Changing atrial fibrillation to a regular heart rhythm using specific drugs has no advantage over controlling the heart rate in elderly people

Atrial fibrillation is an irregularity of the heart rhythm that increases the risk of stroke and death, particularly in the elderly. One way of treating it is to change this abnormal rhythm to a normal sinus rhythm using specific drugs or by delivering an electric shock to the heart. Alternatively, the irregular rhythm is 'accepted' but the heart rate - which is often fast - is controlled using different drugs. In the studies identified in this review, drugs to create normal heart rhythm offered no benefit over drugs to control the pulse rate. People treated with drugs to control rhythm were more likely to require hospitalisation and to suffer adverse effects.

Authors' conclusions: 

There is no evidence that pharmacological cardioversion of atrial fibrillation to sinus rhythm is superior to rate control. Rhythm control is associated with more adverse effects and increased hospitalisation. It does not reduce the risk of stroke. The conclusions cannot be generalised to all people with atrial fibrillation. Most of the patients included in these studies were relatively older (>60 years) with significant cardiovascular risk factors.

Read the full abstract...

Atrial fibrillation is the commonest cardiac dysrhythmia. It is associated with significant morbidity and mortality. There are two approaches to the management of atrial fibrillation: controlling the ventricular rate or converting to sinus rhythm in the expectation that this would abolish its adverse effects.


To assess the effects of pharmacological cardioversion of atrial fibrillation in adults on the annual risk of stroke, peripheral embolism, and mortality.

Search strategy: 

We searched the Cochrane Controlled Trials Register (Issue 3, 2002), MEDLINE (2000 to 2002), EMBASE (1998 to 2002), CINAHL (1982 to 2002), Web of Science (1981 to 2002). We hand searched the following journals: Circulation (1997 to 2002), Heart (1997 to 2002), European Heart Journal (1997-2002), Journal of the American College of Cardiology (1997-2002) and selected abstracts published on the web site of the North American Society of Pacing and Electrophysiology (2001, 2002).

Selection criteria: 

Randomised controlled trials or controlled clinical trials of pharmacological cardioversion versus rate control in adults (>18 years) with acute, paroxysmal or sustained atrial fibrillation or atrial flutter, of any duration and of any aetiology.

Data collection and analysis: 

One reviewer applied the inclusion criteria and extracted the data. Trial quality was assessed and the data were entered into RevMan.

Main results: 

We identified two completed studies AFFIRM (n=4060) and PIAF (n=252). We found no difference in mortality between rhythm control and rate control relative risk 1.14 (95% confidence interval 1.00 to 1.31).

Both studies show significantly higher rates of hospitalisation and adverse events in the rhythm control group and no difference in quality of life between the two treatment groups.

In AFFIRM there was a similar incidence of ischaemic stroke, bleeding and systemic embolism in the two groups. Certain malignant dysrhythmias were significantly more likely to occur in the rhythm control group. There were similar scores of cognitive assessment.

In PIAF, cardioverted patients enjoyed an improved exercise tolerance but there was no overall benefit in terms of symptom control or quality of life.

Health topics: