Tidal volume, cardiac output and functional residual capacity determine end-tidal CO2 transient during standing up in humans

Abstract

In man assuming the upright position, end-tidal P(CO(2)) (P(ETCO(2))) decreases. With the rising interest in cerebral autoregulation during posture change, which is known to be affected by P(ETCO(2)), we sought to determine the factors leading to hypocapnia during standing up from the supine position. To study the contribution of an increase in tidal volume (V(T)) and breathing frequency, a decrease in stroke volume (SV), a ventilation-perfusion (V/Q) gradient and an increase in functional residual capacity (FRC) to hypocapnia in the standing position, we developed a mathematical model of the lung to follow breath-to-breath variations in P(ETCO(2)). A gravity-induced apical-to-basal V/Q gradient in the lung was modelled using nine lung segments. We tested the model using an eight-subject data set with measurements of V(T), pulmonary O(2) uptake and breath-to-breath lumped SV. On average, the P(ETCO(2)) decreased from 40 mmHg to 36 mmHg after 150 s standing. Results show that the model is able to track breath-to-breath P(ETCO(2)) variations (r(2)= 0.74, P P 0.05). Model parameter sensitivity analysis demonstrates that the decrease in P(ETCO(2)) during standing is due primarily to increased V(T), and transiently to decreased SV and increased FRC; a slight gravity-induced V/Q mismatch also contributes to the hypocapnia. The influence of cardiac output on hypocapnia in the standing position was verified in experiments on human subjects, where first breathing alone, and then breathing, FRC and V/Q were controlled.

Figure 1. Diagram of the P_ETCO2 model

The left panel represents circulation with an arterial volume V_a, a venous volume V_v, a lung capillary volume V_cap, and circulating stroke volume per breath SV. The right panel represents ventilation with a functional residual capacity FRC, a respiratory dead space V_D, and a tidal volume V_T. The distribution of SV and V_T as shown here are for the upright position; in the supine position SV and V_T are equally distributed over apical and basal lung segments (1–9).

Figure 2. Individual P_ETCO2 recordings and model simulations during lying down and standing

Plots of breath-to-breath P_ETCO2 of each individual subject. Subjects 1–8 are represented in panels A–H, respectively. Each panel contains a plot of breath-to-breath P_ETCO2 measurements (•) during 150 s supine and 150 s of standing, and a model simulation (^) of the same period. Arrows indicate posture change from supine to standing at time zero.

Figure 3. Pooled data of 300 s of P_ETCO2 registration in 8 subjects, 150 s supine and 150 s standing, plotted against the output of a model run of the same time period

The number of data points is 583, representing the total of 583 breaths. A, pooled data of computed P_ETCO2 (M-P_ETCO2) plotted against measured P_ETCO2. B, pooled data scatter diagram of differences between measured P_ETCO2 and computed P_ETCO2 (M-P_ETCO2) against their mean. Horizontal lines indicate mean ± 1.96 s.d.

Figure 4. P_ETCO2 and Q during controlled breathing in 7 healthy subjects

Results of supine and standing (A) and inflated and deflated leg cuffs (B) protocols. Symbols represent average end-tidal P_CO2 (top) and cardiac output (Q, bottom) during 5 min. The lines link the results of a particular subject. Asterisk indicates P < 0.01.

Figure 5. Parameter sensitivity analysis of SV and FRC

Each line represents a model run where after 200 breaths under baseline conditions the input is changed from its baseline value by –10 to +10%, in steps of 2%. In A the input is SV of the heart, in B the input is FRC. Note difference in ordinate scale.

Figure 6. Analysis of effect of each input variable on P_ETCO2

Parameter sensitivity analysis of model input (T_resp, V_T, V̇_O2) and model parameters (V_D and RQ). Model input is changed from baseline by –10 to +10% (see Methods) and model output (P_ETCO2) determined after 900 s.

Figure 7. Effect of active-standing induced variations in model input variables on P_ETCO2

Each thin line indicates the output of a model run starting with supine settings and with parameter settings varied at 200 s as is likely to occur on standing up: Q, output when Q was reduced by 40%, V_T, output when V_T was increased by 20%, FRC, output when FRC was increased by 20%, V_D, output when V_D was increased by 70 ml, V/Q, output when V/Q shifted from equal distribution to model settings for a gravitationally induced V/Q mismatch. The thick line labelled ALL indicates model P_ETCO2 levels when all of these changes occurred simultaneously.