This study tested the hypothesis that preconditioning, by reducing catabolite accumulation during ischemia, reduces osmotic swelling and myocardial necrosis during subsequent reperfusion. Farm pigs were randomly allocated to one of three groups of treatment: a control group undergoing a 48-min coronary occlusion (CO) of the middle left anterior descending artery, a preconditioned group (2 cycles of 5-min CO and 5-min reperfusion before the 48-min CO), or an intracoronary perfusion group receiving a substrate-free anoxic buffer perfusion into the area at risk between minutes 5 and 10 of the prolonged CO. Animals were killed after 30 min (n = 23) or 6 h (n = 31) of reperfusion. Compared with the control group, both ischemic preconditioning and washout of ischemic by-products by transient anoxic perfusion reduced myocardial edema after 30 min of reperfusion (P < 0.002) by 35 and 32%, respectively, but only ischemic preconditioning reduced final infarct size (by 55%, P < 0.006). Myocardial lactate content before reperfusion, measured in an additional series of 12 experiments, was reduced by 35% in animals receiving preconditioning or intracoronary perfusion. Thus ischemic preconditioning has a marked protective effect against reperfusion edema, and this effect can be explained by reduced catabolite accumulation during ischemia. However, there is no evidence from this study indicating that reduced catabolite accumulation and limited reperfusion edema explain the important anti-infarct effect of ischemic preconditioning.