OUP user menu

The risk of antiarrhythmic drugs in atrial fibrillation: 20 years of controversies

Jean-Yves Le Heuzey
DOI: http://dx.doi.org/10.1093/europace/eup175 840-841 First published online: 22 June 2009

For about 20 years, the safety of antiarrhythmic drugs used in the maintenance of sinus rhythm in atrial fibrillation patients remains questionable. Immediately after the publication of the CAST trial in 1989, Coplen, in a meta-analysis published in 1990, showed that quinidine treatment was more effective than no antiarrhythmic therapy in suppressing recurrences of atrial fibrillation but appeared to be associated with increased total mortality.1 In the past 20 years, this major safety problem has been evoked for all other antiarrhythmic drugs, even if it was already clear at that time that the safety profile of quinidine itself could be unfavourable, when compared with other antiarrhythmic drugs, mainly due to the risk of QT prolongation and torsade de pointes induced by the drug.

In a large, unselected population-based cohort of patients discharged with first time atrial fibrillation and subsequently treated with flecainide, propafenone, sotalol or amiodarone, Andersen et al. found no increased risk of death.2 The results are based on the data obtained in 141 500 patients included in this nationwide registry conducted in Denmark. In this unselected population of patients discharged after a first hospitalization for atrial fibrillation, the main findings were that antiarrhythmic drug therapy was not associated with increased risk of death. Furthermore, deaths occurring in the treatment groups were fewer than observed in the population not receiving antiarrhythmic drugs. The authors concluded that this indicates adequate patient selection by the treating physicians from a safety perspective. This major concern shows that cardiologists, after 20 years, are very careful about the use of these drugs in these patients.

So, the main questions that can be asked are the following: is the bad safety profile of these drugs a myth or a reality? Do we have to consider that these good results are due to safe drugs or to safe prescriptions or both? After the publication of AFFIRM trial results3 showing no difference, in terms of survival, between the strategies of rhythm and rate control, several publications dealt with the problem of proarrhythmic events, survival, and mortality.46 The overall risk of adverse arrhythmic events upon exposure to antiarrhythmic drugs in the AFFIRM study was reasonably low.4 The authors concluded that strict criteria for the safe use of antiarrhythmic drugs were successful in minimizing proarrhythmic events. In another paper, which was an analysis of cause-specific mortality in the AFFIRM study, Steinberg5 concluded that management of atrial fibrillation with the rhythm-control strategy conferred no advantage over rate control strategy in cardiac or vascular mortality and may be associated with an increased non-cardiovascular death rate. The results obtained by Andersen and the Danish group, in real life, clearly demonstrate that it is not the case. Annualized mortality rates were 2.54, 4.25, 4.29 and 7.42 per year per 100 person-years for flecainide, propafenone, sotalol, and amiodarone, respectively. Multivariate Cox proportional hazard models did not show increased risk of death associated with any of the antiarrhythmic drugs. Sinus rhythm is either an important determinant of survival or a marker for other factors associated with survival that were not recorded, determined, or included in the survival model of AFFIRM. In the paper dealing with relationship between sinus rhythm, treatment, and survival in the AFFIRM study,6 the authors concluded that currently available antiarrhythmic drugs were not associated with improved survival, which could suggest that any beneficial antiarrhythmic effects of antiarrythmic drugs are offset by their adverse effects.

From these data it is possible that if an effective method for maintaining sinus rhythm with fewer adverse effects would be available, it might be beneficial. Of course, the hope of all arrhythmologists is that ablation procedures could be this beneficial method with few adverse effects. It is necessary, to date, to have data concerning mortality of these patients by comparing antiarrhythmic drug therapy and ablative procedures as first line therapies. The safety of antiarrhythmic drugs in this indication is now very well established, mainly because of the 20 year experience of the cardiologists, but also by the pharmacovigilance processes. In most of the developed countries, these processes are correctly functioning. Device vigilance is also quickly improving but ‘procedure vigilance’ is, to date, not so developed! It will be mandatory in the future to have more details about the real safety of ablative procedures, taking into account not only the data coming from experienced centres but also from all other centres performing ablation of atrial fibrillation. A real comparison of safety between two therapeutic strategies cannot omit a part of the data. Atrial fibrillation is a serious disease which can alter quality of life, increase mortality, and which is a major burden for healthcare system.7 Nevertheless, the mortality rate of atrial fibrillation remains low, around 3% per year. Very risky procedures have not their place in the treatment of a disease whose spontaneous risk is moderate, even if this risk is real.

The data obtained by the Danish group are really reassuring but it can be concluded from their work and from the experience of 20 year use of antiarrhythmic drugs that cardiologists must remain very cautious. In fact they are and it is probably the reason why these results are reassuring. It is a major concern to know that class I antiarrhythmic drugs must be avoided in patients with congestive heart failure, complete left bundle branch block, and/or coronary artery disease. It is clear that nowadays these contra-indications are perfectly known by doctors, but that was not the case 20 years ago. The negative inotropic effect of these drugs has been clearly demonstrated. In the CAST study, the inclusion criteria concerned post-myocardial infarction patients but it is reasonable to extent the contra-indication of class I antiarrhythmic drugs to all coronary artery disease patients. Finally, the risk of transformation of atrial fibrillation into 1/1 flutter is now prevented by the coprescription of a drug slowing the atrioventricular conduction, mainly a betablocker. The guarantee of a good safety of a class I antiarrhythmic treatment is based more on a good choice for the indication and a strict respect of contra-indications than on the surveillance itself. Nevertheless, it is necessary to check the electrocardiogram to be sure that the drug does not induce a major prolongation of QRS duration. For sotalol the main safety concern is QT prolongation and the risk of torsade de pointes which can be increased, for example, after diarrhoea. Patients must be warned about this risk. In patients with heart failure plus atrial fibrillation, the only drug which can be used, to date, is amiodarone as emphasized in the guidelines on atrial fibrillation management.8,9 The problem of amiodarone safety is the more complex one, as it is mainly related to extracardiac deleterious effects. The higher annualized mortality rate observed by Andersen was that of amiodarone, i.e. 7.42 per year per 100 persons-years. This high rate is also probably due to the fact that the patients for whom amiodarone was chosen had more severe underlying heart diseases.

The opinion we have on the safety of antiarrhythmic drugs will probably change in the future with the use of dronedarone. The results of the ATHENA trial10 are very interesting, as for the first time it has been possible to demonstrate a favourable effect of an antiarrythmic drug in terms of cardiovascular mortality. The results of ATHENA clearly show that dronedarone reduces the incidence of hospitalizations due to cardiovascular events or deaths in patients with atrial fibrillation. It is possible to hope that these favourable effects of dronedarone would announce the end of the controversies! Nevertheless, the problem of severe heart failure remains a major one as dronedarone will have to be avoided in this type of patient.

Conflict of interest: J-Y.L.H. has received grants from Meda and Sanofi-aventis.


  • The opinions expressed in this article are not necessarily those of the Editors of Europace or of the European Society of Cardiology.