 (15.5%) 1 (1.2%) 1.34 (1.60) 4.3 (2.4)§
NOTE. All data are presented as number of patients and percent of group unless otherwise stated. Abbreviations: POAT, postoperative atrial tachyarrhythmia; SD, standard deviation; LVEF, left ventricular ejection fraction; PTCA, percutaneous transluminal coronary angioplasty; CABG, coronary artery bypass graft; ACE, angiotensin-converting enzyme; COPD, chronic obstructive pulmonary disease; MID, minimally invasive direct; OP, off-pump; ICU, intensive care unit. *p ⫽ 0.06. †p ⫽ 0.092. ‡p ⫽ 0.057. §p ⫽ 0.05.
MAGNESIUM IN OFF-PUMP CABG
527
Table 3. Data Comparing Patients Who Received Magnesium With Those Who Did Not Preoperative Variable
Magnesium (n ⫽ 42)
No Magnesium (n ⫽ 66)
Mean age (y) (⫾SD) Sex: male LVEF (%) (⫾SD) ⱕ40% Previous procedure PTCA CABG Coronary lesions Left anterior descending artery Left circumflex artery Right coronary artery/patent ductus arteriosus Preoperative medicines -blocker Calcium channel blocker Digoxin Diuretics ACE inhibitors Coexisting disease Myocardial infarction Congestive heart failure Hypertension Tobacco/COPD Renal disease POAT Surgical procedure MID CABG OP CABG sternotomy No. bypasses performed Intraoperative medicines -blockers Lidocaine Calcium blockers Magnesium dose (g) (⫾SD) Postoperative electrolyte levels (mEq/L) (SD) Initial magnesium Initial potassium Magnesium at POAT Hypomagnesemia (⬍1.8 meq/L) (%) Day of POAT Time to extubation Operating room Same day ICU Postoperative day 1 Other Mean ICU stay (d) (SD) Mean hospital stay (d) (SD)
62.7 (11.9) 33 (78.6%) 49.9 (11.6) 9 (21.4%)
64.4 (13.2) 40 (60.6%) 52.4 (11.6) 12 (18.2%)
16 (38.1%) 1 (2.4%)
20 (30.3%) 9 (13.6%)
40 (95.2%) 22 (52.4%)
63 (95.4%) 38 (57.6%)
25 (59.5%)
34 (51.5%)
36 (85.7%) 16 (38.1%) 2 (4.8%) 7 (16.7%) 6 (14.3%)
61 (92.4%) 27 (40.9%) 4 (4.8%) 15 (22.7%) 15 (22.7%)
20 (30.3%) 3 (7.1%) 22 (52.3%) 16 (38.1%) 3 (7.1%) 5 (12%)
21 (31.8%) 9 (13.6%) 32 (485%) 29 (43.9%) 3 (4.5%) 19 (29%)*
23 (54.7%) 19 (45.2%) 1.9 (1.3)
32 (48.5%) 34 (51.5%) 1.8 (1.1)
20 (47.6%) 13 (30.9%) 0 5 (3)
20 (30.3%) 12 (18.1%) 2 (3.0%) 0
did not (12% v 29%). After adjusting for age and previous CABG surgery, use of intraoperative magnesium still was found to be associated with a significant decrease in the incidence of POAT ( p ⫽ 0.03). No other covariate, including type of surgery, contributed significantly to the model. Although the use of intraoperative -adrenergic blockers was not a significant covariate in the logistic regression model, the authors looked at the combined use of intraoperative -adrenergic blockers and magnesium. In this sample, patients receiving intraoperative magnesium and -blockers had a 5% (1 of 20) incidence of POAT, compared with a 19% (8 of 42) incidence for patients receiving either agent alone and a 33% (15 of 46) incidence for patients receiving neither agent ( p ⬍ 0.05). DISCUSSION
2.37 (0.53) 4.17 (0.49) 1.80 (0.06) 3 (9%) 2.0 (0.7) 18 (42.9%) 17 (40.5%) 6 (14.3%) 1 (2.4%) 1.5 (2.1) 5.0 (4.9)
1.86 (0.40)* 4.22 (0.39) 1.80 (0.19) 17 (35%)* 2.1 (0.7) 31 (47.0%) 22 (33.3%) 13 (19.7%) 0 1.2 (0.8) 4.9 (3.5)
NOTE. All data are presented as number of patients and percent of group unless otherwise stated. Abbreviations: SD, standard deviation; LVEF, left ventricular ejection fraction; PTCA, percutaneous transluminal coronary angioplasty; CABG, coronary artery bypass graft; ACE, angiotensin-converting enzyme; COPD, chronic obstructive pulmonary disease; POAT, postoperative atrial tachyarrhythmia; MID, minimally invasive direct; OP, off-pump; ICU, intensive care unit. *p ⱕ 0.05.
There were no significant differences found regarding preoperative and postoperative hematocrit values or blood use. The incidence of POAT was significantly less in patients who received intraoperative magnesium compared with those who
The overall incidence of POAT after OP CABG surgery in this study is comparable to that reported for conventional CABG surgery and is associated with an increased hospital stay.1-4,13-16 Of patients not receiving magnesium, 35% were hypomagnesemic immediately after surgery. Patients who received intraoperative magnesium had a significantly lower incidence of POAT. These results suggest a possible role of prophylactic magnesium sulfate during OP CABG surgery to reduce the incidence of POAT. Because of the variability of magnesium doses administered, it is not possible to recommend the optimal magnesium dose. This study also suggests a potential additive benefit if magnesium and -adrenergic antagonists are given together. Patients receiving magnesium and -adrenergic antagonists had a lower incidence of POAT than patients receiving either agent alone. Conclusions regarding -adrenergic blockers are difficult to make because of the variability in dosing, timing of administration, and pharmacokinetics of the different agents. Previously reported benefits of magnesium supplementation in animals and in patients undergoing conventional CABG surgery include reductions in atrial and ventricular arrhythmias as well as reduction in ischemic myocardial injury.7-11,17-22 The mechanism of these benefits has not been stated clearly; however, several articles have shown an effect on the transmembrane movement of other ions, specifically an inhibition of calcium influx into myocardial and vascular cells.6,20,23,24 This effect on ion movement may be an important part of the antiarrhythmic activity. Magnesium has been shown to possess class IV (calcium channel blocking activity) and class I (sodium channel blocking activity) antiarrhythmic properties, which result in an increase in conduction time through the atrioventricular node and accessory pathways as well as an increase in the refractoriness of the conduction system.20,25 A review of the literature shows variability in the dosing and timing of intravenous magnesium administration. A reduction in ventricular arrhythmias (ventricular tachycardia and ventricular fibrillation) is seen whether intravenous magnesium is given before or after CPB and whether it is included in the cardioplegia or given only after removal of the aortic cross-clamp.9,11,26-29 Dosing regimens in investigations addressing POAT are variable.7,8,10,11 Regimens range from 2 to 3 g of magnesium after CPB, to supplementation before and after CPB, to 22 to 30 g over the first 4 days after surgery.7-11,27 Intraoperative magnesium supplementation has resulted in significantly higher serum
528
magnesium levels and a ⬎50% reduction in POAT.7,8 Two studies have shown greater reductions in POAT when magnesium was administered before and after CPB compared with administration only after CPB.10,11 Although relatively small doses of intravenous magnesium given after CPB (2 to 3 g) are associated with a reduction in ventricular arrhythmias, it does not appear to reduce the incidence of POAT.9-11,27 In one study, the results did not show a reduction in POAT but instead reported a decrease in severity, defined by the duration of POAT.9 The regimen in this study was 2.5 g of magnesium every 8 hours for 4 days, which started after completion of surgery on arrival in the ICU. In the present study, a single dose of intravenous magnesium, administered before manipulation of the native coronary arteries, was associated with a seemingly prolonged benefit as seen by the reduction in POAT in the early postoperative period. Terzi et al30 showed a similar prolonged benefit in thoracic surgical patients. A dose of 4 g of supplemental magnesium was given on the day of surgery (2 g during the operation and 2 g within 6 hours after surgery on the same day). Patients who received supplemental magnesium had fewer atrial tachyarrhythmias in the postoperative period.30 Both studies have shown a prolonged benefit for several days after thoracic surgery. The prolonged benefit of magnesium administration may be due to the slow equilibrium of magnesium across tissues as well as the long half-life of ⬎40 days.20,25 From the literature, it is unclear what is the optimal dose of magnesium supplementation; however, it does seem that in medical and surgical patients, the timing of administration may be important.10,11,17-22 One problem in the assessment of therapeutic effects of magnesium is the absence of a reliable and readily available monitor for intracellular magnesium levels.20,25 Less than 1% of magnesium exists in the blood, and only a portion of this is ionized and active.20,25 Neither ionized nor total serum magnesium levels correlate well with cellular stores.20,25 This poor correlation is due to the low percentage of magnesium in the blood and the slow equilibrium of magnesium across tissue membranes. Reports have shown that buccal cell magnesium concentration may correlate with myocardial cellular levels and that 24-hour urine samples may predict total body stores, but neither analysis is commonly done or is practical in the perioperative setting.20,25 Although the authors showed that patients who received magnesium supplementation had higher serum magnesium levels, they were unable to show a relationship between serum magnesium levels and POAT (Table 2). Although hypomagnesemia may be associated with arrhythmias, it is not known which serum magnesium level confers protection from arrhythmogenic activity.12,20,25,31-33 Studies have shown that 30% to 70% of cardiac surgery patients have low serum magnesium levels perioperatively.20,25 Three patterns of hypomagnesemia have been described in conventional CABG surgery, including perioperative hemodilution, intraoperative cellular depletion, and postoperative cellular depletion.34 Cellular depletion is thought to be due to the effects of cardioplegia, sympathetic stimulation (endogenous or exogenous catecholamines), and perioperative diuretics.34 The cause of hypomagnesemia in OP CABG surgery patients may be multifactorial and may include hemodilution, blood loss, blood transfusions, and increases in catecholamines during
MASLOW ET AL
surgery. This issue was not examined in the present study, however. Preoperative magnesium levels were not obtained. It is not possible to eliminate nutritional deficiencies as a cause of hypomagnesemia, and it is not possible to know what role this has in the incidence of POAT. The safety of magnesium supplementation has been shown by a reduced need for temporary pacing in the postoperative period10 and an increase in left ventricular stroke work in CABG surgery patients.28,35 Magnesium supplementation has been used safely in medical patients with severely reduced systolic function with heart failure.28,35 Time to extubation was not different or significantly reduced when compared with placebo.10,27 Although most prior literature suggests beneficial effects of magnesium therapy in cardiac surgical patients, 2 studies have shown increases in atrial and ventricular arrhythmias.36,37 One study showed that an increase in atrial arrhythmias was associated with the highest serum magnesium levels, whereas in another, a high serum level was associated with an increased number of defibrillations and an increased amount of energy needed to defibrillate the heart after CPB.36,37 The former study randomized patients after completion of surgery and documented hypomagnesemia.36 -Adrenergic antagonists have been shown to reduce the incidence of POAT in patients undergoing conventional CABG surgery.38-41 The data in the present study showed a trend toward a reduced incidence of POAT after use of intraoperative -blockers. On further examination, a lower incidence of POAT was found in patients receiving -blockers and magnesium than in patients receiving either medication given alone. This finding suggests a possible additive benefit when -blockers and magnesium are used in combination. With the exception of the study by Fanning et al,8 previous studies showing the benefits of supplemental magnesium did not report on the use of -adrenergic blockers in their patients.7,9-11 Given the variability of the agents, the dosing, and the timing of administration, however, the benefits of -adrenergic blockers in this setting can only be speculated. There are limitations to this study, including the retrospective data collection, lack of randomization, and variable doses of magnesium. Because of the design of this retrospective analysis, it is not possible to do more than speculate on a benefit of intraoperative magnesium supplementation in the reduction of POAT. The reduction in POAT is consistent with that seen in conventional CABG surgery patients who were enrolled in prospective, randomized, blinded studies, however. The number of patients evaluated is relatively small, especially when considering the many variables that coexist, including but not limited to age, hypertension, and valve disease. Although there were no significant differences in these variables between the magnesium-treated and not-treated patients, a larger sample size would add more weight to the results. The small sample size may have prevented the detection of a statistically significant benefit of -adrenergic blockers. The small numbers also limited the logistic regression models that could be considered in examining the relationships of intravenous magnesium and other variables with the incidence of POAT. Another limitation was the lack of a clinical management protocol. Although routine postoperative care involved
MAGNESIUM IN OFF-PUMP CABG
529
maintaining serum potassium within the normal range, monitoring of magnesium levels was not performed routinely in patients from the first day after surgery and beyond. Only patients who had POAT had additional magnesium levels obtained. A problem with the retrospective study was the lack of control over dosing of postoperative cardiac medications and heart rate control. This control is especially important in this situation because POAT occurred with a peak incidence 24 to 72 hours after surgery. Nearly all patients in this study were started on -adrenergic antagonists (atenolol or metoprolol) postoperatively. Consistent monitoring of magnesium activity would have been useful to analyze the changes in serum magnesium levels. As stated earlier, however, it is not clear what the relationship is between serum levels and benefits. A limitation that is consistent across the literature is the lack of a readily available tool to assess cellular activity of magnesium. In conclusion, the present data suggest that intraoperative
magnesium supplementation may be associated with a reduction in POAT in OP CABG surgery patients. This suggestion complements previously reported data showing a reduction in POAT in conventional CABG surgery patients. Although supplementation was begun before performance of coronary bypasses, these data do not establish the optimal dose and timing of magnesium administration. A significant number of the patients not receiving magnesium supplementation showed hypomagnesemia immediately after surgery. Based on these data, it is concluded that supplemental magnesium should be considered part of the intraoperative management of OP CABG surgery patients. The combined use of -adrenergic antagonists may decrease POAT further in these patients. Given the variability of -adrenergic blocker administration, however, this conclusion remains speculative. The results of this study and those found in other investigations of POAT suggest a large prospective multidose magnesium trial is called for to clarify the timing and dose of intraoperative magnesium during conventional CABG surgery and OP CABG surgery.
REFERENCES 1. Olshansky B: Management of atrial fibrillation after coronary bypass graft. Am J Cardiol 78:27-34, 1996 2. Aranki SF, Shaw DP, Adams DH, et al: Predictors of atrial fibrillation after coronary artery surgery: Current trends and impact on hospital resources. Circulation 94:390-397, 1996 3. Creswell LL, Schuesslopo RB, Rosenbloom M, Cox JL: Hazards of postoperative atrial arrhythmias. Ann Thorac Surg 56:539-549, 1993 4. Almassie GH, Schowalter T, Nicolosi AC, et al: Atrial fibrillation after cardiac surgery: A major morbid event? Ann Surg 226:501-511, 1997 5. Balser JR: Pro: All patients should receive pharmacologic prophylaxis for atrial fibrillation after cardiac surgery. J Cardiothorac Vasc Anesth 13:98-100, 1999 6. Mayer DB, Miletich DJ, Feld JM, Albrecht RF: The effects of magnesium salts on the duration of epinephrine-induced ventricular tachyarrhythmias in anesthetized rats. Anesthesiology 71:923-928, 1989 7. Colquhoun IW, Berg GA, el-Fiky M, et al: Arrhythmia prophylaxis after coronary artery surgery: A randomised controlled trial of intravenous magnesium chloride. Eur J Cardiothorac Surg 7:520-523, 1993 8. Fanning WJ, Thomas CS Jr, Roach A, et al: Prophylaxis of atrial fibrillation with magnesium sulfate after coronary artery bypass grafting. Ann Thorac Surg 52:529-533, 1991 9. Jensen BM, Alstrup P, Klitgard NA: Magnesium substitution and postoperative arrhythmias in patients undergoing coronary artery bypass grafting. Scand Cardiovasc J 31:265-269, 1997 10. Wistbacka JO, Koistinen J, Karlqvist KE, et al: Magnesium substitution in elective coronary artery surgery: A double-blind clinical study. J Cardiothorac Vasc Anesth 9:140-146, 1995 11. Casthely PA, Yoganathan T, Komer C, Kelly M: Magnesium and arrhythmias after coronary artery bypass surgery. J Cardiothorac Vasc Anesth 8:188-191, 1994 12. Aglio LS, Stanford GG, Medd R, et al: Hypomagnesemia is common following cardiac surgery. J Cardiothorac Vasc Anesth 5:201208, 1991 13. Cohn WE, Sirois CA, Johnson RG: Atrial fibrillation after minimally invasive coronary artery bypass grafting: A retrospective matched study. J Thorac Cardiovasc Surg 117:298-301, 1999 14. Abreu JE, Reilly J, Salzano RP, et al: Comparison of frequencies of atrial fibrillation after coronary artery bypass grafting with and
without the use of cardiopulmonary bypass. Am J Cardiol 83:775-776, 1999 15. Tamis JE, Vloka E, Malhotra S, et al: Atrial fibrillation is common after minimally invasive coronary artery bypass surgery. J Am Coll Cardiol 31:118A, 1998 16. Saatvedt K, Fiane AE, Sellevold O, Nordstrand K: Is atrial fibrillation caused by extracorporeal circulation? Ann Thorac Surg 68:931-933, 1999 17. Herzog WR, Schlossberg ML, MacMurdy KS, et al: Timing of magnesium therapy affects experimental infarct size. Circulation 92: 2622-2626, 1995 18. Christensen CW, Rieder MA, Silverstein EL, Bencheff NE: Magnesium sulfate reduces myocardial infarct size when administered before but not after coronary reperfusion in a canine model. Circulation 92:2617-2621, 1995 19. Antman EM: Magnesium in acute MI: Timing is critical. Circulation 92:2367-2372, 1995 20. Gomez MN: Magnesium and cardiovascular disease. Anesthesiology 89:222-240, 1998 21. Chang C, Varghese PJ, Downey J, Bloom S: Magnesium deficiency and myocardial infarct size in the dog. J Am Coll Cardiol 5:280-289, 1985 22. Leor J, Kloner RA: An experimental model examining the role of magnesium in the therapy of acute myocardial infarction. Am J Cardiol 75:1292-1293, 1995 23. Yoshimura M, Oshima T, Matsuura H, et al: Extracellular Mg2⫹ inhibits capacitative Ca2⫹ entry in vascular smooth muscle cells. Circulation 95:2567-2572, 1997 24. Prielipp RC, Butterworth JF, Roberts PR, et al: Magnesium antagonizes the actions of lysophosphatidyl choline (LPC) in myocardial cells: A possible mechanism for its antiarrhythmic effects. Anesth Analg 80:1083-1087, 1995 25. Fawcett WJ, Haxby EJ, Mala DA: Magnesium: Physiology and pharmacology. Br J Anaesth 83:302-320, 1999 26. Yurvati AH, Sanders SP, Dullye LJ, et al: Antiarrhythmic response to intravenously administered magnesium after cardiac surgery. South Med J 85:714-717, 1992 27. Karmy-Jones R, Hamilton A, Dzavik V, et al: Magnesium sulfate prophylaxis after cardiac operations. Ann Thorac Surg 59:502-507, 1995
530
28. Caspi J, Rudis E, Bar I, et al: Effects of magnesium on myocardial function after coronary artery bypass grafting. Ann Thorac Surg 59:942-947, 1995 29. Shakerinia T, Ali IM, Sullivan JA: Magnesium in cardioplegia: Is it necessary? Can J Surg 39:397-400, 1996 30. Terzi A, Furlan G, Chavocci P, et al: Prevention of atrial tachyarrhythmias after non-cardiac thoracic surgery by infusion of magnesium sulfate. Thorac Cardiovasc Surg 44:300-303, 1996 31. Vyvyan HA, Hayne PN, Cutfield GR: Magnesium flux and cardiac surgery: A study of the relationship between magnesium exchange, serum magnesium levels and postoperative arrhythmias. Anaesthesia 49:245-249, 1994 32. Kalman JM, Munawai M, Howes LG, et al: Atrial fibrillation after coronary artery bypass grafting is associated with sympathetic activation. Ann Thorac Surg 60:1709-1715, 1995 33. Zaman AG, Alamgir R, Richens T, et al: The role of signal averaged P wave duration and serum magnesium as a combined predictor of atrial fibrillation after elective coronary artery bypass surgery. Heart 77:527-531, 1997 34. Satur CMR, Anderson JR, Jennings A, et al: Magnesium flux caused by coronary artery bypass operation: Three patterns of deficiency. Ann Thorac Surg 58:1674-1678, 1994
MASLOW ET AL
35. Sueta CA, Clarke SW, Dunlap SH, et al: Effect of acute magnesium administration on the frequency of ventricular arrhythmia in patients with heart failure. Circulation 89:660-666, 1994 36. Parikka H, Toivonen L, Pellinen T, et al: The influence of intravenous magnesium sulphate on the occurrence of atrial fibrillation after coronary artery bypass operation. Eur Heart J 14:251-258, 1993 37. Hecker BR, Lake CL, Kron IL, et al: Influence of magnesium ion on human ventricular defibrillation after aortocoronary bypass surgery. Am J Cardiol 55:61-64, 1985 38. Jacquiet L, Evenpoel M, Marenne F, et al: Hemodynamic effects and safety of sotalol in the prevention of supraventricular arrhythmias after coronary artery bypass surgery. J Cardiothorac Vasc Anesth 8:431-436, 1994 39. Fuller JA, Adams GG, Buxton B: Atrial fibrillation after coronary artery bypass grafting: Is it a disorder of the elderly. J Thorac Cardiovasc Surg 97:8211-8215, 1989 40. Leitch JW, Thomson D, Barrad DK, Harris PJ: The importance of age as a predictor of atrial fibrillation and flutter after coronary artery bypass grafting. J Thorac Cardiovasc Surg 100:338-342, 1990 41. Mendes LA, Connelly GP, McKenney PA, et al: Right coronary artery stenosis: An independent predictor of atrial fibrillation after coronary artery bypass surgery. J Am Coll Cardiol 25:198-202, 1995
reviewed. Twenty-four patients (22%) had POAT. Forty-two patients (39%) received intraoperative magnesium. In patients receiving intraoperative magnesium, the incidence of POAT was significantly decreased (12% v 29%; p ⴝ 0.03). In these patients, initial postoperative serum magnesium was significantly higher (2.37 mEq/L v 1.86 mEq/L; p F 0.01). In patients not receiving intraoperative magnesium, 35% had hypomagnesemia (serum magnesium F 1.8 mEq/L) compared with 9% of patients receiving magnesium ( p F 0.01). Patients who received intraoperative magnesium and -adrenergic blockers had a lower incidence of POAT (5%) than patients who received only one (19%) or neither (33%) (p F 0.05). Conclusions: Intraoperative magnesium supplementation is associated with a decrease in POAT after off-pump CABG surgery. The combination of a -blocker and magnesium may reduce POAT further. It is recommended that intraoperative magnesium supplementation be part of the care of patients undergoing off-pump CABG surgery. Copyright r 2000 by W.B. Saunders Company
P
complications of CPB. Several surgical approaches have been described. One approach accomplishes CPB surgery through a small anterior thoracotomy (minimally invasive direct [MID] CABG surgery). A second surgical approach is by a median sternotomy incision (off-pump [OP] CABG sternotomy). Although there are a variety of approaches to OP revascularization, they all have in common the avoidance of cardiac cannulation, CPB, and cardioplegic arrest. Despite the avoidance of CPB, the incidence of POAT in OP CABG surgery, regardless of surgical approach, remains substantial and not significantly different from conventional CABG, when patient populations are matched for a variety of predictors of POAT, including age.13-16 The benefits of magnesium administration to reduce POAT in OP CABG surgery patients is unexplored. To examine whether prophylactic magnesium might reduce the incidence of POAT after OP CABG surgery, the authors retrospectively examined the incidence of early POAT (within 7 days of surgery) in patients after MID CABG surgery or OP CABG sternotomy.
OSTOPERATIVE ATRIAL TACHYARRHYTHMIAS (POAT) (atrial fibrillation, atrial flutter, supraventricular tachycardia) occur in 20% to 30% of patients undergoing conventional coronary artery bypass graft surgery (CABG) (cardiopulmonary bypass [CPB] with cardioplegic arrest).1-4 POAT is a cause of and contributes to significant morbidity, including neurologic injury and hypotension.2-4 POAT is associated with increased intensive care unit (ICU) and hospital stays and increased health care costs in this population.5 Several investigations have shown a benefit when magnesium sulfate is used for prophylaxis against POAT.6-11 Studies have shown that perioperative serum magnesium levels may be reduced in 70% of patients undergoing conventional CABG surgery using hypothermic CPB.12 It is postulated that decreased magnesium stores may be associated with increased incidence of atrial and ventricular tachycardias12 and that prophylactic administration of magnesium may be effective in reducing the incidence of tachyarrhythmias after cardiac surgery.6-11 CABG surgery without the use of CPB and cardioplegic arrest has become an increasingly popular way to avoid
From the Department of Anesthesia, Rhode Island Hospital, Providence, RI; and Biometrics Center, and Departments of Anesthesia and Critical Care and Surgery, Beth Israel Deaconess Medical Center, Boston, MA. Address reprint requests to Andrew D. Maslow, MD, Department of Anesthesia, Rhode Island Hospital, Davol Building, Room 128, 593 Eddy St, Providence, RI 02903. Copyright r 2000 by W.B. Saunders Company 1053-0770/00/1405-0006$10.00/0 doi:10.1053/jcan.2000.9485 524
KEY WORDS: arrhythmias, magnesium, cardiac surgery, coronary artery surgery (off-pump)
METHODS The authors sought to determine whether prophylactic magnesium might reduce the incidence of POAT after OP CABG surgery by median sternotomy or MID CABG surgery. According to the literature, the incidence of POAT does not appear to be different for the 2 surgical approaches.13-16 Preliminary analysis in this study showed the incidence to be similar. Both groups were combined to form one group labeled OP CABG (off-pump CABG surgery). The hospital course of 124 consecutive patients who had undergone nonemergent OP CABG surgery at this institution between 1997 and 1998 was retrospectively examined; this includes the first 64 patients who underwent MID CABG surgery and first 60 patients who underwent OP CABG sternotomy. Data were collected from medical record
Journal of Cardiothoracic and Vascular Anesthesia, Vol 14, No 5 (October), 2000: pp 524-530
MAGNESIUM IN OFF-PUMP CABG
review. Exclusion criteria for analysis included a history of atrial tachyarrhythmias and all preoperative antiarrhythmic medication for ventricular tachycardias (eg, amiodarone, procainamide, disopyramide) with the exception of -adrenergic antagonists and calcium channel blockers taken for control of angina or hypertension. Because digoxin has not been shown to affect the incidence of POAT after cardiac surgery, the use of digoxin in patients with reduced systolic function was not considered an exclusionary criterion as long as there was no history of atrial tachyarrhythmia. A total of 108 patients met the criteria for evaluation. Demographic variables collected included age, sex, ejection fraction, history of myocardial infarction, distribution of coronary artery disease, coexisting diseases, preoperative cardiac medications, and previous revascularization procedure. Intraoperative data collected included invasive monitoring, anesthetic technique, use and dosage of magnesium, use of -adrenergic blockers, use of calcium channel blockers, and the surgical procedure. Additional perioperative data included preoperative hematocrit, hematocrit on admission to the ICU, need for perioperative blood transfusion, serum magnesium and potassium levels on admission to the ICU, serum magnesium level at the time POAT was diagnosed, time of extubation, duration of ICU, and duration of hospital stay. In particular, it was recorded whether an early (within 7 days of surgery) POAT occurred. The timing of POAT and the magnesium level before treatment were recorded. POAT was defined as a documented nonsinus rhythm originating from the atrial tissue that lasted ⬎10 minutes or required pharmacologic or electric (cardioversion) therapy. Patient characteristics for the entire cohort of 108 patients are summarized as mean ⫾ standard deviation or number (and percent) of patients. A preliminary analysis examined whether the data analysis should be conducted separately by type of surgery or could be combined across the 2 types of surgery. Important variables, such as incidence of POAT, number of patients receiving magnesium supplementation, and median dose, were comparable between surgeries. Patient characteristics were summarized and compared between patients who did and did not experience POAT. Categoric variables were compared using chi-square or Fisher’s exact test, and continuous variables were compared using a two-sample t-test or Wilcoxon rank sum test, as appropriate. Similarly, characteristics of patients who did and did not receive intraoperative magnesium were compared. A p value ⱕ0.05 was considered statistically significant. The study examined whether the incidence of POAT was decreased in patients who received intraoperative magnesium, while controlling for demographic and intraoperative covariables, including type of surgery, using multiple logistic regression. Demographic and intraoperative variables that tended to be related (ie, p ⬍ 0.25) to the incidence of POAT or use of magnesium in univariate analysis were considered as potential covariates. A forward stepwise selection technique was used, and variables that contributed significantly to the model ( p ⱕ 0.05) were kept. RESULTS
Between 1997 and 1998, 124 patients had one of the CABG procedures; 64 underwent OP CABG surgery by an anterior thoracotomy incision (MID CABG), and 60 underwent OP CABG surgery by median sternotomy (OP CABG sternotomy). Sixteen patients were excluded from the final analysis because of a history of atrial tachyarrhythmia or ventricular arrthythmia requiring treatment with amiodarone, procainamide, or disopyramide. Six patients had been receiving digoxin before surgery because of a history of congestive heart failure. Of these 6, 3 underwent MID CABG surgery (left ventricular ejection fraction, 15%, 50%, and 55%) and 3 underwent OP CABG
525
sternotomy (left ventricular ejection fraction, 25%, 30%, and 40%). Two patients received magnesium (4 g each). None of the 6 patients had POAT. These patients were included in the data analysis because there was no preoperative history of arrhythmias. The anesthetic technique consisted of an induction agent (thiopental, propofol, or etomidate), maintenance with a fentanyl analog (fentanyl or sufentanil), isoflurane, and muscle relaxation (vecuronium or pancuronium). Of patients, 81% were extubated on the same day of surgery. The mean hospital stay was 5 days. These data are summarized in Table 1. Of 108 patients remaining for this study, 24 (22%) experienced POAT. Forty-two patients (39%) received intraoperative magnesium, with a median dose of 5 g of intravenous magnesium. Magnesium was administered before manipulation of the native coronary arteries in all patients. Forty patients (37%) received intraoperative -adrenergic blockers. Of these, 16 received esmolol, 22 received metoprolol, and 2 received labetalol. -Adrenergic blockers were administered for elevated heart rates (⬎95 beats/min) or if coronary ischemia was suspected. When the data from the 2 surgical techniques (MID CABG v OP CABG) were compared, there were no significant differences seen with respect to postoperative serum magnesium levels (2.13 mEq/L v 2.02 mEq/L), administration of intraoperative magnesium (42% v 36%), timing of POAT (2.1 v 2.1 days), or incidence of POAT, whether supplementation was administered or not (MID CABG, 13/55 [24%] v OP CABG 11/53 [21%]). Although there were differences seen in left ventricular ejection fraction (54% v 49%) and number of vessels bypassed (1.0 v 2.4), these differences were not found to be relevant to the occurrence of POAT. The data for the 2 surgical types are combined and shown in Table 1. For the study population, the preoperative and postoperative hematocrit values were 39% (4.9%) and 33% (4.4%). Approximately 30% of patients received packed red blood cells during the first 3 postoperative days. Patients undergoing OP CABG received an average 640 mL (⫾380 mL) of cell-saved blood. Table 2 displays data comparing patients who experienced POAT with patients who did not. Preoperative demographics, anesthetic technique, and surgical procedure were not different between the 2 groups. POAT occurred on postoperative day 2 in 62.5% of patients and on postoperative days 1 and 3 in 16.7% each. Of 24 patients experiencing POAT, 5 received magnesium (21%) compared with 37 of 84 whose hospital course was not complicated by POAT (44%). Of 24 patients (21%) experiencing POAT, 5 received intraoperative -blockers compared with 35 of 84 whose postoperative course was not complicated by POAT (41%; p ⫽ 0.09). Patients who experienced POAT tended to be older (68.0 v 62.5 years; p ⫽ 0.06). There was no statistically significant difference in initial postoperative magnesium or potassium levels between patients with and patients without POAT. Patients with POAT had significantly longer hospital stays (7.5 v 4.3 days; p ⫽ 0.0004). Preoperative and postoperative hematocrit values, transfusion requirements, and cell-saved volume were similar between the 2 groups. Table 3 compares data between patients who did and patients who did not receive intraoperative magnesium. Preoperative demographics, anesthetic technique, and surgical procedure
526
MASLOW ET AL
Table 1. Summary of Patient Data Preoperative Variable
Mean age (y) ⫾ SD Sex Male Female Ejection fraction (%) (⫾SD) 40% Previous procedure PTCA CABG (conventional) Coronary lesions (stenosis ⱖ70%) Left anterior descending artery Left circumflex artery Right coronary artery Posterior descending artery Preoperative medicines -blocker Calcium channel blocker Digoxin Diuretics ACE inhibitors Coexisting disease Myocardial infarction Congestive heart failure Hypertension Tobacco/COPD Renal insufficiency (Cr ⱖ 2 mg/dL) Surgical procedure MID CABG OP CABG sternotomy No. of bypasses (mean grafts ⫾ SD) ⱖ3 bypasses Other intraoperative medicines -blocker Calcium channel blocker Intraoperative magnesium Yes No Dose of magnesium (g) (SD) Median Mean Time of extubation Operating room ICU (same day) Postoperative day 1 Other Mean ICU stay (d) (SD) % ⬎1 day Mean hospital stay (d) (SD) Atrial fibrillation
63.7 (⫾12.7) 73 (67.6%) 35 (32.4%) 51.4 (⫾11.6) 21 (19.4%) 36 (34.3%) 10 (9.3%) 103 (95.4%) 60 (55.6%) 22 (20.4%) 40 (37.0%) 97 (89.8%) 43 (39.8%) 6 (5.6%) 22 (20.4%) 21 (19.4%) 41 (38.0%) 12 (11.1%) 54 (50.0%) 45 (41.7%) 6 (5.5%) 55 (50.9%) 53 (49.1%) 1.9 (1.2) 34 (31.5%) 40 (37.0%) 2 (1.9%) 42 (38.9%) 66 (61.1%) 5 6.5 (2.3) 39 (36.1%) 49 (45.4%) 19 (17.6%) 1 (0.9%) 1.3 (⫾1.4) 10 (9.3%) 5.0 (⫾4.1) 24 (22.2%)
NOTE. All data are presented as absolute number of patients and percent of group (n ⫽ 108) unless otherwise stated. Abbreviations: SD, standard deviation; PTCA, percutaneous transluminal coronary angioplasty; CABG, coronary artery bypass graft; ACE, angiotensin-converting enzyme; COPD, chronic obstructive pulmonary disease; Cr, creatinine; MID, minimally invasive direct; OP, off-pump; ICU, intensive care unit.
were not different between the 2 groups. Of the 108 patients, 42 (39%) received intraoperative magnesium. The initial postoperative serum magnesium level was greater in patients receiving magnesium (2.4 mEq/L v 1.9 mEq/L; p ⬍ 0.05). In patients who did not receive intraoperative magnesium, the incidence of
postoperative hypomagnesemia (⬍1.8 mEq/L) was 35%, compared with 9% in patients who received intraoperative magnesium. The timing of POAT was similar whether patients received intraoperative magnesium or not (2.0 v 2.1 days). Table 2. Data Comparing Patients Experiencing Atrial Fibrillation With Patients Who Did Not Preoperative Variable
Mean age (y) (⫾SD) Sex: male Mean LVEF (%) (⫾SD) ⱕ40% Previous procedure PTCA CABG Coronary lesions Left anterior descending artery Left circumflex artery Right coronary artery/patent ductus arteriosus Preoperative medicines -blocker Calcium channel blocker Digoxin Diuretics ACE inhibitors Coexisting disease Myocardial infarction Congestive heart failure Hypertension Tobacco/COPD Renal disease Surgical procedure MID CABG OP CABG sternotomy No. bypasses performed Intraoperative medicines -blockers Calcium channel blockers Intraoperative magnesium Electrolyte levels (mEq/L) (SD) Postoperative magnesium Postoperative potassium Magnesium at POAT Time of extubation Operating room Same day ICU Postoperative day 1 Other Mean ICU stay (d) (SD) Mean hospital stay (d) (SD)
POAT (n ⫽ 24)
68.0 (14.4) 15 (62.5%) 52.7 (8.9%)
No POAT (n ⫽ 84)
62.5 (12.0)* 58 (69.0%) 51.0 (12.3%) 15 (17.8%)
3 (12.5%) 7 (29.2%)
29 (34.5%) 9 (10.7%)
1 (4.2%) 22 (91.7%)
81 (96.4%) 45 (53.6%)
15 (62.5%)
45 (53.6%)
13 (54.2%) 22 (91.7%) 12 (50.0%) 0 4 (16.7%)
75 (89.3%) 31 (36.9%) 6 (7.1%) 18 (21.4%) 18 (21.4%)
3 (12.5%) 9 (37.5%) 2 (8.4%) 11 (45.8%) 11 (45.8%)
32 (38.1%) 10 (11.9%) 43 (51.2%) 34 (40.5%) 3 (3.6%)
13 (54.2%) 11 (45.8%) 1.8 (1.1)
42 (50%) 42 (50%) 1.9 (1.2)
5 (20.8%) 0 5 (20.8%)
35 (41.7%)† 2 (2.4%) 37 (44.0%)‡
1.87 (0.30) 4.22 (0.44) 1.76 (0.17)
2.15 (0.57) 4.19 (0.44)
12 (50.0%) 6 (25.0%) 6 (25.0%) 0 1.08 (0.65) 7.5 (7.0)
37 (44.0%) 33 (39.4%) 13 (15.5%) 1 (1.2%) 1.34 (1.60) 4.3 (2.4)§
NOTE. All data are presented as number of patients and percent of group unless otherwise stated. Abbreviations: POAT, postoperative atrial tachyarrhythmia; SD, standard deviation; LVEF, left ventricular ejection fraction; PTCA, percutaneous transluminal coronary angioplasty; CABG, coronary artery bypass graft; ACE, angiotensin-converting enzyme; COPD, chronic obstructive pulmonary disease; MID, minimally invasive direct; OP, off-pump; ICU, intensive care unit. *p ⫽ 0.06. †p ⫽ 0.092. ‡p ⫽ 0.057. §p ⫽ 0.05.
MAGNESIUM IN OFF-PUMP CABG
527
Table 3. Data Comparing Patients Who Received Magnesium With Those Who Did Not Preoperative Variable
Magnesium (n ⫽ 42)
No Magnesium (n ⫽ 66)
Mean age (y) (⫾SD) Sex: male LVEF (%) (⫾SD) ⱕ40% Previous procedure PTCA CABG Coronary lesions Left anterior descending artery Left circumflex artery Right coronary artery/patent ductus arteriosus Preoperative medicines -blocker Calcium channel blocker Digoxin Diuretics ACE inhibitors Coexisting disease Myocardial infarction Congestive heart failure Hypertension Tobacco/COPD Renal disease POAT Surgical procedure MID CABG OP CABG sternotomy No. bypasses performed Intraoperative medicines -blockers Lidocaine Calcium blockers Magnesium dose (g) (⫾SD) Postoperative electrolyte levels (mEq/L) (SD) Initial magnesium Initial potassium Magnesium at POAT Hypomagnesemia (⬍1.8 meq/L) (%) Day of POAT Time to extubation Operating room Same day ICU Postoperative day 1 Other Mean ICU stay (d) (SD) Mean hospital stay (d) (SD)
62.7 (11.9) 33 (78.6%) 49.9 (11.6) 9 (21.4%)
64.4 (13.2) 40 (60.6%) 52.4 (11.6) 12 (18.2%)
16 (38.1%) 1 (2.4%)
20 (30.3%) 9 (13.6%)
40 (95.2%) 22 (52.4%)
63 (95.4%) 38 (57.6%)
25 (59.5%)
34 (51.5%)
36 (85.7%) 16 (38.1%) 2 (4.8%) 7 (16.7%) 6 (14.3%)
61 (92.4%) 27 (40.9%) 4 (4.8%) 15 (22.7%) 15 (22.7%)
20 (30.3%) 3 (7.1%) 22 (52.3%) 16 (38.1%) 3 (7.1%) 5 (12%)
21 (31.8%) 9 (13.6%) 32 (485%) 29 (43.9%) 3 (4.5%) 19 (29%)*
23 (54.7%) 19 (45.2%) 1.9 (1.3)
32 (48.5%) 34 (51.5%) 1.8 (1.1)
20 (47.6%) 13 (30.9%) 0 5 (3)
20 (30.3%) 12 (18.1%) 2 (3.0%) 0
did not (12% v 29%). After adjusting for age and previous CABG surgery, use of intraoperative magnesium still was found to be associated with a significant decrease in the incidence of POAT ( p ⫽ 0.03). No other covariate, including type of surgery, contributed significantly to the model. Although the use of intraoperative -adrenergic blockers was not a significant covariate in the logistic regression model, the authors looked at the combined use of intraoperative -adrenergic blockers and magnesium. In this sample, patients receiving intraoperative magnesium and -blockers had a 5% (1 of 20) incidence of POAT, compared with a 19% (8 of 42) incidence for patients receiving either agent alone and a 33% (15 of 46) incidence for patients receiving neither agent ( p ⬍ 0.05). DISCUSSION
2.37 (0.53) 4.17 (0.49) 1.80 (0.06) 3 (9%) 2.0 (0.7) 18 (42.9%) 17 (40.5%) 6 (14.3%) 1 (2.4%) 1.5 (2.1) 5.0 (4.9)
1.86 (0.40)* 4.22 (0.39) 1.80 (0.19) 17 (35%)* 2.1 (0.7) 31 (47.0%) 22 (33.3%) 13 (19.7%) 0 1.2 (0.8) 4.9 (3.5)
NOTE. All data are presented as number of patients and percent of group unless otherwise stated. Abbreviations: SD, standard deviation; LVEF, left ventricular ejection fraction; PTCA, percutaneous transluminal coronary angioplasty; CABG, coronary artery bypass graft; ACE, angiotensin-converting enzyme; COPD, chronic obstructive pulmonary disease; POAT, postoperative atrial tachyarrhythmia; MID, minimally invasive direct; OP, off-pump; ICU, intensive care unit. *p ⱕ 0.05.
There were no significant differences found regarding preoperative and postoperative hematocrit values or blood use. The incidence of POAT was significantly less in patients who received intraoperative magnesium compared with those who
The overall incidence of POAT after OP CABG surgery in this study is comparable to that reported for conventional CABG surgery and is associated with an increased hospital stay.1-4,13-16 Of patients not receiving magnesium, 35% were hypomagnesemic immediately after surgery. Patients who received intraoperative magnesium had a significantly lower incidence of POAT. These results suggest a possible role of prophylactic magnesium sulfate during OP CABG surgery to reduce the incidence of POAT. Because of the variability of magnesium doses administered, it is not possible to recommend the optimal magnesium dose. This study also suggests a potential additive benefit if magnesium and -adrenergic antagonists are given together. Patients receiving magnesium and -adrenergic antagonists had a lower incidence of POAT than patients receiving either agent alone. Conclusions regarding -adrenergic blockers are difficult to make because of the variability in dosing, timing of administration, and pharmacokinetics of the different agents. Previously reported benefits of magnesium supplementation in animals and in patients undergoing conventional CABG surgery include reductions in atrial and ventricular arrhythmias as well as reduction in ischemic myocardial injury.7-11,17-22 The mechanism of these benefits has not been stated clearly; however, several articles have shown an effect on the transmembrane movement of other ions, specifically an inhibition of calcium influx into myocardial and vascular cells.6,20,23,24 This effect on ion movement may be an important part of the antiarrhythmic activity. Magnesium has been shown to possess class IV (calcium channel blocking activity) and class I (sodium channel blocking activity) antiarrhythmic properties, which result in an increase in conduction time through the atrioventricular node and accessory pathways as well as an increase in the refractoriness of the conduction system.20,25 A review of the literature shows variability in the dosing and timing of intravenous magnesium administration. A reduction in ventricular arrhythmias (ventricular tachycardia and ventricular fibrillation) is seen whether intravenous magnesium is given before or after CPB and whether it is included in the cardioplegia or given only after removal of the aortic cross-clamp.9,11,26-29 Dosing regimens in investigations addressing POAT are variable.7,8,10,11 Regimens range from 2 to 3 g of magnesium after CPB, to supplementation before and after CPB, to 22 to 30 g over the first 4 days after surgery.7-11,27 Intraoperative magnesium supplementation has resulted in significantly higher serum
528
magnesium levels and a ⬎50% reduction in POAT.7,8 Two studies have shown greater reductions in POAT when magnesium was administered before and after CPB compared with administration only after CPB.10,11 Although relatively small doses of intravenous magnesium given after CPB (2 to 3 g) are associated with a reduction in ventricular arrhythmias, it does not appear to reduce the incidence of POAT.9-11,27 In one study, the results did not show a reduction in POAT but instead reported a decrease in severity, defined by the duration of POAT.9 The regimen in this study was 2.5 g of magnesium every 8 hours for 4 days, which started after completion of surgery on arrival in the ICU. In the present study, a single dose of intravenous magnesium, administered before manipulation of the native coronary arteries, was associated with a seemingly prolonged benefit as seen by the reduction in POAT in the early postoperative period. Terzi et al30 showed a similar prolonged benefit in thoracic surgical patients. A dose of 4 g of supplemental magnesium was given on the day of surgery (2 g during the operation and 2 g within 6 hours after surgery on the same day). Patients who received supplemental magnesium had fewer atrial tachyarrhythmias in the postoperative period.30 Both studies have shown a prolonged benefit for several days after thoracic surgery. The prolonged benefit of magnesium administration may be due to the slow equilibrium of magnesium across tissues as well as the long half-life of ⬎40 days.20,25 From the literature, it is unclear what is the optimal dose of magnesium supplementation; however, it does seem that in medical and surgical patients, the timing of administration may be important.10,11,17-22 One problem in the assessment of therapeutic effects of magnesium is the absence of a reliable and readily available monitor for intracellular magnesium levels.20,25 Less than 1% of magnesium exists in the blood, and only a portion of this is ionized and active.20,25 Neither ionized nor total serum magnesium levels correlate well with cellular stores.20,25 This poor correlation is due to the low percentage of magnesium in the blood and the slow equilibrium of magnesium across tissue membranes. Reports have shown that buccal cell magnesium concentration may correlate with myocardial cellular levels and that 24-hour urine samples may predict total body stores, but neither analysis is commonly done or is practical in the perioperative setting.20,25 Although the authors showed that patients who received magnesium supplementation had higher serum magnesium levels, they were unable to show a relationship between serum magnesium levels and POAT (Table 2). Although hypomagnesemia may be associated with arrhythmias, it is not known which serum magnesium level confers protection from arrhythmogenic activity.12,20,25,31-33 Studies have shown that 30% to 70% of cardiac surgery patients have low serum magnesium levels perioperatively.20,25 Three patterns of hypomagnesemia have been described in conventional CABG surgery, including perioperative hemodilution, intraoperative cellular depletion, and postoperative cellular depletion.34 Cellular depletion is thought to be due to the effects of cardioplegia, sympathetic stimulation (endogenous or exogenous catecholamines), and perioperative diuretics.34 The cause of hypomagnesemia in OP CABG surgery patients may be multifactorial and may include hemodilution, blood loss, blood transfusions, and increases in catecholamines during
MASLOW ET AL
surgery. This issue was not examined in the present study, however. Preoperative magnesium levels were not obtained. It is not possible to eliminate nutritional deficiencies as a cause of hypomagnesemia, and it is not possible to know what role this has in the incidence of POAT. The safety of magnesium supplementation has been shown by a reduced need for temporary pacing in the postoperative period10 and an increase in left ventricular stroke work in CABG surgery patients.28,35 Magnesium supplementation has been used safely in medical patients with severely reduced systolic function with heart failure.28,35 Time to extubation was not different or significantly reduced when compared with placebo.10,27 Although most prior literature suggests beneficial effects of magnesium therapy in cardiac surgical patients, 2 studies have shown increases in atrial and ventricular arrhythmias.36,37 One study showed that an increase in atrial arrhythmias was associated with the highest serum magnesium levels, whereas in another, a high serum level was associated with an increased number of defibrillations and an increased amount of energy needed to defibrillate the heart after CPB.36,37 The former study randomized patients after completion of surgery and documented hypomagnesemia.36 -Adrenergic antagonists have been shown to reduce the incidence of POAT in patients undergoing conventional CABG surgery.38-41 The data in the present study showed a trend toward a reduced incidence of POAT after use of intraoperative -blockers. On further examination, a lower incidence of POAT was found in patients receiving -blockers and magnesium than in patients receiving either medication given alone. This finding suggests a possible additive benefit when -blockers and magnesium are used in combination. With the exception of the study by Fanning et al,8 previous studies showing the benefits of supplemental magnesium did not report on the use of -adrenergic blockers in their patients.7,9-11 Given the variability of the agents, the dosing, and the timing of administration, however, the benefits of -adrenergic blockers in this setting can only be speculated. There are limitations to this study, including the retrospective data collection, lack of randomization, and variable doses of magnesium. Because of the design of this retrospective analysis, it is not possible to do more than speculate on a benefit of intraoperative magnesium supplementation in the reduction of POAT. The reduction in POAT is consistent with that seen in conventional CABG surgery patients who were enrolled in prospective, randomized, blinded studies, however. The number of patients evaluated is relatively small, especially when considering the many variables that coexist, including but not limited to age, hypertension, and valve disease. Although there were no significant differences in these variables between the magnesium-treated and not-treated patients, a larger sample size would add more weight to the results. The small sample size may have prevented the detection of a statistically significant benefit of -adrenergic blockers. The small numbers also limited the logistic regression models that could be considered in examining the relationships of intravenous magnesium and other variables with the incidence of POAT. Another limitation was the lack of a clinical management protocol. Although routine postoperative care involved
MAGNESIUM IN OFF-PUMP CABG
529
maintaining serum potassium within the normal range, monitoring of magnesium levels was not performed routinely in patients from the first day after surgery and beyond. Only patients who had POAT had additional magnesium levels obtained. A problem with the retrospective study was the lack of control over dosing of postoperative cardiac medications and heart rate control. This control is especially important in this situation because POAT occurred with a peak incidence 24 to 72 hours after surgery. Nearly all patients in this study were started on -adrenergic antagonists (atenolol or metoprolol) postoperatively. Consistent monitoring of magnesium activity would have been useful to analyze the changes in serum magnesium levels. As stated earlier, however, it is not clear what the relationship is between serum levels and benefits. A limitation that is consistent across the literature is the lack of a readily available tool to assess cellular activity of magnesium. In conclusion, the present data suggest that intraoperative
magnesium supplementation may be associated with a reduction in POAT in OP CABG surgery patients. This suggestion complements previously reported data showing a reduction in POAT in conventional CABG surgery patients. Although supplementation was begun before performance of coronary bypasses, these data do not establish the optimal dose and timing of magnesium administration. A significant number of the patients not receiving magnesium supplementation showed hypomagnesemia immediately after surgery. Based on these data, it is concluded that supplemental magnesium should be considered part of the intraoperative management of OP CABG surgery patients. The combined use of -adrenergic antagonists may decrease POAT further in these patients. Given the variability of -adrenergic blocker administration, however, this conclusion remains speculative. The results of this study and those found in other investigations of POAT suggest a large prospective multidose magnesium trial is called for to clarify the timing and dose of intraoperative magnesium during conventional CABG surgery and OP CABG surgery.
REFERENCES 1. Olshansky B: Management of atrial fibrillation after coronary bypass graft. Am J Cardiol 78:27-34, 1996 2. Aranki SF, Shaw DP, Adams DH, et al: Predictors of atrial fibrillation after coronary artery surgery: Current trends and impact on hospital resources. Circulation 94:390-397, 1996 3. Creswell LL, Schuesslopo RB, Rosenbloom M, Cox JL: Hazards of postoperative atrial arrhythmias. Ann Thorac Surg 56:539-549, 1993 4. Almassie GH, Schowalter T, Nicolosi AC, et al: Atrial fibrillation after cardiac surgery: A major morbid event? Ann Surg 226:501-511, 1997 5. Balser JR: Pro: All patients should receive pharmacologic prophylaxis for atrial fibrillation after cardiac surgery. J Cardiothorac Vasc Anesth 13:98-100, 1999 6. Mayer DB, Miletich DJ, Feld JM, Albrecht RF: The effects of magnesium salts on the duration of epinephrine-induced ventricular tachyarrhythmias in anesthetized rats. Anesthesiology 71:923-928, 1989 7. Colquhoun IW, Berg GA, el-Fiky M, et al: Arrhythmia prophylaxis after coronary artery surgery: A randomised controlled trial of intravenous magnesium chloride. Eur J Cardiothorac Surg 7:520-523, 1993 8. Fanning WJ, Thomas CS Jr, Roach A, et al: Prophylaxis of atrial fibrillation with magnesium sulfate after coronary artery bypass grafting. Ann Thorac Surg 52:529-533, 1991 9. Jensen BM, Alstrup P, Klitgard NA: Magnesium substitution and postoperative arrhythmias in patients undergoing coronary artery bypass grafting. Scand Cardiovasc J 31:265-269, 1997 10. Wistbacka JO, Koistinen J, Karlqvist KE, et al: Magnesium substitution in elective coronary artery surgery: A double-blind clinical study. J Cardiothorac Vasc Anesth 9:140-146, 1995 11. Casthely PA, Yoganathan T, Komer C, Kelly M: Magnesium and arrhythmias after coronary artery bypass surgery. J Cardiothorac Vasc Anesth 8:188-191, 1994 12. Aglio LS, Stanford GG, Medd R, et al: Hypomagnesemia is common following cardiac surgery. J Cardiothorac Vasc Anesth 5:201208, 1991 13. Cohn WE, Sirois CA, Johnson RG: Atrial fibrillation after minimally invasive coronary artery bypass grafting: A retrospective matched study. J Thorac Cardiovasc Surg 117:298-301, 1999 14. Abreu JE, Reilly J, Salzano RP, et al: Comparison of frequencies of atrial fibrillation after coronary artery bypass grafting with and
without the use of cardiopulmonary bypass. Am J Cardiol 83:775-776, 1999 15. Tamis JE, Vloka E, Malhotra S, et al: Atrial fibrillation is common after minimally invasive coronary artery bypass surgery. J Am Coll Cardiol 31:118A, 1998 16. Saatvedt K, Fiane AE, Sellevold O, Nordstrand K: Is atrial fibrillation caused by extracorporeal circulation? Ann Thorac Surg 68:931-933, 1999 17. Herzog WR, Schlossberg ML, MacMurdy KS, et al: Timing of magnesium therapy affects experimental infarct size. Circulation 92: 2622-2626, 1995 18. Christensen CW, Rieder MA, Silverstein EL, Bencheff NE: Magnesium sulfate reduces myocardial infarct size when administered before but not after coronary reperfusion in a canine model. Circulation 92:2617-2621, 1995 19. Antman EM: Magnesium in acute MI: Timing is critical. Circulation 92:2367-2372, 1995 20. Gomez MN: Magnesium and cardiovascular disease. Anesthesiology 89:222-240, 1998 21. Chang C, Varghese PJ, Downey J, Bloom S: Magnesium deficiency and myocardial infarct size in the dog. J Am Coll Cardiol 5:280-289, 1985 22. Leor J, Kloner RA: An experimental model examining the role of magnesium in the therapy of acute myocardial infarction. Am J Cardiol 75:1292-1293, 1995 23. Yoshimura M, Oshima T, Matsuura H, et al: Extracellular Mg2⫹ inhibits capacitative Ca2⫹ entry in vascular smooth muscle cells. Circulation 95:2567-2572, 1997 24. Prielipp RC, Butterworth JF, Roberts PR, et al: Magnesium antagonizes the actions of lysophosphatidyl choline (LPC) in myocardial cells: A possible mechanism for its antiarrhythmic effects. Anesth Analg 80:1083-1087, 1995 25. Fawcett WJ, Haxby EJ, Mala DA: Magnesium: Physiology and pharmacology. Br J Anaesth 83:302-320, 1999 26. Yurvati AH, Sanders SP, Dullye LJ, et al: Antiarrhythmic response to intravenously administered magnesium after cardiac surgery. South Med J 85:714-717, 1992 27. Karmy-Jones R, Hamilton A, Dzavik V, et al: Magnesium sulfate prophylaxis after cardiac operations. Ann Thorac Surg 59:502-507, 1995
530
28. Caspi J, Rudis E, Bar I, et al: Effects of magnesium on myocardial function after coronary artery bypass grafting. Ann Thorac Surg 59:942-947, 1995 29. Shakerinia T, Ali IM, Sullivan JA: Magnesium in cardioplegia: Is it necessary? Can J Surg 39:397-400, 1996 30. Terzi A, Furlan G, Chavocci P, et al: Prevention of atrial tachyarrhythmias after non-cardiac thoracic surgery by infusion of magnesium sulfate. Thorac Cardiovasc Surg 44:300-303, 1996 31. Vyvyan HA, Hayne PN, Cutfield GR: Magnesium flux and cardiac surgery: A study of the relationship between magnesium exchange, serum magnesium levels and postoperative arrhythmias. Anaesthesia 49:245-249, 1994 32. Kalman JM, Munawai M, Howes LG, et al: Atrial fibrillation after coronary artery bypass grafting is associated with sympathetic activation. Ann Thorac Surg 60:1709-1715, 1995 33. Zaman AG, Alamgir R, Richens T, et al: The role of signal averaged P wave duration and serum magnesium as a combined predictor of atrial fibrillation after elective coronary artery bypass surgery. Heart 77:527-531, 1997 34. Satur CMR, Anderson JR, Jennings A, et al: Magnesium flux caused by coronary artery bypass operation: Three patterns of deficiency. Ann Thorac Surg 58:1674-1678, 1994
MASLOW ET AL
35. Sueta CA, Clarke SW, Dunlap SH, et al: Effect of acute magnesium administration on the frequency of ventricular arrhythmia in patients with heart failure. Circulation 89:660-666, 1994 36. Parikka H, Toivonen L, Pellinen T, et al: The influence of intravenous magnesium sulphate on the occurrence of atrial fibrillation after coronary artery bypass operation. Eur Heart J 14:251-258, 1993 37. Hecker BR, Lake CL, Kron IL, et al: Influence of magnesium ion on human ventricular defibrillation after aortocoronary bypass surgery. Am J Cardiol 55:61-64, 1985 38. Jacquiet L, Evenpoel M, Marenne F, et al: Hemodynamic effects and safety of sotalol in the prevention of supraventricular arrhythmias after coronary artery bypass surgery. J Cardiothorac Vasc Anesth 8:431-436, 1994 39. Fuller JA, Adams GG, Buxton B: Atrial fibrillation after coronary artery bypass grafting: Is it a disorder of the elderly. J Thorac Cardiovasc Surg 97:8211-8215, 1989 40. Leitch JW, Thomson D, Barrad DK, Harris PJ: The importance of age as a predictor of atrial fibrillation and flutter after coronary artery bypass grafting. J Thorac Cardiovasc Surg 100:338-342, 1990 41. Mendes LA, Connelly GP, McKenney PA, et al: Right coronary artery stenosis: An independent predictor of atrial fibrillation after coronary artery bypass surgery. J Am Coll Cardiol 25:198-202, 1995