OBJECTIVE The aim was to determine whether ischaemic preconditioning occurs through the actions of ATP sensitive potassium (KATP) channels during ischaemia and whether pharmacological antagonism of these channels can reverse the protective effect of preconditioning. METHODS 31 New Zealand White rabbits were instrumented for coronary occlusion and reperfusion. Control animals were subjected to ischaemia (30 min) and reperfusion (120 min) only. Study animals were divided into three experimental groups: (1) those receiving intravenous glibenclamide (0.3 mg.kg-1) prior to the 30 min ischaemia; (2) those receiving 5 min preconditioning ischaemia and 10 min reperfusion prior to the 30 min ischaemia; or (3) those receiving preconditioning in the presence of glibenclamide prior to the 30 min ischaemia. The resulting infarct and myocardium at risk were visualised with Indian ink and tetrazolium staining and quantified using computer assisted planimetry. RESULTS Rabbits which were preconditioned showed a 63% reduction in infarct size [14.8(SEM 2.2)% of risk region] in comparison to non-preconditioned controls [39.8(2.1)%]. When rabbits were preconditioned in the presence of glibenclamide the protection was reversed [31.3(5.1)%] using a dose of glibenclamide which by itself did not alter necrosis [37.7(5.4)%]. CONCLUSIONS Infarct size reduction in the rabbit via ischaemic preconditioning is reversible with the coadministration of glibenclamide.