Postoperative atrial fibrillation (AF) occurs in 18% to

Transcription

1 Oral Reduces Incidence of Postoperative Atrial Fibrillation Kushagra Katariya, MD, Eduardo DeMarchena, MD, and Hooshang Bolooki, MD Division of Cardiothoracic Surgery, University of Miami School of Medicine, Miami, Florida Background. Atrial fibrillation (AF) is a common occurrence after heart operations that use cardiopulmonary bypass. It can cause life-threatening complications as well as delay discharge and increase hospitalization costs. The purpose of this study was to evaluate the effect of orally administered low-dose amiodarone on the incidence of new onset postoperative AF. Methods. In this prospective study, 226 consecutive adult patients (group A) who had various heart operations utilizing cardiopulmonary bypass between April and November of 1998 at the University of Miami/ Jackson Memorial Hospital, were given oral amiodarone (200 mg three times a day), starting immediately after arrival in the intensive care unit until the day of hospital discharge. The incidence of new AF in this group of patients was assessed and compared with a historical group of 239 patients (group B) who had had cardiac operations with cardiopulmonary bypass in the preceding 9 months at the same institution. Results. Preoperative patient characteristics and procedure types were similar in the two groups. Among the 226 patients in group A, 13 (5.7%) had history of AF. Of the remaining 213 patients, new-onset AF occurred postoperatively in 10 (4.7%). Among the 239 patients in group B, 16 (6.7%) had history of AF. Of the remaining 223 patients, 44 (19.7%) developed new-onset AF (p < 0.001). A patients had a shorter length of hospital stay than those in group B (6.5 versus 7.8 days) but had a similar incidence of complications other than AF (23 of 226 patients in group A versus 24 of 239 in group B). The drug was well tolerated. Conclusions. Postoperative low-dose amiodarone given orally to patients who had cardiopulmonary bypass was well tolerated and appeared to reduce the incidence of new-onset AF and decrease the length of hospital stay. (Ann Thorac Surg 1999;68: ) 1999 by The Society of Thoracic Surgeons Presented at the Thirty-fifth Annual Meeting of The Society of Thoracic Surgeons, San Antonio, TX, Jan 25 27, Address reprint requests to Dr Bolooki, Division of Thoracic and Cardiovascular Surgery, University of Miami School of Medicine, PO Box (R-114), 1611 NW 12 Ave, ET 3072, Miami, FL Postoperative atrial fibrillation (AF) occurs in 18% to 53% of patients who have cardiac operations that use cardiopulmonary bypass [1]. The loss of normal atrial contraction reduces the cardiac output in most patients, and a rapid ventricular response increases myocardial oxygen consumption. An increased risk ratio for stroke or peripheral embolization also has been associated with postoperative AF [2 4]. Although prophylaxis of postoperative AF has been evaluated in many studies using multiple agents such as beta blockers, digoxin, diltiazem, quinidine, disopyramide, sotalol, and amiodarone, this arrhythmia remains an important cause of increased length of intensive care unit and hospital stay after heart operations [5 7]. is a class III anti-arrhythmic drug that has proved effective for prevention and treatment of AF. It can be taken safely orally after a high-dose preloading in patients with or without structural heart disease [8 11]. Previous studies in patients who had cardiac operations had used preoperative loading periods lasting up to 7 days [1, 8]. The purpose of this study was to evaluate the prophylactic effect of a short course of low-dose amiodarone, given orally after cardiac operations that used cardiopulmonary bypass, on the incidence of new-onset AF. Material and Methods Two hundred twenty-six patients who had heart operations with cardiopulmonary bypass at the University of Miami, Jackson Memorial Hospital during the 8-month period from April 1, 1998, through November 1, 1998, were studied prospectively after receiving oral amiodarone in a dose of 200 mg three times a day starting immediately after arrival in the intensive care unit until the day of discharge from the hospital. Patients were included in the study if they were in normal sinus rhythm at a rate greater than 65 beats per minute at the completion of operation and within a few hours after arrival in the intensive care unit. Patients were excluded from the study if they were dependent on a pacemaker (permanent or temporary) or had bradycardia (heart rate less than 65 beats per minute) for more than 12 hours. Patients who had heart transplants, pediatric patients, and those who had coronary operations without cardiopulmonary bypass were not included. The surgical procedures were done using a centrifugal pump and membrane oxygenation. Myocardial preservation was achieved with antegrade and retrograde cold blood cardioplegia and epicardial ice slush. A warm dose of cardioplegia was given a few minutes before removal of 1999 by The Society of Thoracic Surgeons /99/$20.00 Published by Elsevier Science Inc PII S (99)

2 1600 KATARIYA ET AL Ann Thorac Surg AMIODARONE FOR POSTOPERATIVE AF 1999;68: aortic cross-clamp. Postoperatively, all patients were cared for in an intensive care unit for 24 to 72 hours, and when sinus rhythm was maintained and a lower level of care was sufficient, they were transferred to a telemetry unit. treatment was discontinued on the evening before the discharge day. The routine postcardiac surgical protocol to prevent AF in our hospital was followed, which includes administration of beta blockers (metoprolol) when heart rate is more than 65 beats per minute in the presence of hemodynamic stability. The incidence of new-onset AF and other nonfatal complications as well as the length of hospital stay (LOS) in group A was assessed and was compared with a historical group of 239 consecutive patients (group B) who had had cardiac operations with cardiopulmonary bypass at the same institution and by the same surgical team during the preceding 9-month period (July 1, 1997, through March 31, 1998). New-onset AF was defined as documented episodes of supraventricular tachyarrhythmia or atrial arrhythmia lasting more than 5 minutes and leading to persistent or recurrent episodes of AF. History of AF was defined as a documented atrial fibrillation within the preceding 4 weeks requiring antiarrhythmic treatment or electrical cardioversion. In patients who had AF despite amiodarone prophylaxis (group A), additional doses of oral amiodarone (300 to 600 mg) were given. In addition, intravenous diltiazem or a short-acting beta blocker ventricular rate control was given. Intravenous procainamide was used followed by electrical cardioversion if rate control treatment was not successful and hemodynamic compromise ensued. Those patients who did not revert to sinus rhythm within a few hours despite therapy were given anticoagulation treatment with intravenous heparin. Follow-up was done during outpatient visits at 1 to 6 weeks postoperatively. Rhythm strip or 12-lead electrocardiograms were obtained when indicated. Hospital readmission records were reviewed when patients required rehospitalization to assess cardiac rhythm. Data are presented as mean standard deviation. Statistical analysis of data were done by Student s t test for unpaired observations, and results were considered significant at 95% confidence interval p value less than The study was done according to institutional protocols for use of Food and Drug Administration approved drugs in a nonrandomized prospective trial. Patients were informed of the use of amiodarone on a routine basis after April 1, Results Patient characteristics and procedures performed in both groups are shown in Tables 1 and 2. There were no significant differences among patient characteristics and types of operative procedures among patients in both groups. Atrial fibrillation was documented in 19 patients (19 of 218, 8.7%) taking amiodarone therapy (group A) and in 59 patients (59 of 239, 25%) in the historical group (group B) ( p 0.001). A history of paroxysmal AF was Table 1. Patient Characteristics Variable Age (y) Sex (male/female) 125/ /111 Left ventricular ejection fraction New York Heart Association class History of recent atrial 13 (5.7) 16 (6.7) fibrillation, n (%) Preoperative betablocker Chronic obstructive pulmonary disease Diabetes mellitus Chronic renal failure Bypass time (min) Cross-clamp time (min) a p 0.05 comparing groups A and B. present in 13 of 226 patients (5.7%) in group A and in 16 of 239 patients (6.7%) in group B. New-onset AF occurred postoperatively in 10 of the remaining 213 patients (4.6%) in group A and in 44 of 223 patients (19.7%) in group B; this difference was statistically significant ( p 0.001) (Table 3). Atrial fibrillation occurred at a mean of days postoperatively in group A compared with days postoperatively in group B ( p 0.05). The maximum ventricular rate response during AF in group A was beats per minute as opposed to beats per minute in group B ( p 0.05). The duration of AF was hours in group A patients versus hours in group B patients ( p 0.001). Of the 10 patients in group A who had new-onset AF Table 2. Types of Procedures Procedure CABG 138 (61%) 140 (58%) Valve operation 48 (21%) 55 (23%) Aortic Mitral Tricuspid 1 Double 3 3 Ross procedure 2 3 CABG/valve 20 (8.8%) 24 (10%) Aortic aneurysm 3 4 Left ventricular aneurysm 3 2 CABG/carotid 6 3 Atrial septal defect 5 6 Ventricular septal defect 1 3 Atrial myxoma 2 2 a p 0.05 for comparison with group B. CABG coronary artery bypass graft.

3 Ann Thorac Surg KATARIYA ET AL 1999;68: AMIODARONE FOR POSTOPERATIVE AF 1601 Table 3. Characteristics of Atrial Fibrillation Table 4. Postoperative Complications Characteristic (n 226) Complication History of recent AF a 13 (5.7%) 16 (6.7%) New onset AF b 10 (4.69%) 44 (19.73%) AF after discharge a 2 (0.98%) 6 (3.3%) Time of occurrence after surgery (days) c Duration of AF (h) b Maximum ventricular rate/min c a p b p between groups. c p AF atrial fibrillation. postoperatively, 3 patients (30%), after a short interval (30 minutes) of AF, converted spontaneously to sinus rhythm, but the other 7 patients (70%) were treated with anti-arrhythmic agents. One of these patients required electrical cardioversion and 1 patient had persistent AF with controlled ventricular rate. Of the 44 patients in group B who had new-onset postoperative AF, 8 (18.1%) converted spontaneously, after a short interval, to sinus rhythm, but the other 36 patients (81.8%) were treated with anti-arrhythmic agents. Four of these 36 patients (9.1%) required electrical cardioversion and 7 patients (15.9%) remained in AF with controlled ventricular rate. Among 13 patients in group A with history of recent AF preoperatively, 4 remained in normal sinus rhythm postoperatively, and in 9 postoperative AF was treated according to protocol. In 7 of them, sinus rhythm was attained. Among 16 patients in group B who had a history of recent AF, 1 remained in sinus rhythm postoperatively and 10 reverted to sinus rhythm with treatment. Five patients, however, remained in atrial fibrillation (4 of these patients had mitral valve disease). There was no significant difference in the incidence of nonfatal postoperative complications between the two groups (Table 4). Postoperative complications, other than AF occurred in 23 of 226 patients (10.1%) in group A and in 24 of 239 patients (10.0%) in group B ( p 0.05). The average length of stay for patients in group A was 6.5 days compared with 7.8 days for patients in group B. In group A the average length of stay for the 207 patients without AF was 6.1 days, whereas the length of stay was 6.4 days for the 181 patients without AF from group B ( p 0.05). For the 19 patients with AF (with history or new onset) from group A, the average length of stay was 11 days compared with 11.7 days for the 59 patients with AF (with history or new onset) from group B ( p 0.05) (Table 5). There were nine postoperative deaths (3.6%) in group A, of whom 1 patient had AF. In comparison, there were eight postoperative deaths (3.3%) in group B of whom, again, only 1 patient had new-onset AF. In group A, 8 patients (3.5%) had significant bradycardia (heart rate less than 65 beats per minute) after initiation of amiodarone administration, and the drug was discontinued after one to two doses. None of these Reexploration for 3 4 bleeding Cerebrovascular accident 1 0 or stroke Pneumonia or 3 4 respiratory failure Sternal infection or 8 7 dehiscence Bowel infarction 0 1 Permanent pacemaker 2 2 Ventricular tachycardia 6 5 Renal failure 1 0 Peripheral embolus 1 1 Total 23 (10.1%) 24 (10.0%) a p 0.05 compared with the historical group. patients had AF. Two of the 8 patients, both of whom had had cardiac valve replacement, required permanent pacemaker placement. Bradycardia resolved without incident after discontinuation of amiodarone in the other 6 patients. In comparison, 2 patients (0.83%) in group B, both of whom had had mitral valve replacement, needed implantation of a permanent pacemaker (Table 4). After discharge from the hospital, 2 other patients (0.98%) from group A developed AF, whereas 6 patients (3.3%) from group B developed AF after hospital discharge ( p 0.05). One of the 2 patients from group A who returned to the hospital with AF after discharge was successfully treated and attained normal sinus rhythm. Conversion to sinus rhythm was recorded in all 6 patients from group B who returned to the hospital with AF. Comment Atrial fibrillation, the most common form of supraventricular arrhythmia after cardiac operations, can lead to hemodynamic instability, peripheral embolization, or if left untreated, might evolve into life-threatening cardiac arrhythmia. The frequent consequence is prolonged hospitalization and increased medical costs. Atrial fibrillation commonly occurs in the first few days postoperatively and may revert to sinus rhythm spontaneously [2]. Table 5. Length of Hospital Stay Length of stay for all patients (days) Patients with AF (days) Patients without AF (days) AF atrial fibrillation. (n 226)

4 1602 KATARIYA ET AL Ann Thorac Surg AMIODARONE FOR POSTOPERATIVE AF 1999;68: Multiple factors have been implicated in its occurrence, including metabolic changes, temperature flux, fluid and electrolyte changes, high circulating catecholamine levels (particularly in patients treated with beta blockers preoperatively), lack of adequate atrial protection with cardioplegia during aortic cross-clamp, pericarditis, anesthetic agents, and preoperative use of digoxin, beta blocker, or diuretic agents, as well as surgical trauma or intensive care unit stress [2 8, 10, 12]. A higher incidence of AF in patients who have had beta blocker therapy that was withdrawn before the cardiac operation suggests that the discontinuation is a contributing factor to this complication [13]. Age has an important effect, and patients over age 60 years are more susceptible to AF after cardiac operations [3]. increases the refractory period of atrial and ventricular muscle as well as the atrioventricular node. It has mild beta blocker and calcium channel blocker activity in addition to its class III anti-arrhythmic activity. It has been effective in acute as well as chronic AF. Cardiac toxicity from amiodarone is uncommon [11, 14, 15]. The incidence of amiodarone-induced ventricular arrhythmia is low even in the presence of structural heart disease [16]. It has no negative inotropic effect, is a powerful afterload-reducing agent, and has coronary vasodilatory effects [17]. Our study showed that in all types of cardiac surgical patients amiodarone administration postoperatively effectively reduced the incidence of postoperative AF. This study confirmed the findings of others regarding the low complication rate from oral amiodarone in cardiac operations, because more than 96% of our patients tolerated it well. It is not known whether noncardiac toxic effects of amiodarone are dose related or dose unrelated. No acute pulmonary toxic effects of amiodarone were noted in these patients postoperatively. The half-life of amiodarone is 13 to 103 days [15, 16]. It can effectively reduce the incidence of AF after discharge from the hospital because of its long half life, although that was not borne out in our study because of the small number of patients with that complication. Bradycardia was seen in only 8 patients in the amiodarone group after an initial dose of the drug. Although this necessitated discontinuation of the drug per protocol, the bradycardia probably was unrelated to amiodarone use, especially in the 2 patients who required permanent pacemaker placement after valve replacement operations. In previous studies, amiodarone was given primarily to patients who had coronary bypass operations [2 4]. Recent studies have involved more heterogeneous populations of patients and especially have included patients with valvular heart disease. In a double-blind randomized trial, in 124 patients who had coronary bypass, Daoud and associates [1] showed a definite preventive effect of amiodarone. The drug had been given for 7 days preoperatively at a dosage of 600 mg per day and was continued at 200 mg per day postoperatively until the day of discharge. Postoperative AF occurred in 25% of patients in the amiodarone group and in 53% of patients in the placebo group ( p 0.003). Patients in the amiodarone group were hospitalized for fewer days than were patients in the placebo group (6.5 versus 7.9 days, p 0.04). Furthermore, total hospitalization costs were significantly less, and the incidence of AF after discharge was lower for the amiodarone group than for the placebo group. This study relied on large-dose preloading with amiodarone (median, 5.0 g) preoperatively and maintenance of a low dose after use of cardiopulmonary bypass. The incidence of AF in the placebo group of 53% is much higher than our study results. We believe the low AF incidence in our patients was from the routine use of metaprolol. Another factor contributing to the finding of low incidence of AF in our patients probably was the assessment of new-onset AF. When all the patients who had AF postoperatively are considered, the overall incidence of AF (with a history and new onset) in group A was 8.7% compared with 25% in the historical group B ( p 0.001). Hohnloser and colleagues [14] performed a prospective placebo-controlled study of intravenous amiodarone as prophylaxis against AF after heart operations in 77 patients. The amiodarone infusion began after the completion of the surgical procedure and significantly reduced the incidence of AF. However, in that study electrocardiographic monitoring was done only during the first 48 hours postoperatively, and amiodarone was discontinued in 18% of patients because of side effects. Intravenous amiodarone was associated with greater side effects in the form of drug intolerance (hypotension) and proarrhythmic complications [18]. We used a low-dose regimen and a duration of therapy that fits most patients in our practice. Most of our patients received their cardiac surgical procedure 24 to 48 hours after hospital admission. We also hypothesized that prophylaxis of AF after cardiac operation, compared with its treatment, might require a lower preloading dose of amiodarone. Because AF usually occurs about 2 days postoperatively, the regimen used in this study would allow the patients to receive more than 1 g of amiodarone before AF episodes. The results of our study showed a significant reduction in the incidence of postoperative AF from 19.7% to 5.7% ( p 0.001). The length of stay in these patients was also shorter compared with the control group (6.5 versus 7.8 days), which could result in cost savings. Postoperative new-onset AF was seen in almost one fifth of all patients in our historical control group despite the routine use of prophylactic therapy with beta blockers (metoprolol) in almost all patients with heart rates of 65 beats per minute or more in the immediate postoperative period. The postoperative protocol for beta blocker use was maintained for group A patients receiving amiodarone if postoperative heart rate remained higher than 65 beats per minute. Therefore, group A patients received amiodarone as well as metaprolol. We found that there was a short duration of AF (11.0 hours) in patients treated with amiodarone (group A). Clinical observation indicated that the shorter duration of AF was mainly due to quick response to additional therapy, such as intravenous calcium channel blockers

5 Ann Thorac Surg KATARIYA ET AL 1999;68: AMIODARONE FOR POSTOPERATIVE AF 1603 (diltiazem) or procainamide. It is well known that amiodarone enhances the effects of these drugs [18]. These results are similar to those reported by others [1, 7]. A limitation of our study is that it was a pilot project and was not randomized. However, we believe that a trial of various doses of this drug with different loading schedules both before and after cardiac operations is needed. Appropriate dosing and preloading regimens may be selected and used in randomized trials. We believe this preliminary report can guide future studies on this subject. They should involve all cardiac surgical procedures in patients older than 60 years of age and focus on new-onset AF, its duration, and response to therapy. We thank Jared Weiner, BS, Araceli Buendia, RN, and Sharon Campbell, RN, for their effort in accumulation of clinical data and Mrs Ely Gando for typing the manuscript. References 1. Daoud EG, Strickberger SA, Man KC, et al. Preoperative amiodarone as prophylaxis against atrial fibrillation after heart surgery. N Engl J Med 1997;337: Ormerod OJM, McGregor CGA, Stone DL, Wisbey C, Petch MC. Arrhythmias after coronary bypass surgery. Br Heart J 1984;51: Leitch JW, Thomson D, Baird DK, Harris PJ. The importance of age as a predictor of atrial fibrillation and flutter after coronary artery bypass grafting. J Thorac Cardiovasc Surg 1990;100: Caretta Q, Mercanti CA, De Nardo D, et al. Ventricular conduction defects and atrial fibrillation after coronary artery bypass grafting: multivariate analysis of preoperative, intraoperative and postoperative variables. Eur Heart J 1991; 12: Johnson LW, Dickstein RA, Fruehan CT, et al. Prophylactic digitalization for coronary artery bypass surgery. Circulation 1976;53: Yilmaz AT, Demírkliliç U, Arslan M, et al. Long-term prevention of atrial fibrillation after coronary artery bypass surgery: comparison of quinidine, verapamil and amiodarone in maintaining sinus rhythm. J Cardiac Surg 1996;11: McAlister HF, Luke RA, Whitlock RM, Smith WM. Intravenous amiodarone bolus versus oral quinidine for atrial flutter and fibrillation after cardiac operations. J Thorac Cardiovasc Surg 1990;99: Butler J, Harriss DR, Sinclair M, Westaby S. prophylaxis for tachycardia after coronary bypass surgery: a randomized double-blind placebo controlled trial. Br Heart J 1993;70: Chun S, Sager P, Stevenson WG, Nademanee K, Middlekauff HR, Singh B. is highly effective in maintaining NSR in refractory atrial fibrillation/flutter. J Am Coll Cardiol 1993;21:203A. 10. Middlekauff HR, Wiener I, Stevenson WG. Low-dose amiodarone for atrial fibrillation. Am J Cardiol 1993;72:75F 81F. 11. Nicklas JM, McKenna WJ, Stewart RA, et al. Prospective, double-blind, placebo controlled trial of low dose amiodarone in patients with severe heart failure and asymptomatic frequent ventricular ectopy. Am Heart J 1991;122: Aranki SF, Shaw DP, Adams DH, et al. Predictors of atrial fibrillation after coronary artery surgery: current trends and impact on hospital resources. Circulation 1996;94: Nattel S, Rangno RE, VanLoon G. Mechanism of propranolol withdrawal phenomena. Circulation 1979;59: Hohnloser SH, Meinertz T, Dammbacher T, et al. Electrocardiographic and antiarrhythmic effects of intravenous amiodarone: results of a prospective, placebo controlled study. Am Heart J 1991;121: Hondeghem LM. Class III agents: amiodarone, bretylium and sotalol. In: Zipes DP, Jalife J, eds. Cardiac electrophysiology: from cell to bedside. 2nd ed. Philadelphia: WB Saunders, 1995: Hamer AWF, Arkles LB, Johns JA. Beneficial effects of low dose amiodarone in patients with congestive cardiac failure: a placebo-controlled trial. J Am Coll Cardiol 1989;14: CôtéP, Bourassa MG, Delaye J, Janin A, Forment R, David P. Effects of amiodarone on cardiac and coronary hemodynamics and on myocardial metabolism in patients with coronary artery disease. Circulation 1979;59: Desai AD, Chun S, Sung RJ. The role of intravenous amiodarone in the management of cardiac arrhythmias. Ann Intern Med 1997;127: DISCUSSION DR MARKO I. TURINA (Zurich, Switzerland): With the application of amiodarone, sometimes one observes disturbing late arrhythmias, especially of the torsade type. That is the reason why some of these trials that involved preoperative loading with the amiodarone have been interrupted. Have you observed this type of arrhythmia in patients who had amiodarone therapy? DR ALVAN W. ATKINSON (Raleigh, NC): I have two questions. One, did you notice any response to treatment in the patients who did develop atrial fibrillation? Did they respond quicker with diltiazem or cardioversion than other patients, and did they maintain their sinus rhythm better after cardioversion? Second, did you notice any interaction or side effects with intravenous or oral diltiazem, as that is part of our standard protocol for radial artery harvesting, and would you have to modify the dosage of either drug if you used them as a standard of care in your hospital? DR REX DE L. STANBRIDGE (London, England): I would like to ask you first how you detected atrial fibrillation. Did you have the patient on continuous 24-hour monitoring, on a Holter, throughout their stay at the hospital, or was it for a shorter period of time only? Was it Holter-monitored or was it just monitored on a nurse bay? Second, oral amiodarone often takes a long time to act, to be absorbed, and I am surprised that with your administration it actually worked quickly enough by 2 days postoperatively. DR BOLOOKI: I thank the discussants for their pertinent questions. Doctor Turina, in response to your question, postoperative dysrhythmia treated with amiodarone was mainly in the form of bradycardia, for which we immediately stopped the drug. As I indicated, 8 patients received one or two doses of amiodarone before bradycardia occurred. We did not restart the drug in those patients. We observed the patients and used pacemaker therapy for that group. Two of them received permanent pacemakers. A number of valve patients were in our series, including those 2 patients who needed pacemakers; they had mitral valve replacement. We did not see any lifethreatening proarrhythmia, such as torsades de pointes or