Hidden Overshoot in Cerebral Blood Flow: Implications for Accurate Assessment of Cerebral CO 2 Reactivity

The impact of temporal and individual differences in respiratory and cerebrovascular responses on the assessment of cerebrovascular carbon dioxide (CO 2 ) reactivity (CVR) using the Douglas bag method remains unclear. We hypothesized that accurate CVR assessment requires attaining a respiratory steady-state during CO 2 inhalation, followed by stabilization of middle cerebral artery mean blood velocity (MCA V mean ), mediated by changes in arterial CO 2 tension. To test this hypothesis, we conducted three 8-minute CO 2 inhalation trials (at 0%, 3.5%, and 5%) and analyzed minute-by-minute responses in 32 healthy participants. Twelve participants (37.5%) exhibited an overshoot in MCA V mean immediately after the onset of CO 2 inhalation. When averaged across all participants, CVR reached an apparent steady-state within 2–3 minutes. In contrast, in the overshoot group, MCA V mean increased acutely and thereafter decreased toward a plateau at approximately 6-7 minutes, whereas ventilation increased progressively throughout the exposure. The magnitude of the MCA V mean overshoot was significantly associated with central respiratory chemoreflex sensitivity (r = 0.560, P = 0.0009), but not with CVR (r = −0.206, P = 0.258). These findings indicate that inter-individual variability in CVR estimates obtained using the Douglas bag method may be attributable to inter-individual differences in cerebrovascular regulation arising from respiratory responses to CO 2 inhalation. Consequently, assessing CVR during the early phase of CO 2 inhalation may over- or underestimate steady-state cerebrovascular responsiveness because of inter-individual differences in respiratory or cerebrovascular control. Accordingly, we recommend a minimum CO 2 inhalation duration of 6–7 minutes to ensure attainment of a steady-state when assessing CVR.