BackgroundIt has been previously documented that the circulatory interaction resulting from the co‐activation of the arterial chemoreflex and the exercise pressor reflex is different when the chemoreflex is evoked by hypoxia vs. hypercapnia. While the interaction effect on sympathoexcitation is simply additive during hypercapnic exercise, it is hyper‐additive during hypoxic exercise. We hypothesize that a relative hypoperfusion and compromised O2 delivery (DO2) to O2‐sensitive brainstem areas/cardiovascular control centers may trigger the Cushing’s mechanism and thereby account, in part, for this potentiated sympathetic activation during exercise in hypoxia. However, although hypoxia, hypercapnia, and exercise are recognized for their strong influence on the cerebrovascular control, the interaction between these stressors and the resulting impact upon regional brain perfusion remains unknown.PurposeTo evaluate the cerebrovascular consequence of the interaction of hypoxia, or hypercapnia, with exercise in humans.MethodOn separate occasions, six healthy volunteers rested and performed rhythmic handgrip exercise (1 Hz, 50% MVC, 3 min) in normoxia (NormRest or NormExercise: SpO2 ~97%, PETO2 ~83 mmHg, PETCO2 ~36 mmHg), normocapnic hypoxia (HypoRestor HypoExercise: SpO2 ~84%, PETO2 ~49 mmHg, PETCO2 ~36 mmHg), and normoxic hypercapnia (HyperRest or HyperExercise: SpO2 ~98%, PETO2 ~111 mmHg, PETCO2 ~46 mmHg). Doppler ultrasound was used to quantify blood flow in the contralateral internal carotid artery (ICA‐BF) and ipsilateral vertebral artery (VA‐BF), and DO2 to the anterior (via ICA‐BF) and posterior (via VA‐BF) brain regions.ResultsExercise alone (i.e., ΔNormExercise‐NormRest) significantly increased ICA‐BF (+52 ± 24 mL· min‐1), VA‐BF (+27 ± 13 mL ·min‐1), and anterior (+11 ± 5 mL ·min‐1) and posterior DO2 (+6 ± 3 mL· min‐1). Hypoxia alone (i.e., ΔHypoRest‐NormRest) significantly increased ICA‐BF (+54 ± 17 mL· min‐1) and VA‐BF (+15 ± 5 mL ·min‐1), but did not affect anterior and posterior DO2 (p ≥ 0.23). Hypercapnia alone (i.e., ΔHyperRest‐NormRest) significantly increased ICA‐BF (+135 ± 27 mL· min‐1), VA‐BF (+54 ± 11 mL· min‐1), and anterior (+28 ± 5 mL ·min‐1) and posterior DO2 (+11 ± 2 mL ·min‐1). During hypoxic exercise (i.e., ΔHypoExercise‐NormRest), the responses of ICA‐BF, VA‐BF, and regional DO2 were significantly lower compared to the sum of the responses elicited by each stressor alone (ICA‐BF: 48 ± 20 vs. 106 ± 37 mL ·min‐1; VA‐BF: 24 ± 8 vs. 41 ± 10 mL ·min‐1; anterior DO2: 0 ± 4 vs. 12 ± 7 mL· min‐1; posterior DO2: 1 ± 1 vs. 5 ± 2 mL· min‐1). During hypercapnic exercise (i.e., ΔHyperExercise‐NormRest), the observed cerebrovascular responses did not differ from the summated responses (p ≥ 0.12).ConclusionWhile the interaction between hypercapnia and exercise is additive in terms of regional cerebral blood flow and DO2, the interaction between hypoxia and exercise is hypo‐additive. The resulting hypoperfusion and compromised DO2 to the brainstem (i.e., VA‐BF and posterior DO2), via the Cushing’s mechanism, might therefore contribute to the potentiated sympathoexcitation previously observed during hypoxic exercise.