THE OXYGENDISSOCIATION CURVEIN ANEMIA OF VARIOUSTYPES1

By A. C. KENNEDYANDD. J. VALTIS 2

(From the University Department of Medicine, The Royal Infirmary, Glasgow)

(Submitted for publication March 1, 1954; accepted June 23, 1954)

It is a clinical observation of long standing thatthe patient who has become anemic gradually, isoften well compensated for the lower hemoglobinlevel he possesses. Thus it is not uncommon foran individual with a hemoglobin level well below50 per cent, due to pernicious anemia or hypo-chromic anemia, to carry on his normal activities.It is generally believed that the compensatory ad-justment is in part due to increased utilization ofoxygen by the tissues and in part to circulatoryadaption. It has also been suggested, from stud-ies of the gas transport of red cells in anemia(1-4), that it may in part be due to displacementof the oxygen dissociation curve to the right ofnormal whereby the amount of oxygen releasedto the tissues is increased; the number of patientsobserved in these investigations is small probablybecause of the difficulties and time-consuming na-ture of the methods. The fullest report is that ofRichards and Strauss (2) who indicated a shiftto the right of the oxygen dissociation curve ap-parent, however, only at the abnormal plasma pHof 7.64. Only the reports of Dill and his co-workers (3) and Isac, Matthes, and Yamanaka(4), demonstrate a shift to the right of the oxygendissociation curve at the standard plasma pH of7.4. There are no observations, as far as we areaware, on the oxygen dissociation curve at variousstages after recovery from anemia. Many of thepatients in these early reports had received wholeblood transfusions before the blood gas studiesand, as we have shown, this produces an alterationin the oxygen dissociation curve of the anemic re-cipient (5). Various hypotheses, many of themcontradictory, have been offered regarding thecause of the beneficial shift of the oxygen dissocia-

1Much of the cost of apparatus and materials wasprovided by the Rankin Fund of the University of Glas-gow.

2Enabled to undertake this work by scholarships fromthe E.C.A. for Greece and from the University ofThessaloniki.

tion curve in anemia but none of them affords asatisfactory explanation for the phenomenon.

For these reasons a further and more extensivestudy of the oxygen dissociation curve in anemiaappeared warranted. The present paper describesthe oxygen dissociation curves of twenty-nine in-dividuals suffering from various types of anemia.Observations on the position of the oxygen dis-sociation curves at various times during and afterrecovery from the anemia are also presented.

MATERIAL ANDMETHODS

The subjects of this study comprised eleven patientswith megaloblastic anemia in relapse and three in thera-peutic remissions three patients with nutritional hypo-chromic anemia, three patients with hypochromic anemiasecondary to chronic hemorrhage, two patients withaplastic anemia, four patients with anemia associatedwith chronic nephritis, acute leukemia, reticulosis, andscurvy respectively, and three patients with hemolyticanemia. Many of the observations were made beforespecific therapy was administered and subsequent curveswere performed at different periods during and afterrecovery. None of the patients had received blood trans-fusions before the blood gas studies were performed, withthe exception of one patient with aplastic anemia oflong standing (Table II, Case 12).

The collection of blood samples, the equilibration intonometers with the desired gas tensions, the gas analy-ses, and the plasma pH (pHs) determinations were per-formed on venous blood samples removed without stasisby the methods described elsewhere (5). The subjectswere resting for at least one hour before the removal ofthe blood samples. The cell pH (pHc) was determinedfrom the Henderson-Hasselbalch equation given below.In the case of the severely anemic subjects the bloodsamples were first concentrated by removal of plasmaunder oil to bring the packed cell volume approximatelyto normal.

pHc = pK'c + log (BHCOs) blood

where (BHCO.) blood = the total CO, of oxygenatedblood at 40 mm. Hg partial pressure of CO, and (HCO,)blood = H2CO, content of blood calculated from theformula (H2CO,) = a (pCO,) where "a" is the solu-bility coefficient of carbon dioxide in blood (from data

1372

S3II0Q

1&

. .493

4),4

I S4-1

s >'

Ir

I, 4,

0

42

-I .

4.1

4,4IZ ._w

V44 ROa 43A0 *

4ci-wt, A

4,Ei mw0 Mv)

04'04,Eu 4,

coCdaI,oo.

4,.1

co 4,A

400

1. ig 8I: >f IC

co -a SI

a1Ee

I cd-

103mI a '10=

I .

. 0

4,d

0so

0 45:U

'04,4,la 0

¢< o

o N | | %O 00 00 0o 0|.0

S =I lUt Nd 0; 1' - 00 1%Q 00 k. 1-

i~~~~~~~~'C4eI|U|I t ~)0

('4

gleZ;t_e N _ N "~~~0C4

- 6 t04C - -0V4O 0 oV K0

If)C44 ('m(4 m "0 eqC4m

m k '%t'.t-.o

° | 0t-|%'- 1f 1 .1 . 0 00 0

o~ U, 0\ o o o - UI-1

0 00 0

U V-4 V-4 V~~~~E~ ('1)-4 .wC4 0 (1'

4 Iq (E (

c; 0 ( 0I v410 - 00

o ___ ___i~~~~~~~~~1~~Uo tn __ot- "do1011'0

8 I-l."olo l 1~0 (4.)Io 000,-40('0___I -o I__ __I_ ____dl I I I' l u 4V)0 ('4zuw;

~~~~~~~~~~~~~~~~~~~Wq

.We~~~~~~~~~~~~~~~~~~~~~~~~Cd=C-1;16C;e"Xl C

4,> I .')I + u V I t, lo 1

=1v~~~~~~~~~ 001 WWoin oCo

§~~~~~~~~~~~~~~~~~~~~~~~~-1- V-41,% W41-W41

V-4~ 1Os~~~~I1- 1 -'-1-

oI4~4.. .4'1 -- 1

flt I c l 1~~~~NW) colmo~-o° lo YU,)

rArI )I >l Cd >lbIe^ -Ie e1ee m

t T m | ~~~toB8 .81e !t

a

4,0

a)C-

.0

-484)

4. 1

.0

04)

I'0

0 q)4.1

0.1O

4,

Ico o

I sv

U0 V'Rf cd

0.4,40

I

T

4a

0U

gO]Ip0 I

0. ='-'

0u)

<0 003

ES0-8MI t 4)

co~8 30 04

0= .4

0Am

.cma0l 0R - q.Ci .-

0 o' o009

1Mdt>3~e

ES w

0

q

.40~ C:

I w0 isK;~

a 0

C5 Ov

.0 .0

0

*d CoY

0 00'0X4J4o

. w 9 >

0.4 .404.C

W.=0

4.J

odoz.Elco on

c:C : S

caod . .03 >

0d4

>c coe0

I~0j ,.4-J Cm

I =

Io-2I M=CO 4.0d

4-ClSa

W~~~~~~~~~~~orta .606t.wo

C~~~~~~4 _____-~~~~~~~~~~~~~~~~~4 C

i lsm O 0O 00W m~~m~rt.. C4 O 0

1o - -W'

o~~~~~~~~~~c N oc

31X14+1r N 106k- ON

140I n I III 100

C4~~ ~ ~ ~~~4

I W)J I I I I I I

-~~~ 31) lid) K0~~~~Wi W)

I I I I I en

o8-YOO! !)D 1S[ -u

4 _ _ _I _ _ _ I _ _ _ _ _ _ _ _ _ _ _ _1_1_ _

ChC4M C4 C4 eq~~~~~31) 44 4 ~ _ _-i I1 131or l x1 1W) 1i W)

. OI I I v u) 31 1) 1

I eq eq~~~~~~4 C4 4 v14 0000C

-~~~~~~~~~~~~~~~~~C!d-. L.1 +

.I I

goII T CI o I O

0.E. IUC1)444444I".+ I o~o+° -I , mee I I I I I

$t~~qu!04'4!4!-00q -- e--Nn1Ne0 -* q+3' I ~- -~ ---- -I - - --~t- ---t~ -tIt______I I ___1 I___________,,OD __ ___^O t|_ |__oD

I __I___X~ 1003)31 1X~rta 1a~ 1m;

| 6| ! > OD ! 4s-

! 1)() !'eW44" 4

N !

I 0D 0 I. Weq

.14

04.41

I

II

1

OXYGENDISSOCIATION CURVEIN. ANEMIA OF VARIOUS TYPES

4,

0

u

.4

0a

..

co

004i

I .3z

as4)

C

co1'Q0

wc

.2

z

co(4

*EIco

0

I..4

.*

&.

+j

w¢

to

co

.

w

.Y 0

.f-I:

II.4toa

*EOu'E

0

= 0

.bI

0

4)0

la

U)

. .44

.r-i

,la,()M

0 eq~~~~~~~~~~~~0

k--

= :;: !;4 g4 mCf° t 13 ]R ' '

m~~~~g __ __ ___ _____I Ieq|<

-~u II I

0-_ _~ ~ ~ ~~~~~~~~~~~~1

______ I__ _ _ __ __

__ ___ __ __I:_-___ fcVl QOCxfice 1

3).* E)

;e 1x~~~n 1n 1o !" I !In n W) W) 1- T

C4m C14 C4 m CR~4~4~ U) U) if

) W) In °

'I) U) ~ ~ ~ ~ ~ W)(4 W

_ __ _ _ 0i t W)__ _ _ _ f jO 0

__ _ _co__ t _ _ _ _ _ _ _ _

LI ~~~~~~~W) I, coIf

pit __ _ _ _ 10 __ 10 0_ _H____i£-e ie ___ ___ ___ ___

0 o v.,'* C I.m I- m I-i 0l)~ _~1ViaD-l|TI.tIza It]2~~~ 02!A>1A !34,32+<if)Cf)

1 I

1375

.2rra4ZINc,

P.

Ira

.2ee

1-4

W04m -2g 1.

t3

t

4eV

t34

1.QA.04

I

I

A. C. KENNEDYAND D. J. VALTIS

4).4)

4)4AI0U-Sa_

v00

4-.a

0

4-0co

0.9

4)

E..co

*d .2

. 0

A3._U)0 0

or ax

0C3 ,

. n.*- 8

*0 A--4- .

.-4 a

4-004)

0

00

I.,=. mg

.0 0

44)blo C.

4.- .

bOa.O*"

4)

0

- -* e

r.

cd 0,.2.&2

.. U)

--.>..

0 ._w U

0d6t6

o l l¢ Ic |O |o | | II~~~U)Q 0 4%4 c0 0oco co to t

I ~~If I C4 eIC ~I | S

U) V -

U4 IQ~~IO- 1- k-1W 1N 1 1___ __ __IagzR >[|> ..t 2,> ; S> ¢

O'V8~ ~

4. ... 4..

gs 6

L04 m~I~f U).. CU) t.

if4 ) .. U

o~~~~~~~~~~~~~~~~~~~c

U) - '-~~~~~ '-~4) 0 0 0 0

0% I IoW_4 _,- if

i376

.0U

0

._

4-)4)_8

.1IA

.itI

I I W)u_,1 I

OXYGENDISSOCIATION. CURVE IN ANEMIA OF VARIOUS TYPES

TABLE III

Oxyhemoglobin saturation of laked blood solutionsderived from patients with anemia

PerniciousPernicious Hypo- anemiaanemia in chromic after

relapse anemia treatmentNormal (Table I, (Table II, (Table I,

blood Case 8) Case 4) Case 10)

p,02 mm. Hg HbO2Volume %

15 28 29 29 2820 50 50 48 50.530 78 80 78 7940 86 87 85 85

by Dill in the paper of Keys, Hall, and Barron (6) andDill, Graybiel, Hurtado, and Taquini (7)).

pK'c was obtained from the alignment chart and pro-cedure given in the paper of Keys, Hall, and Barron (6)modified in such a manner for the pK'c to be 5.98 fornormal blood (8, 9) giving a cell pH of 7.20 to 7.25.The other values for pK'c derived from this alignmentchart were changed proportionately.

Correction of the position of the curve according to theplasma pH and cell pH was carried out in curves drawnon logarithmic co-ordinates by a slight modification (10)of the procedure outlined by Dill in the paper of Keys,Hall, and Barron (6).

Buffered solutions of laked blood of pH 7.4 were pre-

pared according to the methods described by Brooks (11)and Darling and Roughton (12).

zI0c

8C

R0 636K

WE 4C

U.)

2C.0

0 20 40 bO 80OXYGEN PRESSURE MM Hq

FIG. 1. OXYGEN DISSOCIATION CURVES FROM ANEMICAND NORMALSUBJECTS

Curve a-Addisonian pernicious anemia in relapse, hy-pochromic anemia, secondary anemia. At serum pH7.4

Curve b-Addisonian pernicious anemia in relapse, hy-pochromic anemia, secondary anemia. At cell pH 7.22

Curve c-Normal blood at serum pH 7.4 and cell pH7.22

RESULTS

The results are presented in Tables I, II, andIII and in Figure 1. The position of the oxygendissociation curve is indicated by the partial pres-sure of oxygen at which 50 per cent of the hemo-globin is oxygenated.

In Table I are shown the mean values obtainedfrom six normal individuals, which have beenpresented in detail elsewhere (10), and the indi-vidual data of eleven patients with megaloblasticanemia in relapse and three in therapeutic re-

mission. Of the patients in relapse the position ofthe oxygen dissociation curve at the plasma andcell pH of the subject (column 0) was displacedto the right of normal in all but one (case 3).After correction to the standard plasma pH of 7.4(Column P and curve a of Figure 1) this shift tothe right was more pronounced except in cases

5 and 6. In these two cases the curve was nearlynormal at standard plasma pH. Serial observa-tions (cases 4, 8-11) show that the displacementto the right of the oxygen dissociation curve didnot alter during the earlier response to treatment,although by the time the red cell count had reachedabout the four million per cubic mm. level, betweenfour and six weeks, some lessening of the dis-placement was apparent. In the two patients inwhom the initial oxygen dissociation curves were

normal at standard plasma pH (cases 5 and 6) a

shift to the right developed during the early stagesof treatment. The presence of reticulocytes (col-umn H) did not appear to influence the positionof the oxygen dissociation curve. The curve was

normal in position in the three patients in thera-peutic remission (cases 12-14). The cell pH (col-umn N) was determined in seven of the cases, ofwhom three were in relapse, two were duringtreatment, and two were in therapeutic remis-sion. Of the cases in relapse, the difference be-tween the cell and plasma pH was increased inthose showing a shift of the oxygen dissociationcurve to the right at standard plasma pH (cases1 and 2), and normal in the case showing a nor-

mal oxygen dissociation curve at standard plasmapH (case 3). In the patients (cases 4 and 5) inwhom the cell pH was determined when the redcell count had reached between three and fourmillions per cubic mm., at which time the oxygen

dissociation curve was still slightly displaced to

/a

/b/

1,~~~~

I~~~~~~~~~~~~~~~~~~~~~~~~~

1377

A. C. KENNEDYAND D. J. VALTIS

the right at standard plasma pH, the difference be-tween the cell and plasma pH was normal. Aftercorrection to standard cell pH the position of theoxygen dissociation curve was unaltered (columnQ). The cell pH was normal in the two patientsin therapeutic remission (cases 12 and 13) bothof whomshowed a normal position of the oxygendissociation curve.

In Table II are shown the results obtained inpatients with nutritional hypochromic anemia,chronic post-hemorrhagic anemia, secondary ane-mia, aplastic anemia, and hemolytic anemia. Withthe exception of the case of anemia associated withchronic nephritis (case 8), in which there was anacidosis, the plasma pH of the subjects was nor-mal; in the majority of cases the values were in thelower range of normal. The position of the oxy-gen dissociation curve at the plasma and cell pHof the subject was displaced to the right of normalin all the patients' with the exception of one patientwith post-hemorrhagic anemia (case 5), one pa-tient with aplastic anemia (case 11), and all threepatients with hemolytic anemia (cases 13-15).The displacement to the right was still present af-ter correction to the standard plasma pH of 7.4(Figure 1, curve a) excepting the patient withchronic nephritis whose curve was to the rightof normal at blood plasma pH but normal in posi-tion at standard plasma pH. The cell pH was lowin six of the seven cases in which it was determinedand normal in the remaining case (case 14), apatient with hemolytic anemia. After correctionto the standard cell pH the position of the oxygendissociation curve was normal in the two patientswith hemolytic anemia (cases 13, 14) and in onepatient with aplastic anemia (case 11) but wasstill slightly to the right (Figure 1, curve b) in thepatients with hypochromic anemia (cases 1, 4),secondary anemia (case 9), and the remaining pa-tient with aplastic anemia (case 12).

Table III shows that the oxygen dissociationcurves of hemoglobin solutions (buffered solutionsof laked blood) from patients with pernicious ane-mia and hypochromic anemia are the same as thatof hemoglobin solutions of normal blood.

DISCUSSION

Odaira (1) states that in anemia there is amarked acidosis and that as a result there is ashift to the right of the oxygen dissociation curve.

The techniques employed by Odaira are open tocriticism, however, and many authors (3, 4, 13-16) have failed to demonstrate an acidosis in per-nicious anemia but have, in fact, shown a tend-ency to an alkalosis. Stadie and Martin (17), ob-tained a normal oxygen dissociation curve in asingle patient with pernicious anemia but they donot give the plasma pH at which the curve wasperformed. Richards and Strauss (2) reportedthat the oxygen dissociation curves in six patientswith pernicious anemia and five patients withsecondary anemia were in the normal position atthe normal plasma pH of the blood (7.4), and thatonly at the abnormal plasma pH of 7.64 was therea shift to the right. Several of the cases studiedby Richards and Strauss had received transfusionsof whole blood within a few days prior to the de-termination of the oxygen dissociation curves.In discussing the cause of the shift observed atplasma pH 7.64, they suggest that it could be due,in the patients with pernicious anemia, to an in-creased concentration of hemoglobin within the redcell, which would have the result, at this highplasma pH, of increasing the value (- log r) asdeveloped by Van Slyke, Wu, and McLean (18,19) and so increasing the difference between theplasma and cell pH. The hypothesis is admittedto be invalid in hypochromic anemia in which thereis a shift of the oxygen dissociation curve to theright. Dill and his co-workers (3) report that inone patient with pernicious anemia the oxygendissociation curve performed before treatment wasnormal but the subsequent curves during treatmentwith liver orally, and single curves in four othercases of pernicious anemia during treatment, weredisplaced to the right of normal. No curves weredetermined after recovery from anemia. Dill andhis co-workers by indirect methods found the valuerH to be smaller than normal and so concludethat the difference between plasma and cell pH isincreased, the cell pH being relatively more acidthan normal. Both Dill and his co-workers andHenderson (20) were unable to relate the ob-served shift in the oxygen dissociation curve toaltered osmotic or electrolytic relationships be-tween cell and plasma. Isac, Matthes, and Yama-naka (4) found that the oxygen dissociation curvesin four patients with pernicious anemia, four pa-tients with secondary anemia, one patient withaplastic anemia, and one patient with hemolytic

1378 i." i"'.. 1. '.I

OXYGENDISSOCIATION CURVEIN ANEMIA OF VARIOUS TYPES

anemia were displaced to the right of normal.These authors, using indirect methods like Dilland his co-workers, were unable to demonstrateany decrease in the cell pH.

The present observations confirm that there isa shift to the right of the oxygen dissociation curveat standard plasma pH in the majority of patientswith hypochromic anemia,, secondary anemia, andmegaloblastic anemia in relapse, thus providingfor the anemic patient an increased yield of oxy-gen to the tissues. The time taken for the shiftin the curve to lessen following treatment was inthe patients with megaloblastic anemia some fourto six weeks but it should be noted that many ofthese patients were the subjects of an investigationof the value of oral vitamin B12 in the therapy ofpernicious anemia and were not always showingoptimal hematological responses. It is not ap-parent why two of the patients with megaloblasticanemia had a normal oxygen dissociation curveinitially but developed a shift to the right duringthe course of treatment. A similar developmentwas observed by Dill and his co-workers. Thepresent investigation does not confirm that thereis a shift to the right of the oxygen dissociationcurve in hemolytic anemia (4).

The beneficial effect of the displacement of theoxygen dissociation curve is obvious but themechanism responsible for it is not apparent. Areduction of the plasma pH of the subject woulddisplace the curve to the right but there is noevidence that acidosis is present and moreover thedisplacement of the curve is still present after cor-rection to standard plasma pH. The presence ofcarboxyhemoglobin causes a shift to the left ofthe oxygen dissociation curve (17) and the posi-tion of the oxygen dissociation curve in the normalsubject may be influenced to a small extent by thecarboxyhemoglobin present (21). Wehave foundsmaller amounts of carboxyhemoglobin in theanemic subjects than in normal subjects but thequantities involved are so small that they couldnot produce the observed displacement of the oxy-gen dissociation curve. Alteration in the concen-tration of hemoglobin within the red cell also can-not be the factor responsible for the displacementof the oxygen dissociation curve since the dis-placement is demonstrable in both pernicious ane-mia and hemoglobin deficiency anemia. The effectof the increased plasma volume in anemia can

likewise be excluded as the cause of the displace-ment since artificial dilution with fresh plasmafrom the same individual does not influence theposition of the oxygen dissociation curve (2, 5).

It is now accepted (22, 23) that many differentkinds of hemoglobin, at least some of which mayinfluence the position of the oxygen dissociationcurve, may be present in pathological states. Itis unlikely, however, that an abnormality of he-moglobin can be responsible for the observed dis-placement of the oxygen dissociation curve inanemia since the oxygen dissociation curves ofhemoglobin solutions of anemic blood are the sameas those of normal blood.

The position of the oxygen dissociation curveis dependent basically upon the cell pH and it isunfortunate that there is no entirely satisfactorymethod by which the cell pH can be determined di-rectly. Hampson and Maizels (24) using a directmethod at room temperature (glass electrode de-terminations on red cells hemolysed by repeatedexposure to low temperature) found that the dif-ference between the cell pH and plasma pH inpernicious anemia was greater than normal whilein hypochromic anemia it was less than normal.Maizels (25), in referring to cell pH determina-tions, expresses doubts about his results partlybecause of the many manipulations involved andpartly because the determinations were made atroom temperature in conditions far removed fromthose in which red cells exist in the body. Theindirect method for determining cell pH, in whichthe value "r" as developed by Van Slyke, Wu, andMcLean (18) (- log r = pH. - PHC) is deter-mined, or the nomogram developed by Dill (see 6)is used, is superior to the direct method particu-larly as regards comparative studies, although theHenderson-Hasselbalch equation is not a strictphysical chemical equation. If the indirect method,however, is applied to anemic blood where the cellphase is small, there is the disadvantage that ex-perimental errors are multiplied. To avoid thisdisadvantage the cell pH was determined afterconcentrating the anemic blood samples by re-moval of plasma under oil to bring the packedcell volume into the normal range. This manipu-lation permits more accurate study.

Using this indirect method the cell pH was lowrelative to the plasma pH in cases of perniciousanemia in relapse and cases of hypochromic anemia

1379

A. C. KENNEDYAND D. J. VALTIS

which showed a displacement of the oxygen dis-sociation curve to the right. On the other hand,in cases of pernicious anemia in therapeutic re-

mission having the oxygen dissociation curve nor-

mal in position, the cell pH was normal. The cellpH was also normal in the case of pernicious ane-

mia in relapse which showed a normal oxygen dis-sociation curve. In the cases in which the cellpH was low correction of the oxygen dissociationcurve to the standard cell pH largely, but not com-

pletely, corrected the displacement to the right,suggesting that while a low cell pH is the mainfactor in producing the displacement there is alsoa further factor. In both cases of aplastic anemiathe cell pH was also found to be low. Correctionof the position of the oxygen dissociation curve inthese cases to the standard cell pH wholly elimi-nated the displacement of the curve in the case ofrecent onset and partly corrected it in the case

of longstanding. This patient had received numer-

ous blood transfusions over several years and itwould be unwise to draw any conclusions fromexamining this individual's oxygen dissociationcurve since it would appear that one was in factexamining the oxygen dissociation curve of many

donors. The cell pH was low in one of the twocases of hemolytic anemia in which the cell pHwas determined and normal in the other. In theformer case, correction of the position of the curve

to the standard cell pH wholly corrected the slightdisplacement of the oxygen dissociation curve thatwas present at the cell pH of the subject. Thecases of pernicious anemia studied during re-

covery showed that the oxygen dissociation curve

is still slightly abnormal when the red cell countis about the 4 million per cubic mm. level, at whichtime there is still a degree of macrocytosis. Inthe cases in complete therapeutic remission, wherethere is no macrocytosis, there is no displacementof the oxygen dissociation curve.

While a lowered cell pH thus accounts for themajor part of the displacement of the oxygen dis-sociation curve the results suggest that there is inaddition a further factor. Measurements of thered cell diameter thickness ratio were not madein this study but the observed relationship betweendisplacement of the oxygen dissociation curve andshape of the red cell in stored red blood cells (5,10) indicates that such measurements might use-

fully be done in further studies in gas transport inanemia.

SUMMARY

1. The oxygen dissociation curves in twenty-nine cases of anemia have been studied.

2. In ten of the eleven cases of megaloblasticanemia in relapse and in nine of the ten cases ofhypochromic anemia and secondary anemia therewas a slight displacement of the oxygen dissocia-tion curve to the right. No shift to the right waspresent in three cases of hemolytic anemia or inone of two cases of aplastic anemia.

3. The displacement of the oxygen dissociationcurve persisted during the period of recovery inthe cases of megaloblastic anemia but was notpresent in cases in therapeutic remission. Theposition of the curve was not influenced by thepresence of reticulocytes.

4. The major part of the displacement appearsto be due to an alteration in the cell pH.

ACKNOWLEDGMENT

The authors wish to express their gratitude to Profes-sor L. J. Davis for his advice throughout this study.

REFERENCES

1. Odaira, R., The change in the reserve alkali and oxy-gen dissociation curve of blood in clinical and ex-perimental anemias. Tohoku J. Exper. Med., 1923,4, 243.

2. Richards, D. W., Jr., and Strauss, M. L., Oxy-hemo-globin dissociation curves of whole blood in ane-mia. J. Clin. Invest., 1927, 4, 105.

3. Dill, D. B., Bock, A. V., Van Caulaert, C., Fblling,A, Hurxthal, L. M., and Henderson, L. J., Bloodas a physicochemical system. VII. The composi-tion and respiratory exchanges of human blood dur-ing recovery from pernicious anemia J. Biol.Chem., 1928, 78, 191.

4. Isac, C., Matthes, K., and Yamanalk, T., Unter-suchungen uiber den Transport des Sauerstoffs immenschlichen Blut; der Sauerstofftransport imBlute bei verschiedenen Krankheiten. Arch. f.exper. Path. u. Pharmakol., 1938, 189, 615.

5. Valtis, D. J., and Kennedy, A. C., Defective gas-transport function of stored red blood-cells. Lancet,1954, i, 119.

6. Keys, A., Hall, F. G., and Barron, E. S. G., The posi-tion of the oxygen dissociation curve of humanblood at high altitude. Am. J. Physiol., 1936, 115,292.

1380

OXYGENDISSOCIATION CURVE IN ANEMIA OF VARIOUS TYPES

7. Dill, D. B., Graybiel, A., Hurtado, A., and Taquini,A. C., Der Gasaustausch in den Lungen im Alter.Ztschr. f. Altesfoschung, 1940, 2, 20.

8. Dill, D. B., Daly, C., and Forbes, W. H., The pK' ofserum and red cells. J. Biol. Chem., 1937, 117, 569.

9. Dill, D. B., Edwards, H. T., and Consolazio, W. V.,Blood as a physicochemical system. XI. Man atrest. J. Biol. Chem., 1937, 118, 635.

10. Valtis, D. J., and Kennedy, A. C., The causes andprevention of defective function of stored red bloodcells after transfusion. Glasgow M. J., 1953, 34,521.

11. Brooks, J., The oxidation of haemoglobin to methae-moglobin by oxygen. II. The relation between therate of oxidation and the partial pressure of oxygen.

Proc. Roy. Soc., 1935, 118, 560.12. Darling, R. C., and Roughton, F. J. W., The effect

of methemoglobin on the equilibrium between oxy-gen and hemoglobin. Am. J. Physiol., 1942, 137,56.

13. Barr, D. P., and Peters, J. P., Jr., III., The carbondioxide absorption curve and carbon dioxide ten-sion of the blood in severe anemia. J. Biol. Chem.,1921, 45, 571.

14. Dautrebande, L., L'alcalose paradoxale de l'an~miepernicieuse. Compt. rend. Soc. de biol., 1925, 93,1031.

15. Connery, J. E., and Jolliffe, N., Studies on the "aciddeficit" in pernicious anemia, with report of a case

showing return of free acid. Am. J. M. Sc., 1931,181, 830.

16. Emerson, C. P., Jr., and Helmer, 0. M., Reaction(pH) and carbon dioxide content of the venous

plasma in pernicious anemia. Arch. Int. Med.,1935, 55, 254.

17. Stadie, W. C., and Martin, K. A., The elimination ofcarbon monoxide from the blood. J. Clin. Invest.,1925, 2, 77.

18. Van Slyke, D. D., Wu, H., and McLean, F. C.,Studies of gas and electrolyte equilibria in the blood.V. Factors controlling the electrolyte and waterdistribution in the blood. J. Biol. Chem., 1923, 56,765.

19. Van Slyke, D. D., Factors Affecting the Distributionof Electrolytes, Water and Gases in the AnimalBody; Lectures, Delivered at Rutgers Universityunder the Luther Laflin Kellogg Foundation.Philadelphia, J. B. Lippincott, 1926, 62 p.

20. Henderson, L. J., Blood; A Study in General Physi-ology. New Haven, Yale University Press, 1928,397 p.

21. Riley, R. L., Lilienthal, J. L., Jr., Proemmel, D. D.,and Franke, R. E., The relationships of oxygen,carbon dioxide, and hemoglobin in the blood ofman: Oxyhemoglobin dissociation under variousphysiological conditions. J. Clin. Invest., 1946,25, 139.

22. Darling, R. C., Smith, C. A., Asmussen, E., and Co-hen, F. M., Some properties of human fetal andmaternal blood. J. Clin. Invest., 1941, 20, 739.

23. Eastman, N. J., Geiling, E. M. K., and De Lawder,A. M., Fetal blood studies. IV. The oxygen andcarbon-dioxide dissociation curves of fetal blood.Bull. Johns Hopkins Hosp., 1933, 53, 246.

24. Hampson, A. C., and Maizels, M., The difference ofpH between plasma and red cells. J. Physiol.,1927-28, 64, XX.

25. Maizels, M., Phosphate, base and hemolysis in storedblood. Quart. J. Exper. Physiol., 1943, 32, 143.

1381