Dynamic cerebral autoregulation during cognitive task: effect of hypoxia

Changes in cerebral blood flow (CBF) subsequent to alterations in the partial pressures of oxygen and carbon dioxide can modify dynamic cerebral autoregulation (CA). While cognitive activity increases CBF, the extent to which it impacts CA remains to be established. In the present study we determined whether dynamic CA would decrease during a cognitive task and whether hypoxia would further compound impairment. Fourteen young healthy subjects performed a simple Go/No-go task during normoxia and hypoxia (inspired O fraction = 12%), and the corresponding relationship between mean arterial pressure (MAP) and mean middle cerebral artery blood velocity (MCA V ) was examined. Dynamic CA and steady-state changes in MCA V in relation to changes in arterial pressure were evaluated with transfer function analysis. While MCA V increased during the cognitive activity ( P < 0.001), hypoxia did not cause any additional changes ( P = 0.804 vs. normoxia). Cognitive performance was also unaffected by hypoxia (reaction time, P = 0.712; error, P = 0.653). A decrease in the very low- and low-frequency phase shift (VLF and LF; P = 0.021 and P = 0.01) and an increase in LF gain were observed ( P = 0.037) during cognitive activity, implying impaired dynamic CA. While hypoxia also increased VLF gain ( P < 0.001), it failed to cause any additional modifications in dynamic CA. Collectively, our findings suggest that dynamic CA is impaired during cognitive activity independent of altered systemic O availability, although we acknowledge the interpretive complications associated with additional competing, albeit undefined, inputs that could potentially distort the MAP-MCA V relationship. NEW & NOTEWORTHY During normoxia, cognitive activity while increasing cerebral perfusion was shown to attenuate dynamic cerebral autoregulation (CA) yet failed to alter reaction time, thereby questioning its functional significance. No further changes were observed during hypoxia, suggesting that impaired dynamic CA occurs independently of altered systemic O availability. However, impaired dynamic CA may reflect a technical artifact, given the confounding influence of additional inputs that could potentially distort the mean arterial pressure-mean middle cerebral artery blood velocity relationship.