Controlled coronary sinus occlusion was shown to retard necrosis of ischemic myocardium. To elucidate this mechanism, regional myocardial blood flow measurement was performed with and without coronary sinus pressure elevation to 30 mmHg (CS30). Colored microspheres were injected into left and right coronary arteries after coronary perfusion of the left anterior descending (LAD) coronary artery was stopped in seven isolated canine hearts with induced atrioventricular block, either paced at 120 beats/min by direct right ventricular stimulation [beating heart (B)] or during asystole induced by stopping pacing [nonbeating heart (NB)]. Regional myocardial blood flow in the LAD perfused area in the control state in the NB with normal coronary sinus pressure (NB-CScont; 0.27 +/- 0.13 ml . min(-1). g(-1); means +/- SE) was significantly greater than those in B-CScont (0.19 +/- 0.09 ml . min(-1). g(-1); P < 0.05) and in NB with CS30 (NB-CS30; 0.19 +/- 0.09 ml . min(-1). g(-1); P < 0.05). Regional myocardial blood flow of the LAD area in B with CS30 (B-CS30; 0.23 +/- 0.10 ml . min(-1). g(-1)) was significantly greater in comparison with that at B-CScont and NB-CS30 (P < 0.05). The augmentative effect of the LAD area regional myocardial blood flow was observed only in the periphery of the ischemic region but not in its center. Cardiac contraction and CS30 impede regional myocardial blood flow in the ischemic bed independently. The coexistence of these two factors enhances regional myocardial blood flow. In conclusion, coronary sinus pressure elevation in B may participate in augmenting collateral flow.