Perspectives on ATHENA - Cardiovascular Outcomes versus Symptoms - Atrial Fibrillation


Atrial fibrillation (AF) is an arrhythmia with severe consequences for patients in terms of quality of life and survival. Achieving sinus rhythm (SR) through the use of electrocardioversion or rate control drugs is the first-choice strategy for most AF patients. However, maintaining patients in SR has a low success rate, and patients often revert to AF. Antiarrhythmic drugs are frequently used to maintain sinus rhythm, but are also associated with proarrhythmia and a higher risk of morbidity and mortality. Dronedarone is an antiarrhythmic drug that has been evaluated in a number of trials and may have benefits over other antiarrhythmic drugs in terms of a lower proarrhythmic potential. The ATHENA trial has assessed the efficacy and safety of dronedarone in treating patients with AF. Dronedarone was shown to reduce the number of hospitalisations or deaths due to cardiovascular events in patients with AF. Its benefit goes beyond purely achieving SR. Moreover, as ATHENA enrolled typical AF patients and examined events that have a direct impact on quality of life, it can be argued that it better represented the real-world setting. This article examines the evidence from ATHENA in light of other AF trials, and discusses the validity of the outcome measures used.

Disclosure: HJGM Crijns has received lecture fees and consultation fees from sanofi-aventis.



Citation:European Cardiology 2009;5(1):24–6

Correspondence: HJGM Crijns, Department of Cardiology, University Hospital Maastricht, Postbus 5800, 6202 Maastricht, The Netherlands. E:

Open access:

The copyright in this work belongs to Radcliffe Medical Media. Only articles clearly marked with the CC BY-NC logo are published with the Creative Commons by Attribution Licence. The CC BY-NC option was not available for Radcliffe journals before 1 January 2019. Articles marked ‘Open Access’ but not marked ‘CC BY-NC’ are made freely accessible at the time of publication but are subject to standard copyright law regarding reproduction and distribution. Permission is required for reuse of this content.

Atrial fibrillation (AF) is the most common arrhythmia. It takes the form of a rapid and irregular heartbeat and pulse rate, and predominantly affects the older population. AF is particularly prevalent in patients with cardiac disease. Patients suffering from AF largely fall into two categories: the elderly with cardiovascular (CV) complications or the young with intrinsic electrical disease. An estimated 2.3 million people in the US and 10–12 million people in Europe are believed to be affected, and these numbers are expected to increase by 2.5- to three-fold over the next 50 years.1,2

Two studies in particular have examined the prevalence and incidence of AF in large populations.3,4 The Framingham Heart Study showed a prevalence of AF at initial screenings of five per 1,000 in those 50–59 years of age, rising to 18 per 1,000 in those 60–69 years of age. This study reported two yearly incidence rates of approximately 1.9 and 0.9 per 1,000 person-years for men and women 50–59 years of age, respectively. The Framingham Heart Study also reported an overall incidence of AF of approximately three per 1,000 person-years in men and two per 1,000 person-years in women 55–64 years of age.3 The Renfrew-Paisley study reported similar prevalence and incidence rates from a cohort of 15,406 men and women 45–64 years of age living in the west of Scotland. The prevalence rates in subjects 50–59 years of age were 7.2 per 1,000 in men and 2.9 per 1,000 in women compared with five per 1,000 in the Framingham study for the same age group. By 60–69 years of age, prevalence rates increased to 10.2 and 7.4 per 1,000 in men and women, respectively.4

AF is associated with an increased long-term risk of stroke, heart failure and all-cause mortality.4,5 AF causes symptoms such as fatigue and dyspnoea, which can have a negative impact on quality of life scores in patients and which may be worse in women than in men.4 In recent years there has been a two- to three-fold increase in hospitalisations for AF in the US, with the oldest age groups experiencing the greatest increase. This trend is expected to continue.6 AF is also associated with a 5% risk of stroke per year – two- to seven-fold higher than in people without AF. Strokes in AF patients are also more severe and more likely to result in permanent disability.7 Overall, AF negatively affects quality of life; significant improvements in quality of life scores are frequently associated with restoration of sinus rhythm (SR) by cardioversion.8

The majority of AF patients are the elderly with CV complications. AF patients are associated with considerable costs as a result of the treatment for AF (assessment, hospitalisations and pharmacological intervention, particularly with anticoagulants) and the associated co-morbidities. For reasons of cost, effective management of AF and quality of life, high-risk elderly patients with CV complications should be the focus of medical attention.2,6


The vast majority of high-risk patients with AF present with co-morbidities, particularly hypertension. It is widely recognised that in order to improve prognosis in AF patients and to minimise stroke risk, the first focus should be on antithrombotic treatment and management of co-morbidities before embarking on other treatment options.9

Traditionally, treatment of AF aims to return and then maintain the heart in a state of SR. The return to SR brings various benefits to the patient, including the relief of symptoms, improved haemodynamic status, reduced embolic risk, elimination of the need for atrioventricular node (AVN)-blocking drugs for rate control and a reduction in the risk of mechanical dysfunction and electrophysiological remodelling.10,11 Unfortunately, it is difficult for patients to maintain SR, with only 30–60% of patients remaining in SR after one year.12 Restoration of SR is often the first-choice therapy for AF despite the low success rate. Occasionally, the use of antiarrhythmic drugs can lead to episodes of proarrhythmia.13,14 For patients with severely symptomatic AF, continued treatment with rhythm control is unavoidable.

A second option for AF patients is rate control. The RAte Control versus Electrical cardioversion (RACE) study evaluated the outcomes of both strategies in patients with persistent AF. Rate control patients (n=256) were given negative chronotropic drugs and oral anticoagulation; rhythm control patients (n=266) were given serial electrocardioversion, antiarrhythmic drugs and oral anticoagulation as needed. The treatment strategy was not found to affect the quality of life scores of patients after a mean follow-up period of 2.3 years. Neither strategy demonstrated superiority over the other. Rate control is now considered an acceptable alternative to rhythm control in patients with recurrent AF.15–17

The Atrial Fibrillation Follow-up Investigation of Rhythm Management (AFFIRM) study enrolled 4,060 patients with AF and showed that rate control is possible in most patients with AF. Beta-blockers were shown to be the most effective drugs for rate control, followed by calcium channel blockers, both with or without digoxin.18 RACE, AFFIRM and other studies, including Pharmacological Intervention in Atrial Fibrillation (PIAF) and Strategies of Treatment of Atrial Fibrillation (STAF), have demonstrated that rate control is an acceptable alternative to rhythm control in patients with recurrent AF. These trials also highlighted the necessity of continuing antithrombotic treatment even when long-term SR was achieved.15,19

Antiarrhythmic drug therapy forms the mainstay of treatment for AF patients with CV complications by preventing recurrent AF after SR has been restored. The use of antiarrhythmic drugs (class 1A, 1C and III drugs) such as amiodarone – the leading antiarrhythmic drug currently in use – although effective in maintaining SR has been linked to an increase in adverse effects and mortality.20,21 The stroke-related outcomes for high-risk patients with AF can be significantly improved with concomitant anticoagulation treatment.9 Antithrombotic treatment for these high-risk patients, in accordance with the 2001 American College of Cardiology, American Heart Association and European Society of Cardiology (ACC/AHA/ESC) guidelines for the management of AF, was associated with significantly less chance of thromboembolism or the combined end-point of CV death, thromboembolism or major bleeding compared with high-risk patients undertreated with anticoagulation according to the guidelines. Interestingly, overtreatment of anticoagulation according to the guidelines was not associated with a higher chance of major bleeding.22

As for any medical treatment, the success of the drug in treating a patient can be measured by a reduction in hospitalisations and mortality. A placebo-controlled, double-blind, parallel-arm Trial to assess the efficacy of dronedarone 400mg twice daily (BID) for the prevention of cardiovascular Hospitalisation or death from any cause in patiENts with Atrial fibrillation/atrial flutter (ATHENA) has taken the approach of looking at a composite end-point that is clinically relevant to the older AF patient category: CV hospitalisations and all-cause mortality. Although episodes of stroke, duration of AF, percentage of patients in SR or time to first recurrence of AF are typical end-points in AF trials, CV hospitalisations and all-cause mortality end-points, which are often used in heart failure and coronary heart disease trials, are more clinically relevant and appropriate end-point measures for clinicians in terms of their practice.

A CV hospitalisation end-point can easily be translated into clinical practice. If the application of a drug reduces CV hospitalisations, more beds would be available for other patients and/or fewer beds would be needed in a ward. Clinicians as well as managers could benefit from a reduction in the strain on medical service provision; likewise, it would be realistically relevant to the patient in terms of less time in hospital and better quality of life.


ATHENA was designed as a large outcomes trial to examine the efficacy and safety of dronedarone in the treatment of AF or atrial flutter (AFL). ATHENA was a prospective, randomised, placebo-controlled, double-blind, multinational, multicentre, parallel-group trial evaluating the effects of 400mg BID dronedarone versus placebo (ratio 1:1), on top of current standard therapy, over a minimum treatment and follow-up duration of 12 months in 4,628 patients with paroxysmal or persistent AF/AFL ( #NCT 00174785).

The patients selected for the trial represent the majority of AF patients: from the higher age group, with or without additional risk factors such as arterial hypertension (with ongoing therapy with at least two antihypertensive drugs of different classes), diabetes, prior stroke or transient ischaemic attack, systemic embolism, left atrium diameter greater than or equal to 50mm by M-mode echocardiography or left ventricular ejection fraction less than 0.40 by 2D echocardiography. The main exclusion criteria were: permanent AF; unstable haemodynamic situation such as recently decompensated heart failure; congestive heart failure New York Heart Association (NYHA) class IV; planned major non-cardiac or cardiac surgery; acute myocarditis; bradycardia <50bpm and/or a PR interval >0.28 seconds; and significant sinus node disease in the past, if not treated with a pacemaker.23

The mean follow-up period for the ATHENA trial was 21±5 months. Discontinuation rates were similar, with 696 (30.2%) of patients discontinuing dronedarone compared with 716 (30.8%) discontinuing placebo. Adverse events were the stated reason for discontinuing for 12.7% of patients in the dronedarone group versus 8.1% in the placebo group (p<0.001). The primary end-point (first hospitalisation due to CV events or death from any cause) occurred in 734 patients (31.9%) in the dronedarone group and in 917 patients (39.4%) in the placebo group. The hazard ratio (HR) for dronedarone was 0.76 (95% confidence interval [CI] 0.69–0.84; p<0.001).

First hospitalisations due to CV events revealed that 675 patients (29.3%) in the dronedarone group were admitted to hospital compared with 859 patients (36.9%) in the placebo group, a highly significant reduction (HR 0.74, 95% CI 0.67–0.82; p<0.001). Closer inspection of the data indicates that this reduction in hospitalisations was mainly driven by the reduction in AF (p<0.001) and by a reduction of acute coronary syndrome (p<0.03).

In total, there were 116 deaths (5.0%) in the dronedarone group and 139 (6.0%) in the placebo group, leading to an HR of 0.84 for dronedarone (95% CI 0.66–1.08; p=0.18). Deaths were categorised into death from cardiac arrhythmia, death from non-arrhythmic cardiac causes, death from non-cardiac vascular causes and death from non-CV events. Interestingly, although the reduction in death from any cause was not significant, a significant decrease in death from CV causes was observed in the dronedarone group, with 63 deaths (2.7%) versus 90 deaths (3.9%) in the placebo group (HR 0.71, 95% CI 0.51–0.98; p=0.03). There was also a significant decrease in deaths from cardiac arrhythmia, with 26 deaths (1.1%) in the dronedarone group compared with 48 deaths (2.1%) in the placebo group (HR 0.55: 95% CI 0.34–0.88; p=0.01).24

ATHENA Outcomes

The reduction in arrhythmic deaths strongly suggests that dronedarone harbours an extremely low proarrhythmic potential. This could be a significant advantage in treating AF in the future and reducing the chance of death from arrhythmia. However, it is clear that dronedarone treatment confers a great benefit not only in the reduction of arrhythmic deaths but also in a reduction of hospitalisations due to any cardiovascular event. Taking these two end-points together (any hospitalisation due to any CV event or death from any cause), dronedarone treatment was associated with a reduction of total events from 1,668 (71.7%, placebo) to 1,253 (54.5%, dronedarone).

A number of adverse events, including bradycardia, QT-interval prolongation, diarrhoea, nausea, rash and an increase in serum creatinine levels, were significantly higher in patients taking dronedarone than in the placebo group. Nevertheless, discontinuation rates were similar, with 696 (30.2%) patients discontinuing dronedarone compared with 716 (30.8%) discontinuing placebo. Adverse events were the stated reason for discontinuing for 12.7% of patients in the dronedarone group versus 8.1% in the placebo group (p<0.001).

Bradycardia is an intrinsic side effect not only of dronedarone but of all drugs that reduce high heart rates such as beta-blockers, non-dihydropyridine calcium antagonists and other antiarrhythmic drugs, such as amiodarone, which is part of the reason patients with bradycardia were excluded from ATHENA. Other class-related side effects are pulmonary symptoms, interstitial lung disease and abnormalities of thyroid function; these were not significantly more common with dronedarone than with placebo. This may suggest that dronedarone has a more benign side-effect profile than amiodarone. Gastrointestinal effects such as nausea and diarrhoea were the most common complaints, but were transient. The study confirmed the findings of the Anti-arrhythmic Trial with Dronedarone in Moderate to Severe CHF Evaluating Morbidity Decrease (ANDROMEDA) that serum creatinine levels rise by about 0.1mg/dl following initiation of dronedarone treatment. The elevation has a rapid onset, reaches a plateau after seven days and is reversible if treatment is discontinued. This change in creatinine levels has been shown to be the result of inhibition of the tubular secretion of creatinine, not of a reduction in the glomerular filtration rate. About 29% of the patients in the ATHENA trial had a history of CHF; unstable CHF patients were excluded. In patients with stable CHF, outcome benefits were similar to those of the entire group. A post hoc analysis recently presented at HRS 2009 showed that patients with stable CHF stage III at baseline experienced a 44% relative risk reduction for CV hospitalisation or death compared with placebo (p=0.0028) and a 43% relative risk reduction for CV hospitalisation (p=0.0065). Dronedarone is safe to use in all except unstable heart failure patients.25

Many clinicians may be forgiven for believing that achieving SR would serve as a prognostic indicator in light of recent trials in this field. However, maintaining SR is not everything. Certainly, if a patient can be kept in SR, it is beneficial for his or her prognosis, as was demonstrated in the AFFIRM trial, but SR is only a marker for survival, not the cause of it. Improved outcomes were achieved through maintenance of SR despite the increased antiarrhythmic-associated adverse event implications. If SR can be maintained in patients without the associated adverse events of antiarrhythmic drugs, as appears to be the case with dronedarone, not only would patients benefit in terms of fewer hospitalisations and longer life, but also clinicians would need to treat fewer proarrhythmic and extra-cardiac complications in AF patients.

Although it is difficult to compare the results from ATHENA with results from other AF trials due to different primary end-points, it is not impossible. For example, many heart failure trials take a composite end-point into account. RACE and AFFIRM also calculated composite secondary end-points that could be compared. However, in order for a more direct comparison, a trial has recently completed (DIONYSOS) that compared the benefits of treatment with amiodarone and dronedarone. The publication of the results of this trial are eagerly anticipated.

Summary and Conclusions

ATHENA demonstrated that use of dronedarone significantly reduced the risk of hospitalisation from cardiovascular events or death in those patients at highest risk, with paroxysmal or persistent AF or AFL. Dronedarone was associated with a significant reduction in both the rate of death from cardiovascular causes and the rate of death from cardiac arrhythmia. Several trials have now demonstrated the efficacy of dronedarone in terms of maintaining SR and the associated benefits. Moreover, there is now evidence that the benefit of dronedarone goes beyond purely achieving SR. ATHENA shows that both prevention of recurrent AF and rate control during the arrhythmia are important in reducing hospitalisations for CV events. The need for concomitant treatment with anticoagulants has also been demonstrated. Evidence for a lower proarrhythmic potential within dronedarone is mounting and may be a major contributing factor in the reduction of hospitalisations. The reduction in hospitalisations in the ATHENA trial did not translate into a reduction in mortality from all causes, but did have a significant effect on arrhythmia-related deaths. The merits of looking at first hospitalisation and all-cause mortality have been argued in terms of relevance for the clinician as well as for the patient. It is important for clinicians and future trials to consider using similar end-points in trials to those used in ATHENA to allow more direct comparisons of results in terms of clinically relevant outcomes such as fewer hospitalisations and improved quality of life.


  1. Miyasaka Y, et al., Circulation, 2006;114:119–25.
    Crossref | PubMed
  2. Go As, et al., JAMA, 2001;285:2370–75.
    Crossref | PubMed
  3. Kannel WB, et al., N Engl J Med, 1982;306:1018–24.
    Crossref | PubMed
  4. Stewart S, et al., Heart, 2001;86:516–21.
    Crossref | PubMed
  5. Stewart S, et al., Am J Med, 2002;113:359–64.
  6. Wattigney WA, et al., Circulation, 2003;108:711–16.
    Crossref | PubMed
  7. Jorgenson HS, et al., Stroke, 1996;27:1765–9.
    Crossref | PubMed
  8. Hansson A, et al., BMC Cardiovasc Disord, 2004;4:13.
    Crossref | PubMed
  9. Nieuwlaat R, et al., Eur Heart J, 2005;26:2422–34.
    Crossref | PubMed
  10. Khan IA, Eur Heart J, 2004;25:1274–6.
    Crossref | PubMed
  11. Fuster V, et al., Eur Heart J, 2006;27:1979–2030.
    Crossref | PubMed
  12. Van Gelder IC, et al., Arch Intern Med, 1996;156:2585–92.
    Crossref | PubMed
  13. Ehrlich JR, Nattel S, Drugs, 2009;69(7):757–74.
    Crossref | PubMed
  14. Camm AJ, Int J Cardiol, 2008;127(3):299–306.
    Crossref | PubMed
  15. Hagens VE, et al., Card Electrophysiol Rev, 2003;7:118–21.
    Crossref | PubMed
  16. Hagens VE, et al., J Am Coll Cardiol, 2004;43(2):241–7.
    Crossref | PubMed
  17. Hagens VE, et al., Am Heart J, 2005;149(6):1106–11.
    Crossref | PubMed
  18. Olshansky B, Curr Cardiol Rep, 2004;6(5):351–3.
    Crossref | PubMed
  19. Crijns HJ, Drugs, 2005;65:1651–67.
    Crossref | PubMed
  20. Lafuente-Lafuente C, et al., Arch Intern Med, 2006;166:719–28.
    Crossref | PubMed
  21. The AFFIRM Investigators, Circulation, 2004;109:1509–13.
    Crossref | PubMed
  22. Nieuwlaat R, et al., Am Heart J, 2007;153:1006–12.
    Crossref | PubMed
  23. Hohnloser SH, et al., J Cardiovasc Electrophysiol, 2008;19:69–73.
    Crossref | PubMed
  24. Hohnloser SH, et al., N Engl J Med, 2009;360:668–78.
    Crossref | PubMed
  25. Kober L, et al., N Engl J Med, 2008;358:2678–87.
    Crossref | PubMed