A mathematical model has been developed ro invesligate the hemodynamic effects of different modes of coronary sinus interventions. To compare the predictions of this model with experimentally derived results, six open chested dogs with coronary artery occlusions were studied. Aortic pressure, coronary venous pressure, and LAD pressure distal to the occlusion were measured. Throughout each experiment, retroperfusion of arterial blood into the coronary sinus, synchronized to diastole (SRP). was performed at flow rates of 25, 50, 75, and 100 mllmin. In each dog, two modes of superimposed coronary sinus occlusion were investigated: mode I , the coronary sinus was occluded only during diastole; mode 2, coronary sinus occlusion was extended until coronary sinus pressure reached a plateau. The validity of the.model was confirmed by a satisfactory agreement between the experimentally and mathematically derived Request reprints from Wolfgang Schreiner, values for the rise of coronary sinus pressure during coronary sinus occlusion for different SRP-flow rates. Therefore, the model could be used to obtain a theoretical estimate for retrograde flow from the venous compartment into the capillaries, a quantity therapeutically important although inaccessible to measurement. In mode 2, retrograde intramyocardial flow was large, even at low retroperfusion flow, and could not be increased funher by using high retroperfusion flow rates. Conversely, in mode I , retrograde intramyocardial flow was generally lower than in mode 2 but could be markedly enhanced by increasing retroperfusion flow rates.