434

WATER AND ELECTROLYTE BALANCE INDISEASE

By JAMES CONWAY, M.B., J. LEE, M.B., M.R.C.P., and W. O. SYKES, M.Sc.Charing Cross Hospital Medical School

Many medical and surgical emergencies arecomplicated by derangement of water and electro-lyte balance, the successful management of whichdepends upon the rapid assessment of essentialclinical and laboratory data, and the prompt in-stitution of a certain pre-arranged routine oftreatment. An attempt is made in this review toillustrate the application of physiological principlesto specific clinical problems.The major part (35 1.) of the water content of

the body is an integral part of the cell structure,the remainder (15 1.) is extracellular fluid compris-ing intersitial fluid and plasma. The compositionof the fluid in the two compartments is different.Potassium and phosphate are the main cation andanion respectively of the cells, while sodium andchloride preponderate in the extracellular fluid.The difference in ionic composition between thetwo compartments is well maintained in health,the distribution of water between the cells and thefluid bathing them being regulated by the osmoticpressure of solutes. An increased concentration ofsubstances on one side of the membrane will drawwater from the other.The extracellular fluid acts as a buffer between

the cells and the external environment with whichit is in constant contact through the lungs, skin,gut and kidney. It suffers greater proportionalchanges in volume than the cellular fluid.The concentration of solutes in body fluids is

regulated by receptors in the hypothalamus sen-sitive to certain changes in the osmotic pressure ofthe plasma. The reabsorption of water by therenal tubules is then controlled by the productionof the antidiuretic hormone of the posteriorpituitary.'The mechanism whereby the volume of body

fluids, particularly the extracellular fluid, is regu-lated, is not known. A mechanism analogous tothat controlling osmotic ptessure has been postu-lated,2 whereby a receptor sensitive to changes inblood volume or flow stimulates the production ofa hormone, possibly of the adrenal cortex, which,

by adjusting the amount of salt reabsorbed, willregulate blood volume and indirectly extracellularfluid volume.

Disturbances of water and electrolyte balancewill be considered under two headings: (a) Waterand salt balance; (b) potassium balance.

Water and Salt BalanceDisturbances of electrolyte balance affect

primarily the volume of the extracellular fluids,while excess or loss of water will be distributeduniformly throughout the body. When the dis-turbances persist, or become more severe, cellularfunction is affected and further changes are, inconsequence, superimposed on pre-existing abnor-malities.

Marriott3 stressed the importance of evaluatingseparately losses of water and salt, and on Table ithe signs and symptoms of pure water and puresalt are presented.

Losses of water and salt do not occur in-dependently, but against the background of thisseparation of the two distinct processes, deviationsfrom the normal fluid and electrolyte balance indisease may be divided artificially into five clinicalstates: (i) Primary loss of sodium; (2) primaryexcess of sodium; (3) primary excess of saline;(4) primary retention of water; and (5) primaryloss of water.

Primary Loss of SodiumLoss of electrolyte from the extracellular fluid

involves chiefly sodium and chloride ions. Re-tention of CO2 allows easy replacement of lostanions (C1-) by bicarbonate ions (HCO3-) withoutill effect. The cations (Na+) however can only bereplaced from the exterior, therefore only a smallloss can be tolerated. Loss of sodium will tend toreduce the osmotic pressure of the extracellularfluid, with the result that water will pass into thecells and the kidneys will excrete a dilute urine.Thus the first compensatory mechanism is tomaintain the osmotic pressure of body fluids at the

Protected by copyright.

on February 14, 2020 by guest.

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.27.311.434 on 1 S

eptember 1951. D

ownloaded from

September 1951 CONWAY, ET AL.: Water and Electrolyte Balance in Disease 435

expense of extracellular fluid volume. The plasma,being one-third of this fluid, will suffer a simul-taneous reduction in volume.The degree of deficiency of sodium cannot be

gauged with accuracy from the plasma sodium, forthe essential feature of this condition is the re-duction of extracellular volume, which may bemeasured indirectly by the reduction in plasmavolume. This diminution in circulatory fluid isreflected in an increase in packed cell volume andconcentration of plasma proteins. A full andvivid description of the effects of sodium deficiencyis given by McCance in the Gouldstonian Lecturesof I936.4Sodium deficiency occurs classically in Addison's

disease where the ability of the renal tubules toreabsorb sodium is impaired and even in thepresence of a deficiency the kidney continues toexcrete sodium.When considerable quantities of fluid are lost

from the gut in persistent vomiting, intestinalsuction, pancreatic or biliary fistulae and severediarrhoea, the loss of sodium becomes important.Excessive sweating will also carry away largequantities of sodium. When allowed free accessto water these cases will develop sodium de-ficiency.The peripheral circulatory failure which is an

essential feature of diabetic acidosis results fromthe continued loss of sodium in the urine. In renaldisease, when tubular exceeds glomerular damage,the ability of the kidneys to conserve sodium isimpaired, and a condition clinically similar toAddison's disease may appear. These patientstherefore cannot withstand a dietary deficiency ofsodium, and treatment of hypertension with a lowsodium diet may thus be dangerous.

Primary Retention of SodiumThis state, the reverse of that just described,

must lead by the retention of water to the ex-pansion of extracellular fluid volume at the expenseof ingested water and intracellular fluid. Thecondition arises when hypertonic sodium solutionsare ingested and access to water is restricted. Thiswill occur on the ingestion of sea water by the ship-wrecked, or after infusions of normal saline (.9 percent.). The fluid balance of a patient whose re-quirements are met solely by the administration ofnormal saline may be given as follows:-

Daily fluid requirement, 3 pt. or I,500 ccof saline.IntakeWater .. .. .. ,500 cc.NaCI .. ..... . 3.. 5 gm.

LossLungs ... .cc.ICnevitableSkin .. .. 400 cc.

Total .. ,000 cc. Lost as purewater.

Thus I3.5 gm. of NaCI remains to be excretedby the kidneys with 500 cc. water, i.e. 2.7 per cent.NaCI. The kidneys are unable to accomplish this,and the resulting sodium retention will draw waterfrom the cells causing intracellular dehydration.The administration of larger quantities of saline

to cover inevitable losses and renal requirementshas its own danger. When' isotonic saline is givenit does not stimulate a prompt diuresis as doeswater ; it is excreted over a period of days. Muchof the saline is retained expanding the extracellularfluid and giving rise to dangers of heart failure andpulmonary oedema.

TABLE I

Comparison of Effects of Water and Salt DepletionManifestation Pure Water Depletion Pure Salt Depletion

Dehydration + ++ primary or simple - ++ secondary or extracellularThirst + + AbsentLassitude + + + +Orthostatic fainting Absent till late ++ +Urine volume Scanty Normal till lateNaCl in urine Often + Always absent except in Addison's diseaseVomiting Absent May be + + +Cramps Absent May be +++Plasma NaCI Slight increase or normalBlood urea + + ++-Haemoconcentration Not till late and slight -++Blood viscosity Normal till late Increased +++Blood pressure Normal till late Fall + + +Water absorption Rapid SlowMode of death ? due to rise of osmotic pressure Peripheral circulatory failure

Reproduced by kind permission of Dr. H. L. Marriott and the Editor of the British Medical Journal

Protected by copyright.

on February 14, 2020 by guest.

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.27.311.434 on 1 S

eptember 1951. D

ownloaded from

436 POSTGRADUATE MEDICAL JOURNAL September 1951

PRIMARY PRIMARY PRIMARY PRIMARY PRIMARYNORMAL LOSS OF SODIUM SCESSOF SALINE EXCESS OF WATER LOSSOFW^

BDY FLUIDS HYPOTONIC HYERTNI ISOTONIC HYPTONIC HYPERTONC

INTRACELUJR INCREASD DECREASED NORMAL INCREASED DECREASED

EXTRACELLULAR E DECREASED INCREASED INCREASED INCREASED DECREASEDCAUSES- CAUSES CAUSES- CAUSS CAUSES ,

ADDSONS DISEASEINFUSIONS OF OEDEMA WATER VER

VOMITING WITH NORMAL SALINE ) NEPHRITIS INTOXICATION DEPRIVATIONDIARRHOEA FREEINTESTINAL ACESS DRINKING SEA b) CONGESTIVE EXCESSNE ANURIASUCTION TO WATER HEART FAIURE INFUSIONS UNTREATED

FISTULAE WAR OF GLUCOSESWEATING ANURIA TREATEDCHRONIC NEPHRITIS WITH FORCEDDIABETC ACIDOSIS FLUIDS.

Normal saline therefore should never be the solesource of fluid to any patient unless there is ademonstrable deficiency of sodium. This appliesparticularly to patients in the first week afteroperation in whom the ability to excrete sodiumchloride has been shown to be greatly reduced.5If it is necessary after an operation to maintain thepatient by intravenous fluids, the salt contentshould be strictly related to the actual daily loss bythe kidneys.Primary Retention of Isotonic FluidThe retention of isotonic fluid must be regarded

as the cause of oedema and therefore separatedfrom primary retention of either sodium or water.In Type 2 nephritis the continued loss of plasmaprotein leads to retention of salt and water in thebody. The cause of oedema in congestive heartfailure is difficult to explain. Neither the rise invenous pressure nor the increased permeability ofcapillary membranes due to anoxia can explain thepresence of the oedema. Studies of renal functionin heart failure show that there is a reduction in therenal blood flow and in the glomerular filtrationrate; the effect of these is to increase abnormallythe reabsorption of sodium by the tubules.6'7

It is important to realize that sodium is retainedwith water in the oedema; patients are 'brine-

logged' and not waterlogged.8 The success of therigid restriction of sodium (0.5 gm./day) in heartfailure thus becomes apparent. Water restrictionin oedema will produce an increase in the con-centration of sodium in extracellular fluid and awithdrawal of water from the cells. This cellulardehydration is responsible for the paradox ofsevere and distressing thirst existing in a grosslyoedematous patient. The efficiency of mercurialdiuretics in heart failure has been shown to be dueto their ability to increase the excretion of sodium.9Ion-exchange resins are being developed which,by their ability to absorb sodium ions, can preventtheir absorption from the gut even after the normalamount is taken in the diet.10

Primary Retention of WaterIn normal circumstances the kidneys are capable

of rapid elimination of large quantities of ingestedwater, and the danger of water intoxication appearsonly as a result of artificial or diseased states whichprevent the normal prompt diuresis. In practicethis state occurs when renal function has beengrossly depressed. As water accumulates withinthe body it will be distributed throughout allfluid compartments, increasing their volume butdecreasing their osmotic pressure. The signs andsymptoms of an excess of water appear to be due

Protected by copyright.

on February 14, 2020 by guest.

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.27.311.434 on 1 S

eptember 1951. D

ownloaded from

September I951 CONWAY, ET AL.: Water and Electrolyte Balance in Disease 437

to changes within the cells. There is restlessness,delirium, coma, irregular muscle twitching, crampsand sometimes convulsions. If water requirementis over estimated and excessive infusions of iso-tonic glucose given, or liberal amounts are takenorally, a condition analagous to water intoxicationmay occur.

Primary Loss of WaterLoss of water from the body will lead to a pro-

gressive increase in the osmotic pressure of allcompartments of body fluid. The cellular loss offluid is an important cause of the changes inpotassium balance which will be considered in thenext section. Throughout a period of deprivationthe inevitable loss daily of about I,ooo cc. of purewater will cortinue, aggravating the water deficit.The kidneys, in an attempt to conserve water,

will concentrate the urine maximally to a specificgravity of 1032, and will pass a steady minimalamount of urine-about 500 cc.-until renal bloodflow, and hence the glomerular filtrate, is reduced,with ensuing pre-renal uraemia.To reduce the body requirement of water two

principles of management apply-to eliminate theintake of electrolyte and to reduce the productionof urea to a minimum by giving a diet of adequatecalorific value but no protein.

These principles are directly applicable to thetreatment of cases of anuria. A programme oftreatment must be prepared and be ready for usewhenever a case of anuria appears; it may bemodelled on those of Muirheadl' and Bull.12

It is important to realize that when the kidneysare producing no urine the body is unable to loseelectrolyte. It is essential therefore that no electro-lytes whatsoever should be given. The water re-quired to cover daily inevitable losses (I,ooo cc.)should be given with glucose. The practice ofstimulating diuresis by giving infusions of electro-lytes, such as sodium sulphate, can only cause adangerous expansion of the extracellular fluid.

TreatmentThe procedure outlined here applies to adults

and excludes the special difficulties of parenteraltherapy to infants. In every case the followingquestions must be examined separately:-

(I) How much water must be given?(2) How much salt must be given?(3) At what rate should the fluid be given?(4) Should correction be made for disturbances

of acid-base balance?

How Much Water Must be Given?(a) Replace current losses. These are likely to be

of the order of 2,500 cc. per day, allowing an in,evitable loss of I,ooo cc. through the skin and

lungs and a renal loss of I,500 cc. To this mustbe added the volume of any abnormal loss of fluidfrom diarrhoea, vomiting, fistulae, etc.

(b) Estimate and replace deficit. To establish thequantity of water lost the following estimationsmay be made:--(i) Loss of body weight. Changesin the day to day estimation of body weight can betaken to reflect variations in body water. There-fore, neglecting the loss of weight due to an in-adequate diet, a loss of i lb. body weight will beequivalent to a deficit of 500 cc. of water. (2)Packed cell volume, haemoglobin. In the absence ofblood loss the red blood cells may be used toindicate variations in plasma volume. (3) Plasmaprotein level may similarly be used to detecthaemoconcentration and the reduction in plasmavolume. The change in the volume of each of thesewill be proportional to the diminution of circu-latory. fluid, and may therefore be used to give anapproximate estimation of the deficit thus:-Hbi Plasma Pr, Extracellular Fluid

Hb2 Plasma Pr2 Normal Vol. (15L)Normal Value,

Abnormal Value2(4) The specific gravity of the plasma is easily

measured"3 and offers a ready method of deter-mining water deficit. If the normal is taken asI.027 then each o.oI may be assessed as a deficitof 200 cc. of water.The value of these tests is limited for it is

necessary to assume an absolute figure of nor-mality and neglect the range. They are furtherlimited in longstanding cases and those with com-plications, particularly of bleeding, where altera-tions in red cells and plasma proteins occursimultaneously. No single investigation is re-liable in assessing a patient's condition. In aparticular case all possible evidence must begathered together and interpreted against the back-ground of thorough clinical examination. Theestimation of the blood urea and alkali reserve areuseful guides to the severity of the condition.

How Much Salt Must be Given?In all cases of dehydration the most important

question to decide is whether the estimated waterdeficit must be replaced as normal saline or wateralone. In practice this is extremely difficult be-cause no direct measurement of the deficiency ofsodium can be made. Estimation of serum sodiumconcentration cannot be used to indicate thequantity of sodium lost, and the chemical estima-tion takes too long to be of immediate use inassessing a case. A deficit of sodium can be in-

Protected by copyright.

on February 14, 2020 by guest.

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.27.311.434 on 1 S

eptember 1951. D

ownloaded from

438 POSTGRADUATE MEDICAL JOURNAL September 1951

ferred yhen the renal excretion of chlorides isreduced. If a good volume of urine is passedan excretion rate below 3 gm./l. (Fantus test)indicates a deficiency. However the limitations ofthis test must be realized. In Addison's diseaseand in renal failure excretion of chloride con-tinues in spite of the deficiency. When sodiumions are lost in excess of chloride ions (e.g. indiarrhoea and in intestinal fistula) to maintainacid-base balance the kidney will excrete chlorideions in the presence of a plasma deficiency ofsodium. Conversely, after surgical operations theability of the kidney to excrete salt is greatly im-paired, and a reduced urinary chloride does notindicate deficiency.5

Estimation of the amount of salt to be givenmust depend upon the losses indicated by thehistory and from clinical observation of signs andsymptoms (Table i). The reduction of systolicblood pressure to a figure between go and Ioomm. Hg. suggests a moderate loss of sodium,equivalent to 4 to 6 1. of isotonic saline. If sodiumdeficiency is present, fluid replacement should bemade initially with normal saline intravenously,and the amount required should be decided bythe response of the patient. All urine passedshould be collected and the choride contentestimated. When this approaches 3 to 5 gm./l.and the clinical condition has improved, isotonicsaline should be replaced by half isotonic saline andglucose to make up the estimated deficit of fluid.

Rate of AdministrationIt is advisable to divide the day into eight

hourly. periods, before each of which clinicalassessment of the patient's state is made andtreatment ordered accordingly, a careful record ofwhich is essential.

First period. Give one-third of the daily fluidrequirement (800 cc.) to which add one-half of theestimated deficit. In a moderately severe case,with a deficit of 6 1., the fluid prescribed for theinitial period should be about 4 1. Give i 1. inthe first hour and thereafter 500 cc./bour, usingnormal saline in the presence of sodium deficiency,otherwise half normal saline and glucose.

Second period. Reassess the condition of thepatient, taking particular note of the state ofthe circulation (venous P., B.P., pulse, respira-tion rate, urinary excretion (volume and chlo-.rides), haemoconcentration (Plasma S.G., Hb. andP.C.V.). The disproportion between the deficitof salt and water should have been corrected duringthe first period. The volume of fluid to be givenwill consist of one-third of the daily requirementand one-half of the initial deficit of water. In allbut the most severe cases give this orally with

glucose or food, otherwise intravenously as glucosesaline.

Third period. The clinical state of the patientshould be assessed and if a deficit still exists theamount should be added to the normal require-ment (800 cc.) for that period.Correction for Acid-Base Balance

Control of the reaction of the blood is achievedrapidly by the lungs and kidneys and significantdisturbance either of acidosis or alkalosis cannotoccur unless such loss of water and electrolyte hasoccurred as to render the kidneys incapable ofcorrecting the abnormalities. Therefore it is ofthe utmost importance to restore any salt andwater lost. Isotonic saline should be given to casesof alkalosis or acidosis except those in the mostsevere acidosis, when one-sixth molar sodiumlactate should be given at the ratio of 0.5 1. to every15 vols. per cent.. depression in the alkali reservebelow 60 vols. per cent. The correction of severesodium deficiency by normal saline alone canproduce significant acidosis and therefore i 1. in 5of saline should be replaced by sodium lactate,

Potassium Balance

Recognition has only been given during the lastfew years to the fact that there are clinically im-portant conditions besides familial periodicparalysis and Addison's disease in which severechronic disturbances of potassium distribution andbalance may occur. Balance studies, supplementedin some instances by muscle analyses, performedon man and experimental animals, has revealed aloss of cellular potassium in numerous conditions,including diabetic coma, diarrhoea, vomiting,water deprivation, dietary potassium deficiency,alkalosis, shock, adreno-cortical tumours andnephritis.A normal diet contains 2 to 4 g. of potassium

per diem, the ion being widely distributed, inconcentrations of the order of 0.3 per cent. inanimal and vegetable foods. Normally the kid-neys play the major part in the elimination of thision, and are not effective in conserving it when thebody is deficient in potassium.A typical concentration of potassium in intra-

cellular water is i50 m.Eq./l. (600 mg./Ioo cc.)while the interstitial fluid contains a concentra-tion close to that in plasma, about 5 m.Eq./l.(20 mg./Ioo cc.). Most of the body potassium(order of I50 g.), therefore, resides in the tissues;about half is in muscle and only about 1/50 in theextracellular fluid. Work with isotopic potassiumhas shown that there is an equilibrium between theintra- and extracellular potassium, but the precisemechanism by which potassium is maintained at

Protected by copyright.

on February 14, 2020 by guest.

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.27.311.434 on 1 S

eptember 1951. D

ownloaded from

September 1951 CONWAY, ET AL.: Water and Electrolyte Balance in Disease 439

its relatively high concentration in cells remains tobe elucidated.

The Detection and Effects of Disturbances ofPotassium Distribution and BalanceA deficiency of potassium is here understood to

involve such a depletion of the body content asobviously could not occur without loss of cellpotassium.

If balance studies reveal a retention of potassiumwhile the patient is under treatment, it may be con-cluded that a deficiency previously existed. Suchstudies are too elaborate and protracted forordinary clinical practice and in any case onlyreveal a deficit in retrospect.The published work has shown that a deficiency

is reflected in loss of potassium from muscle, andin clinical research the analysis of a muscle biopsyhas been used to provide evidence.

Potassium lost from the cells may be replacedto a variable degree by sodium. 4,'1 The availableevidence suggests that cell potassium loss alonedoes not produce symptoms, but replacement bysodium may do so.1'415 In clinical medicine thelevel of potassium in the extracellularfluid appearsto be the most important factor in the productionof symptoms and signs. Fluctuations in theextracellular potassium concentration are in mostcases secondary to changes in the cell. Inabnormal states the extracellular or serum levelof potassium may be high, normal or low, de-pending on a combination of factors:-

(I) Movement of potassium in or out of cells.In a large number of conditions potassium hasbeen shown to leave the cells for the extracellularfluid. Cellular dehydration is common to most ofthese conditions. The withdrawal of water, orthe impaired nutrition of the cells appears to beresponsible for failure to maintain the normalintracellular potassium concentration. However,dehydration is not the only factor concerned, sincepotassium also leaves the cells in anuria and acutenephritis. Potassium will move back into the cellswhen the abnormal condition affecting the cell isremoved. This usually takes place early in thetreatment of an acute illness, when the clinicalsigns of a reduction in potassium level must beexpected.A variety of natural and synthetic substances,

including insulin and adrenaline, will produceacute effects on the distribution of potassium in thebody, resulting in small reductions in the serumpotassium concentration. Data in this field arefragmentary and the mechanisms by which theactions are exerted not understood.

(2) The state of renal function. If there isimpairment of renal function any potassium lost

from the cells may accumulate in the extracellularfluid and reach a toxic level. When renal functionis normal, potassium liberated from cells is pro-gressively excreted ; and alleviation of the causalcondition will result in withdrawal of potassiumby the cell from the extracellular fluid, in whichit may then fall to a dangerous level.

(3) Dietary potassium. A normal meal containsabout 2 g. of potassium and when this amount isgiven as a single dose to a normal fasting personthe serum potassium will rise approximately 0.75m.Eq./l. (3 mg./Ioo cc.) within three hours.16It is, however, generally agreed that large amountsof potassium can be given by mouth without toxicsymptoms developing, provided kidney function isnormal.The intestinal fluid contains 5 m.Eq./l. (20

mg./Ioo cc.) of potassium and therefore anychronic diarrhoea and vomiting will lead to adeficiency of potassium even in the absence ofdehydration.

Rats on a potassium-deficient diet lose potas-sium from the muscle cell with partial replacementby sodium.

Clinical DiscussionSince it is the extracellular concentration of

potassium that determines the onset of symptoms,it is important to know this level, and there-fore methods of detecting its fluctuations are dis-cussed. The flame photometer, necessary equip-ment for the most rapid method of determination,is not yet widely available, and the chemicalmethods are time consuming. With large devia-tions from the normal serum potassium levels theE.C.G. shows characteristic changes and providesan indirect method for the detection of grossabnormalities. The method would seem to bespecially advantageous for serial observations.17At low levels the QT interval is prolonged, theRST segment depressed and the T wave lowered,flattened and prolonged. At high levels the PRinterval is prolonged, the P wave may disappear,the QRS complex is prolonged, the T wave in-creases in amplitude and becomes peaked andeventually there is intraventricular block. Probablythe single most important change in the E.C.G.with high levels of potassium is the alteration inamplitude of the T wave with peaking.18 There isindividual variation in the E.C.G. response tothe serum potassium level and small departuresfrom the normal are not detectable by thismethod.19,20 Early signs of a fall in serumpotassium which may appear are an alteration ofpersonality, consisting of irritability together withapathy.21 There may be asthenia and diminutionof reflexes and tone. Eventually there may beextreme muscular weakness or virtual paralysis,

Protected by copyright.

on February 14, 2020 by guest.

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.27.311.434 on 1 S

eptember 1951. D

ownloaded from

440 POSTGRADUATE MEDICAL JOURNAL September 1951

bradycardia and shallow respiration which leads tocyanosis.

There is, at present, no easy way of detectingactual or incipient potassium deficiency. Inclinical work interest is naturally focused on theextracellular concentration of potassium, and inthe absence of a flame photometer the E.C.G. isvaluable. In practice it is necessary to assume apotassium deficiency to exist when the necessaryconditions for the production of the abnormalityare present, and treatment should be instituted onthe lines described in the concluding section.

Diabetic Coma

Atchley et al.22 demonstrated a loss of bodypotassium in the development of diabetic coma.They allowed two cases of diabetes to relapse intocoma and estimated by balance studies the deficitsof water, sodium, potassium and chloride. Theloss of water and electrolytes increased as acidosisdeveloped. In one case a typical daily negativepotassium balance was go m.Eq. (3.5 g.) and atypical negative sodium balance was 50 m.Eq.(I.I g.). During recovery there was retention ofwater and electrolytes. They indicated that theinitial polyuria is an important cause#f electrolyteloss.

Danowski23 studied eight cases of diabetic comaand found that all had lost potassium. Oraladministration of potassium salt led to retention ofthe cation in each case. In only one case wasthere a definite paralysis and even in this there wasrecovery with oral potassium.There is a significant loss of potassium during

the development of diabetic acidosis, and this losswill be facilitated if, during therapy, excess fluids aregiven24 or any substance that will promote diuresis.

Conditions Involving Loss of Intestinal JuicesIntestinal juices contain approximately the same

concentration of potassium as the extracellularfluid and the large volume lost in diarrhoea,vomiting and intestinal fistulae will lead to a lossof potassium. This loss will be seriously aggra-vated by the concurrent dehydration.This has been shown by Tobler,26 who analysed

the muscles of babies who had died from diarrhoeaand vomiting, and found a deficit Of 30 per cent. inintracellular potassium. Darrow26 reported a de-ficit of 40 per cent. in potassium on a fat-free solidbasis in the same circumstance. These findingsare in accord with Darrow's26 balance studies. Itis of interest to note that in the muscle compositionstudy referred to above, Darrow found only in afew cases replacement of the lost potassium bysodium, such as occurs in experimental animals

given excess DOCA or on a potassium-deficientdiet.

In fatty diarrhoea, calcium is one of the manyelectrolytes lost. The loss of both potassium andcalcium ions may account for the relative rarity oftetany in steatorrhoea in spite of a low blood cal-cium in most cases. Engel27 has shown that insteatorrhoea, when the serum levels of both ionsare low, the administratibn of potassium mayprecipitate tetany.Renal Disease

(a) Chronic nephritis. In this condition anegative potassium balance may result from theinability of the kidneys to conserve base.

(b) Acute nephritis and anuria. There ispotassium loss from the cells and accumulation inthe extracellular fluid which may lead to potassiumintoxication. During the diuresis which accom-panies recovery from anuria, there is a rapid ex-cretion of potassium in addition to other electro-lytes, and this is likely to produce a considerableloss of potassium.Haemorrhage, Burns, Shock

In all these conditions potassium is lost from thecells. In burs the destruction of tissue willcontribute to the release of intracellular potassiumand so add to the general loss of body potassium.In a severe case the coincident failure of renalfunction may lead to a potassium intoxication.

TreatmentIf conditions which produced an intracellular

potassium deficiency are allowed to continue, thecells will not take up potassium. Only by re-moving the underlying cause, for example, indehydration by replacing fluid and electrolyte willpotassium administration restore the normal intra-cellular concentration. The relatively low con-centration of potassium normally present in extra-cellular fluid cannot be exceeded with safety, andtherefore potassium should only be given whenkidney function is normal so that if none is re-tained by the cells the cation can be excreted.

Oral administration is a therapeutically efficientmethod of administration23 and is safer than theintravenous or intramuscular method. If it is im-practicable to feed a diet rich in potassium, forexample milk, fruit juices or meat broth enrichedwith potassium chloride, it may be possible toadminister a solution by stomach tube.

If oral administration is impracticable as, forexample, in pyloric stenosis, one may have re-course to subcutaneous injection. Darrow recom-mends for this 'K-lactate,' containing 4.0 g.sodium chloride, 2.7 g. potassium chloride and

Protected by copyright.

on February 14, 2020 by guest.

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.27.311.434 on 1 S

eptember 1951. D

ownloaded from

September 1951 CONWAY, ET AL.: Water and Electrolyte Balance in Disease 441

52 cc. of molar sodium lactate (15.3 g.) per litre.This solution should be administered slowly,80 cc./kg. over more than four hours, and pre-ferably in eight hours. If other solutions of potas-sium are thought to be preferable, then theamount of potassium given may range from 30 toI5o,m.Eq. (I.2 to 6 g./day) in an adult, and themaximum concentration of potassium in the solu-tion should be 80 m.Eq. (3.2 g./l).28 Intravenousinjection of potassium salts carries considerablerisk of potassium intoxication.The amount of potassium given in cases of

suspected potassium deficiency should be suchthat if none entered the cell the serum level wouldnot be dangerously elevated; 0.25 g. of potassiumchloride per kg. body weight per diem, orally, is asafe dose. Such a dose will not result in potassiumintoxication even if there is no deficiency.There is as yet no generally approved method

of treatment in cases of acute high serumpotassium, but intravenous calcium and glu-cose29,30,31 have been used. It is possible that acombination of intravenous glucose and sub-cutaneous insulin might be more effective.32Orally administered ion-exchange resins mayprove to be of value in cases of longstanding eleva-tion of serum potassium such as occurs in anuria,where the normal physiological mechanism for theelimination of potassium is inoperative.

ConclusionIn patients suffering from disturbances of water

and electrolyte metabolism attention should bepaid to the balance of both sodium and potassiumions, but laboratory tests remain inadequate topresent a complete and unequivocal picture of apatient's condition; the fundamental changes areobscured as a result of the operation of correctivemechanisms. The immediate changes arisingfrom disturbances of potassium metabolism are:-

(I) Accumulation of potassium in the extra-cellular fluid when renal function is reduced.

(2) Reduction of extracellular potassium when,on recovery, the cells take up potassium.Methods of therapy consist of (a) replacement of

water and such ions as are grossly deficient,thereby (b) facilitating the operation of the mostimportant corrective mechanism, that provided bythe kidneys. Where renal function is severely im-paired special difficulties arise, notably in thetreatment of a dangerously high level of serumpotassium. The well developed technique ofintravenous therapy, used when oral administra-tion is impracticable, is hazardous when used forgiving potassium salts.The value of regulating sodium and water

balances is established and to this may be addedthe importance of attention to potassium meta-bolism.

REFERENCES

'VERNEY, E. B. (1946), Lancet, ii,. 739, 78i.2BORST, J. G. G. (1948), Acta Med. Scand., Suppl. 207.3MARRIOTT, H. L. (1946), Brit. med. J., i, 245, 285, 328.'McCANCE, R. A. (1936), Gouldstonian Lectures, Lancet, i, 643,

704, 765, 823.'WILKINSON, A. W." BILLING, B. H., NAGY, G., and

STEWART, C. P. (1949), Lancet, i, 640.'MERRILL, A. J. (1946), J. Clin. Invest., 25, 389.'MERRILL, A. J., and CARGILL, W. H. (1947), J. Clin. Invest.,

26, 1190."SCHEMM, F. R. (1942), Ann. Intern. Med., 17, 952.9REID, H. A., and HUGHES, W. (1949), Lancet, i, 593."0DOCK, W., and FRANK, N. R. (I950), Amer. Heart J., 40, 638."MUIRHEAD, E. E., HALEY, A. E., HABERMAN, S., and

HILL, J. M. (1948), Blood, Special Issue No. 2, p. o01."BULL, G. M., JOEKES, A. M., and LOWE, K. G. (1949),

Lancet, ii, 229.3VAN SLYKE, D. D., HILLER, A., PHILLIPS, R. A., HAMIL-

TON, P. B., DOLE, V. P., ARCHIBALD, R. M., andEDER, H. A. (g95o), J. Biol. Chem., 183, 331.

"HEPPEL. L. A. (I939), Am. J. Phys., 127, 385."FERREBEE, J. W., PARKER, D., CARNES, W. H., GERITY,

M. K., ATCHLEY, D. W., and LOEB, R. F. (1941), Ibid.,135, 230.

"KEITH, N. M., and OSTERBERG, A. E. (1946), Proc. StaffMeeting Mayo Clinic, 21, 385.

7HOWARD, J. E., and CAREY, R. A. (I949), J. Clin. End., 9, 691."'THOMSON, W. A. R. (1939), Lancet, i, 808."TARAIL, R. (1948), Am. J. Med., 5, 828.'NADLER, C. S., BELLET, S., and LANNING, M. (1948),

Ibid., 5, 838."HAWKINS, C. F., HARDY, T. L., and SAMPSON, H. H.

(1951), Lancet, i, 3x8.2'ATCHLEY, D. W., LOEB, R. F., RICHARDS, JUN., D. W.,

BENEDICT, E. M., and DRISCOLL, M. E. (I933), J.Clin. Invest., 12, 297.

'"DANOWSKI, T. S., PETERS, J. H., RATHBUN, J. C.,QUASHNOCK, J. M., and GREENMAN, L. (1949), Ibid.,28, 1.

"HOLLER, J. W. (1946), J.A.M.A., 131, xi86.'"TOBLER, L. (I9xo), Arch. f. exper. Path. u Pharmakol., 73, 566."6DARROW, D. C. (1946), J. of Paed., 28, SI5."ENGEL, F. L., and MARTIN, S. P. (1948), 'Abstracts of

Southern Society for Clinical Research,' Am. J. Med., 4, 455."ELKINTON, J. R., and TARAIL, R. (I95o), Ibid., 9, 200."'FINCH, C. A., SAWYER, C. G., and FLYNN, J. M. (1946),

Ibid., i, 337.'°GOVAN, C. D., JUN., and DARROW, D. C. (x946),J. of Paed.,

28, 541."aFENN, W. 0. (1940), Phys. Rev., 20, 377."BYWATERS, E. G. (I944), J.A.M.A., 124, 1103.

Protected by copyright.

on February 14, 2020 by guest.

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.27.311.434 on 1 S

eptember 1951. D

ownloaded from