Quantification of pulmonary vascular occlusion in dogs by use of the diffusing capacity

The purpose of these experiments was to quantify stagnant intrapulmonary blood caused by a pulmonary arterial occlusion (PAO). The hypothesis was that the diffusing capacity of the lung for CO (DLCO) would be altered little by PAO when measured with the usual inspired concentrations (0.3%) of CO, since stagnant blood distal to the occlusion takes up CO for 20 s or more before significant CO backpressure would develop. However, higher levels of CO (i.e., greater than or equal to 3%) would equilibrate faster with capillary blood (within 5-10 s), and DLCO measured 10-20 s subsequent to the high CO exposure would reflect only the DLCO in the unoccluded regions. Thus the fractional reduction in DLCO measured with 3% CO, with respect to that measured with 0.3% CO, should be related to the fractional occlusion of the pulmonary artery in a predictable way. We occluded the right pulmonary artery (RPAO), the left pulmonary artery (LPAO), or the left lower lobar artery (LLPAO) and found that DLCO measured during rebreathing a 0.3% CO mixture was 80, 87, and 94%, respectively, of the preocclusion value, whereas the DLCO measured during rebreathing a 3.3% CO mixture was 59, 73, and 87% of the preocclusion value. A computer model was developed to predict the reduction in DLCO at different levels of CO exposure that would be caused by varying fractions of PAO. Our data indicated that RPAO corresponded to a 42% vascular occlusion, LPAO a 35% occlusion, and LLPAO a 20% occlusion. Measurement of DLCO using low and high concentrations of CO might be useful in assessing the fraction of vascular bed occluded and in following noninvasively the course of vascular occlusion in a variety of pulmonary diseases.