INTRODUCTION

Top Abstract Introduction References

VENTILATORY ACCLIMATIZATION to altitude involves a progressive rise in ventilation and a progressive fall in the arterial partial pressure of CO₂. The processes underlying this phenomenon are incompletely understood. Part of the response involves an elevation in the acute hypoxic ventilatory response (AHVR) (10, 11, 22, 30), and there is evidence from animal work that this arises in the carotid body (CB) (5). However, the actual mechanism associated with this has not been fully elucidated.

In view of the abundance of dopamine in the CB (9), and the finding that exogenous administration of dopamine decreases the ventilatory response to hypoxia (2, 3, 8, 21, 26, 27, 29), one possibility is that the increase in AHVR during acclimatization to altitude is related to some reduction in dopaminergic activity. This has recently been investigated in cats by Tatsumi et al. (23). They compared the influence of dopaminergic blockade on hypoxic responses before and after acclimatization and found that peripheral dopaminergic blockade increased ventilatory and CB responses to hypoxia before, but not after, acclimatization. They concluded that the activity of inhibitory dopaminergic mechanisms within the CB is decreased after sustained hypoxic exposure and that this lessening of inhibition may contribute to the enhancement of AHVR in acclimatization to altitude.

The purpose of the present study was to investigate whether such a reduction in the activity of inhibitory dopaminergic mechanisms could also be detected in humans. The study compared the effects of dopaminergic blockade on AHVR with and without a prior period of sustained hypoxia. To determine whether there are any changes in the sensitivity to dopamine, the study also compared the inhibitory effects on AHVR of low-dose dopamine infusions with and without prior sustained hypoxia.

	METHODS

Subjects. Eleven healthy adults, seven men and four women, were studied. Their average age was 23.4 ± 1.1 (SD) yr, average weight was 70.3 ± 2.9 kg, and average height was 177.2 ± 2.5 cm. All subjects received written and verbal descriptions of the experiment before they gave their consent. Subjects were naive as to the precise purpose of the study and the particular protocol employed on the day. The study was approved by the Central Oxford Research Ethics Committee.

Protocols. Each subject was studied both with and without an antecedent period of 8 h of isocapnic hypoxia under each of the following three pharmacological conditions: 1) control, with no drug administered; 2) dopamine; and 3) domperidone. This gave a total of six different protocols. These were undertaken on 6 different days in random order. Each protocol involving prolonged hypoxia was followed by a period of at least 1 wk in which no experiments were conducted on that subject; the remaining experiments were separated by at least 24 h. In each case, the purpose was to determine AHVR.

1

1

Hypoxic exposures. The sustained hypoxic exposures were obtained by varying the gas composition in a purpose-built chamber. In the chamber, subjects were free to move around if they desired. Most spent their time sitting comfortably watching television or reading. Respired gas was sampled via a nasal catheter and analyzed by mass spectrometry. The PCO₂ and PO₂ signals were sampled by computer, and the end-tidal values were determined breath by breath. These values were used in a feedback algorithm to adjust the chamber composition every 5 min to force the end-tidal values of the subject toward the desired levels. A pulse oximeter was used to monitor the arterial saturation in the subject, and there was a separate oxygen monitor with an alarm inside the chamber (for a full description of chamber and gas-controlling system, see Ref. 13).

Measurement of AHVR. Measurements of AHVR were made outside the chamber by using an end-tidal forcing system. This system required subjects to breathe through a mouthpiece with the nose occluded. The mouthpiece was connected in series with a turbine volume-measuring device (15) and a pneumotachograph to measure respiratory volumes and flows. Gas was continuously sampled at the mouth and analyzed for PO₂ and PCO₂ by mass spectrometry. All experimental data were recorded in real time at a sampling frequency of 50 Hz by a computer that also determined PET_O2 and PET_CO2 together with the inspiratory and expiratory volumes and durations. The end-tidal gas values were then passed, breath by breath, to a second computer, which controlled a fast-responding gas-mixing system. The computer controlling the gas-mixing system compared the actual end-tidal values with the desired values, and, to obtain the desired end-tidal profile, made appropriate adjustments to the inspired gas on a breath-by-breath basis. A pulse oximeter was attached to a finger to monitor oxygen saturation, and electrodes were placed on the chest to obtain the electrocardiogram. The details of the forcing procedure and the gas-mixing system are described in greater detail elsewhere (14, 20).

Data analysis. The breath-by-breath data sequence used to assess AHVR was 14 min in length. It was composed of the 2-min period immediately before the square-wave hypoxic stimulation began, together with the data from the 6 repeats of the 120-s hypoxic square wave (each repeat consisted of 60 s at PET_O2 = 50 Torr and 60 s at PET_O2 = 100 Torr). The number of breaths in this data sequence varied between 131 and 385. To obtain numerical values for AHVR, a model of the respiratory response to hypoxia was fitted to each of the individual data sequences. The model used was model 3 as described by Clement and Robbins (6). This single-compartment model has four parameters: hypoxic sensitivity, Gp (l/min); residual ventilation in the absence of hypoxia, c (l/min); a pure delay for the lag between the hypoxic stimulus at the lung and the ventilatory response, T (s); and a time constant,  (s). These parameters were estimated by nonlinear regression [for full details, see Clement and Robbins (6)]. The values for Gp from the model were taken as the estimates of AHVR.

	RESULTS

None of the subjects had any side effects from the dopamine infusion or from the administration of domperidone. Blood pressure did not change substantially in any subject during the administration of dopamine.

Quality of gas control in the chamber. The quality of the gas control obtained in the chamber is illustrated in Fig. 1. PET_CO2 and PET_O2, averaged on a 5-min basis, are shown individually for each of the three exposures in each subject. PET_O2 was maintained constant at close to 55 Torr in all exposures. PET_CO2 was kept almost constant throughout the 8 h at the individual prehypoxic control values.

View larger version (39K): [in this window] [in a new window]   Fig. 1.   Gas control during 8 h of isocapnic hypoxia in chamber averaged on a 5-min basis. All 3 exposures are shown for all 9 subjects. A: end-tidal PO2 (PETO2; target value 55 Torr). B: end-tidal PCO2 (PETCO2; target values differ between experiments).

Quality of gas control during measurement of AHVR. An example of one experiment to determine AHVR is shown in Fig. 2. PET_CO2 was reasonably constant throughout. PET_O2 was lowered and raised abruptly at the onset and relief of hypoxia, with some overshoot at the relief of hypoxia. These findings were typical of all determinations of AHVR.

View larger version (16K): [in this window] [in a new window]   Fig. 2.   Breath-by-breath ventilatory response (A) and breath-by-breath end-tidal gas control (B and C) for a single measurement of acute hypoxic ventilatory response from 1 subject (subject 1015, posthypoxia, with domperidone).

Ventilatory responses. Figure 3 shows an example of one set of results for each protocol from one subject (subject 1015). This subject demonstrated repeatable responses to the successive square waves of hypoxia for all protocols except for the posthypoxic dopamine experiment. All other subjects generally demonstrated repeatable responses to the hypoxic square waves, although, as for subject 1015, there were sets of data for which the hypoxic responses were less steady. Also shown in the figure is the fit of the model to the data. In general this describes the data well.

View larger version (28K): [in this window] [in a new window]   Fig. 3.   Breath-by-breath ventilatory responses to square-wave hypoxia for each experiment for 1 subject (subject 1015). A: control. B: during infusion of dopamine. C: after administration of domperidone. , Responses before 8 h of isocapnic hypoxia; , after 8 h of isocapnic hypoxia; solid lines, fitted model response.

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

1

View larger version (14K): [in this window] [in a new window]   Fig. 4.   Effect of domperidone () and dopamine () on hypoxic sensitivity before and after 8 h of isocapnic hypoxia in absolute values (A) and relative to control (B). Values are average of all subjects; error bars, SE.

View larger version (12K): [in this window] [in a new window]   Fig. 5.   Effect of domperidone () and dopamine () on residual ventilation (c) before and after 8 h of isocapnic hypoxia. Values are average of all subjects; error bars, SE.

	DISCUSSION

This study found no evidence that a reduction in dopaminergic inhibition underlay the increase in AHVR that was observed after sustained hypoxic exposure. Rather, it appeared that the effects of both dopamine infusion and domperidone administration increased in proportion to the overall increase in AHVR after hypoxic exposure, an observation supported by a significant interactive term between prior hypoxia and drug in the ANOVA. Although the 95% CI for the increase in effect of domperidone after hypoxic exposure of 0.11 to 0.92 l · min¹ · %¹ includes zero, and it is therefore impossible to exclude some reduction in the effect of domperidone after hypoxic exposure, the lower limit of the CI is small compared with the rise in AHVR after hypoxic exposure of 0.62 l · min¹ · %¹ (95% CI 0.41-0.83 l · min¹ · %¹). This result suggests that little of the increase in AHVR after sustained hypoxia could have been brought about by a decrease in peripheral dopaminergic inhibition. This finding is in contrast to the findings of Tatsumi et al. (23) in cats, in which a marked reduction in the effect of domperidone on AHVR was detected.

A second subsidiary finding was that low-dose dopamine, but not domperidone, altered (reduced) the estimated level of ventilation in hyperoxia obtained as part of the model-fitting process.

Experimental issues relating to the present study. The technique used in the present study to assess AHVR was to switch PET_O2 repeatedly between 50 and 100 Torr in a series of square waves with a period of 2 min. This technique has the theoretical advantage over an incremental exposure to hypoxia that the results should be relatively unaffected by a varying baseline ventilation. It was also associated with a relatively low total exposure to hypoxia that should have helped to minimize the degree of hypoxic ventilatory decline that developed over the period of the measurement. The mean control value for AHVR of 0.72 ± 0.07 (SE) l · min¹ · %¹ compares favorably with a value of 0.74 l · min¹ · %¹ for 97 subjects from a recent study (24) with (an admittedly imperfect) protocol using incremental hypoxia. The value also compares well with a resting value of 180 for parameter "A" in the equation of Weil et al. (28), which was obtained by using a protocol involving incremental hypoxia (this equates to a sensitivity of ~0.66 l · min¹ · %¹). However, the value compares less well with the mean "V40" of 19.4 liters/ (min × m²) of Kronenberg et al. (17), obtained by using incremental hypoxia, which equates to a sensitivity of ~1.45 l · min¹ · %¹.

1

1

Comparisons with other studies. As noted above, the results from the study by Tatsumi et al. (23) appear dissimilar from ours. This arises presumably from either a difference in methods between the studies or a species difference.

Dopamine in the CB. The physiology of dopamine at the CB is complex and has recently been reviewed by Gonzalez et al. (12). In this review, Gonzalez et al. propose a model where, in response to hypoxia, dopamine is released locally by the type 1 cell at the site of the synapse with the carotid sinus nerve (CSN) ending. Dopamine then acts as an excitatory neurotransmitter at low-affinity receptors on the CSN ending. In addition to these dopamine receptors, there are also high-affinity dopamine autoreceptors on the type 1 cell, which are inhibitory and modulate the release of dopamine by the type 1 cell. With this model, low doses of dopamine inhibit neural activity through the high-affinity autoreceptors, whereas high doses of dopamine stimulate the CSN directly through the low-affinity receptors. Similarly, low doses of dopamine antagonists will enhance transmitter release through their effects on the high-affinity autoreceptors, whereas high doses of antagonists will inhibit the CSN discharge directly through the low-affinity receptors.

Effect of dopamine on c. The finding that dopamine depressed c, whereas domperidone did not affect c, is intriguing. The value for c is a hypothetical value for ventilation in the total absence of any hypoxic stimulation and thus needs to be treated with the caution that such hypothetical values deserve. Nevertheless, it is worth considering how this might be interpreted in terms of the model of Gonzalez et al. (12) for the actions of dopamine at the CB. The observation of a reduction in c with low-dose dopamine implies that there is some residual transmitter release from the CB that stimulates the CSN in the absence of hypoxia. However, the absence of any effect of domperidone on c suggests that, without hypoxia, the level of local release of dopamine at the synapse of the type 1 cell with the CSN is inadequate to provide any endogenous stimulation of the more remote autoreceptors on the type 1 cell.

Summary. Our results in humans do not support the idea that the increased ventilation after prolonged hypoxia is primarily related to a reduction in dopaminergic inhibition at the CB, as has been suggested by Tatsumi et al. (23) from experimental work in cats.