We have shown that the free radical spin trap DMPO (5,5-dimethyl-1-pyrroline-N-oxide) reduces reperfusion-induced arrhythmias in a dose-dependent manner in the isolated perfused rat heart subjected to 10 min regional ischemia and 3 min reperfusion. At its optimal concentration (1,000 mumol/L) DMPO, added to the perfusate 5 min prior to ischemia, reduced (p less than 0.05) the incidence of reperfusion-induced irreversible ventricular fibrillation from 83 (10 of 12) to 33% (4 of 12). When hearts were subjected to ischemia (10 min) and reperfusion, with DMPO (1,000 mumol/L) added to the perfusion fluid only 2 min before reperfusion, comparable protection was observed. To ascertain whether or not DMPO achieved an absolute reduction in vulnerability to arrhythmias irrespective of the duration of ischemia, hearts (12 for each group) were also subjected to 5, 10, 20, 30, or 40 min of ischemia; DMPO (1,000 mumol/L) was added to the perfusate either 5 min before ischemia or 2 min before reperfusion. In each instance a bell-shaped time-response profile was obtained. In the DMPO-free controls this gave a maximal vulnerability to arrhythmias after 10 min of ischemia. In the DMPO-treated hearts this curve was shifted to the right, with a peak vulnerability at 20 min. These results indicate that the primary action of DMPO is to exert a delaying effect which extends the duration of ischemia that can be tolerated before the heart becomes vulnerable to reperfusion-induced arrhythmias. However, this effect is achieved during the reperfusion period and not during the preceding period of ischemia. The precise mechanism by which this free radical spin trapping agent achieves this unusual effect remains to be resolved, but in studies with light-inactivated DMPO, this protective effect was lost, indicating that its ability to be oxidized, possibly by superoxide or hydroxyl radicals, may be critical to its mechanism of action.