Pflfigers Arch. 359, 147--155 (1975) 9 by Springer-Verl~g 1975

Blood Affinity for Oxygen in Experimental Hemorrhagic Shock with Metabolic Acidosis F. Leeompte, H. Aberkane, E. Azoulay, M. Muffat-Joly, and J. J. Pocidalo Unit4 de Recherche de R6animation-INSERM (Usa) Paris, France Received April 22, 1975

Summary. This study was designed to evaluate, in vivo, the effect of a severe non-respiratory acidosis on hemoglobin oxygen transport. Oxygen affinity of hemoglobin, Bohr effect, Hill's number and red cell 2,3-DPG were evaluated during experimental hemorrhagic shock in dogs. Three periods were considered: control, hypotension (mean arterial pressure 60 mm Hg for 2 hr 30 min) and blood replacement. There was no significant change in erythrocyte 2,3-DPG following hemorrhagic hypotension but ATP increased significantly. n, the Hill number (2.6), was not changed by in vivo acidosis (pH 7.1). Respiratory Bohr coefficient (Bcou) corresponding to pile variations was drastically reduced (control Bee2 ~ 0.55, acidosis Beoe ~-- 0.31, blood replacement Bees = 0.35). Ps0(~.a)was not modified significantly by hemorrhagic acidosis. It is unlikely Sha~ variations of blood affinity for oxygen play a major role in oxygen delivery during early experimental hemorrhagic shock. Key words: 2qon-Respiratory Acidosis -- Oxygen Transport -- Oxygen Dissociation Curve -- 2,3-DPG -- Hemorrhagic Shock. Hemorrhagic shock is characterized b y failure of oxygen transport from environmental atmosphere to tissues, involving oxygen debt and cellular hypoxia responsible for death [9]. Simultaneously a metabolic, mainly lactic, acidosis occurs [8], which possibly changes the oxygen affinity of hemoglobin, shifting the oxyhemoglobin dissociation curve (ODC) to the right. Until recent work indicated the influence of 2,3-diphosphoglyeerate (2,3-DPG) on oxygen affinity, the respiratory function of blood in shock received little attention. The inverse relationship between oxygen affinity of hemoglobin and intraerythrocyte 2,3-DPG concentration is now well documented [5, 7]. A close correlation has been demonstrated recently between plasma p H and 2,3-DPG in non-respiratory acid-base disorders [ 2 - - 4 ] . These opposite effects of p H on the hemoglobin affinity for 10.

148

F. Lecompte et al.

oxygen are of g r e a t physiological interest, especially i n shock with s u s t a i n e d blood acidosis. T h e p u r p o s e of the p r e s e n t s t u d y i n dogs with hemorrhagic shock was to i n v e s t i g a t e the effect of m e t a b o l i c acidosis o n t h e ODC, to delineate the respective role of h e m o g l o b i n ligands (particul a r l y I-I+ ions a n d 2,3-DPG) i n the r e g u l a t i o n of o x y g e n affinity, a n d to consider t h e possible effects of a n y change o n o x y g e n t r a n s p o r t .

Material and Methods Procedure. Ten healthy mongrel dogs, weighing 15 to 22 kg, were prepared as described previously [20]. The dogs were anesthetized and respiratory quotient (RQ), arterial pit (pHa), O3 and COs partial pressures (Pao2 and Paeoe) were continuously controlled by mechanical ventilation. Arterial and venous catheters were positioned under X-ray control. The experimental model selected for hemorrhagic shock was close to that of Wiggers [26]. Blood was allowed to flow via the arterial catheter at a rate of 50 ml. rain-1 into a reservoir at variable height. The hemorrhage was stopped when the arterial pressure reached 60 mm Hg. This level was maintained throughout the experiment by changing the height of the reservoir; at that time the procedure included two phases: first, hemorrhage continued until 50 to 60 ~ of blood volume was shed; secondly, blood flowed back to the dog spontaneously and progressively. The hemorrhagic hypotension was maintained until spontaneous reinfusion which balanced almost 400/0 of maximal hemorrhage, representing 20 to 25 ~ of the total blood pool; this took about two and a half hours. The remaining blood was then reinfused at the same flow rate as during bleeding. Thus three steps could be determined for the study: Step 0: control period, before hemorrhage. Step 1: end of hemorrhagic hypotension. Step 2: about 30 rain after the initial blood volume was restored. In each step simultaneous samples were withdrawn: from the pulmonary artery (mixed venous blood) to determine ODC, Bohr coefficient (B) and pressure for half saturation of hemoglobin with oxygen (Pa0), and from the abdominal aorta for measurements of the arterial blood acid-base status. Two series of experiments were achieved in two different groups of dogs: one to establish the ODC prior to and at the end of the hemorrhagic hypotension (5 dogs), the other to evaluate the Bohr coefficient and Ps0 in each of the three periods (5 dogs). Measurements, 1. Acid-base status and blood gases: pH and blood gas partial pressures (Poe and Peoe) were determined as described previously [23]. Arterial and mixed venous oxygen contents (Caoe and C~oe) were measured using a eoulometric method (Lexington Instrument C~ [24]. Arterial blood lactate was determined enzymatieally [12]. 2. Erythrocytic parameters: Hemoglobin, 2,3-DPG and ATP concentrations, red cell count and hematocrit were determined as described previously [23]. 3. The oxygen dissociation curve was established by equilibrating each sample with water--saturated gas mixtures of known 02 and CO2 contents (WSsthoff gas-mixing pump). Four mixtures were utilized with constant Fcoe (0.055) and varying -Foe (0.04, 0.05, 0.06, 0.07). Poe and Pcoe were calculated using the barometric pressure. After a 20 rain equilibration pile, Cos and hemoglobin concentration were determined in duplicate in each sample.

Blood Affinity for Oxygen in Shock The binding mixture The

149

oxyhemoglobin saturation (S) was calculated using the maximum oxygen /cmax capacity ~ HbO,/ measured after equilibration of a sample with a gas containing 95 ~ 02 and 5 ~ CO~. Hill number, n, was then determined using the equation: S log 1 - - S

--nlogPo~-I-apH~+b,

where a and b are constants, and ptI~7 the tonometrie pH. 4. Ps0 and Bohr effect: S was measured at fixed Po2 (~28.5 Tort) and varying Peo2 ( ~ 2 0 , 38, and 76 Torr) using the methods just described. Ps0 was calculated at actual p H from S the linear equation log 1 -- S -- n (logPo2 -- log Ps0), n being the value identified in the former experimental group, namely 2.6. The Bohr coefficient (Peo~ variations) was the slope of the equation log Pa0 ~ - - B p ~ -~ K (K is a constant). Ps0 could then be expressed at standard p H (7.4).

Statistical Analysis. Mean values and standard error of the mean ( S D / ] / ~ were compared during the three periods using the "t" test for paired data, and a difference was considered to be significant if the P value was less than 0.05. Regression equations were calculated using the least-square method and significance of the correlation coefficients was read from Fisher's and Yates' tables.

Results T a b l e 1 s h o w s t h e v a r i a t i o n s o f e r y t h r o c y ~ i c p a r a m e t e r s in b o t h sets of e x p e r i m e n t s . Table 1. Mean values (~) and standard error of the mean (SD/~/~) of ery~hrocy~ic parameters before hemorrhage (0), after hypotension (1), and after blood replacement (2). Determinations were performed in 10 dogs at steps 0 and 1, and in 5 dogs at step 2

Hb g" 100m1-1 Red cell count 9 [ z 1 - 1 Hematoerit % MCHC g" 100ml -1 2,3-DPG tool 9tool -1 Hb ATP tool 9tool -1 Hb

0 Control

1 Hemorrhagic hypotension

2 After blood replacement

16.54 0.80 6453.108 4-316.10 a 47.9 4- 2.5 34.9 4- 0.2 1.072 4- 0.04 0.135 4- 0.011

15.22 * 4- 0.80 6340.103 ~=313.10 a 46.5 • 2.5 32.8 ** 4- 0.3 1.048 4- 0.04 0.150"* 4- 0.013

17.13 4- 2.03 7039.10 a 4-758.10 a 52 4- 6 32.7 * 4- 0.3 1.01 4- 0.03 0.147" 4- 0.017

~

* P < 0.05; ** P < 0.01.

150

F. Lecompte et al.

Mean corpuscular hemoglobin concentration (MCHC) decreased (P < 0.05) after hemorrhage a n d remained low after blood reinfusion while hemoglobin concentration (P < 0.01) a n d h e m a t o c r i t (n.s.) decreased only in the early stage. 2,3-DPG variation was n o t significant t h r o u g h o u t the experiment b u t A T P was significantly increased at steps 1 (P < 0.01) and 2 (P < 0.05). These results remained valid even when the gToups were tested separately. Table 2 summarizes the d a t a concerning acid-base status a n d blood lactate concentration. The h y p e r l a c t a t e m i a of the metabolic acidosis was n o t completely corrected after blood reinfusion. Table 3: A good arterial oxygenation level is evidenced here. The decrease of Pq)o2 a n d S~ou during hypovolemia resulted from the increase o f the difference in arterio-venous 03 content, which was induced b y a fall in cardiac output. The m a x i m u m oxygen combining capacity of 1 g

Table 2. Acid-base status and blood lactate. ~Iean values and standard error of the mean (SD/}/n). The three periods and the number of dogs are the same as in Table 1

0 Control 1 hemorrhagic I-Iypotension 2 After blood replacement

pHa

Paeo~ Torr

ttCO 3m ~ . 1-~

Lactate raM. 1-1

7.41 i 0.01 7.14 4- 0.03 7.24 • 0.04

32.8 ~_• 0.7 29.1 ~ 0.8 26.4 q- 1.6

20.98 :t: 0.93 9.84 :1:0.74 11.46 =~ 1.33

2.1 J: 0.8 7.2 ~: 1.5 4 =]=0.9

Table 3. Respiratory status. Mean values and standard error of the mean (SD/]~-) of PaOu and P~o~, oxygen partial pressures in arterial and mixed venous blood (pulmonary artery), respectively, of S~o~ (hemoglobin oxygen saturation of mixed max venous blood), and the ratio Cl~bOJ[Hb ] which represents the maximum oxygen combining capacity of lg of hemoglobin

0 Control

Pvo~ Torr

S~o~ ~

CHboJ[l{b] ml 02" g-1 ttb

83.7 1.7 88.6 :]: 2.1 86.9 :]: 4.0

43.2 :~ 1.8 26.3 :]: 1.3 33.2 :1:2.0

68.9 :[: 2.1 18.5 :]: 1.6 32.6 :J: 9.7

1.34 -~0.02 1.33 :]:0.02 1.30 ~:0.04

• 1 Hemorrhagic hypotension 2 After blood replacement

max

-Pao~ Torr

Blood Affinity for Oxygen in Shock

151

WrHbl~ h e m o g l o b i n /cmax x }ibOa//l_ A! was t h e s a m e in t h e t h r e e p e r i o d s : p r o b a b l y n o t m u c h f u n c t i o n a l l y a b n o r m a l h e m o g l o b i n was f o r m e d d u r i n g acidosis. T a b l e 4 i n d i c a t e s t h a t t h e H i l l n u m b e r , e v a l u a t e d in t h e e n t i r e first set o f dogs, was n o t modified b y acidosis (likewise, no difference could be d e t e c t e d b e t w e e n t h e i n d i v i d u a l values of n f r o m calculations perf o r m e d in each dog). Correlation coefficients of t h e regression e q u a t i o n s were h i g h l y significant. A t t i m e 0, t o n o m e t r i e pI-I ( p H i ) equalled p H a ; a t t i m e 1, i t b e c a m e s l i g h t l y m o r e acid. T a b l e 5 gives t h e values of t h e ]3ohr coefficient d e t e r m i n e d in t h e whole second set o f dogs. There was a significant a n d i m p o r t a n t decrease following h e m o r r h a g e which p e r s i s t e d a f t e r reinfusion, t)50, expressed a t a c t u a l p g (Ps0(~)), was increased a t steps 1 a n d 2, b u t P5o(7.4) r e m a i n e d ~nehanged. N o significant c o r r e l a t i o n could be f o u n d b e t w e e n Ps0 a n d ery~hrocytie p a r a m e t e r s , p a r t i c u l a r l y 2 , 3 - D P G a n d MCHC. Table 4. Hill's number (n) prior to and after metabolic acidosis. ~ was calculated from the regression equation: 8 log 1 -- S -- n log Pc2 ~- ap}I~, -k b, a and b being constants. 5 Individual values were obtained at each step. r is the correlation coefficient. pHi, is the tonometrie pH, p H a is the arterial blood p H at sampling time

0 Control 1 Hemorrhagic hypotension

pHa

pHT

n

r

P~o Torr

7.408 • 0.027 7.118 ~= 0.048

7.402 -b 0.025 7.056 • 0.035

2.66

0.98

2.63

0.94

28.2 J= 0.9 43.58 =k 1.90

Table 5. ]3ohr coefficient (]3) prior to and after hemorrhagic hypotension and after blood replacement. ]3 is the slope of the regression line: tog P~o ~ --]3pH ~- K. P~o is calculated at p H a (Ps0) or a~ p i t 7.4 (P50(7.4)); r is the correlation coefficient pHa ]3 m q- SD/~/n

r

Pso Torr

P5o(7.4) Tort

• SD/~/~ ~ =LSD/V~ 0 Control

7.406

-- 0.55

-- 0.971

7.084

-- 0.31 *

-- 0.965

7.239 4-0.035

-- 0.39"

-- 0.955

• 1 Hemorrhagic hypotension 2 After blood replacement

•

* P ~ 0.05; *** P ~ 0.001.

29.23 J= 0.58 37.51 *** • 1.31 34.02"** ~ 1.31

29.47 -b 0.77 30.51 ~: 0.44 29.47 =]= 0.41

152

F. Lecompte et al.

Discussion This study demonstrates that a drastic acidosis generated by hemorrhagic shock produces a rightward shift of the ODG without changing its slope since n does not vary. The Bohr coefficient was significantly decreased, while neither erythrocy~e 2,3-DPG nor Ps0(7.4)varied significantly. The lack of significant variatiorm in erythrocyte 2,3-DPG after a 2-hour hemorrhagic shock was surprising. I t is well known that, in vivo, sustained acidosis, whether respiratory [1,15,16] or not [2,11], results in a decrease of erythrocyte 2,3-DPG. However, Iqaylor et al. [19] found no change of this metabolite during hemorrhagic sheck in the rhesus monkey; but it should be noted that no significant modification of p H a occurred in their experiments, owing to a complete ventilatory compensation. On the other hand Proctor et al. [21] gave results close to eurs, since they documented no 2,3-DPG variation throughout hemorrhagic shock in the baboon, despite a fall in pH similar to the present data. Although an influence of a lactic acidosis in red cell energetic metabolism cannot be excluded [17] in our experimental conditions, pH was probably not sufficiently depressed during a time long enough to induce a measurable decrease in the erythrecyte 2,3-DPG. Intraerythrecytic ATP concentration increased slightly but significantly; this weuld mean that the energy linked to the adenylic system was preserved. Such data were already found in hypercapnic dogs [1] and guinea pigs [13]. The exact mechanism ruling ATP preservation is unknown; possible changes in the rate of synthesis as well as in the rate of utilization should be considered. In the present work a non-respiratory acidosis induced no change in the mean Hill number, thus the slope of the ODG was not modified despite sustained acidosis. A similar conclusion ensued in hypercapnie acidosis [1]. pH independence of the slope of the oxygen equilibrium curve has long been accepted as one ef the characteristic features of mammalian hemoglobins. Recently Tynma et al. [25] demonstrated that the maximum slope of Hill's regression line decreases while pH increases; moreover the magnitude of a ODC shift induced by pH depends upon the oxygen saturation. Such results agree with the recent observation of Garby et al. [10] and Meier et al. [14] according to which the magnitude of the Bohr effect in human whole blood markedly depends upon the range of So2. In the present experimental conditions, i.e. pH ranging from 7.4 to 7.1 and Sos corresponding to the medium part of the ODC, the use of a linear regression equation, whose correlation coefficient is very high, was considered justified. Hence the absence of a significant variation of the mean n, when calculated in these conditions, allowed to conclude that pH variations, which actually correspond to a severe pathological acidosis, do not modify the shape of the ODG.

Blood Affinityfor Oxygenin Shock

153

The low value of the Bohr effect induced by fixed acid (BvA) is well known. Wrannc et al. [27] found the A log Pso/A pH ratio in fresh whole blood, when titrated with NaOH and HC1 at constant Pco2, to be 0.37, i.e. lower than when pH was changed by Co2 variation (Bco~ ---- 0.52). Such a BVA coefficient agrees with the value of 0.40 reported by Naeraa et al. [18]. In the present work, the Bohr coefficient (Bco2) was identified by in vitro CO~ titration of blood whose acid-base balance had been previously changed, in vivo, towards metabolic acidosis. During this acidosis, the Bco~ value (0.31) was significantly lower than the control value (0.55), at least in the present experimental conditions. However, two points need further consideration: --first, Bco2 was evaluated at pile. But for an accurate determination of the actual Bohr effect, knowledge of the red cell pH (pHi) would be necessary [6]. Since the hydrogen ion distribution across the red cell membrane is subjected to the Donnan equilibrium, the relationship between the intraeellular (Bi) and extracellular (Be) Bohr coefficients must likewise be dependent on this phenomenon [6,14]. Bi could not be calculated in the present work because it was not possible to measure pHi in dog blood using the classical freezing-thawing hemolysis technic; freezing of the red cells produces micro-crystals in the crythrolysate. It would be of interest to know whether metabolic acidosis also decreases the Bohr effect when referring to pHi. If it were so, the reduction of Bco2 induced by fixed acid (at constant 2,3-DPG concentration) could be best explained by diminution of carbamate formation resulting from acidosis [22] ; secondly, there might be Bohr effect variations depending on the oxygenation level [10,14]. It cannot be assumed that the Bco2 value, identified in acidotic blood, is valid over the whole range of oxygen saturations. Its use should be limited to the saturation range as defined in the first experimental series where n was found unchanged by acidosis. It would be of interest to study the highest and the lowest part of the ODC, especially in human blood, for the reasons just explained. I)50 expressed at actual arterial pH was increased (Table 5) during shock and after blood replacement; this was expected since a nonrespiratory pH decrease induced a rightward shift of the ODC. When standardized at pH 7.4, using the previously determined Beco2 value, Ps0 remained unchanged. But it cannot be assumed that the affinity of HbO~ is constant during the lactic acidosis of hemorrhagic shock, without determining first the 1)50 expressed at p h i instead of P50(7.4), through a Bico~ factor. However, because of the conservation of 2,3-DPG concentration, intraerythroeytie oxygen affiniby of hemoglobin was probably not much modified in the course of metabolic acidosis in shock. The anaerobic metabolism yielding lactic acid and consequently metabolic acidosis is commonly considered to be a result of the reduced

154

F. Lecompte e$ al.

oxygen t r a n s p o r t in hemorrhagic shock. F r o m previous studies carried, out in dogs, there is evidence of a consistant decrease in oxygen consumption, suggesting t h a t total o x y g e n delivery is inadequate [9]. The p a r t t a k e n b y the changes in hemoglobin oxygen t r a n s p o r t is p r o b a b l y v e r y slight as witnessed b y the conservation of b o t h the ODC shape a n d red cell 2,3-DPG concentration. F u r t h e r investigations are required before definite conclusions can be drawn. I t would be interesting to k n o w w h a t happens t o the respirat o r y B o h r effect in blood rendered acid b y lactate overproduction: first the intracellular p H and secondly low saturation levels (since a fall in cardiac o u t p u t considerably lowers S~o2) should be considered.

Re~erences 1. Aberkane, H., Azoulay, E., Lecompte, F., Muffat-Joly, M., Pocidalo, J. J.: Transport de l'oxyg~ne par l'h6moglobine au cours de l'hypercapnie aigu~ exp6rimentale. ]3ull. Physic-Path. Resp. 11, 179--192 (1975) 2. Astrup, P. : Red cell pH and oxygen affinity of hemoglobin. New Engl. J. IVied. 281, 202--203 (1970) 3. Astrup, P., Rorth, M., Thorshauge, C.: Dependancy on acid-base status of oxyhemoglobin dissociation and 2,3-diphosphoglycerate level in human erythrocyte. In rive studies. Scand. J. clin. Lab. Invest. 26, 47--52 (1970) 4. ]3ellingham, A.J., Defter, g.C., Lenfant, C.: Regulatory mechanisms of hemoglobin oxygen affinity in acidosis and alcalosis. J. din. Invest. 50, 700--706 (1971) 5. ]3enesch, R., ]3enesch, R. E. : The effect of organic phosphates from the human erythrocyte on the allosteric properties of hemoglobin. ]3iochem. biophys. Res. Commun. 26, 162--167 (1967) 6. ]3ursaux, E., Freminet, A., Poyart, C. F.: The Bohr effect, the Donnan equilibrium and the estimation of Ps0 in human whole blood. Bull. Physic-Path. Resp. 8, 755--768 (1972) 7. Chanutin, A., Curnish, R. R. : Effect of organic and inorganic phosphates on the oxygen equilibrium of human eryChroeytes. Arch. ]3iochem. ]3iophys. 121, 96--102 (1967) 8. Cloutier, C. T., Lowery, ]3. D., Carey, L. C.: Acid-base disturbances in hemorrhagic shock. Arch. Surg. 98, 551--558 (1969) 9. Crowell, J. V., Smith, E. E. : Oxygen-deficit and irreversible hemorrhagic shock. Amer. J. Physiol. 206, 313--316 (1964) 10. Garby, L., Robert, M., Zaar, ]3.: Proton and earbamino-linked oxygen affinity of whole blood. Aeta physiol, scand. 84, 482--492 (1972) 11. Guest, G. M.: Organic phosphates from the blood and mineral metabolism in diabetic acidosis. Amer. J. Dis. Child. 64, 401--408 (1942) 12. Hohorst, H. J.: L(+) lactate det~ermination with lactic deshydrogenase and DPI~. In: H.U. ]3ergmeyer: Methods of enzymatic analysis, lgew York: Academic Press 1965 13. Jacey, M.J., Schaefer, K. E.: The effects of chronic hypercapnia on blood phosphofructokinase activity and the adenine nueleotide system. Respir. Physiol. 16, 267--272 (1972)

Blood Affinity for Oxygen in Shock

155

14. Meier, U., B6ning, D., Rubenstein, It. J. : Oxygenation dependent variations of the Bohr coefficient related to whole blood and erythrocyte pit. Effect of lactic and carbonic acid. Pfl/igers Arch. 349, 203--213 (1974) 15. Messier, A. A., Schaefer, K. E. : The effect of chronic hypercapnia on oxygen affinity and 2,3-DPG. Respir. Physiol. 12, 291--296 (1971) 16. Messier, A.A., Schaefer, K . E . : The Bohr effect in chronic hypercapnia. Respir. Physiol. 19, 26--34 (1973) 17. Minakami, S., Yoshikawa, H.: Studies on erythrocyte glycolysis. III. The effects of active cation transport, p H and inorganic phosphate concentration on erythroeyte glycolysis. J. Biochem. 59, 145--150 (1966) 18. Naeraa, N., Petersen, E. S., Boye, E., Severinghaus, J. W.: pH and molecular C02 components of the Bohr effect in human blood. Seand. J. clin. Lab. Invest. 18, 96--102 (1966) 19. Naylor, B.A., Welch, M. H., Schafer, W.A., Guenter, C.A.: Blood affinity for oxygen in hemorrhagic and endotoxic shock. J. appl. Physiol. 8~, 829--833 (1972) 20. Poyart, C., Gaudebout, C., Blayo, M.C., Vallois, J.M., Pocidalo, J. J.: Adequate cardiorespiratory functions in mechanically ventilated dogs. Respir. Physiol. 9, 318--329 (1970) 21. Proctor, H. J., Lentz, T. R., Johnson, G. : Alterations in Baboon erythrocyte 2,3-diphosphoglycerate concentration associated with haemorrhagio shock and resuscitation. Ann. Surg. 174, 923--931 (1971) 22. Rossi-Bernardi, L., Roughton, F. J . W . : The specific influence of carbon dioxide and carbamate compounds on the buffer power and Bohr effect in human haemoglobin solution. J. Physiol. (Lend.) 189, 1--29 (1967) 23. Sinet, M., Azoulay, E., Blayo, M. C. : Aftinit6 du sang pour l'oxyg~ne et 2,3-DPG ehez l'homme, la femme et la femme enceinte. Bull. Physic-Path. Resp. 10, 419--421 (1974) 24. Sinet, M., Merlet, C., Joubin, C., Blayo, M.C.: M~thode coulom6triqne de mesure du contenu en oxyg6ne sanguin. Validation par comparaison a v e c l a m6thode manom6trique de Van Slyke. Roy. Europ. Et. Clin. Biol. 18, 1007-1010 (1972) 25. Tyuma, I., Kamigawara, Y., Imai~K.: pH dependence of the shape of the hemoglobin-oxygen equilibrium curve. Biochim. biophys. Acta (Amst.) 810, 317--320 (1973) 26. Wiggers, C. J.: Physiology of Shock. New York: The Commonwealth Fund. 1950 27. Wrarme, B., Woodson, R. D., Better, J. C. : Bohr effect: interaction between t{+, COs and 2,3-DPG in fresh and stored blood. J. appl. Physiol. 39, 749--754 (1972) F. Lecompte INSERM (U,~) HSpital Claude Bernard 10, Avenue de la Porte d'Aubervilliers F-75019 Paris, France