Search

Competing Interests: Competing interests: LH has received research funding support from Bayer PLC. VM has received lecture and meeting attendance support from Chiesi, GSK and AstraZeneca.

Ventilatory acclimatization to hypoxia is associated with an increase in ventilation under conditions of acute hyperoxia (VEhyperoxia) and an increase in acute hypoxic ventilatory response (AHVR). This study compares 48-h exposures to isocapnic hypoxia (protocol I) with 48-h exposures to poikilocapnic hypoxia (protocol P) in 10 subjects to assess the importance of hypocapnic alkalosis in generating the changes observed in ventilatory acclimatization to hypoxia. During both hypoxic exposures, end-tidal PO2 was maintained at 60 Torr, with end-tidal PCO2 held at the subject's prehypoxic level (protocol I) or uncontrolled (protocol P). VEhyperoxia and AHVR were assessed regularly throughout the exposures. VEhyperoxia (P < 0.001, ANOVA) and AHVR (P < 0.001) increased during the hypoxic exposures, with no significant differences between protocols I and P. The increase in VEhyperoxia was associated with an increase in slope of the ventilation-end-tidal PCO2 response (P < 0.001) with no significant change in intercept. These results suggest that changes in respiratory control early in ventilatory acclimatization to hypoxia result from the effects of hypoxia per se and not the alkalosis normally accompanying hypoxia.

The ventilatory response to isocapnic hypoxia is biphasic, which makes any experimental assessment of the relationship between the acute (peak) ventilatory response and the level of hypoxia difficult. This study explored whether one particular protocol could be useful for determining this relationship. The protocol consisted of exposing subjects to seven different levels of isocapnic hypoxia, each of which lasted 50 sec. In order to test whether the order of the hypoxic exposure had any effect on the outcome, the steps were performed both in increasing and decreasing severity of hypoxia, and the ventilatory responses compared. Twelve subjects were studied, and each test was repeated four times in each subject. PETCO2 was held at 2 mmHg above resting throughout. The ventilations obtained at the lowest level of PETO2 employed were clearly different between the two protocols. However, provided that these ventilations were excluded, no significant differences were present between the results from the ascending and descending exposures (ANOVA). This finding suggests that the rate of change of PO2 in these protocols was sufficiently slow for a full ventilatory response to develop, but also sufficiently fast to prevent significant ventilatory depression from occurring.

In the preceding companion paper (L. S. G. E. Howard and P.A. Robbins, J. Appl. Physiol. 78: 1092-1097, 1995), we showed that ventilation rises during 8 h of isocapnic hypoxia. In the present study we report the changes that occur in the ventilatory response to acute hypoxia (AHVR) over 8 h of both isocapnic and poikilocapnic hypoxia. Ten subjects completed the study. Each was seated inside a chamber in which the inspired gas could be controlled so as to maintain the desired end-tidal gases (sampled via nasal catheter) constant. Three 8-h protocols were compared: 1) isocapnic hypoxia, at an end-tidal PO2 of 55 Torr with the end-tidal PCO2 held at the subject's resting value; 2) poikilocapnic hypoxia, at the same end-tidal PO2; and 3) control, where the inspired gas was air. AHVR was measured before and at 20 min and 4 and 8 h after the start of the experiment. A sequence of hypoxic square waves and sawtooth inputs was imposed by an end-tidal forcing system, with the subject breathing through a mouthpiece. End-tidal PCO2 was held constant at 1-1.5 Torr above resting. Values for hypoxic sensitivity (Gp; 1.min-1.%-1) and hypoxia-independent ventilation (Vc; l/min) were calculated for each test of AHVR. Both Gp and Vc increased significantly during both hypoxic exposures in relation to control (P < 0.001, analysis of variance). Over the 8-h period, increases in Gp were 87% in isocapnic hypoxia and 44% in poikilocapnic hypoxia, and increases in Vc were 89% in isocapnic hypoxia and 84% in poikilocapnic hypoxia. There were no significant differences between the isocapnic and poikilocapnic exposures. We conclude that Gp and Vc rise mainly as result of hypoxia per se and not the associated alkalosis.

Almost all studies of the effects of prolonged hypoxia on ventilation (VE) in humans have been performed with the end-tidal PCO2 (PETCO2) left uncontrolled. The purpose of this study was to compare the effects of 8 h of hypoxia with PETCO2 held constant with 8 h of hypoxia with PETCO2 left uncontrolled. Ten subjects completed the study. Each was seated inside a chamber in which the inspired gas could be controlled so as to maintain the desired partial pressures of end-tidal gases (sampled via nasal catheter) constant (see L.S.G.E. Howard et al. J. Appl. Physiol. 78:1088-1091, 1995.). Three 8-h protocols were employed: 1) isocapnic hypoxia, at an end-tidal PO2 of 55 Torr with PETCO2 held at the subject's resting value; 2) poikilocapnic hypoxia, at the same end-tidal PO2; and 3) control, where the inspired gas was air. VE was measured (over 3 min) at 0 and 20 min and at hourly intervals between 1.5 and 7.5 h. There was a rise in VE during isocapnic hypoxia [from an initial VE of 16.2 +/- 1.3 (SE) l/min to a final VE of 24.8 +/- 1.6 l/min], which was significant compared with poikilocapnic hypoxia and control values (P < 0.001, analysis of variance). There was no significant progressive rise in VE during poikilocapnic hypoxia compared with control values. These results show that isocapnic hypoxia produces a progressive increase in VE when sustained over an 8-h period. The onset of this response is faster than has been noted in studies of the progressive rise in VE associated with the poikilocapnic hypoxia of altitude.

This study examined whether the form of the acute ventilatory response to different levels of end-tidal PO2 (PETO2) could be determined from a progressive series of steps in PETO2 at constant end-tidal PCO2. Seven levels of PETO2 were employed each lasting 2 min. The steps were performed in both ascending and descending order to test whether the ventilatory responses were affected by the order of the hypoxic exposures. These exposures were carried out both with and without a prior 20 min period of isocapnic hypoxia (PETO2 = 50 mmHg). Each protocol was undertaken 6 times in each of 6 subjects. With prior exposure to hypoxia, the order in which the steps were performed affected the ventilatory response (P < 0.005, ANOVA). Without prior exposure to hypoxia, this finding did not quite reach significance (P < 0.056), unless one particular abnormal subject was excluded (P < 0.001). It is concluded that the order of the hypoxic exposures in this particular test affects the form of the hypoxic response.

This study assessed whether the form of the peripheral chemoreflex response to hypoxia depends on the magnitude of the stimulus. Two amplitudes of square-wave hypoxic stimulation were employed: small amplitude (SO) PETO2 from 63.2 to 54.9 Torr, and large amplitude (LO) PETO2 from 73.0 to 48.0 Torr. Each was studied at two levels of PETCO2: 2 Torr above resting PETCO2 (EC), and 7 Torr above resting PETCO2 (HC). Each protocol was repeated 6 times on 5 subjects. To assess the form of the response, a simple first-order model was fitted to the data which incorporated a pure delay (Td) and time constant (tau). Average parameter values (sec) were: ECSO tau = 4.07, Td = 6.69; ECLO tau = 8.82, Td = 4.91; HCSO tau = 5.22, Td = 7.08; HCLO tau = 9.96, Td = 4.39. ANOVA demonstrated modest but significant differences for loge(tau) (P < 0.01) and Td (P < 0.02) between the two hypoxic step magnitudes, with tau longer and Td shorter for the larger step size, but no differences were found between the parameter values for the two CO2 levels. We conclude that the form of the response of the peripheral chemoreflex to hypoxia depends upon the magnitude of the stimulus.