Electrolyte and Water Metabolism: Physiologic Considerations*

Electrolyte and Water Metabolism: Physiologic Considerations * EUGENE Y. BERGER, M.D.t J. MURRAY STEELE, M.D.t

LIFE without water cannot exist, yet until recent years the estimate of the water content of the living has been limited to measurements in the dead organism. The determination of the water content of the human being by desiccation was reported almost 100 years ago by Bischoff in three subjects.! Volkman2 presented data in 1874 which consisted of organ weights as a percentage of body weight (his own and other observations), and from the percentage water of each of the organs the total body water was calculated. There seem to be no other data on the water content of adult man until 1945, when Mitchell and co-workers 3 reported 67.8 per cent water in a 35 year old, 70.5 kilogram man who apparently died of rheumatic heart disease and congestive heart failure. Data on the water content of man, dead at that, are meager.

BODY WATER AND SUBDIVISIONS

Body Water Measurements

The first successful measurement of water in living man was made in 1934 by Von Hevesy and Hofer,4 who fed deuterium oxide to a man. By estimating the degree of dilution of deuterium oxide after sufficient time had been allowed for equilibrium to take place, it appeared that From the Research Service, Third Medical Division, Goldwater Memorial Hospital, Welfare Island, and Department of Medicine, New York University College of Medicine, New York.

* Studies on which this article is based were supported in part by the Public Health Research Ins!itute of the City of New York. t Instructor in Medicine, New York University College of Medicine; Research Assistant, Research Service, Goldwater Memorial Hospital, New York. t Professor of Medicine, New York University College of Medicine; Director, New York University Research Service, Goldwater Memorial Hospital, New York. 829

830

Eugene Y. Berger, J. Murray Steele

the man contained 62 per cent water. Moore 5 in 1946 arrived at a figure of 72.5 per cent in a single male subject to whom he had administered deuterium oxide intravenously. In 1947 Pace, Kline, Schachman and Harfenist,6 using tritium oxide, obtained a figure for body water in one male of 64.7 per cent. Hollander and colleagues7 in 1949 measured body water by dilution of deuterium oxide in two normal subjects. The values obtained were 54.9 and 58.5 per cent of body weight. Data available in this laboratory suggested that antipyrine, one of the antipyretics which Dr. B. B. Brodie was studying, appeared to be equally distributed throughout body water. Further study in animals and man confirmed this notion. 8 , 9 The measurement of body water was somewhat simplified by this method. A gram of antipyrine is injected intravenously. After equilibrium has occurred, three blood samples are analyzed at hourly or longer intervals. The concentration of antipyrine is plotted on the logarithmic scale of semilogarithmic paper against time on the arithmetic scale, and the resultant straight line is extrapolated back to the time of injection to obtain the concentration of antipyrine if equilibrium was instantaneous. The amount of antipyrine administered, divided by the concentration in plasma water, indicates the amount of water in which the antipyrine is dissolved-in this case, total body water. When the volume of dilution of antipyrine was compared with that of deuterium oxide by simultaneous measurements in eight normal subjects, agreement was fairly good. Antipyrine space averaged 1.2liters less than deuterium oxide; the greatest difference was 3.1 liters. Subsequently, a metabolite of antipyrine, 4-acetyI4-amino-antipyrine (NAAP), was also found to be distributed evenly throughout the water of various tissues of the body.IO Simultaneous measurements of antipyrine and NAAP spaces in twelve subjects agreed fairly well. Thus four substances, deuterium oxide, tritium oxide, antipyrine, and NAAP, are distributed evenly throughout tissue water and are diluted to the same degree in the body. Measurement of the specific gravity of the body also yields information as to the amount of water in the body,ll· 12. 13 The body can be regarded as composed of a mixture of fat of low density and fat-free tissue of relatively high density. From the measurement of the specific gravity of the body as a whole (by weighing the subject in air and under water), the proportion of either fat or fat-free tissue can be calculated. The sum of all the fat-free tissue has been termed "lean body mass." The lean body mass was shown by Behnke's studies and later by Pace and Rathbun l4 to exhibit considerable constancy as to water content (average 73 per cent). Furthermore, a number of mammalian species exhibited little variation from this figure. Water content was calculated from measurement of the specific gravity

Electrolyte and Water Metabolism

831

and from dilution of antipyrine in nine subjects. 16 The agreement was good. Calculation of the water content of the "lean body mass" from the antipyrine figure affirmed the constancy of the proportion of water in lean body mass, yielding an average figure of 71.8 per cent. Osserman, Pitts, Welham and Behnke 16 extended measurements of specific gravity and antipyrine space to a large number of young men, 81 in all. The correlation of specific gravity and proportion of water was excellent. The more obese subjects of low specific gravity had of course proportionately less water. Recently, additional confirmation of the validity of both specific gravity and antipyrine dilution as measures of body water in cattle has been obtained by comparing the values obtained by these two methods with those obtained by a fat extraction procedure after slaughter of the animalP Thus the amount of water in the body estimated by methods based on the dilution of chemical substances agrees well with the amount estimated by a method based on an entirely different principle, namely, the measurement of the specific gravity of the body. That these divergent procedures agree well lends confidence to the belief that any one of them is a measure of total body water within the limits of error of the method chosen. The results of recent measurements of total body water are presented here, and it turns out that the amount of water is somewhat less than has been generally thought. Body water was measured by the dilution of antipyrine in a group of hospital patients and personnel.l8 In 51 males, body water ranged from 40 to 68 per cent and averaged 53 per cent. In 31 females (this was the first time estimations of body water had been made in women) body water ranged from 30 to 53 per cent and averaged 45 per cent. Osserman and co-workers,16 in Bethesda, using the same technic-namely, dilution of antipyrine-obtained a higher average than ours. Their average figure was 61 per cent, range 43 to 73, in a group of muscular young men. Using deuterium oxide, Moore and his group19 in Boston have measured the total water content in a group of 17 young males and 11 young females. The males averaged 62 per cent body water, range 56 to 70, and the females, 52 per cent, range 46 to 55. The average values for peJ;centage of body water found by Osserman's group (61 per cent) and Moore's group (62 per cent) in males are higher than the value obtained in this laboratory (53 per cent). Presumably our value is lower because of the nature of the subjects. Many of them were older and lived in a chronic disease hospital and were presumably less well muscled. The range of the proportion of water in the body is, however, wide in all the studies reported. The variation is apparently due, for the most part, to variation in body fat. That fat is the great variable in body composition is illustrated by four subjects chosen from the group of subjects studied in our laboratory because their lean

Eugene Y. Be:rge:r, J. Murray Steele

832

body masses were the same (39 to 41 kg.), but their body weight varied from 60 to 106 kg. (Fig. 64). Man has been generally thought to be approximately 70 per cent water. The values given here are obviously much lower and present considerable variability. It is evident that for a man to contain 70 per cent water means that he has practically no fat. When tissues · of the body are dried, the loss in weight is approximately 75 per cent, but these tissues are carefully cut free of fat or the fat is extracted with organic solvents before consistent values for water content are obtained. 20 The effect of fat in tissues on water and electrolyte analyses

25 0

WATER

WATER

WATER

WATER

:SOLlDS~

§SOLlDS~

~SOLlDS~

~SOLlDS~

F

F

M

M

Fig. 64. Val;iation of body fat in subjects with the same water content.

has long been recognized. Moulton21 (1923) quotes Pfeiffer (1887), who quotes Hosslin: "It has been known for a long time that fat deposition in the tissues exerts a great influence on the water content of the same; the organism of the fattened animal contains less water, the richer it is in fat. The question is, according to Voit, in part, one of the replacement of water by fat but chiefly one of deposition of water-free fat in the tissues so that after the removal of the latter the organism shows a nearly normal water content." Clearly, a consideration of the fat content explains the low and variable values of body water which are obtained when water is measured in vivo. Extracellular Fluid Measurements

These low figures for total body water warranted re-examination of the question as to the distribution of water inside and outside cells.

Electrolyte and Water Metabolism

833

Several methods for measuring extracellular fluid have been in use, namely, thiocyanate, bromide, radio sodium and radio chloride.22 , 23, 24, 26 In these studies bromide was used as a measure of what has been termed the chloride or extracellular fluid space. The extracellular space was measured by dilution of sodium bromide, and t0tal body wlitter was measured by dilution of antipyrine in 82 subjects. 26 The intracellular water was then calculated as the difference between the total water and the extracellular water. The figure arrived at in this way averaged 15 liters in males and 11 liters in females. Such values for intracellular water seemed unreasonably small. Indeed, it may be calculated that this is not enough intracellular water to account for the minimum amount of intracellular water which is in muscle alone. 26 In the light of more precise knowledge of total body water it became clear that the existing methods for measuring extracellular fluid grossly overestimated that body of water. This fact was not unexpected. It has become increasingly apparent the chloride enters many cells of the body. For instance,. chloride ions enter the cells of the gastric and intestinal mucosa, of sweat and salivary glands, of the renal tubules, of the liver to be excreted in the bile, and the red blood cells. A more accurate measure of extracellular fluid was plainly needed. Kruhoffer27 in 1946 published work from the University of Copenhagen which indicated that in the nephrectomized rabbit and dog the volume of distribution of inulin was smaller than that of thiocyanate. Inulin is known not to enter the red blood cell nor the cell of the kidney tubule. 28 For these reasons inulin was selected as a likely substance for the measurement of extracellular fluid. The difficulties of using inulin in the living animal are, however, great; they center around its slow diffusion and rapid excretion, but many of them have been alleviated by the work of Gaudino, Schwartz and Levitt. 29 , 30 The procedure involves maintaining a constant infusion of inulin for a sufficient period of time to allow the inulin to come into equilibrium throughout the volume of water to which is has access. The urinary bladder is catheterized and, when equilibrium is complete, is emptied of' urine, which is discarded. The infusion is stopped at that moment, and all the subsequent urine which is excreted over the next forty-eight hours is collected and analyzed for inulin. As inulin is not metabolized, the inulin excreted after the infusion is stopped, represents the amount of inulin which was in the body at the end of the infusion. The inulin space is then calculated by dividing this amount of inulin by the plasma concentration of inulin at the end of the infusion, which represents the concentration of inulin when it was in equilibrium throughout its volume of distribution. In four subjects the dilution of bromide (Le., chloride space) was

834

Eugene Y. Berger, J. Murray Steele

measured simultaneously with the dilution of inulin. 26 The average chloride space was 17 liters, the inulin 10 liters, a value in agreement with that of other investigators.29, 31 If the volume of dilution of inulin is accepted as a measure of the anatomical extracellular space, then the amount of extracellular fluid is, as seemed likely, much less than had been previously thought. Since total body water in these four subjects averaged 35 liters as measured by the dilution of antipyrine, 25 liters are left over for intracellular water on the basis of inulin space, a much more reasonable figure than the 18 liters obtained by subtracting the chloride space. Subsequently, other compounds have been found to be diluted in the body to the same extent as inulin. These are mannitol,32 thiosulfate 33 and ferrocyanide. 26 , 34 The observations made in adult man are in close agreement. With due consideration of the inherent errors of the methodology, an average adult man contains 10 liters of extracellular water. Extracellular fluid may be partitioned into that portion which is in the vascular bed and that portion which is in the lymphatic vessels and between the cells. Circulating volume can be measured with red cells tagged with isotopic iron or phosphorus. 3S , 36 Plasma volume measurements which depend upon marking the albumin with dye (T-1824)37 or radioiodine 38 probably include along with plasma the lymph of the portal system. 39 The interstitial water may then be calculated as the difference between the circulating volume and the extracellular fluid. (The cerebrospinal fluid is not included in these measurements, since inulin does not penetrate the blood-brain barrier. 40 ) Summary of Measurements

At this point it may be well to summarize the newly found values for the various body compartments by giving values for the composition of an average man, due regard being held for the wide variability in amount of body fat. Average man is 60 per cent water, 28 per cent fat, and 12 per cent solids. The water may be divided into portions which are inside and outside the cells. Approximately 75 per cent of the total water is inside and 25 per cent outside the cells. Of the extracellular volume, one-third appears to lie within the blood vessels and twothirds between the cells. The present concept is compared to the classical values proposed by Gamble41 (Fig. 65). McCance and Widdowson42 have recently measured body water and subdivided it. They used thiocyanate as a measure of extracellular fluid and urea as a measure of total body water. The relationships of solids, water and fat are defined, but the methods used are so different from those used in the work discussed here that comparison of results is difficult.

Electrolyte and Water Metabolism

835

THE CHLORIDE ION

There is a large body of literature on measurement of extracellular fluid by the dilution of thiocyanate or bromide. The dilution of these substances in the body gives reproducible values for a volume of fluid, but it is obvious from the present discussion that this volume of fluid is not the same as the extracellular space. What is it, then? Thiocyanate and bromide were accepted as measures of the extracellular space because

BODY COMPOSITION

...

BO~~OW'

p.-_ _ _ _ _ _. ,

EXTRACELLULAR FLUID

EXTRACELLULAR FLUID

75

INTRACELLULAR

50

INTRACELLULAR FLUID

FLUID

25

FAT-and-SOLlDS SOLIDS

o

GAMBLE

AUTHORS' CONCEPT Fig. 65.

they are distributed in tissues and body fluids in proportion to their chloride content,22, 23, 43, 44. 45 the assumption being that chloride was, with certain exceptions, extracellular. Precisely how much of the chloride is included in the intracellular water has not been appreciated until the recent measurements of extracellular fluid with inulin. The best interpretation is that thiocyanate and bromide are not diluted in a certain volume of fluid, but are diluted in the total chloride of the body. Their dilution becomes a measure of the total body chloride in the same way that the dilution of an isotope is a measure of the total quantity of the isotopically normal element46-for example, the dilution of radioactive potassium to meaBure total body potassium. 47 Total body chloride may

836

Eugene Y. Berger, J. Murray Steele

be calculated from the relationship: Total body chloride: Administered bromide: : Plasma chloride: Plasma bromide

Total body chloride has been measured by the dilution of bromide in 52 subjects and found to average 1967 milliequivalents (mEq.)46 From the volume of the extracellular fluid (10 liters) and the concentration of chloride in it (110 mEq. per liter) is is readily apparent that the extracellular fluid accounts for only 1100 mEq. of the total chloride in the body. There remains then about 900 mEq. of chloride in the body which must lie within the cells. Therefore almost half of the body chloride is intracellular. If the 900 mEq. of intracellular chloride is evenly distributed throughout approximately 25 liters of intracellular water,26 the intracellular concentration of chloride may be estimated as 36 mEq. per liter. Of course there is considerable variability in the chloride concentration in the intracellular water of various tissues. In muscle cell water the concentration of chloride seems negligible, whereas in skin it is at least 110 mEq. per liter. 48 The concentration of chloride in red cell water is 75 mEq. per liter.49 In view of this variability it is difficult to interpret the meaning of the "average" value of 36 mEq. per liter. It is evident from these data that the volume of dilution of bromide as it is ordinarily calculated does not represent in any sense a volume of water in which the chloride ion is dissolved, since the concentration of chloride varies widely from tissue cell to tissue cell. EDEMA

The water content of the body, although normally constant, is clearly not static. Each day about two and a half liters of the existing 35 liters are exchanged for exogenous water in the diet. Thus half the water of the body is exchanged every ten days. (Moore 19 has found that the biological half life of deuterium oxide in the human is 9.3 days.) Yet, in the face of this continuous exchange of water, the body maintains a substantially constant water content by balancing the intake against the output. The intake of water comprises the food and water which is ingested. Food provides water from two sources-first, from its water content and, second, from its combustion. Carbohydrate yields 55 gm., fat 107 gm. and protein 41 gm. of water per 100 gm. of each burned. The output of water comprises the losses from the kidney, skin, colon and lung. The daily amounts of water taken in from these various sources and how they are excreted by the different organs of the body are tabulated in Table l. The water from foods and their combustion roughly balances the output of the skin, lung and colon so that the amount of water drunk each day about equals the urinary output.

8.'37

Electrolyte and Water Metabolism

Man gets into difficulties when, during this continuous exchange of water, he fails to maintain his status quo. The difficulties may occur in either direction; he may leave his happy medium by either having too much or not having enough. Water deficits occur either when there is not enough to drink, as for a castaway at sea, or when there are excessive losses, such as sweating at high temperatures. A water deficit may also occur as the result of a disturbance in posterior pituitary function (diabetes insipidus). Water excesses as pure water do not ordinarily occur except under experimental circumstances,61 since the normal kidney almost never permits dilution of body fluids. However, in many clinical conditions there is an increase in the water content of the body, particularly in the extracellular fluid, but the water contains salt to maintain the osmotic pressure. It must be remembered that even when there are losses of water, man does not lose pure Table 1 WATER BALANCE*

70 Kg. Adult INTAKE

EXCRETION

As fluids. . . . . . . . . . . . . . . . . . 1500 cc. In solid foods. . . . . . . . . . . . . .. 800 cc. From combustion .......... , 30 cc.

Urine ...................... , 1500 cc. Vaporization ~kin. . . . . . . . . . . . . . . . . . . . .. 600 cc. Lungs..... . . . . . . . . . . . . . . .. 400 cc. Feces. . . . . . . . . . . . . . . . . . . . . .. 100 cc. Total ....................... 2600 cc.

Total ....................... 2600 cc.

* From Dauphinee: Clinical Nutrition,50 Chap. 12. (Permission to reprint this table in part has been kindly granted us by the author and the publishers.) water, for it always contains a certain amount of electrolyte. For ex ample, sweat contains salt in hypotonic solution, and when man slakes his thirst after sweating by drinking water he may still be in difficulty because of a salt deficit. 62 Science has to tell him to take his salt tablet. In fact, man controls his water content far better than he does his salt content. Thirst is a compelling passion, but no man scours the countryside for a salt lick. When disease characterized by edema sets in, the problem is one of salt, not of water. Although the patient with congestive heart failure or cirrhosis of the liver may eliminate a test dose of water somewhat more slowly than the normal,63. 64, 66 drinking large amounts of water daily has been advocated with success in the treatment of congestive heart failure. 66 More specifically, the difficulty is one of sodium excretion, and not chloride. The kidney of the edematous patient can excrete ammonium, potassium, or calcium chloride with ease and sometimes with an increased amount of water.67 However, the urine of the patient who is actively accumulating edema fluid contains relatively small amounts

838

Eugene Y. Berger, J. Murray Steele

of sodium irrespective of the amount or form of sodium ingested. Under normal circumstances the kidney regulates the sodium content of the body by balancing the output against the intake. Why the kidney fails to do so in certain diseases is not well understood. It might be well at this point to define more precisely the problem of edema. Although the expansion of the intracellular water is a possibility, most of the disturbances in intracellular water are those of dehydration. Clinical edema is chiefly the expansion of the extracellular space, and it may even occur at the expense of intracellular water or from a redistribution of water in the body. For example, the edema of urticaria, or pulmonary edema of left ventricular heart failure, represents an expansion of the extracellular fluid without a gain in body water. Edema may also entail an increase in body water, but the forces causing it come into equilibrium and the edema ceases to accumulate: for example, the edema of venous or lymphatic obstruction or the edema of mild heart failure or certain instances of nephrosis in which weight remains stationary once edema has accumulated. Starling's definition of the relationships which govern the transfer of water from the blood across the capillary membrane to the interstitial spaces and back again place the formation of edema such as described on firm physiologic grounds. 58 However, the forces causing edema may not come into equilibrium; under these circumstances the patient continually gains weight from day to day as long as he is eating an ordinary diet. The kidney fails to balance the output of sodium against the intake, and the reason for such behavior is a point of controversy. Hormonal Disturbance in Edema Formation

The net transfer of sodium by the tubule of the kidney is from the lumen through the cell to the surrounding blood stream. Many factors affect this transfer (filtration rate, and so forth), hormonal control being one of them. The evidence for a hormonal disturbance in those diseases in which the accumulation of edema fluid forms a major clinical complication has been meager and equivocal. Thorn 59 in 1940 observed the exacerbation of ascites and edema during the premenstrual period in a patient with glomerulonephritis and in a patient with cirrhosis of the liver which suggested that gonadal hormones might at least affect the course of edema in these diseases. In 1942 Futcher and Schroeder, 60 in their demonstration of the impaired excretion of sodium chloride in congestive heart failure, suggested that the cause might be increased venous pressure of renal anoxia or "an endocrine mechanism possibly involving the adrenal cortex." In the last two years more substantial, although indirect, evidence has been forthcoming. Parrish 61 found in four of ten patients with congestive heart failure increased excretion of urinary corticoids which

Electrolyte and Water Metabolism

839

prolonged the life of adrenalectomized rats. Deming and Luetscher 62 report a greatly increased sodium-retaining activity (by a bio-assay method) in the urine of patients with massive edema due to nephrosis or heart failure. Albert and Smith,63 after the administration of ACTH, desoxycorticosterone or cortisone to normal subjects, found in five suhjects who developed marked peripheral edema all the hemodynamic changes of heart failure except a decreased cardiac output. Investigation undertaken in this laboratory afforded further evidence of a hormonal disturbance in certain disea"es in which edema fluid accumulated. Desoxycorticosterone affects the transfer of sodium across the kidney and the sweat gland,64 and it has been recently demonstrated that it also influences the transfer of sodium across the salivary gland and colon. 65 There is evidence, too, that the sweat and salivary glands in the patient who is accumulating edema act in a fashion which resembles excessive desoxycorticosterone-like activity.67 Investigations in this laboratory have now shown that the colon of the edematous patient also acts in a fashion which resembles excessive desoxycorticosterone-like activity. The measurement of the transfer of sodium across the sweat gland and the salivary gland is relatively simple, because the concentration of sodium is normally of sufficient magnitude to demonstrate a possible decrease. It is difficult, however, to document a decrease in excretion of fecal sodium in man because the amount of fecal sodium is normally small and variable. 68 In order to insure sufficient amounts of sodium in the stool to be able to measure the difference between the normal subject and the edematous patient a cation exchange resin was fed. 69 In order to interpret the data properly the mechanics of the action of the resin in relation to the transfer of sodium across the wall of the gut must be considered. 65 The exchange of ions on· the surface of the resin is almost immediate, and the pattern of electrolytes on its surface is dependent on the number and nature of the electrolytes in the surrounding medium. 70 . 71 The amount of sodium affixed to the resin is thus dependent on the amount of sodium available to it. When iom; leave the lumen of the colon, a new equilibrium is established between the sodium on the resin and that in solution in the colon contents. The balance between the transfers of sodium across the colon mucosa and the "attraction" of the resin for the sodium eventually determines how much sodium is retained on the resin and hence the amount of sodium appearing i~ the stool. The transfer of sodium across the colon mucosa in the normal subject was thus compared with the transfer in which the edematous patient by comparison of the amounts of sodium removed by the resin in the stool. Forty-five grams of carboxylic resin in the ammonium or potassium form, or a combination of both, was fed daily to two groups of patients.

840

Eugene Y. Berger, J. Murray Steele

The first, considered the normal control group, were patients on the wards of a chronic disease hospital without evident disturbance in their degree of hydration. The second group were patients who were actively accumulating edema fluid because of congestive heart failure or cirrhosis of the liver. Both groups received the same ward diet which contained 50 to 75 mEq. of sodium daily. The edematous patients never excreted more than 5 and usually less than 1 mEq. of sodium per day in the urine. The plasma sodium concentration of these patients was essentially normal (135 to 140 mEq. per liter) except for one patient (FF, Table 2) whose plasma sodium ranged from 128 to 132 mEq. per liter. In the control group of eight subjects who received carboxylic resin, the mean fecal sodium excretion of seventy-four observations was 42.7 mEq. per gram of nitrogen (Table 2). In the group of eight patients who were accumulating edema fluid because of cirrhosis of the liver or congestive heart failure, the mean sodium excretion of fifty-three observations was 20.8 mEq. per gram of nitrogen. The data indicated that under similar conditions of diet and resin administration the net transfer of sodium from the lumen of the colon to the surrounding blood stream is greater in the edematous patient than in the normal. The resin is consequently less effective in retaining sodium in the stool in the edematous patient than in the normal subject. Three patients of the group who were accumulating edema were also studied with respect to sodium excretion by the sweat and salivary glands. Two patients were suffering from congestive heart failure and one from cirrhosis of the liver. In each patient the concentration of sodium in the sweat and saliva was much reduced as compared to normal values (Table 3). In ten samples of hand sweat in seven normal subjects taken at various times, the mean concentration of sodium was 39 mEq. per liter, range 15 to 75 mEq. per liter. In two patients accumulating edema the concentrations of sodium in the sweat were 8 and 13 mEq. per liter. In thirty samples of saliva taken at various times in nine normal subjects the mean concentration of sodium was 35 mEq. perliter, range 10 to 61 mEq. per liter, whereas in three edematous patients the concentrations of sodium were 5, 7 and 8 mEq. per liter. The edematous patient, then, as compared to the normal, limits the transfer of sodium across the tubular epithelium of the kidney, the epithelium of the colon, and the epithelium of the sweat and salivary glands in such a fashion as to limit the loss of sodium from the body. To return, then, to the question of why the kidney of the edematous patient fails to balance the output of sodium against the intake. Alterations in filtration rate,72 renal arterial and venous pressure/3, 74, 76 oxygen tension 76 and plasma volume 73 are factors which have been investigated as individual entities and have been demonstrated to influence the renal excretion of sodium. The controversy resides in the relative importance

Electrolyte and Water Metabolism

841

Table 2 FECAL EXCRETION OF SODIUM AND POTASSIUM DURING RESIN ADMINISTRATION Comparison of the Normal with the Edematous Patient NORMAL Pt. --

A

B

C D D E

G

F H F

Resin Form

EDEMATOUS

No. SOdi-1 P?tasDiagnoSlUm Pt. Obs. urn sist

--

NH4 NH4 NH4 NH, K-NH4 NH4 NH4 NH4 NH4 K-NH4

9 10 S 7 7 5 6 S 7 7

mEq./gm. Nitrogent 65.4 25.3 42.7 26.3 33.2 53.3 3S.0 53.3 40.2 74.4 43.6 119.1 90.0 52.S 53.2 53.1 69.S 63.9 64.2 61.3

Mean

74

42.7

65.1

--

AA CC AA

BB

DD DD EE FF

GG

EE HH

CL CHF CL CHF CL CL CHF CHF CHF CHF CHF

Resin Form

No. SOdi-1 P?tasObs. urn SlUm

--

K-NH4 NH4 NH4 K-NH4 NH4 K K-NH4 NH4 NH4 NH4 NH4 Mean

2 3 4 2 16 3 2 2 1 3 15 53

Standard Deviation ±20.4 ±26.S Standard Deviation Standard Error ± 2.4 ± 3.1 Standard Error * CHF, Congestive heart failure; CL, cirrhosis of the liver. t Mean of stated number of observations.

mEq./gm. Nitrogent 12.2 54.3 13.4 121.7 14.5 33.S 15.4 31.0 20.0 61.S 20.0 9S.4 20.5 97.S 20.S SO.2 22.1 3S.2 24.7 57.5 25.9 67.6 20.S 66.7

±12.0 ±23.S ± 1.6 ± 3.3

Table 3 NET SODIUM EXCRETION Comparison of the Normal with the Edematous Patient --~--

COLON (Patient KIDNEY on Resin mEq./ Adminisday tration) mEq./gm. Nitrogen Congestive failure ..................... 0.6-2.7 Congestive failure ..................... 0.1-5.0 Cirrhosis ............................ " 0.3-2.5 Normals: Number of observations ........... " Mean .. .... . '" ...................... Range. _..... . . . . . . . . . . . . . . . . . . . . . . . .

----

SWEAT SALIVARY GLAND GLAND mEq./ mEq./ liter liter

22 15 20

S 13

5 7 S

53 40 25-64

10 39 15-75

30 35 10-61

of these factors and in what manner they play a part in the development of edema. The retention of salt in congestive heart failure has received considerable attention in the recent literature. The forward theory of congestive failure places the defect in the reduction of cardiac output

Eugene Y. Berger, J. Murray Steele which results in a reduced renal plasma flow and glomerular filtration rate. 78 , 79 The disturbed cardiovascular dynamics found in congestive heart failure are absent in cirrhosis of the liver, and the renal plasma flow and glomerular filtration rate are, furthermore, within normal limits.80 Nevertheless, as has been indicated in both these diseases, when the patient is actively accumulating fluid, there is evidence for difficulties in the transfer of sodium across other tissues than the nephron. From the fact that desoxycorticosterone has the ability to induce a similar disturbance in the transfer of sodium across the kidney, colon, salivary and sweat glands, it is inferred that the accumulation of edema in both cirrhosis of the liver and congestive heart failure is due in part to excessive desoxycorticosterone-like activity. Desoxycorticosterone-like activity occurs as a normal physiologic process when there are excessive losses of sodium from the body or when the dietary intake of sodium is sharply restricted 81 , 82, 83 and is manifested by limitation of the escape of sodium from the body through various membranes. Why the mechanism for limiting the escape of sodium from the body should go awry in diseases in which there is active accumulation of edema fluid is not known. Whether the biochemical stimulus is the same for the normal conservation of salt as it is for disease is not known. It is likewise unknown whether the difficulty is an excess of a desoxycorticosterone-like substance or a deficit of a possible antagonist. Whether the initiating mechanism is the same or different in various diseases is likewise not clear. Nonetheless, in these edematous states, all four organs of the body across which the transfer of sodium can be measured continue to limit the escape of sodium from the body. The accumulation of edema in these widely divergent clinical states would seem, then, to be mediated through some disturbance in hormonal balance. SUMMARY

1. Total body water has been measured in living man by dilution of several substances. About 60 per cent of the body weight of the average man is water. There are wide variations in water content, chiefly dependent upon variation in stores of body fat. 2. Inulin and other substances are perhaps better measures of extracellular fluid than thiocyanate, bromide or radiochloride. If measurements with inulin are accepted, then the extracellular fluid volume is considerably less than has been thought and comprises about a quarter of total body water, leaving three-quarters for intracellular water. 3. Nearly half of the total chloride of the body is within the cells on the basis of inulin dilution as a measure of extracellular fluid. 4. Evidence is presented that in certain patients accumulating edema, the sweat and salivary glands, the colon and the kidney, all tend to

Electrolyte and Water 111etabolism

843

limit the escape of sodium from the body. This evidence, it is thought, indicates that a hormonal imbalance is involved in edema formatioll. REFERENCES 1. von Bischoff, E.: Einige Gewichts-und Trocken-Bestimmungen der Organe des menschlichen Korpers. Z. ration Med. 20: 75, 1863. 2. Volkmann, A. W.: Untersuchungen iiber das Mengenverhaltness des Wassers und der Grundstoffe des menschlichen Korpers. Berichte u. d. Verhandl. d. sachs. Gesellsch. d. Wissenschaften zu Leipzig Math.-physiche Classe, 26: 202, 1874. 3. Mitchell, H. H., Hamilton, T. S., Steggerda, F. R. and Bean, H. W.: The Chemical Composition of the Adult Human Body and Its Bearing on the Biochemistry of Growth. J. BioI. Chem. 158: 625, 1945. 4. von Hevesy, G. and Hofer, E.: Elimination of Water from the Human Body. Naturc, 134: 879, 1934. 5. Moor, F. D.: Determination of Total Body Water and Solids with Isotopes. Science, 104: 157, 1946. 6. Pace, N., Kline, L., Schachman, H. K. and Harfenist, M.: Studies on Body Composition: Use of Radioactive Hydrogen for Measurement in Vivo of Total Body Water. J. BioI. Chem. 168: 459, 1947. }. Hollander, V., Chang, P. and CoTui: Deuterium Oxide and Thiocyanate Spaces in Protein Depletion. J. Lab. & Clin. Med. 34: 680, 1949. 8. Brodie, B. B., Axelrod, J., Soberman, R. and Levy, B. B.: The Estimation of Antipyrine in Biological Materials. J. BioI. Chem. 179: 25, 1949. 9. Soberman, R., Steele, J. M. and others: The Use of Antipyrine in the Measurement of Total Body Water in Man. J. BioI. Chem. 179: 31, 1949. 10. Brodie, B. B., Berger, E. Y., Steele, J. M. and others: Use of N-acetyl 4 Aminoantipyrine in the Measurement of Total Body Water in Dog and Man. Proc. Soc. Exper. BioI. & Med. 77: 794, 1951. 11. Behnke, A. R., Jr.: Physiologic Studies Pertaining to Deep Sea Diving and Aviation, Especially in Relation to the Fat Content and Composition of the Body. Harvey Lectures, 37: 198, 1941-1942. 12. Behnke, A. R, Jr.: Physiologic Studies Relating to Deep Sea Diving and Aviation, Especially in Relation to the Fat Content andComposition of the Body. Bull. New York Aead. Med. 18: 561, 1942. 13. Morales, M. F., Rathbun, E. N., Smith, R. E. and Pace, N.: Studies on Body Composition. Il. Theoretical Considerations Regarding the Major Body Tissue Components, with Suggestions for Application to Man. J. BioI. Chem. 158: 677, 1945. 14. Pace, N. and Rathbun, E. N.: Studies on Body Composition. Ill. The Body Water and Chemically Combined Nitrogen Content in Relation to Fat Content. J. BioI. Chem. 158: 685, 1945. 15. Messinger, W. J. and Steele, J. M.: Relationship of Body Specific Gravity to Body Fat and Water Content. Proc. Soc. Exper. BioI. & Med. 70: 316,1949. 16. Osserman, E. F., Pitts, G. C., Welham, W. C. and Behnke, A. R.: In Vivo Measurement of Body Fat and Water in a Group of Normal Men. J. Applied PhysioI. 2: 633, 1949-1950. 17. Kraybill, H. F., Hankins, O. G. and Bitter, H. L.: Body Composition of Cattle; Estimation of Body Fat from Measurement in Vivo of Body Water by Use of Antipyrine. J. Applied PhysioI. 3: 681, 195(}-1951. 18. Steele, J. M., Berger, E. Y., Dunning, M. F. and Brodie, B. B.: Total Body Water in Man. Am. J. PhysioI. 162: 313,1950. 19. Schloerb, P. R. and others: The Measurement of Total Body Water in the Human Subjcct by Deuterium Oxide Dilution. J. Clin. Investigation, 29: 1296, 1950. 20. Hastings, A. B.: The Electrolytes of Tissues and Body Fluids. Harvey Lp,ct.. 36: 91, 194(}-1941.

844

Eugene Y. Berger, J. Murray Steele

21. Moulton, C. R.: Age and Chemical Development in Mammals. J. BioI. Chem. 57: 75, 1923. 22. Crandall, L. A., Jr. and Anderson, M. X.: Estimation of the State of Hydration of the Body by the Amount of Water Available for the Solution of Sodium Thiocyanate. Am. J. Digest. Dis. 1: 126, 1934. 23. Brodie, B. B., Brand, E. and Lashin, S.: The Use of Bromide as a Measure of Extracellular Fluid. J. BioI. Chem. 130: 555, 1939. 24. Kaltreider, N. L., Meneely, G. R., Alien, J. R. and Bale, W. F.: Determination of the Volume of the Extracellular Fluid of the Body with Radioactive Sodium. J. Exper. Med. 74: 569, 1941. 25: Winkler, A. W., £lkinton, J. R. and Eisenman, A. J.: Comparison of Sulfocyanate with Radioactive Chloride and Sodium in the Measurement of Extracellular Fluid. Am. J. Physiol. 139: 239, 1943. 26. Berger, E. Y., Steele, J. M. and others: Estimation of Intracellular Water in Man. Am. J. Physiol. 162: 318, 1950. 27. Kruhoffer, P.: Inulin as an Indicator for the Extracellular Space. Acta physiol Scandinav. 11: 16, 1946. 28. Smith, H. W.: The Kidney, Structure and Function in Health and Disease. New York, Oxford University Press, 1951. 29. Gaudino, M., Schwartz, I. L. and Levitt, M. F.: Inulin Volume of Distribution as a Measure of Extracellular Fluid in Dog and Man. Proc. Soc. Exper. BioI. & Med. 68: 507, 1948. 30. Gaudino, M. and Levitt, M. F.: Inulin Space as a Measure of Extracellular Fluid. Am. J. Physiol. 157: 387, 1949. 31. Schwartz, 1. L., Schacter, D. and Freinkel, N.: The Measurement of Extracellular Fluid in Man by Means of a Constant Infusion Technique. J. Clin. Investigation, 28: 111, 1949. 32. Schwartz, I. L., Breed, E. S. and Maxwell, M. H.: Comparison of the Volume of Distribution, Renal and Extrarenal Clearances of the Inulin and Mannitol in Men. J. Clin. Investigation, 29: 517, 1950. 33. Schwartz, I. L.: Measurement of Extracellular Fluid by Means of a Constant Infusion Technique without Collection of Urine. Am. J. Physiol. 160: 526. 1950. 34. CalcagilO, P. L., Husson, G. S. and Rubin, M. I.: Measurement of "Extracellular Fluid Space" in Infants by Equilibration Technic Using Inulin and Sodium Ferrocyanide. Proc. Soc. Exper. BioI. & Med. 77: 309, 1951. 35. Gibson, J. G., 11, Peacock, W. C., Seligman, A. M. and Sack, T.: Circulating Red Cell Volume Measured Simultaneously by the Radioactive Iron and Dye Methods. J. Clin. Investigation, 25: 838, 1946. 36. Nachman, H. M., James, G. W., Ill, Moore, J. W. and Evans, E. I.: A Comparative Study of Red Ccll Volumes in Human Subjects with Radioactive Phosphorus Tagged Red Cells and T-1824 Dye. J. Clin. Investigation, 29: 258, 1950. 37. Gibson, J. G., 11, and Evans, W. A., Jr.: Clinical Studies of the Blood Volume. I. Clinical Application of a Method Employing the Azo Dye "Evans Blue" and the Spectrophotometer. J. Clin. Investigation, 16: 301, 1937. 38. Crispell, K. R., Porter, P. and Nieset, R. T.: Studies of Plasma Volume Using Human Serum Albumin Tagged with Radioactive Iodine. J. Clin. Investigation, 29: 513, 1950. 39. Peters, J. P.: The Role of Sodium in the Production of Edema. New England J. Med. 239: 353, 1948. 40. Berger, E. Y.: Unpublished data. 41. Gamble, J. L.: Chemical Anatomy, Physiology and Pathology of Extracellular Fluid. Cambridge, Mass., Harvard University Press, 1947. 42. McCance, R. A. and Widdowson, E. M.: A Method of Breaking down the Body Weights of Living Persons into Terms of Extracellular Fluid, Cell Mass and Fat, and Some Applications of It to Physiology and Medicine. Proc. Roy. Soc., SB, 138: 1951. 43. Wallace, G. B. and Brodie, B. B.: The Distribution of Administered Iodide

Electrolyte and Water Metabolism 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68.

845

and Thiocyanate in Comparison with Chloride, and Their Relation to Body Fluids. J. Pharmacol. & Exper. Therap. 61: 397, 1937. Wallace, G. B. and Brodie, B. B.: Distribution of Administered Bromide in Comparison with Chloride and Its Relation to Body Fluids. J. Pharmacol. & Exper. Therap. 65: 214, 1939. Weir, E. G. and Hastings, A. B.: The Distribution of Bromide and Chloride in Tissues and Body Fluids. J. BioI. Chem. 129: 547,1939. Dunning, M. F., Steele, J. M. and Berger, E. Y.: The Measurement of Total Body Chloride. Proc. Soc .. Exper. BioI. & Med. 77: 854,1951. Hevesy, G.: Radioactive Indicators. New York, Interscience Publishers, Inc., 1948. Eisele, C. W. and Eichelberger, L.: Water, Electrolyte and Nitrogen Content of Human Skin. Proc. Soc. Exper. BioI. & Med. 58: 97, 1945. Peters, J. P.: Body Water. Springfield, Ill., Charles C Thomas, 1935, p. 99. Dauphinee, J. A.: Sodium, Potassium, and Chloride Malnutrition, Including Water Balance and Shock, in Joliffe, N., Tisdall, F. F. and Cannon, P. R.: Clinical Nutrition. New York, Paul B. Hoeber, Inc., 1950, Chap. 12. Adolph, E. F.: Water Metabolism. Ann. Rev. Physiol. 9: 381, 1947. Marriott, H. L.: Water and Salt Depletion. American Lecture Series. Springfield, IlL, Charles C Thomas, 1950. Fremont-Smith, F.: The Mechanism of Water Diuresis in Man. J. Clin. Investigation, 9: 7, 1930. Fremont-Smith, F.: The Mechanism of Edema Formation. New England J. Med. 206: 1286, 1932. Ralli, E. P., Leslie, S. H., Stueck, G.H. and Laken, B.: Studies of the Serum and Urine Constituents in Patients with Cirrhosis of the Liver during Water Tolerance Tests. Am. J. Med. 9: 157, 1951. Schemm, F. R.: A High Fluid Intake in the Management of Edema, Especially Cardiac Edema. H. Clinical Observations and Data. Ann. Int. Med. 21: 937, 1945. Goodman, L. and Gilman, A.: The Pharmacological Basis of Therapeutics. New York, Macmillan Company, 1941, Chap. 35. Starling, E. H.: Mercer's Company Lectures on the Fluids of the Body. London, Archibald Constable, 1909. Thorn, G. W. and Emerson, K. W.: Role of Gonadotropic and Adrenotropic Hormones in the Production of Edema. Ann. Int. Med. 14: 757, 1940. Futcher, P. H. and Sehroeder, H. A.: Studies on Congestive Heart Failure. 11. Impaired Renal Excretion of Sodium Chloride. Am. J. M. Se. 204: 52, 1945. Parrish, A. E.: The Bioassay of Adrenal Corticoids in the Urine of Patients with Congestive Heart Failure. J. Clin. Investigation, 28: 45, 1949. Deming, G. B. and Luetscher, J. A., Jr.: Increased Sodium-Retaining Corticoid Excretion in Edema, with Some Observations on the Effects of Cortisone in Nephrosis. J. Clin. Investigation, 29: 808, 1950. Albert, R. E. and Smith, W. W.: The hemodynamic Effects of ACTH, DCA and Cortisone Compared with Congestive Heart Failure. Proc. National Meeting Am. Federation Clin. Research, Atlantic City, N. J., May 1,1951. Conn, J. W.: Electrolyte Composition of Sweat. Arch. Int. Med. 83: 416,1949. Berger, E. Y., Quinn, G. P. and Homer, M. A.: Effect of Desoxycorticosterone on the Colon: Its Relation to the Action of Cation Exchange Resins in Man. Proe. Soc. Exper. BioI. & Med. 76: 601,1951. White, A. G., Gonion, H. and Leiter, L.: Studies in Edema. H. The Effect of Congestive Heart Failure on Saliva Electrolyte Concentrations. J. Clin. Investigation, 29: 1445, 1950. Eisenmenger, W ..1., Blondheim, S. H., Bongiovanni, A. M. and Kunkel, H. G.: Electrolyte Studies on Patients with Cirrhosis of the Liver. J. Clin. Investigation, 29: 1491, 1950. Clark. G. W.: Studies in the Mineral Metabolism of Adult Man. Univ. Cal.

846 69. 70. 71. 72. 73. 74. 75. 76. 77. 78.

79. 80. 81. 82. 83.

Eugene Y. Berger, J. Murray Steele

Pub. Physiol. 5: 195, 1928; quoted by Shol, A. T.: Mineral Metabolism. New York, Reinhold Publ. Corp., 1939, p. 333. Irwin, L., Berger, E. Y., Rosenberg, B. and Jackenthal, R.: The Effect of a Cation Exchange Resin on Electrolyte Balance and Its Use in the Edematous State ..T. Clin. Investigation, 28: 1403, 1949. Bauman, W. C. and Eichhorn, J.: Fundamental Properties of a Synthetic Cation Exchange Resin. J. Am. Chem. Soc. 69: 2830, 1947. Kunin, R.: Ion Exchange. Anal. Chem. 21: 87, 1949. Berliner, R. W.: Renal Excretion of Water, Sodium, Chloride, Potassium, Calcium, and Magnesium. Am. J. Med. 9: 541,1950. Blake, W. D., Wegria, R. P., Keating, R. P. and Ward, H. P.: Effect of Increased Venous Pressure on Renal Function. Am. J. Physiol. 157: 1, 1949. Farber, S. J., Alexander, J. D. and Eichna, L. W.: Renal Hemodynamics and Salt and Water Excretion during Induced Congestion of the Inferior Vena Cava of Man. J. Clin. Investigation, 30: 638, 1951. Selkurt, E. E.: Effect of Pulse Pressure and Mean Arterial Pressure Modification on Renal Hemodynamics and Electrolyte and Water Excretion. Circulation, 4: 541, 1951. Berger. E. Y., Galdston, M. and Horowitz, S. A.: The Effect of Anoxic Anoxia on the Human Kidney. J. Clin. Investigation, 28: 648,1949. Welt, L. G. and Orloff, J. : The Effects of an Increase in Plasma Volume on the Metabolism and Excretion of Water and Electrolytes by Normal Subjects. J. Clin. Investigation, 30: 751, 1951. Warren, J. V. and Stead, E. A., Jr.: Fluid Dynamics in Chronic Congestive Failure; An Interpretation of the Mechanisms Producing Edema, Increased Plasma Volume and Elevated Venous Pressure in Certain Patients with Prolonged Congestive Failure. Arch. Int. Med. 73: 138, 1944. Merrill, A. J.: Edema and Decreased Renal Blood Flow in Patients with Chronic Congestive Heart Failure: Evidence of "Forward Failure" as the Primary Cause of Edema. J. Clin. Investigation, 25: 389, 1946. Epstein, F. H., Lesser, G. T. and Berger, E. Y.: Renal Function in Decompensated Cirrhosis of the Liver. Proc. Soc. Exper. BioI. & Med. 75: 822, 1951. Daly, C. and Dill, D. B.: Salt Economy in Humid Heat. Am. J. Physiol. 118: 285, 1937. McCance, R. A.: The Effect of Salt Deficiency in Man on the Volume of the Extracellular Fluids, and on the Composition of Sweat, Saliva, Gastric Juice, and Cerebrospinal Fluid. J. Physiol. 92: 208, 1938. Conn, J. W.: The Mechanism of Acclimatization to Heat. Advances in Int. Med. 3: 373, 1949.