|
|
||||||||
Ann Thorac Surg 2003;75:223-230
© 2003 The Society of Thoracic Surgeons
a Division of Cardiovascular and Thoracic Surgery, Mayo Clinic, Scottsdale, Arizona, USA
b Section of Biostatistics, Mayo Clinic, Rochester, Minnesota, USA
* Address reprint requests to Dr Lanza, Division of Cardiovascular and Thoracic Surgery, Mayo Clinic Hospital, 5777 East Mayo Boulevard, Phoenix, AZ85054, USA
Presented at the Thirty-eighth Annual Meeting of The Society of Thoracic Surgeons, Fort Lauderdale, FL, Jan 2830, 2002.
| Abstract |
|---|
|
|
|---|
METHODS: We identified 31 patients who received prophylactic LDOA (200 mg by mouth every 8 hours) while hospitalized and 52 patients who received no prophylactic treatment. The groups were comparable for sex, age, comorbidities, and surgical procedure.
RESULTS: Twenty of 83 patients (24%) had postoperative atrial fibrillation: 17 of 52 patients (33%) without prophylaxis and 3 of 31 (9.7%) with prophylaxis (odds ratio, 0.221; 95% confidence interval, 0.059 to 0.829; p = 0.0253). The median total hospital charge was $30,800 (range, $20,400$96,900) for 50 patients without prophylaxis and $26,700 (range, $11,000$55,900) for 31 patients with prophylaxis (p = not significant). Patients receiving LDOA had lower accumulated charges per day of hospital stay (p = 0.0011).
CONCLUSIONS: LDOA prophylaxis significantly reduces the incidence of atrial fibrillation after pulmonary resection. Its use in this population may be cost-effective. Results of this pilot study provide a rationale for a prospective randomized trial.
| Introduction |
|---|
|
|
|---|
Previously authors have examined alternative strategies for minimizing the effect of postoperative AF. Amar and colleagues [1] recently published results of a prospective, double-blind, placebo-controlled trial using intravenous and oral diltiazem therapy in the postoperative period. The incidence of AF was reduced from 25% to 15% among patients who received diltiazem therapy. However, no difference in length of stay or cost could be identified. Despite recommendations to consider routine use of prophylaxis after pulmonary resections in patients older than 60 years, most thoracic surgeons currently use a therapeutic approach and do not routinely use prophylaxis for AF after pulmonary resections.
In cardiac surgery, multiple authors have reported the high efficacy and safety of oral amiodarone therapy for reducing the incidence of postoperative AF [24], and the use of prophylaxis to reduce postoperative AF has become more prevalent. Some have shown that its prophylactic use reduces length of stay and costs among patients undergoing cardiac operations.
We hypothesized that this approach should also prove effective in reducing the incidence of AF after pulmonary resection. Although low-dose oral amiodarone (LDOA) is not yet approved by the U.S. Food and Drug Administration for use in treating or preventing AF, published literature documenting its efficacy and safety for this application support its use.
| Material and methods |
|---|
|
|
|---|
All patients undergoing resection for cancer underwent mediastinal lymph node dissection. All patients had continuous cardiac monitoring during their hospital stay. Electronically stored rhythm histories were reviewed to identify time of initial AF, time to reach a heart rate (HR) less than 110 beats per minute (BPM), and time to normal sinus rhythm (NSR). Patients in whom AF developed had a confirmatory electrocardiogram when necessary during the initial episode, and select patients at risk had serial concentrations of cardiac enzymes measured to rule out myocardial infarction (MI). The use of therapeutic antiarrhythmic drugs for control of AF or electrical cardioversion, as well as recurrence of AF during hospitalization or readmission related to AF, was recorded. Complications related to the arrhythmia were identified.
To evaluate the association between LDOA prophylaxis and postoperative AF, data were analyzed for potential confounding preoperative risk factors, operative procedure variables, and associated morbidity. Cardiovascular risk factors considered for AF were presence of previous arrhythmia, heart diseases, use of preoperative heart medication, hypertension, previous MI, recent MI (within the past 6 months), and a previous cardiac operation or angioplasty. Preoperative electrocardiograms were analyzed for rhythm abnormalities. From the preoperative echocardiogram, ejection fraction was recorded. Preoperative pulmonary studies were also analyzed. The actual values were included in the data assessment, as well as the presence of desaturation greater than 2% between the values before and after exercise in preoperative oximetry, percentage of predicted forced expiratory volume in 1 second (FEV1%) less than 60, and percentage of predicted diffusion capacity of the lung for carbon monoxide (DLCO%) less than 50 (Table 1).
|
Statistical analysis
The association of postoperative AF (coded as "No" or "Yes") with the use of LDOA prophylaxis was assessed with logistic regression analysis (postoperative AF was the binary dependent variable). Various potential confounding risk factors (e.g., age and previous AF) were included in separate models; not all potential risk factors could be included in a single model because of the limited number of patients with postoperative AF. A linear multiple regression analysis was used to assess the association of hospital charges with the use of LDOA prophylaxis after first transforming these charges to a logarithmic scale. Part A, the hospital charge, and part B, the nominal charges for the operative procedure, were included in part C, the total hospital charges. Duration of hospital stay was included in the regression models, and an interaction term (LDOA prophylaxis by duration) was used to test for the association between the charge accumulation per day and the use of LDOA prophylaxis. Univariate associations between specific clinical variables or outcomes were based on univariate logistic regression models (e.g., postoperative AF vs a specific cardiovascular risk factor) and summarized as an odds ratio (OR), with 95% confidence interval (CI), computed from the parameter estimates of the logistic regression models.
| Results |
|---|
|
|
|---|
|
|
|
In 2 of 3 patients in whom AF developed while LDOA was used for prophylaxis, therapeutic amiodarone was administered intravenously, and LDOA was resumed after 12 hours of HR control. In the other patient, conversion to NSR was spontaneous. In all 3 patients, use of LDOA was extended until postoperative day 14. AF recurred in 1 patient at 26 hours (duration, 12 hours) and in another patient at 72 hours (duration, 8 hours). No patient presented with complications related to AF.
The 17 patients in the no-prophylaxis group in whom AF developed were treated with intravenous (IV) diltiazem (14 patients), IV amiodarone (2 patients), or oral amiodarone (400 mg every 8 hours) plus IV digoxin (1 patient). Diltiazem was administered as an IV bolus (0.25 mg/kg) over 20 minutes followed by a second IV bolus followed by a continuous infusion. After the HR was controlled, therapy was changed to oral diltiazem and was continued for 4 to 6 weeks. Digoxin was administered as a loading dose of 1 mg followed by a maintenance oral dose of 0.25 mg/d continued for 4 to 6 weeks. Amiodarone therapy was given as an IV dose of 150 mg over 30 minutes, followed by a continuous infusion, and then was converted to LDOA. Among the 14 patients treated with diltiazem, treatment was changed to administration of esmolol in 1 patient and procainamide in another patient because of persistent AF. Two patients required electrical cardioversion. In 1 of these 2, conversion to NSR did not occur. Anticoagulation was initiated, which resulted in bleeding complications requiring admission to the intensive care unit. In 1 patient, heart block developed, requiring pacemaker insertion. After conversion to NSR, AF recurred in 5 of 17 patients with no prophylaxis (29.4%) within a median of 49 hours (range, 14 to 216 hours).
In the group of 52 patients with no prophylaxis, complications developed in 16 (16/52; 31%); 7 of them (7/16; 43.8%) also had AF. There were no postoperative deaths. The following complications occurred in those 7 patients: respiratory failure in 1; pericarditis in 1; renal insufficiency in 1; MI in 1; prolonged air leak in 1; atelectasis, respiratory failure, gastrointestinal tract bleeding, and renal insufficiency in 1; and reoperation for bleeding, atelectasis, and pneumonia in 1. In the group of 31 patients with LDOA prophylaxis, complications developed in 9 (9/31; 29%), and 1 of them (1/9; 11%) had AF. There were 2 postoperative deaths not related to LDOA prophylaxis: 1 patient died of delayed (6 weeks) postpneumonectomy bronchopleural fistula at a different hospital after having received induction chemotherapy and radiation therapy for a stage III lung cancer and having an uneventful early postoperative recovery; a second patient died of a recurrent MI in the postoperative period after having a preoperative MI treated with percutaneous transluminal coronary angioplasty and stent placement 6 weeks preoperatively (Table 5).
|
|
| Comment |
|---|
|
|
|---|
The incidence of AF after pulmonary resection in the current series of patients is similar to that in other published reports [57]. LDOA prophylaxis was shown in our study to be associated with a statistically significant reduction in the occurrence of postoperative AF. The odds ratio for postoperative AF in patients receiving prophylactic LDOA therapy compared with those who did not receive prophylaxis was 0.221 (95% CI, 0.059 to 0.829; p = 0.0253). Moreover, with the use of prophylactic amiodarone, tachycardia was controlled faster and the duration of the arrhythmia was shorter. In addition, the median length of stay for the patients who developed AF while receiving LDOA (5 days) was shorter than the median for the study population of 83 patients (7 days) and was shorter than the median length of stay for the 17 patients who developed AF without prophylaxis (9.5 days). Although the overall total hospital charges were not statistically different between the prophylaxis and the control groups in this pilot study, the results indicated a slower rate of accumulation of hospital charges per hospital day, and the regression model predicted important cost savings at the median length of stay of 7 days in a typical patient. Together these results suggest that this prophylaxis strategy would be cost-effective in this population.
Various other prophylaxis strategies for AF after pulmonary resection have been described in the past. Digoxin use has been reviewed extensively: although it may be useful for controlling the ventricular rate after AF occurs, the narrow therapeutic window and ineffectiveness in preventing the onset of AF have resulted in abandonment of its use for prophylaxis. Moreover, an increase of postoperative AF and associated mortality with digitalis prophylaxis after pulmonary resection was reported for patients with no previous cardiac disease [8]. In a small, prospective, randomized trial comparing use of digoxin with use of calcium-channel blockers as prophylaxis for AF after pneumonectomy, AF developed in 32% of 31 patients receiving prophylactic digoxin, compared with 0% of 31 receiving diltiazem (p < 0.003). Because prophylactic therapy with diltiazem was considered a safe and effective means of decreasing the overall incidence of supraventricular arrhythmias after pneumonectomy, the use of digoxin was discouraged [9]. In a more recent, randomized, double-blind, placebo-controlled trial [1], AF developed in 15% of 167 patients receiving IV diltiazem prophylaxis, compared with 25% of 163 receiving placebo after pulmonary resection (p = 0.03). Because no difference in length of stay and cost was demonstrated, a therapeutic approach for AF after pulmonary resection continues to be the choice in most U.S. centers.
Other antiarrhythmics have also been investigated for AF prophylaxis. In a small, prospective, randomized trial after pulmonary resection, IV flecainide reduced the incidence of postoperative AF to 7% in 15 patients, compared with an incidence of 47% in 15 control patients receiving digoxin [10]. Cost-effectiveness was not examined. The significant proarrhythmia potential of flecainide limits its potential for use in atrial fibrillation prophylaxis.
Amiodarone is a class III antiarrhythmic that has been used for the treatment of ventricular arrhythmias [11] and recommended for the treatment of AF in patients with a low ejection fraction [12]. Recent publications have shown its efficacy and safety in preventing AF after cardiac operations. In a double-blind, randomized study, Daoud and colleagues [2] demonstrated a significant reduction in postoperative AF, costs, and length of stay for patients who received preoperative as well as postoperative oral amiodarone for elective cardiac operations. Guarnieri and associates [3] demonstrated in another randomized, double-blind study the efficacy and safety of IV amiodarone (1 g/d for 2 days) in reducing the incidence of AF after cardiac operations, but they were unable to demonstrate shorter hospital stays with its use. In a nonrandomized study using historical controls, Katariya and coworkers [4] demonstrated the efficacy and safety of LDOA therapy (200 mg 3 times daily) starting postoperatively and continuing until dismissal. In addition to reduction of postoperative AF in patients undergoing cardiac operations, length of stay was also shortened in patients receiving oral amiodarone in that study. Although oral amiodarone is absorbed slowly in humans, and its onset of action for treatment of ventricular arrhythmias may take 2 to 3 days, it is likely that much lower tissue concentrations are required for the prevention and treatment of atrial arrhythmias. As such, prolonged oral loading of amiodarone may not be necessary when used in this setting.
Long-term therapy and high doses of amiodarone have been associated with pulmonary toxicity, which may bias surgeons against its use after pulmonary resection [13, 14]. Amiodarone pulmonary toxicity unrelated to surgery has been described in 1% to 10% of patients receiving long-term therapy with high cumulative doses for ventricular arrhythmias [15]. Interstitial fibrosis with diffuse alveolar damage and phospholipidosis are identified by pathology studies [16]. The most common clinical presentation is a subacute or chronic infiltrate associated with cough, dyspnea, and weight loss. Lung toxicity from amiodarone is clinically reversible after the drug is stopped [17]. Van Mieghem and colleagues [18] reported the occurrence of acute respiratory distress syndrome (ARDS) associated with amiodarone use for AF prophylaxis after pulmonary resections. In that prospective study, 100 patients with pneumonectomy and 200 patients with lobectomy were to receive IV amiodarone (150 mg IV over 20 minutes followed by continuous infusion of 1,200 mg per 24 hours for 3 days), IV verapamil, or placebo. After 64 lobectomy and 32 pneumonectomy patients had been treated, the study was interrupted because ARDS occurred in 3 patients who had undergone right pneumonectomy. No patients with lobectomy or left pneumonectomy experienced pulmonary toxicity.
A retrospective review of 310 lobectomy, 108 right pneumonectomy, and 134 left pneumonectomy patients in a preceding 4-year period revealed an overall 7.6% incidence of postoperative ARDS among patients with right pneumonectomy and a 1.6% risk among patients with other types of resection. In the study population, postoperative ARDS developed in 3 of 11 patients with right pneumonectomy who received IV amiodarone. The authors concluded that amiodarone may be implicated in the development of ARDS after pulmonary operations and suggested avoiding the use of amiodarone to treat arrhythmia after pulmonary resection. Because of the higher amiodarone doses administered intravenously; because toxicity was observed exclusively in patients with right pneumonectomy, a group that is at high risk of postoperative pulmonary dysfunction; and because pathologic confirmation for amiodarone toxicity was not obtained, the conclusions of this study appear premature.
More recently, Ciriaco and coworkers [19] have reported the use of IV amiodarone for the treatment of supraventricular arrhythmia after lung resection. They concluded that amiodarone was safe and highly effective in establishing and maintaining sinus rhythm. Neither pulmonary toxicity or increase in the incidence of postoperative pulmonary complication was observed in any of our patients treated with LDOA prophylaxis.
Although there was no difference in complications unrelated to AF, admission to an intensive care unit because of AF-related complications occurred only in the group without prophylaxis. While none of the patients in either group was readmitted after dismissal because of AF, the median hospital stay was longer for the 17 AF patients without prophylaxis (10 days) than for the 3 AF patients receiving LDOA prophylaxis (5 days). Shorter duration of hospitalization may result in lower cost of treatment. Furthermore, the half-life of amiodarone may be of benefit by providing a protective antiarrhythmic effect up to 2 weeks after therapy is stopped.
Cost may be a relevant factor for the routine prophylactic use of amiodarone after pulmonary surgery. Oral amiodarone is significantly less expensive than parenteral amiodarone. The approximate retail cost of a 20-day course of LDOA (sixty 200-mg tablets) is $220; a 1-week course of prophylaxis therefore costs approximately $77 (personal communication, Mayo Clinic Hospital pharmacy). As such, perioperative prophylactic use of LDOA adds minimal costs to the hospitalization.
In our study, the median total hospital charge and the median duration of hospital stay were not statistically different between groups, although both trended lower in patients receiving LDOA. Patients with LDOA prophylaxis accumulated charges more slowly throughout their hospital stay, as reflected by the flatter slope of the log of accumulated hospital charges plotted against days of hospital stay. As the duration of stay increased, this group of patients had smaller accumulated billed charges than the group with no prophylaxis. Similar findings of lower hospitalization cost were reported for patients who received oral amiodarone prophylaxis for cardiac operations [2].
In summary, LDOA significantly reduces the incidence of AF following pulmonary resection. In our pilot study, no patient experienced significant complications from its prophylactic use. In patients for whom prophylaxis fails, the duration of AF was reduced and the length of hospital stay was shortened. Prophylactic therapy was associated with a slower rate of accumulation of hospital charges during the hospital stay. Our findings may be biased by the retrospective, nonrandomized nature of this study. Results of our pilot study support proceeding with a prospective randomized trial.
| Discussion |
|---|
|
|
|---|
DR LANZA: Well, certainly none that we could detect. We had to stop amiodarone therapy in one patient during hospitalization for nausea. Whether it was attributable to that or not, it is hard to say, but that is the only time we had to stop amiodarone therapy during the postoperative course.
DR KOHMAN: And do all the surgeons at your institution now use this?
DR LANZA: No, not yet.
DR WALTER WEDER (Zurich, Switzerland): I would like to make a very brief comment. We recently observed severe pulmonary toxicity in a patient under amiodarone 2 weeks after onset of treatment. The patient had to be intubated for several days until we made the diagnosis based on the findings in the bronchio-alveolar lavage, the computed tomography, and by exclusion of other factors. He recovered under high-dose steroids but is still on oxygen 3 months later.
DR LANZA: Certainly amiodarone has a potential for pulmonary toxicity. In previous reports, toxicity was reported in patients undergoing pulmonary resection, but that was using extremely high doses of intravenous amiodarone. In patients receiving low-dose amiodarone, which is widely used for atrial fibrillation, the incidence of pulmonary complications is minimal.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
S. C. Tomaszek, S. D. Cassivi, K. R. Shen, M. S. Allen, F. C. Nichols III, C. Deschamps, and D. A. Wigle Clinical Outcomes of Video-Assisted Thoracoscopic Lobectomy Mayo Clin. Proc., June 1, 2009; 84(6): 509 - 513. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. E. Roselli, S. C. Murthy, T. W. Rice, P. L. Houghtaling, C. D. Pierce, D. P. Karchmer, and E. H. Blackstone Atrial fibrillation complicating lung cancer resection J. Thorac. Cardiovasc. Surg., August 1, 2005; 130(2): 438 - 444. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Sedrakyan, T. Treasure, J. Browne, H. Krumholz, C. Sharpin, and J. van der Meulen Pharmacologic prophylaxis for postoperative atrial tachyarrhythmia in general thoracic surgery: Evidence from randomized clinical trials J. Thorac. Cardiovasc. Surg., May 1, 2005; 129(5): 997 - 1005. [Abstract] [Full Text] [PDF] |
||||
![]() |
V. Shrivastava, B. Nyawo, J. Dunning, and G. Morritt Is there a role for prophylaxis against atrial fibrillation for patients undergoing lung surgery? Interactive CardioVascular and Thoracic Surgery, December 1, 2004; 3(4): 656 - 662. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. A. Vaporciyan, A. M. Correa, D. C. Rice, J. A. Roth, W. R. Smythe, S. G. Swisher, G. L. Walsh, and J. B. Putnam Jr Risk factors associated with atrial fibrillation after noncardiac thoracic surgery: analysis of 2588 patients J. Thorac. Cardiovasc. Surg., March 1, 2004; 127(3): 779 - 786. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |