Atrial fibrillation (AF) occurs in a high proportion of patients after cardiac surgery and is associated with increased morbidity and longer hospital stay. Beta-blockers and amiodarone have been shown to reduce the incidence, but AF still occurs in up to 25% despite pre-treatment. The mechanisms of AF after cardiac surgery are presumably multifactorial. The transient nature of postoperative AF suggests a reversible trigger in patients with susceptible underlying electrophysiological substrates such as abnormal automaticity and conduction delay due to atrial incisions, ischemia and preexisting disease). These could result in atrial premature beats (APBs) and prolonged atrial activation causing lengthening of the P wave. Prophylactic atrial pacing (single- or multi-site) is reported to be effective in patients at high risk for postoperative AF. The mechanisms are probably a combination of preventing bradycardia-induced arrhythmias, overdrive suppression of APBs, eliminating compensatory pauses after APBs and reduction of dispersion of refractoriness. By reducing non-uniform and asynchronous activation resulting from anatomic or functional block, multi-site pacing could improve local excitability and reduce the window of opportunity for AF initiation. We found that the incidence of AF after coronary bypass surgery (CABG) was significantly reduced in patients who received prophylactic biatrial overdrive pacing (BiA) compared with single site left atrial (LA) or right atrial (RA) pacing or no pacing. (BiA 12.5% versus LA 36.4%; RA 33.3% or control 41.9%; P < 0.05). BiA pacing was associated with the greatest reduction of P wave dispersion compared with single site pacing or control (BiA 42 +/- 8%; LA 13 +/- 6%; RA 10 +/- 9%; P < 0.05). Prophylactic postoperative BiA pacing is thus a reasonable and attractive strategy for reducing the risk for postoperative AF.