Recent studies suggested that distal coronary sinus (CS) pacing may prevent atrial fibrillation (AF) by reducing site dependent intraatrial conduction delay. The aim of this study was to investigate the effect of high right atrial (HRA) and distal CS pacing on local conduction delay in patients with AF. The study population consisted of 10 patients with persistent AF after transvenous atrial defibrillation and 10 control subjects. The local conduction delays along the anterolateral right atrium (RA), in the CS, and at the right atrial septum (RAS), and the incidence of AF in response to an atrial extrastimulus during HRA and distal CS pacing at a drive cycle length of 400, 500, and 600 ms were evaluated. In patients with AF, distal CS and HRA pacing are associated with more prominent and similar extent of conduction delay within the atria, without any significant difference in the dispersion of conduction delay and susceptibility to AF induction (70% vs 60%, P = 0.9). In normal controls, distal CS pacing reduces the conduction delay at the RAS and CS ostium and decreases the dispersion of conduction delay and the propensity for AF induction (0% vs 50%, P = 0.03) compared to HRA pacing. The pacing drive cycle length has no significant effect on conduction delay in patients with AF and normal controls (P > 0.05). Compared to normal controls, patients with AF have significantly longer conduction delay at the RAS and along the anterolateral RA during HRA and distal CS pacing. The result of this study demonstrates that the effect of HRA and distal CS pacing on the local atrial conduction delay in patients with and without AF differ significantly. These patients with AF may have more diffuse atrial anisotropy causing the changes in conduction, and pacing from distal CS in these patients dose not reduce the propensity for AF.