The effects of glucose-insulin-potassium (GIK) infusion and glucose (G) infusion started 30 min after experimental coronary occlusion and the combination of GIK and propranolol (P) started 3 hours after coronary occlusion on the development of myocardial infarction were studied in 37 dogs. Fifteen minutes after the coronary occlusion, epicardial electrocardiograms were recorded at 10-15 sites; 24 hours later transmural specimens were obtained from the same sites for determination of myocardial creatine phosphokinase (CPK) activity and the evaluation of morphologic changes. In the control group (normal saline infusion) the relationship between S-T-segment elevation (mv) 15 min after occlusion and CPK activity (IU/mg of protein) 24 hours later was: log CPK = −0.064 S-T + 1.24; r = 0.81. In the GIK group, the infusion was begun 15 min following epicardial mapping, and sites with the same S-T-segment elevations showed less CPK depression than did the control group: log CPK = −0.022 S-T + 1.25. The G group also showed less CPK depletion than the control group but to a somewhat lesser extent than the GIK group (log CPK = −0.030 S-T + 1.20). The group receiving GIK and P 3 hours after occlusion also showed less CPK depression than did the control group (log CPK = −0.034 S-T + 1.26). Histologic analysis in 24-hour specimens showed that sites which exhibited S-T-segment elevation 15 min after occlusion showed normal histology in 3% of specimens obtained from control dogs, while the other 97% showed early signs of myocardial infarction. However, in the GIK group, 36% of the specimens with S-T-segment elevation prior to the infusion were histologically normal 24 hours later, while in the G group 30% were normal, and in the GIK and P group 17% were normal. Electron microscopy confirmed the morphologic changes observed by light microscopy. Thus, in the presence of experimental coronary occlusion, GIK exerts a protective effect against myocardial ischemia and reduces the extent of myocardial necrosis. G alone acts similarly but to a lesser degree, while a beneficial effect can also be demonstrated when GIK and P are started 3 hours after the onset of coronary occlusion.