Critique of conceptual basis of diffusing capacity estimates: a finite element analysis.

We present a simple geometric model of a pulmonary capillary segment containing a variable number of red blood cells. The pattern of CO transfer from alveolar air to capillary red blood cells in this model is accurately computed by a finite element method and used to explore conceptual flaws in the Roughton-Forster (RF) and morphometric methods of estimating pulmonary diffusing capacity for CO. The CO uptakes calculated by the finite element method at two alveolar O2 tensions are introduced into the RF model to determine whether the anatomically defined membrane component of diffusing capacity for CO (DmCO) and pulmonary capillary blood volume (Vc) are recovered. The same capillary model is also subjected to standard morphometric analysis. Results are compared at different levels of capillary hematocrit (Hct). The RF method accurately recovers DmCO and Vc at a low Hct but modestly overestimates DmCO and underestimates Vc at higher Hct; errors arise because conductance of the tissue-plasma membrane for CO varies with alveolar O2 tension. The morphometric method seriously overestimates DmCO because the true tissue-plasma resistance to diffusion is underestimated and the effective membrane utilized for diffusion is overestimated; these errors are accentuated by a low Hct.