Hypoxic tachycardia is not a result of increased respiratory activity in healthy subjects.

Animal data suggest that hypoxic tachycardia is secondary to hyperpnoea, and for years this observation has been extrapolated to humans, despite a lack of experimental evidence. We addressed this issue in 17 volunteers aged 29 ± 7 (SD) years. A transient hypoxia test, comprising several nitrogen-breathing episodes, was performed twice in each subject. In the first test, the subject breathed spontaneously (spontaneous breathing). In the second test, the subject was repeatedly asked to adjust his or her depth and rate of breathing according to visual (real-time inspiratory flow) and auditory (metronome sound) cues, respectively (controlled breathing), to maintain respiration at the resting level during nitrogen-breathing episodes. Hypoxic responsiveness, including minute ventilation [Hyp-VI; in liters per minute per percentage of blood oxygen saturation ( S p O 2 )], tidal volume [Hyp-VT; in litres per S p O 2 ], heart rate [Hyp-HR; in beats per minute per S p O 2 ], systolic [Hyp-SBP; in millimetres of mercury per S p O 2 ] and mean blood pressure [Hyp-MAP; in millimetres of mercury per S p O 2 ] and systemic vascular resistance [Hyp-SVR; in dynes seconds (centimetres)-5 per S p O 2 ] was calculated as the slope of the regression line relating the variable to S p O 2 , including pre- and post-hypoxic values. The Hyp-VI and Hyp-VT were reduced by 69 ± 25 and 75 ± 10%, respectively, in controlled versus spontaneous breathing (Hyp-VI, -0.30 ± 0.15 versus -0.11 ± 0.09; Hyp-VT, -0.030 ± 0.024 versus -0.007 ± 0.004; both P < 0.001). However, the cardiovascular responses did not differ between spontaneous and controlled breathing (Hyp-HR, -0.62 ± 0.24 versus -0.71 ± 0.33; Hyp-MAP, -0.43 ± 0.19 versus -0.47 ± 0.21; Hyp-SVR, 9.15 ± 5.22 versus 9.53 ± 5.57; all P ≥ 0.22), indicating that hypoxic tachycardia is not secondary to hyperpnoea. Hyp-HR was correlated with Hyp-SVR (r = -074 and -0.80 for spontaneous and controlled breathing, respectively; both P < 0.05) and resting barosensitivity assessed with the sequence technique (r = -0.60 for spontaneous breathing; P < 0.05). This might suggest that the baroreflex mechanism is involved.