Determination of distribution of diffusing capacity in relation to blood flow in the human lung

Abstract

A method for appraising the distribution of diffusing capacity of the lungs (D(L)) in relationship to pulmonary capillary blood flow ([unk]Q(C)) in normal human subjects was derived from measurements of oxygen diffusing capacity (D(LO2)) and carbon monoxide diffusing capacity (D(LCO)) performed during breath holding. This method utilizes the fact that the observed D(LO2) is considerably reduced in value if uneven distribution of D(L) with respect to [unk]Q(C) (uneven D(L)/[unk]Q(C)) is present. In contrast, D(LCO) is barely affected by uneven D(L)/[unk]Q(C), and from its measured value one can calculate the value D(LO2) would have if no uneven D(L)/[unk]Q(C) were present (true D(LO2)). Once observed D(LO2) and true D(LO2) are known, the degree of uneven D(L)/[unk]Q(C) in the lung can be calculated. In five normal, resting, sitting subjects average values for true D(LO2) were 57 ml per (minute x mm Hg), and the directly measured D(LO2) was 33 ml per (minute x mm Hg). These values could be explained if one-half of total [unk]Q(C) were distributed to approximately 15% of total D(L). These measurements did not permit the determination of the alveolar to end capillary O(2) gradient, but calculations demonstrate that an important factor in determining its size may be the pattern of uneven D(L)/[unk]Q(C) present in the lungs. Estimations of the alveolar-end capillary O(2) gradient from measurements of D(LCO) or D(LO2) that do not take into account uneven D(L)/[unk]Q(C) may underestimate its size.