TAKE-HOME MESSAGE
- In this population-based observational study involving 6039 individuals with incident atrial fibrillation,
amiodarone exposure was not significantly associated with an increased risk of interstitial lung disease or primary lung disease after a median follow-up period of 4.2 years. After inverse probability treatment weighting, amiodarone exposure was associated with a 35% reduced risk of all-cause mortality compared with nonexposure.
- These findings highlight the safety of intermediate- to long-term amiodarone use, without evidence of increased mortality risks.
- Written by
Since its inception in the 1960s initially as an antianginal drug but later popularized as a potent antiarrhythmic agent by Argentinian cardiologist Dr. Mauricio Rosenbaum, amiodarone has become the most widely employed antiarrhythmic agent worldwide. The compassionate use of amiodarone in the USA and Europe employed it as salvage therapy at a wide range of dosing (from high dose [800–2000 mg/day] to low [200 mg/day] dose loading, with variable dosing maintenance regimens) for life-threatening ventricular arrhythmias in the pre-defibrillator era. Recognition of multiorgan system toxicities associated with medium- and long-term amiodarone use emerged in the 1980s. Of these, the most prominent was pulmonary toxicity diagnosed as interstitial lung disease. This appeared to be dose related and was occasionally fatal. Clinical experience with amiodarone therapy for atrial tachyarrhythmias, such as atrial fibrillation, reinforced these observations. Knowledge of the multiorgan toxicity of the drug solidified. In an AF population with less likelihood of advanced acquired cardiovascular disease, cause-specific mortality analysis in the AFFIRM trial suggested an increased risk of noncardiovascular mortality with rhythm-control therapy. This was due to pulmonary and oncologic causes, most commonly lung cancer. Amiodarone was the dominant agent used for rhythm control.1 Subsequent individual rhythm-control drug analyses from the AFFIRM trial using propensity-score matched cohorts for 64 demographic and clinical variables showed a nonsignificant increase in the risk of all-cause mortality with amiodarone.2
In the current report from Tsaban and colleagues almost two decades after the AFFIRM trial, the authors revisit this subject.3 The rationales for this observational analysis were the standardization of the use of low-dose amiodarone for long-term use in their AF population and the availability of a large AF population with detailed drug exposure records followed for long periods necessary for the study of long-term toxicities. Using real-world experience from a national insurance database in Israel, the authors compared the rates of interstitial lung disease and all-cause mortality in patients with AF without prior disease who were exposed to amiodarone from initiation of therapy with those of patients who did not. The latter presumably received rate control therapies. The large cohorts in each arm (over 45,000 patients) were matched for three demographic variables and AF duration.
In summary, the authors did not elicit an increased risk of interstitial lung disease and noted a reduced risk of mortality in the amiodarone-treated cohort in the intention-to-treat analysis. The secondary outcome of primary lung cancer was not worsened by amiodarone exposure. The on-treatment analysis indicated a small increased risk for interstitial lung disease with amiodarone. Post hoc comparison of clinical and other covariates suggested comparability of the two cohorts. The authors suggest that the reduced risk of all-cause mortality in the amiodarone-treated cohort negates the likelihood of any impact of amiodarone on all-cause and, inferentially, noncardiovascular mortality. In a retrospective observational cohort comparison, the absence of strong matching techniques to ensure comparability of these cohorts is important to the robustness of the results. As an example, an important variable is the use of anticoagulation therapy in both AF cohorts; this was significantly lower in the rate cohort and is particularly necessary when rhythm control is not required. In fact, in prior AFFIRM trial reports, the use of anticoagulation reduced the rates of total mortality. As such, mortality rates in the comparator arm of this analysis may be skewed to higher values, masking differences. A major limitation of real-world analyses in comparison to prospective randomized trials is highlighted by these data. At the present time, safety surveillance for pulmonary toxicity associated with the long-term use of amiodarone should be considered good medical practice.
References