Respiration Physiology (1978) 35, 79-87 © Elsevier/North-Holland Biomedical Press

THE CONTRIBUTION OF THE REFLEX HYPOXIC DRIVE TO THE HYPERPNOEA OF EXERCISE

R.A. STOCKLEY Department o] Medicine, University o[Birmingham, Queen Elizabeth Hospital, Edgbaston, Birmingham B15 2TH, England

Abstract. Oxygen breath tests have been applied to six normal subjects at rest and during steady state exercise on a bicycle ergometer (200, 400 and 600 kpm) to estimate the contribution of the reflex hypoxic drive to total ventilation. The subjects were changed without their knowledge from air to 100°o oxygen for 4-5 breaths, the fall in ventilation was recorded and expressed as a percentage of total ventilation (hypoxic drive). The average reflex hypoxic drive was 16.2"L (SE + 2.6; n = 35) this remained unaltered during all levels of steady state exercise. The m a x i m u m fall in ventilation occurred earlier in exercise (P < 0.0005). At rest the m a x i m u m fall occurred on average 31.2 s (SE +_ 2.6) from the start of the first breath of oxygen. During the first level of steady state exercise the m a x i m u m fall occurred 19.4 s (SE +_0.5) from the start of the first breath of oxygen. It is concluded that the contribution of the reflex hypoxic drive to the total ventilation is unaltered by exercise although a substantial part of the hyperpnoea can be accounted for by the presence of this drive, Chemoreceptors Exercise Oxygen breath tests

Respiratory drive Ventilation

The cause of the increase in ventilation during exercise has been the subject of much investigation. In particular the role of the peripheral chemoreceptors seems far from clear. Lugliani et al. (1971) found a normal ventilatory response to exercise in a group of patients who had undergone bilateral carotid body resection. They suggested that the carotid body provided little contribution to the hyperpnoea of exercise. However, Weil et al. (1972) demonstrated that the ventilatory response to increasing Accepted.[br publication 22 May 1978 79

80

R.A. S T O C K L E Y

hypoxia was enhanced by exercise. Theoretical consideration of the data suggested that, at a normal oxygen tension, the peripheral chemoreceptors provided a 37"0 hypoxic drive at rest and this increased to 59°o in exercise. The only direct means of assessing the hypoxic drive is by the oxygen breath test. Dejours el al. (1958) found an 8-10°,o hypoxic drive at rest in 3 subjects using a single breath oxygen test. When the same subjects were studied in exercise a 5-61~; hypoxic drive was found. Recently Stockley (1977) has shown that the single breath oxygen test used by Dejours probably resulted in an underestimation of the hypoxic drive at rest. The present study was undertaken to assess the reflex hypoxic drive at rest and at 3 levels of steady state exercise using the modified technique of Stockley (1977) in order to measure the contribution of this drive to the hyperpnoea of exercise.

Methods Six normal male subjects between 22 and 28 years of age were studied. None had any clinical evidence of cardiopulmonary disease and all gave informed consent. The average body surface area of the subjects was 1.82 m 2 (SD + 0.08). None of the subjects was starved but all were studied at least 2 hours after a meal.

O X Y G E N B R E A T H TESTS A T REST

The method has been described in detail elsewhere (Stockley, 1977). Briefly, the subjects were seated in an armchair and breathed through a low resistance solenoid controlled valve. Tidal volume was obtained from the integrated output of a Fleisch pneumotachograph connected to a bag in box system. When ventilation was judged by eye to be constant in tidal volume and frequency, it was recorded for 30 s and then the inspired gas was changed suddenly from air to oxygen for 20 s (between 4 and 6 breaths) without the subjects' knowledge. Ventilation was recorded for a further 40 s from the start of the first breath of oxygen. Six to eight oxygen breath tests were performed whilst the subjects listened to background music with headphones and interference was kept to a minimum. Between oxygen tests the subjects breathed room air for 5-10 minutes.

O X Y G E N B R E A T H TESTS IN EXERC1SE

The subjects were seated on a bicycle ergometer and studied at 3 levels of steady state exercise (200, 400 and 600 kpm) with 30 minutes rest between. They pedalled for 7 minutes before any oxygen tests were performed. Ventilation was again observed until it was judged by eye to be constant in tidal volume and frequency. It was

REFLEX HYPOXICDRIVE-

EXERCISE

81

recorded for 30 s before the subjects were changed from air to oxygen and recording was continued for 40 s from the start of the first breath o f oxygen. For each oxygen test the subjects received the same number of whole breaths of oxygen as during the resting oxygen tests (4~6 breaths). Between 6 and 8 oxygen tests were performed during each level of exercise. The subjects continued to exercise and breathe air through the valve between oxygen tests until the mixed expired gas composition returned to that observed before the first oxygen test for each level of exercise (1 2 min).

OXYGEN UPTAKE AND HEART RATE

Before the oxygen breath tests both at rest and during exercise the subjects breathed air through the valve for 7 minutes. The ventilation was recorded for the last minute and the expired gas was passed through a 6.5-1itre mixing chamber. Oxygen concentration was measured by a paramagnetic analyser (Servomex) calibrated with air and nitrogen before each experiment and carbon dioxide was measured by an infrared analyser (Godart Capnograph) calibrated with carbon dioxide free air and a known carbon dioxide mixture. The oxygen uptake was then calculated, corrected to STPD and also for body surface area. Heart rate was recorded at the same time using a continuous ECG monitor.

ANALYSIS OF VENTILATION TRACES

Each oxygen breath test was analysed in the same way. The control ventilation was obtained from the tidal volume and frequency of all breaths in the 30 s period prior to oxygen breathing. The resting traces were then analysed as a 2 or 3 breath advancing mean. The exact number of breaths depended on the frequency of breathing but occupied a time as close to 10 s as possible. The exercise traces for each subject were also analysed as a 2 or 3 breath advancing mean, the number of breaths being the same as that used for the resting oxygen tests. The lowest value for any 2 or 3 consecutive breaths in the control period was determined as well as the lowest value breathing oxygen timed at its midpoint from the start of oxygen breathing. Corresponding values for the 6 8 oxygen tests of each part of the experiment for an individual were then averaged. The significance of any fall in ventilation breathing oxygen was determined by comparing the lowest values in the control period with those breathing oxygen using a paired 't' test.

82

R.A. STOCKLEY

Results RESPONSE TO EXERCISE All subjects showed a progressive rise in ventilation, with increasingly

strenuous

exercise. The average

h e a r t r a t e and o x y g e n u p t a k e r e s u l t s f o r all six s u b j e c t s a r e

s h o w n in t a b l e 1.

REPRESENTATIVE E X P E R I M E N T

Figure 1 summarises the ventilation results of 6 oxygen tests at rest and in the second level of exercise in one of the subjects. The lower part of the figure shows the results at rest (oxygen uptake = 136 ml/min/m2). The control ventilation was 6.68 l/rain (SE +_ 0.24). When the oxygen tests were analysed as a 2 breath advancing mean, ventilation fell to 5.41 1/min (SE _+ 0.22) and this occurred on average 30.4 s (SE + 3.4) after the start of oxygen breathing. This was 19°% below the control level and was significantly different (P < 0.025) from the variability of ventilation breathing air (9~'%). The upper part of fig. 1 shows the result of 6 oxygen tests also analysed as a two breath advancing mean in the second level of exercise (oxygen uptake = 30

25

I!IERc,sE2 I ,18, 20 c E

L_ L

-20

i

-10

I

i REST 119%1

L

i

J

i

1o

20

30

40

o. TIME Is}

Fig. 1.1. Representative experiment. The vertical axes are the inspired ventilations (Vl) with different scales and the horizontal axis is the time in seconds. The vertical solid line indicates zero time at which oxygen breathing commenced, the horizontal dotted lines are the control values for VI breathing air and the percent figures indicate the hypoxic drive. The lower halt" of the figure is the result of 6 oxygen tests at rest and the upper half is the result of 6 oxygen tests in the same subject in exercise. The solid circles indicate the midpoint of the maximum fall in "VI.The bar lines are +_ 1 SE.

REFLEX HYPOXIC D R I V E

83

EXERCISE

441 ml/min/m-~). The control ventilation was 26.67 1/min (SE +_0.24). The variability of ventilation in the control period was 8°/,o and following the breathing of oxygen there was a significant fall in ventilation (P < 0.005), 18','~; below the control level to 21.86 l/rain (SE _+ 1.11). The maximum fall occurred, 21.2 s (SE + 2.1), after the start of oxygen breathing and was earlier than at rest (P < 0.05).

HYPOXIC DRIVE

All the subjects had a significant fall in ventilation following the breathing of oxygen. The hypoxic drive varied from 5-23 o at rest and the average was 16.2 o, jJo (SE _+ 2.6). However the hypoxic drive in the group as a whole remained unaltered by exercise (table 1) ranging from 8-231~'Jo in the most strenuous exercise. The variability of control ventilation was also unaltered by exercise ranging from 3-10'~i at rest / . m the most severe exercise (average = R 6 ° (average = 7 SE + 1.1) and 5 11 o.... SE + 0.9).

TABLE 1 Results of oxygen tests 9o~ (ml/min/m 2)

Heart rate (beats/min)

VI (l/min)

Midpoint of fall in "~l (s)

Hypoxic drive ("i,)

Rest (n = 35)

123 (4.3)

68.0 (2.5)

6.43 (0.30)

31.2 (0.9)

16.2 (2.6)

Exercise l (n = 35)

297 (11.5)

84.0 (5.0)

15.91 (0.64)

19.4 (0.5)

17.0 (2.1)

Exercise 2 (n = 35)

462 (13.7)

99.3 (6.9)

23.29 (0.82)

19.4 (1.3)

16.0 (1.8)

Exercise 3 (n =41)

642 (29.1)

120.0 (5.2)

32.50 (1.97)

17.3 (0.7)

15.5 (2.3)

The average results from the oxygen tests in all subjects are given for each phase of the experiments. Heart rate, ~/o~ (corrected to STPD), inspired control ventilation (~h corrected to BTPS), hypoxic drive and midpoint of maximum fall in ventilation after the start of oxygen breathing are shown. Figures in parentheses are + 1 SE. (n = number of oxygen tests analysed).

THE RISE OF Pao:

No direct measurments were made of the rise of Pao:. However, the oxygen analyser continually monitored mixed expired gas composition. During the oxygen tests at

84

R.A. STOCKLEY

30

REST--*1

+X~ EST

\

TPME (s)

~,gp 1 ____

20

P ~ 0"0005

1--~2

NS

1---'3

P < 0"0125

2 +.._.._

3

10 t

@

250 500 ~'o 2 (ml/min/m21

750

Fig. 2. Timing of maximum fall in ventilation. The vertical axis is the time in seconds from the start of oxygen breathing and the horizontal axis is the oxygen uptake corrected to sIPL~. Each point is the average result of all the oxygen tests performed in each phase of the experiment lk~r the six subjects (table 1). The bar lines are _+ 1 SE and the significance of differences are shown.

rest this rose to values greater than 4nv,,o in all subjects and in exercise it exceeded 80%. This suggested a rise of Pao, to about 40 kPa (300 torr) at rest and at least 80 kPa (600 torr) in exercise in subjects with normal lungs,

T I M I N G IN THE FALL IN V E N T I L A T I O N

The maximum fall in ventilation at rest occurred on average 31.2 s (SE + 0.9) after the start of oxygen breathing (table 1). The fall occurred progressively earlier in exercise. The results are summarised in fig. 2 together with the significance of the changes between each successive part of the experiment.

Discussion

The results of the oxygen breath tests at rest were similar to the previous findings of Stockley (1977). Meaningful and repeatable results can be obtained from as few as five successful oxygen tests at rest using the present method, thus making it possible to study the effect of a changing situation upon hypoxic drive (Stockley, 1977). Estimation of the hypoxic drive at rest is largely unaffected by secondary factors (Dejours, 1962). However, there are several factors that could affect estimation of hypoxic drive in exercise. Firstly, a change in arterial Pox with exercise would in itself alter the bypoxic

REFLEX HYPOX1C DRIVE

EXERCISE

85

drive. Patients with heart and lung disease often have such a change during exercise and for this reason only subjects without cardiopulmonary disease were studied. Dejours (1964) had shown that there is no appreciable change in Pao, when normal subjects exercise at comparable levels of work to those studied here. Secondly, it is known that anaerobic respiration occurs in severe exercise with increasing production of lactic acid. This in itself provides a substantial respiratory drive and accounts for much of the lower ventilation seen when normal subjects breathe oxygen in strenuous exercise (Bannister and Cunningham, 1954). For this reason the subjects were studied at levels of exercise below the anaerobic threshold, when changes in blood lactate are negligible (Dejours, 1964). The effect of carbon dioxide is more difficult to assess. The arterial Pco: rises by up to 0.5 kPa (3.25 torr) in mild exercise (Dejours, 1964). However, Cunningham (1974) emphasised that the actual level of Pco~ is unimportant in transient tests that lead to complete withdrawal of a chemoreceptor stimulus, such as the oxygen breath test. Nevertheless a rise of Paco~ resulting from the fall in ventilation during an oxygen breath test could stimulate the chemoreceptors and result in an underestimation of the reflex hypoxic drive. Stockley (1977) argued that estimation of the hypoxic drive at rest was largely unaffected by a change of carbon dioxide tension since any such change would not affect ventilation for at least 30 s from the start of oxygen breathing. This is largely because of a 10-s delay between the start of oxygen breathing and the fall in ventilation, together with a 20-s delay between any change in CO, and the response of the central chemoreceptors (the peripheral receptors being relatively insensitive to CO2 changes in the presence of hyperoxia). In moderate exercise Ventilation does not change for at least 4 s from the start of oxygen breathing (Dejours, 1962). No studies are available on the effect of exercise on the lung to central chemoreceptor time. However, even if alveolar CO, tension rose as soon as ventilation started to fall, the response time would have to be reduced by 50~;i to have much effect on the fall in ventilation. It therefore seems likely that the majority of the fall in ventilation during oxygen breathing in exercise is unaffected by a change in Paco ~ although this possibility cannot be excluded. The method of analysis of ventilation traces selects the lowest value breathing oxygen to assess the fall in ventilation. This value is partially dependent on the variability of ventilation and results cannot be compared between situations which affect the overall variability of control ventilation. In the present study, exercise had no effect on the variability of ventilation in any individual subject or the group as a whole. The rise in Po, breathing oxygen is possibly more substantial in exercise than at rest because, although the same number of whole breaths of oxygen are given, the tidal volume is increased. However, breathing oxygen for 20 s at rest raised the Po~ in excess of 30 kPa (225 torr) in subjects with normal lungs (Stockley, 1977), and this is sufficient to exceed the chemoreceptor threshold suggested by Dejours (1962) of 22.7 kPa (170 torr). However, the chemoreceptor threshold may be raised in exercise and therefore it is possible that breathing oxygen in the studies reported

86

R.A. STOCKLEY

here may have failed to exceed it. The ventilatory relationship to Po, at rest and in exercise is hyperbolic (Weft et al., 1972) and the Pao: probably rises in excess of 80 kPa (600 torr) during the exercise oxygen tests. It is therefore likely that, even if the chemoreceptor threshold is not exceeded in the present studies, the majority of the hypoxic drive will have been abolished. Therefore it seems likely that the results obtained here provide a good estimate of the reflex hypoxic drive in exercise. It is possible that secondary factors may have had more effect upon the fall in ventilation during exercise than at rest. Perhaps further information could be obtained using the more complicated method of Downes and Lambertsen (1966) in which any change of CO, tension can be prevented and a slightly greater rise of Po, can be achieved by driving ventilation with inhaled CO,. The results of the present study suggest that no great increase in hypoxic drive occurs in exercise. This is in disagreement with the theoretical consideration of Weil et al. (1972) and confirms the reservation held by these authors. Dejours et al. (1958) found no increase in hypoxic drive during exercise using a single breath oxygen test. However, Stockley (1977) found a much higher hypoxic drive at rest using a prolonged oxygen test and suggested this was mainly due to the greater rise in Pa o . The results of the oxygen tests reported here at rest confirms the findings of Stockley (1977). The hypoxic drive shown by the oxygen tests in light to moderate exercise is higher than that found with a single breath oxygen test by Dejours et al. (1958) and would also suggest that a greater rise of Pao: is necessary to obtain a good estimate of the hypoxic drive. It is concluded that the reflex hypoxic drive does not increase with exercise although it does provide a stimulus proportional to the increase in total ventilation. It is possible that secondary factors could have led to a relative underestimation of the drive in exercise when compared with that at rest and this warrants further investigation using more complicated techniques.

Acknowledgements I would like to thank Miss J. Blackhall for technical assistance, Professor J. M. Bishop and D r K. D. Lee for encouragement and comments on the manuscript and Mrs Jean Bonner for her typing.

References Bannister, R.G, and D.J.C. Cunningham (1954). The effects on thc respiration and performance during exercise of adding oxygen to the inspired air. J. Physiol. (London) 125:118 137. Cunningham, D.J.C. (1974). Integrative aspects of the regulation of breathing: A personal view. In: Respiratory Physiology, edited by J. G. Widdicombe. London, Butterworth University Park Press, pp. 304~369. Deiours, P.. Y. Labrousse, J. Raynaud, F. Girard and A. Teillac (1958). Stimulus oxyg6ne de la

R E F L E X HYPOXIC DRIVE

EXERCISE

87

ventilation au repos et au cours de l'exercice musculaire fi basse altitude (50 m) chez l'homme. Rev. Fr. Etud. Clin. Biol. 3: 105-123. Dejours, P. (1962). Chemoreflexes in breathing. Physiol. Rev. 42:335 358. Dejours, P. (1964). Control of respiration in muscular exercise. In: Handbook of Physiology. Section 3. Respiration. Vol. 1, edited by W. O. Fenn and H. Rahn. Washington D.C., American Physiological Society, pp. 631 648. Downes, J. J. and C. J. Lambertsen (1966). Dynamic characteristics of ventilatory depression in man on abrupt administration of O. J. Appl. Physiol. 21 : 447-453. Lugliani, R., B. J. Whipp, C. Seard and K. Wasserman (1971). Effects of bilateral carotid body resection on ventilatory control at rest and during exercise in man. N. Engl. J. Med. 285:1105-1111. Stockley, R. A. (1977). The estimation of the resting reflex hypoxic drive to respiration in normal man. Re,spir. Physiol. 31:217 230. Weil, J.V., E. Byrne-Quinn, I.E. Sodal, J.S. Kline, R.E. McCullough and G . F . Filley (1972). Augmentation of chemosensitivity during mild exercise in normal man. J. Appl. Physiol. 33 : 813-819.