Dronedarone Flatlines .... Again

Written By

Tiffany Shin, PharmD
Maria Pruchnicki, PharmD, BCPS, CLS

Reviewed By

Dawn Havrda, Pharm.D., BCPS
David Parra, Pharm.D., BCPS (AQ Cardiology)


Connolly SJ, Camm AJ, Halperin JL, et al. Dronedarone in high-risk permanent atrial fibrillation. N Engl J Med. 2011; 365 : 2268-76.

Dronedarone (Multaq®), a guideline recommended drug therapy option for paroxysmal and persistent atrial fibrillation (AF), has failed to deliver on the initial hopes and hype as the safer alternative to amiodarone.  Recent studies have shown that dronedarone is associated with severe liver injury, pulmonary fibrosis, pneumonitis, as well as an increased risk of stroke and death.1-6 With the addition of a boxed warning prohibiting its use in patients with permanent AF, dronedarone may now need some “life support" of its own.7

Dronanderone was initially approved to reduce the risk of cardiovascular hospitalizations in patients with paroxysmal or persistent AF based on the ATHENA trial (A Placebo-Controlled, Double-Blind, Parallel Arm Trial to Assess the Efficacy of Dronedarone 400mg BID for the Prevention of Cardiovascular Hospitalizations or Death from any Cause in Patients with Atrial Fibrillation/Atrial Flutter).8,9  No other antiarrhythmic can claim to decrease hospitalizations in this population.9  Results from a subgroup analysis of those patients with permanent AF who participated in ATHENA demonstrated similarly positive outcomes and provided an impetus to validate this finding.10

Enter PALLAS, a multicenter, randomized, double-blind, placebo-controlled study designed to assess the efficacy of dronedarone to reduce major vascular events and hospitalizations in high risk patients with permanent AF.11  However, hopes of reproducing the ATHENA subgroup analysis were quickly dashed.10  The planned interim analysis revealed a two-fold increase in the number of cardiovascular events in the dronedarone group compared to placebo, and the trial was terminated.5,6,11,12  While unexpected, this finding was not unprecedented.  The 2008 ANDROMEDA study (Antiarrhythmic Trial with Dronedarone in Moderate to Severe CHF Evaluating Morbidity Decrease) in heart failure patients was also stopped early due to increased mortality.13   Both ANDROMEDA and PALLAS prompted boxed warnings to be added to the product labeling.  Dronedarone is now contraindicated in 1) patients with NYHA Class IV heart failure or symptomatic decompensated heart failure requiring recent hospitalization, and 2) patients in permanent AF defined as patients who will not or cannot be converted to normal sinus rhythm (NSR).1,6

PALLAS was conducted at 489 sites in 37 countries and was sponsored by the manufacturer Sanofi-Aventis.  The study had two co-primary outcomes and several secondary endpoints (see Table 1).   The inclusion criteria were:

  • Permanent AF or atrial flutter AND
  • Age ≥ 65 years PLUS one of the following
    • Coronary artery disease (CAD)
    • Symptomatic heart failure (NYHA Class I/II with hospitalization in the previous year but not the most recent month)
    • Left ventricular ejection fraction (LVEF) ≤ 40%   OR
  • Combination of Age ≥ 75 years, hypertension, and diabetes

Patients with paroxysmal/persistent AF, implantable cardioverter-defibrillator, bradycardia, or a QTc interval > 500 msec were excluded.

The study population included 3236 participants randomized to receive dronedarone 400mg orally twice daily or placebo.  Baseline characteristics were similar (mean age 75 years) and nearly 70% had permanent AF ≥ 2 years.  Many participants had CAD, LVEF ≤ 40%, and previous stroke or transient ischemic attack.  The mean CHADS2 score was 2.8 in the dronedarone group and 2.9 in the placebo group.  Approximately 54% of patients in each group had NYHA Class II/III heart failure.  Median follow-up at study termination was 3.5 months.

On July 5, 2011 the data monitoring committee recommended the study be stopped given the clear, consistent, and persistent evidence of net harm (Table 1).  Specifically, there was a 2.3 fold increase in the composite rate of stroke, myocardial infarction, systemic embolism, or death from cardiovascular causes; the most prevalent contributors were stroke and death from cardiovascular causes.  It should be noted that time in therapeutic range (INR = 2.0 - 3.0) for those patients receiving vitamin K antagonists was statistically lower in the dronedarone group compared to placebo (55.6% vs. 58.6%, p=0.02).  Although this small difference in INR control may not be clinically important, it may have contributed to the increased incidence of stroke observed in the dronedarone group.  Similarly, unplanned hospitalizations for cardiovascular cause or death occurred more frequently in the dronedarone-treated patients, likely due to the higher rate of cardiovascular events and heart failure hospitalizations.  Arrhythmic death was more frequent in the dronedarone group as well as significant QT prolongation, suggesting proarrhythmic properties not previously recognized. 

Table 1




Hazard Ratio (95% CI)

P Value


N = 1619

N= 1617

No. of Events (%)




Co-primary outcome: Composite of stroke, myocardial infarction, systemic embolism, or death from cardiovascular causes

43 (2.65)

19 (1.18)




Co-primary outcome:
Unplanned hospitalization for a cardiovascular cause or death

127 (7.84)

67 (4.14)




Death from cardiovascular cause

21 (1.3)

10 (0.62)




Death from any cause

25 (1.54)

13 (0.8)




Heart failure episode or hospitalization for heart failure

115 (7.1)

55 (3.4)




Hospitalization for heart failure

43 (2.66)

24 (1.48)





23 (1.42)

10 (0.62)




Arrhythmic death

13 (0.8)

4 (0.25)




To understand the apparently discordant results of the ATHENA and PALLAS trials, comparing the studies is helpful.14  ATHENA and PALLAS differed in the population studied (i.e. intermittent vs. permanent AF) as well as several patient characteristics.  Participants in ATHENA were younger (72 vs. 75 years old at baseline) and less likely to have CAD (30 % vs. 41%) or NYHA class II/III heart failure (21% vs. 54%) when compared to those enrolled in PALLAS.6,8,11  The different populations alone may explain some of the increased cardiovascular events noted in PALLAS.14  Indeed, ANDROMEDA had already established that dronedarone should not be used in heart failure patients.13

While dronedarone shouldn’t be used in patients with permanent AF, other uses are still being studied.  HARMONY (A Study to Evaluate the Effect of Ranolazine and Dronedarone When Given Alone and in Combination in Patients With Paroxysmal Atrial Fibrillation) is a phase 2 randomized, double-blind, placebo-controlled, pacemaker-patient-based trial that began enrollment in January 2012.15,16  This proof-of-concept study uses a surrogate endpoint of AF burden (defined as percentage of total time spent in atrial tachycardia/AF).16 If the results are positive, further studies with clinical outcomes will be needed to determine the safety and efficacy of this novel combination.

PALLAS has identified yet another patient population who shouldn’t receive dronedarone.  For patients with intermittent AF and a moderately healthy heart, dronedarone might be a reasonable therapeutic option so long as they remain in NSR.12  An ECG should be obtained every 3 months and dronedarone should be stopped immediately if the patient converts to AF and cannot be cardioverted.6  Previously unrecognized adverse effects now associated with dronedarone use - severe liver injury, pulmonary fibrosis, pneumonitis, and QTc prolongation – require vigilant monitoring and should be reported to the FDA’s MEDWATCH program.  Needless to say, for a drug specially designed to have less toxicity, dronedarone’s list of adverse effects is getting rather lengthy.  Amiodarone has little to fear of being replaced any time soon.  Dronedarone’s future seems precarious.  If it flatlines again, there is a chance it will never recover.  Should we write a “do not resuscitate” (DNR) order now?  What are your thoughts?



Dronaderone is a great example of how we should never get too excited about drugs that claim to be better than what is currently available. Our healthcare system, as a whole, is sometimes too quick to jump on the newest drug that gets approved with minimal evidence regarding long-term safety and efficacy. For me, this is the final "nail in the coffin" for dronaderone. In my own clinical experience, dronaderone is often ineffective. Amiodarone gets a bad rap but it is a very efficacious drug with low-risk of proarrhytmia. Yes, long-term use is associated with several toxicities but if prospective monitoring is performed, many of these toxicities can be avoided and/or managed without causing significnat harm to the patient. At best, dronaderone will turn out to be a niche drug and it remains to be seen just how small that niche will be.