Vol. 83, Issue 6, 2098-2104, December 1997

Ventilation and hypoxic ventilatory responsiveness in Chinese-Tibetan residents at 3,658 m

Linda S. Curran¹, Jianguo Zhuang², Shin Fu Sun², and Lorna G. Moore^1,3

¹ Department of Anthropology, University of Colorado at Denver, Denver 80217-3364; ² Tibet Institute of Medical Sciences, Lhasa, Tibet Autonomous Region, China 850000; and ³ Cardiovascular Pulmonary Research Laboratory, University of Colorado Health Sciences Center, Denver, Colorado 80262

ABSTRACT

INTRODUCTION

MATERIALS AND METHODS

RESULTS

DISCUSSION

ACKNOWLEDGEMENTS

FOOTNOTES

REFERENCES

ABSTRACT

Curran, Linda S., Jianguo Zhuang, Shin Fu Sun, and Lorna G. Moore. Ventilation and hypoxic ventilatory responsiveness in Chinese-Tibetan residents at 3,658 m. J. Appl. Physiol. 83(6): 2098-2104, 1997.When breathing ambient air at rest at 3,658 m altitude, Tibetan lifelong residents of 3,658 m ventilate as much as newcomers acclimatized to high altitude; they also ventilate more and have greater hypoxic ventilatory responses (HVRs) than do Han ("Chinese") long-term residents at 3,658 m. This suggests that Tibetan ancestry is advantageous in protecting resting ventilation levels during years of hypoxic exposure and is of interest in light of the permissive role of hypoventilation in the development of chronic mountain sickness, which is nearly absent among Tibetans. The existence of individuals with mixed Tibetan-Chinese ancestry (Han-Tibetans) residing at 3,658 m affords an opportunity to test this hypothesis. Eighteen men born in Lhasa, Tibet, China (3,658 m) to Tibetan mothers and Han fathers were compared with 27 Tibetan men and 30 Han men residing at 3,658 m who were previously studied. We used the same study procedures (minute ventilation was measured with a dry-gas flowmeter during room air breathing and hyperoxia and with a 13-liter spirometer-rebreathing system during the hypoxic and hypercapnic tests). During room air breathing at 3,658 m (inspired O₂ pressure = 93 Torr), Han-Tibetans resembled Tibetans in ventilation (12.1 ± 0.6 vs. 11.5± 0.5 l/min BTPS, respectively) but had HVR that were blunted (63 ± 16 vs. 121 ± 13, respectively, for HVR shape parameter A) and declined with increasing duration of high-altitude residence. During administered hyperoxia (inspired O₂ pressure = 310 Torr) at 3,658 m, the paradoxical hyperventilation previously seen in Tibetan but not Han residents at 3,658 m (11.8 ± 0.5 vs. 10.1 ± 0.5 l/min BTPS) was absent in these Han-Tibetans (9.8 ± 0.6 l/min BTPS). Thus, although longer duration of high-altitude residence appears to progressively blunt HVR among Han-Tibetans born and residing at 3,658 m, their Tibetan ancestry appears protective in their maintenance of high resting ventilation levels despite diminished chemosensitivity.

control of breathing; hypoxic ventilatory depression; high altitude

INTRODUCTION

LIFELONG TIBETAN RESIDENTS at 3,658-m altitude, breathing ambient air at rest, ventilate as much as newcomers acclimatized to 3,658 m, unlike immigrant Han (Chinese) long-term residents at 3,658 m (9, 34). Tibetans residing at 3,658 m also have greater hypoxic ventilatory responses (HVRs) than do acclimatized Han who came to 3,658 m as children (34). However, it has also been found that Tibetan lifelong residents at 4,400 m who were measured within 3 days after descent to 3,658 m ventilate as much as do Tibetans who were lifelong residents at 3,658 m, despite having blunted ventilatory responses similar to Han residents at 3,658 m (4). These previous studies suggest the existence of population variation in the effects of lifelong hypoxia on ventilation (E) and ventilatory control. These findings are of interest because hypoventilation has been implicated in the development of chronic mountain sickness among native high-altitude residents (5, 9, 12, 14, 17, 28). Coincidentally, Tibetans have very low prevalence of chronic mountain sickness when compared with high-altitude residents in China (22, 33) and the Americas (24, 32). Whether Tibetans' ability to maintain higher E is inherited is unknown, but previous studies have demonstrated that relevant components of O₂ transport may be inherited, e.g., studies of pulmonary function in Andeans (20) and Chinese (10), studies of HVR in Tibetans (1), and familial studies of HVR (1, 3, 11, 18). Although E involves many pathways and components and is likely to be subject to complex determination, the frequent observation that decreased HVR accompanies lower resting E in at least some long-term high-altitude residents (14, 15, 25, 30) has previously led some researchers to implicate diminished chemosensitivity to hypoxia as the likely cause. The existence of individuals with mixed Tibetan-Chinese ancestry (Han-Tibetans) who were born and have resided at 3,658 m all their lives affords an opportunity to test the hypothesis that Tibetan ancestry confers an advantage in the maintenance of high resting E during lifelong hypoxic exposure. We anticipated values intermediate between the two parental populations for the primary variables [E, HVR shape parameter A (HVR A), hypercapnic ventilatory response S (HCVR S), hyperoxic E] in these Han-Tibetans, and we hypothesized that there might be a lesser decline in their HVR  A as a function of residence duration than was previously found in Han (34). To determine whether Han-Tibetans resemble Tibetans or Han, or are intermediate between parental populations in ventilatory characteristics, such as HVR and HCVR, we measured 18 lifelong Han-Tibetan residents at 3,658 m for comparison with 27 Tibetan and 30 Han residents at 3,658 m who were studied previously at the same altitude (3,658 m) by the same research team and with the use of the same study procedures. Because the Han-Tibetans were born and raised at high altitude, their exposure to hypoxia is similar to that of the Tibetans and longer than that of the Han group, affording the opportunity to investigate the relative roles of Han and Tibetan ancestry in their levels of resting E without the complication of differing duration of hypoxic exposure.

MATERIALS AND METHODS

Subjects for this study were 18 Han-Tibetan male residents from Lhasa, Tibet Autonomous Region, China (altitude 3,658 m). Subjects granted informed consent for study procedures that had been approved by the Human Research Committees of the University of Colorado Health Sciences Center and by the Tibet Institute of Medical Sciences. All were 20- to 31-yr-old and were judged healthy by history, physical examination, resting electrocardiogram, and the ratio of 1-s forced expiratory volume (FEV₁) to forced vital capacity (FVC). All of the subjects had been born and raised at 3,658-m altitude, where they worked as hospital or clerical workers. None of the group engaged in exercise more strenuous than routinely riding bicycles for transportation. Of the Han-Tibetans, ~60% (11/18) had smoked for an average of 4.0 ± 0.7 pack · yr (packs/day × years smoked). The Han-Tibetan subjects were older and taller than the subjects in the two comparison groups, and Han-Tibetans tended to be heavier than the Han subjects (Tables 1 and 2). Blood pressure (diastolic/systolic pressure) was similar among the three groups (Tibetans, 113 ± 2/71 ± 3 mmHg; Han-Tibetans, 100 ± 3/73 ± 2 mmHg; and Han group, 104 ± 2/69 ± 1 mmHg), as were resting heart rates (Tibetans, 73 ± 3 beats/min; Han-Tibetans, 72 ± 3 beats/min; and Han group, 78 ± 3 beats/min).

Table  1.   Group characteristics Characteristic Tibetans* (n = 27) P Han-Tibetans (n = 18) P Han* (n = 30) P* Height, cm 166 ± 1  0.05 170 ± 1  <0.01 165 ± 1  NS Weight, kg 55 ± 1  NS 59 ± 1  <0.01 53 ± 1  NS Duration of residence at 3,658 m, yr 23 ± 1  NS 25 ± 1  <0.01 14 ± 1  <0.01 Smoking history; pack · yr 1.2 ± 0.6  <0.01 4.0 ± 0.7  <0.01 1.5 ± 0.5  NS  O2, ml/min STPD 278 ± 10  <0.01 224 ± 12  NS 250 ± 10  <0.05  O2, ml · min1 · kg1 STPD 5.0 ± 0.2  <0.01 3.8 ± 0.2  <0.01 4.8 ± 0.2  NS  CO2, ml/min STPD 238 ± 10  <0.05 206 ± 12  NS 202 ± 9  <0.05  CO2, ml · min1 · kg1 STPD 4.3 ± 0.2  <0.01 3.5 ± 0.2  NS 3.8 ± 0.2  <0.05 Respiratory quotient 0.85 ± 0.03  NS 0.93 ± 0.03  <0.01 0.81 ± 0.03  NS  E, l/min BTPS 11.5 ± 0.5  NS 12.1 ± 0.6  <0.01 10.1 ± 0.5  <0.05 f, breaths/min 19.1 ± 1.8  <0.05 25.1 ± 2.2  <0.01 16.5 ± 1.7  <0.05 PETO2, Torr 66 ± 1  <0.01 71 ± 1  <0.01 67 ± 1  NS  E/O2, l BTPS · min1 · liters 1 STPD 42.8 ± 2.4  <0.01 57.9 ± 2.8  <0.01 41.0 ± 2.3  NS HVR  0.47 ± 0.05  NS  0.35 ± 0.06  NS  0.40 ± 0.05  NS Hyperoxic E, l/min BTPS 11.8 ± 0.5  <0.01 9.8 ± 0.6  NS 10.1 ± 0.5  <0.01 HCVR B, Torr 22.5 ± 5.8  <0.01  24.9 ± 7.1  <0.01 18.3 ± 5.5  NS Values are means ± SE; n, no. of subjects. O2, O2 consumption; CO2, CO2 production; E, minute ventilation; f, respiratory frequency; PETO2, end-tidal PO2; HVR, hypoxic ventilatory response; HCVR B, hypercapnic ventilatory response x-intercept; pack · yr, packs/day × years of smoking; NS, not significant. * Data from Ref. 34. P, Han vs. Tibetans. 0.05 < P < 0.10.

Table  2.   Individual characteristics for Han-Tibetans Subject No. Age, yr Hb, g/dl PETCO2, Torr SaO2, %  HVR A HPETCO2, Torr HCVR S  2 23 17.1 34.2 89.7 112 33.6 0.41  3 30 18.1 30.1 92.0 102 29.5 0.41 14 25 18.5 30.7 88.8 100 18.4 0.39 26 24 18.3 31.2 89.7 46 28.8 1.06 27 23 15.6 29.0 86.3 55 27.2 0.84 28 29 16.4 29.6 91.6 46 29.3 0.69 29 31 19.3 33.5 90.9 69 31.5 0.25 30 26 17.6 32.4 91.3 116 31.3 0.23 31 28 17.4 30.5 89.9 132 31.8 0.57 42 25 18.1 23.5 91.0 57 30.6 0.86 54 30 16.7 33.3 90.0 27 35.7 1.30 57 30 18.9 32.4 91.8 84 31.3 0.89 58 31 21.5 29.8 91.0 81 28.0 0.98 62 30 23.9 27.3 92.8 29 31.1 0.60 63 29 19.9 20.5 92.5 20 13.3 1.61 64 29 19.8 32.4 93.3 21 31.8 1.44 65 30 18.9 31.6 92.0 5 31.5 0.93 67 23 17.2 27.5 89.8 33 27.3 0.79 Han-Tibetans 27 ± 1  18.5 ± 0.3  30.0 ± 0.7  91.1 ± 0.7   63 ± 16  29.0 ± 1.0  0.74 ± 0.13  Tibetans 23 ± 1* 17.8 ± 0.3 32.2 ± 0.6* 89.3 ± 0.6 121 ± 13* 29.9 ± 0.8  1.44 ± 0.11* Han 23 ± 1* 18.1 ± 0.2  31.6 ± 0.5 90.2 ± 0.5   81 ± 12  30.8 ± 0.7  1.10 ± 0.10* Hans vs. Tibetans NS NS NS NS <0.05 <0.05 <0.05 Values in 3 rows at bottom are means ± SE. Hb, hemoglobin; PETCO2, end-tidal PCO2; HVR A, hypoxic ventilatory response shape parameter; HPETCO2, hyperoxic PETCO2; SaO2, arterial O2 saturation; HCVR S, hypercapnic ventilatory response slope. * P < 0.05,  0.05 < P < 0.10, compared with Han-Tibetans. Data from Ref. 34.

RESULTS

The two comparison groups of 30 Han and 27 Tibetans were well matched for age and body size in the previous study design (34), but the Han-Tibetans in this study were taller than either of the other two groups and were heavier than the Han subjects (Table 1). Han-Tibetans were on average 4 yr older than the Tibetans or Han subjects (Table 2), but they were similar to Tibetans in duration of residence at high altitude (Table 1). Han-Tibetans resemble Han subjects living at 3,658 m in FVC, whether expressed in absolute terms or corrected for height (Tibetans, 4,956 ± 106 ml BTPS; Han-Tibetans, 4,392 ± 130 ml BTPS; and Han, 4,389 ± 102 ml BTPS). Hemoglobin values were similar in Han-Tibetans and Han (Table 2), but Han-Tibetans tended to differ from Tibetans (P = 0.097).

Han-Tibetans resemble Tibetans living at 3,658 m in E while breathing room air (PI_O2 = 93 Torr) at rest (Table 1, Fig. 1), whether or not it is corrected for Han-Tibetans' greater height. Han-Tibetans resemble Han in HVR A values (Table 2, Fig. 2). However, E/ was similar for all three groups (Table 1). Values for Sa_O2 were similar between and within each of the three samples, whether or not the encumbrance of a mouthpiece was present (see Table 2 for on-mouthpiece values). Sa_O2 off-mouthpiece values were 90.0 ± 0.4 for Tibetans, 91.7 ± 0.3 for Han-Tibetans, and 90.8 ± 0.3% for Han (data not shown). In both Han-Tibetans and Han men, increasing duration of high-altitude residence is associated with declining HVR, with similar slopes (Fig. 3). Han-Tibetans also resemble the Han group in their lack of hyperventilatory response to hyperoxia (Table 1). The PET_CO2 during hyperoxia was similar in all three groups (Table 2). Han-Tibetans' respiratory frequency was greater than that in the Han or Tibetan groups (Table 1). Han-Tibetans also had greater E per unit O₂ and higher PET_O2 than either the Han or Tibetan groups (Table 1). PET_CO2 was lower in Han-Tibetans than in Tibetans and tended to be lower than in the Han group (P = 0.077; Table 2, Fig. 4).

HCVR was lower in Han-Tibetans than in the Tibetan or Han groups (Table 2). The x-intercept B was also reduced in Han-Tibetans relative to both Tibetan and Han groups (Table 1). There was no decline in HCVR with increasing age in Han-Tibetans (data not shown).

DISCUSSION

The major finding of this study was that Han-Tibetans residing at 3,658 m had levels of resting E that were similar to those of Tibetans living at 3,658 m (Table 1, Fig. 1), despite having lower HVR (Table 2, Fig. 2). Thus, with lifelong exposure to hypoxia, both Tibetans (E = 11.5 ± 0.5 l/min BTPS) and Han-Tibetans (E = 12.1 ± 0.6 l/min BTPS) residing at 3,658 m are able to maintain levels of resting E similar to acclimatized newcomers to 3,658 m [n = 16, residence duration = 4 ± 1 yr, E = 10.6 ± 0.5 l/min BTPS (34)]. Han-Tibetans accomplish their high resting E despite a progressive decline in HVR with lengthening duration of high-altitude residence (Fig. 3).

We previously reported a decrease in HVR with increasing duration of high-altitude residence in Han subjects residing at 3,658 m (34) and in Tibetan subjects residing at 4,400 m who were studied within 3 days of descent to 3,658 m (4). Duration of residence at high altitude (4,400 m) was greater in Tibetans (25 ± 1 yr) than in the Han group (9 ± 1 yr). However, Tibetans living at 4,400 m were still able to maintain resting E levels similar to those of newcomers acclimatized to high altitude, whereas the Han group experienced decreased levels of resting E. The possibility exists that blunting of hypoxic drive in Tibetans does not become evident until more advanced age and/or longer residence duration at 3,658 m or that Tibetans may require residence at altitudes >3,658 m to blunt ventilatory drive. Nevertheless, Tibetan residents at 4,400 m, like Han-Tibetan residents at 3,658 m, appear to maintain resting E despite blunted HVR.

Our approach for investigating the relative roles of Tibetan and Han ancestry in levels of resting E and hypoxic drive was to recruit individuals born in Tibet at 3,658 m to a Tibetan mother and Han father for the purposes of comparison with both Tibetans and Han individuals residing at 3,658 m. While we sought to recruit subjects who were as similar as possible in characteristics known to influence E and ventilatory responsiveness, the hybrid nature of this population and/or their favored social status resulted in the Han-Tibetans being taller than individuals in either of the other two samples and heavier than members of the Han group. Their existence is also a unique historical phenomenon, the outcome of government-sanctioned intermarriage during the first decade of Chinese presence in Tibet. Such unions are now relatively rare in Tibet, with the result that the Han-Tibetan cohort we studied is somewhat older than our Han or Tibetan groups. Adjusting the analysis for body mass index, body surface area, or height did not change either the nature or statistical significance of our findings (data not shown).

Another difference between Han-Tibetans and the two comparison groups was in the extent of smoking, with Han-Tibetans' smoking histories exceeding values for both the Han and Tibetan groups (Table 1). However, there were no significant differences between smokers and nonsmokers in hemoglobin, Sa_O2, FEV₁/FVC ratio, or end-tidal gases in any of the three groups. There was only a trend (0.5 < P < 1.0) toward lower breathing frequency and FVC in Han-Tibetan smokers. We concluded that smoking is not likely to have influenced the Han-Tibetans' E or ventilatory responsiveness.

We used multiple measures to assess E and ventilatory responsiveness. Because E, Sa_O2, PET_O2, and E per unit O₂ were equal or greater in the Han-Tibetans compared with the Tibetan group, we concluded that Han-Tibetans ventilated at least as much as did Tibetans. Furthermore, Han-Tibetan E was greater than Han values, despite a duration of high-altitude residence nearly three times longer (Table 1). We found that although Han-Tibetans' HVR A values were similar to those of the Han, E/ did not differ among the three groups (Table 1). The reasons why one but not the other measure of HVR differed among the three groups was not immediately apparent, but the two measures were significantly correlated for Han-Tibetans (r = 0.82, P < 0.05). The agreement between these two measures of hypoxic ventilatory sensitivity and the well-recognized association between HVR and HCVR (2, 8) supported the likelihood of blunted ventilatory responsiveness in Han-Tibetans when compared with Tibetans. Higher E in the Han-Tibetans compared with Han was due to increased respiratory frequency, which exceeded even Tibetan values (Table 1). The possibility of hyperventilation in Han-Tibetans is rendered less likely by the absence of differences in Sa_O2 or breathing frequency on and off the mouthpiece. PET_CO2 was lower in Han-Tibetans compared with Tibetans and tended to be lower compared with the Han group (Table 2, Fig. 4). The ventilatory equivalent per unit O₂ was higher in Han-Tibetans than in either the Han or Tibetan groups (Table 1), indicating greater alveolar ventilation per unit metabolic rate.

Given the similarity in PET_O2 and PET_CO2 between Tibetan and Han individuals in the previous study (34), the CO₂ production differences in end-tidal gases between Han-Tibetans and both of these groups were unexpected. Han-Tibetans compared with Tibetans had lower relative levels of O₂ but similar E (Table 1), which contributed to their higher PET_O2 and higher calculated ventilatory equivalent for O₂. In the absence of direct measurements, it is not possible to discern whether the Han-Tibetan sample had a narrower alveolar-arterial O₂ gradient, such has been found in Tibetan but not Han residents at 3,658 m (35), although there was no discernible lung disease or decrement in Han-Tibetans' Sa_O2 at a given PO₂. Han-Tibetan subjects compared with Han subjects had similar levels of CO₂ but higher E and lower O₂, which resulted in a lower PET_CO2 and higher calculated respiratory quotient, respectively.

Problems for the interpretation of our study results stemmed from considerations relating to the numbers of subjects who could be examined during a given time period. Although all subjects were studied with the use of the same equipment and by many of the same investigative personnel, the Han and Tibetan groups living at 3,658 m were studied during 1987-1989, a study of 10 Han-Tibetans was conducted in 1991, and the remaining 8 Han-Tibetans were studied in 1994. The similarity in anthropometric measurements between the two groups of Han-Tibetans, as well as their similar levels of resting E, HVR, and PET_CO2, convinced us that the differences we observed in comparison with Han and Tibetans were more likely attributable to Han-Tibetans' mixed population ancestry than to different study times.

We consider the most likely explanation for the lower hypoxic ventilatory sensitivity observed in the Han-Tibetans to be their lifelong residence at 3,658 m. This conclusion is supported by our observation of decreasing HVR with increasing years of high-altitude residence in Han-Tibetans (Fig. 3). In this regard, Han-Tibetans resemble Tibetans living at 4,400 m (4), Han long-term residents at 3,658 m (34), and Sherpas (6). There are a number of possible explanations for the progressive blunting of ventilatory sensitivity. If the development of peripheral chemoreceptor response to hypoxia is prompted by increased O₂ availability in the transition from intrauterine to extrauterine life (7, 13, 21, 27), and if Han-Tibetans resembled the Han subjects in having poorer neonatal oxygenation, this might have led to decreased ventilatory sensitivity to hypoxia from birth. However, the similarity in average HVR at age 20 yr (Fig. 3) between Tibetans and Han-Tibetans (147.5 and 136.0, respectively) does not support this hypothesis, although the slopes of the HVR vs. duration of high-altitude residence relationship are quite similar for Han and Han-Tibetan subjects (2.36 and 2.92, respectively). An altered peripheral chemosensory drive as a consequence of the Han-Tibetans' lifelong exposure to hypoxia is more consistent with their blunted hypoxic and hypercapnic responses, which are approximately one-half the values we previously reported for Tibetans living at 3,658 m (Table 2). A similarly blunted peripheral chemosensory drive may apply to Tibetans residing at 3,658 m, although their narrow range of residence duration did not allow us to confirm this. Tibetan residents at 4,400 m have blunted hypoxic ventilatory drives (4) yet maintain E levels similar to those of Tibetans living at 3,658 m.

Our observation that Han-Tibetans did not hyperventilate while breathing hyperoxic gas mixtures differs from previous reports for lifelong high-altitude residents of the Rocky Mountains, Andes, and Himalayas (6, 15, 16, 26), as well as our own previous study of Tibetan residents at 3,658 m (34). However, a similar lack of hyperoxic hyperventilation was found in our study of Tibetan residents at 4,400 m (4). A clear consensus is lacking on the mechanisms for hyperventilation during exposure to hyperoxia, but these may be due to a central depressant effect of hypoxia that is effectively opposed by the peripheral chemosensory drive to maintain higher levels of resting E. If so, the lack of hyperventilatory response to hyperoxia in the Han-Tibetans could indicate an irreversible central depressant effect of hypoxia, presumably also an effect of lifelong exposure to hypoxia and possibly poorer neonatal oxygenation as well.

These findings support the hypothesis that Tibetan ancestry confers an advantage in maintaining resting E under conditions of sustained hypoxic exposure. This advantage appears to involve compensation for impaired chemosensitivity to hypoxia so that Han-Tibetan lifelong residents at 3,658 m have resting E levels that are similar to Tibetans who are lifelong residents at 3,658 m and 4,400 m and similar to acclimatized newcomers. The mechanism for this protected E remains to be investigated, but the fact that Tibetans living at 4,400 m and Han-Tibetans living at 3,658 m evidence blunted hypoxic and hypercapnic responses but still maintain similar levels of resting E appears to implicate mechanisms other than central or peripheral chemoreceptors for ventilatory control among high-altitude residents with Tibetan ancestry.

The nature of Han-Tibetans' resemblance to their parent populations suggests several avenues for future study. For the most part, Han-Tibetans show a pattern of stronger resemblance to one or the other of the two parent populations, rather than intermediate values in the array of ventilatory variables we measured. To the extent that such traits are polygenic and located in the nuclear rather than mitochondrial genome, Han-Tibetans' ventilatory characteristics should tend to be intermediate between those of the parent populations, but they are not. There are a number of possible explanations for this. E may be controlled by a smaller number of factors having a pattern of simple Mendelian (dominant/recessive) inheritance, with Han-Tibetans being heterozygous for most of these traits. Should this be so, the next generation (offspring of Han-Tibetans) could be expected to include more variation in ventilatory phenotypes when pairing of recessive genes occurs. Another possibility is that some ventilatory traits may be influenced by the maternally transmitted mitochondrial genome. Approximately 20% of the oxidative phosphorylation enzymes are coded for by genes located within the mitochondrial genome, including complexes I-IV of the electron transport chain and complex V of ATP synthase (19, 29). In this case, Han-Tibetans should resemble their maternal population, because no paternal contribution to the mitochondria is present after conception. Therefore, the ventilatory characteristics of Han-Tibetans having Han mothers should differ from those with Tibetan mothers. A third possibility is that the contribution of developmental factors to adult phenotypes may be more substantial and complex than is presently appreciated. In particular, the interaction of genetic and environmental variables during the neonatal period, especially as these influence the development of hypoxic and hypercapnic ventilatory sensitivity, should be considered.

ACKNOWLEDGEMENTS

This study was made possible by the cooperation of the subjects and by the assistance of the Bureau of Health of the Tibet Autonomous Region.

FOOTNOTES

Address for reprint requests: L. S. Curran, Dept. of Anthrolpology, C. B. 103, PO Box 173364, University of Colorado-Denver, Denver, CO 80217-3364 (E-mail: LCURRAN{at}CASTLE.CUDENVER.EDU).

Received 29 July 1996; accepted in final form 7 August 1997.

REFERENCES