OBJECTIVE To determine whether pacing would have a protective effect similar to preconditioning on reperfusion-induced ventricular arrhythmias and whether this protective effect would be induced by pacing-induced myocardial ischemia. METHODS Isolated rat hearts (n = 36) were perfused by the Langendorff technique and the working heart mode. Global ischemia was induced for 10 mins followed by reperfusion for 15 mins. The control group had no pacing, while in the other groups, the hearts were electrically paced three times at 300, 400 and 600 beats/min, respectively, for 5 mins with 1 min intervals between pacings. RESULTS The incidence of reperfusion-induced ventricular fibrillation in controls was 100%, whereas that in the 300, 400 and 600 beats/min groups was reduced to 67% (not significant), 50% (P < 0.03) and 30% (P < 0.01), respectively. The mean duration of reperfusion-induced ventricular fibrillation was also significantly reduced in the 300, 400 and 600 beats/min groups compared with the control group. Coronary flow during pacing in the 400 and 600 beats/min groups, but not the 300 beats/min group, was reduced significantly compared with controls. In the 300 beats/min group, oxygen tension of the coronary effluent during pacing was decreased significantly, and proton production and lactate release of coronary effluent during pacing were increased significantly compared with the control group. CONCLUSIONS Pacing exerted a protective effect similar to preconditioning against reperfusion-induced ventricular arrhythmias. Pacing-induced mild myocardial ischemia, in which coronary flow was maintained, may be involved in the mechanism of this protective effect.