Many investigators use in vitro models of global ischemia to examine the effects of preconditioning, often with recovery of contractile function as the end-point. Such models are relevant to myocardial protection during cardiac surgery. However, there is still debate as to whether preconditioning preserves ventricular contraction secondary to limitation of infarction or by a direct effect on stunning. Since infarct size is the original end-point against which protection by preconditioning is measured, our aims were, first, to validate global ischemic preconditioning by measuring infarct size after subsequent regional ischemia and, second, to correlate limitation of infarct size with mechanical function. After stabilization, seven isolated buffer perfused rabbit hearts were subjected to 5 minutes of global "no-flow" ischemia followed by 10 minutes of reperfusion ("global preconditioning"). Seven control hearts were allowed to stabilize for an additional 15 minutes at constant flow. Subsequently, regional ischemia was induced in both groups for 45 minutes followed by 2 hours of reperfusion. Left ventricular and coronary perfusion pressures were measured throughout. Myocardial infarct size was measured using triphenyltetrazolium staining and expressed as a percentage of the area at risk outlined with fluorescent microspheres. The ratio of infarct to risk zone was reduced from 47.6 +/- 7.3% in control hearts to 16.4 +/- 5.4% (p = 0.005) in preconditioned hearts, confirming the efficacy of global preconditioning. In addition, preconditioning led to a better preservation of systolic function, which correlated significantly with limitation of infarct size (r = 0.75, p = 0.002). Global preconditioning may account for the successful use of cross-clamp fibrillation during cardiac surgery.