OUP user menu

Hospitalization: a relevant endpoint in atrial fibrillation management?

Jean Yves Le Heuzey
DOI: http://dx.doi.org/10.1093/europace/eur178 1061-1062 First published online: 28 June 2011

This editorial refers to ‘Impact of dronedarone on hospitalization burden in patients with atrial fibrillation: results from the ATHENA study’ by C. Torp-Pedersen et al., on page 1118.

The goals of treatment in patients with atrial fibrillation are numerous: to maintain physiological rhythm, to preserve haemodynamic input of atrial systole, to suppress symptoms, or to make them more acceptable, to improve quality of life, to decrease professional absenteeism, to improve exercise performance, to decrease the burden of atrial fibrillation, to increase the time to first recurrence, to avoid thromboembolic accidents, to avoid progressive cognitive cerebral deterioration, to avoid the evolution towards chronicity, to prevent the occurrence of heart failure, to increase life expectancy, or, if possible… to cure the patient!

Among all these goals, improvement of quality of life is a key issue.1 One of the best ways to increase quality of life is to decrease the number of hospitalizations. This endpoint, i.e. hospitalization, is considered as very relevant in some categories of patients, for example in patients with heart failure. In atrial fibrillation, the first trial which attempted to demonstrate a decrease in hospitalization rates was the ATHENA (A placebo-controlled, double-blind trial to assess the efficacy of dronedarone for the prevention of cardiovascular Hospitalization or dEath from any cause in patieNts with Atrial fibrillation and flutter) trial.2 In this issue of the Journal, Torp-Pedersen et al.3 address the impact of dronedarone on hospitalization burden in patients with atrial fibrillation. ATHENA was a randomized, double-blind, placebo-controlled trial conducted in 551 centres in 37 countries. Subjects were followed up for a minimum of 1 year. The aim of the post-hoc analysis published by Torp-Pedersen et al. was to evaluate the number of first hospitalizations per treatment group, the number of hospitalizations after the first recurrence of atrial fibrillation or flutter, the total number of hospitalizations, duration of hospital stay, and hospitalization burden overtime.

The paper clearly shows that dronedarone reduces the risk of cardiovascular hospitalization and the total hospitalization burden. As previously mentioned, even if hospitalization is a classical endpoint in trials testing the efficacy of drugs used in heart failure, it is not so usual in atrial fibrillation. Some authors have discussed the relevance of hospitalization in atrial fibrillation patients. Indeed many reservations exist: different practices according to the country, different levels of severity of hospitalization (e.g. differences between admissions to intensive care units and regular wards). Another concern is the possibility of multiple hospitalizations that raises the question as to whether it is necessary to censor the data after the first hospitalization. Furthermore, while time to first hospitalization is an important parameter, duration of hospitalization should also be taken into account. Finally, the reason for hospitalization is also a key issue and hospitalizations for cardiovascular and non-cardiovascular reasons as well as for a specific cardiovascular reason should be distinguished. It is also necessary to differentiate between hospitalization due to atrial fibrillation itself and due to concomitant disease.

In fact it is very important, when designing a trial in which it is planned to analyse hospitalizations, to pre-specify the main reasons for hospitalizations. It is mandatory to accurately define these reasons, as it has been made in ATHENA4: myocardial infarction, unstable angina, stable angina, or atypical chest pain; percutaneous coronary, cerebrovascular or peripheral interventions, cardiovascular surgery, atrial fibrillation and other supraventricular rhythm disorders, ventricular arrhythmias, and non-fatal cardiac arrest, worsening congestive heart failure, including pulmonary oedema or dyspnoea of cardiac origin, syncope, implantation of a pacemaker, implantable cardioverter-defibrillator, or any other cardiac device, TIA or stroke, pulmonary embolism or deep vein thrombosis, blood pressure-related disturbances, major bleeding (requiring two or more units of blood or any intracranial haemorrhage), and cardiovascular system-related infection. This list has been pre-defined in the ATHENA trial, indicating that the results are valid but it is mandatory to pre-specify the type of hospitalization.

The duration of hospitalization is also a major determinant in cost of care in these patients because hospitalizations account for almost half the total cost of care.5 In ATHENA, it can be demonstrated that the reduction in AF-related hospitalizations was not simply due to fewer cardioversions, but also to a decrease in the severity of recurrent atrial fibrillation episodes.

If we conclude that reasons for cardiovascular hospitalization in ATHENA were pre-specified, it would be also interesting to clearly characterize and pre-specify the reasons for non-cardiovascular hospitalizations. For example, it is well known that during the development of dronedarone the problem of lever toxicity has not been clearly observed. In the DIONYSOS (Dronedarone vs. Amiodarone for the Maintenance of Sinus Rhythm in Patients with Atrial Fibrillation) trial,6 the rate of elevation of alanine aminotransferase levels in the dronedarone group was higher than in the previous trials, but similar to that observed in the case of amiodarone. It is known that in the ATHENA trial, cardiovascular mortality but not total mortality, was significantly decreased. Thus, it would have been interesting to analyse the reasons for non-cardiovascular hospitalizations in ATHENA.

Reservations have been made about the relevance of hospitalizations in the trials analysing the effects of antiarrhythmic drugs in atrial fibrillation. Indeed, it remains difficult to accurately characterize the reason for hospitalization in a multicentre trial, considering that there are different clinical practices in different countries. Nevertheless, it is plausible to assume that if the reasons for hospitalization are clearly pre-specified, hospitalization could be considered a relevant endpoint.

Conflict of interest: J.Y.L. received research grants from Sanofi-Aventis and was the principal investigator of DIONYSOS.