A dynamic analysis of the ventilatory response to hypoxia in man.

1. The dynamics of the ventilatory response to isocapnic hypoxia were studied in seven healthy subjects using four different levels of hypoxia, (inspired oxygen pressures, PI,O2 equal to 110, 100, 80 and 60 mmHg) successively increasing and decreasing stepwise. 2. Five such progressions were performed for each subject, corresponding to five different durations of the steps (t) ranging between 0.33 and 5.00 min. The overall duration of one test (T) was taken as the sum of the seven successive PI,O2 hypoxic steps (t) plus one step t of air breathing. Thus, the values of T ranged between 2.6 and 40.0 min. 3. End-tidal CO2 pressure was maintained constant (+/- 1 mmHg) throughout the test by manipulation of inspired CO2 pressure. 4. We measured, as a function of T, (i) the magnitude of the loops formed by the ventilatory response curves (PA,O2-VE) as measured by their surface area (S), (ii) the magnitude of ventilatory response to each rising hypoxic step, and (iii) the difference between resting VE and VE observed at PA,O2 equal to 50 mmHg (delta V50). On average, we found one maximum in absolute value of S at T = 8 min and one minimum at T = 12 min, along with two maxima of ventilatory response at T values of 8 and 24 min. 5. The same measurements were made on tidal volume response curves (PA,O2-VT) and ventilatory frequency response curves (PA,O2-f): on average we observed two non-significant peaks in the progression with T of VT and S(VT) and two significant peaks in that of delta VT,50 for T = 8 and T = 24 min. No significant peak was observed in the progression with T of f curve parameters. 6. These results are discussed together with the current dynamic model of the ventilatory control system, which includes a central neural controller with no dynamics of its own and a linear response to chemoreceptor inputs. We discuss the physiological meaning of a negative loop area in relation to the previously described depressant effect of hypoxia upon the brain stem. 7. We conclude that the dynamics of the controlling neuronal network are responsible for the observed singularities which result from differential sensitivity properties of the controller. We propose the existence of discrete excitatory states of the controller as a possible explanation of the shape of the steady-state response curve to hypoxia and of the loop variations.