This study was designed to investigate the anti-arrhythmic effect of diosgenin preconditioning in myocardial reperfusion injury in rat, focusing on the involvement of the nitric oxide (NO) system and mitochondrial ATP-dependent potassium (mitoKATP) channels in this scenario. After isolation of the hearts of male Wister rats, the study was conducted in an isolated buffer-perfused heart model. Global ischemia (for 30 min) was induced by interruption of the aortic supply, which was followed by 90-min reperfusion. Throughout the experiment, the electrocardiograms of hearts were monitored using three golden surface electrodes connected to a data acquisition system. Arrhythmias were assessed based on the Lambeth convention and were categorized as number, duration and incidence of ventricular tachycardia (VT), ventricular fibrillation (VF), and premature ventricular complexes (PVC), and arrhythmic score. Additionally, lactate dehydrogenase (LDH) levels in coronary effluent were estimated colorimetrically. Diosgenin pre-administration for 20 min before ischemia reduced the LDH release into the coronary effluent, as compared with control hearts (P < 0.05). In addition, the diosgenin-receiving group showed a lower number of PVC, VT and VF, a reduced duration and incidence of VT and VF, and less severe arrhythmia at reperfusion phase, in comparison with controls. Blocking the mitoKATP channels using 5-hydroxydecanoate as well as inhibiting the NO system through prior administration of L-NAME significantly reduced the positive effects of diosgenin. Our finding showed that pre-administration of diosgenin could provide cardioprotection through anti-arrhythmic effects against ischemia-reperfusion (I/R) injury in isolated rat hearts. In addition, mitoKATP channels and NO system may be the key players in diosgenin-induced cardioprotective mechanisms.