NPC Archive Item: Dronedarone▼ increases the risk of heart failure, stroke and death compared to placebo in people with permanent AF

NOTE – This is an archive post from the NPC and has not been updated since first publication. Therefore, some hyperlinks may no longer be working.
MeReC Rapid Review NPC Logo

28 November 2011

The PALLAS study found that, in patients with permanent atrial fibrillation (AF) at high risk of vascular events, dronedarone▼ more than doubled the risk of the composite of stroke, myocardial infarction (MI), systemic embolism, or death from cardiovascular (CV) causes compared to placebo (an extra 4.6 events per 100 patient years). It also almost doubled the risk of the composite of unplanned hospitalization for a CV cause or death (an extra 12.4 events per 100 patient years). These data were included in the recent review of dronedarone’s safety by the European Medicines Agency(EMA) which also highlighted potential adverse effects on the liver, lung and kidney.

Level of evidence:
Level 1 (good quality patient-oriented evidence) according to the SORT criteria.

Action
NICE guidance recommends dronedarone as a second line option for the treatment of non-permanent AF but only in people who meet certain criteria. Dronedarone is contraindicated in patients with permanent AF. Healthcare professionals should follow the NICE guideline taking into account the additional recommendations from the EMA (summarised in MeReC Rapid Review No. 4451) and the revised SPC. To support safer use, the MHRA recommends that patients should have their treatment reviewed at the next routine appointment to ensure that they remain eligible for dronedarone treatment according to the revised prescribing information, including new restrictions on use of dronedarone (see ‘so what? section below). Dronedarone should be prescribed only after other treatment options have been considered, and should be started and monitored only by a specialist.

What is the background to this?
AF which resolves spontaneously within 7 days (usually within 48 hours) is known as paroxysmal AF. Persistent AF does not spontaneously terminate, or lasts longer than 7 days, but is potentially amenable to cardioversion back to sinus rhythm. Established AF that has not terminated, has terminated but recurred, or for which cardioversion has not been attempted is known as permanent AF.

Dronedarone was previously compared with placebo in with paroxysmal or persistent AF or flutter in the ATHENA trial. In that study, over a mean of 21 months, dronedarone reduced the risk of the composite of first hospitalisation for CV events or death from any cause (hazard ratio (HR) 0.76, 95% confidence interval (CI) 0.69 to 0.84, p<0.001). Benefits were seen in subgroups within ATHENA, including patients in whom permanent AF developed during the trial.

What did this study set out to do?
In view of the results from ATHENA, the PALLAS investigators hypothesised that dronedarone might have benefits in AF additional to restoration of sinus rhythm, and might be useful in people with permanent AF. They therefore enrolled patients with a 6-month or longer history of permanent atrial fibrillation and risk factors for major vascular events. The mean CHADS2 score was 2.8. Patients were randomised to receive dronedarone 400 mg twice daily or placebo, in addition to their current treatment (88% were receiving rate-control medication, one third were receiving digoxin, and 84% were receiving a vitamin K antagonist). The first coprimary outcome was composite of stroke, MI, systemic embolism, or death from CV causes. The second coprimary outcome was unplanned hospitalization for a CV cause or death.

The intention was to enrol 10,800 patients over a 2-year period with 1 year of additional follow-up. However, after 3,236 patients had undergone randomisation with a median follow-up of 3.5 months, the study was stopped for safety reasons: there was a clear excess of the coprimary endpoints in the dronedarone group.

What does this study claim?
There was a highly statistically significant doubling in the rate of the first coprimary outcome in the dronedarone group (8.2 versus 3.6 events per 100 patient years, HR 2.29, 95%CI 1.34 to 3.94, p=0.002). This increase was driven by increased rates of stroke (HR 2.32, 95%CI 1.11 to 4.88, p=0.02) and death from CV causes (HR 2.11, 95%CI 1.00 to 4.49, p=0.046). There was also a substantial increase in risk of heart failure (rate of heart failure episode or hospitalisation 23.2 versus 10.7 per 100 patient years, HR 2.16, 95%CI 1.57 to 2.98, p<0.001).Together, these increased rates explain the near doubling in the rate of the second coprimary outcome (25.3 versus 12.9 events per 100 patient years, HR 1.95, 95%CI 1.45 to 2.62, p<0.001).

So what?
As reported in MeReC Rapid Review No. 4451 and highlighted in the October edition of Drug Safety Update, the EMA’s Committee for Medicinal Products for Human Use (CHMP) has reviewed the evidence relating to CV, hepatic and pulmonary safety of dronedarone. This review included data from the PALLAS study. The review concluded that the benefits of treatment continue to outweigh the risks for the maintenance of sinus rhythm after successful cardioversion in a limited population of patients with paroxysmal or persistent AF. However, in light of safety concerns dronedarone should be prescribed only after other treatment options have been considered, and should be started and monitored only by a specialist. A number of further contraindications for use and recommendations for monitoring were also endorsed by the CHMP: these are discussed in MeReC Rapid Review No. 4451 and are summarised as follows:

  • Treatment with dronedarone should be restricted to patients with paroxysmal or persistent AF when sinus rhythm has been obtained. It is no longer indicated for use in patients in whom AF is still present.
  • Dronedarone must not be used in patients with permanent AF, heart failure or left ventricular systolic dysfunction.
  • Prescribers should consider discontinuing treatment if AF reoccurs.
  • Switching treatment from amiodarone to dronedarone should be done cautiously by a specialist.
  • Dronedarone must not be used in patients who have had previous liver or lung injury following treatment with amiodarone.
  • Patients on dronedarone should have their lung and liver function as well as their heart rhythm regularly monitored. Liver function should be monitored especially closely during the first few weeks. Kidney function should also be monitored during the first week of treatment

To support safer use, the MHRA recommends that patients should have their treatment reviewed at the next routine appointment to ensure that they remain eligible for dronedarone treatment according to the revised prescribing information, including new restrictions on use.

Study details
Connolly SJ, et al. Dronedarone in high-risk permanent atrial fibrillation. New Engl J Med 10.1056/NEJMoa1109867, published early online 14 November 2011

Design Double-blind randomised controlled trial conducted at 489 sites in 37 countries. Allocation was concealed.

Patients 3236 patients with permanent AF or flutter of at least 6 months’ standing (69% had more than a 2-year history), who had no evidence of intervening sinus rhythm and for whom there was no plan to restore sinus rhythm. Eligible patients were at least 65 years of age (mean age 75 years) with at least one of the following risk factors:

  • coronary artery disease
  • previous stroke or transient ischaemic attack (TIA)
  • symptomatic heart failure (NYHA class II or III symptoms and admission to hospital for heart failure in the previous year but not in the most recent month)
  • a left ventricular ejection fraction of 40% or less
  • peripheral arterial disease
  • the combination of an age of 75 years or older, hypertension, and diabetes.

Intervention and comparison Dronedarone 400 mg twice daily or placebo. Median follow-up 3.5 months.

Outcomes and results The rate of the first coprimary outcome (composite of stroke, MI, systemic embolism, or death from CV causes) in the dronedarone group was more than double that in the placebo group (8.2 versus 3.6 events per 100 patient years, HR 2.29, 95%CI 1.34 to 3.94, p=0.002). This increase was driven by increased rates of stroke (HR 2.32, 95%CI 1.11 to 4.88, p=0.02) and death from CV causes (HR 2.11, 95%CI 1.00 to 4.49, p=0.046). The HR for MI or unstable angina was not significantly increased (HR 1.89, 95%CI 0.80 to 4.45, p=0.14), nor was the HR for MI (HR 1.54, 95%CI 0.26 to 9.21, p=0.63), although this may in part be dues to the small numbers of events observed (a total of 23 and 5 respectively) in the duration of the study. The HR for death from any cause was 1.94 (95%CI 0.99 to 3.79, p=0.049) and for death from arrhythmia it was 3.26 (95%CI 1.06 to 10.0,p=0.03).

There was also a substantial increase in risk of heart failure (rate of heart failure episode or hospitalisation 23.2 versus 10.7 per 100 patient years, HR 2.16, 95%CI 1.57 to 2.98, p<0.001).The rate of the second coprimary outcome (unplanned hospitalization for a CV cause or death) was nearly doubled in the dronedarone group compared to the placebo group (25.3 versus 12.9 events per 100 patient years, HR 1.95, 95%CI 1.45 to 2.62, p<0.001).

Study sponsorship Sanofi Aventis

Further information on AF can be found on NHS Evidence and in the atrial fibrillation e-learning section of NPC.

Please comment on this rapid review using our feedback form.

Make sure you are signed up to NPC Email updates — the free email alerting system that keeps you up to date with the NPC news and outputs relevant to you