Adjustments in oxygen transport during head-out immersion in water at different temperatures MARIE-LUCE CHOUKROUN AND PIERRE VARENE Laboratoire de Physiologie, Universite de Bordeaux II, 33076 Bordeaux
CHOUKROUN,
MARIE-LUCE,
AND
PIERRE
VARENE.
Ad&&-
ments in oxygen transport during head-out immersion in water at different temperatures. J. Appl. Physiol. 68(4): 14751480, 1990.-Respiratory gas exchange was investigated in human subjects immersed up to the shoulders in water at different temperatures (T, = 25,34, and 40°C). Cardiac output (Qc) and pulmonary tissue volume (Vti) were measured by a rebreathing technique with the inert gas Freon 22, and O2 consumption (VO*) was determined by the closed-circuit technique. Arterial blood gases (Pa+ Pa& were analyzed by a micromethod, and alveolar gas (PAL,) was analyzed during quiet breathing with a mass spectrometer. The findings were as follows. 1) Immersion in a cold bath had no significant effect on &c compared with the value measured at T, = 34OC, whereas immersion in a hot bath led to a considerable increase in Qc. Vti was not affected by immersion at any of the temperatures tested. 2) A large rise in metabolic rate Vo2 was only observed at T, = 25°C (P < 0.001). 3)Arterial blood gases were not significantly affected by immersion, whatever the water temperature. 4) O2 transport during immersion is affected by two main factors: hydrostatic Above neutral temperature, Q2 pressure and temperature. transport is improved because of the marked increase in Qc resulting from the combined actions of hydrostatic counter pressure and body heating. Below neutral temperature, O2 transport is altered; an increase in O2 extraction of the tissue is even calculated. water temperature; rial blood gases
cardiac output;
oxygen consumption;
Cedex, France
25°C a decrease of VC and at T, = 40°C an increase of VC. At 25°C we observed a marked increase in 02 consumption (VOW), whereas in this case the only change in heart rate (HR) was a slight bradycardia. This last phenomenon seemsto have been first reported by Cannon and Keatinge (8), but it does not appear to have been investigated any further. The present study was therefore designed to investigate the effect of T, during head-out immersion on respiratory gas exchange. Three T, were chosen: 34°C (thermoneutral bath), 25OC (cold bath), and 40°C (hot bath). We measured Q, Vo2, and arterial blood gases(Pao, and Pace,). MATERIALS
AND METHODS
Subjects Eleven healthy volunteers (4 females, 7 males) were selected for the study. Informed consent was obtained. Subjects others than laboratory staff received a financial allowance. Their anthropometric data are given in Table 1. Most of the subjects are nonsmokers; only two are light smokers (subjects AML and BL). All have normal pulmonary function tests.
arte-
Procedure
alongwithpulmonaryand cardiocirculatory readjustments during head-out immersion in water, has been extensively studied at thermoneutral temperature (34°C). The hydrostatic. counter pressure causes an increase in cardiac output (Qc), central blood volume, and pulmonary arterial pressure (2, 4, 13) that is associated with ventilatory adjustments such as a marked reduction in functional residual capacity and a slight drop in vital capacity (VC) (1, 13, 23). Some authors (3, 11, 21) have observed alterations in gas exchange: a fall in arterial POT (Pao,) and an increase in alveolar-arterial difference (PAo,-Pao,), attributed to modifications in the distribution of the ventilation-perfusion (V/Q) ratio during immersion. However, humans are rarely immersed in water at thermoneutral temperature. Immersion generally takes place in colder water. As previously described by Kurss et al. (20) we have found that the temperature of the water (T,) during immersion affects pulmonary capacities (10); in comparison, using a thermoneutral bath, we measured at T, =
RESPIRATORYGASEXCHANGE,
0161-7567/90
$1.50 Copyright
Immersion was carried out in a vertical, cylindrical plastic tank containing -600 liters of water. The desired T, was obtained by mixing cold and hot water. Because of the large volume of water, T, remained approximately constant (+OJ”C) during the experiments. The subjects stood upright in the tank, immersed up to the shoulders, more precisely to the Louis-angle level. Body temperature was measured with a rectal probe (T,,), and cardiac activity was recorded on an electrocardiogram throughout the experimental session. On the same one-half day, a given subject was immersed at all three T, 34, 25, and 40°C. The T, were chosen in random order, and at least 20 min elapsed before the measurements. The order of the measurements in each situation was the following: 1) determination of vo2 and expired ventilation (VE), 2) collection of blood samples for blood-gas analysis, and 3) determination of Qc and pulmonary tissue volume (Vti). Before immersion, control values of each variable were obtained from the subjects standing in thermally neutral air.
0 1990 the American
Physiological
Society
1475
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (130.130.211.199) on July 30, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
1476
OXYGEN
TABLE 1. Anthropometric characteristics participating in the experiment Subjects
Sex
Age, yr
Height, m
AML MHB BC SD ES RC BJ LP LO BL BB
F F F M M M M M M M F
27 28 37 27 21 20 26 25 20 19 22
1.67 1.68 1.57 1.60 1.81 1.84 1.70 1.75 1.85 1.78 1.55
Means
2
SD
2525
1.71kO.11
TRANSPORT
of subjects
Weight, kg 49 53 47 57 68 96 65 63 65 72 45 62-1-14
Body Area, m* 1.53 1.59 1.44 1.59 1.73 2.19 1.86 1.77 1.86 1.89 1.40 1.71k0.23
Measurements Cardiac output. Qc was measured by a noninvasive method involving rebreathing of an inert gas, Freon 22 (CHF&l), according to Bonde-Petersen et al. (5). The rebreathing apparatus consisted of a 3-liter rubber bag connected to a three-way stopcock. The bag was filled with 2 liters of a gas mixture containing 4.9% Freon 22, 6.75% Ar, 39.8% O2 in NP. After a resting time of 20 min the subject was connected to the rebreathing bag at the end of a complete expiration. The rebreathing maneuver was maintained for 15 s at a respiratory rate of 20 breaths/min. During the rebreathing, the fractional concentrations of Freon, Ar, and 02 were analyzed continuously by a mass spectrometer (Spectralab M VG Gas Analysis Instrument) and recorded on paper (Gould ES 1000). The mass spectrometer was programmed to analyze a mixture of five gases: 02, COZ, Nz, Ar, CHF&l whose total concentration equals one. The calibration curves for each gas were determined with gas mixtures whose composition is known with an accuracy generally better than 0.1%. In the rebreathing bag, Freon 22 concentration was measured with a resolution of 0.010%. The data were collected only for the breaths during which there was minimal change in Ar concentrations. The first breaths were discarded because of the likelihood of incomplete gas mixing. The maneuver was terminated after 15 s to avoid recirculation. The pulmonary capillary blood flow (Qc) and the pulmonary tissue plus capillary blood volume (Vti + Vc) were calculated from the disappearance curve of Freon 22 according to the equation of Cander and Forster (7). The regression line was fitted with the expiratory and inspiratory Freon values as suggested by Sackner et al. (28). By knowing the Bunsen solubility coefficients of Freon 22 in blood and tissues, Qc was derived from its slope and (Vti + Vc) .from its intercept value. The Bunsen coefficients for Freon 22 have been determined in our laboratory (29). They are ablood = 0.804 t 0.049 cm3 STPD of Freon 22cmm3 blood*ATA-1 at 37°C = 0.635 t 0.091 cm3 STDP of Freon 22 crne3 and atissue tissue. ATAB at 37°C. Vti was calculated after subtraction of Vc, estimated from mean values previously measured in our laboratory
DURING
IMMERSION
on normal subjects under similar conditions (9). O2 consumption. iTo was measured in the steady state by the closed-circuit technique before the rebreathing maneuver. vo2 was expressed in STPD conditions. VE was also calculated from the tidal volume and the respiratory frequency given by the spirometer (Expirograph Gould-Godard) and was expressed in BTPS conditions. Arterial blood gases.Arterial blood gases were analyzed in blood samples collected from the hyperemized earlobe immediately before rebreathing by the subjects. The partial pressures in arterial blood, (Pao,, Pace,) were determined with a blood gas analyzer (Corning 173) and the arterial oxygen content (Cao,) with another blood gas analyzer (Co-Oxymeter, Instrumentation Laboratory). Appropriate corrections were made for Pao, and Pace, as a function of Tree Hemoglobin concentration ([Hb]) was also measured. The (PAo,-ho2) was calculated from Pao,, determined by micromethod, and the mean alveolar POP (PA~J. PAN, was estimated from the end-tidal 02 fraction measured with the mass spectrometer during quiet breathing before the rebreathing maneuver. Statistical analysis. All data obtained for each situation (air, Tw = 34”C, Tw = 25°C and T, = 40°C) were compared with Student’s t test for paired samples. RESULTS Temperature
With respect to the temperature moneutral bath, the mean core (AT,) were -0.55 t 0.42OC during bath and +l.Ol t 0.53OC during bath. Hemodynamic Data (Table
recorded in the thertemperature changes immersion in the cold immersion in the hot
2)
For the subjects standing in air, the mean &c was 4.75 t 0.53 dm3* min-‘. This corresponds to a cardiac index (CI) of 2.80 t 0.35 dm3* mine1 l m-? The mean HR was 95 t 14 beatsmin-‘, giving a mean stroke volume (SV) of 42 t 7 cm3. The mean Vti calculated after subtraction of the Vc was 456 & 177 cm3. Immersion in thermoneutral water (34°C) increased &c significantly (+18%, P < 0.05). This was essentially because of a marked rise in SV (+52%, P < 0.005). HR fell significantly (P < 0.01). Vti was not affected by thermoneutral immersion. The value of Qc during immersion in the cold water V = 25°C) was comparable with the value at T, = 34’C. Cardiac frequency was somewhat reduced but not to the extent previously observed in a similar situation (10). Vti was unchanged. Immersion in the hot water (T, = 40°C) induced a considerable increase in Qc (+41%, P < 0.005) because of a rise in HR, whereas SV was less than that observed in the thermoneutral bath, and Vti was unchanged. Gas Exchange Data (Table 3)
The mean 60, measured, when the subjects were standing in air, was 0.260 t 0.056 dm3 STPDemin-‘.
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (130.130.211.199) on July 30, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
OXYGEN
TABLE
TRANSPORT
DURING
1477
IMMERSION
2. Hemodynamic data in air and during immersion at different temperatures Immersion Air Qc, dm3 - min-’
CI, dm” . min-’ HR, beats
l
4.75kO.53 t
- m-’
T, = 34°C Ps
5.61tl.27 0.05w’NS 4
2.80k0.35
t
t
Vti,
cm3
cm3
75t13
456tl77 tNS
71t13
P 5 0.0001-
64t14
69t17
55&11
MC--
P 5 0.05 + NS-
513t145
565k130
551&275
463&126
411-+260
-+NS416t139 -1NS
,-NS-
NS
-
3. Gas exchange data in air and during immersion at different temperatures Immersion
Air T, = 34°C %JO~, dm3 STPD
\jE,
llOt15
P 5 0.0001___)
4
TABLE
3.89-eO.44
Af--
Ps 0.005-‘NS
547t154 tNSd’NS
l
3.43kO.64
4 (Vc + Vti),
--)
P 5 0.005
Ps 0.01-‘NS
42k7
6.71t1.50 P 5 0.005
4 SV, cm”
T, = 40°C
5.90t1.45 ,b
3.27k0.54
95-el4
min-l
T, = 25°C
l
min-’
dm3 BTPS .min-’
0.260t0.056 -
T, = 25°C
0.239t0.053 NS
A+
0.364tO.
160
P < 0.001 +-
8.3Ok 1.52
7.7kl.22
T, = 40°C 0.258kO.047 NS -
11.74k3.81
10.76k5.34
-NS-tPs0.005+-NSI Pa+
Torr
Torr
(PAo,-ho,),
Pace,, Torr
Hb, g/100
cm” blood
Cao,, cm3/100
cm” blood
91.50t9.52 -NS-t
94.27t9.15
22.5OklO.95 -NS--NSF’NS-
20.25k10.20
P s 0.05 91.27t10.05
89.32t9.90
24.52t14.62
22.87-1-8.92
NSF’NS-
36.60t3.75 -
NS --
35.62k3.52
15.91kl.89 c--------------
NS -
21.60-2.50 -
NS A-
15.63t1.88 t
21.22k2.53
Immersion in the thermoneutral bath led to a nonsignificant fall in VOW. There wag a marked rise in metabolic rate during cold immersion (VO, at 25°C = 0.364 t 0.160 dm3 STPD . minT1, P c 0.001). However, in the hot bath the change in Voz was not statistically significant with respect to that measured at T, = 34°C. VE was not affected by thermoneutral immersion. However, there was a significant rise in VE when the water temperature deviated from 34’C. In fact, there w.as considerable individual variation in the change in VE with change in bath temperature, especially for immersion at T, = 40°C.
34.50t3.90 NS --NS-
5 0.005
16.02&l --
21.68k2.20 NS -1NS-
35.25t4.27
.79
15.68Ifr1.67 NS A
21.27k2.32
Pao. and Paoo, and the Cao were not significantly altered by immersion at any of the temperatures studied. (PAO,-ho2) was not affected by the bath temperature. The only change in [Hb] was a slight but significant rise during immersion in the cold bath (16.02 g/100 cm” blood at T, = 25°C vs. 15.63 g/100 cm” blood at T, = 34°C; P < 0.005). DISCUSSION
Methodology
The noninvasive methods, such as Freon 22 rebreathing chosen here to measure Qc, are based on the as-
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (130.130.211.199) on July 30, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
1478
OXYGEN
TRANSPORT
sumption that the Qc equals to the pulmonary capillary shunt being neglected. blood flow, wit h the right-to-left This is generally true in normal subjects, the difference caused by the shunt being less than that introduced by the scatter of all the techniques. Our control Qc or CI values were sma ller than those reported by workers using methods such as acetylene invasive (2) or noninvasive rebreathing (4) and CO2 rebreathing (13). It is generally adm itted that the Qc values obta .ined with the rebreath1% technique are smaller with Freon 22 rather than acetylene (5), which is the reference gas. Nevertheless, because of its low toxicity and of its safe handling, we chose to use Freon 22 instead of acetylene, more especially as we were mainly interested in the variations of Qc in function of the bath temperature rather than its absolute value. Moreover another explanation for these differences in Qc might be put forward: one-th ird of our group was females, whereas in the above studies they consisted of male subjects in whom CI is higher. (Vti + Vc) was calculated from the intercept of the disappearance curve of Freon 22, time 0 of this curve being identified by the rise of the Freon concentration during the first inspiration. This method is somewhat approximate, but our control values in air were in good agreement with those obtained by authors using an acetylene-rebreathing method (28). The calculation of Qc and (Vti + Vc) necessitates to know the Bunsen solubility coefficients of Freon 22 in blood and tissue ( CYB,ati). The solubility of an inert gas is influenced by the tempe ratu re of the solvent. In our experiment, the changes in T re were too small to induce significant measurable changes of the solubility of Freon in pulmonary blood or tissue. Then the variations of with the temperature have been neablood and atissue, glected in this work. Another point to be discussed concerns the estimation of [Hb] and Ca 0, contents from a sample of capillary blood. As it is generally admitted, despite a careful sampling (hyperemized earlobe) the given Pao, and Cao, values were probably underestimated, especially in cold bath when a considerable skin vasoconstriction takes exists place. N ‘evertheless, because a close correlation between arterial and arterialized capillary Paz, or pco2, these values were sufficient ly reliable to support our discussion. Results Thermoneutral immersion. Numerous studies on the hemodynamic readjustments during thermoneutral immersion have confirmed the prediction of Gauer and Thron (14) that there is an increase in Qc associated with a rise in SV, with the HR being almost unchanged (2). These changes are attributed to improved venous return induced by a shift of blood from the periphery to the intrathoracic regions because of the hydrostatic counter pressure. There is, however, some lack of agreement on the extent of the increase; Arborelius et al. (2) reported a 30-35% rise in Qc, Begin et al. (4) a 20-40% increase, Farhi and Linnarsson (13) an increase of 40% during immersion up to the xiphoid and of 66% during immersion up to the chin, whereas we observed only a
DURING
IMMERSION
rise of 18%. According to Farhi and Linnarsson (13) the discrepancies can be explained by differences in the position of the body in the water (sitting, standing, or supine) and in the level of immersion (hips, xiphoid, shoulders, or neck). Cohen et al. (11) have suggested that in some subjects, immersion increases the shunt component, and so by using an inert gas rebreathing method, we may have underestimated the rise in Qc induced by the hydrostatic counter pressure. However, in our experiment, (PA@ao,) did not change significantly during immersion, and we concluded that the increase in shunt at T, = 34°C was not important, and so this would have had no effect on the determination of Qc. Hood et al. (17) and McArdle et al. (22) have not found such changes in Qc during immersion, whereas Rennie et al. (25) even observed a fall, but their experiments were not performed at exactly thermoneutral temperature. In our study, the rise in Qc was caused by a large rise in SV, with a marked fall in HR. This change in HR is generally attributed to a baroreflex induced by the increase in blood pressure related to the rise in Qc. Vti was not affected by thermoneutral immersion. Farhi and Linnarsson (13) found changes in equivalent lung tissue volume, which was largely attributed to alterations in Vc. Our values of Vti, after subtraction of Vc, which is known to be affected by immersion (9), are in good agreement with these findings. They indicate that there is no extravasation of fluid during . immersion. The slight but nonsignificant drop in iJo2 was probably caused by reduced metabolism in the immersed subjects. During immersion, postural muscles are quiescent because of the supporting hydrostatic counter pressure. The results on arterial blood gaseswere more surprising. Thermoneutral immersion up to the neck has generally been reported to lead to a decrease in Pao, associated with an increase in the (PAo.,-Pa& (11, 21). This iS attributed to an alteration ofAventilation-perfusion ratio distribution (3, 21). We did not observe such variations in Pao, or (PAo,-ho,). Two explanations can be found. First, the time elapsed between the measurements of Tie,, by use of the closed-circuit technique, and the blood sampling is -2 or 3 min. Then because of the possibility of 02 lingering in poorly ventilated zones, these variations may be avoided by high PAN, levels. However a second explanation can be found in the study of Prefaut et al. (24); they reported that “gas exchange modifications during immersion depend on position of end expiratory level in relation to closing volume,” this closing volume being largely dependent on the age of the subjects. In the present study our subjects, except for one, were ~28 yr old, and so owing to a reduced closing volume this effect is probably not very significant in our conditions. Then the most important effect during immersion was a rise in Qc leading to an improvement in gas exchange. Although Cao, was unchanged, Qc was significantly increased, whereas Tj02 was slightly lower during immersion which should lead to a rise in venous 02 content (Cvo,) and venous PO:! (PvoJ. Although the increase in Cvo, calculated at T, = 34°C was not statistically significant (Cvo, air = 15.76 t 2.45 cm3/100 cm3 blood; Cvo,
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (130.130.211.199) on July 30, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
OXYGEN
TRANSPORT
34°C = 16.76 t 2.34 cm3/100 cm3 blood), this improvement is consistent with the value measured by Farhi and Linnarsson (13) for immersion up to the xiphoid or chin. Immersion in cold water. During cold stimulation, like during immersion of a foot in cold water according to Cooper et al. (12), vasoconstriction should result in a rise in blood pressure and a drop in HR without any significant alteration in Qc. Our results are in agreement with these predictions. At T, = 25”C, Qc was not significantly increased compared with the value obtained at T, = 34OC. This was essentially caused by a lower cardiac frequency in response to the cold-induced vasoconstriction. However, this fall in cardiac frequency measured at T = 25°C was not as marked as that observed in a prWevious study (10). This may have been caused by individual differences in response. Keatinge and Evans (18) only observed such a fall in cool water (25°C); in colder water (15°C 5°C) the HR was markedly increased. Along with this cardiovascular readjustment during cold immersion,, the metabolic response leads to a large increase in Vo2. This last observation was reported by Cannon and Keatinge (8); the metabolic rate rose as soon as the T, dropped below 34°C the thermoneutral temperature in water. On the basis of these two observations, it can be calculated that there will be a larger arteriovenous difference in O2 (Pao,-Pvo,) during cool immersion (T, = 25°C) than during thermoneutral immersion (Pao,Pvo, 34°C = 4.28 t 0.87 cm3/100 cm3 blood; Pao,-Pvo, 25°C = 6.11 t 2.06 cm3/100 cm3 blood, P c 0.05). Because Pao, and Cao, are unchanged at T, = 25°C there should be a decrease in mixed venous POT. We calculated a mean decrease of Cvo, of 8.5%, which corresponds to an enhanced O2 extraction of the tissue (Pao,-Pvo,)/Ca,, (28% at T, = 25°C for 20% at T, = 34°C P 5 0.05). The alterations in ventilation observed at T, = 25°C did not induce a significant change in Pace,. In fact, the ventilatory response in cold water (T, = 25°C) varied considerably among the subjects. Some subjects hyperventilated and reduced their Pace,, whereas others showed little change in VE. The higher [Hb] in the blood during cold immersion is of interest. This can be related to cold diuresis. This hemoconcentration effect has also been investigated by Harrison et al. (16), who found that “skin cooling from normothermia caused hemoconcentration” because of the rise in capillary pressure resulting from vasoconstriction. If we consider the duration of our experimentation, the increase in Hb is more likely related to this last mechanism. Immersion in the hot bath. The rise in Qc results from both the hydrostatic counterpressure and the effect of body heating. In recumbent subjects wearing heated suits, whose hemodynamics are similar to those during head-out immersion, Rowe11 et al. (27) measured a 38% increase in Qc when the skin temperature was maintained at 34°C and a 112% increase at 40.9”C. Weston et al. (30), using a Doppler ultrasound technique, found a 121% increase in Qc during immersion at T, = 39OC. Our experimental conditions were comparable to the above CT = 40°C; AT,, = +l.Ol t 0.53”C), and the changes we obierved in &c (+41%), perhaps slightly underestimated,
DURING
1479
IMMERSION
are essentially in agreement with these results. The underestimate may stem from an enhanced shunt at T, = 4O”C, which is not taken into account by the inert gas method we used for measuring Qc. This excess Qc cannot be attributed to metabolic requirements, as \jo2 is almost constant, but is more likely to result from peripheral vasodilation at the higher temperature. It induces a decrease in the (Pao,-Pvo,) at T, = 4O’C. Because Cao, is unaffected by the bath temperature, an improvement in Cvo, (and hence Pvo,) may be expected at the higher temperature (Cvo, 34°C = 16.76 t 2.34 cm3/100 cm3 blood; Cvo; 40°C = 17.17 t 2.13 cm3/100 cm3 blood, NS). However, in front of this increase in Qc the question of a change in the distribution of V/Q ratio arises. Although there was little intersubject variation in the rate of increase of &, there was considerable variation in TjE: six subjects had a raised TjE, whereas in five there was little change. This may lead to an uneven distribution of v/Q ratio, such as an increase in the areas of low v/Q, which would affect gas exchange. From our measurements an answer cannot be given, but we did not observe a significant increase in (PAo,-Pa*,) at T, = 4O”C, although Pao, was slightly but not significantly reduced. In hot water, the most important effect was an improvement in mixed venous O2 content in relation to the increased Q. In conclusion, we demonstrated an interaction between two main factors affecting 02 transport during immersion: temperature and hydrostatic pressure. Farhi and Linnarsson (13) describe how “below neutral temperature, thermal and hydrostatic effects act in opposition.” Our results indicate that during cold immersion the effects of temperature outweigh the effects of hydrostatic counterpressure. The rise of Pvo, related to the increase of Qc during thermoneutral immersion is cancelled out at T, = 25°C and enhanced O2 extraction of the tissue is even observed (P 5 0.05). During immersion in hot water, heat-induced cutaneous vasodilation is the main factor, which leads to a marked increase in HR and Qc; 02 transport is improved, although in some subjects there may be a redistribution of the V/Q ratio. The authors thank Pierre Techoueyres for technical assistance and Nadine Capdeville for typing the manuscript. Address for reprint requests: M. L. Choukroun, Laboratoire de Physiologie, UER III, Universite de Bordeaux-II, 146 rue Leo Saignat, 33076 Bordeaux Cedex, France. Received
22 February
1989; accepted
in final
form
27 October
1989.
REFERENCES E., G. GUTNER, G. TORRI, AND H. RAHN. Respiratory during submersion and negative pressure breathing. J. Appl. Physiol. 21: 251-258, 1966. ARBORELIUS, M., JR., U. I. BALLDIN, B. LILJA, AND C. E. G. LUNDGREN. Hemodynamic changes in man during immersion with head above water. Aerosp. Med. 43: 592-598, 1972. ARBORELIUS, M., JR., U. I. BALLDIN, B. LILJA, AND C. E. G. LUNDGREN. Regional lung function in man during immersion with the head above water. Aerosp. Med. 43: 701-707,1972. BEGIN, R., M. EPSTEIN, M. A. SACKNER, R. LEVINSON, R. DOUGHERTY, AND D. DUNCAN. Effects of water immersion to the neck on pulmonary circulation and tissue volume in man. J. Appl. Physiol. 40: 293-299, 1976. AGOSTONI,
mechanics
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (130.130.211.199) on July 30, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
1480
OXYGEN
TRANSPORT
5. BONDE-PETERSEN, F., P. NORSK, AND Y. SUZUKI. A comparison between Freon and acetylene rebreathing for measuring cardiac output. Aviat. Space Environ. Med. 51: 1214-1221, 1980. 6. BOYER, J. T., J. R. E. FRASER, AND A. E. DOYLE. The haemodynamic effects of cold immersion. Clin. Sci. Lond. 19: 539~550,196O. 7. CANDER, L., AND R. E. FORSTER. Determination of pulmonary parenchymal tissue volume and pulmonary blood flow in man. J. Appl. Physiol. 14: 541-551, 1959. 8. CANNON, P., AND W. R. KEATINGE. Metabolic rate and heat loss of fat and thin man in heat balance in cold and warm water. J. Physiol. Lond. 154: 329-344, 1960. 9. CHOUKROUN, M. L., H. GUENARD, AND P. VARENE. Pulmonary capillary blood volume during immersion in water at different temperatures. Undersea Biomed. Res. 10: 331-342, 1983. 10. CHOUKROUN, M. L., C. KAYS, AND P. VARENE. Effects of water temperature on pulmonary volumes in immersed human subjects. Respir. Physiol. 75: 255-266, 1989. 11. COHEN, R., W. H. BELL, H. A. SALTZMANN, AND J. A. KYLSTRA. Alveolo-arterial oxygen pressure difference in man immersed up to the neck in water. J. Appl. Physiol. 30: 720-723, 1971. 12. COOPER, K. E., S. MARTIN, AND P. RIBEN. Respiratory and other responses in subjects immersed in cold water. J. Appl. Physiol. 40: 903-910,1976. 13. FARHI, L. E., AND D. LINNARSSON. Cardiopulmonary readjustments during graded immersion in water at 35OC. Respir. Physiol. 30: 35-50, 1977. 14. GAUER, 0. H., AND H. L. THRON. Postural changes in the circulation. In: Handbook of Physiology. Circulation. Washington, DC: Am. Physiol. Sot., 1965, sect. 2, vol. III, p. 2409-2439. 15. GOFF, L. G., H. F. BRUBACH, H. SPECHT, AND N. SMITH. Effect of total immersion at various temperatures on oxygen uptake at rest and during exercise. J. Appl. Physiol. 9: 59-61, 1956. 16. HARRISON, M. H., R. J. EDWARDS, L. A. COCHRANE, AND M. J. GRAVENEY. Blood volume and protein responses to skin heating and cooling in resting subjects. J. Appl. Physiol. 54: 515-523, 1983. 17. HOOD, S. K., R. H. MURRAY, C. W. URSCHEZ, J. A. BOWERS, AND J. K. GOLDMANN. Circulatory effects of water immersion upon human subjects. Aerosp. Med. 39: 579-584, 1968. 18. KEATINGE, W. R., AND M. EVANS. The respiratory and cardiovascular responses to immersion in cold and warm water. Q. J. Exp. Physiol. 46: 83-94, 1961.
DURING
IMMERSION
19. KNIGHT, D. R., AND S. M. HORVATH. Effect of hydrostatic pressure on plasma concentrations of norepinephrine during cold water immersion. Undersea Biomed. Res. 14: l-10, 1987. 20. KURSS, D. I., C. E. G. LUNDGREN, AND A. J. PASCHE. Effect of water temperature and vital capacity in head-out immersion. Underwater Physiology VII, edited by A. J. Bachrach and M. M. Matcen. Bethesda, MD: Undersea Med. Sot., 1981, p. 297-301. (Proc. 7th Symp. Underwater Med.) 21. LOLLGEN, H., G. VON NIEDING, H. KREKELLER, U. SMIDT, K. KOPPENHAGEN, AND H. FRANK. Respiratory gas exchange and lung perfusion in man during and after head out water immersion. Undersea Biomed. Res. 3: 49-56, 1976. 22. MCARDLE, W. D., J. R. MAGEL, AND G. R. LESMES. Metabolic and cardiovascular adjustments to work in air and water at 18, 25 and 33°C. J. Appl. Physiol. 40: 85-90,1976. 23. PREFAUT, C., H. E. LUPI, AND R. ANTHONISEN. Human lung mechanics during water immersion. J. Appl. Physiol. 40: 320-323, 1976. 24. PREFAUT, C., M. RAMONATXO, Q. BOYER, AND G. CHARDON. Human gas exchange during water immersion. Respir. Physiol. 34: 307-318,1978. 25. RENNIE, D. W., P. DI PRAMPERO, AND P. CERETELLI. Effects of water immersion in cardiac output, heart rate and stroke volume of man at rest and during exercise. Med. Dell0 Sport 24: 223-228, 1971. 26. ROBERTSON, H. T., H. P. MCKENNA, AND M. P. HLASTALA. Influence of temperature and hemoglobin saturation on partitions coefficients for three gases used of inert gas elimination studies (Abstract). Physiologist 25: 268, 1982. 27. ROWELL, L. B., G. L. BRENGELMANN, AND J. A. MURRAY. Cardiovascular responses to sustained high skin temperature in resting man. J. Appl. Physiol. 27: 673-680, 1969. 28. SACKNER, M. A., G. MARKWELL, N. ATKINS, S. J. BIRCH, AND R. - J. FERNANDEZ. Rebreathing techniques for pulmonary capillary blood flow and tissue volume. J. Appl. Physiol. 49: 910-915, 1980. 29. VARENE, N., M. L. CHOUKROUN, R. MARTHAN, AND P. VARENE. Solubility of Freon 22 in human blood and lung tissue. J. Appl. Physiol. 66: 2468-2471, 1989. 30. WESTON, C. F., J. P. O’HARE, J. M. EVANS, AND R. J. M. CORRALL. Haemodynamic changes in man during immersion in water at different temperatures. Clin. Sci. Lond. 73: 613-616, 1987.
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (130.130.211.199) on July 30, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
CHOUKROUN,
MARIE-LUCE,
AND
PIERRE
VARENE.
Ad&&-
ments in oxygen transport during head-out immersion in water at different temperatures. J. Appl. Physiol. 68(4): 14751480, 1990.-Respiratory gas exchange was investigated in human subjects immersed up to the shoulders in water at different temperatures (T, = 25,34, and 40°C). Cardiac output (Qc) and pulmonary tissue volume (Vti) were measured by a rebreathing technique with the inert gas Freon 22, and O2 consumption (VO*) was determined by the closed-circuit technique. Arterial blood gases (Pa+ Pa& were analyzed by a micromethod, and alveolar gas (PAL,) was analyzed during quiet breathing with a mass spectrometer. The findings were as follows. 1) Immersion in a cold bath had no significant effect on &c compared with the value measured at T, = 34OC, whereas immersion in a hot bath led to a considerable increase in Qc. Vti was not affected by immersion at any of the temperatures tested. 2) A large rise in metabolic rate Vo2 was only observed at T, = 25°C (P < 0.001). 3)Arterial blood gases were not significantly affected by immersion, whatever the water temperature. 4) O2 transport during immersion is affected by two main factors: hydrostatic Above neutral temperature, Q2 pressure and temperature. transport is improved because of the marked increase in Qc resulting from the combined actions of hydrostatic counter pressure and body heating. Below neutral temperature, O2 transport is altered; an increase in O2 extraction of the tissue is even calculated. water temperature; rial blood gases
cardiac output;
oxygen consumption;
Cedex, France
25°C a decrease of VC and at T, = 40°C an increase of VC. At 25°C we observed a marked increase in 02 consumption (VOW), whereas in this case the only change in heart rate (HR) was a slight bradycardia. This last phenomenon seemsto have been first reported by Cannon and Keatinge (8), but it does not appear to have been investigated any further. The present study was therefore designed to investigate the effect of T, during head-out immersion on respiratory gas exchange. Three T, were chosen: 34°C (thermoneutral bath), 25OC (cold bath), and 40°C (hot bath). We measured Q, Vo2, and arterial blood gases(Pao, and Pace,). MATERIALS
AND METHODS
Subjects Eleven healthy volunteers (4 females, 7 males) were selected for the study. Informed consent was obtained. Subjects others than laboratory staff received a financial allowance. Their anthropometric data are given in Table 1. Most of the subjects are nonsmokers; only two are light smokers (subjects AML and BL). All have normal pulmonary function tests.
arte-
Procedure
alongwithpulmonaryand cardiocirculatory readjustments during head-out immersion in water, has been extensively studied at thermoneutral temperature (34°C). The hydrostatic. counter pressure causes an increase in cardiac output (Qc), central blood volume, and pulmonary arterial pressure (2, 4, 13) that is associated with ventilatory adjustments such as a marked reduction in functional residual capacity and a slight drop in vital capacity (VC) (1, 13, 23). Some authors (3, 11, 21) have observed alterations in gas exchange: a fall in arterial POT (Pao,) and an increase in alveolar-arterial difference (PAo,-Pao,), attributed to modifications in the distribution of the ventilation-perfusion (V/Q) ratio during immersion. However, humans are rarely immersed in water at thermoneutral temperature. Immersion generally takes place in colder water. As previously described by Kurss et al. (20) we have found that the temperature of the water (T,) during immersion affects pulmonary capacities (10); in comparison, using a thermoneutral bath, we measured at T, =
RESPIRATORYGASEXCHANGE,
0161-7567/90
$1.50 Copyright
Immersion was carried out in a vertical, cylindrical plastic tank containing -600 liters of water. The desired T, was obtained by mixing cold and hot water. Because of the large volume of water, T, remained approximately constant (+OJ”C) during the experiments. The subjects stood upright in the tank, immersed up to the shoulders, more precisely to the Louis-angle level. Body temperature was measured with a rectal probe (T,,), and cardiac activity was recorded on an electrocardiogram throughout the experimental session. On the same one-half day, a given subject was immersed at all three T, 34, 25, and 40°C. The T, were chosen in random order, and at least 20 min elapsed before the measurements. The order of the measurements in each situation was the following: 1) determination of vo2 and expired ventilation (VE), 2) collection of blood samples for blood-gas analysis, and 3) determination of Qc and pulmonary tissue volume (Vti). Before immersion, control values of each variable were obtained from the subjects standing in thermally neutral air.
0 1990 the American
Physiological
Society
1475
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (130.130.211.199) on July 30, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
1476
OXYGEN
TABLE 1. Anthropometric characteristics participating in the experiment Subjects
Sex
Age, yr
Height, m
AML MHB BC SD ES RC BJ LP LO BL BB
F F F M M M M M M M F
27 28 37 27 21 20 26 25 20 19 22
1.67 1.68 1.57 1.60 1.81 1.84 1.70 1.75 1.85 1.78 1.55
Means
2
SD
2525
1.71kO.11
TRANSPORT
of subjects
Weight, kg 49 53 47 57 68 96 65 63 65 72 45 62-1-14
Body Area, m* 1.53 1.59 1.44 1.59 1.73 2.19 1.86 1.77 1.86 1.89 1.40 1.71k0.23
Measurements Cardiac output. Qc was measured by a noninvasive method involving rebreathing of an inert gas, Freon 22 (CHF&l), according to Bonde-Petersen et al. (5). The rebreathing apparatus consisted of a 3-liter rubber bag connected to a three-way stopcock. The bag was filled with 2 liters of a gas mixture containing 4.9% Freon 22, 6.75% Ar, 39.8% O2 in NP. After a resting time of 20 min the subject was connected to the rebreathing bag at the end of a complete expiration. The rebreathing maneuver was maintained for 15 s at a respiratory rate of 20 breaths/min. During the rebreathing, the fractional concentrations of Freon, Ar, and 02 were analyzed continuously by a mass spectrometer (Spectralab M VG Gas Analysis Instrument) and recorded on paper (Gould ES 1000). The mass spectrometer was programmed to analyze a mixture of five gases: 02, COZ, Nz, Ar, CHF&l whose total concentration equals one. The calibration curves for each gas were determined with gas mixtures whose composition is known with an accuracy generally better than 0.1%. In the rebreathing bag, Freon 22 concentration was measured with a resolution of 0.010%. The data were collected only for the breaths during which there was minimal change in Ar concentrations. The first breaths were discarded because of the likelihood of incomplete gas mixing. The maneuver was terminated after 15 s to avoid recirculation. The pulmonary capillary blood flow (Qc) and the pulmonary tissue plus capillary blood volume (Vti + Vc) were calculated from the disappearance curve of Freon 22 according to the equation of Cander and Forster (7). The regression line was fitted with the expiratory and inspiratory Freon values as suggested by Sackner et al. (28). By knowing the Bunsen solubility coefficients of Freon 22 in blood and tissues, Qc was derived from its slope and (Vti + Vc) .from its intercept value. The Bunsen coefficients for Freon 22 have been determined in our laboratory (29). They are ablood = 0.804 t 0.049 cm3 STPD of Freon 22cmm3 blood*ATA-1 at 37°C = 0.635 t 0.091 cm3 STDP of Freon 22 crne3 and atissue tissue. ATAB at 37°C. Vti was calculated after subtraction of Vc, estimated from mean values previously measured in our laboratory
DURING
IMMERSION
on normal subjects under similar conditions (9). O2 consumption. iTo was measured in the steady state by the closed-circuit technique before the rebreathing maneuver. vo2 was expressed in STPD conditions. VE was also calculated from the tidal volume and the respiratory frequency given by the spirometer (Expirograph Gould-Godard) and was expressed in BTPS conditions. Arterial blood gases.Arterial blood gases were analyzed in blood samples collected from the hyperemized earlobe immediately before rebreathing by the subjects. The partial pressures in arterial blood, (Pao,, Pace,) were determined with a blood gas analyzer (Corning 173) and the arterial oxygen content (Cao,) with another blood gas analyzer (Co-Oxymeter, Instrumentation Laboratory). Appropriate corrections were made for Pao, and Pace, as a function of Tree Hemoglobin concentration ([Hb]) was also measured. The (PAo,-ho2) was calculated from Pao,, determined by micromethod, and the mean alveolar POP (PA~J. PAN, was estimated from the end-tidal 02 fraction measured with the mass spectrometer during quiet breathing before the rebreathing maneuver. Statistical analysis. All data obtained for each situation (air, Tw = 34”C, Tw = 25°C and T, = 40°C) were compared with Student’s t test for paired samples. RESULTS Temperature
With respect to the temperature moneutral bath, the mean core (AT,) were -0.55 t 0.42OC during bath and +l.Ol t 0.53OC during bath. Hemodynamic Data (Table
recorded in the thertemperature changes immersion in the cold immersion in the hot
2)
For the subjects standing in air, the mean &c was 4.75 t 0.53 dm3* min-‘. This corresponds to a cardiac index (CI) of 2.80 t 0.35 dm3* mine1 l m-? The mean HR was 95 t 14 beatsmin-‘, giving a mean stroke volume (SV) of 42 t 7 cm3. The mean Vti calculated after subtraction of the Vc was 456 & 177 cm3. Immersion in thermoneutral water (34°C) increased &c significantly (+18%, P < 0.05). This was essentially because of a marked rise in SV (+52%, P < 0.005). HR fell significantly (P < 0.01). Vti was not affected by thermoneutral immersion. The value of Qc during immersion in the cold water V = 25°C) was comparable with the value at T, = 34’C. Cardiac frequency was somewhat reduced but not to the extent previously observed in a similar situation (10). Vti was unchanged. Immersion in the hot water (T, = 40°C) induced a considerable increase in Qc (+41%, P < 0.005) because of a rise in HR, whereas SV was less than that observed in the thermoneutral bath, and Vti was unchanged. Gas Exchange Data (Table 3)
The mean 60, measured, when the subjects were standing in air, was 0.260 t 0.056 dm3 STPDemin-‘.
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (130.130.211.199) on July 30, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
OXYGEN
TABLE
TRANSPORT
DURING
1477
IMMERSION
2. Hemodynamic data in air and during immersion at different temperatures Immersion Air Qc, dm3 - min-’
CI, dm” . min-’ HR, beats
l
4.75kO.53 t
- m-’
T, = 34°C Ps
5.61tl.27 0.05w’NS 4
2.80k0.35
t
t
Vti,
cm3
cm3
75t13
456tl77 tNS
71t13
P 5 0.0001-
64t14
69t17
55&11
MC--
P 5 0.05 + NS-
513t145
565k130
551&275
463&126
411-+260
-+NS416t139 -1NS
,-NS-
NS
-
3. Gas exchange data in air and during immersion at different temperatures Immersion
Air T, = 34°C %JO~, dm3 STPD
\jE,
llOt15
P 5 0.0001___)
4
TABLE
3.89-eO.44
Af--
Ps 0.005-‘NS
547t154 tNSd’NS
l
3.43kO.64
4 (Vc + Vti),
--)
P 5 0.005
Ps 0.01-‘NS
42k7
6.71t1.50 P 5 0.005
4 SV, cm”
T, = 40°C
5.90t1.45 ,b
3.27k0.54
95-el4
min-l
T, = 25°C
l
min-’
dm3 BTPS .min-’
0.260t0.056 -
T, = 25°C
0.239t0.053 NS
A+
0.364tO.
160
P < 0.001 +-
8.3Ok 1.52
7.7kl.22
T, = 40°C 0.258kO.047 NS -
11.74k3.81
10.76k5.34
-NS-tPs0.005+-NSI Pa+
Torr
Torr
(PAo,-ho,),
Pace,, Torr
Hb, g/100
cm” blood
Cao,, cm3/100
cm” blood
91.50t9.52 -NS-t
94.27t9.15
22.5OklO.95 -NS--NSF’NS-
20.25k10.20
P s 0.05 91.27t10.05
89.32t9.90
24.52t14.62
22.87-1-8.92
NSF’NS-
36.60t3.75 -
NS --
35.62k3.52
15.91kl.89 c--------------
NS -
21.60-2.50 -
NS A-
15.63t1.88 t
21.22k2.53
Immersion in the thermoneutral bath led to a nonsignificant fall in VOW. There wag a marked rise in metabolic rate during cold immersion (VO, at 25°C = 0.364 t 0.160 dm3 STPD . minT1, P c 0.001). However, in the hot bath the change in Voz was not statistically significant with respect to that measured at T, = 34°C. VE was not affected by thermoneutral immersion. However, there was a significant rise in VE when the water temperature deviated from 34’C. In fact, there w.as considerable individual variation in the change in VE with change in bath temperature, especially for immersion at T, = 40°C.
34.50t3.90 NS --NS-
5 0.005
16.02&l --
21.68k2.20 NS -1NS-
35.25t4.27
.79
15.68Ifr1.67 NS A
21.27k2.32
Pao. and Paoo, and the Cao were not significantly altered by immersion at any of the temperatures studied. (PAO,-ho2) was not affected by the bath temperature. The only change in [Hb] was a slight but significant rise during immersion in the cold bath (16.02 g/100 cm” blood at T, = 25°C vs. 15.63 g/100 cm” blood at T, = 34°C; P < 0.005). DISCUSSION
Methodology
The noninvasive methods, such as Freon 22 rebreathing chosen here to measure Qc, are based on the as-
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (130.130.211.199) on July 30, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
1478
OXYGEN
TRANSPORT
sumption that the Qc equals to the pulmonary capillary shunt being neglected. blood flow, wit h the right-to-left This is generally true in normal subjects, the difference caused by the shunt being less than that introduced by the scatter of all the techniques. Our control Qc or CI values were sma ller than those reported by workers using methods such as acetylene invasive (2) or noninvasive rebreathing (4) and CO2 rebreathing (13). It is generally adm itted that the Qc values obta .ined with the rebreath1% technique are smaller with Freon 22 rather than acetylene (5), which is the reference gas. Nevertheless, because of its low toxicity and of its safe handling, we chose to use Freon 22 instead of acetylene, more especially as we were mainly interested in the variations of Qc in function of the bath temperature rather than its absolute value. Moreover another explanation for these differences in Qc might be put forward: one-th ird of our group was females, whereas in the above studies they consisted of male subjects in whom CI is higher. (Vti + Vc) was calculated from the intercept of the disappearance curve of Freon 22, time 0 of this curve being identified by the rise of the Freon concentration during the first inspiration. This method is somewhat approximate, but our control values in air were in good agreement with those obtained by authors using an acetylene-rebreathing method (28). The calculation of Qc and (Vti + Vc) necessitates to know the Bunsen solubility coefficients of Freon 22 in blood and tissue ( CYB,ati). The solubility of an inert gas is influenced by the tempe ratu re of the solvent. In our experiment, the changes in T re were too small to induce significant measurable changes of the solubility of Freon in pulmonary blood or tissue. Then the variations of with the temperature have been neablood and atissue, glected in this work. Another point to be discussed concerns the estimation of [Hb] and Ca 0, contents from a sample of capillary blood. As it is generally admitted, despite a careful sampling (hyperemized earlobe) the given Pao, and Cao, values were probably underestimated, especially in cold bath when a considerable skin vasoconstriction takes exists place. N ‘evertheless, because a close correlation between arterial and arterialized capillary Paz, or pco2, these values were sufficient ly reliable to support our discussion. Results Thermoneutral immersion. Numerous studies on the hemodynamic readjustments during thermoneutral immersion have confirmed the prediction of Gauer and Thron (14) that there is an increase in Qc associated with a rise in SV, with the HR being almost unchanged (2). These changes are attributed to improved venous return induced by a shift of blood from the periphery to the intrathoracic regions because of the hydrostatic counter pressure. There is, however, some lack of agreement on the extent of the increase; Arborelius et al. (2) reported a 30-35% rise in Qc, Begin et al. (4) a 20-40% increase, Farhi and Linnarsson (13) an increase of 40% during immersion up to the xiphoid and of 66% during immersion up to the chin, whereas we observed only a
DURING
IMMERSION
rise of 18%. According to Farhi and Linnarsson (13) the discrepancies can be explained by differences in the position of the body in the water (sitting, standing, or supine) and in the level of immersion (hips, xiphoid, shoulders, or neck). Cohen et al. (11) have suggested that in some subjects, immersion increases the shunt component, and so by using an inert gas rebreathing method, we may have underestimated the rise in Qc induced by the hydrostatic counter pressure. However, in our experiment, (PA@ao,) did not change significantly during immersion, and we concluded that the increase in shunt at T, = 34°C was not important, and so this would have had no effect on the determination of Qc. Hood et al. (17) and McArdle et al. (22) have not found such changes in Qc during immersion, whereas Rennie et al. (25) even observed a fall, but their experiments were not performed at exactly thermoneutral temperature. In our study, the rise in Qc was caused by a large rise in SV, with a marked fall in HR. This change in HR is generally attributed to a baroreflex induced by the increase in blood pressure related to the rise in Qc. Vti was not affected by thermoneutral immersion. Farhi and Linnarsson (13) found changes in equivalent lung tissue volume, which was largely attributed to alterations in Vc. Our values of Vti, after subtraction of Vc, which is known to be affected by immersion (9), are in good agreement with these findings. They indicate that there is no extravasation of fluid during . immersion. The slight but nonsignificant drop in iJo2 was probably caused by reduced metabolism in the immersed subjects. During immersion, postural muscles are quiescent because of the supporting hydrostatic counter pressure. The results on arterial blood gaseswere more surprising. Thermoneutral immersion up to the neck has generally been reported to lead to a decrease in Pao, associated with an increase in the (PAo.,-Pa& (11, 21). This iS attributed to an alteration ofAventilation-perfusion ratio distribution (3, 21). We did not observe such variations in Pao, or (PAo,-ho,). Two explanations can be found. First, the time elapsed between the measurements of Tie,, by use of the closed-circuit technique, and the blood sampling is -2 or 3 min. Then because of the possibility of 02 lingering in poorly ventilated zones, these variations may be avoided by high PAN, levels. However a second explanation can be found in the study of Prefaut et al. (24); they reported that “gas exchange modifications during immersion depend on position of end expiratory level in relation to closing volume,” this closing volume being largely dependent on the age of the subjects. In the present study our subjects, except for one, were ~28 yr old, and so owing to a reduced closing volume this effect is probably not very significant in our conditions. Then the most important effect during immersion was a rise in Qc leading to an improvement in gas exchange. Although Cao, was unchanged, Qc was significantly increased, whereas Tj02 was slightly lower during immersion which should lead to a rise in venous 02 content (Cvo,) and venous PO:! (PvoJ. Although the increase in Cvo, calculated at T, = 34°C was not statistically significant (Cvo, air = 15.76 t 2.45 cm3/100 cm3 blood; Cvo,
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (130.130.211.199) on July 30, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
OXYGEN
TRANSPORT
34°C = 16.76 t 2.34 cm3/100 cm3 blood), this improvement is consistent with the value measured by Farhi and Linnarsson (13) for immersion up to the xiphoid or chin. Immersion in cold water. During cold stimulation, like during immersion of a foot in cold water according to Cooper et al. (12), vasoconstriction should result in a rise in blood pressure and a drop in HR without any significant alteration in Qc. Our results are in agreement with these predictions. At T, = 25”C, Qc was not significantly increased compared with the value obtained at T, = 34OC. This was essentially caused by a lower cardiac frequency in response to the cold-induced vasoconstriction. However, this fall in cardiac frequency measured at T = 25°C was not as marked as that observed in a prWevious study (10). This may have been caused by individual differences in response. Keatinge and Evans (18) only observed such a fall in cool water (25°C); in colder water (15°C 5°C) the HR was markedly increased. Along with this cardiovascular readjustment during cold immersion,, the metabolic response leads to a large increase in Vo2. This last observation was reported by Cannon and Keatinge (8); the metabolic rate rose as soon as the T, dropped below 34°C the thermoneutral temperature in water. On the basis of these two observations, it can be calculated that there will be a larger arteriovenous difference in O2 (Pao,-Pvo,) during cool immersion (T, = 25°C) than during thermoneutral immersion (Pao,Pvo, 34°C = 4.28 t 0.87 cm3/100 cm3 blood; Pao,-Pvo, 25°C = 6.11 t 2.06 cm3/100 cm3 blood, P c 0.05). Because Pao, and Cao, are unchanged at T, = 25°C there should be a decrease in mixed venous POT. We calculated a mean decrease of Cvo, of 8.5%, which corresponds to an enhanced O2 extraction of the tissue (Pao,-Pvo,)/Ca,, (28% at T, = 25°C for 20% at T, = 34°C P 5 0.05). The alterations in ventilation observed at T, = 25°C did not induce a significant change in Pace,. In fact, the ventilatory response in cold water (T, = 25°C) varied considerably among the subjects. Some subjects hyperventilated and reduced their Pace,, whereas others showed little change in VE. The higher [Hb] in the blood during cold immersion is of interest. This can be related to cold diuresis. This hemoconcentration effect has also been investigated by Harrison et al. (16), who found that “skin cooling from normothermia caused hemoconcentration” because of the rise in capillary pressure resulting from vasoconstriction. If we consider the duration of our experimentation, the increase in Hb is more likely related to this last mechanism. Immersion in the hot bath. The rise in Qc results from both the hydrostatic counterpressure and the effect of body heating. In recumbent subjects wearing heated suits, whose hemodynamics are similar to those during head-out immersion, Rowe11 et al. (27) measured a 38% increase in Qc when the skin temperature was maintained at 34°C and a 112% increase at 40.9”C. Weston et al. (30), using a Doppler ultrasound technique, found a 121% increase in Qc during immersion at T, = 39OC. Our experimental conditions were comparable to the above CT = 40°C; AT,, = +l.Ol t 0.53”C), and the changes we obierved in &c (+41%), perhaps slightly underestimated,
DURING
1479
IMMERSION
are essentially in agreement with these results. The underestimate may stem from an enhanced shunt at T, = 4O”C, which is not taken into account by the inert gas method we used for measuring Qc. This excess Qc cannot be attributed to metabolic requirements, as \jo2 is almost constant, but is more likely to result from peripheral vasodilation at the higher temperature. It induces a decrease in the (Pao,-Pvo,) at T, = 4O’C. Because Cao, is unaffected by the bath temperature, an improvement in Cvo, (and hence Pvo,) may be expected at the higher temperature (Cvo, 34°C = 16.76 t 2.34 cm3/100 cm3 blood; Cvo; 40°C = 17.17 t 2.13 cm3/100 cm3 blood, NS). However, in front of this increase in Qc the question of a change in the distribution of V/Q ratio arises. Although there was little intersubject variation in the rate of increase of &, there was considerable variation in TjE: six subjects had a raised TjE, whereas in five there was little change. This may lead to an uneven distribution of v/Q ratio, such as an increase in the areas of low v/Q, which would affect gas exchange. From our measurements an answer cannot be given, but we did not observe a significant increase in (PAo,-Pa*,) at T, = 4O”C, although Pao, was slightly but not significantly reduced. In hot water, the most important effect was an improvement in mixed venous O2 content in relation to the increased Q. In conclusion, we demonstrated an interaction between two main factors affecting 02 transport during immersion: temperature and hydrostatic pressure. Farhi and Linnarsson (13) describe how “below neutral temperature, thermal and hydrostatic effects act in opposition.” Our results indicate that during cold immersion the effects of temperature outweigh the effects of hydrostatic counterpressure. The rise of Pvo, related to the increase of Qc during thermoneutral immersion is cancelled out at T, = 25°C and enhanced O2 extraction of the tissue is even observed (P 5 0.05). During immersion in hot water, heat-induced cutaneous vasodilation is the main factor, which leads to a marked increase in HR and Qc; 02 transport is improved, although in some subjects there may be a redistribution of the V/Q ratio. The authors thank Pierre Techoueyres for technical assistance and Nadine Capdeville for typing the manuscript. Address for reprint requests: M. L. Choukroun, Laboratoire de Physiologie, UER III, Universite de Bordeaux-II, 146 rue Leo Saignat, 33076 Bordeaux Cedex, France. Received
22 February
1989; accepted
in final
form
27 October
1989.
REFERENCES E., G. GUTNER, G. TORRI, AND H. RAHN. Respiratory during submersion and negative pressure breathing. J. Appl. Physiol. 21: 251-258, 1966. ARBORELIUS, M., JR., U. I. BALLDIN, B. LILJA, AND C. E. G. LUNDGREN. Hemodynamic changes in man during immersion with head above water. Aerosp. Med. 43: 592-598, 1972. ARBORELIUS, M., JR., U. I. BALLDIN, B. LILJA, AND C. E. G. LUNDGREN. Regional lung function in man during immersion with the head above water. Aerosp. Med. 43: 701-707,1972. BEGIN, R., M. EPSTEIN, M. A. SACKNER, R. LEVINSON, R. DOUGHERTY, AND D. DUNCAN. Effects of water immersion to the neck on pulmonary circulation and tissue volume in man. J. Appl. Physiol. 40: 293-299, 1976. AGOSTONI,
mechanics
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (130.130.211.199) on July 30, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
1480
OXYGEN
TRANSPORT
5. BONDE-PETERSEN, F., P. NORSK, AND Y. SUZUKI. A comparison between Freon and acetylene rebreathing for measuring cardiac output. Aviat. Space Environ. Med. 51: 1214-1221, 1980. 6. BOYER, J. T., J. R. E. FRASER, AND A. E. DOYLE. The haemodynamic effects of cold immersion. Clin. Sci. Lond. 19: 539~550,196O. 7. CANDER, L., AND R. E. FORSTER. Determination of pulmonary parenchymal tissue volume and pulmonary blood flow in man. J. Appl. Physiol. 14: 541-551, 1959. 8. CANNON, P., AND W. R. KEATINGE. Metabolic rate and heat loss of fat and thin man in heat balance in cold and warm water. J. Physiol. Lond. 154: 329-344, 1960. 9. CHOUKROUN, M. L., H. GUENARD, AND P. VARENE. Pulmonary capillary blood volume during immersion in water at different temperatures. Undersea Biomed. Res. 10: 331-342, 1983. 10. CHOUKROUN, M. L., C. KAYS, AND P. VARENE. Effects of water temperature on pulmonary volumes in immersed human subjects. Respir. Physiol. 75: 255-266, 1989. 11. COHEN, R., W. H. BELL, H. A. SALTZMANN, AND J. A. KYLSTRA. Alveolo-arterial oxygen pressure difference in man immersed up to the neck in water. J. Appl. Physiol. 30: 720-723, 1971. 12. COOPER, K. E., S. MARTIN, AND P. RIBEN. Respiratory and other responses in subjects immersed in cold water. J. Appl. Physiol. 40: 903-910,1976. 13. FARHI, L. E., AND D. LINNARSSON. Cardiopulmonary readjustments during graded immersion in water at 35OC. Respir. Physiol. 30: 35-50, 1977. 14. GAUER, 0. H., AND H. L. THRON. Postural changes in the circulation. In: Handbook of Physiology. Circulation. Washington, DC: Am. Physiol. Sot., 1965, sect. 2, vol. III, p. 2409-2439. 15. GOFF, L. G., H. F. BRUBACH, H. SPECHT, AND N. SMITH. Effect of total immersion at various temperatures on oxygen uptake at rest and during exercise. J. Appl. Physiol. 9: 59-61, 1956. 16. HARRISON, M. H., R. J. EDWARDS, L. A. COCHRANE, AND M. J. GRAVENEY. Blood volume and protein responses to skin heating and cooling in resting subjects. J. Appl. Physiol. 54: 515-523, 1983. 17. HOOD, S. K., R. H. MURRAY, C. W. URSCHEZ, J. A. BOWERS, AND J. K. GOLDMANN. Circulatory effects of water immersion upon human subjects. Aerosp. Med. 39: 579-584, 1968. 18. KEATINGE, W. R., AND M. EVANS. The respiratory and cardiovascular responses to immersion in cold and warm water. Q. J. Exp. Physiol. 46: 83-94, 1961.
DURING
IMMERSION
19. KNIGHT, D. R., AND S. M. HORVATH. Effect of hydrostatic pressure on plasma concentrations of norepinephrine during cold water immersion. Undersea Biomed. Res. 14: l-10, 1987. 20. KURSS, D. I., C. E. G. LUNDGREN, AND A. J. PASCHE. Effect of water temperature and vital capacity in head-out immersion. Underwater Physiology VII, edited by A. J. Bachrach and M. M. Matcen. Bethesda, MD: Undersea Med. Sot., 1981, p. 297-301. (Proc. 7th Symp. Underwater Med.) 21. LOLLGEN, H., G. VON NIEDING, H. KREKELLER, U. SMIDT, K. KOPPENHAGEN, AND H. FRANK. Respiratory gas exchange and lung perfusion in man during and after head out water immersion. Undersea Biomed. Res. 3: 49-56, 1976. 22. MCARDLE, W. D., J. R. MAGEL, AND G. R. LESMES. Metabolic and cardiovascular adjustments to work in air and water at 18, 25 and 33°C. J. Appl. Physiol. 40: 85-90,1976. 23. PREFAUT, C., H. E. LUPI, AND R. ANTHONISEN. Human lung mechanics during water immersion. J. Appl. Physiol. 40: 320-323, 1976. 24. PREFAUT, C., M. RAMONATXO, Q. BOYER, AND G. CHARDON. Human gas exchange during water immersion. Respir. Physiol. 34: 307-318,1978. 25. RENNIE, D. W., P. DI PRAMPERO, AND P. CERETELLI. Effects of water immersion in cardiac output, heart rate and stroke volume of man at rest and during exercise. Med. Dell0 Sport 24: 223-228, 1971. 26. ROBERTSON, H. T., H. P. MCKENNA, AND M. P. HLASTALA. Influence of temperature and hemoglobin saturation on partitions coefficients for three gases used of inert gas elimination studies (Abstract). Physiologist 25: 268, 1982. 27. ROWELL, L. B., G. L. BRENGELMANN, AND J. A. MURRAY. Cardiovascular responses to sustained high skin temperature in resting man. J. Appl. Physiol. 27: 673-680, 1969. 28. SACKNER, M. A., G. MARKWELL, N. ATKINS, S. J. BIRCH, AND R. - J. FERNANDEZ. Rebreathing techniques for pulmonary capillary blood flow and tissue volume. J. Appl. Physiol. 49: 910-915, 1980. 29. VARENE, N., M. L. CHOUKROUN, R. MARTHAN, AND P. VARENE. Solubility of Freon 22 in human blood and lung tissue. J. Appl. Physiol. 66: 2468-2471, 1989. 30. WESTON, C. F., J. P. O’HARE, J. M. EVANS, AND R. J. M. CORRALL. Haemodynamic changes in man during immersion in water at different temperatures. Clin. Sci. Lond. 73: 613-616, 1987.
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (130.130.211.199) on July 30, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
