Measuring the human ventilatory and cerebral blood flow response to CO₂: A technical consideration for the end-tidal-to-arterial gas gradient

Our aim was to quantify the end-tidal-to-arterial gas gradients for O₂ (P_ET-PaO₂) and CO₂ (Pa-P_ETCO₂) during a CO₂ reactivity test to determine their influence on the cerebrovascular (CVR) and ventilatory (HCVR) response in subjects with (PFO+, n = 8) and without (PFO-, n = 7) a patent foramen ovale (PFO). We hypothesized that 1) the Pa-P_ETCO₂ would be greater in hypoxia compared with normoxia, 2) the Pa-P_ETCO₂ would be similar, whereas the P_ET-PaO₂ gradient would be greater in those with a PFO, 3) the HCVR and CVR would be underestimated when plotted against P_ETCO₂ compared with PaCO₂, and 4) previously derived prediction algorithms will accurately target PaCO₂. P_ETCO₂ was controlled by dynamic end-tidal forcing in steady-state steps of -8, -4, 0, +4, and +8 mmHg from baseline in normoxia and hypoxia. Minute ventilation (V_E), internal carotid artery blood flow (Q_ICA), middle cerebral artery blood velocity (MCAv), and temperature corrected end-tidal and arterial blood gases were measured throughout experimentation. HCVR and CVR were calculated using linear regression analysis by indexing V_E and relative changes in Q_ICA, and MCAv against P_ETCO₂, predicted PaCO₂, and measured PaCO₂. The Pa-P_ETCO₂ was similar between hypoxia and normoxia and PFO+ and PFO-. The P_ET-PaO₂ was greater in PFO+ by 2.1 mmHg during normoxia (P = 0.003). HCVR and CVR plotted against P_ETCO₂ underestimated HCVR and CVR indexed against PaCO₂ in normoxia and hypoxia. Our PaCO₂ prediction equation modestly improved estimates of HCVR and CVR. In summary, care must be taken when indexing reactivity measures to P_ETCO₂ compared with PaCO₂.