HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Ren, X. Articles by Robbins, P. A. Search for Related Content PubMed PubMed Citation Articles by Ren, X. Articles by Robbins, P. A.

Ventilatory responses to hypercapnia and hypoxia after 6 h passive hyperventilation in humans

Xiaohui Ren and Peter A. Robbins

MS 8553 Received 27 July 1998; accepted after revision 3 November 1998.

	ABSTRACT

Top Abstract Introduction Methods Results Discussion References

1. Acute exposure to hypoxia stimulates ventilation and induces hypocapnia. Long-term exposure to hypoxia generates changes in respiratory control known as ventilatory acclimatization to hypoxia. The object of this study was to investigate the degree to which the hyperventilation and hypocapnia can induce the changes known as ventilatory acclimatization to hypoxia, in the absence of the primary hypoxic stimulus itself.

2. Three 6 h protocols were each performed on twelve healthy volunteers: (1) passive hypocapnic hyperventilation, with end-tidal CO₂ pressure (P_ET,CO2) held 10 Torr below the eupnoeic value; (2) passive eucapnic hyperventilation, with P_ET,CO2 maintained eucapnic; (3) control.

3. Ventilatory responses to acute hypercapnia and hypoxia were assessed before and half an hour after each protocol.

4. The presence of prior hypocapnia, but not prior hyperventilation, caused a reduction in air-breathing P_ET,CO2 (P < 0·05, ANOVA), and a leftwards shift of the ventilatory response to hypercapnia (P < 0·05). The presence of prior hyperventilation, but not prior hypocapnia, caused an increase in the ventilatory sensitivity to CO₂ (P < 0·05). No significant effects of any protocol were detected on the ventilatory sensitivity to hypoxia.

5. We conclude that following 6 h of passive hyperventilation: (i) the left shift of the V_E-P_ET,CO2 relationship is due to alkalosis and not to hyperventilation; (ii) the increase in slope of the V_E-P_ET,CO2 relationship is due to the hyperventilation and not the alkalosis; and (iii) ventilatory sensitivity to hypoxia is unaltered.

	INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

On going to altitude, the fall in environmental P_O2 triggers a rise in ventilation and a fall in alveolar P_CO2. Over a subsequent period of hours to days, ventilation progressively rises further in a process known as ventilatory acclimatization to hypoxia (VAH). The relative roles of hypoxia, of the initial increase in ventilation, and of the associated hypocapnia in driving this process remain to be fully determined. The general purpose of this study was to examine the effects of the latter two factors (hyperventilation and hypocapnia) on the respiratory control system in the absence of the first factor (hypoxia) over a time course in which the initial stages of acclimatization are thought to occur.

More specifically, during VAH, there is a leftwards shift of the relationship between ventilation (_E) and end-tidal P_CO2 (P_ET,CO2), together with an increase in the slope (acute hypercapnic ventilatory response, AHCVR) of this relationship (Chiodi, 1957; Michel & Milledge, 1963; White et al. 1987). There is also an increase in the acute ventilatory sensitivity to hypoxia (acute hypoxic ventilatory response, AHVR) (White et al. 1987; Sato et al. 1992; Howard & Robbins, 1995). While it is well established that hyperventilation and/or hypocapnia can affect the _E-P_ET,CO2 relationship in the absence of hypoxia (Brown et al. 1948, 1950; Smith et al. 1962; Eger et al. 1968; Dempsey et al. 1975), the details remain unresolved. In particular, it remains unclear whether hyperventilation in the absence of hypocapnia does (Smith et al. 1962) or does not (Eger et al. 1968) cause a leftwards shift in the _E-P_ET,CO2 relationship, and whether hyperventilation (with or without hypocapnia) increases (Brown et al. 1948, 1950) or does not increase (Eger et al. 1968) the slope of the _E-P_ET,CO2 relationship. Furthermore, while there exist studies emphasizing the crucial importance of sustained hypoxia in generating the increase in AHVR (Engwall & Bisgard, 1990; Howard & Robbins, 1995), we are unaware of any studies to determine whether a sustained period of hyperventilation with or without hypocapnia can also affect this sensitivity. The specific purpose of this study was to answer the above questions, firstly by examining the effects of prior hyperventilation both with and without hypocapnia on the _E-P_ET,CO2 relationship, and secondly by determining whether a period of prior hyperventilation with or without hypocapnia affects the magnitude of AHVR.

	METHODS

Top Abstract Introduction Methods Results Discussion References

Subjects

Twelve subjects (10 men, 2 women) volunteered to take part in the study. Their average age was 26 ± 10 years (mean± S.D.), their average weight was 67 ± 9 kg, and their average height was 174 ± 9 cm. All were healthy and none gave a history of respiratory symptoms or illnesses. All were non-smokers except for one subject (subject 1081) who smoked about five cigarettes per day. The requirements of the study were fully explained in both written and verbal forms to all subjects, but they remained naive as to the precise purpose of the study. Each subject gave informed consent prior to the study. The study was approved by the Central Oxford Research Ethics Committee.

Protocols

In order to study the effects of a sustained period of hyperventilation with and without an associated hypocapnia, three different protocols were employed. The first protocol (hypocapnic hyperventilation, protocol HH) employed a 6 h exposure to mechanical hyperventilation (tidal volume, 1-1·5 l; frequency, 10-15 breaths min^-1) during which P_ET,CO2 was allowed to fall to 10 Torr below the subjects' control values. The second protocol (eucapnic hyperventilation, protocol EH) employed a 6 h exposure to hyperventilation (tidal volume and frequency as for protocol HH) during which P_ET,CO2 was maintained at the subjects' control values by adding CO₂ to the inspirate. The third protocol (control, protocol C) was a control protocol where the subject breathed room air for 6 h through the nose mask.

The effects of the three different exposures were determined by making a number of measurements before and half an hour after the 6 h periods. A venous blood sample was taken to determine pH and standard bicarbonate. Then a value for air-breathing P_ET,CO2 was determined using a nasal catheter connected to a mass spectrometer. Following this, the ventilatory responses to hypoxia and hypercapnia were determined using an end-tidal forcing system. The particular procedure used is illustrated in Fig. 1. End-tidal PO₂ (P_ET,O2) was held at 100 Torr for the first 5 min. Immediately following this, six square waves in P_ET,O2 were generated, each with a period of 120 s and with P_ET,O2 stepping between 60 s at 50 Torr and 60 s at 100 Torr. Finally P_ET,O2 was raised to 200 Torr for the final 10 min while the ventilatory responses to hyperoxic hypercapnia were determined. P_ET,CO2 was held at 1-2 Torr above the subject's control value, as determined at the beginning of the experimental day, during all but the last 5 min of the procedure. For the final 5 min of hyperoxia, P_ET,CO2 was elevated to 10 Torr above the subject's control value in order to determine the hypercapnic response.

		View larger version [in this window] [in a new window]
Figure 1. Schematic for end-tidal PO2 and end-tidal PCO2 during measurement of AHVR and AHCVR

Prior to undertaking the three main experimental protocols, the subjects made one or two preliminary visits to the laboratory during which they were familiarized with the apparatus. For the main experiments, the protocols were undertaken in random order, and, for each subject, they were separated from one another by a period of at least one week. Each experiment started at 08.00- 09.00 h, and there was a break at 12.30 h of less than 20 min during which a light lunch was served. Female subjects only took part in the experiments within the first 14 days of their menstrual cycles. Subjects were requested to refrain from alcohol and caffeine-containing drink from the evening before each experimental day.

Technique for mechanical hyperventilation

A ventilator (Siemens 900B) was used to conduct the hyperventilation. Subjects were seated semi-recumbently and connected to the ventilator via a close-fitting nasal mask. Tidal volume and breathing frequency were set on the ventilator, and were chosen so that P_ET,CO2 would fall by at least 10 Torr. Inspiratory flow occupied 33 % of the respiratory cycle and there was no pause before expiration. Gases supplying the ventilator came from an air pump and a gas cylinder containing CO₂. The inspiratory gas was humidified and heated to 30°C before it was passed to the subject.

Airway pressure and flow were measured through a pressure transducer and a pneumotachometer incorporated in the ventilator. Gas was sampled continuously at the rate of 150 ml min^-1 through a catheter in the nose mask near one of the subject's nostrils and analysed for P_CO2 and P_O2 (Nomocap 200OXY, Datex). All of these variables were displayed and recorded on an electrostatic recorder (ES1000, Gould).

The P_ET,CO2 of the subject was regulated by adjusting the flow of CO₂ from the CO₂ cylinder into the ventilator. The breath-by-breath values for P_ET,CO2 recorded from the subject provided the necessary feedback to the experimenter for this purpose.

Measurements of AHVR and AHCVR

The ventilatory responses to acute hypoxia and hypercapnia were determined using a dynamic end-tidal forcing system. The subject was seated in an upright position and breathed through a mouthpiece with his/her nose occluded by a clip. Respiratory volumes were measured by a turbine volume-measuring device (Howson et al. 1986) fixed in series with the mouthpiece. Respiratory flows and timing information were obtained with a pneumotachograph. The total dead space associated with the apparatus was 100 ml. Gas was sampled continuously from this dead space close to the mouth at a rate of 80 ml min^-1 and analysed by mass spectrometry for P_O2 and P_CO2. The data were recorded by a computer. In addition, the computer was used to pick out the ends of inspiratory and expiratory phase from the flow data to determine inspiratory and expiratory durations, volumes, P_ET,O2 and P_ET,CO2. A pulse oximeter was attached to the forefinger to monitor the O₂ saturation of the blood.

Before the experiment started, a 'forcing function' was calculated which consisted of the predicted inspired gas compositions on a second-by-second basis that would be required to produce the desired levels of P_ET,O2 and P_ET,CO2 in the subject. This forcing function was entered into a controlling computer and used to generate the initial gas mixtures. During the course of the experiment, actual values for P_ET,O2 and P_ET,CO2 were passed to the controlling computer from the data-acquisition computer, and the deviations of these actual values from the desired values were used to adjust the new inspired gas mixtures using an integral-proportional feedback control scheme. The controlling computer adjusted the inspired partial pressures of N₂, O₂ and CO₂ through a fast gas-mixing system (Howson et al. 1987). This control scheme has been described in more detail elsewhere (Robbins et al. 1982).

Model fitting

In order to quantify AHVR from the data, the responses to the six square waves of hypoxia were fitted by a single compartment model as described by Clement & Robbins (1993, model 3). In this model, total _E is divided into central (_c) and peripheral (_p) components. In our assessments of AHVR, isocapnia was maintained, and thus _c is assumed to be constant. As the hypoxic stimulus varied over time, _p is represented by a dynamic first-order model:

(d_p/dt) + _p = G_p[1 - S(t - T_d)], (1)

where is the time constant, t is time, T_d is the time delay, Gp is the hypoxic sensitivity, and S is the saturation, calculated from P_O2 using the haemoglobin desaturation function described by Severinghaus (1979).

The fitting technique used was as described by Clement & Robbins (1993). First, eqn (1) was solved by integration assuming a constant input over a single breath i to give a difference equation describing _p for breath i in terms of S, breath duration (T_B), and the value of _p for the previous breath (i - 1), as well as in terms of the parameters of the model:

(_p)_i =

G_p [1 - S(t - T_d)] - {G_p [1 - S(t - T_d)] - (_p)_{i - 1}} exp[-(T_B)_i/]. (2)

From this,

(_E)_i = (_p)_i + _c. (3)

Equations (2) and (3) now enable the parameters of the model to be estimated by non-linear regression. This was undertaken by using the Numerical Algorithms Group (Oxford, UK) FORTRAN library routine E04FDF to minimize the sum of squares of the residuals. All of the parameters were constrained to be > 0, and the dynamic parameters were constrained to be < 30 s. The squared residual was weighted by (T_B)_i to avoid the fit being weighted towards periods of higher frequency _E.

Statistical analysis

The statistical significance of the results was assessed using analysis of variance (ANOVA). First, the variance was partitioned by the random factor of subject, and by the fixed factors of protocol (HH, EH and C) and time (a.m. and p.m.), together with the interactive terms. If the interactive term for protocol × time was significant, indicating that the different protocols had differing effects on the changes occurring over time, then the protocol term was divided to reflect separately the effects of hyperventilation and the effects of hypocapnia, and the ANOVA was then repeated (using type III sum of squares). Confidence intervals (95 %) for the effects of hyperventilation and hypocapnia were calculated from the second ANOVA. ANOVA was carried out using the SPSS statistical package (Chicago, IL, USA).

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Subjects

All of the twelve subjects completed the study and none reported discomfort from the hyperventilation or from the measurement of AHVR and AHCVR. Of the subjects, eleven provided data which were suitable for analysis. Subject 1057 produced very variable control responses between the three protocols - part of this may have been due to technical problems - and so the data from this subject were excluded from further analysis.

Passivity of ventilation and quality of gas control on the ventilator

Passivity of ventilation was judged from the record of airway pressure (Datta et al. 1991; Roberts et al. 1995). In all subjects, this was smooth and positive for the vast majority of the time, indicating an absence of spontaneous respiratory effort.

Figure 2 shows P_ET,CO2 recorded from each of the eleven subjects on the ventilator, sampled every 5 min. The upper plots show the range of normal values whilst the lower plots show the deviations in P_ET,CO2 from the control/target values over the 6 h period. In general a good quality of control over P_ET,CO2 was achieved.

		View larger version [in this window] [in a new window]
Figure 2. End-tidal PCO2 (upper panels) and deviation in end-tidal PCO2 from control/target values (lower panels) for each subject for the 6 h exposures

Blood samples

The three protocols did not differ significantly with respect to their effects on any of the measurements or derived variables relating to the venous blood samples. Mean (± S.D.) values were 61·3 (± 18·1) % for venous oxygen saturation, 51·0 (± 6·1) Torr for venous P_CO2, 25·7 (± 1·2) mequiv l^-1 for standard bicarbonate and 7·42 (± 0·02) for calculated arterial pH (based on the concomitant P_ET,CO2).

Spontaneous air-breathing P_ET,CO2 and isocapnic _E in euoxia and hyperoxia

Measurements of spontaneous air-breathing P_ET,CO2 obtained by use of a nasal catheter before and half an hour after the three protocols are shown in Table 1. Values are averages from the last 2 min of the 5 min measurement period. The three different protocols had significantly different effects (P < 0·001, ANOVA). There was a significant fall in P_ET,CO2 associated with prior hypocapnia of 2·3 Torr (95 % confidence interval (CI), 0·6-3·9 Torr; P < 0·05), but any effect of prior hyperventilation was not significant (fall of 0·7 Torr; 95 % CI, -0·2 to 1·6 Torr).

Table 1. End-tidal P_CO2 during air breathing and isocapnic ventilation in euoxia and hyperoxia

PET,CO2 (Torr)

E,100 (l min-1)

E,200 (l min-1)

Protocol HH

Protocol EH

Protocol C

Protocol HH

Protocol EH

Protocol C

Protocol HH

Protocol EH

Protocol C

Subject

Before

After

Before

After

Before

After

Before

After

Before

After

Before

After

Before

After

Before

After

Before

After

0981

34·9

33·7

35·4

36·1

35·9

36·9

16·2

16·6

20·3

15·2

16·4

16·5

18·0

19·8

24·2

15·6

22·5

18·8

1004

41·2

38·8

39·8

38·5

41·7

42·4

11·8

22·3

11·0

12·6

13·9

13·0

13·8

30·8

11·6

15·1

15·7

14·7

1023

41·1

41·0

43·2

41·9

43·0

44·5

18·8

22·7

13·5

15·6

16·5

12·5

19·6

27·7

14·0

19·3

20·4

13·7

1055

37·1

34·6

37·3

36·5

36·4

37·3

12·6

16·7

14·7

12·9

13·9

11·6

14·0

19·6

15·7

14·8

14·8

13·8

1066

40·4

37·6

39·3

39·0

38·1

37·4

12·3

13·1

12·9

14·0

12·2

9·4

13·1

16·6

13·6

15·9

11·6

9·3

1069

36·6

33·2

35·6

34·4

35·7

36·7

10·5

9·4

9·9

9·2

9·4

7·8

9·8

10·7

12·1

11·2

10·3

8·4

1076

41·0

36·7

33·3

37·8

36·4

39·0

12·1

12·8

9·0

10·1

8·3

9·1

11·9

14·6

9·5

9·2

9·0

10·1

1078

40·4

38·6

40·5

40·2

42·5

42·1

10·5

14·9

11·2

10·7

11·2

11·7

11·3

15·6

10·7

11·4

12·7

15·3

1079

37·6

35·1

37·0

37·8

37·4

39·2

9·4

16·5

10·1

9·1

11·7

8·7

7·5

17·8

12·3

11·3

13·3

10·4

1080

41·3

39·5

41·1

41·8

42·2

43·3

16·9

22·1

11·8

15·6

10·7

13·2

18·8

26·4

20·0

14·3

13·0

16·7

1081

38·7

40·8

39·7

42·7

39·7

42·6

16·4

21·0

13·7

12·5

12·1

18·8

17·6

23·3

14·7

11·2

13·4

19·4

Mean

39·1

37·2

38·4

38·8

39·0

40·1

13·4

17·1

12·6

12·5

12·4

12·0

14·1

20·3

14·4

13·6

14·2

13·7

s.e.

0·7

0·8

0·9

0·8

0·9

0·9

0·9

1·3

0·9

0·7

0·8

1·0

1·2

1·9

1·3

0·9

1·2

1·1

Values for _E under euoxic (P_ET,O2 = 100 Torr) and hyperoxic (P_ET,O2 = 200 Torr) conditions obtained as part of the process of measuring AHVR and AHCVR when P_ET,CO2 was elevated by 1-2 Torr are also shown in Table 1. There were significantly different effects between the protocols for both sets of data (P < 0·005). The effect of prior hypocapnia was to elevate _E by 3·7 l min^-1 (95 % CI, 1·4-6·1 l min^-1; P < 0·005) under euoxic conditions and by 7·0 l min^-1 (95 % CI, 3·8-10·1 l min^-1; P < 0·001) under hyperoxic conditions. No significant effects of prior hyperventilation were detected for either the euoxic or hypocapnic data.

Acute hypoxic ventilatory response (AHVR)

Figure 3 illustrates the ventilatory responses and end-tidal gas tensions for the hypoxic square waves imposed before and after each protocol. In this figure, the six hypoxic square waves for each assessment of AHVR for a given protocol were ensemble averaged for each subject, and these responses were then averaged across all subjects. This figure suggests that both P_ET,CO2 and P_ET,O2 were well controlled during these measurements. No changes were apparent in the sensitivity of the ventilatory response to hypoxia after any of the three protocols, although there appeared to be a vertical shift in the responses following protocol HH, consistent with the increase in euoxic and hyperoxic ventilation associated with prior hypocapnia.

		View larger version [in this window] [in a new window]
Figure 3. Ensemble averages of the 6 hypoxic square waves for all 11 subjects for end-tidal PCO2 (top panels), end-tidal PO2 (middle panels) and ventilation (bottom panels) Interrupted lines indicate the response before the 6 h exposure, continuous lines indicate the responses after the 6 h exposure.

An example of the fit of the respiratory model to the breath-by-breath data is given in Fig. 4. The figure shows that the model is capable of providing a good fit to the experimental data. Individual estimates and mean values for Gp and _c are given in Table 2. In line with the above observations, there was a significant increase in _c associated with prior hypocapnia of 6·1 l min^-1 (95 % CI, 3·3-8·9 l min^-1; P < 0·005). No significant effect of prior hyperventilation was detected for _c (fall of 0·4 l min^-1; 95 % CI, -2·8 to 3·6 l min^-1). There were no changes that differed significantly between the protocols for any of the parameters G_p, or T_d. The mean value (± S.D.) for was 10·2 (± 6·1) s and for T_d it was 4·3 (± 2·1) s.

		View larger version [in this window] [in a new window]
Figure 4. Example of fit of respiratory model (line) to breath-by-breath data () Data for subject 1004 at start of protocol HH.

Table 2. Estimated parameters for AHVR for hypoxic sensitivity (G_p) and residual ventilation in hyperoxia (_c)

Gp (l min-1)

c (l min-1)

Protocol HH

Protocol EH

Protocol C

Protocol HH

Protocol EH

Protocol C

Subject

Before

After

Before

After

Before

After

Before

After

Before

After

Before

After

0981

72·2

76·4

118·7

55·2

74·1

68·0

17·6

19·1

22·0

13·8

19·3

17·5

1004

97·0

158·1

60·0

76·1

101·2

80·6

12·7

23·6

8·7

10·2

12·4

11·8

1023

176·7

157·8

139·7

145·2

123·7

74·9

16·9

21·4

12·5

13·5

16·8

10·8

1055

30·1

26·7

35·0

42·1

35·4

40·4

10·8

19·3

12·8

11·8

13·5

10·1

1066

72·7

98·0

87·3

98·1

65·7

52·5

10·5

12·9

11·9

12·7

10·1

7·9

1069

87·7

54·5

50·9

46·3

67·3

56·9

8·6

8·1

10·1

8·5

8·4

5·7

1076

59·5

83·7

55·5

57·1

50·6

59·7

10·1

12·1

7·8

7·4

6·6

6·9

1078

63·8

66·9

75·3

47·4

60·5

78·8

9·2

10·8

8·8

9·6

11·4

10·2

1079

70·8

80·7

53·0

46·1

65·9

50·8

6·2

16·3

9·6

7·7

9·9

7·6

1080

124·4

123·0

147·4

133·7

102·1

108·1

17·2

22·6

14·2

11·0

10·3

13·0

1081

79·3

94·3

65·1

42·3

51·4

89·3

14·4

19·6

12·5

9·5

11·3

17·5

Mean

84·9

92·7

80·7

71·8

72·6

69·1

12·2

16·9

11·9

10·5

11·8

10·9

s.e.

11·6

12·2

11·5

11·3

7·9

5·9

1·2

1·6

1·2

0·7

1·1

1·2

Acute hypercapnic ventilatory responses (AHCVR)

The slopes and intercepts for the _E-P_ET,CO2 responses are given in Table 3, and the mean responses are illustrated in Fig. 5. The changes in both slope (P < 0·05) and intercept (P < 0·05) differed significantly between the protocols. There was a significant increase in slope of 0·4 l min^-1 Torr^-1 (95 % CI, 0·0-0·8 l min^-1 Tor^-1; P < 0·05) associated with prior hyperventilation, but no significant effect of prior hypocapnia was detected (rise of 0·1 l min^-1 Torr^-1; 95 % CI, -0·3 to 0·5 l min^-1 Torr^-1). For the intercept, there was no significant effect associated with prior hyperventilation (rise of 1·3 Torr; 95 % CI, -0·5 to 3·0 Torr), but there was a significant fall associated with prior hypocapnia (2·8 Torr; 95 % CI, 0·3-5·3 Torr; P < 0·05).

		View larger version [in this window] [in a new window]
Figure 5. VE-PET,CO2 responses averaged across all 11 subjects for protocol HH, protocol EH and protocol C Open symbols and dashed lines indicate responses before the 6 h exposures, filled symbols and continuous lines indicate the responses after the 6 h exposures.

Table 3. Slopes and intercepts for the _E-P_ET,CO2 responses

CO2 sensitivity (l min-1 Torr-1)

Intercept (Torr)

PET,CO2 at E = 15 l min-1 m-2 (Torr)

Protocol HH

Protocol EH

Protocol C

Protocol HH

Protocol EH

Protocol C

Protocol HH

Protocol EH

Protocol C

Subject

Before

After

Before

After

Before

After

Before

After

Before

After

Before

After

Before

After

Before

After

Before

After

0981

1·98

3·03

2·51

2·51

2·46

2·27

29·7

32·6

30·7

34·2

31·4

32·1

44·6

42·3

42·4

45·9

43·3

45·1

1004

2·94

4·10

2·73

2·92

3·01

2·47

36·8

30·1

35·0

37·3

39·1

36·8

47·4

42·2

47·9

46·2

47·8

49·3

1023

3·36

4·05

3·08

4·20

3·04

2·68

39·7

38·7

41·2

41·0

39·0

40·3

48·8

46·2

51·1

48·3

49·1

51·7

1055

1·34

2·67

1·52

2·20

1·78

1·53

28·1

32·0

29·3

33·1

30·4

29·6

50·6

43·3

49·2

46·8

47·4

49·4

1066

2·14

2·76

2·21

2·85

2·25

2·25

36·2

34·6

36·1

36·5

36·0

36·6

48·9

44·5

48·5

46·1

48·0

48·7

1069

1·57

1·63

1·42

1·56

1·48

1·28

32·1

31·8

30·0

30·9

31·5

31·9

46·4

45·6

45·8

45·4

46·6

49·4

1076

1·59

2·40

2·25

2·85

1·83

1·78

35·2

37·3

35·1

35·6

34·0

33·8

52·9

49·0

47·6

45·4

49·3

49·6

1078

1·58

1·44

1·75

1·46

1·48

1·47

35·6

31·6

38·2

35·6

36·7

34·7

51·4

49·0

52·4

52·6

53·5

51·6

1079

1·87

1·84

1·67

1·81

1·57

1·81

36·6

30·6

32·8

33·8

31·6

34·4

49·0

43·3

46·8

46·6

46·5

47·3

1080

2·68

1·90

2·30

2·28

2·55

2·40

37·8

35·3

37·6

36·5

38·4

37·9

46·9

44·4

46·8

49·2

49·8

48·1

1081

1·70

1·77

1·08

1·48

1·29

1·98

36·3

33·7

33·0

39·3

37·2

38·1

53·7

50·5

60·5

59·4

60·2

53·1

Mean

2·07

2·51

2·05

2·37

2·07

1·99

34·9

33·5

34·4

35·8

35·0

35·1

49·1

45·5

49·0

48·4

49·2

49·4

s.e.

0·20

0·28

0·18

0·25

0·19

0·14

1·1

0·8

1·1

0·9

1·0

1·0

0·9

0·9

1·4

1·3

1·3

0·7

One previous study has reported values for the intercept of the _E-P_ET,CO2 response as a value for P_ET,CO2 at _E = 15 l min^-1 m^-2 body surface area rather than as a value at _E = 0 l min^-1 (Eger et al. 1968). For the purposes of comparison, values of P_ET,CO2 at _E = 15 l min^-1 m^-2 body surface area are also given in Table 3. Differences between the protocols were significant (P < 0·005). There was a significant decrease in P_ET,CO2 with prior hypocapnia of 3·0 Torr (95 % CI, 1·6-4·5 Torr; P < 0·005), but there was no significant effect of prior hyperventilation (fall of 0·8 Torr; 95 % CI, -1·7 to 3·3 Torr).

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

Our results demonstrate that half an hour after a 6 h period involving both hypocapnia and hyperventilation, there is a reduction in P_ET,CO2 associated with spontaneous ventilation while breathing air, and an elevation in _E in euoxia and hyperoxia when P_ET,CO2 is held 1-2 Torr above the original control values. These findings are broadly consistent with those that might have been predicted from the literature (Brown et al. 1948, 1950; Eger et al. 1968; Dempsey et al. 1975). However, the present study also extends previous findings in the following ways. (1) The reduction in P_ET,CO2 and elevation in _E described above were found to be due to the presence of prior hypocapnia rather than prior hyperventilation. (2) The presence of prior hyperventilation resulted in a modest increase in CO₂ sensitivity that was not affected by the presence or absence of prior hypocapnia. (3) Neither hypocapnic hyperventilation nor eucapnic hyperventilation affected the ventilatory sensitivity to hypoxia.

Assessment of shifts in the relationship between _E and P_ET,CO2

The _E response to CO₂ is normally considered to be a linear function of P_ET,CO2 at levels of CO₂ above eucapnia. The response can thus be described by a slope and a constant that describes the position of the line. Changes in slope are unambiguous. However, in all cases where there is a change of slope, descriptions of any simultaneous change in position depend upon where the shift is measured. For example, the same response line might be described as shifted to the right (or shifted downwards) below the point at which it crosses the original response, not shifted at all at the crossing point, and shifted to the left (or shifted upwards) above the crossing point. In this study, we report several measures of shift including: (1) the calculated change in P_ET,CO2 at a _E of zero (value B of the Lloyd equation (Lloyd et al. 1958)), (2) the change in P_ET,CO2 under air-breathing conditions, (3) the change in _E under conditions of very mild hypercapnia (1-2 Torr above initial control values), and (4) the change in P_ET,CO2 at _E = 15 l min^-1 m^-2 (for comparison with the study by Eger et al. (1968) which used this measure).

In a physiological sense, the measure of shift that would be most useful is the one that is unaffected by physiological changes that affect purely the ventilatory sensitivity to CO₂. Thus it would seem that measures of shift in the hypercapnic region are inappropriate. However, it is not clear whether measures of shift at around the spontaneous level of ventilation or measures of shift at some theoretical lower level of ventilation (e.g. _E = 0) best meet this purpose. Fortunately, in the present study, the same conclusions may be drawn from all the measures of shift that were employed; namely, that there was a significant leftwards shift associated with prior hypocapnia, but not with prior hyperventilation (in the absence of hypocapnia).

Comparisons with previous studies of shifts in the _E-P_ET,CO2 relationship following hyperventilation

The results for hypocapnic hyperventilation in the present study are consistent with previous studies. Brown et al. (1948) reported that, in two out of three volunteers, hyperoxic _E was higher following 24 h of passive mechanical ventilation. This result was extended to two further subjects in a subsequent study (Brown et al. 1950). Similar results are apparent from a study by Eger et al. (1968) and Dempsey et al. (1975).

Previous results concerning the effect of a period of eucapnic hyperventilation on the position of the _E-P_ET,CO2 response are scarce and far from conclusive. Smith et al. (1962), in a study of two patients with neuromuscular weakness who had been ventilated artificially for clinical reasons, provided evidence that hyperventilation without concomitant hypocapnia can increase subsequent spontaneous _E. In their study, the ventilator settings and the size of an external dead space were manipulated together to vary the level of passive hyperventilation whilst maintaining P_ET,CO2 constant. Subsequent spontaneous _E off the ventilator was related to the prior level of _E on the ventilator. In some contrast to the study by Smith et al. (1962), Eger et al. (1968) found no residual effects in two subjects of a period of 8 h of eucapnic hyperventilation at 30 l min^-1. In both of these studies, there were only a couple of subjects and those of Smith et al. (1962) were clearly abnormal. Our study of eleven subjects now provides clear evidence that the predominant effect of a prior period of hyperventilation on the position of the _E-P_ET,CO2 response arises from the associated hypocapnia and not from the hyperventilation itself.

Mechanism underlying the shift in _E response to CO₂

The finding that the shift is related to the hypocapnia and not the hyperventilation suggests that it is an acid-base effect, and the finding that the shift is present under hyperoxic conditions suggests that it is likely to be related to the stimulus at the central chemoreceptors. One possible mechanism is that the 6 h period of hypocapnia is sufficiently long to induce a fall not only in the bicarbonate concentration of the interstitial fluid of the brain, but also in the bicarbonate concentration of the cerebrospinal fluid (Fencl, 1986). Upon restoration of P_ET,CO2 to near normal levels, the restoration of the bicarbonate concentration in the cerebrospinal fluid will be fairly slow as it depends upon diffusional exchange with the brain interstitial space and formation of new cerebrospinal fluid at the choroid plexus (Fencl, 1986). As the central chemoreceptors are superficial structures, the stimulus at the central chemoreceptors is influenced, in part, by cerebrospinal fluid composition (Pappenheimer et al. 1965). Thus, other things being equal, the bicarbonate concentration will be lower than in the control condition, and therefore, provided the P_CO2 is the same, the pH will be lower, providing a greater respiratory stimulus.

It is also possible that a degree of bicarbonate depletion could arise from the fact that hypocapnia causes an increase in lactate production within the brain (Fencl, 1986).

Changes in the slope of the _E response to CO₂ following a period of hyperventilation

There are relatively few observations with which to compare our finding of an increased _E-P_ET,CO2 response slope, and those that do exist do not appear to be consistent. On the one hand, our study is consistent with two previous studies carried out on three subjects (Brown et al. 1948) and two subjects (Brown et al. 1950), respectively. On the other hand, the study by Eger et al. (1968) found no effect of hyperventilation on the _E-P_ET,CO2 response slope in four subjects. The present study now provides data on a larger set of subjects, and thus a firmer statistical basis from which to work. In addition, the fact that the increase in slope was not affected by whether or not there was prior hypocapnia suggests that it results directly from the hyperventilation and not from the hypocapnia.

Beyond the observation that the increase in slope is due to the hyperventilation per se, and not the hypocapnia, it is difficult to conclude much about the mechanism. One possibility is that the change arises from increased afferent feedback during the period of hyperventilation. However, even if this possibility were correct, it remains unclear whether it would be feedback via the vagus or feedback from the muscle spindles of the chest wall that is important in generating the response.

Ventilatory sensitivity to hypoxia

We were unable to detect any effect of sustained hyperventilation/hypocapnia on AHVR. We are unaware of any studies with which this result may be compared directly. However, studies of the effect of sustained hypoxia on AHVR in both humans (8 h; Howard & Robbins, 1995) and goats (4 h; Engwall & Bisgard, 1990) have found no difference between hypoxic exposures that were hypocapnic compared with those that were eucapnic. This would appear to be consistent with the results from the present study, although in the studies involving hypoxia the degree of hypocapnia would have been considerably less.

Implications of the study for understanding ventilatory acclimatization to hypoxia

A major motivation for the present study was to increase our understanding of the processes involved in early VAH by studying the hyperventilation and the hypocapnia that normally accompany VAH in the absence of the primary stimulus of hypoxia.

One feature of VAH is that there is a leftwards shift of the _E-P_ET,CO2 response (Kellogg et al. 1957; Chiodi, 1957; Michel & Milledge, 1963; White et al. 1987). The finding of a shift in the _E-P_ET,CO2 response relationship with hypocapnic hyperventilation, but not eucapnic hyperventilation, clearly demonstrates the potential of hypocapnia to cause this effect. However, it should be appreciated that the degree of hypocapnia in the present study was substantially greater than would normally arise through acute exposure to hypoxia. This may explain why the effect of hypocapnia in causing a left shift appeared rapidly in the present study, but was not present after 8 h (Fatemian & Robbins, 1998) or even 48 h (Tansley et al. 1998) of hypoxia where the degree of hypocapnia induced was mild.

A second feature of VAH is the increase in slope of the _E-P_ET,CO2 response relationship (Chiodi, 1957; Michel & Milledge, 1963; White et al. 1987). Our study demonstrates that hyperventilation can induce an increase in _E sensitivity to CO₂. However, the increase was modest for a level of hyperventilation that greatly exceeds that normally found during the acclimatization to modest hypoxia. Thus it seems unlikely that it is the hyperventilation that generates the entire increase in slope of the _E-P_ET,CO2 response relationship in VAH, although one caveat is that the hyperventilation of the present study was passive, whereas the hyperventilation of VAH is active.

It has been suggested that a leftwards shift generated by an acid-base effect at the central chemoreceptors would be expected to increase the slope of the _E-P_ET,CO2 response relationship because of the logarithmic relationship between P_CO2 and pH (Severinghaus et al. 1963, 1966; Pappenheimer et al. 1965; Ward et al. 1995), although other studies have suggested that the increase in slope was too great to be accounted for by this factor alone (Michel & Milledge, 1963; Forster et al. 1971; Dempsey et al. 1975; Tansley et al. 1998; Fatemian & Robbins, 1998). The present findings support the idea that the leftwards shift is dissociated from the increase in slope, as there were no differences in the change in slope between the eucapnic and hypocapnic protocols. In this respect, it is also worth noting that recent studies involving eucapnic and poikilocapnic exposures to hypoxia may generate increases in CO₂ slope without causing any shift in the intercept of the _E-P_ET,CO2 relationship with the axis for _E = 0 (Fatemian & Robbins, 1998; Tansley et al. 1998).

The results from the present study suggest that there is no effect of hypocapnia and hyperventilation on AHVR early on in VAH. However, hypocapnia that is of considerably longer duration will engender some renal compensation for the alkalosis, and it would seem likely that this would alter AHVR later on in VAH by a direct effect at the carotid bodies.

	REFERENCES

Top Abstract Introduction Methods Results Discussion References

Brown, E. B. Jr, Campbell, G. S., Johnson, M. N., Hemingway, A. & Visscher, M. B. (1948). Changes in response to inhalation of CO2 before and after 24 h of hyperventilation in man. Journal of Applied Physiology 1, 333-338.
Brown, E. B. Jr, Hemingway, A. & Visscher, M. B. (1950). Arterial blood pH and PCO2 changes in response to CO2 inhalation after 24 h of passive hyperventilation. Journal of Applied Physiology 2, 544-552.
Chiodi, H. (1957). Respiratory adaptations to chronic high altitude hypoxia. Journal of Applied Physiology 10, 81-87.
Clement, I. D. & Robbins, P. A. (1993). Dynamics of the ventilatory response to hypoxia in humans. Respiration Physiology 92, 253-275	[Medline]
Datta, A. K., Shea, S. A., Horner, R. L. & Guz, A. (1991). The influence of induced hypocapnia and sleep on the endogenous respiratory rhythm in humans. The Journal of Physiology 440, 17-33	[Abstract]
Dempsey, J. A., Forster, H. V., Gledhill, N. & DoPico, G. A. (1975). Effects of moderate hypoxemia and hypocapnia on CSF [H+] and ventilation in man. Journal of Applied Physiology 38, 665-674	[Medline]
Eger, E. I. I., Kellogg, R. H., Mines, A. H., Lima-Ostos, M., Morrill, C. G. & Kent, D. W. (1968). Influence of CO2 on ventilatory acclimatization to altitude. Journal of Applied Physiology 24, 607-615	[Medline]
Engwall, M. J. A. & Bisgard, G. E. (1990). Ventilatory responses to chemoreceptor stimulation after hypoxic acclimatization in awake goats. Journal of Applied Physiology 69, 1236-1243	[Medline]
Fatemian, M. & Robbins, P. A. (1998). The human ventilatory response to CO2 after 8 h of isocapnic/poikilocapnic hypoxia. Journal of Applied Physiology 85, 1922-1928	[Abstract/Full Text]
Fencl, V. (1986). Acid-base balance in cerebral fluids. In Handbook of Physiology, section 3, The Respiratory System, vol. 2, Control of Breathing, ed. Fishman, A. P., pp. 115-140. American Physiology Society, Bethesda, MD, USA.
Forster, H. V., Dempsey, J. A., Birnbaum, M. L., Reddan, W. G., Thoden, J., Grover, R. F. & Rankin, J. (1971). Effect of chronic exposure to hypoxia on ventilatory response to CO2 and hypoxia. Journal of Applied Physiology 31, 586-592	[Medline]
Howard, L. S. G. E. & Robbins, P. A. (1995). Alterations in respiratory control during eight hours of isocapnic and poikilocapnic hypoxia in humans. Journal of Applied Physiology 78, 1098-1107	[Medline]
Howson, M. G., Khamnei, S., McIntyre, M. E., O'Connor, D. F. & Robbins, P. A. (1987). A rapid computer-controlled binary gas-mixing system for studies in respiratory control. The Journal of Physiology 394, 7P.
Howson, M. G., Khamnei, S., O'Connor, D. F. & Robbins, P. A. (1986). The properties of a turbine device for measuring respiratory volumes in man. The Journal of Physiology 382, 12P.
Kellogg, R. H., Pace, N., Archibald, E. R. & Vaughan, B. E. (1957). Respiratory response to inspired CO2 during acclimatization to an altitude of 12,470 feet. Journal of Applied Physiology 11, 65-71.
Lloyd, B. B., Jukes, M. G. M. & Cunningham, D. J. C. (1958). The relation between alveolar oxygen pressure and the respiratory response to carbon dioxide in man. Quarterly Journal of Experimental Physiology 43, 214-227.
Michel, C. C. & Milledge, J. S. (1963). Respiratory regulation in man during acclimatisation to high altitude. The Journal of Physiology 168, 631-643.
Pappenheimer, J. R., Fencl, V., Heisey, S. R. & Held, D. (1965). Role of cerebral fluids in control of respiration as studied in unanesthetized goats. American Journal of Physiology 208, 436-450.
Robbins, P. A., Swanson, G. D. & Howson, M. G. (1982). A prediction-correction scheme for forcing alveolar gases along certain time courses. Journal of Applied Physiology 52, 1353-1357.	[Medline]
Roberts, C. A., Corfield, D. R., Murphy, K., Calder, N. A., Hanson, M. A., Adams, L. & Guz, A. (1995). Modulation by 'central' PCO2 of the response to carotid body stimulation in man. Respiratory Physiology 102, 149-161.
Sato, M., Severinghaus, J. W., Powell, F. L., Xu, F.-D. & Spellman, M. J. Jr (1992). Augmented hypoxic ventilatory response in men at altitude. Journal of Applied Physiology 73, 101-107	[Medline]
Severinghaus, J. W. (1979). Simple, accurate equations for human blood O2 dissociation computations. Journal of Applied Physiology 46, 599-602.	[Medline]
Severinghaus, J. W., Bainton, C. R. & Carcelen, A. (1966). Respiratory insensitivity to hypoxia in chronically hypoxic man. Respiration Physiology 1, 308-334	[Medline]
Severinghaus, J. W., Mitchell, R. A., Richardson, B. W. & Singer, M. M. (1963). Respiratory control at high altitude suggesting active transport regulation of CSF pH. Journal of Applied Physiology 18, 1155-1166.
Smith, A. C., Spalding, J. M. K. & Watson, W. E. (1962). Ventilation volume as a stimulus to spontaneous ventilation after prolonged artificial ventilation. The Journal of Physiology 160, 22-31.
Tansley, J. G., Fatemian, M., Howard, L. S. G. E., Poulin, M. J. & Robbins, P. A. (1998). Changes in respiratory control during and after 48 h of isocapnic and poikilocapnic hypoxia in humans. Journal of Applied Physiology 85, 2125-2134	[Abstract/Full Text]
Ward, M. P., Milledge, J. S. & West, J. B. (1995). The ventilatory response to hypoxia and carbon dioxide. In High Altitude Medicine and Physiology, ed. Ward, M. P., Milledge, J. S. & West, J. B., pp. 69-97. Chapman and Hall Medical, London.
White, D. P., Gleeson, K., Pickett, C. K., Rannels, A. M., Cymerman, A. & Weil, J. V. (1987). Altitude acclimatization: influence on periodic breathing and chemoresponsiveness during sleep. Journal of Applied Physiology 63, 401-412	[Medline]

Acknowledgements

This study was supported by The Wellcome Trust. X. Ren holds an Overseas Research Students Award and is supported by a K. C. Wong Scholarship. We wish to thank Professor A. Guz for help with the procedures for hyperventilation. We are grateful to Mr D. O'Connor for his skilled technical assistance.

Corresponding author

P. A. Robbins: University Laboratory of Physiology, Parks Road, Oxford OX1 3PT, UK.

Email: peter.robbins{at}physiol.ox.ac.uk

This article has been cited by other articles:

				J. Duffin Role of acid-base balance in the chemoreflex control of breathing J Appl Physiol, December 1, 2005; 99(6): 2255 - 2265. [Abstract] [Full Text] [PDF]

				S. Donoghue, M. Fatemian, G. M. Balanos, A. Crosby, C. Liu, D. O'Connor, N. P. Talbot, and P. A. Robbins Ventilatory acclimatization in response to very small changes in PO2 in humans J Appl Physiol, May 1, 2005; 98(5): 1587 - 1591. [Abstract] [Full Text] [PDF]

				R. W. Putnam, J. A. Filosa, and N. A. Ritucci Cellular mechanisms involved in CO2 and acid signaling in chemosensitive neurons Am J Physiol Cell Physiol, December 1, 2004; 287(6): C1493 - C1526. [Abstract] [Full Text] [PDF]

				J. H. Mateika, C. Mendello, D. Obeid, and M. S. Badr Peripheral chemoreflex responsiveness is increased at elevated levels of carbon dioxide after episodic hypoxia in awake humans J Appl Physiol, March 1, 2004; 96(3): 1197 - 1205. [Abstract] [Full Text] [PDF]

				S. Zhang and P. A. Robbins Methodological and physiological variability within the ventilatory response to hypoxia in humans J Appl Physiol, May 1, 2000; 88(5): 1924 - 1932. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Ren, X. Articles by Robbins, P. A. Search for Related Content PubMed PubMed Citation Articles by Ren, X. Articles by Robbins, P. A.