Amiodarone Significantly Decreases Atrial Fibrillation in Patients Undergoing Surgery for Lung Cancer

CARDIOTHORACIC ANESTHESIOLOGY: The Annals of Thoracic Surgery CME Program is located online at http://cme.ctsnetjournals.org. To take the CME activity related to this article, you must have either an STS member or an individual non-member subscription to the journal.

Amiodarone Significantly Decreases Atrial Fibrillation in Patients Undergoing Surgery for Lung Cancer Lars P. Riber, MD, PhD, Thomas D. Christensen, MD, DMSc, Henrik K. Jensen, MD, DMSc, Anette Hoejsgaard, MD, and Hans K. Pilegaard, MD Department of Cardiothoracic and Vascular Surgery and Institute of Clinical Medicine, and Department of Cardiology and Institute of Clinical Medicine, Aarhus University Hospital, Aarhus, Denmark

Background. Postoperative atrial fibrillation occurs in 5% to 65% of patients undergoing thoracic surgery. Although postoperative atrial fibrillation often is regarded as a temporary, benign, operation-related problem, it is associated with a twofold to threefold increase in risk of adverse events, including transient or permanent stroke, acute myocardial infarction, and death. Methods. A total of 254 consecutively eligible enrolled patients undergoing surgery for lung cancer were included in this randomized, controlled, double-blinded trial. Patients received 300 mg of amiodarone or placebo intravenously over 20 minutes immediately after surgery and an oral dose of 600 mg of amiodarone or placebo twice daily during the first 5 postoperative days.

Results. The patients in the amiodarone prophylaxis group had a reduction in the risk of atrial fibrillation of 23% (12 to 31); number needed to treat was 4.4 (3.1 to 7.8). A total of 38 in the control group and 11 in the amiodarone group experienced atrial fibrillation (p < 0.001). Adverse effects were observed in 10 patients equally distributed in both trial arms. Conclusions. Postoperative prophylaxis with a high dose of oral amiodarone after an intravenous bolus infusion is a safe, practical, feasible, and effective regimen for patients with lung cancer undergoing surgery. It significantly reduced the incidence of postoperative atrial fibrillation. (Ann Thorac Surg 2012;94:339 – 46) © 2012 by The Society of Thoracic Surgeons

B

cular Surgery, Aarhus University Hospital, we previously conducted a consecutive study in 100 patients and found an incidence rate of 30% for AF in patients with lung cancer undergoing lobectomy (unpublished data). Atrial fibrillation is most common on the second to third postoperative day [3, 6 – 8, 11, 23], but may occur at any postoperative day. Atrial fibrillation may cause hypotension, heart failure, and embolism to the brain causing apoplexies. Several studies have reported that postoperative AF has a worse short and long-term prognosis after lung surgery [3, 13, 18]. The mortality after pneumonectomy in lung cancer patients experiencing AF is up to 25% [18]. Cardiac arrhythmia significantly increases the length of hospital stay [19] and thus increases hospital cost [14], although results are not consistent [24]. Furthermore, the onset of AF often causes anxiety in patients, diminishes mobilization, increases medical costs, and induces more tests and examinations. Several studies have attempted to find predictors for AF after lung surgery without any clear conclusions. However, some associations have been postulated; surgery on the left lung [12, 16, 21], operation of

ailey first described cardiac arrhythmias related to lung surgery in 1943 [1]. Supraventricular arrhythmias are dominant, especially atrial fibrillation (AF). The etiology is unknown and possibly multifactorial. A combination of postoperative hyperadrenergic activity and atrial dilatation are most likely the best pathophysiological explanation, but increased vagus tonus, atrial inflammation, pulmonary hypertension, dilatation of the right ventricle, and hypoxemia or infection at the remaining bronchus tip may also contribute [1–5]. The prevalence and frequency of AF have been investigated in several retrospective studies [3, 6 –22], and the incidence of AF after pneumonectomy is as high as 46% [6]. In our center, Department of Cardiothoracic and VasAccepted for publication Dec 14, 2011. Presented at the Forty-eighth Annual Meeting of The Society of Thoracic Surgeons, Fort Lauderdale, FL, Jan 28 –Feb 1, 2012. Winner of the J. Maxwell Chamberlain Memorial Award for General Thoracic Surgery. Address correspondence to Dr Riber, Department of Cardiothoracic and Vascular Surgery & Institute of Clinical Medicine, Aarhus University Hospital, Brendstrupgaardsvej 100, 8200 Aarhus N, Denmark; e-mail: [email protected].

© 2012 by The Society of Thoracic Surgeons Published by Elsevier Inc

0003-4975/$36.00 doi:10.1016/j.athoracsur.2011.12.096

GENERAL THORACIC

J. MAXWELL CHAMBERLAIN MEMORIAL PAPER FOR GENERAL THORACIC SURGERY

GENERAL THORACIC

340

CHAMBERLAIN MEMORIAL PAPER RIBER ET AL AMIODARONE PROPHYLAXIS PREVENTS AF

Ann Thorac Surg 2012;94:339 – 46

Table 1. Baseline Characteristics of Patients Undergoing Lung Cancer Surgery

Characteristics

Amiodarone Control Group Group ␹2 Test (n ⫽ 122) (n ⫽ 120) p Value

Age, years (min, median, 46, 66, 82 35, 67, 84 max) Preoperative use of 17 14 ␤-blockers Male 60 56 Hypercholesterolemia 21 16 Hypertension 48 35 Diabetes mellitus 13 8 Current smoker 43 47 Chronic obstructive 13 11 pulmonary disease Previous myocardial 2 2 infarction Hgb (mmol/L) (min, 5.8, 8.3, 9.9 5.6, 8.2, 10.4 median, max) Creatinine (␮mol/L) (min, 33, 66, 278 7, 66, 163 median, max) Thyroid-stimulating hormone 0, 1, 6 0, 1, 14 (mU/L) (min, median, max) Hgb ⫽ hemoglobin;

min ⫽ minimum;

0.4 0.7 0.7 0.5 0.1 0.3 0.7 0.7 1.0 0.9 0.2 0.5

max ⫽ maximum.

malignant disease [9], male sex [9], impaired lung function [25], increased age [7, 11, 12, 23], the extent of tissue removed [3, 7, 11, 15], known cardiac disease [8, 14, 16], pleuropneumonectomy [12], former arrhythmias or acute myocardial infarction [14]. The treatment of postoperative AF varies and includes treatment with digoxin, beta-blockers, calcium antagonists, or class III antiarrhythmic drugs as amiodarone. Several studies have attempted to investigate whether antiarrhythmic drugs can prevent the risk of postoperative AF after lung surgery in patients with lung cancer [4, 21, 24, 26 –29]. Some centers have implemented prophylactic treatment in order to reduce the risk of postoperative AF [8]. No drug has yet shown superiority, but the most promising antiarrhythmic drug is amiodarone (Cordarone). Shorttime amiodarone prophylaxis is characterized by a very low risk of side effects and few contraindications. The most feared noncardiac side effect is pulmonary toxicity, which often manifests chronically or subchronically, but acute attacks have been described in the literature. The most frequent cardiac side effect is bradycardia [23]. Several studies have not found any toxicity regarding amiodarone [8, 23]. A retrospective case-control study by Lanza and colleagues [14] found a significantly reduced risk of AF after orally administered amiodarone and suggested that a randomized, controlled trial should be performed. No study has yet provided any clear evidence regarding the effect of prophylactic amiodarone or which regime should be used. Our hypothesis was that prophylactic amiodarone significantly reduces the risk of AF, why the aim of this study was to compare prophylactic amiodarone with placebo.

Material and Methods The study design was a randomized, controlled, doubleblinded trial. Among the 1,250,000 inhabitants in our region, approximately 200 patients are referred annually to Aarhus University Hospital, Denmark to undergo surgery for lung cancer. Consecutive patients were screened for eligibility and enrolled in the study. Inclusion criteria were the following: (1) Elective lobectomy or pneumonectomy; (2) age above 18 years; (3) willingness to participate; and (4) provision of informed consent. Exclusion criteria were the following: (1) Previous heart or lung surgery; (2) resting heart rate below 40 beats per minute; (3) hypotension with systolic blood pressure below 80 mm Hg; (4) atrialventricular blockage of any degree or sick sinus syndrome; (5) preoperative AF or flutter; (6) known previous AF or flutter lasting more than 1 month; (7) hepatic dysfunction (alanine-aminotransferase level more than twice the upper normal limit); (8) hyperthyroidism; (9) pregnancy; (10) breastfeeding; (11) in treatment with monoamine oxidase inhibitors (MAOI); (12) QTc interval longer than 440 ms for men and 460 ms for women; or (13) known adverse reactions to amiodarone. The regional scientific ethics committee approved the study. The study complied with the Helsinki II declaration and all patients consented to participate. The study was conducted according to the standards of good clinical practice and was monitored and approved by the good clinical practice unit at Aarhus University Hospital.

Patients Two hundred and fifty-four eligible, consecutively enrolled patients undergoing elective lung cancer surgery were included. Based on a computerized prospective randomization schedule they were assigned to receive either the intervention (amiodarone) or placebo. Randomization was performed in variable blocks and stratified by age (ⱕ65 years and ⬎65 years) and preoperative use of ␤-blockers.

Table 2. Perioperative Data of Patients Undergoing Lung Cancer Surgery Perioperative Characteristics Right side lobectomy Bilobectomy Pneumonectomy Video-assisted thoracic surgery Reoperation T3–T4 tumor stage Operation duration (minutes) (minimum, median, maximum) Intubation time (minutes) (minimum, median, maximum)

Amiodarone Group (n ⫽ 122)

Control Group (n ⫽ 120)

␹2 test p Value

71 4 2 17

68 4 4 24

0.9 1.0 0.4 0.2

5 17a 55, 128, 352

4 23b 53, 139, 350

1.0 0.4 0.9

93, 182, 566

106, 193, 409

0.5

a

b 118 patients, 119 patients, as some patients had metastatic, carcinoid, or non-Hodgkin’s tumors.

Tumor stage is calculated on a smaller sample.

341

CHAMBERLAIN MEMORIAL PAPER RIBER ET AL AMIODARONE PROPHYLAXIS PREVENTS AF

Table 3. Relative and Absolute Effects Including Numbers Needed to Treat Endpoints Atrial fibrillation Symptomatic atrial fibrillation CI ⫽ confidence interval;

Amiodarone Group

Control Group

RD (95% CI)

NNT (95% CI)

Relative Risk (95% CI)

9% 2.5%

32% 9.2%

23% (13–33) 6.7% (0.9–13)

4.4 (3.1–7.8) 15 (8–116)

3.5 (1.9–6.5) 3.7 (1.2–12)

NNT ⫽ number needed to treat;

RD ⫽ risk difference.

Each patient received a randomization number, which was recorded and sent by fax to the pharmacy at Aarhus University Hospital, along with the patient’s Civil Registration Number. The pharmacy decoded the number, prepared the appropriate infusion and pills, and forwarded them (together with a sealed opaque envelope containing the randomization assignment) to the Department of Cardiothoracic and Vascular Surgery. Baseline characteristics are shown in Table 1.

Intervention Immediately after surgery and transfer to the ICU the study participants received a bolus infusion of 300 mg of amiodarone or placebo (5% aqueous dextrose solution) over 20 minutes, together with their first oral dose of either 600 mg amiodarone or placebo. The oral dose of 600 mg amiodarone or placebo was administered twice daily (8 a.m. and 8 p.m.) for 5 days. Patients in the intervention group received a total of 6.3 g amiodarone. If the patients were unable to swallow the capsules, a nurse from another department opened them and dissolved the content in sterile water. If possible, patients on anticoagulation (eg, acetylsalicylic acid, warfarin) treatment discontinued this 5 days preoperatively except for heparin treatment and continued their other regular medication, including ␤-blockers, until the day of surgery. Anticoagulation therapy and additional medications like ␤-blockers were resumed on the first postoperative day. Patients were continuously monitored during surgery and in the ICU until transferred to the ward. Holter monitoring was continued for the first 5 postoperative days. Additional electrocardiograms were obtained in case of clinical signs of AF (eg, irregular pulse, fainting, confusion, shortness of breath) or subjective manifestations (eg, palpitations, dizziness, fatigue, lightheadedness). The study drug was discontinued and appropriate treatment was initiated if AF or adverse reactions such as drop in systolic blood pressure below 80 mm Hg or drop in heart rate below 40 beats per minute developed. Patients were routinely discharged on the fourth to seventh postoperative day to either their own home or their local hospital. The patients’ local hospital provided discharge summaries for the study. On postoperative day 30, patients were contacted by a study physician and asked if they had developed any signs of AF. During the same call, patients were asked to contact their general practitioner to have an electrocardiogram performed; this was subsequently sent to the study staff. If the patient reached a study endpoint before postoperative day 30, the observation period stopped. All operations were performed in general anesthesia with propofol and fentanyl combined with epidural morphine

analgesia through an anterior muscle-sparing thoracotomy or by video-assisted thoracoscopic surgery. All patients were intubated with a Carlens double-lumen tube.

Endpoints The primary endpoint was time to AF; secondary endpoint was time to symptomatic AF. We calculated the risk difference and number needed to treat with 95% confidence intervals. In addition, the frequency of possible drug-related side effects (ie, bradycardia or hypotension) was examined during the first 30 postoperative days. Patients were evaluated for occurrence of endpoints on postoperative days 7 and 30.

Atrial Fibrillation The definition of AF was fast, irregular, eddy current activation of the atrium with neutralization of its contractions. Electrocardiogram showing totally irregular ventricle rhythm with a frequency around 150 to 200 beats per minute, displaying a line flickering with narrow QRS complexes. Symptomatic AF was defined as the patient’s subjective discomfort (palpitations, angina pectoris, shortness of breath, or fainting) or the need for acute intervention due to hypotension or heart failure. Electrocardiogram was performed in all cases to verify the diagnosis. Asymptomatic AF was defined as random AF detected on electrocardiogram or by Holter monitoring. Only AF lasting more than 5 minutes was defined as an event. In case of AF, possible contributing factors (ie, pleura effusion, electrolyte derangement, fever, volume overload) were examined and corrected and the opaque sealed envelope was opened to reveal the prophylactic regime of the patient, and to treat the patient in accordance with the following department guidelines: (1) hemodynamically stable patients on amiodarone continued treatment and were subsequently Table 4. Postoperative Atrial Fibrillation in Patients Undergoing Lung Cancer Surgery Atrial Fibrillation Characteristics Atrial fibrillation, n (%) Symptomatic, n (%) Asymptomatic, n (%) Time of occurrence after surgery, hours (minimum, median, maximum) Atrial fibrillation duration, hours (minimum, median, maximum)

Amiodarone Group (n ⫽ 122)

Control Group (n ⫽ 120)

␹2 Test p Value

11 (9%) 3 (27%) 8 (73%) 62, 96, 140

38 (32%) 11 (29%) 27 (71%) 20, 59, 225

0.001 0.029 0.001 0.135

0.1, 3.6, 12

0.1, 0.6, 50

0.701

GENERAL THORACIC

Ann Thorac Surg 2012;94:339 – 46

GENERAL THORACIC

342

CHAMBERLAIN MEMORIAL PAPER RIBER ET AL AMIODARONE PROPHYLAXIS PREVENTS AF

DC-converted within 48 hours; (2) hemodynamically unstable patients on amiodarone were given an intravenous dose of amiodarone or subacute DC-converted; (3) patients in the placebo group received oral amiodarone treatment or intravenous amiodarone if hemodynamically unstable, and then they were DC-converted within 48 hours. After transfer to the local hospital and subsequent discharge, treatment followed local guidelines of the department responsible for the patient.

Statistical Analyses and Sample Size Baseline data and perioperative characteristics were compared using the Wilcoxon Mann-Whitney rank sum test for variables. The Pearson ␹2 test and the Student t test were used to compare proportions. Intention-to-treat analyses of primary and secondary endpoints were conducted in both groups. Kaplan-Meier

Ann Thorac Surg 2012;94:339 – 46

curves were used to display time free of AF data, and Cox proportional hazards regression was used to estimate relative risks and to calculate p values and 95% confidence intervals, which are presented in the parentheses subsequent to the result (Tables 1,2,4). The proportional hazard assumption was evaluated using Schoenfeld residuals. The 7 and 30 day incidence rates, risk differences, and number needed to treat were calculated with corresponding 95% confidence intervals. Data were analyzed using Stata 9 software (StataCorp, College Station, TX). The level of type I error was set to 0.05 and type II to 0.20, allowing for 80% power. The a priori hypothesis that AF risk would be reduced from 30% in the placebo group to 15% in the intervention group required inclusion of 121 patients in each trial arm. The estimated exclusion rate after initial inclusion was 5% (12 patients). Thus, a total of 254 consecutive patients was needed.

Consecutive patients enlisted for lobectomy (n=386)

Excluded (n=144) Met the exclusion criteria (n=89) Refused participation (n=5) Subacute operation (n=38) No lobectomy (n=10) Preoperative atrial fibrillation (n=2)

Included and randomized (n=242)

Allocated to amiodarone prophylaxis (n=122)

Discontinued intervention (n=5) Postoperative nodal rhythm (n=1) Asystole (n=1) Bradycardia (n=1) Hypotension (n=1) AV-blockage grade 2 (n=1)

Intention-to-treat analysis (n=122) Per-protocol analysis (n=117) Fig 1. Trial flow chart. (AV ⫽ atrioventricular; n ⫽ numbers.)

Allocated to placebo prophylaxis (n=120)

Discontinued intervention (n=5) Mors (n=1) Sinus arrest (n=1) Bradycardia (n=1) Hypotension (n=1) AV-blockage grade 3 (n=1)

Intention-to-treat analysis (n=120) Per-protocol analysis (n=115)

CHAMBERLAIN MEMORIAL PAPER RIBER ET AL AMIODARONE PROPHYLAXIS PREVENTS AF

343

One patient in the placebo group died during the trial due to an acute myocardial infarction. No patients developed transient or persistent stroke. Five patients in each group experienced possible side effects that led to discontinuation of the study medication. There was no difference between the 2 groups, no side effects could be traced to the active prophylactic regime, and no pulmonary toxicity was found.

Comment

Fig 2. Symptomatic and asymptomatic atrial fibrillation in the intervention and the control group.

Results The trial flowchart is shown in Figure 1. From September 2008 to March 2011, 386 patients underwent lobectomy or pneumonectomy at Aarhus University Hospital, Denmark. A total of 132 patients were excluded subsequently because of the inclusion or exclusion criteria (n ⫽ 89), refusal to participate (n ⫽ 5), or subacute operation (n ⫽ 38). Of the 254 remaining patients, 10 patients did not undergo a lobectomy or pneumonectomy due to inoperability and 2 had preoperative AF (n ⫽ 2) and were thus excluded after the initial inclusion; thus, perioperative exclusion rate was 4.7%. A total of 122 patients were randomized to the intervention group. The study drug was discontinued in 10 participants equally distributed in the trial arms, due to the possibility of adverse reactions in terms of hypotension, bradycardia, nodal rhythm and AV-blockage. Baseline data are shown in Tables 1 and 2. All data were equally distributed in the 2 groups. Patients in the placebo group had a 3.5 (1.9 to 6.5) times higher risk of developing AF compared with patients in the amiodarone group. The overall number needed to treat was 4.4 (3.1 to 7.8), and the absolute risk reduction was 23% (13 to 33), decreasing from 32% to 9% (Table 3). Absolute numbers are displayed in Figure 2 and time free of AF is shown in a Kaplan-Meyer curve (Fig 3). The distribution of symptomatic AF also differed significantly between the 2 trial arms, with a relative risk of 3.7 (1.2 to 12) for patients in the placebo group. The overall number needed to treat was 15 (8 to 116) and the risk reduction was 6.7% (0.9 to 13), decreasing from 9.2% to 2.5%. Eleven patients receiving amiodarone and 38 patients receiving placebo developed AF (p ⬍ 0.001) (Table 4). Of the patients with AF, 27% had symptomatic AF in the amiodarone group versus 29% in the placebo group. There was no statistically significant difference between the 2 groups with respect to outcome labeled “time of occurrence after surgery” or “atrial fibrillation duration.” Before hospital discharge, all episodes of AF were converted to sinus rhythm by means of additional antiarrhythmic medication or electrical cardioversion. All patients had sinus rhythm at day 30. Atrial fibrillation occurred at a median of 68 hours (0.1 to 225 hours) after surgery with a median duration of 45 minutes (0.1 to 50 hours).

In this randomized, controlled double-blinded trial we found that our regimen of prophylactic amiodarone administered to patients undergoing lobectomy, bilobectomy, or pneumonectomy significantly reduced the risk of postoperative AF. The results were statistically significant, stressing the efficiency of using high-dose oral amiodarone for 5 postoperative days after an initial intravenous loading bolus. These results may therefore be expected to lead to an improved short-term and long-term prognosis after lung cancer surgery. The use of prophylactic amiodarone after cardiac surgery has been well documented in the literature [30] but no randomized controlled trial has previously been performed in order to obtain a feasible, practical prophylactic regime for patients undergoing lobectomy, bilobectomy, or pneumonectomy due to lung cancer. Postoperative AF generally occurred on the second or third postoperative day, which is the time when the blood concentration of amiodarone should be maximal in our regime. Accordingly, we have documented a clinically feasible way to significantly reduce the risk of AF after lung cancer surgery. Postoperative time to onset and duration of AF were similar in both trial arms. The duration of AF is probably more influenced by the attending physician’s urge to intervene rather than by the effect of the medication. Most of our patients were operated through an anterior thoracotomy and only a few patients by video-assisted thoracoscopic surgery. In our population of patients undergoing video-assisted thoracoscopic surgery the risk of AF seems to be reduced as the risk of AF in the placebo group was 17% compared with overall risk of 32%; the risk in the amio-

Fig 3. Kaplan-Meier survival estimates in each trial arm.

GENERAL THORACIC

Ann Thorac Surg 2012;94:339 – 46

GENERAL THORACIC

344

CHAMBERLAIN MEMORIAL PAPER RIBER ET AL AMIODARONE PROPHYLAXIS PREVENTS AF

darone group was 12% compared with 9% overall. Due to a very low number of patients, only 24 patients, in the placebo group and 17 patients in the amiodarone group it was statistically insignificant. Larger studies including more patients are thus needed. Serious prophylactic-related side effects were not encountered; however, our study of 254 patients was too small to observe rare side effects. The strengths of this study are the reliable study design and the feasibility of drug administration within the routine of a general ward setting. The setup did not interfere with routine admission and discharge schedules and was thus easy to implement and did not require any changes in the preoperative medications. This study was designed to look for all attacks of AF lasting more than 5 minutes. The establishing of the AF diagnosis after hospital discharge required that the patients had symptomatic AF, for which they visited their general practitioner. This might reduce the validity of our AF data. However, our study population was Holter monitored for 5 days, ascertained from medical discharge summaries, and from a telephone interview with the treating physician at day 30. At this time, all patients had an electrocardiogram performed and the result was forwarded to, and interpreted by, a study physician blinded to the prophylactic regime. Accordingly, there is a small risk that we have overlooked some asymptomatic cases in the late postoperative phase after hospital discharge. However, randomization makes it unlikely that risk underestimation would differ statistically between the 2 groups. Moreover, no potential confounders were found. The authors wish to acknowledge Vibeke Laursen, the project nurse who participated in enrolling and randomizing patients and in conducting follow-up.

References 1. Bailey CC. Cardiac arrhythmias following pneumonectomy. N Engl J Med 1943;229:356 –9. 2. Kirsh MM, Rotman H, Behrendt DM, Orringer MB, Sloan H. Complications of pulmonary resection. Ann Thorac Surg 1975;20:215–35. 3. von Knorring J, Lepäntalo M, Lindgren L, Lindfors O. Cardiac arrhythmias and myocardial ischemia after thoracotomy for lung cancer. Ann Thorac Surg 1992;53:642–7. 4. Ritchie AJ, Bowe P, Gibbons JR. Prophylactic digitalization for thoracotomy: a reassessment. Ann Thorac Surg 1990; 50:86 – 8. 5. Cerney C. The prophylaxis of cardiac arrhythmias complicating pulmonary surgery; a preliminary report. J Thorac Surg 1957;34:105–10. 6. Curtis JJ, Parker BM, McKenney CA, et al. Incidence and predictors of supraventricular dysrhythmias after pulmonary resection. Ann Thorac Surg 1998;66:1766 –71. 7. Asamura H, Naruke T, Tsuchiya R, Goya T, Kondo H, Suemasu K. What are the risk factors for arrhythmias after thoracic operations? A retrospective multivariate analysis of 267 consecutive thoracic operations. J Thorac Cardiovasc Surg 1993;106:1104 –10. 8. Ciriaco P, Mazzone P, Canneto B, Zannini P. Supraventricular arrhythmia following lung resection for non-small cell lung cancer and its treatment with amiodarone. Eur J Cardiothorac Surg 2000;18:12– 6.

Ann Thorac Surg 2012;94:339 – 46

9. Keagy BA, Lores ME, Starek PJ, Murray GF, Lucas CL, Wilcox BR. Elective pulmonary lobectomy: factors associated with morbidity and operative mortality. Ann Thorac Surg 1985;40:349 –52. 10. Polanczyk CA, Goldman L, Marcantonio ER, Orav EJ, Lee TH. Supraventricular arrhythmia in patients having noncardiac surgery: clinical correlates and effect on length of stay. Ann Intern Med 1998;129:279 – 85. 11. Rena O, Papalia E, Oliaro A, et al. Supraventricular arrhythmias after resection surgery of the lung. Eur J Cardiothorac Surg 2001; 20:688 –93. 12. Harpole DH, Liptay MJ, DeCamp MM Jr, Mentzer SJ, Swanson SJ, Sugarbaker DJ. Prospective analysis of pneumonectomy: risk factors for major morbidity and cardiac dysrhythmias. Ann Thorac Surg 1996;61:977– 82. 13. Amar D, Burt M, Reinsel RA, Leung DH. Relationship of early postoperative dysrhythmias and long-term outcome after resection of non-small cell lung cancer. Chest 1996; 110:437–9. 14. Lanza LA, Visbal AI, DeValeria PA, Zinsmeister AR, Diehl NN, Trastek VF. Low-dose oral amiodarone prophylaxis reduces atrial fibrillation after pulmonary resection. Ann Thorac Surg 2003;75:223–30. 15. Breyer RH, Zippe C, Pharr WF, Jensik RJ, Kittle CF, Faber LP. Thoracotomy in patients over seventy years. J Thorac Cardiovasc Surg 1981;81:187–93. 16. Mowry FM, Reynolds EW Jr. Cardiac rhythm disturbances complicating resectional surgery of the lung. Ann Intern Med 1964;61:688 –95. 17. Beck-Nielsen J, Sørensen HR, Alstrup P. Atrial fibrillation following thoracotomy for non-cardiac diseases, in particular cancer of the lung. Acta Med Scand 1973;193:425–9. 18. owka MJ, Pairolero PC, Trastek VF, Payne WS, Bernatz PE. Cardiac dysrhythmia following pneumonectomy. Clinical correlates and prognostic significance. Chest 1987;91: 490 –5. 19. Wahi R, McMurtrey MJ, DeCaro LF, et al. Determinants of perioperative morbidity and mortality after pneumonectomy. Ann Thorac Surg 1989;48:33–7. 20. Ghosh P, Pakrashi BC. Cardiac dysrhythmias after thoracotomy. Br Heart J 1972;34:374 – 6. 21. Borgeat A, Biollaz J, Bayer-Berger M, Kappenberger L, Chapuis G, Chioléro R. Prevention of arrhythmias by flecainide after noncardiac thoracic surgery. Ann Thorac Surg 1989;48:232– 4. 22. Massie E, Valle AR. Cardiac arrhythmias complicating total pneumonectomy. Ann Intern Med 1946;26:231–9. 23. Barbetakis N, Vassiliadis M. Is amiodarone a safe antiarrhythmic to use in supraventricular tachyarrhythmias after lung cancer surgery? BMC Surgery 2004;4:7. 24. Amar D, Roistacher N, Rusch VW, et al. Effects of diltiazem prophylaxis on the incidence and clinical outcome of atrial arrhythmias after thoracic surgery. J Thorac Cardiovasc Surg 2000;120:790 – 8. 25. Stougard J. Cardiac arrhythmias following pneumonectomy. Thorax 1969;24:568 –72. 26. Shields TW, Ujiki GT. Digitalization for prevention of arrhythmias following pulmonary surgery. Surg Gyn Obst 1968;126:743– 6. 27. Van Mieghem W, Coolen L, Malysse I, Lacquet LM, Deneffe GJ, Demedts MG. Amiodarone and the development of ARDS after lung surgery. Chest 1994;105:1642–5. 28. Amar D, Roistacher N, Burt ME, et al. Effects of diltiazem versus digoxin on dysrhythmias and cardiac function after pneumonectomy. Ann Thorac Surg 1997;63:1374 – 81. 29. Juler GL, Stemmer EA, Connolly JE. Complications of prophylactic digitalization in thoracic surgical patients. J Thorac Cardiovasc Surg 1969;58:352– 60. 30. Zebis LR, Christensen TD, Thomsen HF, et al. A practical regimen for amiodarone use in preventing postoperative atrial fibrillation. Ann Thorac Surg 2007;83:1326 –31.

CHAMBERLAIN MEMORIAL PAPER RIBER ET AL AMIODARONE PROPHYLAXIS PREVENTS AF

345

DISCUSSION DR KEMP KERNSTINE (Duarte, CA): I want to thank Dr Mack and the Society for allowing me to review this paper. I also want to congratulate the authors for completing this prospective trial. There is a great deal of difficulty and effort and forethought that goes into these. I also wanted to thank them for providing me with a copy of their manuscript and their PowerPoint presentation ahead of this meeting. As the authors describe, atrial fibrillation is a significant health care problem in the postoperative resection patient, resulting in significant morbidity, mortality, and health care resource utilization, and worthy of investigation. Their prospective doubly blinded randomized clinical trial has demonstrated that amiodarone reduces the risk of atrial fibrillation by 23%, without any identifiable increased risk. There are a few questions that I have. Number one, your endpoint, atrial fibrillation for 5minutes, appears to be clear in your manuscript, but were there any patients in which it was difficult to confirm atrial fibrillation from some other supraventricular or ventricular arrhythmia? If so, how was atrial fibrillation confirmed? The second is, your group chose a dose of 300 mg given over 20 minutes immediately after surgery, together with 600 mg of amiodarone given orally, and then 600 mg b.i.d. [twice a day] for the next 5 days. In 2009 at the 45th Annual Meeting of the Society of Thoracic Surgeons in San Francisco, Tisdale presented the results of their prospective trial at the University of Indiana [Tisdale JE, Wroblewski HA, Wall DS, et al. Ann Thorac Surg 2009;88:886 –95], used an amiodarone dose of 1,050 mg started at the time of induction and then continued for the next 24 hours. They then continued 400 mg b.i.d. for the next 6 days until the time of discharge. They found a similar result to yours; a reduced risk of atrial fibrillation, from approximately 32% down to 14%, an 18% reduction. How did you decide on the dose and the length of time for the use of amiodarone? Do we change the dose according to body weight or surface area? And finally, in 1994, Van Mieghem [Van Mieghem W, Coolen L, Malysse I et al. Amiodarone and the development of ARDS after lung surgery. Chest 1994;105:1642–5] published the findings of an aborted prospective clinical trial evaluating the use of amiodarone to reduce the risk of atrial fibrillation in patients who underwent lung resection. They found in the treatment arm a higher rate than expected of ARDS, especially in those patients who had pneumonectomy. This so impressed those that put together the thoracic surgery practice guidelines for the use of amiodarone, they recommended against its use in pneumonectomy patients outside of clinical trials. Pneumonectomy was performed in only 2 of your patients. Do we now recommend the use of amiodarone prophylaxis in patients undergoing a pneumonectomy? After all, pneumonectomy patients are at the greatest risk of postoperative atrial fibrillation. Thank you. DR RIBER: Thank you, Dr Kernstine, for your remarks. You asked how we differentiated atrial fibrillation from other supraventricular arrhythmias. All our patients were on Holter monitor and we had the normal staff in the cardiac department going through all the Holter monitors. We believe that we have a good cardiac department and they gave us the right numbers of atrial fibrillation. But it is always the discussion what is what, but these were evaluated by 2 excellent cardiologists. Regarding Tisdale’s paper and why we settled on a different postoperative way of handling it with a lower intravenous dose.

This is mainly due to the fact that Tisdale had 24 hours of infusion. We believe that it is not in the best interests for patients to have an infusion for a long time, as this diminished the possibility to mobilize. The bolus we used was therefore given over 20 minutes. This increased the fast-track mobilization and diminished the length of stay at the intensive care unit. Furthermore, a long infusion time extend the time the patients have to be monitored due to the risk of arrhythmias. Twenty minutes gives a more feasible time. Furthermore, we chose this scheme since we performed a similar project in patients undergoing coronary artery bypass grafting in 2007, where we showed that this was an effective regime for those patients as well to diminish the risk of atrial fibrillation. We have also reviewed the relevant literature and found that a bolus infusion giving a high blood dose of amiodarone continued by oral prophylaxis seemed to be effective even though no randomized trials had been performed. It was previously shown to be effective for coronary artery bypass grafting patients so we thought it would be equally effective for patients undergoing operation for lung cancer, and, as our study shows, it seems that we were correct. Regarding the third question about the early trials. Van Mieghem infused approximately 4 g of amiodarone. We believe that the high volume of intravenous administrated amiodarone could be the reason why they experienced adverse events. Our trial is though too small to give a scientific answer on how to handle pneumonectomy patients. Studies show that one-fourth of the patients who experience atrial fibrillation while undergoing pneumonectomy will actually die in-hospital, why correct handling of these patients and avoidance of atrial fibrillation is critical. Maybe we should look into the recommendations regarding postoperative handling of pneumonectomy patients, but this trial is not powered to present valid data confirming amiodarone as a good prophylaxis, but at the same time we did not find any adverse events regarding the regime used, why amiodarone very well could have a place in prophylaxis for atrial fibrillation in patients undergoing pneumonectomy. DR GLENN WHITMAN (Baltimore, MD): I, too, thought that was an excellent presentation. I have 2 questions. One, it seemed that only 10% of the patients received a perioperative ␤ blocker. Was there any additive effect of amiodarone with the ␤ blocker? Secondly, one of the problems associated with new onset, postoperative atrial fibrillation occurs when it persists for more than 2 to 3 days. At that point, one may require Coumadin therapy for 1 to 2 months. Was there a difference between groups regarding duration of atrial fibrillation, and requirement for discharge anticoagulation? DR RIBER: Regarding the second question, nobody was discharged with atrial fibrillation. One of the patients did, though, experience atrial fibrillation after discharge from our hospital but that patient was in sinus rhythm as well after discharge from the local hospital. Regarding Coumarin treatment, we have the same guidelines in Denmark but none of our patients were candidates for it as they were in sinus rhythm. I forgot the first question. Can you repeat the first question? DR WHITMAN: The first question dealt with the issue of whether a ␤-blocker had any separate effect from amiodarone, or an additive effect.

GENERAL THORACIC

Ann Thorac Surg 2012;94:339 – 46

GENERAL THORACIC

346

CHAMBERLAIN MEMORIAL PAPER RIBER ET AL AMIODARONE PROPHYLAXIS PREVENTS AF

DR RIBER: Well, from the studies regarding ␤-blocker and amiodarone, we know that it has an additive effect, but we haven’t particular looked into this in our study. Most of our patients aren=t in ␤-blocker treatment as their lung function doesn=t allow it. This is furthermore probably one of the reasons why amiodarone is more effective in diminishing the risk of atrial fibrillation in patients undergoing lung surgery than for patients undergoing coronary bypass, as these patients most commonly are in ß-blocker therapy. DR JAMES EDGERTON (Plano, TX): I congratulate you also on an important study well done, with significant contributions. My question is also about your dosing. Your slides went by pretty fast. I was trying to pick up when your peak incidence of onset of atrial fibrillation was. Although you will obtain high serum levels with the intravenous dosing you suggested, most electrophysiologists would tell us that we are not going to get effective myocardial levels of amiodarone until the patient has received at least 5 g on board, and 10 g is better. If I did my math right, you are not going to achieve 5 g until the middle of your fourth postop day, and you are not going to achieve 10 until about your ninth postop day. So I am just wondering if you think that if you started dosing the patients as outpatients, a week earlier, your results may have been better. DR RIBER: Thank you for an excellent question. Most of the background in the literature regarding atrial fibrillation is done on patients going through cardiac bypass grafting. So far there has not been a lot of studies investigating atrial fibrillation and thoracic surgery. When I did my PhD, I went through all those papers regarding atrial fibrillation and cardiac surgery, and these studies showed no sign of that an early preoperative start with prophylaxis had an enhanced effect. Furthermore, a regime like that would not be feasible in a clinical setting and no study has yet shown superiority regarding early prophylactic start. Oh, I forgot to answer one of Dr Kernstine’s questions: “Did we change the dose according to body weight or surface area?” We did not even though we know that that oral treatment has a bioavailability between 15 and 85%. So maybe we can make an even more effective prophylaxis regime, if we are looking into how high a blood dose the patients have during the postoperative phase, but that would be costly and not feasible in a clinical setting and I really don’t think that it would change the effect that much eventually. DR RALPH J. DAMIANO (St. Louis, MO): Congratulations on an excellent study. I have 2 quick questions. First of all, since you showed no reduction in either length of stay or adverse events with prophylactic treatment with amiodarone, can you really recommend it? You basically showed no effect except that you decreased the incidence of postoperative atrial fibrillation, but you had no effect, at least from what I could see, on any other objective endpoints. So is it worth the cost of giving prophylactic amiodarone and overtreating the 70% of patients who never get atrial fibrillation? That is my first simple question. The next question is, did you measure QT interval, because amiodarone has a pro arrhythmic effect that can result in fatal ventricular arrhythmias. How many patients had QT prolongation and when would you stop the amiodarone? I didn’t see that it was ever stopped, but certainly we see that problem with the use of amiodarone in adult cardiac surgery patients. DR RIBER: Thank you for the questions. Regarding the QT, we stopped if the QT length was more than 460 msec in women and

Ann Thorac Surg 2012;94:339 – 46

440 in males. We had some patients where we discontinued the study drug due to the possibility of adverse events, but, as you see, there was just as many in the placebo group as in the active treatment group. Maybe amiodarone has some adverse events, but this study was too small to identify them. On the other hand, we didn’t find any data showing that amiodarone prophylaxis was harmful in our regime. Regarding “Why?”, that’s the question I always get, and if you ask the patients, they know why: they don’t like to experience atrial fibrillation. Furthermore, as I mentioned before, I have done a similar project on patients undergoing coronary artery bypass grafting. I just made a 6-year follow-up on those patients and found that atrial fibrillation actually is correlated to worse long-term outcome regarding death. Furthermore, this follow-up study showed that amiodarone was safe even after 6 years of follow-up. If this population experience the same, I=ll tell you why we should try to diminish the risk of atrial fibrillation for other reasons than to avoid subjective symptoms in the patients. I believe that atrial fibrillation is a poor prognostic factor regarding long-term outcome, and the patients will gladly avoid having atrial fibrillation anyway. But I don=t think it will be cost effective in a short perspective, but in the long run it depends upon how much a life is worth. DR KEITH B. ALLEN (Kansas City, MO): I enjoyed your talk. We published a similar benefit with oral amiodarone in open heart patients, but your dose of 600 mg orally b.i.d. seems high. In our study, 400 mg wasn’t tolerated very well by about a third of the patients. Can you comment on how many of your patients actually were able to keep 600 mg down? There is a lot of nausea with that size dose. DR RIBER: Thank you. We didn’t have a big problem regarding nausea and vomiting despite the high dose, and we do not have it when we are using it as prophylaxis for our coronary artery bypass grafting patients or as atrial fibrillation treatment for coronary valve operated patients either. If you look into the literature, some suggest that amiodarone is correlated to gastrointestinal problems but most data are not conclusive. Furthermore, most patients have nausea after surgery due to intubation. If a patient vomited any pills they got an extra capsule. So they all received the full oral prophylaxis. DR WHITMAN: A couple of points. One, I disagree fully with Ralph that because you didn’t show a difference there wasn’t a difference. I don’t think this study was powered to show a difference, and I think atrial fibrillation, in and of itself, is a great surrogate for what might be a stroke in this population. Secondly, our cardiologists tell us that the prolongation in the QT interval with amiodarone is much more benign than with other drugs, and we allow our patients to go into the 550 to even higher millisecond range before we will get upset with the QT prolongation with amiodarone. And finally, we also load with amiodarone IV [intravenous] the first night after surgery, and then we switch to 600 b.i.d. until they get a total of approximately 7 g orally or 3 to 5 g total intravenously before we feel that they are loaded. We also use 600 mg b.i.d. as our oral load the following day when the patients are taking PO, and, without any data, it seems to be well tolerated. DR RIBER: Thank you for your comment, and I must fully agree. Furthermore, if a patient not in prophylaxis experiences atrial fibrillation in our department, we use the same scheme for treating them and it often works brilliantly.