Pump perfusion of blood in coated systems elicits shear-induced platelet aggregation, which may be prevented by administration of substances that block the binding of von Willebrand factor to glycoprotein Ib receptors on the platelets. The effects of pumping on platelets are masked in uncoated circuits because of the dominant influence of blood-material contact.

Pressure-cross-sectional area (P-A) relations of a (thick-walled) arteriole and (thin-walled) small vein (both maximally dilated), embedded in cardiac muscle in both static systole and diastole at slack length and at 90% of maximal length (Lmax), were calculated. The elastic properties of cardiac muscle and vessel wall per se were taken into account. Muscle fibers and vessels were assumed to run in parallel. The muscle tissue (fibers + collagen) was assumed to be incompressible, homogeneous, nonlinearly elastic, and transversely isotropic. Cross-fiber stress-strain relations were assumed to be proportional to those in fiber direction. It is predicted that cardiac muscle in diastole has little effect on the P-A relation of the arteriole but strongly affects that of the small vein. In systole, the myocardium strongly affects the P-A relations of both vessels. Isometric transition from static diastole to static systole (isometric "contraction") was found to reduce arteriolar and venous area (at constant pressures of 35 and 7 mmHg, respectively) by approximately 50 and 40, respectively. Contraction with a 14% shortening was found to reduce these areas by 48 and 32%, respectively. The differences in the results for the two vessels were found to be determined mainly by their difference in the ratio of outer to inner radius. Furthermore, it was found that the area reductions are much larger for contractions (with or without shortening) than for muscle stretch per se. It is concluded that the change in elastic properties and, more specifically, development of stress in cross-fiber direction of the cardiac muscle during contraction causes the area reductions of coronary vessels.

Pressure-flow relations were calculated for a symmetrical, maximally dilated, crystalloid-perfused coronary vascular network embedded in cardiac muscle in (static) diastole and (static) systole at two muscle lengths: slack length and 90% of maximal muscle length (Lmax). The calculations are based on the "time-varying elastance concept." That is, the calculations include the mechanical properties of the vascular wall and the (varying) mechanical properties of the myocardial tissue (in cross-fiber direction). We found that, at any given perfusion pressure, coronary flow is smaller in systole than in diastole. Relative reduction in vascular cross-sectional area, which forms the basis of flow impediment, was largest for the smallest arterioles. At a constant perfusion pressure of 62.5 mmHg, the transition from (static) diastole to (static) systole at constant muscle length ("isometric contraction") was calculated to reduce flow by 74% (from 18.9 to 5.0 ml x min(-1) x g(-1)) and by 64% (from 12.6 to 4.6 ml x min(-1) x g(-1)) for the muscle fixed at slack length and 90% of Lmax, respectively. At this perfusion pressure, contraction with 14% shortening (from 90% of Lmax in diastole to slack length in systole) was calculated to reduce flow by 61% (from 12.6 to 5.0 ml x min(-1) x g(-1)). Increasing muscle length from slack length to 90% of Lmax decreases coronary flow by 34% in diastole and by 8% in systole. We conclude that modeling cardiac contraction on the basis of the time-varying elastic properties of the myocardial tissue can explain coronary flow impediment and that contractions, with or without shortening, have a larger effect on coronary flow than changes in muscle length.

Extracorporeal circulation may have adverse effects on vascular reactivity. To reduce such effects, we recently coated a tube connecting the carotid and the distal femoral artery of rats with albumin. When we partially occluded this perfusion line, the reduction of flow was followed by a marked increase, which seemed not to be caused by autoregulation but by release of a vasodilator at the site of occlusion. In the present study, we investigated whether this vasodilator could originate from platelets aggregating under the influence of increased shear stress at the site of occlusion. Blood distal to the site of occlusion indeed contained numerous platelet aggregates that were not present before occlusion. Continuous recording with a photometric device showed that aggregation in the tube started before flow increased and ended before flow decreased again. Blockade of serotonin S1- and S2-receptors with methiothepin prevented the flow response. Estimated shear stress (231 +/- 17 dyn/cm2) and shear rate (6,370 +/- 478 s-1) at the site of occlusion were of the magnitude known to elicit platelet aggregation. Others have recently demonstrated that shear-induced platelet aggregation is mediated by binding of von Willebrand factor to platelet glycoprotein Ib, which is inhibited by aurintricarboxylic acid. This drug (35 mg/kg iv) completely abolished both platelet aggregation and flow increase in our experiments. These results suggest that the vasodilation during partial tube occlusion is mediated by serotonin released from platelets that aggregate as a result of high shear stress.

It was calculated how cardiac contraction influences the luminal cross-sectional area of a maximally dilated coronary arteriole (37-micron inner diameter at a pressure of 35 mmHg) that is accompanied by two equal venules (45-micron inner diameter at a pressure of 17 mmHg), forming a so-called "triad." It was found that, during a contraction with 14% cardiac muscle shortening, arteriolar area is virtually unaffected (increase of 4%) at the expense of a large (55%) decrease in venular area. For comparison, the areas of an unaccompanied arteriole and an unaccompanied venule were calculated to be reduced by 45 and 36%, respectively, demonstrating the "protective effect" on accompanied arterioles in a triad. During contraction, the overall resistance of a system consisting of one arteriole in series with two parallel venules of equal length was calculated to increase about twice as much for nonaccompanied vessels (resistance increases by a factor of 2.8) than for vessels in a triad arrangement (resistance increased by a factor of 1.4). The calculations show that the extravascular (intramyocardial) pressure, which determines vascular area, is not an independent variable as in the intramyocardial pump and waterfall models but depends on the vascular "loading" conditions. Thus the small venular pressure together with the large venular compliance causes the extravascular pressure to remain low during contraction, thereby protecting the stiff arteriole at high pressure. We conclude that the triad arrangement of intramyocardial coronary vessels attenuates the increase in coronary resistance during cardiac contraction and thus has an important functional advantage.