Amiodarone pulmonary toxicity

Amiodarone

pulmonary

toxicity

Of the side effects that complicate amiodarone therapy, pulmonary fibrosis is potentially the most serious. Therefore, the development of techniques to predict the onset of this troublesome reaction would be of great practical value. Reports of 39 patients who developed pulmonary toxicity with amiodarone were evaluated for clues to precipitating factors and information on the response to corticosteroid treatment. The majority of patients were betng given maintenancedoses >400 mg/day. Patients appeared to improve after withdrawal of amiodarone, both with and without corticosteroid treatment. in addition, a case report is presented of a patient who developed pulmonary changes that disappeared when amiodarone was withdrawn and did not recur when amtodarone was reinstituted. Data from sequential pulmonary function tests and cumulative amiodarone dosage in 35 patients were also examined to determine their value in predicting pulmonary comptications. Pulmonary function tests did not appear to be useful in predicting the likelihood of an individual patlent’s developtng pulmonary complications. Although none of the available information identifies the mechanism mediating amiodarone pulmonary toxicity, the frequency of the complication probably can be reduced by timely reductions in maintenance dosage. (AM HEART J 106:906, 1963.)

Louis Rakita, M.D., Samuel M. Sobol, M.D., Nelson Mostow, M.D., and Thomas Vrobel, M.D. Cleveland, Ohio, and San Francisco, Calif.

Amiodarone has been reported to cause a variety of cardiac’s 2 and extracardiac 4 side effects. Of the extracardiac side effects, the most serious is pulmonary toxicity, a usually reversible but sometimes fatal complication of respiratory failure.5 Amiodarone has been implicated in pulmonary toxicity, even though there is no direct link, because (1) no other cause could be found for the reactions that occurred; (2) radiographic lesions cleared when amiodarone was discontinued or the dose reduced; (3) lung biopsies revealed changes similar to those seen in other drug reactions; (4) as in other druginduced pulmonary reactions, regression may follow steroid administration; and (5) specific therapy for other possible pulmonary disorders failed to reverse the process. We summarize current knowledge of the pulmonary complications of amiodarone based on our own experience and on a review of the literature and solicited unpublished information.6 In addition, we present new data relating to two aspects of management of patients receiving amiodarone therapy: the results of serial pulmonary function testing and the

From the Department of Hospital and Case Western Letterman Army Medical University of California, San Reprint Hospital,

requests: Louis 3395 Scranton

Medicine, Cleveland Metropolitan General Reserve University, the Cardiology Service, Center, and the Department of Medicine, Francisco.

Rakita, M.D., Rd., Cleveland,

Cleveland OH 44109.

Metropolitan

General

relationship to cumulative amiodarone dose, and the results of readministration of amiodarone in a patient with presumed amiodarone-induced pulmonary toxicity. REVIEW OF PUBLISHED

CASES

There were no reports of pulmonary complications from amiodarone in the extensive international literature published from 19627 to 1980. Additionally, there have been no reports in several recent large studies.*-” The first mention of the possible association of pulmonary toxicity with amiodarone was made in 1980 by Rotmensch et al.,12 who found that pulmonary infiltrates observed on x-ray studies resolved when amiodarone was withdrawn and steroids administered. These investigators saw this complication in only one additional patient in the more than 400 they treated.13 Since those cases, there have been reports of 37 additional patients worldwide with pulmonary complications associated with amiodarone therapy; data are given in Table I. Heger et a1.5 noted pulmonary infiltrates on chest x-ray studies in two of 45 patients, and found fibrosing alveolitis on lung biopsy in one of these patients, who subsequently died of severe respiratory insufficiency. A third patient, who died of congestive heart failure, was found at autopsy to have similar pulmonary histologic changes. Two other patients developed pleural effusions. We summarized the findings and clinical course in

Volume Number

106 4, Part 2

six patients who developed pulmonary complications among 432 patients then being given amiodarone in the United States,14 including-the two previously reported by Heger et al5 All six patients had received 1600 mg amiodarone daily for 1 to 8 months. Biopsy specimens in four patients showed interstitial fibrosis and pneumonitis. Two patients died; both had severe congestive heart failure in addition to pulmonary parenchymal changes. The remaining four patients received corticosteroid therapy and had radiographic resolution from 1.5 to 4 months after withdrawal of amiodarone. In 70 patients given amiodarone ~500 mg/day for ~1 month, Marchlinski et all5 noted evidence of pulmonary toxicity in four patients after 6 to 11 months. One of these patients died of severe respiratory insufficiency and underlying severe ischemic heart disease. The remaining three patients recovered in 2 to 3 months after amiodarone was withdrawn and corticosteroids were given. McGovern et a1.16reported interstitial pneumonitis in four of 80 patients given a mean dose of amiodarone 550 mg after 5 to 36 months (mean 14 months) of therapy. Kudenchuk et al6 noted radiographically defined pulmonary fibrosis in one of 70 patients given a mean dose of 564 mg/day for an average of 6.7 months. Individual cases17-19have been reported that document resolution of amiodarone-associated pulmonary infiltrates after corticosteroid treatment. Resolution has also been noted on withdrawal of amiodarone alone or on reduction of the dose without the addition of steroids.6~16,21-23 Two deaths occurred despite steroid therapy among five additional patients with biopsy-proved pulmonary fibrosis, but the data presented were too limited to fully assess the relationship to amiodarone.24 Clinical findings. Early symptoms in most patients with pulmonary toxicity included exertional dyspnea, nonproductive cough, and weight loss, and only occasionally low-grade fever, but no sweats or chills. Several patients have had associated muscle weakness. Pleuritic pain has occurred but is uncommon. Physical findings usually include rales or decreased breath sounds. Rarely, a pleuritic friction rub may be heard. Uncommonly, patients have demonstrated neuropathy14 or clinical thyroid dysfunction.15 The white blood cell count in patients with pulmonary complications is often normal to moderately elevated, but may be markedly elevated.*6*20 Eosinophilia has not been found. The erythrocyte sedimentation rate is elevated in most cases. Liver function studies may be normal, but lactic dehydrogenase may be mildly elevated.20

Amiodarone

Radiographic diffuse bilateral patchy alveolar gest pulmonary culosis.ls Rarely, thickening may

pulmonary

toxicity

907

findings characteristically consist of interstitial changes and diffuse or infiltrates. Chest x-ray findings sugedema,Q l4 pneumonitis,‘l or tuberpleural effusion or areas of pleural be found.3*22 Relationship to dosage. Amiodarone-induced pulmonary toxicity may be, at least in part, dose related. Although six patients developed infiltrates with only 400 mg/day,12, 16,17,*O 33 well-documented cases have occurred in patients receiving maintenance doses of ~600 mg/day. Marchlinski et all5 noted that their four cases occurred exclusively in a group of 23 patients receiving amiodarone maintenance doses >400 mg/day, whereas no pulmonary abnormalities occurred among 47 patients given maintenance doses of 5400 mglday. Suiuez et al.‘l describe a patient who tolerated amiodarone 200 mgjday uneventfully for 5 years, but developed respiratory symptoms and pulmonary infiltrates within 8 months of progressive dose increases to 800 mg/day for recurring arrhythmias. Rotmensch et all3 remark that their two cases of pulmonary fibrosis occurred in patients receiving 400 to 800 mg/day, whereas more than 400 other patients, usually given 200 to 400 mg/day, had no pulmonary complications. Rotmensch also points to the rarity of reports of pulmonary toxicity in Europe, where amiodarone has been used for more than 20 years,25 but generally in maintenance doses of 200 to 400 mg/day. This compares with the larger number of cases occurring among fewer patients in the United States. As the drug is still investigational in the United States, only resistant arrhythmias have been treated, and generally with higher doses. Finally, resolution of pulmonary complications with cessation of therapy follows a time course that precedes the complete elimination of amiodarone from tissues. This can be inferred from the recognized long elimination half-life of the drug26 and the reports of continued arrhythmia suppression well after resolution of radiologic changes.ls Other predisposing factors. Possibly other as yet unidentified factors play a mediating role, for example, individual genetic or metabolic features that affect patient susceptibility, incidental viral exposure, unrecognized drug interaction, or preexisting pulmonary disease. Whether the severity of the underlying cardiac disease and the preexistence of congestive heart failure contribute to susceptibility also is not clear. Those few patients who have died after the appearance of pulmonary toxicity have generally had severe cardiac dysfunction in addition

908

Rakita et al.

Table

I. Reports of pulmonary toxicity in patients given amiodarone

Reference

American

Patients

(n)

Rotmensch et al.*‘~”

2

Heger et al.> Sobol et al.14

6

1

Riley et al.” Wright et al.*O Morera et a1.24 Rasmussen et a1.23

1 1 5 I

Marchlinski et al.“’ Waxman et aLI*

5

Schlaeffer et al.** McGovern et al.16

4

Kudenchuk

9

1

et al.*

Zaher et a1.19 Suarez et al.“’ Rakita et a1.‘2 CAD, Coronary

artery

1 1 1 disease: CHD, congestive

heart disease; MVP,

Case

Diagnosis

1 2 3 4 5 6 7 8 9 10 11 12 to 16 17

CAD MVP CAD CAD CAD CHD CAD RHD CAD -

18 19 20 21 22 23 24 25 26 27 28 29 to 30 31 to 36 37 38 39

-

mitral

to arrhythmia, and a history of congestive heart failure. The fatal outcome may have resulted from a combination of the pulmonary insult superimposed on an already severely compromised cardiopulmonary system. These few instances, however, cannot be taken as an indication that amiodarone should be avoided in patients with congestive heart failure; a substantial percentage of all patients in most series have congestive failure, yet respond well to the drug without pulmonary effects or worsening of the faiiure. Indeed, patients with both severely impaired cardiac performance and high-grade ventricular arrhythmias are those most at risk of sudden death and thus most likely to benefit from amiodarone. Response to corticoeterdd therapy. The role of corticosteroid therapy in the management of amiodarone-induced pulmonary toxicity remains unde-

valve prolapse;

CAD WPW CAD WPW

RHD,

October, 1983 Heart Journal

Age ivri

50 67 56 61

55 40 67 64 71 53 to 75

55 44 67 73 54 60 56 49 50 59 -

-

-

CAD WPW CAD

46 56 63

rheumatic

heart disease; WPW, Wolf-Parkinson-White

Sex

M M M M M F M F M M M M M M M F M F M F -

M F M

fined. Steroids are advocated in other drug-induced pulmonary disorders,z7 and have been used in about half of reported cases related to amiodarone. In some instances steroid administration may have contributed to or hastened improvement.12* 17~1s~20~2* In some patients, however, resolution of amiodarone-induced pulmonary changes have been demonstrated without steroid therapy.6*‘6,2*-22 If the mechanism of production of pulmonary changes is not immunologic, the action of corticosteroids in this condition is not clear. If the inflammatory changes and fibrosis noted on biopsyI are secondary to a metabolic alteration, steroid-mediated suppression of chronic inflammation might be benefmial. Until further clarification, it would seem advisable to recommend a short course of corticosteroid therapy as well as discontinuation of amiodarone in pulmonary complications.

Volume Number

106 4, Part 2

Amiodarone

pulmonary

toxicity

909

Amiodarone Maximum dose

in&day) 400 -

Maintenance dose Cmglday)

Total dose 694

400 600 to 800 800 600 600 600 800 600 400 400 <600

128.2 14.6 31.2 142.2 134.6 44.4 15.4 (?)292.0 -

-

-

-

1400 1400 1400 1400 1400 600 1000 400 400 400 1200 1200 1200 800 800 1400

800 800 600 600 600 200 to 600 1000 400 400 400 600 600 Variable 800 800 600 to 1000

800 1400 600 600 1400 600 400 400 -

PREDICTIVE TESTS

VALUE OF PULMONARY

Duration of therapy 1

12 -

189.6 214.0 123.2 211.6 177.8 90.0 80.4 84.0 472.4 ?120.0 495.8 181.5

FUNCTION

Pulmonary function tests (PFT) in ptitients with amiodarone-related infiltrates have shown hypoxia, restrictive changes with reduction in total lung capacity, and diffusion abnormalities.15* l7 Kudenchuk et al.6 performed serial PFT in 69 patients given generally high doses of amiodarone (mean 564 mg/day) for 6 months, and found that total lung capacity was reduced from 97 % to 91% of predicted value. Reductions >15% occurred in 10 patients. Diffusion capacity decreased by >15% in 19 patients. Furthermore, these investigators suggested that patients with preexisting impairment of diffusion capacity (DLCO <80% of predicted value), total lung capacity <80% of predicted value, and lung changes on x-ray studies might be at increased risk of developing pulmonary toxicity. Prospective studies. To determine the predictive value of PFT in identifying patients at increased

27 6 1 2 a 4 2.5 1.5 12 to 72 7 9 6 11 13 3 3 12 5 36 <12 5 67 8

(no)

Time recovery

of (ma)

0.3 1 Died 4 Died Died 1.5 1.5 3.0 Died after 3 mo 2 Died 3 Recovered 2 Died 2 2 2 1 to 2 18 0.5 to 2

Died Died 7 Recovered 2 2 3

Steroids Yes No Yes No No Yes Yes Yes Yes Yes Yes NO NO Yes Yes NO Yes 1 Yes 3 No

NO Yes NO No

risk of pulmonary toxicity and to establish the effect, if any, of long-term amiodarone treatment on PFT, we prospectively studied 35 patients given amiodarone for ventricular or supraventricular arrhythmias between November 1978 and March 1983. There were 23 men and 12 women, ranging in age from 34 to 80 years. Of these, 24 had coronary artery disease, 10 had idiopathic cardiomyopathy, and one had rheumatic heart disease. The mean duration of amiodarone therapy was 19.3 months (range 4.1 to 45.5 months). Medication was administered as previously reported.2g-32 In most recent cases, loading doses of amiodarone were given intravenously and pharmacokinetics were studied. Pulmonary function tests were done before amiodarone was started in 10 patients and after the severe arrhythmias were controlled in 25 patients. Initial doses of amiodarone in all patients ranged from 400 to 1600 mg/day. Intravenously administered loading doses were used in 12 patients; the

910

Rakita et al.

American

100

October, 1963 tieart Journal

r

W Z

2

N Number

5

20

k

40 i

of patients

r= 0.529 1 SD

I

I

I

I

40

I

I

80

Cumulative

I

I

I

120

,

180

Dose Amiodarone

$56

(gm)

1. FEVJFVC in all patients (0) and in 10 patients (0) in whom pulmonary function studies were done prior to the initial dose of amiodarone. Fig.

N Number

of patients

r= 0.346

I

I

20 Mean

I

I

60 Amiodarone

I

I

100 Cumulative

I

I

140

’

/YJ

356

Dose (gm)

Fig. 2. Total lung capacity in all patients (0) and in 10 (0) patients in whom pulmonary function studies were done prior to the initial doseof amiodarone.

maximum dose was 80 mg/kg

given over 48 hours. This dose, determined from pharmacokinetic studies for individual patients, was designed to maintain serum amiodarone concentrations between 2.0 and 2.5 cLg/ml.31 The 10 patients in whom PFT was done prior to the administration of amiodarone received a mean maximum cumulative dose of 120.8 + 111.8 gm and a mean daily maintenance dose of 380.0 +- 220.1 mg. The 25 patients not tested prior to the first dose were given cumulative doses ranging from 2.8 to 93.8 gm (average 34.7 + 29.0 gm). Tests of vital capacity (FVC), forced expiratory

volume in 1 second (FEV,), total lung capacity (TLC), and the diffusing capacity for carbon monoxide (DLCO) were performed by laboratory personnel. Calculations and interpretations of the data were made by pulmonologists who were unaware of the patient’s clinical status or treatment with amiodarone. The number of studies per patient varied from two to 13, depending, in most instances, on the duration of treatment. A total of 170 studies are included for analysis. Results of PFT. The results of serial PFT for all 35 patients are presented in Figs. 1 to 3. There was no significant correlation between the cumulative dose

Volume Number

106 4, Part 2

Amiodarone

.t 5 s

pulmonary

toxicity

911

E N Number

4

of

Patients

r= 0.30

n

I

C

SD

:

6

I

Mean Amiodarone Cumulative Dose (gm) Fig. 3. Diffusion capacity for carbon monoxide in all patients (0) and in 10 patients

(0)

in whom

pulmonary function studies were done prior to the initial doseof amiodarone.

of amiodarone and FEVJFVC, TLC, or DLCO. Analysis of linear regression yielded F values of 0.34, 0.52, and 0.30, respectively. Data from the 10 patients in whom PFT was done prior to the initiation of amiodarone therapy revealed mean values of FEVJFVC 65.8 +- 9.9%, TLC 5.01 ? 1.44 L, and DLCO 12.0 + 3.6 ml/ mm Hg/min. The values and responses were not significantly different from those in the 25 patients in whom the first PFT was done after the initiation of treatment (Figs. 1 to 3). Of importance is the finding that there were marked variations in individual test results in any one patient over time. The changes appeared to be related more to the clinical state of these seriously ill patients at the time of the test than to cumulative amiodarone dose. The PFT results for this group as a whole were not significantly altered by the total dose of amiodarone. These data are similar to those of Kudenchuk et al., who nonetheless suggested that progressive alterations in DLCO or TLC, along with changes in chest x-ray findings, may serve as important indications for discontinuing the drug or reducing the dosage. These investigators also advised that patients with preexisting lung abnormalities be considered for alternative therapy. Our analysis of the data was different in some respects, but still failed to provide support for this view. Serial changes in pulmonary function tests provide no predictive evidence for the development of pulmonary toxicity. Furthermore, there were many possibilities to account for the changes seen in

these seriously ill patients. Even if all other factors could be excluded, the low incidence of pulmonary complications generally, and in our patients specifically, suggests that changes in pulmonary function tests do not identify patients at risk of pulmonary toxicity. RECHALLENGE

WITH AMIODARONE

In the United States, amiodarone is given only for recurring arrhythmias that are unresponsive to other agents. Therefore, pulmonary toxicity presents a difficult therapeutic dilemma. Should the drug be withdrawn in a patient with a life-threatening arrhythmia who has demonstrated a satisfactory antiarrhythmic response? Can amiodarone be safely reinstituted, possibly at lower doses, in a patient who has shown susceptibility to the pulmonary complications? Amiodarone rechallenge after suspected amiodarone-induced pulmonary toxicity has been reported by Zaher et al.,lg who gave a 46-year-old man amiodarone 800 mg/day for 5 months for ventricular tachycardia when this patient developed respiratory tract symptoms and a unilateral upper lobe infiltrate that failed to respond to antituberculosis therapy. Amiodarone was discontinued, and prednisone 60 mg/day given. Pulmonary symptoms subsided and the radiologic abnormalities resolved within 2 months, but ventricular tachycardia recurred after 6 months. Amiodarone was started at 400 mg/day along with prednisone 10 mg every other day. With this regimen the patient remained asymptomatic. After 9 months, however, ventricular arrhythmia

912

Rakita et al.

American

October. 1983 Heart Journal

Fig. 4. Left, Chest radiograph

at maximum pulmonary involvement. Right, The most recent chest radiograph taken after the reinstitution of amiodarone therapy. The pleural changes at the left base are residual from the time of acute involvement and have not changed since restarting treatment.

recurred, and amiodarone was increased to 600 mg daily. Three months later, respiratory tract symptoms and nonspecific radiographic changes were noted. Amiodarone was continued at 600 mg/day, but prednisone was increased to 40 mg/day. As symptoms and radiographic abnormalities resolved, prednisone was tapered to 10 mg every other day, and over the ensuing 4 months the patient remained asymptomatic with amiodarone 600 mg/day. The authors conclude that low-dose corticosteroids may be beneficial in patients who require amiodarone but have demonstrated susceptibility to amiodarone-related pulmonary complications. The prophylactic role of low-dose corticosteroids is far from proved, however. The low-dose prednisone regimen given this patient did not prevent recurrence of symptoms and radiographic abnormalities. Another patient6 had recurrence of pulmonary changes on rechallenge with amiodarone, but corticosteroids were not tried to determine whether the medication could be continued. Therefore, we present our experience in a patient successfully rechallenged with amiodarone without recurrence of pulmonary complications. Case report. After receiving amiodarone, 600 to 1000 mg/day for approximately 8 months for ventricular tachyarrhythmia, a 63-year-old man was hospitalized because of apparent congestive heart failure. Chest x-ray findings included bilateral pulmonary infiltrates and a fluid-filled bullous lesion in the left upper lobe (Fig. 4). Transbronchial biopsy of the lung showed thickened alveolar septae, with scattered mononuclear cell infiltration and focal prominent hyperplastic pneumocytes, small foci of organizing intraalveolar exudate, aggregates of his-

tiocytes, and mild anthrocosis. At that time, the serum concentration of amiodarone was 3.08 pglml; the cumulative dose was 181.5 gm. Amiodarone was stopped. Corticosteroids were not given, but findings on chest x-ray cleared within two months (Fig. 4) and the patient remained free of significant arrhythmias without antiarrhythmic therapy for 2 months. Then frequent multiform premature ventricular contractions and couplets occurred. Amiodarone serum concentration at this time was 1.87 pug/ml. The patient was given procainamide 500 mg every 4 hours. Ventricular tachycardia was controlled for the following 10 months when it recurred. After more than 1 year without amiodarone, the serum concentration was 0.19 pg/ml. Thus the halflife of amiodarone in this patient was found to be 90 days. In view of the failure of procainamide to prevent ventricular tachycardia, amiodarone was restarted in a dose of 800 mg/day orally. After 11 days the dose was reduced to 200 mg daily. Procainamide then was discontinued and amiodarone continued at a dose of 400 mg. Ventricular tachycardia has been suppressed and pulmonary infiltrates have not recurred during 6 months, although there is evidence of a residual pleural reaction at the left base (Fig. 4). The total amiodarone dose received by the patient since restarting therapy is 54.6 gm, and his serum level is currently 1.33 pglml. MECHANISM OF AMIODARONE-INDUCED COMPLICATIONS

PULMONARY

The mechanism of amiodarone-induced puhnonary toxicity has not been established. Evidence of a drug-induced autoimmune mechanism has been

Volume Number

106 4, Part 2

detected in only a few reported cases.4*14B21 Although most patients had at some time been given procainamide, there is no clinical evidence of drug-induced lupus erythematosus, an entity rarely associated with pulmonary fibrosis. The most promising clues to the pathogenesis of amiodarone-induced pulmonary toxicity lie in the association between the molecular structure of the drug and the electron microscopic findings in lung biopsy specimens from two patients of osmophilic lamellar or granular inclusion bodies within distended lysozomes of macrophages, type II pneumocytes, interstitial cells, and endothelial ce11s.15 Marchlinski et al., who made these observations, noted the similarity of these ultrastructural findings to those described in dermal histiocytes and pericytes of a patient with amiodarone-induced cutaneous pigmentation. 33 Comparable inclusion bodies have also been described in Schwann cells, muscle cells, and fibroblasts in patients with amiodaronerelated neuropathy34,35 and in the corneal epithelial cells of patients with cornea1 microdeposits. In all of these reports the concentric lamellated appearance and osmophilic nature of these lysozomal inclusion bodies were similar and consistent with a lipoid material. Marchlinski et a1.15and others26,36 have cited the amphiphilic nature of the amiodarone molecule, which contains both a highly apolar aromatic ring system and a polar side chain with a positively charged nitrogen atom, and have noted its similarity to other cationic amphiphilic compounds that have been demonstrated both clinically and experimentally to cause lipid storage disturbances.37 A model of such an amphiphilic drug-induced lipoidosis affecting the lung has been described with chlorphentermine,38 which causes an increase in phospholipid in lung tissue and produces multilamellar lysozomal inclusion bodies, chiefly in phagocytes, similar to those described here. It is postulated that this drug binds to phospholipids, thereby inhibiting their normal enzymatic degradation and resulting in their accumulation in lysozomes. As one of the major sites of phospholipid synthesis and turnover, the lung would be expected to show high concentrations of retained phospholipid. The concentration would relate to the steady-state tissue drug concentrations and thus would be dose-dependent. In the proposed model, reversibility of this process would depend on the dissociation rate of the drug from the phospholipid-binding site and therefore be related, at least in part, to the half-life of the drug in body tissues. This model conforms to many of the observations

Amiodarone pulmonary

toxicity

9 13

made in the reported cases of amiodarone pulmonary toxicity, including relationship to dose and reversibility at lower doses. The variability in both time course and total dose to development of pulmonary toxicity may relate to the wide variations in elimination half-life among individual patients. The accumulation of phospholipid would be consistent with the histologic findings of foamy macrophages reported by Marchlinski et a1.15and by Suarez et al.*l Inflammation and fibrosis, which dominated the microscopic findings in other reports> 14,l6 may be secondary or late phenomena, related possibly to a nonspecific reaction to the breakdown of phospholipid-laden macrophages, interstitial cells, and pneumocytes. CONCLUSIONS

Pulmonary toxicity may occur in patients receiving high doses of amiodarone, usually within the first year of treatment, with a reported frequency ranging from none to 5.7% .15 Of the 39 reported cases, nine patients have died and 30 have had resolution of respiratory symptoms and radiologic abnormalities after withdrawal of amiodarone. Half of the patients who died and half of those who survived had been given corticosteroids. Surveillance of patients with serial chest x-ray studies at frequent intervals during the first year of amiodarone exposure seems warranted. Dose reduction or temporary discontinuation of amiodarone is indicated if otherwise unexplained radiographic changes occur, especially in conjunction with respiratory symptoms. In view of the current state of knowledge, we agree with Rotmensch et a1.13 that it would be inappropriate to withhold amiodarone therapy from patients with potentially life-threatening arrhythmias for fear of pulmonary toxicity, regardless of preexisting pulmonary status. Although Kudenchuk et a1.6 consider patients with prior pulmonary changes to be at higher risk of pulmonary toxicity than others, that conclusion is not supported by our data. In any event, pulmonary complications are not predictable in a given patient, and the overall incidence continues to be low in published studies. We did not find serious deterioration of pulmonary function in patients given amiodarone, whether pretreatment pulmonary function tests were abnormal or not. Some data suggest that the risk of pulmonary complications increases with higher maintenance doses. Perhaps more attention should be paid to the size of the maintenance doses and more aggressive attempts made to reduce longterm doses to a minimum once control of arrhythmia is achieved.

914

Rakita et al.

REFERENCES

22.

1. McGovern B, Garan H, Ruskin JN: Sinus arrest during treatment with amiodarone. Br Med J 284:160, 1982. 2. McComb JM, Logan KR, Khan MM, Geddes JS, Adgey AAJ: Amiodarone-induced ventricular fibrillation. Eur J Cardiol 11:381, 3. 4.

5.

6.

7.

8.

9.

10.

23.

1980.

Lubbe WF, Mercer CJ: Amiodarone: Its side effects, adverse reactions and dosage schedules. NZ Med J 95~502, 1982. Harris L, McKenna WJ, Rowland E, Holt DW, Storey GC, Krikler DM: Side effects of long-term amiodarone therapy. Circulation 67:45, 1983. Heger JJ, Prystowsky EN, Jackman WM, Naccarelli GV, Warfel KA. Rinkenbereer RL. Zincs DP: Amiodarone: Clinical efficacy’ and electrlphysiology during long-term therapy for recurrent ventricular tachycardia or ventricular fibrillation. N Engi J Med 305:539, 1981. Kudenchuk PJ. Pierson DJ. Greene HL. Graham EL. Sears GK, Gross SW: Werner JA,’ Carmen CK: Effects of amiodarone on pulmonary function. (Submitted for publication.) Charlier R, Deltour G, Tondeur R, Binon F: Studies in the benzofuran series. VII. Preliminary pharmacological study of 2-butyl-3-(3,5-diiodo-4-ON-diethylaminoethoxybenzoyl)-benzofuran. Arch Intern Pharmacodyn 139:255, 1962. Simpson WT: A review of the therapeutic results and unwanted effects of amiodarone. In Amiodarone in cardiac arrhythmias. Royal Society of Medicine, International Congress and Symposium Series No. 16. London, 1979, Grune & Stratton. Ward DE, Camm AJ, Spurell RAJ: Clinical antiarrhythmic effect of amiodarone in patients with resistant paroxysmal tachycardias. Br Heart J 44:91, 1980. Kaski JC, Girotti LA, Messuti H, Rutilzky B, Rosenbaum MB: Long-term management of sustained recurrent symptomatic ventricular tachycardia with amiodarone. Circulation 64:273,

American

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

AM HEART J 101:374,1981.

12. Rotmensch HH, Liron M, Tupilski M, Laniado S: Possible association of pneumonitis with amiodarone therapy (Letter). AM HEARTJ 13.

35.

100:412,1980.

Rotmensch HH, Belhassen B, Ferguson RK: Amiodarone: Benefits and risks in perspective. AM HEART J 104:1117,

36.

1982.

Sobol SM, Rakita L: Pneumonitis and pulmonary fibrosis associated with amiodarone treatment: A possible complication of a new antiarrhythmic drug. Circulation 65:819, 1982. 15. Marchlinski FE, Gansler TS, Waxman HL, Josephson ME: Amiodarone pulmonary toxicity. Ann Intern Med 97:839, 1982. 16. McGovern B, Garan H, Kelly E, Ruskin JN: Adverse reactions observed during treatment with amiodarone hydrochloride. Br Med J (Accepted for publication.) 17. Riley SA, Williams SE, Coke NJ: Alveolitis after treatment with amiodarone. Br Med J 284:161,1$X%2. 18. Waxman HL, Groh WC, Marchlinski FE, Buxton AE, Sadowski LM, Horowitz LN, Josephson ME, Kastor JA: Amiodarone for control of sustained ventricular tachyarrhythmia: Clinical and electrophysiologic effects in 51 patients. Am J Cardiol 50:1066, 1982. 19. Zaher C, Hamer A, Peter T, Mandel W: Low-dose steroid therapy for prophylaxis of amiodarone-induced pulmonary infiltrates. N Engl J Med 306:779, 1983. 20. Wright AJ, Brackenridge RG: Pulmonary infiltration and bone marrow depression complicating treatment with amiodarone. Br Med J 284:1303, 1982. 21. Suerez LD, Poderoso JJ, Elsner B, hnster AM, Esteva H, Belotti M: Subacute pneumonopathy during amiodarone therapy. Chest 83:566, 1983.

Rakita L, Sobol SM, Mostow ND, Vrobel T: Amiodarone pulmonary toxicity. AM HEART J 106:906, 1983. Rasmusson K, Winkler RJ, Griffin J, Peters F, Mason J: Antiarrhythmic efficacy of amiodarone in recurrent ventricular tachycardia evaluated by multiple electrophysiological and ambulatory ECG recordings. Acta Med Stand 212:367, 1982.

24.

1981.

11. Podrid PJ, Lown B: Amiodarone therapy in symptomatic, sustained refractory atria1 and ventricular tachyarrhythmias.

October, 1983 Heart Journal

14.

37.

Morera J, Vidal R, Morel1 F, Ruiz J, Bernado LL, Laporte JR: Pulmonarv fibrosis and amiodarone (letter). Br Med J 285:895, i982. Gillot P: Huit ans d’experience clinique du traitement de I’angine de poitrine par l’amiodarone. Brux Med 50:657, 1970. Kannan R, Nademanee K, Hendrickson JA, Rostami JS, Singh BN: Amiodarone kinetics after oral doses. Clin Pharmacol Ther 31:438, 1982. Rohlfing BM, Stauffer JL: Drug-induced pulmonary disease. In Preger L, editor: Induced disease: Drugs, irradiation, occupation. New York, 1980, Grune & Stratton, p 45. Schlaeffer F, Gold B, Hirsch M, Keynan A: A case of interstitial pneumonitis during amiodarone therapy. Isr J Med Sci 18:8OS,1982. Rakita L, McGee J, Sobol S, Luria M: Clinical trial of amiodarone hydrochloride for the treatment of refractory arrhythmias. Am J Cardiol 43:359, 1979. Rakita L, Sobol SM: Amiodarone treatment for refractory arrhythmias: Dose ranging and importance of high initial doses. Circulation 64(suppl 4):263, 1981. Mostow N, Rakita L, Vrobel T, Cancilla D, Blumer J: Amiodarone. IV loading for rapid suppression of complex ventricular arrhythmias. Clin Res 31:207A, 1983. Rakita L, Sobol SM: Amiodarone in the treatment of refractory ventricular arrhythmias: Importance and safety of high initial dose therapy. JAMA (Accepted for publication). Delage C, Lagace R, Huard J: Pseudocyanotic pigmentation of the skin induced by amiodarone: A light and electron microscooic studv. Can Med Assoc J 112:1205. 1975. Dudognon P, Hauw JJ, deBaecque C, Derrida JP, Escourolle R, Nick J: Neuropathie au chlorhydrate d’amiodarone. Rev Neurol Paris 135:527, 1979. Lemaire JF, Autret A, Biziere K, Romet-Lemone JL, Gray F: Amiodarone neuropathy: Further arguments for human drug-induced neurohpidosis. Eur Neurol 21:65, 1982. D’Amico DJ, Kenyon KR, Ruskin JN: Amiodarone keratopathy: Drug-induced lipid storage disease. Arch Ophthalmol 99~257, 1981. Liillmann H, Lullmann-Rauch R, Wasserman 0: Druginduced phospholipidosis. II Tissue distribution of the amnhinhilic drug chlorohentermine CRC Crit Rev Toxic01 4:1'85,'1975.

38.

-

_

Lullmann-Rauch R, Reil GH: Chlorphentermine-induced lipoidosis-like ultrastructural alterations in lungs and adrenal glands of several species. Toxic01 Appl Pharmacol 30:408, 1974.

DISCUSSION

Dr. Michat.- Is there any correlation between the development of pulmonary changes and loading doses, which are not commonly used in Europe? Dr. Rakita: We have not found a correlation. We have usedhigh loading dosesand have seenthis complication in only one patient. High dosesover time may be a more critical factor in the majority of patients, which is the reasonwe are seeinga higher incidence of this problem in the United States than in Europe. Some patients do develop toxicity within a month of initiation of therapy, however, so there may be important

Volume Number

106 4, Part 2

factors other than dose, such as the underlying condition of the lungs, the turnover rate of drug within the lungs, or interactions between amiodarone and other drugs. Dr. Singh: Dr. Mandel, how do corticosteroids work, and does the response to them indicate whether the underlying mechanismis immunologic or nonspecific? Dr. Ma&l: I have a follow-up report on the patient Dr. Rakita described in a letter to the New England Journal of Medicine. That patient decided to stop taking corticosteroids and to continue with amiodarone. He developed pulmonary infiltrates again. Another patient with severe pulmonary toxicity responded to corticosteroid therapy and has done well with low maintenance doses. We have had 12 patients who developed pulmonary reactions. One patient, with marked symptomsand abnormalities of pulmonary function tests and blood gases,had a dramatic responsewithin 48 hours of initiation of a high doseof prednisone, (60 mg/day). On the basisof pharmacokinetic data, I suspectthat stopping amiodarone alone would not be adequate except, perhaps,in mild cases.Our practice is to treat significant symptoms with corticosteroids. If amiodaroneis essential,we continue it along with low dosesof steroids. Dr. Rakita: The pulmonary changes disappear and the patient’s status improves even when amiodarone is still measurablein the serum. This makes it difficult to postulate an immunologic reaction. In many patients, the pulmonary changes disappear before the arrhythmia recurs. Our patient had a measurableamiodarone level, 0.19 pg/ml, at the end of 54 weeks when ventricular tachycardia finally recurred. Perhaps drug turnover in the lung is different from that in other tissues. In addition, radiologists and pathologists in the United States are extremely sensitive to finding amiodarone pulmonary reactions. Therefore, it is important that radiologic and pathologic interpretations be made by blinded observers. Dr. Wellens: Pulmonary reactions can occur with low maintenance doses.I know of four casesin the Netherlands, all of which occurred in patients receiving 200 mgfday. Dr. Singh: The function of time is important in assessing these changes,especially in view of the variable half-life of amiodarone, from 10 to 110days. Even a dose of 200 mg may ultimately become“high” becauseof drug cumulation for a patient in whom drug elimination halflife is 100 days. Dr. Heger: Most patients improve very rapidly with a short courseof steroids. We have one patient who, after a year, still requires maintenance oxygen and low-dose corticosteroid therapy; presumably he has permanent pulmonary changes.I would like to know the frequency of permanent changesobserved by other investigators. Dr. Podrid: We have six patients, from a seriesof 240, with clear-cut pulmonary changes.Most of our patients were receiving low doses.We rechallenged two patients who developed pulmonary changeswith 200 mg/day. The

Amiodarone

pulmonary

toxicity

915

first patient had developed pulmonary infiltrates after 2 years of treatment. The infiltrates cleared without the use of corticosteroids in lessthan 2 weeks.Three months later this patient was rechallenged with amiodarone, 200 mg twice a week, and again developed infiltrates. Amiodarone wasstopped, and the infiltrates again cleared, without the use of steroids, in 2 weeks.The other patient developed pulmonary infiltrates after 1% months of amiodarone therapy. The infiltrates cleared in about 3 days, and she has been rechallenged successfully. This patient had a history of idiopathic interstitial pulmonary fibrosis, which had been treated with corticosteroids. Chest x-ray findings were similar during both episodes of pulmonary disease,but the time course with amiodarone was more rapid. What was the time courseto resolution noted by other investigators? How many patients were receiving antiarrhythmic agents concomitantly, drugs that may also causepulmonary changes? Dr. Rakita: The incidence varies, ranging from none to 10%. We do not understand why. There is alsovariability in response to rechallenge and in the pathologic findings. Dr. Reiffel: Among our first 50 patients, six had pulmonary reactions. Clear-cut fibrosis was found in biopsy specimensin one of the two patients who died. Pulmonary reactions resolved within 2 weeks in the remaining four patients, after a reduction in amiodarone dosageand treatment with corticosteroids. Dr. Sobol: Fibrosis and inflammatory changeswerethe predominant pathologic findings in our patients. Marchlinski noted foamy histiocytes and a foamy appearanceof other pneumocytes, however. Although the underlying mechanismis unknown, Marchlinski proposed that the amphiphilic cationic nature of amiodarone induces an abnormality of phospholipid deposition. The lung would be a likely target becauseit has a high level of phospholipid turnover. Inhibition of enzymatic degradation of phospholipid in lung tissue may cause deposition of phospholipid, followed by inflammation and fibrosis. Therefore, patients in whom amiodarone is withdrawn early may have less fibrosis and fewer inflammatory changes.As a result, corticosteroids may not be effective. In contrast, patients with secondary inflammation may respond. Dr. Waxman: We have seen seven or eight patients with pulmonary reactions in a seriesof 140patients. In the first 52 patients, we sawsix cases.Now we are using lower dosesof amiodarone,and have seenonly one or two cases in 100 patients. This change suggeststhat lower doses causelesspulmonary toxicity. In the last three patients, diagnosis was made early by chest x-ray examination. X-ray findings cleared within 3 to 4 months of stopping amiodarone. We have rechallenged two patients with lower dosesof amiodarone,and have not seena recurrence during 7 to 8 months of follow-up. Theoretically, corticosteroids should be helpful only when there is secondary inflammation.

Rakita et al.

American

I have been told of patients with pulmonary toxicity that responded to corticosteroids and discontinuing amiodarone. The chest x-ray findings cleared 3 to 4 months after amiodarone was stopped, but as steroids were tapered the infiltrates recurred. Again the infiltrates resolved as the dose of steroids was increased. If this was a cumulative phenomenon, was there any relationship to the serum concentration of amiodarone? Dr. Rakitu: I do not know. It would be interesting to see amiodarone blood levels in patients with different stages of pulmonary toxicity, from early to late and severe. Dr. Waxman: How often do you recommend performing x-ray studies? Dr. Rakita: Every 3 months. Dr. Singh: When blood levels have been measured, they are not very high, 1.2 to 1.5 lg/ml in several cases.

Side effects and possible amiodarone use

October, 1993 Heart Journal

The crucial issue, according to Holt’s data, may be the very high concentrations of both the drug and its metabolite in the lungs. Therefore, blood levels may not be particularly informative. Dr. Scheinman: We have seen pulmonary toxicity in eight of 270 patients. It is very important to obtain transbronchial biopsy specimens to confirm the diagnosis. One patient had all the “typical” clinical features of amiodarone pulmonary toxicity. On postmortem, however, we found that this patient had had pneumonia and congestive heart failure. Another patient given amiodarone and procainamide showed a resolution of infiltrates when procainamide alone was stopped. Dr. Zipes: Unfortunately, too much information is anecdotal. A central registry for patients who develop pulmonary complications should be established to gather and organize data.

contrtindications

of

With the increasing use of amiodarone, several unwanted effects have been recognized. We revtewed 140 patients treated with amiodarone over a !i-year period in an attempt to identify patients at risk, to assess the incidence of these effects and their possible relation to dose, and to determine their outcome. The most common effect was photosensitivity (57% of patients responding to a questionnaire), whereas asymptomatic cornea1 microdeposits were found in all patients undergoing ophthalmologic examination. In contrast, aymptomatfc eye changes (colored halos) and slate-gray skin pigmentation were rare. Of the metabolic alterations, the rise in hepatic enzymes correlated with dose and plasma drug and metabotfte concentrations (r = 0.59, p < 0.001; f= 0.62, p < 0.001, respectively) but was not associated with clinical disease. This relation to dose was not evident in patients developing clinical thyroid abnormalfties (two hypothyroidism, two hyperthyroidlam), all of whom had normal thyroid function prior to therapy. Four of the five hypothyroid patients were over 70 years of age. No patients developed peripheral neuropathy, but tremor and sleeplessness were common complaints (30% and 26% of patients, respectively) that responded to a decrease in dose. One patient with an abnormal chest x-ray film prior to therapy developed pulmonary fibrosis. We suggest the restricted use of high doses of amiodarone for protracted periods. Patients at particular risk are the older age group (hypothyroidism) and those with abnormal lung function prior to therapy who may be predisposed to pulmonary alveolitia. Most of the observed unwanted effects resolve when amiodarone is decreased in dose or discontinued. (AM HEART J 106~916, 1983.)

Louise Harris, M.B., William J. McKenna, M.D., Edward Rowland, and Dennis M. Krikler, M.D. London, England

From Royal

the Division Postgraduate

of Cardiovascular Medical School.

Based on data previously et al: Amiodarone: Side 1983. Reprint Hospital,

916

Disease,

published in Harris effects of long-term

Department

of Medicine,

L, McKenna WJ, Roland E, therapy. Circulation 67:45,

requests: Dr. L. Harris, Division of Cardiology, Toronto 101 College St., Toronto, Ontario, Canada MSFG 1L7.

General

M.B.,

Amiodarone is a drug with unique pharmacodynamic and pharmacokinetic properties. Its increasing use as an antiarrhythmic agent attests to its potent efficacy but has also been accompanied by the increased recognition of unwanted effects. As with many other drugs, it is possible that these effects can