INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

VENTILATORY ACCLIMATIZATION to high altitude (VAH) involves a progressive rise in minute ventilation (E) and a fall in end-tidal PCO₂ (PET_CO2). The mechanisms that underlie this process are not completely understood. Part of the process involves an elevation of the peripheral chemoreflex sensitivity to hypoxia (6, 16). However, not all of the VAH can be explained by this because, after a period of acclimatization, return to hyperoxia does not bring E back to baseline (5, 10). One interpretation of this is that the residual hyperventilation in hyperoxia arises from the alteration in acid-base status induced by the hypocapnia and that it takes days to recover from this. However, we have observed a sustained elevation of E after exposure to sustained isocapnic hypoxia, in which acid-base changes have been prevented (17, 18). In addition, Forster et al. (7) doubt that, on exposure to acute hyperoxia following VAH, the associated rise in PET_CO2 is quantitatively sufficient to explain the residual hyperventilation via a mechanism involving the central chemoreceptors. Thus an alternative explanation of the persistent hyperventilation in hyperoxia is that carotid body (CB) function is so altered after hypoxia that there is persistent activity, even in hyperoxia.

To investigate this possibility, the acute peripheral chemoreflex response to hypoxia was first suppressed with the use of a hyperoxic gas mixture at various stages of an acclimatization process. Then, on top of this, the effect of low-dose dopamine (DA) administration (3 µg · kg¹ · min¹) was investigated. In humans, low doses of DA substantially reduce CB-induced respiratory activity (23), a finding shown to persist during the early stages of VAH (13). If a component of the ventilatory acclimatization that is not readily reversible by hyperoxia arises (at least in part) through an increase in CB activity, then we would expect DA to become progressively more effective in suppressing E and elevating PET_CO2 as acclimatization proceeds. If, on the other hand, no part of this component arises from the CB, then we would expect no change in the effect of DA with acclimatization.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects. Nine healthy young adults, eight for the study and one reserve, were selected (age = 23-37 yr, all were men). The selection of subjects for the study was based on a number of criteria. Excellent health and physical condition were required. The subjects' motivation to participate was assessed, and psychological tests were employed to try to determine their ability to withstand confinement in a chamber for 5 wk. Medical students, doctors, or other medically related workers who had previous experience in altitude were preferred. All subjects exercised a certain amount on each day of chamber confinement. All subjects received written and verbal descriptions of the experiment before they gave their consent. The study was approved by the Marseilles Research Ethics Committee.

Acclimatization. Eight subjects lived in a hypobaric chamber (COMEX, life chamber type 2500, length of 8 m, available volume = 32 m³) at normal room temperature during 31 days of progressive decompression designed to simulate barometric pressure conditions during an ascent of Mount Everest. Figure 1 illustrates the progression profile. Before subjects entered the chamber, they spent 1 day at Refuge de Cosmique (3,650 m) and 5 days at Observatoire Vallot (4,350 m) to preacclimatize.

View larger version (14K): [in this window] [in a new window]   Fig. 1.   Progression profile for barometric pressure during the simulated ascent of Mount Everest. Arrows indicate the days on which the present study was carried out.

Experimental setup. All experiments were carried out in a hydrosphere (diameter = 5 m, available volume = 65 m³) that was equipped with the necessary technical facilities. Each morning, the hydrosphere was decompressed to reach the barometric pressure appropriate to the level of ascent inside the life chamber; each evening, it was recompressed to ambient pressure outside the chamber. This allowed the experimenters (who had their own oxygen supplies) to enter the hydrosphere in the morning to conduct experiments with the subjects and to leave it in the evening. A lock (COMEX, type 2000) served as the connection (and bathroom) between the life chamber and the hydrosphere. The life chamber was equipped with eight sleeping bunks and a table for meals. At 7,000 m, the subjects were sufficiently familiar with the experiments, thus allowing experimenters to supervise procedures outside the hydrosphere via a microphone.

Protocol. For each subject, the experiment was undertaken twice before any period of acclimatization; twice at 5,000 m [422 mmHg, inspired PO₂ (PI_O2) was 79 Torr], with 4 days between measurements; twice at 7,000 m (324 mmHg, PI_O2 = 58 Torr), again with 4 days between measurements; and, finally, on days 2 and 4 after return to sea level (Fig. 1). Each experiment lasted 40 min, which meant that it took most of the day to complete the experiments on all eight subjects. The order in which the subjects were studied on each day was random, but the protocol was always the same (Fig. 2). Resting pulmonary ventilation was measured during the initial 10 min (period 1: "air"). Hyperoxia [end-tidal PO₂ (PET_O2) of ~200 Torr] was then introduced and maintained throughout the remaining 30 min. The first 10 min of hyperoxia (period 2: "hyperoxia-1") formed an initial control period. Then, the DA infusion (3 µg · kg¹ · min¹ iv) was started and continued for 10 min (period 3: "DA and hyperoxia"). Finally, a second 10-min control period of hyperoxia followed (period 4: "hyperoxia-2"). The dose of DA was kept the same in all experiments on any given subject (and was based on the subjects' weight at sea level before acclimatization). To obtain PET_O2 values of ~200 Torr, the fraction of inspired oxygen in the hyperoxic mixture was progressively increased. The inspired oxygen used was 30.9% at sea level, 58.7% at 5,000 m, and 79.4% at 7,000 m.

View larger version (14K): [in this window] [in a new window]   Fig. 2.   Protocol in terms of end-tidal PO2 (PETO2) for all experiments: sea level (SL), return to sea level (RSL), 1st and 5th day at 5,000 m (1 and 2, respectively), 1st and 5th day at 7,000 m (1 and 2, respectively). Arrows indicate time periods used for data analysis.

Respiratory measurements. Subjects were seated comfortably inside the hydrosphere and breathed through a mouthpiece with the nose occluded. The mouthpiece was connected in series with a valve so that the inspirate could be either ambient air or the appropriate hyperoxic gas mixture from a 150-liter Douglas bag that was filled continuously from a cylinder containing the premixed gas. The expirate would either go to the environment or, during hypobaria, be directed out of the hydrosphere. Expiratory airflow was sensed with a symmetrically disposed Pitot tube flowmeter, and E was determined from this (14). The gas analyzers consisted of a galvanic fuel cell for measurement of oxygen concentration and an infrared carbon dioxide analyzer (model CPX/D, Medical Graphics System). A pulse oximeter was attached to the earlobe to monitor oxygen saturation (model 502, Criticare). Arterial blood pressure was collected over 20-s intervals by an automatic sphygmomanometer (Dinamap 1846 SX P, Criticon) to monitor any adverse effects of the DA infusion on blood pressure. All experimental data were recorded in real time by a computer.

Data analysis. Breath-by-breath data for each experiment were first averaged over 60-s periods. Average values for the four periods (air, hyperoxia-1, DA and hyperoxia, and hyperoxia-2) were then obtained by averaging the data for the 6th-9th min of each period (see Fig. 2). The significance of the effects of altitude, first or second repeat at each altitude, hyperoxia, and DA was assessed with the use of ANOVA, with subjects treated as a random variable and the other factors as fixed effects. A probability of P < 0.05 was taken as statistically significant. Analysis was undertaken by using the statistical software package SPSS 7.0.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

During the preacclimatization period at Observatoire Vallot, one subject developed pulmonary edema and the reserve took his place. Of the 8 subjects entering the hypobaric chamber, three did not complete all three ascents to 8,848 m (see Fig. 1). One subject developed neurological symptoms at 8,000 m and was evacuated, one subject was evacuated after the first ascent (8,848 m), and one subject did not take part in the final ascent. The subjects all recovered within a few minutes of oxygen inhalation. All subjects entering the chamber reached the altitudes of interest for the present study. Subjects that were evacuated were studied on the 2nd and 4th day after leaving the chamber for the return-to-sea-level experiments. No subject felt any side effects from the DA infusions, and there was no change in blood pressure with the DA infusions in any subject at any altitude.

Blood measurements. Base excess, pH, and hemoglobin concentration rose during the ascent; the values for these variables are listed in Table 1.

Ventilatory responses to increasing hypobaria. The acclimatization process was characterized by a progressive increase in E and fall in PET_CO2. This is illustrated in Fig. 3, along with the changes in PET_O2. The increases in E and decreases in PET_CO2 while breathing ambient air were significant between altitudes (P < 0.001). After 2 and 4 days at return to sea level, E was still significantly higher (P < 0.001) and PET_CO2 was significantly lower (P < 0.001) than before any acclimatization had taken place.

View larger version (15K): [in this window] [in a new window]   Fig. 3.   Characteristics of the acclimatization process during ambient air breathing () together with the effects of acute hyperoxia (). A: ventilation. B: end-tidal PCO2 (PETCO2). C: PETO2. Values are average of all subjects; error bars show SE. Ventilation and PETCO2 vary significantly with altitude (P < 0.001, ANOVA). * Ventilation postacclimatization was significantly greater than that preacclimatization (P < 0.001).  PETCO2 postacclimatization was significantly lower than that preacclimatization (P < 0.001).

E in acute hyperoxia. Figure 4 illustrates the breath-by-breath results from one experiment from one subject (subject 9, experiment from the 5th day at 5,000 m). Figure 4 shows that the induction of hyperoxia decreases E and increases PET_CO2. Values for E, PET_O2, and PET_CO2 appear similar between the two hyperoxic periods (hyperoxia-1 and hyperoxia-2). This was generally true for all subjects and altitudes, with no significant differences being detected between the two periods. Therefore, in Figs. 3 and 5, which illustrate the overall responses of the group, average values were used for these two periods.

View larger version (22K): [in this window] [in a new window]   Fig. 4.   Breath-by-breath ventilation (A), PETO2 (B), and PETCO2 values for 1 subject (subject 9) from 1 experiment (5th day at 5,000 m). Arrows indicate time periods used for data analysis.

Effects of DA on E and PET_CO2 during acute hyperoxia. In the single experiment illustrated in Fig. 4, DA appeared to have little effect on either E or PET_CO2. The effect of DA on the data averaged from all subjects from all experimental days is shown in Fig. 5. DA, when administered on top of hyperoxia, had no consistent effect in reversing E further toward its preacclimatization value. For PET_CO2, the group mean values for DA were slightly higher than without DA; this result was consistent at all altitudes and was significant overall (P < 0.05). There was no progressive alteration in the effects of DA on either E or PET_CO2 as acclimatization proceeded.

View larger version (13K): [in this window] [in a new window]   Fig. 5.   Effect of dopamine when administered on top of hyperoxia () compared with the acute effect of hyperoxia alone (). A: ventilation. B: PETCO2. Values are average of all subjects; error bars show SE. Dopamine had no significant effect on ventilation (ANOVA) but did cause a mild elevation in PETCO2 (P < 0.05, ANOVA).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The major finding of this study was that DA, when infused at a low dose in patients undergoing acute hyperoxic exposure, did not suppress E or elevate PET_CO2 in a manner that became progressively more effective as acclimatization proceeded. This finding does not provide any support for the hypothesis that part of the hyperventilation that remains after VAH arises from an increase in CB activity that persists under the newly imposed hyperoxic state.

If a clear effect of DA on E and PET_CO2 that became progressively more marked with acclimatization had been demonstrated, then this study would have provided quite compelling evidence for a peripheral origin for some of the ventilatory stimulation that persists under hyperoxic conditions following VAH. However, because this study did not demonstrate such an effect, our question is this: to what extent can the present study exclude a peripheral origin for some of the persistent hyperpnea under hyperoxic conditions? With respect to this, the major issue is the degree of confidence that we have that DA would be expected to suppress residual activity at the CB under these conditions.

In anesthetized cats, infusion or injection of DA reduces both the chemosensory discharge from the CB (11) and ventilation (11, 24). Of particular note is that these findings were determined under hyperoxic conditions. Similar results have been obtained in hyperoxia for E (3) and phrenic nerve activity (12) in awake and anesthetized goats, respectively. In conscious humans, low-dose DA infusion reduces both the ventilatory sensitivity to hypoxia (1, 2, 4, 15, 20, 22, 23) and the ventilatory sensitivity to carbon dioxide (in euoxia) (15, 21). However, in the case of effects of DA in hyperoxia in humans, the authors are aware of only one previous study [Welsh et al. (23)], which found that low-dose DA infusion had no effect on E.

There are several possible interpretations of the finding of Welsh et al. (23). It is possible that there is a genuine species difference between cats and goats on the one hand and humans on the other hand, such that DA is an effective inhibitor of the CB under hyperoxic conditions in the former but not in the latter. However, there may also be a species difference such that hyperoxia is more effective in reducing carotid chemoreceptor activity in humans than in cats and/or goats, such that there remains little or no chemoreceptor activity in hyperoxic humans for the low-dose DA infusion to suppress. A third possibility is that the result of Welsh et al. arises from a type II statistical error. There was a nonsignificant reduction in E and elevation of PET_CO2 by low-dose infusion of DA in hyperoxia that was around the size of the standard error, favoring this possibility. Furthermore, the present study did manage to detect a small but significant effect of DA on PET_CO2 in hyperoxia, if not on E.

The previous studies cited above all relate to subjects or experimental animals that were not acclimatized to hypoxia. In rats, a 4-wk exposure to hypoxia resulted in an approximate doubling of CB diameter and protein content and a 15-fold increase in DA content (8). These changes, particularly in DA metabolism, certainly raise the possibility that the action of low-dose DA infusion may not be the same as in unacclimatized CB. In support of this notion, Tatsumi et al. (19) found in the anesthetized cat that the ability of domperidone to potentiate the ventilatory response to hypoxia was lost after VAH. However, this finding has not been replicated in subsequent studies in awake goats (9) and humans (13). In relation to the effects of low-dose DA infusion, we know of no studies that used this after a prolonged period of VAH. After a brief period of acclimatization, the percent reductions in both the ventilatory sensitivity to hypoxia and a calculated isocapnic residual ventilation in the absence of hypoxia generated by low-dose DA infusion appear similar to those found before acclimatization.

A further difference in the use of low-dose DA infusion in the present study compared with studies of the unacclimatized state relates to PET_CO2. In the present study, all subjects remained markedly hypocapnic in the hyperoxic exposure compared with their preacclimatization state. Sabol and Ward (15) found that DA was, in terms of percentage, more effective in suppressing euoxic E in hypercapnic conditions than under conditions of eucapnia. Thus one possible explanation for the results from the present study is that DA is relatively ineffective at suppressing CB activity under conditions of hypocapnia.

In summary, we have found no evidence for a persistent stimulus to ventilation arising from the CB under conditions of acute hyperoxia generated by VAH. The results suggest that the hyperventilation that remains in acute hyperoxia essentially arises centrally, but the strength of this conclusion is limited somewhat by uncertainties as to the action of low-dose DA infusion under these particular conditions.