Increased Mortality After Dronedarone Therapy for Severe Heart Failure - ANDROMEDA


Patients with heart failure have higher mortality and morbidity with atrial and ventricular arrhythmias. Dronedarone, a newer antiarrhythmic drug, is similar to amiodarone, and has been shown to be efficacious in the treatment of atrial fibrillation (AF), without an observed proarrhythmic effect. Accordingly, the goal of this trial was to evaluate the efficacy of dronedarone in reducing hospitalization due to congestive heart failure (CHF) in patients with symptomatic heart failure.


Dronedarone is superior to placebo in reducing hospitalization due to CHF in patients with symptomatic heart failure.

Study Design

Patients Screened: 2,402
Patients Enrolled: 627
NYHA Class: II (40.2%), III (56.8%), IV (3%)
Mean Follow Up: 2 months (median)
Mean Patient Age: 71.5 years (mean)
Female: 25

Patient Populations:

• Age ≥18 years
• Hospitalization for new or worsening heart failure
• At least one episode of New York Heart Association functional class III or IV or paroxysmal nocturnal dyspnea within the month before admission
• Wall-motion index ≤1.2 (ejection fraction ~35%)


• Acute myocardial infarction within the past 7 days
• Heart rate <50>
• PR interval >280 msec
• Sinoatrial block or II/III degree heart block not treated with pacemaker
• History of torsades de pointes
• QTc >500 msec
• Serum potassium <3.5>
• Use of class I or III antiarrhythmic drugs
• Drugs known to cause torsades de pointes, or known inhibitors of the cytochrome P450 CYP3A4 system
• Other serious disease
• Acute myocarditis
• Constrictive pericarditis
• Planned or recent (30 days) cardiac surgery or angioplasty
• Clinically significant obstructive cardiac disease
• Acute pulmonary edema within 12 hours
• Pregnancy or lactation
• Expected poor compliance
• Previous treatment with dronedarone
• Participation in another clinical trial

Primary Endpoints:

Death from any cause or hospitalization for worsening heart failure

Secondary Endpoints:

• Death from all causes
• Hospitalization for cardiovascular causes
• Hospitalization for worsening heart failure
• Occurrence of AF or atrial flutter
• Death from arrhythmia
• Sudden death

Drug/Procedures Used:

All patients received either 400 mg twice daily of dronedarone or matching placebo.

Concomitant Medications:

Angiotensin-converting enzyme inhibitors (86.3%), beta-blockers (61.3%), diuretics (94.1%); also implantable cardioverter defibrillators (1.6%)

Principal Findings:

ANDROMEDA was scheduled to run for 2 years, but it was stopped prematurely after 7 months, after interim analysis suggested a higher mortality in the dronedarone group compared with placebo. By this time, 627 patients had been randomized, 310 to dronedarone and 317 to placebo. A history of AF was present in about 38% of the patients, whereas AF was present in about 25% of the patients at randomization. The mean QTc at baseline was 444 msec.

The overall mortality was more than double in the dronedarone group compared with placebo (8.1% vs. 3.8%, hazard ratio [HR] 2.13 (95% CI 1.07-4.25, p = 0.03), mostly due to an increase in cardiovascular mortality (7.7% vs. 2.8%). Death due to heart failure (3.2% vs. 0.6%), and due to arrhythmias or sudden death was more common with dronedarone (3.2% vs. 1.9%). When patients were followed up for an additional 6 months after study drug cessation, the mortality curves seemed to be similar (HR 1.13, 95% CI 0.73-1.74, p = 0.6). Death seemed to be highest in patients with lower wall motion index (or ejection fraction).

The primary composite endpoint of death from all causes or hospitalization from heart failure was similar between the two arms (17.1% vs. 12.6%, HR 1.38, 95% CI 0.92-2.09, p = 0.12). First hospitalization due to cardiovascular causes was more common in the dronedarone group (22.9% vs. 15.8%, p = 0.02). The incidence of heart failure (10% vs. 8.2%), ventricular tachycardia (1.9% vs. 0.6%), ventricular fibrillation (0.3% vs. 0.9%), and supraventricular tachycardia (5.6% vs. 2%) was similar between the two groups, whereas the incidence of increase in serum creatinine was greater in the dronedarone group (2.6% vs. 0).


The results of ANDROMEDA indicate that the use of dronedarone in patients with heart failure, especially those with systolic dysfunction, is associated with a higher incidence of mortality compared with placebo, primarily from death due to CHF and arrhythmias. There also seems to be a higher incidence of increase in serum creatinine with treatment with dronedarone, although whether this resulted in renal failure or other consequences is unknown.

The findings of ANDROMEDA are in contrast to other randomized trials with dronedarone, including ATHENA and EURIDIS and ADONIS, which have not demonstrated a detrimental effect from the use of dronedarone, although ANDROMEDA is the only study to include patients exclusively with symptomatic heart failure and systolic dysfunction. The recently presented ATHENA trial, in fact, demonstrated a beneficial effect on cardiovascular mortality with dronedarone, including a significant reduction in death due to arrhythmias.

It is also interesting to note that post-hoc analysis from the ScD-HeFT trial had demonstrated worse outcomes in patients with class III CHF who received amiodarone compared with placebo. Further focused studies will thus need to determine the safety and efficacy of class III agents in patients with systolic left ventricular dysfunction. Meanwhile, the use of dronedarone in patients with CHF cannot be recommended at this time based on this study.


Kober L, Torp-Pederson C, McMurray JJ, et al., on behalf of the Dronedarone Study Group. Increased mortality after dronedarone therapy for severe heart failure. N Engl J Med 2008;358:2678-87.

Clinical Topics: Arrhythmias and Clinical EP, Heart Failure and Cardiomyopathies, SCD/Ventricular Arrhythmias, Atrial Fibrillation/Supraventricular Arrhythmias, Acute Heart Failure

Keywords: Tachycardia, Supraventricular, Ventricular Fibrillation, Creatinine, Dyspnea, Renal Insufficiency, Tachycardia, Ventricular, Heart Failure, Hospitalization, Ventricular Dysfunction, Left, Death, Sudden, Cardiac

< Back to Listings