Blocking of the K-ATP-channel with glibenclamide has been shown to abolish the infarct-reducing effect of ischemic preconditioning in dog and swine. In the rabbit the results have been divergent purportedly related to anaesthesia. The aim of this study was to investigate the importance of the K-ATP-channel in a rabbit model where anaesthesia was not a confounding factor. Isolated rabbit hearts perfused with a Krebs-Henseleit bicarbonate buffer were subjected to 30 min regional ischemia by ligating a coronary artery, followed by 120 min reperfusion. The preconditioning protocol was 5 min global ischemia and 10 min reperfusion. Glibenclamide (100 mu m) was added to the perfusion solution before the preconditioning ischemia and stopped after 5 min regional ischemia. Infarcts were measured with tetrazolium staining and risk zones with fluorescent microspheres. The main results expressed as percent infarction of the risk zone +/- SEM for the different groups are as follows: control (n = 12) 26.8 +/- 3.2, ischemic preconditioning (IF) (n = 9) 7.3 +/- 1.5, (p < 0.05 vs. control), control +/- glibenclamide (n = 9) 46.9 +/- 7.3 (p < 0.05 vs. control), IP + glibenclamide (n = 10) 38.3 +/- 6.9 (p < 0.05 vs. IF). These results show that glibenclamide treatment aggravates ischemia. Also, under the influence of glibenclamide ischemic preconditioning was no longer effective in reducing infarct size in the isolated perfused rabbit heart.