ATP-sensitive potassium (K+ATP) channels have been shown to play a role in the maintenance of basal coronary vascular tone in vivo. K+ATP channels are also involved in the coronary vasodilator response to adenosine. The aim of this study was to determine the role of K+ATP channels in local metabolically mediated increases in coronary blood flow during cardiac electrical paired pacing without catecholamine effects. In 10 anesthetized closed-chest dogs, coronary blood flow was measured in the left circumflex coronary artery, and myocardial O2 consumption was calculated using the arteriovenous O2 difference. Cardiac interstitial adenosine concentration was estimated from coronary venous and arterial plasma adenosine measurements using a previously described, multicompartmental, axially distributed, mathematical model. Paired stimulation increased heart rate from 57 to 120 beats/min, myocardial O2 consumption 88%, and coronary blood flow 76%. During K+ATP channel blockade with glibenclamide, baseline coronary blood flow decreased in relation to myocardial O2 consumption and thus coronary sinus O2 tension fell. Paired-pulse pacing with glibenclamide resulted in increases in myocardial O2 consumption and coronary blood flow similar to those during control pacing. Coronary venous and estimated interstitial adenosine concentration did not increase sufficiently to overcome the glibenclamide blockade. In conclusion, K+ATP channels are not required for locally mediated metabolic increases in coronary blood flow that accompany myocardial O2 consumption during pacing tachycardia without catecholamines, and adenosine levels do not increase sufficiently to overcome the glibenclamide blockade.