Reduced venous admixture in hemorrhagic hypovolemia: Maintenance of arterial oxygenation by selective pulmonary vascular collapse

In nine anesthetized and ventilated swine, a microcomputer calculated cardiac output, venous admixture (Qsp/Qt) and physiologic deadspace (VD/VT) every 20 sec, utilizing dual oximetry and a gas exchange analyzer. After lung injury with ethchlorvynol (ECV), animals were bled 40% blood volume over 40 min. Mean cardiac output decreased 7.0 to 2.2 L/min (p < .05) accompanied by a decrease in mean Qsp/Qt from 0.28 to 0.14 (p < .05) and an increase in mean VD/VT from 0.39 to 0.54 (p < .05). Arterial Hgb saturation (Sao2) increased from 88 ± 7% to 90 ± 6%. On regression of all data points for each variable, Qsp/Qt had a positive correlation with cardiac output (r = .90), mean arterial pressure (MAP, r = .87), mean pulmonary artery pressure (MPAP, r = .86), and mixed venous Hgb saturation (Svo2, r = .89, p < .001). VD/VT had an inverse correlation with cardiac output (r = −.90), MAP (r = −.82), Qsp/Qt (r = −.83), MPAP (r = −.77), and Svo2 (r = −.92, p < .001). The decreasing Qsp/Qt and increasing VD/VT, with decreasing pulmonary perfusion pressures, were attributed to selective loss of perfusion to alveoli with low ventilation/perfusion ratios. (Crit Care Med 1990; 18:208)