Protection against postinfarction myocardial dysfunction is modest with classic preconditioning (PC). We investigated whether multiple cycles of PC could improve this protection and whether postinfarction dysfunction only depends on the amount of viable tissue. Eighteen rabbits were chronically instrumented with coronary occluders and ultrasonic crystals (segment shortening, SH) in the ischemic zone. A control group underwent 30-min coronary artery occlusion (CAO) with 72-h reperfusion (CAR). In two other groups, PC was induced by six 4-min CAO/4-min CAR cycles (PCx6) or one 5-min CAO/10-min CAR cycle (PCx1). After 72-h CAR, depression in SH was reduced in PCx1 (-68 +/- 7% from baseline) and to a greater extent in PCx6 (-18 +/- 10%) vs. control (-99 +/- 7%; all P < 0.05). Infarct sizes were reduced in PCx1 (15 +/- 2%) and to a greater extent in PCx6 (3 +/- 1%) vs. control (46 +/- 5%; P < 0.05). Contractility of salvaged myocardium was evaluated by calculating the ratio between SH at 72-h CAR and the amount of viable tissue. This index was enhanced in PCx1 (0.39 +/- 0.07, P < 0.05) and to a greater extent in PCx6 (0.82 +/- 0.09) vs. control (0.0 +/- 0.10). This differential effect of PC was not related to changes in apoptosis, endothelial nitric oxide synthase (NOS) expression, or macrophages infiltration but, rather, to blunted inducible NOS expression in PCx6 vs. control and PCx1. Thus multiple cycles of PC induced an almost complete protection against postinfarction dysfunction, potentially involving beneficial effects on salvaged myocardium.