There were 47 studies (3,476 patients) of pharmacological conversion, 30 studies (3,470 patients) of maintenance of sinus rhythm, 63 studies (1,880 patients) of heart rate control, and 11 studies (8,690 patients) of antithrombotic therapy.
No RCTs were identified for out- versus in-patient strategies. However, 2 observational studies with 717 patients or drug trials were identified.
For echocardiography, there were 6 studies evaluating acute cardioversion and 2 studies evaluating maintenance of sinus rhythm; the numbers of patients were not reported. 'Many' trials reported on the use of echocardiography to predict embolic events. Results from a pooled analysis of 3 trials with 167 patients were presented. No trials regarding the use of transoesophageal echocardiography to guide timing of acute cardioversion were identified.
All results were presented as the combined OR with the 95% confidence interval (CI).
Pharmacological conversion.
Compared with control, the OR was 2.9 (95% CI: 1.2, 7.0) for quinidine, 7.0 (95% CI: 0.3, 153.0) for disopyramide, 24.7 (95% CI: 9.0, 68.3) for flecainide, 4.6 (95% CI: 2.6, 8.2) for propafenone, 5.7 (95% CI: 1.0, 33.4) for amiodarone, 0.4 (95% CI: 0.0, 3.0) for sotalol, and 29.1 (95% CI: 9.8, 86.1) for ibutilide or dofetilide.
For quinidine, the OR was 0.4 (95% CI: 0.1, 2.0) versus propafenone, 0.2 (95% CI: 0.1, 0.9) versus amiodarone, and 5.8 (95% CI: 2.4, 14.2) versus sotalol.
For flecainide, the OR was 7.4 (95% CI: 1.9, 28.3) versus procainamide, 5.1 (95% CI: 2.3, 11.0) versus propafenone, and 2.5 (95% CI: 0.2, 29.6) versus amiodarone.
For propafenone versus amiodarone, the OR was 13.1 (95% CI: 2.1, 79.6).
Other comparisons of the efficacy of miscellaneous drugs and combination drugs (with and without DCC) on pharmacological conversion were also detailed in the review.
Maintenance of sinus rhythm. Compared with control, the OR was 4.1 (95% CI: 2.5, 6.7) for quinidine, 3.4 (95% CI: 1.6, 7.1) for disopyramide, 3.1 (95% CI: 1.5, 6.2) for flecainide, 3.7 (95% CI: 2.4, 5.7) for propafenone, and 7.1 (95% CI: 3.8, 13.4) for sotalol.
For quinidine, the OR was 0.7 (95% CI: 0.4, 1.2) versus flecainide, 0.3 (95% CI: 0.1, 0.7) versus propafenone, 0.9 (95% CI: 0.5, 1.5) versus sotalol, and 0.9 (95% CI: 0.1, 16.5) versus amiodarone.
For disopyramide, the OR was 1.8 (95% CI: 0.6, 5.1) versus propafenone, and 0.3 (95% CI: 0.1, 1.0) versus amiodarone.
For flecainide versus propafenone, the OR was 0.9 (95% CI: 0.4, 2.2).
For propafenone versus sotalol, the OR was 0.7 (95% CI: 0.4, 1.1).
For long- versus short-acting quinidine, the OR was 3.5 (95% CI: 0.9, 13.0).
For flecainide versus cibenzoline, the OR was 1.4 (95% CI: 0.5, 4.0).
The results from the trials of rate control were too disparate for meta-analyses. Diltiazem and verapamil were more efficacious than placebo or digoxin in reducing the heart rate at rest and during exercise in patients with AF. Beta-blockers were more efficacious than placebo or digoxin in reducing the heart rate during exercise in patients with AF. Exercise tolerance was decreased in patients on beta-blockers in a number of studies. The effect of beta-blockers on resting heart rate was inconsistent. The evidence was inconclusive regarding the efficacy of digoxin, particularly during exercise.
Warfarin versus placebo for various outcomes.
The OR was 0.30 (95% CI: 0.19, 0.48) for stroke, 0.50 (95% CI: 0.19, 1.35) for peripheral embolism, 1.90 (95% CI: 0.89, 4.04) for major bleeding, 2.01 (95% CI: 1.51, 2.69) for minor bleeding, and 0.62 (95% CI: 0.38, 1.02) for mortality.
Aspirin versus placebo for various outcomes.
The OR was 0.65 (95% CI: 0.43, 0.99) for stroke, 1.02 (95% CI: 0.33, 3.17) for peripheral embolism, 0.81 (95% CI: 0.37, 1.77) for major bleeding, 3.01 (95% CI: 0.12, 7.4) for minor bleeding, and 0.79 (95% CI: 0.51, 1.22) for mortality.
Warfarin versus aspirin for various outcomes.
The OR was 0.64 (95% CI: 0.43, 0.96) for stroke, 1.27 (95% CI: 0.31, 5.16) for peripheral embolism, 1.60 (95% CI: 0.77, 3.35) for major bleeding, 1.80 (95% CI: 1.05, 3.11) for minor bleeding, and 0.96 (95% CI: 0.58, 1.56) for mortality.
Warfarin versus low-dose warfarin and aspirin for various outcomes.
The OR was 0.35 (95% CI: 0.21, 0.59) for stroke, 1.00 (95% CI: 0.17, 5.81) for peripheral embolism, 1.14 (95% CI: 0.55, 2.36) for major bleeding, 1.68 (95% CI: 0.98, 2.86) for minor bleeding, and 1.02 (95% CI: 0.68, 1.52) for mortality.
Warfarin versus indobufen for various outcomes.
The OR was 0.56 (95% CI: 0.25, 1.22) for stroke, 0.63 (95% CI: 0.21, 1.95) for peripheral embolism, 5.13 (95% CI: 0.60, 44) for major bleeding, 6.68 (95% CI: 1.96, 22) for minor bleeding, and 0.93 (95% CI: 0.56, 1.52) for mortality.
LMWH vs placebo for various outcomes.
The OR was 0.34 (95% CI: 0.06, 1.83) for stroke, 0.56 (95% CI: 0.05, 6.44) for peripheral embolism, and 3.93 (95% CI: 0.74, 20) for mortality.
The echocardiography results suggested an inverse association between left-atrial diameter and successful cardioversion. No conclusions were reached with regards to comparisons of initiating therapy in an in-patient versus an out-patient setting.