Factors affecting tissue volume measurements in normal and edematous dog lungs.

To characterize further some of the factors affecting lung tissue soluble-gas rebreathing volume (Vlt), we determined the solubility of acetylene in blood and lung tissue, the influence of the presence of pulmonary edema on tissue solubility, the effects of varying tidal volume (VT), and the tissue volume actually measured in two groups of six anesthetized paralyzed dogs: controls (C) and oleic acid-induced pulmonary edema (OA). Each animal's solubility was used to compute Vlt for comparison with gravimetric lung weight (Ql) and extravascular lung water content (Qwl). Solubility at 37.5 degrees C in blood (0.125 ml X 100 ml-1 X Torr-1) exceeded that in lung tissue (P less than 0.005): C = 0.118 and OA = 0.112 ml X 100 ml-1 X Torr-1 (NS). Vlt, expressed as %Ql, increased with increasing VT (20, 35, and 50 ml/kg) in OA (62.2, 78.9, and 94.7%, respectively, P less than 0.0001) but not in C (92.4, 94.4, and 99.3%, respectively). We conclude that solubility differs in blood and lung tissue but not in normal and edematous lungs, Vlt is not affected by VT in normal dogs but is in those with pulmonary edema, and Vlt measures Ql rather than Qwl.

Gas exchange during conventional and high-frequency pulse ventilation in the surfactant-deficient lung: influence of positive end-expiratory pressure.

High-frequency pulse ventilation (HFPV) was compared to conventional ventilation (CV) in a model of severe respiratory failure induced by serial lung lavages with warm saline in 8 mongrel dogs. Before the lavage, during HFPV at 4 Hz with a pulse volume (PV) of 125 ml, mean PaO2 was 107 torr and mean PaCO2 was 34 torr. After the last lavage, during CV at an inspired oxygen fraction FIO2 of 1.0 and a tidal volume (VT) of 535 ml, the PaO2 averaged 60 torr and PaCO2 was 45 torr. At an FIO2 of 0.21, 20 cm H2O of positive end-expiratory pressure (PEEP) was applied to prevent hypoxemia. The resulting PaO2 was 87 torr; PaCO2 was 40 torr. Peak airway pressure (Ppa) rose from 21 to 51 cm H2O. When ventilation was switched to HFPV on room air, a PV similar to the control levels was associated with severe hypoxemia (PaO2 less than 45 torr, PaCO2 greater than 50 torr). As PV was increased PaO2 improved, reaching 113 torr at a PV of about 470 ml. The corresponding mean airway pressure (Paw) was about 20 cm H2O. Thus, application of PEEP during HFPV at low PV did not improve PaO2 even when measured Paw approximated 20 cm H2O. This suggests that HFPV with high PV is more effective than either CV with PEEP, or HFPV with low PV and PEEP.

Uneven driving pressures and redistribution of intrapulmonary blood flow.

Pulmonary arterial pressure (Ppa) and airway pressure (Paw) change equally when Paw varies widely in normal lungs under zone 2 conditions. To evaluate the mechanism underlying the decrease in Ppa at a given Paw in edematous lungs, we measured the quantity of right-to-left shunting of blood (Qs/QT) with sulfur hexafluoride (SF6) using in situ perfused dog lungs from five control animals, five animals with bronchial occlusion, and eight animals with oleic acid-induced pulmonary edema. Under base-line conditions when positive end-expiratory pressure (PEEP) was increased from 5 to 15 cmH2O, Ppa increased because Paw was fully transmitted to Ppa. After bronchial occlusion at 5 cmH2O PEEP, observed QS/QT was less than predicted (P less than 0.01) on the basis of regional conductances, but at 15 cmH2O observed and predicted QS/QT did not differ (P greater than 0.5). In five of eight dogs with oleic acid-induced pulmonary edema. QS/QT increased when PEEP was increased. We conclude that pressure barriers that create nonuniform driving pressures cause a decrease in Ppa and a redistribution of blood flow.

Influence of hematocrit and temperature on solubility of acetylene and dimethyl ether.

Estimation of pulmonary capillary blood flow (Qc) and tissue volume (Vt) by rebreathing acetylene (Ac) and dimethyl ether (DME) is dependent on the blood-gas partition coefficient (lambda) of these gases. We investigated the effects of hematocrit (Hct) and temperature on lambda. Human and canine blood was equilibrated at different Hct for 40 min at 37.5 degrees C with 1% Ac and 1% DME in air. Gas concentration in the headspace was measured by gas chromatography. lambda Ac increased with increasing Hct, suggesting a slight lipid affinity, whereas lambda DME decreased with increasing Hct, displaying the known blood-water content dependence. The influence of temperature on lambda in the range of 32--47 degrees C was investigated also. lambda Ac and lambda DME decreased as the temperature was raised, in accordance with the Gibbs-Helm-holtz equation. Empirical equations were derived to relate lambda to to Hct and temperature.