co,dm5migu4zj3pb.cloudfront.net/manuscripts/104000/104353/jci6110… · data on the early...

EFFECTS OF CHRONICHYPERCAPNIAON ELECTROLYTEANDACID-BASE EQUILIBRIUM. I. ADAPTATION*

By ADOLFPOLAK, GORDOND. HAYNIE,t RICHARD M. HAYSt ANDWILLIAM B. SCHWARTZ§

(From the Department of Medicine, Tufts University School of Medicine, and the Pratt Clinic-New Englanid Center Hospital, Boston, Mass.)

(Submitted for publication February 15, 1961; accepted March 13, 1961)

Although chronic respiratory acidosis is one ofthe commonest clinical disturbances of acid-basebalance, the process of adaptation which it induceshas not been extensively investigated. Normalman tolerates prolonged exposure to high CO.,tensions poorly and is, therefore, not a suitablesubject for study. Recent observations in therat exposed to a high CO, atmosphere (1, 2)have provided much valuable information, but nobalance data are available in the dog, the animalwhich appears more closely to resemble man inits adjustment to other disturbances of acid-baseequilibrium. For this reason, and in order toprovide a basis for further investigations ofchronic hypercapnia in the dog, the present stud-ies of adaptation were undertaken.

METHODS

Eleven balance studies were carried out on 11 healthyfemale mongrel dogs weighing between 13 and 20 kg.Observations were made during a 4- to 7-day controlperiod and a 6- to 15-day period of exposure to 10 to13 per cent carbon dioxide (CO2 period).

The studies were carried out with the dogs in meta-bolic cages maintained at a temperature of approxi-mately 150 C inside a small, ventilated room. The cageswere covered with a plastic canopy which was left openduring the control period. During the CO2 period thecanopy was closed, and a gas mixture of C02, 02, andcompressed air was delivered into it. The ratio of thegases delivered was regulated manually. The CO2 con-tent of the cages was monitored frequently with aKwik-Check carbon dioxide analyzer (Burrell Corp.),

* Supported in part by grants from the National HeartInstitute (H-759 and HTS-5309), the American HeartAssociation and the Life Insurance Medical ResearchFund.

t Medical Foundation Research Fellow, The MedicalFoundation of Metropolitan Boston, Inc.

t Public Health Research Fellow of the NationalHeart Institute.

§ Established Investigator, American Heart Associa-tion.

which was standardized with gas mixtures whose com-position had been determined by the Scholander method.The oxygen content was determined by a Beckman modelC oxygen analyzer.

On the first day of the CO2 period, the CO2 contentof the cage atmosphere was increased to approximately10 per cent over a period of 3 to 6 hours. Over thenext 1 to 2 days the level was raised slowly to the de-sired range of 11 to 13 per cent. The oxygen concen-tration was maintained at 20 + 3 per cent throughout.

The dogs were removed from the cages each morningfor weighing, feeding, collection of urine and feces, cagecleaning, and (on most mornings) arterial blood sampling.During the CO, period they were also removed in theevening so that the daily diet could be fed in two portions.The total time out of the high CO2 atmosphere on eachof these days was about 30 minutes. In order to obtaindata on the early adaptation to uninterrupted respiratoryacidosis, 9 of the dogs were given their entire diet forthe first day of the CO2 period 2 to 3 hours before theCO2 tension was raised, and they were then left un-disturbed for 24 hours.

A synthetic diet was given, consisting (in per cent) of55 water, 14 casein, 1.5 agar, 8 hydrogenated vegetableoils, 7.5 dextrose, 14 dextrin, and supplementary vita-mins. To each 100 g of diet, 9 mEq of potassium wasadded as neutral phosphate (4 parts K2HPO4, 1 KH2PO4).The diet was homogenized with 1.5 times its weight ofwater shortly before feeding. The homogenate was fedin a pan during the control and recovery periods, butduring the CO2 period it had to be tube-fed because ofanorexia. Rejected food was weighed and any vomituswas either re-fed or weighed and analyzed. The sodium,potassium, chloride, and nitrogen content of the dietwas checked frequently by analysis.

The dogs received 30 g of diet (75 g of homogenate)per kg of body weight per day. In order to study the ef-fects of restriction of sodium and of chloride, the ani-mals were divided into 3 groups whose diet was identicalexcept in the following respects: 1) 5 dogs received 50to 70 mEqof sodium and chloride daily (high-salt group);2) 5 dogs received approximately 5 mEq of sodium and3 mEq of chloride daily, the intrinsic sodium and chlo-ride content of the synthetic diet (low-salt group) ; 3) 1dog was studied on a diet identical with that of the low-salt group except that sufficient potassium chloride wasadded in lieu of potassium phosphate to raise the dailychloride intake to 13 mEq.

1223

CN 4O ~0 0 -1 0 1-1 C' C"001l)4 4 4 4 44 4 4 4 el 4 4

in Vc It ")0 0 00 00 0% V%.00 0%0%0%%0%,ONO 00 00 00 00 00

n m (o t-4 C4 men I'd st t In m en4 - - -4 l- - - 1. - -

t o ON 00 0ll ON C-

0 - O - 00 - I-- -0%0a, 00 a, a,0 .- - -

-4. c 0 00 t-.I)

0 0o in 1)C ' - 00 Vo uI 0 00 m o NC104 00UL -) m £0--el; (S - .\0 tz os o0 rZ X '-0 0O:_ O i __-N C4 (4 m ") m m m en en ene) en N m en 14N) N C4

0-N

> \0 O t '.) ) 0- 4W N en 00 00 t-r 04 00 \0 00000 00 t- L)(In ") ") - - - - en N4 (1 0LN 0N 00" "0 C) Cm mC, tC'.-C f- 0tz--C' C'.C_ 1,- V'. tCtC-.0-- t-- C-z C'.C'.C'. C' C-zt-*

040-40- 0 00 000-.O %Xnr' m 00

C)e t o00%00 O 00 0

00 N a,t§{ wF+tP mP t- q a, m t-w. m (,4 W.

NO 4- 14O i*oe ho0) C0 00 m w m m- m- -

0 000 e t X

0) -404000-0- 0-C t0-4-4 -4-

? " n Lo c°°° w - U- °n °° -

"0) se s S-N-NN

S ")0x 0N V000-C)C'--'1)C')0%000%0"20- t

'N )W mo XocEbXs s :U xi

: * qt tU)%~0~0 %b t~n t , e

ZN

Xoo \0 m\0 0 -t to w .t o m m 0 NMoX

?N

-o 00 0)C'.C'og t-g)g t-8 8 8 8o ON0 000000 1 VC C

- 4

0)

0 t,

cil 0%0% 0 0 00 0% 0 %~0 %C 0 0 0

I o%0 e -000o C'- a, OI m 11 L t U) -t m m 10

CN C NEN N N N("1 40. 14-4 --44 - 40-4 -

00000000000~~'t0000000t0')0C')" 00~0004"o-.C-00 0- 0> rC)N00-40

\0 f-4 0%0 d") in In " O'0 ON 00

N N - -4 N0-.0 " c- N.s -4 '4 t4

*" !t O, M m r. in toC)esz \0 Cz \0 010-0N m N N N N N NeC N

(N Q 0(Ds in C, X o - X - t 0-__-_ m,N eq C1 eq,.:/ _ N _4 t- _ N

, 0, ,,, .o,,C'_.00No t _ ,.

, )0o 0

_N t- U)

00

tS- X t

-- O

000.000§ 00 010011000%'00000

_c oo X In t- t- -4 o

O oOo- O

OoO ooo

ON0s l ON0b as as ON

. .

'o C'

XOx-cEoO sovo \:> v Xo O Xo\° x sO O XO '4O \: \o \o \o \o sQ O (4N C4 (N eq (1s f1 (11 (1 (1

XtotNees o toto s N o tot~N s NN Ns Ns C1NNoc eq e-s c" (,4 No No oo soso f

$uWdttt m m mm m u W W Wt u u m e - m W in IN tN C4 N tN t("t N N0 -40 -40 -4e0 0064 r r_0\%\

-U)t- X0 0 0 _ t, \0 r- °° 0h ° _ ~ In \0 000000 - In- - - - -4 - - C-4 eS C4t. qN. eq(1St mm.msm~t mstt

"V

U)du

"0

0Uv

co

I.

CZ.Z

co"0

0

0

40

10

W

0

10)c0 C

U"0O

0)

0.0_

00

0 -c

T4>

tvO

°*,)000_

CO

z'CO

0

21

C)etz

S

U

zCO

0

u

CO

z

>2

0 cd QWX Cd

0

a4

6.

z

x

Q4Cd

x

"01)

-0)t,-.

"4

t.)

¢ *;

~o )0%)0.

0)00

e4)

I

I

I

I

I

I

tjetb- %00 - 0 00 t- 0

CO C'- C- in W 00 00 00 00 C- -

n s00 vc 00 ~00uX =

_ t1 CS <, < ne~~~~C cl1 C1N N C

0= 00 0o't

O N __

ot CO t t-in00 00 00 C e e-0 01

CZS z 0 0C CC C'o=tt tC C10 0000 0 O- CCCin tCr

t;W~~~~~~~~~tOO b O N °O O -b sb+rt-i 0t t0O=Oo C'sN(4 N-N--- -

__

St~~~~~~~~~~~~~'t 00 \ 00 C" a;ee , _0 --d M m C,s< < b m O X _ X _ e m £

m C4 m m

mOt4~~~~~~~~~~~~~~~~~~~ 00 m kn \ \0 ON tn \0CSmS-m mCSCiO - 0 r cO - -mt

N~~~~~~~~~~~~~~C,N N C, m m m mN INZLtuouo 0_

0e On) 0m \ 00 00C- IS 00,0 00to nN \rt- a,0-0 0 _o 0

ta4~ ~ ~ ~~~~ ~ ~~~"d N '4 m C, in n tn -, t C' m en on -d 'te

0100 _ M e C \ (oN o - 0 t \ t 't

+~~~~~~nU t n~ I m

0 '--e _00 10= Id, 00 oC C 0 M_

Q

=o O > m o b n n n~~~~~(NI00 n t- C1 C4 N X n W In

o^ £\ o-o" i n i U) U) U_1 000o0O ocoooo C)o o 0 o o o o o U

o~ ~~~~~~~~\ \0 \0\oNo\- t- (1 0,\S ON C\_\_\)Co5owO O o oo 0o o 00 o o00 00o 0 GoO O

CCCCCoC00Co oo o oo oCCe0CCCCC -o CC CCCCOCC

% 00 0. C0U,0 -00XX X CX CX 0

t-ooXoo ZX ~oo ZX o.o6o6 t. t:o6X-oo:XZoo:oo-- SboO~ -000t-0. 000' 0CO 00 0-=~~~~~~~~~~~~~~~~~l Ln n tn n in o souQn tn C Ln Ln Ln tf Lntn;m

CoZ

Y ofi os b~~~~~ON Cso N ON C, as _ 1oo 00_ Cs CK C,z f to Ln n to Ul Ln Q~o Lf tn c in U' n tn Ln L

rQ

'CC0~~~~~~~~~~~~~~~~~~~~~

01 o. o. o. o. o. o.0C'-~o. o. o,u0u=u o, o. o. o, o. o.oCCCCCC3CC0C'CC CCCCCCC*++* + * e e

C

0.~~~~~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~~~~~+

0 ~'

Il

POLAK, HAYNIE, HAYS AND SCHWARTZ

The data on acid-base balance obtained in the thirdstudy were fully comparable with those obtained in thestudies on dogs of the low-salt group, and will thereforebe included with them in the Results section. Twenty-twoadditional dogs were studied during 1 day of exposureto CO. in order to obtain further data on bicarbonate ex-cretion. Some animals were fed the synthetic diet whileothers were given a meat diet.

The urine drained into bottles containing mineral oiland thymol chloroform over a siliconized metal surface.The system was tested frequently with water and gaveover 97 per cent recovery. Collections were ended at9 a.m. each day, and volume, pH, and CO2 content weredetermined promptly.

Blood samples of approximately 25 ml were drawnfrom the femoral artery with heparinized syringes.During the CO. period blood samples were taken while

the dogs breathed 12 per cent carbon dioxide in 20 percent oxygen and 68 per cent nitrogen through a non-rebreathing valve mask in order to achieve conditionssimilar to those in the cage. It is apparent, however,that these values for blood pH and pCO2 can be takenonly as approximations.

Feces were collected and pooled for each control, CO2and recovery period. The collections were begun andended 24 hours later than the corresponding experimentalperiod.

The analytic procedures have been described in aprevious paper (3). The electrolyte content of diet,stool, and vomitus was determined on nitric acid ex-tracts. Titratable acid was taken as the sum of thecalculated contributions of phosphate and organic acids(4).

The daily balance was calculated as the net intake

DAYS

FIG. 1. CHANGES IN ACID-BASE EQUILIBRIUM AND SERUM CHLORIDE CONCENTRATION

DURING EXPOSURETO A HIGH CO2 ATMOSPHEREIN A DOGON A LOWSODIUM CHLORIDE IN-

TAKE. The dashed horizontal lines represent the mean daily excretion during the controlperiod. The values for net acid excretion (ammonium + titratable acid - bicarbonate)represent changes from the mean daily excretion during the control period. The valuesfor plasma pH during exposure to CO2 are shown with a dashed line since they are onlyapproximate (see text). Urine pH values are not shown for the 2 days during whichdivided collections were made. The portion of the figure depicting the changes afterwithdrawal from CO2 is discussed in the following paper (5).

1226

ADAPTATION TO CHRONICHYPERCAPNIA1

I,,i[- ii

9 II 12l ! 3 14 D '61A 18

D AYS

FIG. 2. CHANGESIN THE EXCRETION OF URINARY ELECTROLYTESAND NITROGEN DURING EX-POSURETO A HIGH CO. ATMOSPHEREIN A DOGON A LOWSODIUM CHLORIDE INTAKE. The dashedhorizontal lines represent the mean daily excretion during the control period. The portion ofthe figure depicting the changes after withdrawal from CO2 is discussed in the following paper(5).

minus the combined output in the urine and stools. Thechange in net external balance (or "A balance"), towhich frequent reference will be made in presenting anddiscussing the results, was calculated as the differencebetween the balance on an experimental day and themean daily balance during the control period. However,in the case of the dogs on a low-salt intake, which wereoften not in a state of balance until the last control day,the balances of sodium and chloride on that day wereused instead of the mean control values in the calcu-lation of the A balance. In calculating a correction forpotassium excretion due to changes in nitrogen balance,a value of 2.7 mEq potassium per g of nitrogen wasused.

RESULTS

A. Representative studies on low- and high-saltintake

The results of a typical study on a dog (DogH) receiving a low intake of sodium and chlorideare shown in Table I, and the chief parameters

relevant to acid-base and electrolyte balance aredepicted in Figures 1 and 2. Observations madeduring a subsequent recovery period are includedin Table I and Figures 1 and 2 but will be con-sidered in the succeeding paper (5).

Figure 1 shows that the plasma bicarbonate,which at the end of the control period was 22.6mEq per L, had risen to 30.5 mEq per L after 1day of exposure to CO, (Day 6). It then climbedgradually to 37.6 by the end of Day 8, and fluctu-ated between 37.2 and 39.5 mEq per L duringthe rest of the 15-day period. Urinary excretionof ammonium on the first day of the CO., perioddid not change from the mean control level, whiletitratable acid fell, and there was a slight loss ofbicarbonate. Thus, the rise of nearly 8 irEq perL in plasma bicarbonate on the first day of theCO2 period was actually accompanied by a (le-crease in net acid excretion. By contrast, the

122 7


rise of 7 mEqin plasma bicarbonate during the next7 days of the period (7 through 13) was accom-panied by a cumulative increment in net acid ex-cretion of 130 mEq. During the remainder of theCO, period (Days 14 through 20), while theplasma bicarbonate fluctuated between 37.2 and39.5 mEq per L, net acid excretion gradually fellto the control level. It will be seen that ammo-nium excretion rose markedly on the second day ofthe CO2 period and remained significantly ele-vated thereafter, but part (and late in the period,all) of this increment was offset by a moderatesuppression of titratable acid and a slight urinarybicarbonate loss which persisted throughout the

DAYS

FIG. 3. CHANGESIN ACID-BASE EQUILIBRIUM AND SE-

RUM CHLORIDE CONCENTRATIONDURING EXPOSURE TO A

HIGH CO2 ATMOSPHEREIN A DOG ON A HIGH SODIUM

CHLORIDE INTAKE. The dashed horizontal lines repre-

sent the mean daily excretion during the control period.The values for net acid excretion (ammonium + titrata-ble acid-bicarbonate) represent changes from the mean

daily excretion during the control period. The valuesfor plasma pH during exposure to CO2 are shown witha dashed line since they are only approximate (see text).Urine pH is not shown for Day 13, during which dividedurine collections were made. The portion of the figuredepicting the changes after withdrawal from CO2 is dis-cussed in the following paper (5).

[URINE]3 DogI (U)

ChloridemEq/Doy

SodiummEq/Day

60

Potassium 40 _

mEq/Day

Phosphate 40mM/Day 20

KlUirnnan lor;

Gms/Day 5J__IEOrganic Acids 1 0mEq/Doy 9...h1E m~..

Sulphate mEq/Dayl 1'0thmhmEEEEi.UL3.EUDAYS

FIG. 4. CHANGESIN EXCRETION OF URINARY ELECTRO-LYTES AND NITROGEN DURING EXPOSURETO A HIGH CO2ATMOSPHEREIN A DOG ON A HIGH SODIUM CHLORIDE IN-TAKE. The dashed horizontal lines represent the meandaily excretion during the control period. The portionof the figure depicting the changes after withdrawalfrom CO2 is discussed in the following paper (5).

period. The urinary pH (range 6.17 to 6.50) wasslightly above control values throughout theperiod.

Table I and Figure 2 show a sharp rise in theexcretion of chloride, sodium, potassium, and phos-phate on the first day of the CO2 period (Day 6).Plasma chloride fell by 8 mEqper L (Table I).

The urinary chloride output remained signifi-cantly elevated from the second day of the CO.,period through the sixth (Days 7 through 13).averaging 10 mEq above the final control level.During these 7 days the plasma chloride fell by afurther 8 mEq, while the plasma bicarbonate rose7.1 mEq. The urinary sodium excretion returnedto a low level more quickly than did the chlorideexcretion, while the plasma sodium level re-mained steady.

Urinary phosphate, like urinary sodium, hadreturned to its control level after 2 to 3 days ofexposure to CO, and remained close to that levelsubsequently. Urinary potassium fell promptlybelow the control level from its peak on the first

1228

ADAPTATION TO CHRONICHYPERCAPNIA

day and remained depressed for Days 7 through10. Cumulative A potassium balance for the wholeCO2 period totaled + 10 mEq (+ 36 mEq aftercorrection for nitrogen balance). The serum po-

tassium showed no significant change.Table I shows that the nitrogen balance, which

averaged - 2.3 g per day in the control pe-

riod, became somewhat more negative in the CO2period, averaging - 3 g per day. It can be seen

that the output of organic acids and sulfate alsorose, but by the last 4 days of the CO2 period, sul-

fate and organic acid as well as nitrogen had re-

turned approximately to the control range. Thehematocrit fell from 55 to 47 (over 300 ml ofblood was drawn).

The results of a study on Dog I, receiving a

high intake of sodium and chloride, are shown inTable II, and the chief parameters relevant to acid-base and electrolyte balance are depicted in Fig-ures 3 and 4. The changes were closely similarto those seen in the low-salt study (Table I andFigures 1 and 2) and will not be described indetail. Comparisons of the high- and low-saltgroups, instead, are considered below.

B. Collective data

This section summarizes the observations madein the 11 balance studies (5 on a high-salt and 6on a low-salt diet).

1. Acid-base balance. At the end of the controlperiod, the range of plasma bicarbonate concen-

3 4 5DAYS

FIG. 5. EFFECT ON PLASMA BICARBONATE CONCENTRA-TION OF EXPOSURETO A HIGH CO2 ATMOSPHERE. Day - 1

refers to the last day of the control period. The curvedline through the points was drawn by inspection. Notethat the level of sodium chloride intake appears to haveno influence on the plasma bicarbonate values.

tration in the 11 dogs was 21.4 to 24.7 mEq perL (mean 22.8). Figures 5 and 6 show that bythe end of the first day of the CO2 period theplasma bicarbonate had risen by an average of9.1 mEqper L (range 8.0 to 9.8) in the high-saltgroup, and 7.1 mEq per L (range 5.0 to 9.3) inthe low-salt group. This constituted half or more

of the rise that occurred during the entire CO2

E III

A Cumulative balances during exposure to a high CO2 atmosphere

Cumula-Na K corr. tive A

CI- K corr. net acidGroup Dog Days Cl Na 1.3 K for N* N excretiont

mEq mEq mEq g mEqHigh NaCl intake

G 7 - 51 - 9 -44 -68 -45 - 9 +103K 7 - 74 - 43 -41 -24 -14 - 4 + 90I 6 -119 -100 -42 -52 - 7 -17 +125J 6 -64 -31 -40 -18 -18 0 + 29S 8 - 17 + 7 -12 -43 +32 -27 +212

Low NaCI intakeH 15 - 74 + 31 -98 +10 +36 -11 +181A 7 - 79 - 19 -64 -38 - 6 -12 + 96B 7 - 88 - 36 -60 -39 +27 -24 +223D 9 - 83 - 27 -62 -78 -50 -11 +149F 7 - 41 + 2 -43 + 1 +25 - 8 +168

* 2.7 mEq potassium/g nitrogen.t This value is not a balance but represents the change in net acid excretion.

-1229


15A

Plasma ioHCO3

mEq/L 5

Net Acid +4oExcretion 20mEq/Doy -20

DAYS-

HIGH NaCI Group

21o DogK O9Sj

12 3 4 5 6 12 3 4 5 6 7 1 2 3 4 5 6 7 8

LOWjNaC GROUPJ

'5-A

Plasma 10HCOS og

mEq/L 5 / Agela 0. j

Net Aid + 0

Excretion omEq/Doy -20

I

DAYS- 1 2 3 4 5 6 7 8

000 6

i 256

ll0-e

I-

12 34 5 6

* a]12 3 4 5 6 7 8 9

Dvg F

1 2 3 4 5 6 7 1 2 3 4 5 6 7

15

10

5

OD0

15

10

0.40+

a0

10

20-

.15

*10

.5

0404,20020-

FIG. 6. RELATIONSHIP BETWEENTHE RISE IN PLASMA BICARBONATE CONCENTRATIONAND CHANGESIN NET ACID EXCRETION DURING EXPOSURETO A HIGH CO2 ATMOSPHERE.Changes in plasma bicarbonate are calculated from the final control value. Changes innet acid excretion (ammonium + titratable acid - bicarbonate) are calculated from themean control value.

period. Figure 6 shows the urinary net acid ex-cretion (ammonia plus titratable acid minus bi-carbonate). In 4 dogs there was no significantrise in the urinary net acid during the first day,and in 2 of these net acid excretion actually fell.In the remaining 7 dogs net acid excretion roseon the first day, but there was no relationship be-tween the magnitude of the rise in plasma bi-carbonate and the direction or magnitude of thechange in net acid excretion. On the second andsubsequent days of the CO2 period, the rate ofrise in plasma bicarbonate decreased progressively,approaching a plateau between 35 and 38 mEqper L in a week. In every dog net acid excretionafter the first day showed a cumulative incre-ment which greatly exceeded the increment seen

on the first day. With the exception of one ani-mal, the cumulative increase in net acid excretionfor the entire CO2 period (Table III) was fargreater than would be needed to produce the ob-served rise in plasma bicarbonate, based on anassumed extracellular volume of 25 per cent ofthe body weight.

Figure 7 shows the changes in the three com-ponents of urinary net acid excretion separatelyand demonstrates that there was no apparent dif-ference in the response of the high- and low-saltgroups. The ammonium excretion rose abovethe mean control level on the first day in 8 of the11 dogs. By the third day, ammonium excretionhad risen in all dogs and remained elevated abovecontrol levels for the rest of the study. Urine

1230

Iv.0

K

a

IC42c2

ADAPTATIONTO CHRONICHYPERCAPNIA

pH, which was in the range of approximately 5.5to 6.0 during the control period, rose slightly toa range of 6.0 to 6.5.

Titratable acid excretion rose on Day 1 in afew dogs, but in most dogs the excretion of titrata-ble acid was unchanged or fell below controlvalues. From the second day of the CO2 periodonward, titratable acid was suppressed in almostevery dog.

Urinary bicarbonate excretion (Figure 7),which during the control period averaged lessthan 1 mEqper day, rose slightly (usually by 1 to5 mEq) when the dogs were exposed to CO2 andremained slightly elevated. In the 22 additionalbrief studies on bicarbonate excretion, a muchwider range of urinary bicarbonate incrementswas observed during the first day in CO2. In 9animals excretion reached a level of 10 to 25 mEq,and in those dogs there was regularly a suppressionof ammonium and titratable acid excretion.

2. Chloride, sodium, potassium, and phosphate.On the first day of exposure to CO2 the plasma

chloride fell by an average of 8 mEq in the high-salt and 7 mEq in the low-salt group. The "mir-ror image" relationship between the curves de-scribing changes in plasma chloride and plasmabicarbonate was maintained throughout the CO2period in all the animals and is well illustratedin Figures 1 and 3.

Figure 8 shows the corresponding changes inthe daily chloride and sodium balances. Thesechanges can be interpreted as virtually equivalentto changes in the urinary excretion, because in-take and fecal excretion remained nearly con-stant. On the first day of the CO2 period therewas a significant increase in urinary chloride inall but 2 of the 10 dogs (these two dogs were ex-ceptional in that their urine showed no significantchange in electrolyte composition until the secondand fourth days, respectively, of the CO2 period,although their plasma electrolyte levels did showcharacteristic changes on the first day). Subse-quently, there was a progressively diminishingchloruresis, chloride excretion approaching the

+60 0+eHgh NoCl Groupo+50 - o Low NaCl Group

mEq/Doy 0 :{ 2 0 0

20

+20 _ 8 § 0

o o-8

Q 0 00 0 0+10 0 0

+30- ~~~~~~~0

020 ~~~~~0

+10 - 0 -0

+ 10 8

r*CO3is ; ^A *+ *0

1T. A. 0 0 1mEq/Doy jt O0 0 0 0

I- *HCO~~I 00*00 0 J

m________ f

0

1 2 3 4 5 6 7 8D AYS

9 10 11 12 13 14 I

FIG. 7. CHANGESIN THE EXCRETION OF AMMONIUM,TITRATABLE ACID, AND BICAR-

BONATEIN BOTH THE HIGH- AND LOW-SALT GROUPSDURING EXPOSURETO A HIGH CO2ATMOSPHERE. The values represent changes from the mean daily excretion duringthe control period.

1231


HIGH NaCl GROUP LOWNaCI GROUP

AChlorideBalancemEq/Day

ASodiumBalancemEq/Day

+2010

0

l0

2030

40

50

2010

10

0

20

30

4050

) Swl * tls'oi* * k0 08

* * 0 8~~~~~~~~~~~~~~~~~~~~~00so0

_~Z _ _ _ _ 8 , , , , , ,

* 0 0~~~~~~~~~~~O

0ot

20

,0

0

10

20

30

40

50

20+

10

0

1020304050

1 2 3 4 5 6 7 8 1 2 3 4 5 6 7 8 9 10DAYS DAYS

FIG. 8. CHANGESIN CHLORIDE AND SODIUM BALANCEDURING EXPOSURETO A HIGHCO2 ATMOSPHERE. The values for the high-salt group represent changes from themean daily balance during the control period. For the low-salt group, changes arecalculated from the values for the last day of the control period, since earlier in thecontrol period the urinary excretion was falling steadily.

control level toward the end of a week in CO2.Table III shows that cumulative A chloride bal-ance ranged from - 41 to - 119 in 9 of the 10dogs and was not significantly different in thetwo dietary groups.

Urinary sodium excretion (Figure 8) was in-creased initially in every dog except two (thesewere the two dogs in which all the changes in uri-nary electrolyte composition were delayed). Thedogs on a low sodium chloride intake came intoapproximate daily sodium balance by the fourthday of exposure to CO2. In the high-salt groupthe pattern of sodium excretion varied widely,but by the sixth day all animals had returned tocontrol levels of excretion. Table III shows thatthe cumulative A sodium balance for the entireCO2 period was usually somewhat negative, butin only 2 dogs (both of the high-salt group), wasthis deficit substantial. In no dog was sodiumbalance significantly positive. Cumulative chlo-ride loss regularly exceeded the cumulative so-dium loss. The excess chloride calculated fromthe ratio of the concentrations of these two ions

in the extracellular fluid is shown in Table III.Expressed in this way, it is apparent that a sub-stantial selective depletion of chloride occurred inall but one animal.

The plasma potassium, which at the end ofthe control period ranged from 3.5 to 4.6 mEqperL, showed no consistent trend in either dietarygroup during the CO2 period. During the con-trol period all the dogs had remained approxi-mately in potassium balance. On the first dayin CO2 all dogs except two increased potassiumexcretion over the control level by amounts rang-ing from 25 to 65 mEq. (The exceptions werethe two dogs in which all the changes in urinaryelectrolyte composition were delayed.) Thesefigures were not appreciably changed by correctionfor nitrogen balance. During the rest of theCO, period, the potassium excretion fell to thecontrol level or below in 4 of the 5 dogs in eachdietary group. In these dogs, the A potassitimbalance, corrected for nitrogen balance, rangedfrom - 18 to + 36 mEq (Table III). In theremaining dog in each group (Dogs D and G) the

.2I

d

1232


excretion remained somewhat elevated and cumu-

lative A potassium balance was significantly nega-

tive, even after correction for nitrogen balance.Urinary phosphate output rose significantly on

the first day of exposure to CO2 in 8 dogs (theexceptions were the two dogs in which all thechanges in urinary electrolyte excretion were de-layed) by increments ranging from 9 to 25mmoles. On subsequent days, in the majority ofdogs, there was a persistent excess of phosphateexcretion over control values. The dogs in whichthis excess was greatest were also those in whichnitrogen balance was most strongly negative. Inthe dogs which remained in nitrogen balance,phosphate excretion returned to the control range

after the first day. Plasma phosphate was esti-mated at intervals during the study in several dogsand showed no consistent change.

3. Nitrogen, sulfate, and organic acids. In most

dogs of both the high- and low-salt groups, ni-trogen excretion rose, with the result that A mean

daily nitrogen balance during the CO2 period

+ 50

+40LGl

mEq + 30

+20

+10

0

- 10

A. ° 0

o 0 -

. .-

0 lo 20 30 40 50 60 7CANOmEq

+50

+40&NomEq +30

+20

+10

0

- 10

averaged approximately - 1.5 g in both dietarygroups (Table III). The excretion of organicacids tended to increase, as illustrated in Fig-ures 2 and 4, but the changes were small and therewas no significant difference between the twodietary groups. The rise in organic acid excre-

tion was greatest in those animals whose nitrogenbalance was most strongly negative. Sulfate ex-

cretion was measured in 3 low-salt and 3 high-salt dogs. In most instances there was an increasein sulfate excretion of 5 to 20 mEqper day. Thesechanges also appeared to correlate with changes innitrogen balance.

4. Interrelationship of changes in certain uri-nary constituents on the first day. Figure 9A re-

lates the increments in sodium and chloride ex-

cretion. It will be seen that the ratio of A chlo-ride excretion to A sodium excretion was usuallygreater than 1: 1.3, which is the approximateratio of the concentrations of these electrolytesin the extracellular fluid (ECF) represented bythe dashed line. It will be noted that in 2 dogs

+50

+40ANH4mEq + 30

+20

+10

0

0

0 10 20 30 40 5A K.mEa

l0 WD 30 40 50 60 70 4

(&Na + AK),mEq80 90s

* HIGH MAt GRUIP0 LOWNoC GROUP

50 60 70

FIG. 9. INTERREL.ATIONSHIP OF CHANGESIN CERTAIN URINARY CONSTITUENTS DURING TIlE

FIRST DAY OF EXPOSURETO A HIGH CO2 ATMOSPHERE. Data are included from 2 low-salt dogswhich were not studied after the first day of exposure.

B.

@0~~~0 0

0 0

00 0)F T I) 0

C. *

I o00

0

0

0 0* 0 0 0

I° o

I

1233


from the high-salt group the changes in electrolyteexcretion on the first day were negligible, as men-tioned earlier.

Figure 9B relates the increment in ammoniumexcretion to the sum of the increments in sodiumand potassium excretion. It will be seen that Aammonium excretion did not appear to be a func-tion of A sodium plus potassium, and thereforewas, in effect, unrelated to the magnitude of thetotal cation diuresis. There was, furthermore,no evidence that either parameter was influencedby the salt intake.

Figure 9C relates the increments in sodium andpotassium excretion. In the low-salt group, Apotassium excretion regularly exceeded A sodiumexcretion, comprising at least 60 per cent of Asodium plus potassium. Indeed, a significantnatriuresis was seen in this group only when theloss of potassium was large (50 to 65 mEq).When any of the 3 high-salt dogs is comparedwith a low-salt dog whose potassium loss wassimilar, it appears that the high-salt dog lost moresodium; but, with only three pertinent observa-tions in the high-salt group, the data must be in-terpreted with caution.

5. Miscellaneous data. During the CO2 periodthe weight change for the high-salt group averaged- 0.5 kg, and for the low-salt group, - 1.5 kg.The hematocrit showed no consistent change.

The electrolyte losses in the stools were notsignificantly different in the control and CO2 pe-riods. In both groups the daily loss of sodiumaveraged about 4 mEq per day. The loss of po-tassium averaged 1 in the high-salt group, and 6mEqper day in the low-salt group. The chlorideloss was about 1 mEqper day in both groups.

DISCUSSION

These studies indicate that adaptation to chronicrespiratory acidosis in the dog is a gradual proc-ess, is uninfluenced by the presence or absenceof sodium chloride in the diet, and leads to onlyincomplete correction of the extracellular hy-drogen ion concentration. Specifically, the datashow a rise in plasma bicarbonate concentrationover 6 or 7 days-independent of the sodiumchloride intake-followed by a plateau which fallsshort, by 5 to 10 mEq per L, of the level neces-

sary to restore the normal ratio of bicarbonate tocarbonic acid.

The initial steep ascent in bicarbonate concen-tration, which usually comprised half or more ofthe final elevation, was often achieved with littleor no increase in renal acid excretion. The incre-ment in bicarbonate must, therefore, have beendue largely to tissue buffering, a mechanism whoseeffectiveness has previously been demonstratedin the nephrectomized dog (6) and which is nowseen to serve an important function even when thekidneys are intact. Calculation of the red bloodcell contribution to the buffering process (7) in-dicates that no more than one-third of the bicarbo-nate could have been derived from this source, andit follows that other tissues, such as muscle orbone, must have made the major contribution tothe initial rise in extracellular bicarbonate. How-ever, after the first day the cumulative incrementin urinary net acid excretion was usually muchgreater than would be required to account forthe concomitant increase of bicarbonate in the ex-tracellular fluid. It may well be that the excessbicarbonate generated in the kidneys during thisphase was consumed in the neutralization of hy-drogen ions initially buffered in the cells.' Sucha restoration of cellular buffer capacity might beexpected, since a rise in extracellular bicarbonateconcentration will produce a favorable gradientfor the movement of hydrogen ions out of cells.However, it should be noted that this interpre-tation rests upon the assumption that the endog-enous acid load was constant. In view of thenegative nitrogen balance and increased excretionof organic acids and sulfate which regularly oc-curred during hypercapnia, it seems likely that, infact, the endogenous acid load increased, and tothis extent the augmentation of renal acid excre-tion cannot be considered as contributing to thecompensation of the respiratory acidosis. Untilmeans of computing the net acid or alkaline loadresulting from tissue breakdown become available,

1 For convenience the internal changes in buffer storeshave been expressed in terms of hydrogen ion movement.However, the data do not permit any distinction to bemade between the transfer of hydrogen in one directionand the transfer of bicarbonate or hydroxyl ions in theother. The effect on extracellular bicarbonate concen-tration will, of course, be the same in each case.

1234


no final interpretation of the observed increase inrenal acid excretion will be possible.

In each animal the rise in plasma bicarbonatewas accompanied by a nearly equal fall in chloride.The cumulative increment in chloride excretionwas of an order of magnitude sufficient to accountfor the fall2 in plasma concentration-an obser-vation which has also been made in the rat (1).How a high CO2 tension induces the kidney toreject chloride is obscure. As has previously beensuggested (1), it is possible that the chloruresisis due to a primary effect of the high pCO2 in ac-

celerating hydrogen transfer into the glomerularfiltrate, an action that would reduce the electricalgradient along which chloride is thought to move

passively. Alternatively, the elevation of pCO2might have the direct effect of lowering the perme-

ability of the tubular epithelium to chloride. Ifthis were so, the diminished reabsorbability ofchloride might in itself contribute to the acceler-ated transfer of hydrogen ions into the glomeru-lar filtrate (3) and thus to the increased reab-sorption of bicarbonate. Another possibility isthat hypercapnia depresses the proximal reab-sorption of sodium and, hence, of chloride. If, inturn, some of the sodium were then exchanged inthe distal tubule for potassium and ammonium, a

selective urinary loss of chloride would result.Some insight into the interrelationship of chlo-ride and hydrogen ion movement is provided byobservations during recovery from chronic re-

spiratory acidosis (5).In addition to the changes in renal acid and

chloride excretion, there were significant altera-tions in the excretion of other electrolytes. Onthe first day in the CO2 atmosphere, a cation diu-resis occurred, which consisted largely of potas-sium in the low-salt group and of both sodiumand potassium in the high-salt animals. It is notclear whether the cation increment was a conse-

quence of displacement of these ions from cells[analogous to that which occurs in the nephrec-

2 In Dogs K and J the initial fall in serum chlorideconcentration occurred without significant chloruresis.This hypochloremia can, in part, be attributed to thetransfer of chloride into red blood cells, but the magni-tude of the fall suggests that there was, in addition, a

movement of sodium and water from other tissues intothe extracellular space. By the end of the study, how-ever, these animals had lost sufficient chloride to fullyaccount for their hypochloremia.

tomized dog (6)] or whether the kidney was pri-marily responsible.

In the latter part of the study, some of the dogswith access to sodium showed a tendency to re-tain sodium, but in most animals a slight but sig-nificant deficiency persisted to the end of the study.This finding emphasizes that bicarbonate reab-sorption can be greatly enhanced, without con-comitant sodium retention, simply by increasingthe proportion of bicarbonate and decreasing theproportion of chloride which is reabsorbed withthe sodium.

After the first day of exposure to CO2 therewas usually no potassium loss beyond that to beexpected in association with the negative nitrogenbalance. Indeed, the potassium deficit, correctedfor nitrogen loss, was repaired in all but two ofthe animals, possibly as a result of the movementof hydrogen ions out of cells into the decreasinglyacid extracellular fluid. In any event, the evi-dence that the specific deficit was restored by theend of the period of adaptation indicates that po-tassium deficiency cannot be invoked as a factorin the increased capacity for bicarbonate reab-sorption. Moreover, in the two dogs in which sig-nificant potassium depletion persisted throughoutthe CO2 period, the renal response to hypercapniadid not appear to be modified in any way.

While the data demonstrate a suppression oftitratable acid, a rise in urine pH, and a slight in-crease in bicarbonate excretion throughout theperiod of exposure to CO2, it should be notedthat the technique of the study may have intro-duced an artifact into the distribution of theseacid-base moieties in the urine. Each day it wasnecessary to remove the animals for brief periodsfrom the high CO2 atmosphere, and it is possiblethat a modest bicarbonate diuresis at this time wasresponsible for the continued excretion of alkali,The addition of bicarbonate to the 24-hour urinewould, of course, also elevate its pH and diminishits content of titratable acid. Since only the firstday's urine was formed during uninterrupted hy-percapnia, it is only on this day that such an arti-fact can be excluded. However, the finding in anumber of animals that bicarbonate excretion roseappreciably on the first day, despite the fact thatplasma level had risen only moderately, suggeststhat the maximal ability of the kidney to conservebicarbonate had not yet been achieved, a finding

1235


consistent with earlier observations on reabsorp-tive capacity of such animals during acute bicarbo-nate loading (8).

The response of the dog to exposure to a highCO2 atmosphere resembles that of the rat in thata large rise in plasma bicarbonate concentration isaccompanied by an increase in the urinary outputof chloride, ammonium, potassium, and phosphate(1, 2). The only striking difference is the muchlonger period required for adaptation in the dog.However, this latter characteristic provides valu-able evidence that the various adaptive responses,which in the rat occur almost simultaneously, arenot in fact interdependent. In addition to demon-strating the importance of tissue-buffering in theinitial stages, the dog also shows a clear dissocia-tion between the ammonium response and the in-crements in the output of chloride, phosphate, andpotassium. These increments are largely con-fined to the first day, while the rise in ammoniumexcretion is usually delayed and always persistent.Thus, the ammonium response cannot be attrib-uted to the anion diuresis. It is also a matter ofsome interest that the pattern of cation excretionis qualitatively similar to that seen after the in-duction of metabolic acidosis by the administrationof ammonium chloride (9, 10). As in metabolicacidosis, there is initially a diuresis of sodium andpotassium, and only later a rise in ammoniumexcretion.

SUMMARY

Balance studies were carried out in 11 dogs ex-posed to an atmosphere of 11 to 13 per cent car-bon dioxide for periods of 6 to 15 days. Aconsistent pattern of response was found, charac-terized by a sharp rise in plasma bicarbonate con-centration during the first day of exposure, and asubsequent slower rise over the next 5 or 6 daysto final concentrations ranging from 35 to 38mEqper L. During the first day, when approxi-mately one-half of the total rise in plasma bi-carbonate occurred, there was often little or noincrease in renal acid excretion, indicating thattissue buffers played a major role in the initialdefense of the extracellular pH. Subsequently,however, the increment in net acid excretion wasfar in excess of the amount required to accountfor the estimated increase in extracellular bicarbo-

nate. It has been suggested that the excess ofbicarbonate may well have been utilized in theneutralization of hydrogen ions diffusing fromcells as the extracellular bicarbonate concentra-tion rose. The plasma bicarbonate approached aplateau which fell short, by 5 to 10 mEq per L,of the level necessary for full compensation of theacidosis. The sodium chloride content of the dietappeared to exert no significant effect on theadaptive process.

The plasma chloride concentration varied in avirtually reciprocal fashion with the plasma bi-carbonate concentration throughout the study. Ineach dog there was a chloruresis and a cumulativechloride loss of sufficient magnitude readily toaccount for the degree of hypochloremia. A cleardissociation was shown between the increments inchloride and in ammonium excretion.

On the first day in the CO2 atmosphere therewas a marked increase in the excretion of potas-sium and phosphorus and a variable sodium diu-resis. Subsequently, both sodium and potassiumexcretion returned toward or to the control level.At the end of the study there usually remained aslight deficit of sodium, but in most instances, af-ter correction for changes in nitrogen balance,there was no evidence of potassium deficiency.

REFERENCES

1. Levitin, H., Branscome, W., and Epstein, F. H. Thepathogenesis of hypochloremia in respiratory aci-dosis. J. clin. Invest. 1958, 37, 1667.

2. Carter, N. W., Seldin, D. W., and Teng, H. C.Tissue and renal response to chronic respiratoryacidosis. J. clin. Invest. 1959, 38, 949.

3. Bank, N., and Schwartz, W. B. The influence ofanion penetrating ability on urinary acidificationand the excretion of titratable acid. J. clin. In-vest. 1960, 39, 1516.

4. Schwartz, W. B., Bank, N., and Cutler, R. W. P.The influence of urinary ionic strength on phos-phate pK2' and the determination of titratable acid.J. clin. Invest. 1959, 38, 347.

5. Schwartz, W. B., Hays, R. M., Polak, A., and-Hay-nie, G. D. Effects of chronic hypercapnia on elec-trolyte and acid-base equilibrium. II. Recovery,with special reference to the influence of chlorideintake. J. clin. Invest. 1961, 40, 1238.

6. Giebisch, G., Berger, L., and Pitts, R. F. The ex-trarenal response to acute acid-base disturbancesof respiratory origin. J. clin. Invest. 1955, 34,231.

1236


7. Davenport, H. W. The ABC of Acid-base Chem-istry: The Elements of Physiological Blood-gasChemistry for Medical Students and Physicians,4th ed. Chicago, Univ. of Chicago Press, 1958.

8. Sullivan, W. J., and Dorman, P. J. The renal re-

sponse to chronic respiratory acidosis. J. clin.Invest. 1955, 34, 268.

9. Gamble, J. L., Blackfan, K. D., and Hamilton, B.A study of the diuretic action of acid producingsalts. J. clin. Invest. 1924-25; 1, 359.

10. Sartorius, 0. W., Roemmelt, J. C., and Pitts, R. F.The renal regulation of acid-base balance in man.

IV. The nature of the renal compensations in am-

monium chloride acidosis. J. clin. Invest. 1949,28, 423.

1237

co,dm5migu4zj3pb.cloudfront.net/manuscripts/104000/104353/jci6110… · data on the early...

Documents