Diffusional shunting in the canine myocardium.

A test for the existence of a diffusional shunt in the myocardium was performed in closed-chest, chloralose-anesthetized dogs. Coronary blood flow was varied from 0.21 to 1.17 ml/min per g with a roller pump via a stainless steel coronary perfusion cannula. Following intracoronary artery bolus indicator injection, the coronary sinus venous appearance times of hydrogen, plasma protein labeled with indocyanine green dye, and red blood cells labeled with 99mTc were measured. Venous appearance time was calculated, either as the time from injection until the indicator reached 5% of its peak venous concentration or the time from injection until 0.1% of the injected material arrived at the coronary sinus. At low coronary flow rates (0.4 sec dye appearance time) hydrogen appearance preceded by about 1 second the appearance of the intravascular 99mTc-labeled red blood cells and the dyed plasma, indicating that hydrogen traversed an extravascular shunt path from small arteries to veins. As flow was increased, this hydrogen precession was reduced, and at high flows detectable levels of hydrogen arrived in the coronary sinus after the intravascular indicators. The difference in appearance time between hydrogen and the intravascular indicators was related linearly to the dye appearance time (proportional to l/flow), as would be expected with diffusional shunting. These data indicate that there is arterial-venous shunting of hydrogen in the myocardium. Furthermore, diffusional shunting may be important in the delivery of oxygen to, or removal of carbon dioxide from, the heart.