The
III.
Distribution of Body Fluids in
Congestive Heart Failure
Exchanges in Patients during Diuresis
By R. D. SQUIRES, M.D., A. P. CROSLEY, JR., M.D.,
AND
T. 1H. ELKINTON, M.D.
The balance technic was applied in six studies of patients with congestive heart failure during the
diuresis of their edema fluid. From the data so obtained changes in the extra- and intracellular
phases of the body fluids were calculated, as well as transfers of sodium and potassium between the
several phases and between the body and external environment. The results indicate that abnormial-
ities of intracellular fluid exist in this condition.
Downloaded from http://circ.ahajournals.org/ by guest on June 15, 2017
main types of disturbaitee are ^v-oith investigating for such light as may be shed on both
the etiology and the therapy of the condition.
The purpose of the study reported ill this
paper wvas to ascertain the principal changes
in the distribution of body fluids which take
place when a cardiac patient loses the major
portion of his edema by diuresis. By quantitating these changes in respect to the water
and some of the electrolytes of both the extracellular and intracellular phases, it was hoped
to learn something of the abnormalities which
had originally been presenit.
S INDICATED in the first paper in this
series' the edema of cardiac failure has
for many years been considered to be
essentially an isotonic expansion of extracellular fluid. More recently, evidence has accumulated which suggests that the intracellular fluid
is involved as w-ell in the physiopathology of
this condition. The existence of an abnormally
low concentration of sodium in extracellular
fluid and serum, as reported by other workers
A
and
as
documented in the preceding paper,2 inl
itself suggests that the distribution of water between the tvo phases of body fluid must be
abnormal. Furthermore, at least one intracellular electrolyte has been implicated since
EXPERIMENTAL IMATERIAL AND X\IEITDI(SS*
Six studies w-ere conducted in five diffelent pa~
tients with massive peripheral edema due to heart
failure during the diuresis of the major portion of
their edema. On five of the six occasions the dinresis
was effected by the rlepeated administration of
mercurial diuretics. The sixth study mwas of a patient
who had a spontaneous diuresis on beti rest without
the use of mercurial lrugs. One patient (L.).) was
reported reten-
Newman and co-workers3 have
tion of potassium during the loss of cardiac
edema and Fox, Friedburg and White4 have
claimed that the administration of potassium
to some cardiac patients results in the diuresis
of their edema. Fox has postulated that the
abnormal body fluid pattern in these patients
is one of hypotonicity with an excess of extracellular, and a deficit of intracellular, base.
While it is unlikely that such a complex disturl)ance can be reduced to simple terms, the
studied twice, once during the summer heat when
she sweated profusely-, and again three months later
during the cooler weather when sweating wsas not
clinically I)eicel)tilble. Three of the other foul patients wdere studied in the winter time.
The balance technic Adas used for the investigation.
The patients were weighed daily and were maintained on measured amounts of (lialvze(l sodium-frec
milk (Lonala(c) or whole milk. Occasionally othler
fluids were given; water was allowed ad libituml but
was measure(. The urine was collected over 24 hlour
From the Department of lMedicine anled Hospital
of the University of Pennsylvania, Philadelphia, la.
Laboratory facilities and personnel were aided by
grants from the National Heart Institute, United
States Public Health Service, anid Ib the C. Mahlon
Kline Fund of the Department of Medicine.
Research Fellow
During this study, A. P. C.
of the American Heart Association, anird J. I. E.
an Established Investigator of the American Heart
Association.
was
*The laboratory assistance of M\rs. Claire Tissari
and Mrs. Helen Lightman is especially ackimow-le(lge(l
for the studies reported in this paper aos(l in the suc-
a
was
ceeding
868
paper.
Circulationl, lOlum( IV, Derernher, 1.9.51
R. D. SQUIREIS, A. P. CR(OSLEY, J1., AND J. It. lILKINTON.8
Downloaded from http://circ.ahajournals.org/ by guest on June 15, 2017
p)eriods. Representative aliquots of the dialyzed milk
and the daily uines were analyzed for chloride,
sodium, potassium, and nitrogen, and the urine for
(reatinine. Analyses of blood ancl serum for the
l)rincipal electrolytes were performed at the begrinning and end of each experimental period. Feces
were not analyzed since only very small amounts
were passedl (luring the regimen of the liquid dliet and
since formed stools of the size l)assed contained
amounts of the electrolytes which are quantitativel
insiginificant5 in relation to the exchanges under
investigation. The volume of distribution and the
clearance of mannitol or inulin were obtained at the
l)eginning or end of five of the six studies, or at
1)oth times.
Sodium and potassium were (letermined in serum,
urine, and nitric acid digests of milk by means of a
l3aclavl internal standard flame p)hotometei accordino to the methods of Wallace, Holliday, Cushman,
an111d Elkinton.6
(Clhloride swas measured in serumi by the method of
lviseiii.mIi7 and in urine by that of Harv ey The total
carbon ldioxi(le content of serum was measured by
the method of Van Slyke and Stadie.9 Nitrogen in
urline was determinedl by the macr.okjel(lalhl technic.
(Creatinine in urine was determine(l according to
the method of Brod an(l Sirotal" aml(l in serum accor(linlg to the method of Bonsnes an(l Tausskv-." Inulin
Ximd mannitol determinations in urine and serum were
ina(le by the method of Harrison'2 and a modificatio(I of the method of Smith ani(l co- workers 1 resp1)ectiv ely; spaces and clearances being determine(l
acording to the formulas, In-Out/P and I /P,
riesl)e(ctively-, where P is the concentration in l)lasma
an1d (Tl the amiount of the substances in urine.
(Culculation, of Results
The (leriveol ldata were calculate(l froim the balance
hmta a(cord(ing to lrevious iiietllo(ds14 15; for the
sake of (laritv these methods are reviewed here.
Change in total water (AW) is calculated from the
chlange in weight (AWt.) correctedl for the solids lost
a.n(l the food burnecl, 1) the fommula'6
Allr = AWt. + (Se - Si) + (C + F + 0.54P)
where 5, and Si = solids of excreta anol ingesta,
resl)e(tively, an(l C, P, anol F = carbohy(lrate, protein, an(l fat buned, resl)ectively. The metabolic
mixture is estimated as follows. C.ribolhd-(riate (C)
burned is taken as the caibohydiate given. Protein
(I') burned is taken as equivalent to the nitrogen ex(retedl in the urine X 6.25. The fat l)urne(l is calculated from the insensible weight loss (IL) according
to time forim1ula.i'7
F = (IL - 2.12C - 1.69P) /3.78
InI the eases studied here the dlailv insensible weight
loss as measure(l frequently exceeded by a consilernal)le amount the prol)able rate of vap)orization
86;9
of water for purposes of heat expenditure; the cause
of this high rate of insensible weight loss (luring
mercurial diuresis is at present unknown. * But for
this reason, a (laily insensible weight loss of 1 Kg. is
assumeed in these patients and the fat burned calculated oIn the basis of this assumption.
Change in extracellular fluid voluamle (AE) is calculated from the change in concentration of chloride
ill sermi correctedl for seruim water and a Donnan
equilibrium factor of 0.95 ([Cl]E), the balance of
chloride (b)cm, anld anI initial extracellular volume
(E1) either assumIe(l or measuredl, in the following
muanner:
E2 = (El [Cl]E1 + bCI)I[CI]E
AE = E2-El
In the l)atients studie(l here E1 was taken as the
v olume of (listribution (of inulin or mainitol and
AE was c(al(ulate(l oI this basis either forward or
b)ack(1 a1id(1 in1 time a(coirling to whethler the inulin
sI)ace ms(ldetermined at the beginning or end of the
stud(l. I1 l)atient I.). (I) ani inu1iIni spLace was
measuredl at botlh beginning and end of thle study;
the change ini the ('alculate(l (hloridle spa(e differe(l
from that measured foi the iiiulin space by less
than 9 per cent.
Change ill intracellular fluid (AJ) is (aleulated
simply as the difference between the (hianges inl total
water and in1 extracellular fluid:
AI = Al - AE
Fron1 the values for the volumes of extracellular
fluid (E1 and E,), from the concentration of so(lium
andl l)otassium ill extracellular fluid ([Na]E an1d
[K]E), and fiom thieir external balances (bN, and bK),
the exchanges of these two ions between the extraandl intia(ellular l)hases of the total body- fluid's
(ANaE and.xNa1a, AKE and AsK,) are (alclated as
folloxs:
.ANnlE = ELIN'I]E2 - E'[INa E
ANa, = bxa - ANa\ E
AKE = E2[K]E2 - EEUI{IE
A1K = bK- A}K
Change ill intracellular potassiumAlK, is furtlhel ('Orrected by the nitrogets balance (bN) to indicate that
is transportion of intracellular potassium which1
ferined independently of the catabolism or ain.a1)olisi
of tissues, or' "ill excess of nitrogen" (AK,'):
AKLI,, = AK - (2.4b)
m-hIere bN is exl)ressedl ill (mIll.
These exchanges of cationis betwee l)phtases are
basedl omi the assumption that chloride remains extracellular. Although this may not always be soiS 20
* lE(luivoctl evidence for anI immCrease ill inlsellsible
Nv ater loss (lurimig inercurial (iuresis has meemin meporte(l.36 37
870
l~ ~I - -cI
BODY FLUIDS IN CONGESTIVE HEART FAILURE
Ci t- 0
oS
+ C+ -c;
Ci
Ob
to
I o
C CO
ozI.0C:O
10
.
0CO0
O eO10
sIi
I
0
0m
-0
UID
00
0_
CO
10
1
I.
1N
R.
auilu!
CO
+
Downloaded from http://circ.ahajournals.org/ by guest on June 15, 2017
Q)
*~t.
ad
P.I
"
m
CO
Co
C.
C| Ic
1
'4l)
| t t: t
10c~
10
oq)
P'-)
X
Pe3
CO
±
+
1 1
CO
cCl
COC
+
+
III I
U,
CO
-
04
+
1
CO
t.
00
C
cc
t.
-
CO
NC'm
N
_ _CO'
Ic
cr
m
|C4 N _0
C
9
.,-
l
C
U
-i
O00 ~'-_
000C
11
10;
NcN t
_0 CO
eCO
I_4
COIOl
I
O
C C1l
C OCO
c
Coo
0cl]I0
0N
__
0cl]c IC
00
C,
C0
O
00
N
-~~~~~CO
q
ce
CO0
' f 0C00O
IOIOC'I
0Cl
m
IO
4a
I_
C'1
CO COC
=
-
C
0C
C0
Ot-N
N
O
_|
00_III
tI
_0t
c)
U,
|000N tm
"
L 10CICDC
LN
-iCOt-
_I
_
4.
-O00LO
-o
t--
1
-10oo
0
l _
0ko
_II
I_
C)
-
<
|
tN
'-4~~~cq
~~ ~COCOCO
m
Z
++
I
0
C~It4M
0
|
0
r-
O0 m
:1$t21
0,
"
I_
I
M C,
CO
"
k-
o
c~~~~
0
000
__
l
CN
CO
CO
CO
1
C
COCo~ttrC-
o0
to
+
t 4
|_ t
cl]
to |mN
00N I0
1
10
10to
r
m
C:5
-c:4
.
o0
)
._
t~-C'4l
CO
0
to
O
+-
Cq
+
~
C
t
0)
U:
°
1,* I l****
o~
_q
CO
~,
m
Il Il
I
10
0
+++ + +++ + ++++ +
C
o
m C 1D0
-o CD
to
b
00Cb
I
-O
Cm
ee
mC
-
m
c. 0CO
"
±
++
t
COO
o0
C
o0
p00
CD
1
+II+++ ++
c:Cd
CO
10OCQ
o~
CO 0c
mb Co CO CN
CO
CO k- Ci -
CO Cl CO
I t-+
10
_
OcOC m
_IIII
I~00coCoIOCOCO
_ SO
10C
t-
I
-o
m
t
-0
0k
004.
-m0C0
II I0
I0 ICO
000
oIN
0m
m
C
C.
CO
4.
0 3
Q--
cj
I
~
~ ~
~
~
CO0>
1_
~
_ a) s o
(I
a
-
t
Ic4I
~
I
I
Cl
1
N
c-
OI
-C_o
I
o
+
2~~~~~~
0
1
I
CICICs
dz V
H
CIC
I
I
I
- I
C I C£ CI
00
9)
t.
<
I-
II
eo s
to
I
CCII
CO
10C'1IC
II
III
-7'41
O+E
Co I 0 Iooo
I
+E
I ~ ICICICIo~t + I
I ln
CIC<ttIC
10X
S
I
b
l-b
U2
b
I t g -
01
-- +dICOCOCl]O
cl I
", ~
~ N;
~
°q M
-->
I ~ ~~
I
~
~
~Cq Ns
~
~
~
I-
~
.
s-- Cs
Iq
I
C
I-
co
0CDc-4I I
o
cI
I
_.
IIIIIII II
C
IDC\]clim
_
c0m~
C)
C.OcOCO
oq[2>t> |f
N N C,
m to
< t
IC oA
- - )C sC
0
m
0 rl
V-
I~~~~
C~
~~.
0
4
r.
CO
m
0+~~~~~~~~~~~~
4m11CS>
R. D. SQUIRES, A. P. CROSLEY, JR., AND J. R. ELKINTON
Downloaded from http://circ.ahajournals.org/ by guest on June 15, 2017
the probability is against there being enough of a
shift of chloride across the phase boundary to invalidate the direction of the calculated transfers of
cations. Where such an error does occur it will affect
the calculated changes in intracellular sodium
(ANar) to a far greater extent than those of potassium (AKI) because of the predominantly extracellular position of the sodium ion.
Change in osmotically active cell base (AB) is
calculated from the change in total water (zW) and
from the concentration of sodium in extracellular
fluid ([Na] ,); phase calculations derived from the
chloride balance or from other measurements do not
enter into this calculation. Assuming an initial
volume of total body water (W,) to be equal to
0.7 body weight,* and the concentration of osmotically active base in the body fluids ([ ]) to equal
[Na]g + 10 mEq. per liter, the initial and final
total amounts of osmotically active base in the body
fluids (B, and B2) are calculated from the formula:
B, = WI[BI1
B2 = (W, + AW)[BI2
From these values and from the observed balance of
sodium plus potassium (bNa + K) the change in
osmotically active base (ABi) is calculated:
ABi
= (B2- B)
-
bNa+K
The ratio of many of the balances of these two
ions approached more closely than did the excretion rates, the normal ratio of sodium to
chloride in extracellular fluid of 1.3 to 1.0
(fig. 1B). This appeared not to be true for some
of the larger negative balances when the rates
of excretion of the ions exceeded to a marked
degree the rate of excess ingestion of chloride
over sodium.
The amount of potassium excreted each day
was not excessive and a positive balance of poTABLE 2.- Renal Clearances of Creatinine and Inulin
or Mannitol
Avge.
Patient
24 hr.
creatinine
clearance
6/16-20
6/20-24
Crea- Inulin Mantinine
nitol
ml.
per
m n.
mnl.
per
min.
.
6/27
46
54
9/29-10/1
37
9/30
57
72
10/1-3
10/3-6
40
33
10/6
54
74
1/4-7
1/7-10
1/10-12
38
45
39
1/12
68
87
2/15
71
86
E. F.
6/3-5
6/6-7
6/7-9
Ml.
per
min.
40
44
30
32
31
6/24-27
J. IT.
a
-a
12/17
12/18
L. 1)., 1
M. K.
Acute ci learance
ml.
per
mn,.
F. J.
L. D.,
in the initial stages, W1V. Xeas so assumed for the end
of the experiment and the calculation made haekwar(l
in time.
Period
dates
The validity of this calculation has been discussed
elsewhere.2'
RESULTS
The results are presented in tables 1 to 3
and in figures 1 to 4.
Urinary Excretion and Total Balances of
lVater and Electrolytes (tables 1 and 3). During
most of the time of observation each patient
was in a negative water balance which ranged
in magnitude from 6.0 plus to 18.3 liters. Diuresis through the kidney accounted for most
of the electrolyte loss and much of the water
loss. A considerable portion of the water loss,
however, appeared to be due to an unusually
high rate of formation of "insensible" water of
vaporization and sweat which took place in
most of the patients.
The urine contained sodium and chloride in
large and almost equimolar amounts (fig. IA).
The balances of sodium and of chloride waere
negative and were of considerable magnitude.
*
In these studies of patients who were edematous
871
68
86
76
tassium was present, in all of the patients except F.J.; in this patient the lowest intake and
the highest excretion rate of the series resulted
in a negative balance of this ion. Positive nitrogen balances were maintained in all of the
patients except F.J., whose intake of nitrogen
wX-as minimal.
Data on the rate of glomerular filtration in
these patients during the periods of diuresis
are presented in table 2. As measured by the
24 hour clearance of endogenous creatinine the
glomerular filtration rate was markedly de-
C.t
c.
It
C4I
Downloaded from http://circ.ahajournals.org/ by guest on June 15, 2017
COD
BODY FLUIDS IN CONGESTIVE HEART FAILURE'
r
872
R. D. SQUIRES, A. P. CROSLEY, JR., AND J. R. ELKINTON
Io
N
CD
_
C
C
1Pcl
-
1+1
m
cq
e~ ~ ~ ~ ~ ~ ~ l
o C.)
C)
_
bD
_~
CS
C
00
00
0
C)
CO
+
+
+
m
;<
00o
CS
o
o
CO
e
I
s
._
_+
cO
I1+
u
*0
CS
CSC
C,)
0O
t_
~~~~~~~~~~~~~e
.=
coom
+
Downloaded from http://circ.ahajournals.org/ by guest on June 15, 2017
0S00
0d1
e
O
c; CS
Cl-
ce
o
M' CS
c
O~~~~~~~~
00C'N
_~
z
0
cc
-t
U)
C
oo
CO
_I
CC
.e
++
o0
*
Ill
1, ,I
2~
~
~
~
~
~-
~
~
~
~
~
0
r
: -!
a
0
C)
U2
c
~
C)
0
C.)
~ *_
,.,
~
_O
~
C. N m
-)U2
C.)~~~~~~~~~~~~l
C;
C-
-N
0v
-S
bx~~~~~~~~~C
00
00
dzz
r.
Q:
b£ :
s
cd
:S
_
e
-CCOO
t
.__bfo
o
oe
0o
c,
ICE=<
~
ze ~
0o so
+++++~
Xn
c, c,
a, W
c
0
I-4
1.
C
11,_ t
+-
coo~t
c
_
CCO
C.)
COD
873
BODY FLUIDS IN CONGESTIVE HEART FAILURE
874
pressed in the three patients receiving
mercurials in whom this measurement was
made. At the beginning or end of periods of
study, acute clearances of creatinine and of
inulin or mannitol were measured in five of the
six patients. The results indicated somewhat
higher values than those of the 24 hour creatinine clearances in the patients in whom both
acute and chronic clearances were obtained,
but the acute clearances do not invalidate the
interpretation that these diureses occurred in
the majority of periods in the patients who received mercurials; these findings reflected the
metabolic alkalosis as defined in the preceding
paper.2
Calculated Changes in the Phases of Body
Fluids (table 3). The exchanges of water and of
the principal electrolytes as calculated in this
group of patients were remarkably uniform. The
reduction in extracellular water volume ranged
from 4.1 to 11.5 liters. Since the loss of total
body water was from 6.0 plus to 18.3 liters, a
Downloaded from http://circ.ahajournals.org/ by guest on June 15, 2017
Balance
Urinary
sodium,
of
sodium,
UVN0
bNo
meq.per
24 hras.
meq. per
24 hbs.
A
200
400
600
800
Urinary chloride,
meq.
UVc,
per
B
0
-200 -400 -600 -800
Balance of chloride,
bcl
24 hrs.
meq. per
24 hrs.
FIG. 1. Sodium and chloride transfers during diuresis: relation (A) of rates of urinary excretion
and (B) of balances, of the two ions.
Data are plotted for 24 hour periods during which mercurials were given (solid symbols), and for
similar periods in the patient who diuresed without mercurial therapy (open circles). Identical
symbols represent periods in same patient. The solid line defines a ratio of sodium to chloride of
1.0:1.0; the broken line indicates such a ratio of 1.3:1.0 (the normal ratio of these two ions in extracellular fluid).
A. In most of the periods, whether or not mercurials were given, the excretion ratio of sodium
to chloride was closer to 1:1 than to 1.3:1, during diuresis.
B. In many of the periods the ratio of the balances of these two ions approaches more closely
a 1.3:1.0 than a 1:1 ratio. This appears not to be true for some larger negative balances which
greatly exceed the amount of chloride ingested in excess of sodium (in dialyzed milk or as ammonium chloride).
the presence of critically lowered rates of
glomerular filtration. The 24 hour creatinine
clearance was somewhat less depressed in the
patient who had a diuresis without mercurial
therapy.
Serum Electrolyte Concentrations (table 3).
The serum sodium level was normal or slightly
elevated at the onset of diuresis in each patient
and tended to fall slightly during the diuresis
in the five patients who received mercurials.
The serum potassium level remained within
normal limits throughout the studies. The
serum carbon dioxide content was slightly elevated and the chloride level depressed during
definite portion of the water lost (15 to 62 per
cent) came from the intracellular phase.
Large amounts of sodium were lost from the
extracellular phase; the changes in intracellular
sodium content were equivocal. Potassium,
however, was taken up in excess of nitrogen
into the intracellular phase. In the four patients
who were observed continuously for more than
four days, this increment of potassium equalled
4.3 to 8.4 mEq. per Kg. of body weight; these
increments are comparable to those observed in
other types of potassium depletion of moderate
degree.22
5Cry~ ~
R. D. SQUIRES, A. P. CROSLEY, JR., AND J. It. ELKINTON
cent of the sodium lost externally from
the body.
The course of these events in one of the
patients, L.D. (II) is illustrated in figure 2 (ob-
Change in the amount of osmotically active
base present in the total body fluids of each
patient, as calculated from the serum sodium
concentration and the balances of total water,
Patient:
55 i
._.
Therapy:
with congestive failure
oalyzed) milk
E
Digitoxin
0Hq a
PMercurials (HQ)
Arteriosclerotic
L. D. ,
2+
Edema - peripheral
Serum conc.
145 I-
-
Na
K
Downloaded from http://circ.ahajournals.org/ by guest on June 15, 2017
3.s _
~o
o
135 .
_
0
5
meq. per liter
of:
balance
Daily
Sodium
intaksi
-
mew.
output
0
-
50-
Doy
of
lo.
Z
clearance
cc /min.
A
i
100W
100
creatinine
\
-
l 50in
S
Total f luid
study
--"0
!W
-
24 hr. endog.
55.
4,
400
+ 100
kg.
a
Discharged
edema -free
+2001
Potassium
Weight
.l.0AC
---
0
-200
60-
disease
heart
__owNo
N
4a.h
" p
5.5
875
1
1
n
R
-
4
intake
E
3
output
liters
2
urine
U
-
a
|
1
F1
iIF
2
3
4
5
V
a
6
7
FIG. 2. The course of one of the patients, L. D. (II), during diuresis of her edema: observed data.
There are shown the various items of treatment, the clinical degree of edema, the serum levels
of sodium and potassium, the daily balances of these ions, the daily clearance of creatinine, the
weight, and the total intake and output.
The diuresis of the edema took place with no marked changes in the serum levels of sodium and
potassium or in the rate of glomerular filtration as measured by creatinine. The balances of sodium
were markedly negative while those of potassium were positive.
sodium, and potassium, was in the negative
direction in each case. The magnitude of this
"loss" of osmotically active base, which presumably took place within the cells, was large
and was roughly equivalent to 24 to 108 per
served daily balances, serum electrolyte levels,
and weight) and in figure 3 (calculated cumulative changes in the body fluids); the alterations in body fluid patterns are presented diagrammatically in figure 4.
BODY FLUIDS IN CONGEISTIVE HEART FAILURE
87 (;
clearance which are somewhat lower thait the
acute creatiinine clearances, might all be explained by failure to collect all the urine passed
DISCUSSION
Certain of the unusual data obtained in these
studies might well be ascribed to errors ocArteriosclerotic
L. D. ,
Patient:
with congestive
(diolysed) milk
a
Di__
disease
heart
55 9 N
failure
-Low No
Therapy:
W
_ercuiolf,
Hg
Hg
4+
2+ ____
Edema - peripheral
__a
Cumulative change in:
Extracellulor
0-
01
Downloaded from http://circ.ahajournals.org/ by guest on June 15, 2017
Water
A Ect
Sodium
-2
-
400-
AEin
-
A NaE
-4
- 800
-6
liters
-1200
meq
AE
,
-8
Intracellu lar
o-
Potassium
AK
Water
-2-
A1
Sodium
-4.
liters
-
A Na1
Al
-6-
meq.
Total f luid
intake
output (urine)
lt
titers
Day of study
1
2
3
4
5
6
7
FIG. 3. The course of one of the Iuatients, L. D. (11), during diuresis of her edema: calculated
changes in body fluids.
The cumulative changes are shown of extracellular water volume (AECI), extracellular sodium
(ANaE), intracellular water volume (Al), intracellular potassium (A-KI,) and sodium (\Na1), and
iii total osmotically active base (AB,).
During the diuresis of the edema water was lost from both extra- and intracellular phases, extrtacellular sodium diminished and intracellular potassium increased. The decrement of osmotically
active base inactivated within the cells, appeared to be of approximately the same magnitude as
the amount of sodium lost externally from the body.
curring in the balance technic. High rates of
insensible water loss, discrepancies between
total water loss as calculated from weight loss
and the balance of sodium plus potassium, and
the values for 24 hour endogenous creatinine
during the metabolic periods. However, the
very constancy of these findings in most of the
studies militates strongly against, such anl explanation.
In these patients the well knowi fact was
It. 1). SQUIRES, A. P. CROSLEY, JR., AND J. R. ELKINTON8
Downloaded from http://circ.ahajournals.org/ by guest on June 15, 2017
corroborated that in the diuresis of the edema
of congestive heart failure, large amounts of
extracellular sodium and water are lost from
the body.2:325 Our data also agree with those of
Blumgart and his co-workers25 in that chloride
was excreted by the kidneys in almost equal
amounts with sodium. As indicated above, and
as illustrated in figure 2, in many of the periods
the balance of sodium tended to exceed the
balance of chloride and to approach the extracellular ratio of the two ions. The difference
between sodium:chloride ratios of urinary excretion and of balances is due to the fact that
all of the patients were ingesting more chloride
than sodium, either in the dialyzed milk or as
ammonium chloride. Two interpretations are
possible, therefore, of the more nearly equimolar ratio of the ions in urine. One is that the
kidney was making an adjustment to the excess
intake of chloride. The other is that under the
conditions of diuresis the kidney tends to excrete as much chloride as sodium and that the
ratio of the balances of the two ions was shifted
fortuitously toward that of extracellular fluid
by the ingestion of chloride in excess of sodium.
Support for this latter interpretation is found
not only in the data of Blumgart and associates2" and of other workers,23, 24 but also
in the fact that during the periods of the largest
diuresis the ratio of the balances again tended
to approach that of 1:1. In such periods the
excess chloride ingested would have relatively
less effect o01 the net chloride balance.
If this latter interpretation is correct, namely
that a diuresis tends to sweep out chloride in
relatively greater amounts than sodium, at
least a partial explanation has been found for
the hypochloremic metabolic alkalosis present
in these patients and in other cardiacs who have
received mercurial diuretics.2 The question remaiiss unanswxered, however, of why the kidney
does not, under these circumstances, excrete
more bicarbonate and less chloride, and so preserve the normal pattern of electrolyte
concenitrations in extracellular fluid. This might
be accounted for by the specific inhibition by
the mercurial drug of tubular reabsorption of
chloride as well as of sodium; some other explanation is required in the case of the patient
(J.P.) who had a spontaneous diuresis. What-
877
ever the explanation the data appear to indicate
that a relative chlorur-esis is one factor ii the
metabolic alkalosis found in these patients.
It may be presumed that the diureses of
these patients were effected primarily as a result of altered renal tubular transfers of sodium,
chloride, and water. Our data confirm the findings of others2' 26, 27 that under the influence
of mercurial drugs, diuresis of edema may occur
without significant elevation of the glomerular
filtration rate above the so-called "critical"
L.D., IL
Intracellular (I)
Extracellular (E)
HCO3
before
Mercurial
diuresis
after
H20°
10
0
10
20
liters
Fi;. 4. Diagramnmatic representatioti of the
changes in body fluids which occurred in one of the
patients, L. D. (II), during diuresis of her e(demna.
Concentrations of solutes are plotted along the
ordinates, volumes of fluid phases along the abscissas,
therefore areas represent amounts of solutes. Cation:
anion patterns are superimposed on the area of e ich
fluid phase. Cross-hatched area indicates solutes 11o
longer osmotically active. Broken lines indicate approximate normal dimensions of the bo(dN fluids.
Sodium and chloride were lost from the extracellular phase, potassium entered the imit rc.t(ellular
phase, water was lost from both phases, andicl a large
amount of int racellular solute b)ecane osmotically
inactivate I.
level of about 70 ml. per minute.28 Although the
chronic and acute clearances of enidogeniouls
creatinine were somewhat lower than the corresponding acute clearances of inulini, the data
indicate that the glomerulatr filtration rate in
all of these patients was below or in the vicinity
of the "critical" level mentioned above.
Transfers of intracellular fluid clearly were
involved in the removal of the edema from these
patients. Water was lost from cells and concomitantly potassium, when given, was taken
up. This would appear to be a paradoxic situation, especially since intracellular sodium was
878
BODY FLUIDS IN CONGESTIVE HEART FAILURE
not lost in equivalent amounts. Our interpretation of these findings is that the change in total
osmolarity must have been the result of some
Downloaded from http://circ.ahajournals.org/ by guest on June 15, 2017
process of osmotic inactivation within the cell.
The nature of such process can only be conjectured at present, but it is not inconceivable
that it is due to changes in size and degree of
dissociation of various potassium phosphate
and proteinate molecules which take place during processes of cellular metabolism. This hypothesis of changes in intracellular osmolarity
has not been completely substantiated but the
sheer magnitude of the discrepancy between
water shifts and cation transfers demands some
explanation other than error in observation or
calculation of the data. Previous work21 29 and
other workers3'}3- have found evidence of such
a phenomenon in other conditions.
The loss of intracellular water, whatever the
cause, by these patients would hardly fit in
with the theory suggested in the first paper'
that intracellular dehydration might stimulate
the production of antidiuretic hormone (ADH).
It is probable that prior to the diuresis the cells
of the body and the osmoreceptors were overhydrated and antidiuretic hormone production
inhibited to such an extent that an effective
stimulation of antidiuretic hormone did not
take place during the diuresis. No evidence is
presented, of cource, that the antidiuretic hormone played any abnormal role in the water
transfers of these patients. But aside from a
possible effect on a specific cellular locus (the
osmoreceptors of the posterior pituitary), this
evidence of large changes in the osmolarity of
intracellular solutes lends support to the thesis
presented in the first paper' that modifications
of cellular functions are involved in the pathogenesis of congestive failure.
In conclusion, the exchanges in these patients
can be regarded in two ways. It can be considered that the exchanges represent the result
of an abnormal stress placed upon the body by
mercurial drugs or other unidentified diuretic
factors, or they can be taken to represent the
correction of an abnormal pattern of body fluid
distribution already existing in the body. If
the latter interpretation is accepted, the following attributes of that abnormal pattern may be
defined: (1) excess of extracellular water,
sodium, and chloride, (2) deficit of total intracellular potassium, but (3) increased osmolarity
of solutes remaining in cells, and hence (4)
excess of water within the cells.*
SUMMARY AND CONCLUSIONS
Six studies in five patients with edema due
to cardiac failure were conducted during partial
or complete diuresis of their edema. By means
of the balance technic exchanges were calculated of water, sodium, and potassium between
the several phases of body fluid.
Large amounts of water, sodium, and
chloride were lost from the extracellular phase.
Chloride tended to be excreted in equimolar
proportions to sodium.
Water was lost from, and potassium was
taken up into, the intracellular phase of body
fluids. Osmotic activity of solutes within the
cells appeared to be decreased.
It is concluded that the intracellular phase
shares in the body fluid abnormalities of congestive heart failure, and that these intracellular abnormalities consist of potassium depletion, increased osmolarity of solutes present,
and overhydration.
ACKNOWLEDGMENTS
This work was done in the Chemical and Renal
Sections of the Department of M\Iedicine; the help
of the laboratory staffs and of the clinical staff of
the Hospital is gratefully acknowledged.
REFERENCES
1 ELKINTON, J. R., AND SQUIRES, R. D.: The distribution of body fluids in congestive heart
failure. I. Theoretic considerations. Circulation
4: 679, 1951.
2 SQUIRES, R. D., SINGER, R. B.,yMOFFITT, G. R.,
*
Since this paper was written Iseri, Boyle, and
Myers38 have reported similar studies. Using the
balance technic on patients recovering from congestive failure, they found that water was lost from,
and potassium and sodium were taken up into, the
intracellular phase, and that cellular base was inactivated. They, concluded that the stage of congestive
failure is characterized by deficit of cellular potassium
and sodium and increase of cellular osmolarity and
hydration. With the exception of the cellular sodium
transfers, our studies confirm their observations.
R. D. SQUIRES, A. P. CROSLEY, JR., AND J. R. ELKINTON
Downloaded from http://circ.ahajournals.org/ by guest on June 15, 2017
JR., AND ELKINTON, J. R.: The distribution of
body fluids in congestive heart failure. II. Abnormalities in serum electrolyte concentration
and in acid-base equilibrium. Circulation 4:
697, 1951.
3 NEWMAN, E. V.: Functions of the kidney and
metabolic changes in cardiac failure. Am. J.
Med. 7: 490, 1949.
4 Fox, C. L., JR., FRIEDBURG, C. K., AND WHITE,
A. G.: Electrolyte abnormalities in chronic congestive heart failure: effects of administration
of potassium and sodium salts. J. Clin. Investigation 23: 781, 1949.
5 PETERS, J. P.: Body water: the exchange of fluids
in man. Springfield, Ill., Charles C Thomas,
1935. P. 199.
6 WALLACE, W. I., HOLLIDAY, M., CUSHMAN, 1\/I.,
AND ELKINTON, J. R.: The application of the
internal standard flame photometer to the analysis of biological material. J. Lab. & Clin. i\Ied.
37: 621, 1951.
7 EISENMAN, A. J.: A note on the Van Slyke method
for the determination of chlorides in blood and
tissue. J. Biol. Chem. 82: 411, 1929.
8 HARVEY, S. C.: The quantitative determination
of the chlorids in the urine. Arch. Int. Med. 6:
12. 1910.
9 VAN SLYKE, D.
D., AND STADIE, W. C.: The determination of the gases of the blood. J. Biol.
Chem. 49: 1, 1921.
10 BROD, J., AND SIROTA, J. H.: The renal clearance
of endogenous "crleatinine" in man. J. Clin.
Investigation 27: 645, 1948.
11 BONSNES, R. W., AND TAUSSKY, H. H.: On the
calorimetric determination of creatinine by the
Jaffe reaction. J. Biol. Chem. 158: 581, 1945.
12 HARRISON, H. E.: A modification of the diphenylamine method for determination of inulin. Proc.
Soc. Exper. Biol. & M-\ed. 49: 111, 1942.
13 ELKINTON, J. R.: The volume of distribution of
mannitol as a measure of the volume of extracellular fluid, with a study of the mannitol
method. J. Clin. Investigation 26: 1088, 1947.
14 ELKINTON, J. R., AND TAFFEL, _M.: Prolonged
water deprivation in the dog. J. Clin. Investigation 21: 787, 1942.
15 ELKINTON, J. R., AND WINKLER, A. W7.: Transfers
of intracellular potassium in experimental dehydration. J. Clin. Investigation 23: 93, 1944.
16 PETERS, J. P., KYDD, D. M\., AND LAVIETES, P. H.:
A note on the calculation of water exchange.
J. Clin. Investigation 12: 689, 1933.
17 LAVIETES, P. H.: The metabolic measurement of
the water exchange. J. Clin. Investigation 14:
57, 1935.
18 D)EAN, R. B.: Theories of electrolyte distribution
in muscle. Biol. Symp. 3: 331, 1941.
19 WILDE, W. S.: The chloride equilibrium in muscle. Am. J. Physiol. 143: 666, 1945.
879
20 BURCH, G. E., THREEFOOT, S. A., AND RAY, C. T.:
Rates of turnover and biologic decay of chloride
and chloride space in the dog determined with
the long-life isotope, Cl36. J. Lab. & Clin. Med.
35: 331, 1950.
21 ELKINTON, J. R., WINKLER, A. W., AND DANOWSKI, T. S.: Inactive cell base and the measurement of changes in cell water. Yale J. Biol. &
Med. 17: 383, 1944.
22 TARAIL, R., AND ELKINTON, J. R.: Potassium deficiency and the role of the kidney in its production. J. Clin. Investigation 28: 99, 1949.
23 KEITH, N. M., AND WHELAN, M.: A study of the
action of ammonium chlorid and organic mercury compounds. J. Clin. Investigation 3: 149,
1926-27.
24 HARRIS, L., RUBIN, E. L., AND LAWRENCE, J. S.:
Salyrgan and ammonium chloride as diuretics
in cardiac oEdema. Acta med. Scandinav. 83: 23,
1934.
25 BLUMGART, H. L., GILLIGAN, D. R., AND VOLK,
M. C.: Action of diuretic drugs. II. Effect of
diuretic drugs on the acid-base equilibrium of
the blood in patients with cardiac edema. Medical Papers Dedicated to Henry Asbury
Christian. Baltimore, Waverly Press, 1936.
P. 191.
26 FARNSWORTH, E. B.: Electrolyte partition in patients with edema of various origins. Sodium
and chloride. Am. J. Med. 4: 338, 1948.
27 GROSSMAN, J., WESTON, R. E., EDELMAN, I. S.,
AND LEITER, L.: Studies on thiomerin a subcutaneously administerable mercurial diuretic.
Circulation 1: 508, 1950.
28 MERRILL, A. J.: Mechanisms of salt and water
retention in heart failure. Am. J. Med. 6: 357,
1949.
29 ELKINTON, J. R., WINKLER, A. W., AND DANOWSKI, T. S.: Transfers of cell sodium and potassium in experimental and clinical conditions.
J. Clin. Investigation 27: 74, 1948.
30 YANNET, H., AND DARROW, D. C.: The effect of
depletion of extracellular electrolytes on the
chemical composition of skeletal muscle, liver,
and cardiac muscle. J. Biol. Chem. 134: 721,
1940.
31 MIELLORS, R. C., MIUNTWYLER, E., AND MAUTZ,
F. R.: Electrolyte and water exchange between
skeletal muscle and plasma in the dog followving
acute and prolonged extracellular electrolyte
loss. J. Biol. Chem. 144: 773, 1942.
32 SELDIN, D. Wr., AND TARAIL, R.: Effect of hypertonic solution on the metabolism and excretion
of electrolytes. Am. J. Physiol. 159: 160, 1949.
33 GAUDINO, M., AND LEVITT, AI. F.: Influence of
the adrenal cortex on body water distribution
and renal function. J. Clin. Investigation 28:
1487, 1949.
14WELT, L. G., JR., ORLOFF, J., KYDD, 1). ir., AND
880
BODY FLUIDS IN CONGESTIVE HEART FAILURE
OLTMAN, J. E.: An example of cellular hyperosmolality. J. Clin. Investigation 29: 935,
1950.
35SMS, E. A. H., ELT, L. G., ORLOFF, J., AND
NEEDHAM, J. W.: Asymptomatic hyponatremia
in pulmonary tuberculosis. J. Clin. Investigation 29: 1545, 1950.
'6 GABRIEL, G., AND KAHLER, H.: Ueber den Einfluss
des respiratoris chen Quotienten auf die Hiohe
der Gewichtsabgabe. Zur Frage der perspiratio
insensibilis bei Kreislaufkranken. Wien. Arch. f.
inn. Med. 24: 181, 1934.
3 MAGENDANTZ, H., AND STRATMAN, F.: Ueber die
extrarenale Wasserausscheidung bei Herzinsufficienz. Deutsches Arch. f. klin. M\ed. 174: 1,
1933.
8 ISERI, L. T., BOYLE, A. J., AND MYERS, G. B.:
Water and electrol-te balance during recovery
from severe congestive failure
50 milligram sodium diet. Am. Heart J. 40: 706, 1950.
on
a
Downloaded from http://circ.ahajournals.org/ by guest on June 15, 2017
The Distribution of Body Fluids in Congestive Heart Failure: III. Exchanges in Patients
during Diuresis
R. D. SQUIRES, A. P. CROSLEY, JR. and J. R. ELKINTON
Downloaded from http://circ.ahajournals.org/ by guest on June 15, 2017
Circulation. 1951;4:868-880
doi: 10.1161/01.CIR.4.6.868
Circulation is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1951 American Heart Association, Inc. All rights reserved.
Print ISSN: 0009-7322. Online ISSN: 1524-4539
The online version of this article, along with updated information and services, is located on
the World Wide Web at:
http://circ.ahajournals.org/content/4/6/868
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally
published in Circulation can be obtained via RightsLink, a service of the Copyright Clearance Center, not
the Editorial Office. Once the online version of the published article for which permission is being
requested is located, click Request Permissions in the middle column of the Web page under Services.
Further information about this process is available in the Permissions and Rights Question and Answer
document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Circulation is online at:
http://circ.ahajournals.org//subscriptions/