- Email: [email protected]
Respiration and blood gases in the duck exposed to normocapnic and hypercapnic hypoxia
Respiration Physiology (1987) 67, 1-12
1
Elsevier
RSP 01228
Respiration and blood gases in the duck exposed to normocapnic and hypercapnic hypoxia H. Shams and P. Scheid Institut J~r Physiologie, Ruhr-Universitiit Bochum, F.R.G.
(Accepted for publication 30 August 1986)
Abstract. Cardio-respiratory parameters were measured in the unrestrained, unanesthetized duck by whole body plethysmography during exposure to varied levels of inspired hypoxia without or with (3.7 ~o) CO 2. At any level of inspired Po~ (Plo2), ventilation (~/E) was larger with CO 2 inhalation, leading, for PIo2 > 50 Torr, to higher levels of arterial Po2 (Pao2) and 0 2 content (Cao2). Below Pto2 of 50 Torr, the (PI-Pa)Oz difference without CO2 reached a value as low as 5 Torr which was not diminished by further stimulation of ~'E by inhaled CO 2. Without CO 2 inhalation at this deep hypoxic level the ensuing hypoxiainduced respiratory alkalosis was partly compensated by lactacidosis, whereas CO 2 inhalation resulted in markedly lower blood pH leading to significantly lower arterial and venous 02 content (Bohr effect). As a result, the deepest level ofhypoxia tolerated without CO 2 inhalation, 30 Torr, is significantly deeper than that, 36 Torr, tolerated when CO 2 is inhaled. The data suggest that a number of factors contribute to the high hypoxia tolerance in birds, e.g. the effectiveness of parabronchial ventilation and the tolerance of low arterial Pco2 levels, whereby part of the lactacidosis is compensated.
Birds; Blood gases; Hypercapnia; Hypoxia; Respiration.
Birds are generally considered to be more tolerant of hypoxia than mammals (Tucker, 1968), although notable species differences among birds appear to exist (Butler, 1970). These differences between birds and mammals may be the result of a number of structural and functional differences, of which the high inherent gas exchange efficiency of the parabronchial lung (Scheid, 1979) is one example. Of particular interest to us was the possible role of arterial hypocapnia which occurs in animals as a result of the hypoxic ventilatory stimulation and may affect the hypoxia tolerance in many different ways. For one, it is expected to remove part of the hypoxic ventilatory drive, thus limiting arterial oxygenation, as has been shown in man (Forwand Correspondence address: Hashim Shams, Institut for Physiologie, Ruhr-Universitat, Universit~itsstr. 150, D-4630 Bochum, F.R.G. 0034-5687/87/$03.50 © 1987 Elsevier Science Publishers B.V. (Biomedical Division)
2
H. SHAMS AND P. SCHEID
et al., 1968 ; Sutton et al., 1979). On the other hand, hypocapnia may compensate in part the metabolic acidosis expected to occur at deep hypoxia. Thirdly, hypocapnia may reduce brain blood flow, which has been shown in mammals (Vogel et al., 1969; Grote et aL, 1981 ; Nolan and Davies, 1982), but does not appear to occur in the duck, at least not down to arterial Pco2 levels of about 8 Torr (Grubb et al., 1977). During flight, on the other hand, CO2 production may in part compensate for hypoxia-induced hypocapnia, and it would be of interest to study hypoxic effects also when inspired CO 2 counteract hypocapnia. We have measured respiratory and cardiac parameters and blood gases in unrestrained, unanesthetized ducks, at varied levels of hypoxia with or without CO 2 inhalation, in an attempt to identify the role of the respiratory system in the hypoxia tolerance of an avian species, and the effect blood hypocapnia exerts on this tolerance.
Methods
Animals and preparations. Experiments were conducted in eleven adult Pekin ducks, average body weight of 2.77 kg (SD = 0.45). Under local anesthesia, a catheter was inserted into the right brachial artery for sampling arterial blood and for monitoring blood pressure (Statham, type P23D6). Using pressure control, a second catheter was placed through the right cutaneous ulnar vein into the right ventricle for collection of mixed venous blood. A thermocouple was advanced through a third catheter, implanted into the left cutaneous ulnar vein, close to the heart to monitor body temperature. Measuring techniques. Experiments were performed with the unanesthetized bird placed in a lucite chamber (volume, 31 L), which allowed it to stand and move. Vascular catheters exited the chamber via air-tight outlets at the top. The chamber was ventilated with gas mixtures at a rate high enough for the difference in Po2 and Pco2 between the inflowing and chamber gas to be less than 2 Torr. Pure gases were used to produce the gas mixtures using flow-meters, the exact composition being measured by a calibrated mass spectrometer. To measure ventilation, the chamber was sealed by closing gas inflow and outflow ports and the fluctuations in chamber pressure, resulting from changes in temperature and humidity of the tidal air, was recorded using a pressure transducer (Gould Godart BV 17212 Bilthoven, The Netherlands). The method (Drorbaugh and Fenn, 1955) requires measurement of chamber and body temperature as well as chamber humidity (Feuchte-Sensor, Valvo, Hamburg, F.R.G.). For calibration, cyclic changes in chamber volume were produced by a piston pump, with stroke volume of 3-9 ml and frequency of 250 rain 1, and the resulting pressure changes were recorded. The amplitude of the pressure changes were similar to those produced by the tidal breathing of the animal. This calibration was repeatedly performed during the experiment, with the animal inside the chamber. The frequency was sizeably higher than the highest respiratory frequency of the animal so that the calibration could easily be analyzed even though it was
RESPIRATION DURING HYPOXIA IN BIRDS
3
superimposed on the respiratory pressure waves. The frequency response of the calibration device was checked without the animal inside the box. The entire calibration was checked in test experiments with tracheotomized birds by comparing the tidal volume obtained from the box pressure with that obtained by pneumotachography. Agreement was better than 5 ~o. For measurement of gas exchange rates, lVlo~ and l~lco 2, the chamber was sealed for about 5-10 min and the changes in Po2 and Pco2 were recorded by mass spectrometry (see below). The changes in partial pressure were less than 5 Torr during this period. Partial pressures o f O 2 and CO 2 in the chamber were measured by a respiratory mass spectrometer (Centronic Q806, Croydon, England), calibrated before and after each measurement with gases provided by precision gas mixing pumps (Type M 300/a-F, WOsthoff, Bochum, F.R.G.). Partial pressures of 0 2 and CO 2 as well as pH in arterial (Pao2, P a c o ~, pHa) and mixed venous blood samples (Pvo~, PVco~, p H i ) were measured immediately after blood collection with conventional blood electrodes (Radiometer, B MS 3, Copenhagen, Denmark) at 41.5 ° C. Differences in temperature between the animal and the electrode were corrected using correction factors of Severinghaus (1964). Blood content of O2 was analyzed (Lex-O2-Con, Lexington Instruments, Waltham, MA, U.S.A.) and was used to calculate cardiac output by the direct Fick principle. Hematocrit was measured with a microcentrifuge (Compur M 1100, MOnchen, F.R.G.). One side of the plethysmograph facing the experimental equipment was darkened, and the animal was monitored throughout the experiment by a video camera.
Experimental protocol. Several hours after insertion of the catheters, the bird was placed into the recording chamber, and one hour was allowed for the bird to habituate to the experimental situation, during which air was passed through the chamber. Measurements were made in distinct measuring periods, each of which started with interrupting chamber ventilation for 30 sec to record the breathing-induced fluctuations in chamber pressure. Then, ventilation of the chamber was resumed, and arterial and mixed venous blood samples were slowly collected for analysis of blood gases and pH. Thereafter, chamber ventilation was again stopped and the chamber sealed for about 5-10 min to record Po2.and Pco2 in the chamber gas for calculation o f O 2 uptake, IVlo2, and CO 2 output rates, M o o ~. Then, the chamber was ventilated with a new gas mixture, and 40 rain were allowed after reaching a new steady level of Po2 and Pco2 in the chamber gas before a new measuring period was started. Five levels of fractional concentration of 0 2 in the gas ventilating the chamber, FIo2, were applied which were in sequence 0.21; 0.09; 0.07; 0.05; 0.04. In five ducks, there was no CO 2 in this gas, FIco ~ = 0; in the remaining six birds, F l c o 2 = 0.037. A given level of hypoxia was taken as being tolerated if the measuring period could be completed without signs of deterioration of the bird, which were a sharp fall in arterial blood pressure and/or excitement. Indeed these signs of deterioration were typically observed in birds shortly before death.
4
H. SHAMS AND P. SCHEID
Data analysis. Tidal volume (VT) was calculated according to the formula described by Drorbaugh and Fenn (1955), and "V'Ewas calculated as VT'fresp" 02 and CO 2 exchange rates, IVIo2 and l~Ico ~, were calculated from chamber volume and the rate of change in Po2 and Pco~ when the chamber was sealed. The paired t-test was used for statistical analyses. Differences between the mean values at the same PIo~ but different PIco~ were accepted to be significant at the 5 ~o level.
Results
The main results are reported in tables I and 2. The hypoxia level is presented in these tables as the equivalent altitude which would yield the same Po~ in moist inspired gas at the animal's body temperature. Ventilation. As shown in table 1, decreasing PIoz caused ventilation (VE) to increase, due to increases in both tidal volume (VT) and respiratory frequency (fresp)" These effects of hypoxia are similar without and with inspired CO2. At any level of PIo:, inspiring CO2 increased VE, by increasing VT with no significant effect on fresp" Hypoxia tolerance. The lowest level tolerated by the five ducks subjected to normocapnic hypoxia was PIo~ = 30 Torr, while the hypercapnic animals did not reach this low level, tolerating at most 36 Torr PIQ. Metabolism. 0 2 uptake (I~lo2) and C O 2 output rates (/VIco:) did not significantly vary with PIo2, but below PIo~ of 50 Torr, these rates appear to fall, concomitant with a drop in body temperature (Tb). Effects of inspired CO2 are not consistent, and differences to corresponding values without CO 2 are not statistically significant. Blood gases. Blood parameters are listed in table 2. Arterial (Pao2) and mixed venous Po2 (Pvo2) diminish with decreasing Plo2. At any PIo2 level, Pao2 during CO2 inhalation exceeds the value with no CO 2 added. However, this difference becomes insignificant at the deepest hypoxia, when the (PI-Pa)o2 difference fall to levels of about 5 Torr (fig. 1). Content of 02 in arterial (Cao2) and mixed venous blood (CVo2) diminish with falling Plo2, both without or with added CO 2. At any given PIo2 down to about 50 Torr, both Cao~ and CVo~ during COz inhalation exceed the respective levels with no CO2 load. This, however, is reversed below PIo~ of 50 Torr, when a significantly higher Cao: level is attained without added CO 2. In fact, the Cao~ levels at the respective tolerance limits of PIo~, 30 Torr without and 36 Torr with added CO2, are indistinguishable. Arterial (Paco~) and mixed venous Pco~ (Pvco~) without CO 2 inhalation drop markedly with diminishing PIo~, reaching levels which average 6 and 9 Torr, respectively (table 2). Elevation of PIco ~ to 26 Torr yields slightly elevated Paco ~ and Pvco 2 at normoxia and maintains their levels close to normal during hypoxia. That blood Pco~
1.57 0.17
0.78 0.04
0.869 0.093
l~Ico 2 (mmol. min - i )
R
0 (L.min
41.5 0.2
41.1 0.2
178 24
5.4 0.6
0.915 0.073
0.87 0.04
1.51 0.02
1.76 0.08
1.668 0.168
26.7 4.5
66 8
6,310 63.2
40.5 0.1
192 16
6.4 0.6
1.201 0.071
0.82 0.03
1.53 0.07
1.88 0.12
2.291 0.287
39.3 8.8
64 10
8,050 48.8
39.7 0.2
194 22
6.4 1.7
0.279
1.179
0.89 0.06
1.19 0.16
1.40 0.25
4.039 0.303
45.0 4.7
93 9
10,230 36.0
* Significantly different from corresponding value without CO2 at the 5% level.
Tb (:C)
168 10
f~,t (rain
i)
5.2 0.6
Qs (ml)
~)
2,01 0.16
~)
0.956 0.069
VE (L'min
.~Io2 (mmol.min
15.5 1.1
fre~p (rain l)
t)
65 7
0 145.0
VT (ml)
'Altitude" (m): Plo2 (Torr):
Without CO 2 (Plco 2 ca, 2 Torr)
38.8 0.3
219 39
6.2 0.9
1.114 0.035
0.98 0.01
1.24 0.01
1.26 0.01
5.594 1.515
46,6 4.2
123 35
11,580 30.1
41,4 0.2
165 5
4.4 0.4
0,733 0.066
0.80 0.03
1.34 0.08
1.69 0.14
1.758" 0.170
15,8 0.9
111" 8
0 144.7
41.2 0.2
170 10
5.4 0.5
0.919 0.108
0.83 0.05
1.70 0.13
2.10 0.23
3.720* 0.461
24.9 1.8
149' 11
6,310 63.3
36.0 3.4
187* 18
8,050 50.6
40.5 0.3
184 12
5.0 0,5
0,981 0.130
0.80 0.03
1.67 0.15
2.07 0.18
6.523* 0.529
With CO 2 (P1co 2 ca. 26 Torr)
39.0* 0.3
277* 34
4.1 0,5
0,813 0.089
0.76 0.04
0.94 0.19
1.24 0.25
7.289* 0.807
43.4 4.0
172" 20
10,230 35.6
TABLE 1 Cardiorepiratory parameters in ducks exposed to different levels ofhypoxia without or with addition of CO 2. Mean values and SE. Pl02, inspired P02 at the animal's body condition. 'Altitude', equivalent altitude at which Px02 would be identical to the experimental PI02.
x
,.< ©
~J
z
3.92 0.20 1.96 0.14 2.8 0.5 18.3 1.4 23.3 1.7 7.602 0.006 7.550 0.007
7.23 0.18
4.84 0.31
1.9 0.4
30.0 0.6
34.0 1.0
7.456 0.009
7.433 0.006
20.5 0.3
1.8 0.5
t)
Cao: (retool - L
CVo: (mmol - L
PlcQ (Torr)
Paco ~ (Torr)
P f c o _. (Torr)
pHa
pHv
[HCO.C]rl ( m m o l ' l ,
[Lactate]p~ (mmol. L J) 4.9 0.4
17.5 1.1
30.4 1.1
56.3 1.9
PVo~ (Torr)
7.1 0.8
13.9 1.3
7.577 0.015
7.624 0.017
17.6 1.5
13.7 1.1
1.9 0.6
1.44 0.11
3.03 0.21
24.4 0.7
33.6 1.8
13.2 2.4
5.0 1.7
7.419 O.123
7.459 0.111
10.4 0.7
7.0 0.9
1.9 0.4
0.79 0.21
2.05 0.31
15.7 2.1
28.7 1.3
17.4 1.8
2.7 1.6
7.238 0.280
7.289 0.260
8.7 0.3
5.7 0.9
1.5 0.6
0.49 0.09
1.40 0.30
13.1 1.4
25.2 0.6
0.3* 0.1
21.5" 0.3
7.408* 0.008
7.434" 0.008
36.2* 0.6
33.1" 0.9
26.0* 0.2
5.06 0.25
7.40 0.30
58.8 1.4
110.1" 1.7
144.7 1.5
42.2 1.8
30.l 0.6
95.0 1.3
36.0 0.4
Pao~ (Torr)
48£ 0.6
63.2 0.3
145.0 0.6
11,580
Plo2 (Torr)
1(I,230
0
8,050
6,310
0
Altitude (m):
With C O 2
Without CO 2
0.6* 0.2
20.2* 0.5
7.481" 0.006
7.504* 0.009
29.6* 0.4
26.3* 0.7
26.1" 0.3
2.91" 0.22
5.22* 0.27
37.6* 0.9
53.2* 1.1
63.3 0.3
6,310
7.126" 0.051 7.103" 0.072
7.509* 0.006 7.480* 0.010
16.4 0.9
28,3* 0.4
28.2* 1.3
2.3* 0.6
26,1" 0.8
26.2* 0.6
9.8* 1.9
26.1" 0.5
26.1" 0.9
19.7" 0.5
0.59 0.13
14.5 1.0 28.8* 0.9
1.69 0.22
29.2 1.0 42.0* 0.6
1.22* 0.18
35.6 0.2
50,6 0.3
3.88* 0.21
10,230
8,050
TABLE 2 Blood parameters in ducks exposed to different levels of hypoxia without or with addition of CO2. Mean values and SE.
d3 a:
> Z ~J
-e >
:=
7
RESPIRATION DURING HYPOXIA IN BIRDS
(PI - PQ}02
'~
(Po- PI)co~
{Tort )
{Torr)
O O
20
"
10
.
~
~
B
~° 0
40
8O
PIo2
120 (Torr)
0
&O
80
Pro2
120 (Torrl
Fig. 1. Mean values (and SE) of inspired-to-arterial Po2 differences, (Pl - Pa)o2, and arterial-to-inspired Pco: differences, (Pa - PDco2- without (O) and with ( 0 ) inspired CO2.
remains constant at all hypoxic levels, despite increasing ~'E, is explained by the fact that Paco 2 nearly equilibrates to Pxco 2 during hypoxia (table 2 and fig. 1).
Acid-base status. Using measured plasma pH (pHp0 and Pco2 values, bicarbonate concentration, [ H C O 3 - ] p l , was calculated for arterial plasma. For use of a [ H C O 3 3]pl/pHpl diagram, all values were corrected to one temperature (40 °C), and values of pK' (6.082) and CO2 solubility (0.0289 m m o l - L 1.Torr 1) obtained for human plasma at 40 °C (Severinghaus, 1964) were used in calculating [HCO3- Ivy The corresponding mean values are listed in table 2 and plotted in fig. 2. Without CO 2 inhaled, hypoxia (PIo2 = 65 Torr) induces changes in HCO 3 and pH that follow the in vitro buffer line of the duck (Scheipers et al., 1975). Below Pio2 of 65 Torr, the respiratory alkalosis is combined with a metabolic acidosis, whereby, at Plo: = 35 Torr, the normal pHpl is reattained, at a low [ H C O f ]pl of below 5 mmol. L - ~. At the lowest PIo2, 30 Torr, metabolic acidosis supersides, pHpl falling to about 7.26. Compared with these changes, inspiring CO2 leads to somewhat more acidic pH during normoxia, but with hypoxia, HCO3- and pH are kept closer to their control levels, at least at or above PIo2 = 50 Torr, with nearly no metabolic component. In contrast, at PIo2 = 35 Torr, there is a large metabolic change, and pH falls to 7.10, which is distinctly lower than without CO2 inhalation. The metabolic changes are evidenced by the levels of lactate in arterial plasma (table 2) which increase with hypoxia. Without CO2 inhaled, this increase starts at the first level of hypoxia, while during CO 2 inhalation a significant increase is observed only at the lowest PIo2. Hematocrit did not significantly vary during the experiment, the average being 36% (SD = 3~o).
8
H. SHAMS AND P. SCHEID 30
35
[HCO~] /
(mmo[.L 1} p~ 2C3rnmolL'pH' 20
..
~
1~..
~ ~
PCO:
~--~ ._-~-~
/
/ " P[?O
25
,o-
/"
15
26 Torr/.
./ ~ ) ;>h
35~
P'Ico
0 -9r-
35 ~" 0
I
i
[{I] L~
720
: [
730
PHpt I
-
240
750
760
270
Fig. 2. Plot of plasma H C O 3 - concentration [ H C O 3 - ] e l , against plasma pH, prim, for mean values without (©) and with (0) inspired CO2. Experimental data corrected to 40 °C. Numbers at each symbol represent the approximatelevels of P]o2. Curved lines CO2 isopleths; straight line, whole blood bufferline for the duck, with buffer value of/~ = 20.0 mmol. L 1. pH ~ (Scheipers et aL, 1975).
Circulation. The (Ca-C~)o2 differences in table 2 and 1~1o2 values of table l allow calculation of cardiac output ((~) which is listed in table 1. Both without and with CO 2 inhalation, Q increases slightly with diminishing PIo2 , but tends to drop again at deep hypoxia when CO2 is inhaled. This drop is due to a hypoxic reduction in stroke volume (Qs) while cardiac frequency (fcard) rises monotonously (table 1).
Discussion Critique of methods. An important part in this study is the plethysmograph chamber which allows measurement of ventilation in the unanesthetized, virtually unrestrained animal. Problems with this technique derive from pressure drifts, induced by temperature variations. A small leak of the chamber eliminated this drift without the need for a reference chamber (Drorbaugh and Fenn, 1955). The error introduced by this leak in the measurement of tidal volume in the range of respiratory frequencies was less than 0.1%. Calibration with a piston pump at high frequency was mainly responsible for the good agreement between the VT estimates and the direct pneumotachograph measurements. Another problem with the technique is due to the fact that Po2 drops and Pco2 increases while ventilation or exchange rates are measured in the sealed box. The time of sealing the chamber for these measurements was, therefore, kept short enough for partial pressures to change less than 1 Tort when measuring ventilation and less than 5 Tort for measurement of exchange rates. We believe that the disadvantage resulting from sealing the chamber is outweighed by the advantage of measuring in an animal without tracheotomy or face mask.
RESPIRATION DURING HYPOXIA IN BIRDS
9
The tolerance limit observed by Black and Tenney (1980) in the Pekin duck, 11,580 m, is the same as that found by us. The increase in "~E with hypoxia, which is similar to that observed in the Pekin duck (Black and Tenney, 1980), leads to a steep decrease in Paco2, down to 6 Torr at the lowest Plo2 tolerated, 30 Torr. Similarly low levels of Paco 2 in extreme hypoxia have been reported for the Pekin duck and bar-headed goose (Black and Tenney, 1980; Faraci et al., 1984). This extreme hyperventilation exerts effects on the acid-base balance. As shown in fig. 2, mild hypoxia results in respiratory alkalosis which, as hypoxia progresses, is accompanied by a metabolic, i.e. lactic acidosis. Thus, the extent of blood alkalosis is limited, and at the deepest hypoxic level there is a distinct acidosis despite the very low level of arterial Pco2. While Black and Tenney (1980) found similar trends both in the Pekin duck and bar-headed goose, Faraci et al. (1984) reported a slightly alkalotic arterial pH at Pao2 of 28.5 Torr and Paco: of 6.4 Torr in the duck and in the majority of their geese. Respiratory alkalosis was also reported for the pigeon acutely exposed to an altitude of up to 9000 m (Lutz and Schmidt-Nielsen, 1977; Weinstein et al., 1985). This hyperventilation is clearly advantageous for the animal as it maximizes 02 loading of blood in the lung (Dejours, 1982; West, 1983). The efficiency of oxygenation in the lung is shown by the difference in (PI-Pa)o: which becomes progressively smaller with hypoxia, reaching a value of only 5 Torr at PIo2 of 30 Torr. Results similar to these have previously been reported for the Pekin duck and the bar-headed goose (Black and Tenney, 1980; Faraci et al., 1984) and somewhat higher (PI - Pa)o 2 values for the pigeon (Lutz and Schmidt-Nielsen, 1977; Weinstein et al., 1985). This small (PI - Pa)o 2 difference is not primarily due to the peculiarly high gas exchange efficiency of the avian lung, but appears to be related to the exquisitely high parabronchial ventilation relative to 02 uptake. In fact, with such high ventilation, the Po~ drop gas as flows through the parabronchus will become minimal, and will approach the inspired level, whereby the gas exchange systems of the parabronchial and alveolar lungs approach each other (Scheid, 1979). Cardiac output in our animals did not increase nearly as much as in the duck and goose of Black and Tenney (1980). Their values for the duck started at lower, and reached higher, levels than in our study. Correspondingly, for our ducks the "coefficient of oxygen delivery", Cao2/(Ca - C~)o2, is higher in normoxia, but reaches similar values in deep hypoxia. It appears that the stroke volume limits further rise in (~ in hypoxia in our ducks, whereas cardiac frequency shows a monotonous increase with progressing hypoxia. Tachycardia in the spontaneously breathing hypocapnic and hypoxic duck has also been reported by Butler (1970) and Butler and Taylor (1973), and, like in our study, the response was more pronounced during hypercapnic hypoxia. When, on the other hand, hypercapnic hypoxia develops during apnea, as in diving, bradycardia is observed (Butler and Taylor, 1973, 1983). The drop in Mo~ and l~lco~ with deep hypoxia, which has also been observed by Butler (1970), may in part be caused by anaerobiosis in some tissues, but may also reflect the observed drop in body temperature, most likely a result of the extreme hyperEffects ofhypoxia without CO 2 inhalation.
10
H. SHAMS AND P. SCHEID
ventilation. Assuming a Qm value of 2.3 for the metabolic rate, the observed drop in body temperature by 2.5 °C would lead to a reduction in IVlo~by about 20~o, which is nearly the reduction observed. Mammals acutely exposed to hypoxia hyperventilate (Vogel et al., 1969; Nesarajah et aL, 1983; Sidi etal., 1983) but the levels of Pa¢o~ reached appear to be distinctly higher. A notable exception is the singular report of a mountaineer on the top of Mt. Everest who measured alveolar Pco~ as low as 7.5 Torr when breathing ambient air (West, 1983). This value is, however, worth corroborating, with particular attention to the problem of steady state. It seems that arterial hypocapnia is not well tolerated in mammals (Dejours, 1982), the tolerance being much less than in birds. One difference between mammals and birds that might contribute to the higher hypoxia tolerance of birds is their lack of hypocapnic vasoconstriction in many vascular beds, including the brain (Grubb etal., 1977; Wolfenson et al., 1982). In mammals, hypocapnia leads to distinct decrease in brain blood flow (Grote et aL, 1981). Effects of C02 on cardio-respiratory responses to hypoxia. Adding CO 2 to the inhaled air increased the ventilation at all levels of hypoxia, which is in line with the reported respiratory drive of CO2 in birds (cf Scheid and Piiper, 1986) and in mammals. As in normoxia, this drive is mainly on tidal volume, respiratory frequency being unaffected. It is reasonable to expect higher Po2 and Co2 in arterial blood as a result of the CO2-induced stimulation of ventilation, and this is indeed observed with moderate levels of hypoxia, i.e. PIo2 at or above 50 Torr (table 2). Below PIo~ of 50 Torr, however, there is no advantageous effect from the CO 2 drive on ventilation. In fact, with CO 2 inhalation, the (PI - Pa)o~ difference (fig. 1) is not further reduced, indicating that ventilation does not significantly limit 0 2 exchange. In fact, the levels of Pao2 in deep hypoxia are similar without or with CO2. Despite similar Pao2, Cao~ is lower with COg inhalation in this condition, and this is explained by the lower arterial pH which reduces blood affinity by virtue of the Bohr effect. This lower pH during CO2 inhalation is a result mainly of the higher Paco ~ than with CO2-free breathing at comparable hypoxic levels, standard bicarbonate being less reduced with CO2 breathing than without (fig. 2). The effect of CO 2 can thus be summarized to improve arterial oxygenation by increasing ventilation, and concomitantly increasing Pao2 , at low or moderate levels of hypoxia (PIo2 at or above 50 Torr). Below this level ofhypoxia, however, the ventilatory stimulation of CO 2 is without effect on Pao2, but the lower blood pH reduces 0 2 loading to hemoglobin. That this is likely to limit hypoxia tolerance is evident from the ensuing low mixed venous Co~ which is in fact similar at the tolerance limits without or with CO 2 inhalation. That CO2 inhalation at Plo~ of 36 Torr has other disadvantageous effects on the 02 transport is shown by the decreased cardiac output and stroke volume. In man, administration of carbonic anhydrase blockers in acute altitude exposure exerts beneficial effects on symptoms of acute mountain sickness (Forwand et al., 1968; Sutton etal., 1979). That this may be related to the prevention of hypocapnia is
RESPIRATION DURING HYPOXIA IN BIRDS
11
suggested from experiments, in which C O 2 administration during acute altitude exposure has been found to induce a sense of well-being (of. Luft, 1965). Aside from preventing hypocapnia, C O 2 addition at a given altitude stimulates ventilation and thus results in improved blood oxygenation. That the advantageous effect of C O 2 breathing is mainly due to better oxygen loading in the lung is suggested by experiments of Maher et al. (1975) who compared identical levels of arterial Po2 and found that C O 2 breathing aggravated the symptoms of acute mountain sickness. Other, partly advantageous effects o f C O 2 were also observed, e.g. maintained stroke volume (Grover et al., 1976); inhibition of forearm venoconstriction (Weil et al., 1971; Cruz et al., 1976) in man; increased coronary blood flow with increased systemic, splanchnic and renal vasoconstriction in the dog (Koehler et al., 1980). It is thus not easy to decide whether C O 2 breathing in acute altitude exposure in mammals is on balance advantageous or not. Experiments with progressive hypoxia, extending to severe levels, appear to be needed in mammals. Conclusion. One factor that may be of importance for the pronounced high altitude tolerance of birds compared with mammals appears to be the increase in ventilation, that is not impeded by the ensuing extremely low arterial Pco2 and allows arterial blood to nearly equilibrate with inspired Po2- The basis for this difference between birds and mammals relates probably to the hypoxia and hypocapnia sensitivity of the respiratory centers, and is not primarily a result o f the high gas exchange efficiency of the avian parabronchial lung. Addition of CO2 to the inspired gas is advantageous during mild hypoxia, when the effect on stimulating ventilation results in increased blood oxygenation. During extreme hypoxia, however, blood oxygenation is not further improved by addition of C O 2, and adverse effects, mainly due to the uncompensated metabolic acidosis and its effects on blood 0 2 transport, become limiting.
Acknowledgement.The skilled technical assistance of Mrs. R. Hildebrandt is gratefully acknowledged. References Black, C.P. and S.M. Tenney (1980). Oxygen transport during progressive hypoxia in high-altitude and sea-level waterfowl. Respir. Physiol. 39: 217-239. Butler, P.J. (1970). The effect of progressive hypoxia on the respiratory and cardiovascular systems of the pigeon and duck. J. Physiol. (London) 211: 527-538. Butler, P.J. and E.W. Taylor (1973). The effect of hypercapnic hypoxia, accompanied by different levels of lung ventilation, on heart rate in the duck. Respir. Physiol. 19: 176-187. Butler, P.J. and E.W. Taylor (1983). Factors affecting the respiratory and cardiovascular responses to hypercapnic hypoxia, in mallard ducks. Respir. Physiol. 53: 109-127. Cruz, J.C., R.F. Grover, J.T. Reeves, J.T. Maher, A. Cymerman and J,C. Denniston (1976). Sustained venoconstriction in man supplemented with CO2 at high altitude. J. Appl. Physiol. 40: 96-100. Dejours, P. (1982). Mount Everest and beyond: breathing air. In: A Companion to Animal Physiology, edited by C. R. Taylor, K. Johansen and L. Bolis. Cambridge, UK, Cambridge Univ. Press, pp. 17-30. Drorbaugh, J. E. and W. O. Fenn (1955). A barometric method for measuring ventilation in newborn infants. Pediatrics 16: 81-87.
12
H. SHAMS AND P. SCHEID
Faraci, F.M., D.L. Kilgore, Jr. and M.R. Fedde (1984). Oxygen delivery to the heart and brain during hypoxia: Pekin duck vs. bar-headed goose. Am. J. Physiol. 247: R69-R75. Forwand, S.A., M. Londowne, M.J.N. Follansbee and C.J.E. Hansen (1968). Effect of acetazolamide on acute mountain sickness. N. Engl. J. Med. 279: 839-845. Grote, J., K. Zimmer and R. Schubert (1981). Effects of severe arterial hypocapnia on regional blood flow regulation, tissue Po2 and metabolism in the brain cortex of cats. Pfliigers Arch. 391: 195-199. Grover, R.F., J.T. Reeves, J.T. Maher, R.E. McCullough, J.C. Cruz, J.C. Denniston and A. Cymerman (1976). Maintained stroke volume but impaired arterial oxygenation in man at high altitude with supplemental CO 2. Circ. Res. 38: 391-396. Grubb, B., C. D. Mills, J.M. Colacino and K. Schmidt-Nielsen (1977). Effect of arterial carbon dioxide on cerebral blood flow in ducks. Am. J. Physiol. 1: H596-H601. Koehler, R.C., B.W. McDonald and J.A. Krasney (1980). Influence of CO 2 on cardiovascular response to hypoxia in conscious dogs. Am. J. Physiol. 239: H545-H558. Luft, U.C. (1965). Aviation physiology - the effects of altitude. In: Handbook of Physiology. Section 3, Respiration, Vol. II, edited by W.O. Fenn and H. Rahn. Washington, D.C., American Physiological Society, 1099 pp. Lutz, P.L. and K.S. Schmidt-Nielsen (1977). Effect of simulated altitude on blood gas transport in the pigeon. Respir. Physiol. 30: 383-388. Maher, J.T., A. Cymerman, J.T. Reeves, J.C. Cruz, J.C. Denniston and R.F. Grover (1975). Acute mountain sickness: Increased severity in eucapnic hypoxia. Aviat. Space Environ. Med. 46: 826-829. Nesarajah, M.S., S. Matalon, J.A. Krasney and L.E. Farhi (1983). Cardiac output and regional oxygen transport in the acutely hypoxic conscious sheep. Respir. Physiol. 53: 161-172. Nolan, W.F. and D.G. Davies (1982). Brain extracellular fluid pH and blood flow during isocapnic and hypocapnic hypoxia. J. Appl. Physiol. 53: 247-252. Scheid, P. (1979). Mechanisms of gas exchange in bird lungs. Rev. Physiol. Biochem. Pharmacol. 86:137-186. Scheid, P. and J. Piiper (1986). Control of breathing in birds. In: Handbook of Physiology. Section 3. The Respiratory System. Vol. II. Control of breathing. Part 2, edited by A. P, Fishman, N.S. Cherniack, J.G. Widdicombe and S.R. Geiger. Bethesda, MD. American Physiological Society, pp. 815-832. Scheipers, G., T. Kawashiro and P. Scheid (1975). Oxygen and carbon dioxide dissociation of duck blood. Respir. Physiol. 24: 1-13. Severinghaus, J.W. (1964). Blood gas concentrations. In: Handbook of Physiology. Section 3. Respiration. Vol. If, edited by W.O. Fenn and H. Rahn. Washington, D.C., American Physiological Society, pp. 1475-1487. Sidi, D., J. R.G. Kuipers, D. Teitel, M.A. Heymann and A.M. Rudolph (1983). Development changes in oxygenation and circulatory responses to hypoxemia in lambs. Am. J. Physiol. 245: H674-H682. Sutton, J. R., C. S. Houston, A. L. Mansell, M. D. McFadden, P. M. Hackett, J. R. A. Rigg and A. C. P. Powlcs 1979). Effect of acetazolamide on hypoxemia during sleep at high altitude. N. Engl. J. Med. 301: 1329-1331. Tucker, V.A. (1968). Respiratory physiology of house sparrows in relation to high altitude flight. J. Exp. Biol. 38: 55-66. Vogel, J.A., R.I. Pulver and T.M. Burton (1969). Regional blood flow distribution during simulated high-altitude exposure. Fed. Proc. 28:1155-1159. Weil, J.V., E. Byrne-Quinn, D.J. Battock, R.F. Grover and C. Chidsey (1971). Forearm circulation at high altitude. Clin. Sci. 40: 235-246. Weinstein, Y., M.H. Bernstein, P.E. Bickler, D. V. Gonzales, F.C. Samaniego and M.A. Escobedo (1985). Blood respiratory properties in pigeons at high altitudes: effects of acclimation. Am. J. Physiol. 249: R765-R775. West, J. B. (1983). Climbing Mt. Everest without oxygen: an analysis of maximal exercise during extreme hypoxia. Respir. Physiol. 52: 265-279. Wolfenson, D., Y.F. Frei and A. Berman (1982). Blood flow distribution during artificially induced respiratory hypocapnic alkalosis in the fowl. Respir. Physiol. 50: 87-92.
1
Elsevier
RSP 01228
Respiration and blood gases in the duck exposed to normocapnic and hypercapnic hypoxia H. Shams and P. Scheid Institut J~r Physiologie, Ruhr-Universitiit Bochum, F.R.G.
(Accepted for publication 30 August 1986)
Abstract. Cardio-respiratory parameters were measured in the unrestrained, unanesthetized duck by whole body plethysmography during exposure to varied levels of inspired hypoxia without or with (3.7 ~o) CO 2. At any level of inspired Po~ (Plo2), ventilation (~/E) was larger with CO 2 inhalation, leading, for PIo2 > 50 Torr, to higher levels of arterial Po2 (Pao2) and 0 2 content (Cao2). Below Pto2 of 50 Torr, the (PI-Pa)Oz difference without CO2 reached a value as low as 5 Torr which was not diminished by further stimulation of ~'E by inhaled CO 2. Without CO 2 inhalation at this deep hypoxic level the ensuing hypoxiainduced respiratory alkalosis was partly compensated by lactacidosis, whereas CO 2 inhalation resulted in markedly lower blood pH leading to significantly lower arterial and venous 02 content (Bohr effect). As a result, the deepest level ofhypoxia tolerated without CO 2 inhalation, 30 Torr, is significantly deeper than that, 36 Torr, tolerated when CO 2 is inhaled. The data suggest that a number of factors contribute to the high hypoxia tolerance in birds, e.g. the effectiveness of parabronchial ventilation and the tolerance of low arterial Pco2 levels, whereby part of the lactacidosis is compensated.
Birds; Blood gases; Hypercapnia; Hypoxia; Respiration.
Birds are generally considered to be more tolerant of hypoxia than mammals (Tucker, 1968), although notable species differences among birds appear to exist (Butler, 1970). These differences between birds and mammals may be the result of a number of structural and functional differences, of which the high inherent gas exchange efficiency of the parabronchial lung (Scheid, 1979) is one example. Of particular interest to us was the possible role of arterial hypocapnia which occurs in animals as a result of the hypoxic ventilatory stimulation and may affect the hypoxia tolerance in many different ways. For one, it is expected to remove part of the hypoxic ventilatory drive, thus limiting arterial oxygenation, as has been shown in man (Forwand Correspondence address: Hashim Shams, Institut for Physiologie, Ruhr-Universitat, Universit~itsstr. 150, D-4630 Bochum, F.R.G. 0034-5687/87/$03.50 © 1987 Elsevier Science Publishers B.V. (Biomedical Division)
2
H. SHAMS AND P. SCHEID
et al., 1968 ; Sutton et al., 1979). On the other hand, hypocapnia may compensate in part the metabolic acidosis expected to occur at deep hypoxia. Thirdly, hypocapnia may reduce brain blood flow, which has been shown in mammals (Vogel et al., 1969; Grote et aL, 1981 ; Nolan and Davies, 1982), but does not appear to occur in the duck, at least not down to arterial Pco2 levels of about 8 Torr (Grubb et al., 1977). During flight, on the other hand, CO2 production may in part compensate for hypoxia-induced hypocapnia, and it would be of interest to study hypoxic effects also when inspired CO 2 counteract hypocapnia. We have measured respiratory and cardiac parameters and blood gases in unrestrained, unanesthetized ducks, at varied levels of hypoxia with or without CO 2 inhalation, in an attempt to identify the role of the respiratory system in the hypoxia tolerance of an avian species, and the effect blood hypocapnia exerts on this tolerance.
Methods
Animals and preparations. Experiments were conducted in eleven adult Pekin ducks, average body weight of 2.77 kg (SD = 0.45). Under local anesthesia, a catheter was inserted into the right brachial artery for sampling arterial blood and for monitoring blood pressure (Statham, type P23D6). Using pressure control, a second catheter was placed through the right cutaneous ulnar vein into the right ventricle for collection of mixed venous blood. A thermocouple was advanced through a third catheter, implanted into the left cutaneous ulnar vein, close to the heart to monitor body temperature. Measuring techniques. Experiments were performed with the unanesthetized bird placed in a lucite chamber (volume, 31 L), which allowed it to stand and move. Vascular catheters exited the chamber via air-tight outlets at the top. The chamber was ventilated with gas mixtures at a rate high enough for the difference in Po2 and Pco2 between the inflowing and chamber gas to be less than 2 Torr. Pure gases were used to produce the gas mixtures using flow-meters, the exact composition being measured by a calibrated mass spectrometer. To measure ventilation, the chamber was sealed by closing gas inflow and outflow ports and the fluctuations in chamber pressure, resulting from changes in temperature and humidity of the tidal air, was recorded using a pressure transducer (Gould Godart BV 17212 Bilthoven, The Netherlands). The method (Drorbaugh and Fenn, 1955) requires measurement of chamber and body temperature as well as chamber humidity (Feuchte-Sensor, Valvo, Hamburg, F.R.G.). For calibration, cyclic changes in chamber volume were produced by a piston pump, with stroke volume of 3-9 ml and frequency of 250 rain 1, and the resulting pressure changes were recorded. The amplitude of the pressure changes were similar to those produced by the tidal breathing of the animal. This calibration was repeatedly performed during the experiment, with the animal inside the chamber. The frequency was sizeably higher than the highest respiratory frequency of the animal so that the calibration could easily be analyzed even though it was
RESPIRATION DURING HYPOXIA IN BIRDS
3
superimposed on the respiratory pressure waves. The frequency response of the calibration device was checked without the animal inside the box. The entire calibration was checked in test experiments with tracheotomized birds by comparing the tidal volume obtained from the box pressure with that obtained by pneumotachography. Agreement was better than 5 ~o. For measurement of gas exchange rates, lVlo~ and l~lco 2, the chamber was sealed for about 5-10 min and the changes in Po2 and Pco2 were recorded by mass spectrometry (see below). The changes in partial pressure were less than 5 Torr during this period. Partial pressures o f O 2 and CO 2 in the chamber were measured by a respiratory mass spectrometer (Centronic Q806, Croydon, England), calibrated before and after each measurement with gases provided by precision gas mixing pumps (Type M 300/a-F, WOsthoff, Bochum, F.R.G.). Partial pressures of 0 2 and CO 2 as well as pH in arterial (Pao2, P a c o ~, pHa) and mixed venous blood samples (Pvo~, PVco~, p H i ) were measured immediately after blood collection with conventional blood electrodes (Radiometer, B MS 3, Copenhagen, Denmark) at 41.5 ° C. Differences in temperature between the animal and the electrode were corrected using correction factors of Severinghaus (1964). Blood content of O2 was analyzed (Lex-O2-Con, Lexington Instruments, Waltham, MA, U.S.A.) and was used to calculate cardiac output by the direct Fick principle. Hematocrit was measured with a microcentrifuge (Compur M 1100, MOnchen, F.R.G.). One side of the plethysmograph facing the experimental equipment was darkened, and the animal was monitored throughout the experiment by a video camera.
Experimental protocol. Several hours after insertion of the catheters, the bird was placed into the recording chamber, and one hour was allowed for the bird to habituate to the experimental situation, during which air was passed through the chamber. Measurements were made in distinct measuring periods, each of which started with interrupting chamber ventilation for 30 sec to record the breathing-induced fluctuations in chamber pressure. Then, ventilation of the chamber was resumed, and arterial and mixed venous blood samples were slowly collected for analysis of blood gases and pH. Thereafter, chamber ventilation was again stopped and the chamber sealed for about 5-10 min to record Po2.and Pco2 in the chamber gas for calculation o f O 2 uptake, IVlo2, and CO 2 output rates, M o o ~. Then, the chamber was ventilated with a new gas mixture, and 40 rain were allowed after reaching a new steady level of Po2 and Pco2 in the chamber gas before a new measuring period was started. Five levels of fractional concentration of 0 2 in the gas ventilating the chamber, FIo2, were applied which were in sequence 0.21; 0.09; 0.07; 0.05; 0.04. In five ducks, there was no CO 2 in this gas, FIco ~ = 0; in the remaining six birds, F l c o 2 = 0.037. A given level of hypoxia was taken as being tolerated if the measuring period could be completed without signs of deterioration of the bird, which were a sharp fall in arterial blood pressure and/or excitement. Indeed these signs of deterioration were typically observed in birds shortly before death.
4
H. SHAMS AND P. SCHEID
Data analysis. Tidal volume (VT) was calculated according to the formula described by Drorbaugh and Fenn (1955), and "V'Ewas calculated as VT'fresp" 02 and CO 2 exchange rates, IVIo2 and l~Ico ~, were calculated from chamber volume and the rate of change in Po2 and Pco~ when the chamber was sealed. The paired t-test was used for statistical analyses. Differences between the mean values at the same PIo~ but different PIco~ were accepted to be significant at the 5 ~o level.
Results
The main results are reported in tables I and 2. The hypoxia level is presented in these tables as the equivalent altitude which would yield the same Po~ in moist inspired gas at the animal's body temperature. Ventilation. As shown in table 1, decreasing PIoz caused ventilation (VE) to increase, due to increases in both tidal volume (VT) and respiratory frequency (fresp)" These effects of hypoxia are similar without and with inspired CO2. At any level of PIo:, inspiring CO2 increased VE, by increasing VT with no significant effect on fresp" Hypoxia tolerance. The lowest level tolerated by the five ducks subjected to normocapnic hypoxia was PIo~ = 30 Torr, while the hypercapnic animals did not reach this low level, tolerating at most 36 Torr PIQ. Metabolism. 0 2 uptake (I~lo2) and C O 2 output rates (/VIco:) did not significantly vary with PIo2, but below PIo~ of 50 Torr, these rates appear to fall, concomitant with a drop in body temperature (Tb). Effects of inspired CO2 are not consistent, and differences to corresponding values without CO 2 are not statistically significant. Blood gases. Blood parameters are listed in table 2. Arterial (Pao2) and mixed venous Po2 (Pvo2) diminish with decreasing Plo2. At any PIo2 level, Pao2 during CO2 inhalation exceeds the value with no CO 2 added. However, this difference becomes insignificant at the deepest hypoxia, when the (PI-Pa)o2 difference fall to levels of about 5 Torr (fig. 1). Content of 02 in arterial (Cao2) and mixed venous blood (CVo2) diminish with falling Plo2, both without or with added CO 2. At any given PIo2 down to about 50 Torr, both Cao~ and CVo~ during COz inhalation exceed the respective levels with no CO2 load. This, however, is reversed below PIo~ of 50 Torr, when a significantly higher Cao: level is attained without added CO 2. In fact, the Cao~ levels at the respective tolerance limits of PIo~, 30 Torr without and 36 Torr with added CO2, are indistinguishable. Arterial (Paco~) and mixed venous Pco~ (Pvco~) without CO 2 inhalation drop markedly with diminishing PIo~, reaching levels which average 6 and 9 Torr, respectively (table 2). Elevation of PIco ~ to 26 Torr yields slightly elevated Paco ~ and Pvco 2 at normoxia and maintains their levels close to normal during hypoxia. That blood Pco~
1.57 0.17
0.78 0.04
0.869 0.093
l~Ico 2 (mmol. min - i )
R
0 (L.min
41.5 0.2
41.1 0.2
178 24
5.4 0.6
0.915 0.073
0.87 0.04
1.51 0.02
1.76 0.08
1.668 0.168
26.7 4.5
66 8
6,310 63.2
40.5 0.1
192 16
6.4 0.6
1.201 0.071
0.82 0.03
1.53 0.07
1.88 0.12
2.291 0.287
39.3 8.8
64 10
8,050 48.8
39.7 0.2
194 22
6.4 1.7
0.279
1.179
0.89 0.06
1.19 0.16
1.40 0.25
4.039 0.303
45.0 4.7
93 9
10,230 36.0
* Significantly different from corresponding value without CO2 at the 5% level.
Tb (:C)
168 10
f~,t (rain
i)
5.2 0.6
Qs (ml)
~)
2,01 0.16
~)
0.956 0.069
VE (L'min
.~Io2 (mmol.min
15.5 1.1
fre~p (rain l)
t)
65 7
0 145.0
VT (ml)
'Altitude" (m): Plo2 (Torr):
Without CO 2 (Plco 2 ca, 2 Torr)
38.8 0.3
219 39
6.2 0.9
1.114 0.035
0.98 0.01
1.24 0.01
1.26 0.01
5.594 1.515
46,6 4.2
123 35
11,580 30.1
41,4 0.2
165 5
4.4 0.4
0,733 0.066
0.80 0.03
1.34 0.08
1.69 0.14
1.758" 0.170
15,8 0.9
111" 8
0 144.7
41.2 0.2
170 10
5.4 0.5
0.919 0.108
0.83 0.05
1.70 0.13
2.10 0.23
3.720* 0.461
24.9 1.8
149' 11
6,310 63.3
36.0 3.4
187* 18
8,050 50.6
40.5 0.3
184 12
5.0 0,5
0,981 0.130
0.80 0.03
1.67 0.15
2.07 0.18
6.523* 0.529
With CO 2 (P1co 2 ca. 26 Torr)
39.0* 0.3
277* 34
4.1 0,5
0,813 0.089
0.76 0.04
0.94 0.19
1.24 0.25
7.289* 0.807
43.4 4.0
172" 20
10,230 35.6
TABLE 1 Cardiorepiratory parameters in ducks exposed to different levels ofhypoxia without or with addition of CO 2. Mean values and SE. Pl02, inspired P02 at the animal's body condition. 'Altitude', equivalent altitude at which Px02 would be identical to the experimental PI02.
x
,.< ©
~J
z
3.92 0.20 1.96 0.14 2.8 0.5 18.3 1.4 23.3 1.7 7.602 0.006 7.550 0.007
7.23 0.18
4.84 0.31
1.9 0.4
30.0 0.6
34.0 1.0
7.456 0.009
7.433 0.006
20.5 0.3
1.8 0.5
t)
Cao: (retool - L
CVo: (mmol - L
PlcQ (Torr)
Paco ~ (Torr)
P f c o _. (Torr)
pHa
pHv
[HCO.C]rl ( m m o l ' l ,
[Lactate]p~ (mmol. L J) 4.9 0.4
17.5 1.1
30.4 1.1
56.3 1.9
PVo~ (Torr)
7.1 0.8
13.9 1.3
7.577 0.015
7.624 0.017
17.6 1.5
13.7 1.1
1.9 0.6
1.44 0.11
3.03 0.21
24.4 0.7
33.6 1.8
13.2 2.4
5.0 1.7
7.419 O.123
7.459 0.111
10.4 0.7
7.0 0.9
1.9 0.4
0.79 0.21
2.05 0.31
15.7 2.1
28.7 1.3
17.4 1.8
2.7 1.6
7.238 0.280
7.289 0.260
8.7 0.3
5.7 0.9
1.5 0.6
0.49 0.09
1.40 0.30
13.1 1.4
25.2 0.6
0.3* 0.1
21.5" 0.3
7.408* 0.008
7.434" 0.008
36.2* 0.6
33.1" 0.9
26.0* 0.2
5.06 0.25
7.40 0.30
58.8 1.4
110.1" 1.7
144.7 1.5
42.2 1.8
30.l 0.6
95.0 1.3
36.0 0.4
Pao~ (Torr)
48£ 0.6
63.2 0.3
145.0 0.6
11,580
Plo2 (Torr)
1(I,230
0
8,050
6,310
0
Altitude (m):
With C O 2
Without CO 2
0.6* 0.2
20.2* 0.5
7.481" 0.006
7.504* 0.009
29.6* 0.4
26.3* 0.7
26.1" 0.3
2.91" 0.22
5.22* 0.27
37.6* 0.9
53.2* 1.1
63.3 0.3
6,310
7.126" 0.051 7.103" 0.072
7.509* 0.006 7.480* 0.010
16.4 0.9
28,3* 0.4
28.2* 1.3
2.3* 0.6
26,1" 0.8
26.2* 0.6
9.8* 1.9
26.1" 0.5
26.1" 0.9
19.7" 0.5
0.59 0.13
14.5 1.0 28.8* 0.9
1.69 0.22
29.2 1.0 42.0* 0.6
1.22* 0.18
35.6 0.2
50,6 0.3
3.88* 0.21
10,230
8,050
TABLE 2 Blood parameters in ducks exposed to different levels of hypoxia without or with addition of CO2. Mean values and SE.
d3 a:
> Z ~J
-e >
:=
7
RESPIRATION DURING HYPOXIA IN BIRDS
(PI - PQ}02
'~
(Po- PI)co~
{Tort )
{Torr)
O O
20
"
10
.
~
~
B
~° 0
40
8O
PIo2
120 (Torr)
0
&O
80
Pro2
120 (Torrl
Fig. 1. Mean values (and SE) of inspired-to-arterial Po2 differences, (Pl - Pa)o2, and arterial-to-inspired Pco: differences, (Pa - PDco2- without (O) and with ( 0 ) inspired CO2.
remains constant at all hypoxic levels, despite increasing ~'E, is explained by the fact that Paco 2 nearly equilibrates to Pxco 2 during hypoxia (table 2 and fig. 1).
Acid-base status. Using measured plasma pH (pHp0 and Pco2 values, bicarbonate concentration, [ H C O 3 - ] p l , was calculated for arterial plasma. For use of a [ H C O 3 3]pl/pHpl diagram, all values were corrected to one temperature (40 °C), and values of pK' (6.082) and CO2 solubility (0.0289 m m o l - L 1.Torr 1) obtained for human plasma at 40 °C (Severinghaus, 1964) were used in calculating [HCO3- Ivy The corresponding mean values are listed in table 2 and plotted in fig. 2. Without CO 2 inhaled, hypoxia (PIo2 = 65 Torr) induces changes in HCO 3 and pH that follow the in vitro buffer line of the duck (Scheipers et al., 1975). Below Pio2 of 65 Torr, the respiratory alkalosis is combined with a metabolic acidosis, whereby, at Plo: = 35 Torr, the normal pHpl is reattained, at a low [ H C O f ]pl of below 5 mmol. L - ~. At the lowest PIo2, 30 Torr, metabolic acidosis supersides, pHpl falling to about 7.26. Compared with these changes, inspiring CO2 leads to somewhat more acidic pH during normoxia, but with hypoxia, HCO3- and pH are kept closer to their control levels, at least at or above PIo2 = 50 Torr, with nearly no metabolic component. In contrast, at PIo2 = 35 Torr, there is a large metabolic change, and pH falls to 7.10, which is distinctly lower than without CO2 inhalation. The metabolic changes are evidenced by the levels of lactate in arterial plasma (table 2) which increase with hypoxia. Without CO2 inhaled, this increase starts at the first level of hypoxia, while during CO 2 inhalation a significant increase is observed only at the lowest PIo2. Hematocrit did not significantly vary during the experiment, the average being 36% (SD = 3~o).
8
H. SHAMS AND P. SCHEID 30
35
[HCO~] /
(mmo[.L 1} p~ 2C3rnmolL'pH' 20
..
~
1~..
~ ~
PCO:
~--~ ._-~-~
/
/ " P[?O
25
,o-
/"
15
26 Torr/.
./ ~ ) ;>h
35~
P'Ico
0 -9r-
35 ~" 0
I
i
[{I] L~
720
: [
730
PHpt I
-
240
750
760
270
Fig. 2. Plot of plasma H C O 3 - concentration [ H C O 3 - ] e l , against plasma pH, prim, for mean values without (©) and with (0) inspired CO2. Experimental data corrected to 40 °C. Numbers at each symbol represent the approximatelevels of P]o2. Curved lines CO2 isopleths; straight line, whole blood bufferline for the duck, with buffer value of/~ = 20.0 mmol. L 1. pH ~ (Scheipers et aL, 1975).
Circulation. The (Ca-C~)o2 differences in table 2 and 1~1o2 values of table l allow calculation of cardiac output ((~) which is listed in table 1. Both without and with CO 2 inhalation, Q increases slightly with diminishing PIo2 , but tends to drop again at deep hypoxia when CO2 is inhaled. This drop is due to a hypoxic reduction in stroke volume (Qs) while cardiac frequency (fcard) rises monotonously (table 1).
Discussion Critique of methods. An important part in this study is the plethysmograph chamber which allows measurement of ventilation in the unanesthetized, virtually unrestrained animal. Problems with this technique derive from pressure drifts, induced by temperature variations. A small leak of the chamber eliminated this drift without the need for a reference chamber (Drorbaugh and Fenn, 1955). The error introduced by this leak in the measurement of tidal volume in the range of respiratory frequencies was less than 0.1%. Calibration with a piston pump at high frequency was mainly responsible for the good agreement between the VT estimates and the direct pneumotachograph measurements. Another problem with the technique is due to the fact that Po2 drops and Pco2 increases while ventilation or exchange rates are measured in the sealed box. The time of sealing the chamber for these measurements was, therefore, kept short enough for partial pressures to change less than 1 Tort when measuring ventilation and less than 5 Tort for measurement of exchange rates. We believe that the disadvantage resulting from sealing the chamber is outweighed by the advantage of measuring in an animal without tracheotomy or face mask.
RESPIRATION DURING HYPOXIA IN BIRDS
9
The tolerance limit observed by Black and Tenney (1980) in the Pekin duck, 11,580 m, is the same as that found by us. The increase in "~E with hypoxia, which is similar to that observed in the Pekin duck (Black and Tenney, 1980), leads to a steep decrease in Paco2, down to 6 Torr at the lowest Plo2 tolerated, 30 Torr. Similarly low levels of Paco 2 in extreme hypoxia have been reported for the Pekin duck and bar-headed goose (Black and Tenney, 1980; Faraci et al., 1984). This extreme hyperventilation exerts effects on the acid-base balance. As shown in fig. 2, mild hypoxia results in respiratory alkalosis which, as hypoxia progresses, is accompanied by a metabolic, i.e. lactic acidosis. Thus, the extent of blood alkalosis is limited, and at the deepest hypoxic level there is a distinct acidosis despite the very low level of arterial Pco2. While Black and Tenney (1980) found similar trends both in the Pekin duck and bar-headed goose, Faraci et al. (1984) reported a slightly alkalotic arterial pH at Pao2 of 28.5 Torr and Paco: of 6.4 Torr in the duck and in the majority of their geese. Respiratory alkalosis was also reported for the pigeon acutely exposed to an altitude of up to 9000 m (Lutz and Schmidt-Nielsen, 1977; Weinstein et al., 1985). This hyperventilation is clearly advantageous for the animal as it maximizes 02 loading of blood in the lung (Dejours, 1982; West, 1983). The efficiency of oxygenation in the lung is shown by the difference in (PI-Pa)o: which becomes progressively smaller with hypoxia, reaching a value of only 5 Torr at PIo2 of 30 Torr. Results similar to these have previously been reported for the Pekin duck and the bar-headed goose (Black and Tenney, 1980; Faraci et al., 1984) and somewhat higher (PI - Pa)o 2 values for the pigeon (Lutz and Schmidt-Nielsen, 1977; Weinstein et al., 1985). This small (PI - Pa)o 2 difference is not primarily due to the peculiarly high gas exchange efficiency of the avian lung, but appears to be related to the exquisitely high parabronchial ventilation relative to 02 uptake. In fact, with such high ventilation, the Po~ drop gas as flows through the parabronchus will become minimal, and will approach the inspired level, whereby the gas exchange systems of the parabronchial and alveolar lungs approach each other (Scheid, 1979). Cardiac output in our animals did not increase nearly as much as in the duck and goose of Black and Tenney (1980). Their values for the duck started at lower, and reached higher, levels than in our study. Correspondingly, for our ducks the "coefficient of oxygen delivery", Cao2/(Ca - C~)o2, is higher in normoxia, but reaches similar values in deep hypoxia. It appears that the stroke volume limits further rise in (~ in hypoxia in our ducks, whereas cardiac frequency shows a monotonous increase with progressing hypoxia. Tachycardia in the spontaneously breathing hypocapnic and hypoxic duck has also been reported by Butler (1970) and Butler and Taylor (1973), and, like in our study, the response was more pronounced during hypercapnic hypoxia. When, on the other hand, hypercapnic hypoxia develops during apnea, as in diving, bradycardia is observed (Butler and Taylor, 1973, 1983). The drop in Mo~ and l~lco~ with deep hypoxia, which has also been observed by Butler (1970), may in part be caused by anaerobiosis in some tissues, but may also reflect the observed drop in body temperature, most likely a result of the extreme hyperEffects ofhypoxia without CO 2 inhalation.
10
H. SHAMS AND P. SCHEID
ventilation. Assuming a Qm value of 2.3 for the metabolic rate, the observed drop in body temperature by 2.5 °C would lead to a reduction in IVlo~by about 20~o, which is nearly the reduction observed. Mammals acutely exposed to hypoxia hyperventilate (Vogel et al., 1969; Nesarajah et aL, 1983; Sidi etal., 1983) but the levels of Pa¢o~ reached appear to be distinctly higher. A notable exception is the singular report of a mountaineer on the top of Mt. Everest who measured alveolar Pco~ as low as 7.5 Torr when breathing ambient air (West, 1983). This value is, however, worth corroborating, with particular attention to the problem of steady state. It seems that arterial hypocapnia is not well tolerated in mammals (Dejours, 1982), the tolerance being much less than in birds. One difference between mammals and birds that might contribute to the higher hypoxia tolerance of birds is their lack of hypocapnic vasoconstriction in many vascular beds, including the brain (Grubb etal., 1977; Wolfenson et al., 1982). In mammals, hypocapnia leads to distinct decrease in brain blood flow (Grote et aL, 1981). Effects of C02 on cardio-respiratory responses to hypoxia. Adding CO 2 to the inhaled air increased the ventilation at all levels of hypoxia, which is in line with the reported respiratory drive of CO2 in birds (cf Scheid and Piiper, 1986) and in mammals. As in normoxia, this drive is mainly on tidal volume, respiratory frequency being unaffected. It is reasonable to expect higher Po2 and Co2 in arterial blood as a result of the CO2-induced stimulation of ventilation, and this is indeed observed with moderate levels of hypoxia, i.e. PIo2 at or above 50 Torr (table 2). Below PIo~ of 50 Torr, however, there is no advantageous effect from the CO 2 drive on ventilation. In fact, with CO 2 inhalation, the (PI - Pa)o~ difference (fig. 1) is not further reduced, indicating that ventilation does not significantly limit 0 2 exchange. In fact, the levels of Pao2 in deep hypoxia are similar without or with CO2. Despite similar Pao2, Cao~ is lower with COg inhalation in this condition, and this is explained by the lower arterial pH which reduces blood affinity by virtue of the Bohr effect. This lower pH during CO2 inhalation is a result mainly of the higher Paco ~ than with CO2-free breathing at comparable hypoxic levels, standard bicarbonate being less reduced with CO2 breathing than without (fig. 2). The effect of CO 2 can thus be summarized to improve arterial oxygenation by increasing ventilation, and concomitantly increasing Pao2 , at low or moderate levels of hypoxia (PIo2 at or above 50 Torr). Below this level ofhypoxia, however, the ventilatory stimulation of CO 2 is without effect on Pao2, but the lower blood pH reduces 0 2 loading to hemoglobin. That this is likely to limit hypoxia tolerance is evident from the ensuing low mixed venous Co~ which is in fact similar at the tolerance limits without or with CO 2 inhalation. That CO2 inhalation at Plo~ of 36 Torr has other disadvantageous effects on the 02 transport is shown by the decreased cardiac output and stroke volume. In man, administration of carbonic anhydrase blockers in acute altitude exposure exerts beneficial effects on symptoms of acute mountain sickness (Forwand et al., 1968; Sutton etal., 1979). That this may be related to the prevention of hypocapnia is
RESPIRATION DURING HYPOXIA IN BIRDS
11
suggested from experiments, in which C O 2 administration during acute altitude exposure has been found to induce a sense of well-being (of. Luft, 1965). Aside from preventing hypocapnia, C O 2 addition at a given altitude stimulates ventilation and thus results in improved blood oxygenation. That the advantageous effect of C O 2 breathing is mainly due to better oxygen loading in the lung is suggested by experiments of Maher et al. (1975) who compared identical levels of arterial Po2 and found that C O 2 breathing aggravated the symptoms of acute mountain sickness. Other, partly advantageous effects o f C O 2 were also observed, e.g. maintained stroke volume (Grover et al., 1976); inhibition of forearm venoconstriction (Weil et al., 1971; Cruz et al., 1976) in man; increased coronary blood flow with increased systemic, splanchnic and renal vasoconstriction in the dog (Koehler et al., 1980). It is thus not easy to decide whether C O 2 breathing in acute altitude exposure in mammals is on balance advantageous or not. Experiments with progressive hypoxia, extending to severe levels, appear to be needed in mammals. Conclusion. One factor that may be of importance for the pronounced high altitude tolerance of birds compared with mammals appears to be the increase in ventilation, that is not impeded by the ensuing extremely low arterial Pco2 and allows arterial blood to nearly equilibrate with inspired Po2- The basis for this difference between birds and mammals relates probably to the hypoxia and hypocapnia sensitivity of the respiratory centers, and is not primarily a result o f the high gas exchange efficiency of the avian parabronchial lung. Addition of CO2 to the inspired gas is advantageous during mild hypoxia, when the effect on stimulating ventilation results in increased blood oxygenation. During extreme hypoxia, however, blood oxygenation is not further improved by addition of C O 2, and adverse effects, mainly due to the uncompensated metabolic acidosis and its effects on blood 0 2 transport, become limiting.
Acknowledgement.The skilled technical assistance of Mrs. R. Hildebrandt is gratefully acknowledged. References Black, C.P. and S.M. Tenney (1980). Oxygen transport during progressive hypoxia in high-altitude and sea-level waterfowl. Respir. Physiol. 39: 217-239. Butler, P.J. (1970). The effect of progressive hypoxia on the respiratory and cardiovascular systems of the pigeon and duck. J. Physiol. (London) 211: 527-538. Butler, P.J. and E.W. Taylor (1973). The effect of hypercapnic hypoxia, accompanied by different levels of lung ventilation, on heart rate in the duck. Respir. Physiol. 19: 176-187. Butler, P.J. and E.W. Taylor (1983). Factors affecting the respiratory and cardiovascular responses to hypercapnic hypoxia, in mallard ducks. Respir. Physiol. 53: 109-127. Cruz, J.C., R.F. Grover, J.T. Reeves, J.T. Maher, A. Cymerman and J,C. Denniston (1976). Sustained venoconstriction in man supplemented with CO2 at high altitude. J. Appl. Physiol. 40: 96-100. Dejours, P. (1982). Mount Everest and beyond: breathing air. In: A Companion to Animal Physiology, edited by C. R. Taylor, K. Johansen and L. Bolis. Cambridge, UK, Cambridge Univ. Press, pp. 17-30. Drorbaugh, J. E. and W. O. Fenn (1955). A barometric method for measuring ventilation in newborn infants. Pediatrics 16: 81-87.
12
H. SHAMS AND P. SCHEID
Faraci, F.M., D.L. Kilgore, Jr. and M.R. Fedde (1984). Oxygen delivery to the heart and brain during hypoxia: Pekin duck vs. bar-headed goose. Am. J. Physiol. 247: R69-R75. Forwand, S.A., M. Londowne, M.J.N. Follansbee and C.J.E. Hansen (1968). Effect of acetazolamide on acute mountain sickness. N. Engl. J. Med. 279: 839-845. Grote, J., K. Zimmer and R. Schubert (1981). Effects of severe arterial hypocapnia on regional blood flow regulation, tissue Po2 and metabolism in the brain cortex of cats. Pfliigers Arch. 391: 195-199. Grover, R.F., J.T. Reeves, J.T. Maher, R.E. McCullough, J.C. Cruz, J.C. Denniston and A. Cymerman (1976). Maintained stroke volume but impaired arterial oxygenation in man at high altitude with supplemental CO 2. Circ. Res. 38: 391-396. Grubb, B., C. D. Mills, J.M. Colacino and K. Schmidt-Nielsen (1977). Effect of arterial carbon dioxide on cerebral blood flow in ducks. Am. J. Physiol. 1: H596-H601. Koehler, R.C., B.W. McDonald and J.A. Krasney (1980). Influence of CO 2 on cardiovascular response to hypoxia in conscious dogs. Am. J. Physiol. 239: H545-H558. Luft, U.C. (1965). Aviation physiology - the effects of altitude. In: Handbook of Physiology. Section 3, Respiration, Vol. II, edited by W.O. Fenn and H. Rahn. Washington, D.C., American Physiological Society, 1099 pp. Lutz, P.L. and K.S. Schmidt-Nielsen (1977). Effect of simulated altitude on blood gas transport in the pigeon. Respir. Physiol. 30: 383-388. Maher, J.T., A. Cymerman, J.T. Reeves, J.C. Cruz, J.C. Denniston and R.F. Grover (1975). Acute mountain sickness: Increased severity in eucapnic hypoxia. Aviat. Space Environ. Med. 46: 826-829. Nesarajah, M.S., S. Matalon, J.A. Krasney and L.E. Farhi (1983). Cardiac output and regional oxygen transport in the acutely hypoxic conscious sheep. Respir. Physiol. 53: 161-172. Nolan, W.F. and D.G. Davies (1982). Brain extracellular fluid pH and blood flow during isocapnic and hypocapnic hypoxia. J. Appl. Physiol. 53: 247-252. Scheid, P. (1979). Mechanisms of gas exchange in bird lungs. Rev. Physiol. Biochem. Pharmacol. 86:137-186. Scheid, P. and J. Piiper (1986). Control of breathing in birds. In: Handbook of Physiology. Section 3. The Respiratory System. Vol. II. Control of breathing. Part 2, edited by A. P, Fishman, N.S. Cherniack, J.G. Widdicombe and S.R. Geiger. Bethesda, MD. American Physiological Society, pp. 815-832. Scheipers, G., T. Kawashiro and P. Scheid (1975). Oxygen and carbon dioxide dissociation of duck blood. Respir. Physiol. 24: 1-13. Severinghaus, J.W. (1964). Blood gas concentrations. In: Handbook of Physiology. Section 3. Respiration. Vol. If, edited by W.O. Fenn and H. Rahn. Washington, D.C., American Physiological Society, pp. 1475-1487. Sidi, D., J. R.G. Kuipers, D. Teitel, M.A. Heymann and A.M. Rudolph (1983). Development changes in oxygenation and circulatory responses to hypoxemia in lambs. Am. J. Physiol. 245: H674-H682. Sutton, J. R., C. S. Houston, A. L. Mansell, M. D. McFadden, P. M. Hackett, J. R. A. Rigg and A. C. P. Powlcs 1979). Effect of acetazolamide on hypoxemia during sleep at high altitude. N. Engl. J. Med. 301: 1329-1331. Tucker, V.A. (1968). Respiratory physiology of house sparrows in relation to high altitude flight. J. Exp. Biol. 38: 55-66. Vogel, J.A., R.I. Pulver and T.M. Burton (1969). Regional blood flow distribution during simulated high-altitude exposure. Fed. Proc. 28:1155-1159. Weil, J.V., E. Byrne-Quinn, D.J. Battock, R.F. Grover and C. Chidsey (1971). Forearm circulation at high altitude. Clin. Sci. 40: 235-246. Weinstein, Y., M.H. Bernstein, P.E. Bickler, D. V. Gonzales, F.C. Samaniego and M.A. Escobedo (1985). Blood respiratory properties in pigeons at high altitudes: effects of acclimation. Am. J. Physiol. 249: R765-R775. West, J. B. (1983). Climbing Mt. Everest without oxygen: an analysis of maximal exercise during extreme hypoxia. Respir. Physiol. 52: 265-279. Wolfenson, D., Y.F. Frei and A. Berman (1982). Blood flow distribution during artificially induced respiratory hypocapnic alkalosis in the fowl. Respir. Physiol. 50: 87-92.