BACKGROUND Atrial fibrillation and atrial flutter (AF) frequently complicate coronary artery bypass surgery (CABG) and increase hospital stay as well as morbidity. Studies of drug prophylaxis to prevent AF with beta-adrenergic blocking agents administered in fixed doses have had conflicting results. METHODS One hundred patients were randomized to receive metoprolol or placebo following CABG. A dosing algorithm was used to achieve clinically significant beta-adrenergic blockade. RESULTS There was no significant difference between the incidence of AF in the metoprolol (24%) and placebo (26%) groups. However, the incidence of AF in all patients having CABG at this institution declined over the period of the study from 31% to 23% (P < .025), in association with the adoption of a continuous technique of cardioplegia delivery. CONCLUSIONS Metoprolol is not efficacious for the prevention of post-CABG AF even when dosage is titrated to achieve clinical evidence of beta blockade. It is likely that the adoption of a continuous cardioplegia technique caused a reduction in our incidence of post-CABG AF.