Microemboli reduce phase III slopes of CO2 and invert phase III slopes of infused SF6.

We investigated the effect of increasing doses of intravenously infused glass microspheres (mean diameter 125 microns) on gas exchange in anesthetized, heparinized, mechanically ventilated goats (VT = 16-18 ml/kg). Breath-by-breath CO2 expirograms were collected using a computerized system (Study A) during the infusion of a total of 15 g of microspheres. We found a 50% decrease in extravascular lung water by indicator dilution with a corresponding doubling of alveolar dead space (VDalv). Airways deadspace (VDaw) decreased by 13 ml (10%) and mean normalized phase III slope for CO2 decreased from 0.23 to -0.08 L-1 becoming negative in 3 of 5 animals. In a second study (Study B), simultaneous breath-by-breath CO2 and infused SF6 expirograms were collected using an infrared CO2 analyzer and a mass spectrometer. Under baseline conditions VDaw for CO2 was smaller than for SF6 and the ratio of the phase III slope for SF6 to the phase III slope for CO2 was 1.39. Following embolization there were no differences in VDaw between the two gases, however, the phase III slope for CO2 became either slightly negative or extremely flat, while the phase III slope for SF6 became negative in 73% of the breaths (-0.17 L-1, P < 0.05). Negative phase III slopes have been predicted by a single path model when blood flow is confined to the most mouthward generations of the acinus (Schwardt et al., Ann. Biomed. Engin, 19: 679-697, 1991). The agreement between the numerical model and the experimental data is consistent with a serial distribution of blood flow within the acinus.

Modelling steady state pulmonary elimination of He, SF6 and CO2: effect of morphometry.

We studied the influence of acinar morphometry on the shape of simulated expirograms computed from a single path convection-diffusion model that includes a source term for gas evolution from the blood (Scherer et al., J. Appl. Physiol. 64: 1022-1029, 1988). Acinar structure was obtained from published data of 3 different lung morphometries. The simulations were performed over a range of tidal volumes (VT) and breathing frequencies (f) comparable to those observed in a previously reported human study. Airways dead space (VDaw) increased with VT in all the morphometric models tested and in the experimental data. The increase in VDaw with VT was inversely related to the diffusivity of the evolving gas and to the rate of increase in airway cross-section of the most mouthward (proximal) alveolated generations of the models. Normalized phase III slope for all the gases decreased with increasing VT in all the models as was previously reported for healthy human subjects. In the model simulations, the greatest sensitivity of phase III slope to VT was seen with the least diffusible gas using the airway morphometry with the smallest cross-sectional areas in the proximal alveolated generations. We conclude that both VDaw and phase III slope of an evolving gas are sensitive to the geometry of the proximal acinar airways and that this is manifest by their dependence on tidal volume, breathing frequency, molecular diffusivity and alveolar/blood source emission rate. The model simulations indicate that heterogeneity of gas washout is not required to explain the magnitude of the phase III slope in healthy human subjects.

Diffusivity, respiratory rate and tidal volume influence inert gas expirograms.

We modified, and developed software for, a computer-controlled quadrupole mass spectrometer to measure complete breath-by-breath expirograms of helium (He) and sulfur hexafluoride (SF6) exhaled during the infusion of saline saturated with the inert gases. He and SF6 have similar blood solubilities but very different gas phase diffusivities allowing examination of the influence of gas phase diffusivity on steady state inert gas expirograms. We studied six normal human volunteers in nine separate studies and examined the influence of tidal volume (VT) and breathing frequency (f) on the airways dead space (VDaw) and alveolar plateau slope (phase III) for the inert gases and CO2. The experimental data showed a reduction in VDaw with rapid shallow breathing, while phase III slope increased by a factor of two to three. We critically evaluated the data and methodology of these and previously reported studies of continuous and single breath washout of He and SF6. In general the 15 to 20 ml differences in VDaw between He and SF6 were in keeping with previous studies by others. The ratio of phase III slopes of SF6 to He reported by us previously (Scherer et al., J. Appl. Physiol. 64: 1022-1029, 1988) was 3.13. In the current study, which includes the analysis of more than 400 He and SF6 breaths, the ratio of SF6 to He slope was 1.85. The difference between the two studies was largely related to the improved methodology of the current study, particularly for the measurement of He. The results support the conclusion that diffusivity is an important component of both phase II and phase III of the expirogram. However, the difference in phase III between He and SF6 is somewhat less than previously reported.

Sensitivity of CO2 washout to changes in acinar structure in a single-path model of lung airways.

A numerical solution of the convection-diffusion equation with an alveolar source term in a single-path model (SPM) of the lung airways simulates steady state CO2 washout. The SPM is used to examine the effects of independent changes in physiologic and acinar structure parameters on the slope and height of Phase III of the single-breath CO2 washout curve. The parameters investigated include tidal volume, breathing frequency, total cardiac output, pulmonary arterial CO2 tension, functional residual capacity, pulmonary bloodflow distribution, alveolar volume, total acinar airway cross sectional area, and gas-phase molecular diffusivity. Reduced tidal volume causes significant steepening of Phase III, which agrees well with experimental data. Simulations with a fixed frequency and tidal volume show that changes in blood-flow distribution, model airway cross section, and gas diffusivity strongly affect the slope of Phase III while changes in cardiac output and in pulmonary arterial CO2 tension strongly affect the height of Phase III. The paper also discusses differing explanations for the slope of Phase III, including sequential emptying, stratified inhomogeneity, and the issue of asymmetry, in the context of the SPM.

Numerical and experimental study of steady-state CO2 and inert gas washout.

The predictions of a single-path trumpet-bell numerical model of steady-state CO2 and infused He and sulfur hexafluoride (SF6) washout were compared with experimental measurements on healthy human volunteers. The mathematical model used was a numerical solution of the classic airway convention-diffusion equation with the addition of a distributed source term at the alveolar end. In the human studies, a static sampling technique was used to measure the exhaled concentrations and phase III slopes of CO2, He, and SF6 during the intravenous infusion of saline saturated with a mixture of the two inert gases. We found good agreement between the experimentally determined normalized slopes (phase III slope divided by mixed expired concentration) and the numerically determined normalized slopes in the model with no free parameters other than the physiological ones of upper airway dead space, tidal volume, breathing frequency, and breathing pattern (sinusoidal). We conclude 1) that the single-path (Weibel) trumpet-bell anatomic model used in conjunction with the airway convection-diffusion equation with a distributed source term is adequate to describe the steady-state lung washout of CO2 and infused He and SF6 in normal lungs and 2) that the interfacial area separating the tidal volume fron from the functional residual capacity gas, through which gas diffusion into the moving tidal volume occurs, exerts a major effect on the normalized slopes of phase III.

Baroreceptor responses derived from a fundamental concept.

A model is presented that relates the change in baroreceptor firing rate to a step change in blood pressure. This relationship is nonlinear since the alteration in rate of firing depends on the current rate of firing. It is shown that this simple relationship embodies all currently established baroreceptor response modes. The model needs refinement to allow for effects arising from the properties of the tissue matrix in which the receptors are embedded. Further analysis is precluded at present owing to paucity of quantitative experimental data.