Author + information
- Received August 28, 1996
- Revision received March 31, 1997
- Accepted April 16, 1997
- Published online August 1, 1997.
- Steven N. Singh, MDAB,*,
- Susan G. Fisher, PhDAB,
- Prakash C. Deedwania, MDAB,
- Prashant Rohatgi, MDAB,
- Bramah N. Singh, MDAB,
- Ross D. Fletcher, MDAB,
- for the Congestive Heart Failure–Survival Trial of Antiarrhythmic Therapy (CHF-STAT) Investigators (Veterans Affairs Cooperative Study No. 320)1
- ↵*Dr. Steven N. Singh, Veterans Affairs Medical Center, 50 Irving Street NW, Room 1E301, Washington, D.C. 20422.
Objectives. The purpose of this study was to evaluate the pulmonary effects of amiodarone in patients with heart failure, in those with chronic obstructive pulmonary disease (COPD) and in those undergoing a surgical procedure.
Background. Amiodarone has been known to cause pulmonary complications, especially in those with COPD and in those undergoing a surgical procedure.
Methods. Patients receiving vasodilator therapy for congestive heart failure were prospectively randomized to placebo or amiodarone at 800 mg/day for 14 days, 400 mg/day for 50 weeks and then 300 mg/day thereafter. Chest X-ray film and pulmonary function tests with diffuse capacity of carbon monoxide (DLCO) were obtained at baseline and annually. The power to detect a 20% difference in DLCO at 1 year exceeded 90% in all patients and in those with COPD (two-sided alpha = 0.05). The sample allowed a 75% power to detect pulmonary complications (1% vs. 5%) between the two treatment groups.
Results. There was no difference in baseline characteristics between patients randomized to amiodarone (n = 269) or placebo (n = 250). The DLCO measurements at randomization were 18.3 ± 6.9 and 17.7 ± 7.6 ml/min per mm Hg for the amiodarone and placebo groups, respectively (p = 0.3). At 1 and 2 years, DLCO measurements were 17.7 ± 7.0 and 18.3 ± 7.7 ml/min per mm Hg for the amiodarone group and 17.9 ± 7.2 and 18.2 ± 7.2 for the placebo group, respectively. There were no significant differences between the groups, with corresponding p values of 0.73 and 0.96 at years 1 and 2, respectively. Among patients with COPD, DLCO measurements at randomization were 17.9 ± 6.7 and 15.8 ± 6.8 ml/min per mm Hg for the amiodarone and placebo groups, respectively. At years 1 and 2, DLCO measurements were 16.6 ± 7.8 and 17.8 ± 9.5 ml/min per mm Hg for the amiodarone group and 16.5 ± 6.6 and 16.3 ± 7.0 ml/min per mm Hg for the placebo group, with corresponding p values of 0.95 and 0.48, respectively. There was no difference in survival free of noncardiac or perioperative deaths between patients assigned to amiodarone or placebo. Pulmonary fibrosis was diagnosed in four patients (1.1%) treated with amiodarone and in three patients (0.8%) receiving placebo.
Conclusions. Our study shows that amiodarone can be safely used, with an acceptable pulmonary toxicity, in patients with heart failure.
Amiodarone, a very complex antiarrhythmic agent with class III activity, has been used extensively to treat life-threatening arrhythmias. Its usefulness has been hindered by a variety of adverse effects, including pulmonary toxicity and respiratory failure, especially in patients with chronic obstructive pulmonary disease (COPD) [1–3]. It has been suggested that patients undergoing an operation while being maintained on amiodarone may be at risk for postoperative complications, including respiratory distress syndrome [4–6]. However, there is little information available from prospective, well-controlled studies to properly evaluate the extent of such risk. Accordingly, the purpose of this study was to identify the pulmonary effects of amiodarone in patients with heart failure, especially in those with a history of COPD, and in those undergoing an operation while participating in the prospective, randomized, placebo-controlled trial Congestive Heart Failure–Survival Trial of Antiarrhythmic Therapy (CHF-STAT).
The methods have been described previously . In brief, 674 patients with documented congestive heart failure, premature ventricular contractions (PVCs) ≥10/h and left ventricular ejection fraction ≤40% while receiving vasodilator therapy were prospectively randomized to placebo or amiodarone—800 mg/day for 14 days, 400 mg/day for 50 week and then 300 mg/day thereafter until the end of the study. Chest X-ray film and pulmonary function tests, including diffuse capacity of carbon monoxide (DLCO), were obtained at baseline and annually. Chronic obstructive pulmonary disease was defined as forced expiratory volume in 1 s/forced vital capacity (FEV1/FVC) ≤70% at randomization. Deaths were reviewed by a committee in blinded manner and classified as cardiac or noncardiac. The median time of follow-up was 45 months (range 0 to 54). Pulmonary fibrosis was defined according to chest X-ray film findings and the assessment of the principal investigator. Perioperative death was defined as death within 30 days from the operation.
1.1 Statistical analysis
This study had a power >95% to detect a 20% difference in DLCO at 1 year between amiodarone- and placebo-assigned patients (two-sided alpha = 0.05). The power to detect such differences among patients with COPD exceeded 90%. The sample allowed for a 75% power to detect a significant difference in pulmonary complications (1% vs. 5%) between the two treatment groups. Differences between drug groups were examined using the chi-square test for categoric variables and the Student ttest for continuous, independent measures. Data are summarized using frequencies, percentages and mean values ± SD. Changes in DLCO measurements were compared between the two groups by calculating the difference in DLCO levels from randomization to the 1-year assessment for each patient alive and evaluable at 1 year. The mean difference scores were then compared between the groups using the Student ttest. Survival time was calculated from the date of randomization to cardiac arrest. Surviving patients were censored on the last day of known follow-up status. Survival curves were constructed using the Kaplan-Meier method; differences between distributions were tested using the log-rank statistic. A two-sided alpha level <0.05 was considered statistically significant for all analyses.
In this analysis, 519 patients with DLCO measurements at baseline were evaluated: 250 received placebo and 269 received amiodarone. The baseline characteristics of the randomized patients by treatment group are shown in Table 1. There were no significant differences in the number of nonsmokers, current smokers and ex-smokers. The cigarette pack-years in current smokers randomized to placebo was 58 ± 35 and to amiodarone 55 ± 37 (p = 0.39). Chronic obstructive pulmonary disease was present in 82 (33%) and 81 (30%) patients randomized to placebo and amiodarone, respectively (p = 0.51).
The DLCO measurements at randomization for all patients receiving amiodarone and placebo were 18.3 ± 6.9 and 17.7 ± 7.6 ml/min per mm Hg, respectively (p = 0.3). At 1 and 2 years, the DLCO measurements were 17.7 ± 7.0 and 18.3 ± 7.7 ml/min per mm Hg for the amiodarone group and 17.9 ± 7.2 and 18.2 ± 7.2 ml/min per mm Hg for the placebo group (Table 2). These changes were not significant between the groups, with corresponding p values of 0.73 and 0.96 at years 1 and 2, respectively.
Among the 163 patients with COPD at randomization, the DLCO measurements were 17.9 ± 6.7 and 15.8 ± 6.8 ml/min per mm Hg for the amiodarone and placebo groups, respectively. At 1 and 2 years, the DLCO measurements were 16.6 ± 7.8 and 17.8 ± 9.5 ml/min per mm Hg, respectively, for the amiodarone group and 16.5 ± 6.6 and 16.3 ± 7.0 ml/min per mm Hg for the placebo group. There were no significant differences (p = 0.95 and p = 0.48 at years 1 and 2, respectively) between the groups at any point in time, as summarized in Table 2. However, in a matched analysis limited to patients alive at 1 year, there was a slightly greater decrease in DLCO from randomization to 1 year in patients receiving amiodarone compared with those given placebo (1.13 vs. 0.04 ml/min per mm Hg, p = 0.02) (Fig. 1). Also, in patients with COPD who survived at least 1 year, there was a significantly greater decrease in DLCO with amiodarone compared with placebo (2.05 vs. 0.09 ml/min per mm Hg, p = 0.008) (Fig. 1).
There was no evidence of a difference in survival free of noncardiac death between patients assigned to amiodarone and placebo (Fig. 2). There were 16 (6.4%) and 18 (6.7%) deaths in the placebo and amiodarone groups, respectively (p = 0.91). For patients with COPD, there was no detectable difference in survival free of noncardiac death (Fig. 3). There were 10 (10.3%) and 7 (7.1%) deaths in the placebo and amiodarone groups, respectively (p = 0.55). Of note, there was no difference in survival in patients with or without COPD (p = 0.75).
Nineteen patients underwent surgical treatment, including coronary artery bypass graft surgery. Thirteen patients received placebo and six received amiodarone. There was one perioperative death in a patient assigned to placebo. There were no perioperative deaths in the group receiving amiodarone. The median time to operation was 12 months (range 1 day to 44 months).
Pulmonary fibrosis as shown on chest X-ray film was seen in three patients (0.8%) receiving placebo and four patients (1.1%) treated with amiodarone. All three patients on placebo and three of the four on amiodarone died. Other pulmonary abnormalities, including effusion and consolidation, occurred in 1 patient (0.3%) on placebo and six (1.8%) assigned to amiodarone (Table 3). The one patient receiving placebo died. There was also one death among the six patients treated with amiodarone. The incidence of pulmonary complications in patients with and without COPD was similar (2.4 vs. 2.8%).
Our results show that there were no significant differences at any point in time in DLCO measurements in all patients assigned to amiodarone and placebo. In survivors only at 1 year was there a slight but significant difference in DLCO measurements between all patients and those with COPD. There was no difference between the two groups in terms of noncardiac mortality for all patients or those with COPD. In patients undergoing surgical treatment, noncardiac perioperative deaths were similar for the patients receiving amiodarone and placebo. The incidence of pulmonary fibrosis was similar for both groups, but other pulmonary toxicities were reported more frequently with the use of amiodarone.
Because of its amphiphilic nature, amiodarone may interfere with lysosomal phospholipases, leading to phospholipidosis and perhaps pulmonary toxicity . Also implicated is the direct effect on the lungs. It has been suggested that an immunologic reaction may also be producing a hypersensitivity response . In previous studies, the range of pulmonary toxicity varies from 5% to 10%, and death due to pulmonary complications has also been reported in the 5% to 10% range . Initial symptoms may include cough, dyspnea and fever, and chest X-ray film findings include patterns of alveolar and interstitial infiltrates. In some cases, gallium scans may show increased uptake over the lung parenchyma. Pulmonary function testing shows a restrictive pattern and a decrease in DLCO. Some studies have suggested that preexisting lung disease might predispose the patient to lung toxicity, although this has not been verified by other studies [1, 10]. In two other large, recent clinical trials—Gruppo de Estudio de la Sobrevida en la Insuficiencia Cardiaca en Argentina and Basal Antiarrhythmic Study of Infarct Survival—amiodarone’s pulmonary toxicity was also not observed, a finding consistent with our study [11, 12]. Moreover, the Polish Arrhythmia Trial and the Spanish Study on Sudden Death reported only one case each of adverse pulmonary effects [13, 14]. Based on these findings, it is obvious that in placebo-controlled trials in which investigators have no knowledge of the treatment, there was no inherent bias to make judgment errors in side-effect profiles. An 15% incidence of pulmonary toxicity over 3 years was reported in the Conventional Antiarrhythmic Amiodarone in Survivors of Cardiac Arrest Drug Evaluation trial, an open-label study . However, no deaths attributed specifically to pulmonary complication were reported.
Greenspon et al. published the relation between acute respiratory distress syndrome (ARDS) and amiodarone use in patients recovering postoperatively. Of the 18 postoperative survivors on amiodarone, nine (50%) developed ARDS, compared with 0 of 44 patients not on amiodarone and in whom similar surgical procedures were performed. In contrast, Chelimsky-Fellick et al. compared 29 patients on amiodarone with 29 control subjects undergoing heart transplantation. There were no differences between the two groups with respect to assisted ventilation, inotropic support or hospital stay. Moreover, there was no difference in mortality.
Our study did not address ARDS, hospital days or time on assisted ventilation, but it specifically looked at deaths and found no difference between placebo and amiodarone. However, it should be noted that the number of patients receiving amiodarone was small, and these results should be interpreted cautiously. We did not find any relation between the presence of COPD and a predisposition for amiodarone toxicity, except for a small but significant decrease in DLCO at 1 year.
From our study, it appears that amiodarone may be safely used, with an acceptable pulmonary risk side-effect profile, in patients with congestive heart failure.
We acknowledge Patricia Jones for assistance in the preparation of the manuscript.
↵fn1 This study was supported by the Department of Veterans Affairs Cooperative Studies Program of the Medical Research Service (Washington, D.C.), Sanofi Winthrop Recherche (Paris, France) and Wyeth-Ayerst Laboratories (Philadelphia, Pennsylvania).
- acute respiratory distress syndrome
- Congestive Heart Failure–Survival Trial of Antiarrhythmic Therapy
- chronic obstructive pulmonary disease
- diffuse capacity of carbon monoxide
- forced expiratory volume in 1 s/forced vital capacity
- premature ventricular contraction
- Received August 28, 1996.
- Revision received March 31, 1997.
- Accepted April 16, 1997.
- The American College of Cardiology
- Greenspon AJ,
- Kidwell GA,
- Hurley W,
- Mannion J
- Dean PJ,
- Groshart KD,
- Porterfield JG,
- Iansmith DH,
- Golden EB
- Burkart F,
- Pfisterer M,
- Kiowski W,
- et al.
- Ceremuzynski L,
- Kleczar E,
- Krzeminsk-Pakula M,
- et al.
- Chelimsky-Fallick C,
- Middlekauff HR,
- Stevenson WG,
- et al.