Preconditioning (PC) is a potential approach to myocardial protection. We hypothesize that brief ischemia or adenosine given prior to an extended period of warm ischemia may prevent myocardial stunning by altering myocardial metabolism. Using a global ischemia model, 19 dogs were subjected to no PC(control), two episodes of ischemia (2 min of global ischemia followed by 3 min of reperfusion) (IPC), or 30 min of pulmonary artery adenosine infusion (AP), to a maximum of 350 microg/kg/min, followed by 20 min of global warm ischemia on cardiopulmonary bypass. Left ventricular pressure-volume loops and myocardial oxygen consumption (MVO(2)) were measured at baseline and after 60 min of reperfusion, on right heart bypass. All data were compared between baseline and reperfusion. Load independent left ventricular function, defined as preload recruitable stroke work (PRSW), decreased in control and IPC groups (72+/-7%, 71+/-12%, respectively). AP blunted the decrease in PRSW (45+/-9%, p<.05 compared to control). Myocardial energetic conversion efficiency, defined as the slope of the MVO(2)-Stroke work relationship was not significantly changed for controls (2.17+/-0.47 to 1.84+/-0.68) and IPC (2.99+/-0.45 to 2.16+/-0.65), but was for AP (1.16+/-0.88 to 5.71+/-1.66, p<0.04). IPC did not prevent ventricular stunning or alter myocardial energetics. AP reduced ventricular stunning but resulted in worsened myocardial energy efficiency. The benefits to ventricular function of the adenosine pretreatment protocol used in this study were only possible at a cost of higher metabolic requirements.