asimp Na 1950

7/27/2019 asimp Na 1950

1/13

ASYMPTOMATICHYPONATREMIA IN PULMONARYTUBERCULOSIS 1

BY E. A. H. SIMS,2 LOUIS G. WELT,3 JACK ORLOFF,4 AND JAMES W. NEEDHAM

(From the Department of Internal Medicine, Yale UniversitySchool of Medicine,Ne-w Haven, Conn.)

(Submitted for publicationJune 21, 1950;accepted,August 28, 1950)

INTRODUCTION

In the Goulstonian lecture of 1936 McCance (1)first clearly def ined thephysiologicaland chemicalchanges induced by the depletionof salt in normalman. He confirmed Gamble's (2 ) postulate ofthe jealouslyguarded constancy of the concentra-tion of total base in the serum. This was largelyeffectedby the elaborationof a urinevirtually freeof salt soon after th e institution of salt deprivation.More recently Platt (3 ) noted that the concen-tration of total base may fall very slightly in anormal person just beforemaximal tubular reab-sorp tio n ofsalt occurs. This renal defenseislacking in Addison's disease and in those casesof kidney diseasein which salt continuesto be ex-creted in the urine at a timewhen th e concentra-tion of sodium in th e serum has fallen belownormal.

Subsequent to McCance's classical description,Winkler and Crankshaw (4 ) describeda new syn-drome characterizedby renal salt-wasting andsubnormal concentrationsof sodium in the serumin patients with pulmonary disease. Of th e threecases documented, two had far-advanced pul-monary tuberculosisand one had carcinoma of thelung. Besidesexcreting salt under conditionswhich would normally be associatedwith maximalreabsorptionof sodium, these pat ients rapidlyeliminated intravenouslyadministeredsalt andwere unable to maintain the concentrationof so-

dium in the serum at a normal level. Adrenal1 Aided by a grant from the Fluid Research Funds of

the Yale UniversitySchool of Medicine,and by an insti-tutional grant from the National Heart Institute, U. S.Public Health Service.

2Present address: College of Medicine, University ofVermont, Burlington,Vt .

8 Work done in part during the tenure of a Postdoc-torate Fellowship,U. S. Public Health Service.

4Work done in part while a fellow of th e DazianFoundation for Medical Research and during the tenureof a Life Insurance Medical Research Fellowship. Pres-ent address: National Heart Institute, Bethesda, Md .

cortical extractwas without influenceon the per-sistent wastage of salt in one patient.

Thorn, Howard and Dayman (5) analyzed alarge number of cases with pulmonary tubercu-losis and found none with a subnormal concentra-tion of sodium in th e serum. However, tw o ofhis patients did continue to excrete sodium afterthree days on an intakelow in salt. They did notconfirm Winkler's findingsand attributedthe lowconcentrationof chloride in the serum in manyof his patientsto retention ofbicarbonate.' Westover, Stiven,and Garry (6) observed lowconcentrationsof sodium in the serum of 55 of 114patients with tuberculosis. In a smallnumber inwhom balanceswere calculatedtherewas evidencesuggestiveof wastage of salt. More recentlyKolmer and his associates(7) analyzed the chem-ical data of a group of tuberculoussubjectsstatis-tically. The concentrationof total base was some-what lower than normal. It is of interest that thisdepressionwa s better correlatedwith magnitudeof daily temperature variations than with theseverity of the disease as estimated by X-rayexaminations.

Sincenone of these studies clearly definedthechemicalanatomy or elucidatedthe mechanism ofthe syndrome of pulmonary salt-wasting, thisstudy wa s undertaken. -It wa s hoped that by in-vestigatingthe role of the kidneys, the adrenalcortices, as wellas the status of the cellular struc-

tures of the body,an explanationfo r the chronicdepressionof the effective osmotic pressure ofthe extracellular water might be found. Alsoof importance to this study was the developmentof a means of differentiating this conditionfromAddison's disease since both are encountered inpatients with tuberculosis.

EXPERIMENTAL MATERIAL AND PROCEDURE

Ten male patients with advanced tuberculous diseaseand hyponatremia were studied. Four of the ten wereNegroes. One of the patients had miliary tuberculosis

1545

Downloaded on September 28, 2013. The Journal of Clinical Investigation. More information at www.jci.org/articles/view/102396
http://www.jci.org/http://dx.doi.org/10.1172/JCI102396http://dx.doi.org/10.1172/JCI102396http://www.jci.org/

7/27/2019 asimp Na 1950

2/13

E. A. H. SIMS, LOUIS G. WELT, JACK ORLOFF, AND JAMES W. NEEDHAM

0

0z

0 0 Bo4J 4-.01.4 4-i~~~~~~~~~~~.

Ca~~~~~C

- a a ) -oa0. )d ..c I..C

CL c l a a a

0

+..* 4) m a:) m) m) uCl ) ~~~ ~ ~U) U) U) U ) 0

.-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~4 ."2otE ~~e to~ -*i ra ) auE 0'~E. O a-. e~i '3 '3 ' if) 130 ~ ~ ~ ~ ~ ~ ~ ~ ~ 0

0 ei if) if)}* ..iiI SS4 if 0% - '34 _ _ C a ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ W -^ S nI 'C 0% 0 ' C

. ' % ) 0

zo-

in

.azSe U) U) U) U) Ce - C

.tDemX~~X N %m

00~~~~~

C . , ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ 1

.0to 0 0 0 0 0 0 0 0 0

< X I Z'.+ ei 00 0 a ei 0 00 if i H1u C)

i U) U ) U U0Po 0 0 0 00 C) Ouf) 00 0 0 0 0 0 0o

w o o o o2ouooo o o o o oo0oa-C 0 0 c D Qf) C)' i C) Q QC0 0% 0

t~~~~~~~~~~~6m- 0 :z:1 o 0 to o t- o- o w o- o o, , o to to ol ' o Oe

_4 _ 4 ___ 4 _ f ) _4 if00

e 4

C) 0 0o to U t 00

7/27/2019 asimp Na 1950

3/13

ASYMPTOMATIC HYPONATREMIA IN PULMONARY TUBERCULOSIS

with pulmonary infiltration limited to one lobe. On eNegro male with tuberculosis of th e adrenals and Ad -dison's disease was also investigated during a crisis.Normal concentrationsof sodium in th e serum werefound in 14 other patients with moderate or advanced tu-berculosis, in three with advanced bronchiectasis,and infive with various conditionsassociated with malnutri-tion. In addition, th e hospital records of a large numberof patients with abnormally low concentrationsof so-dium in the serum and continued excretionof sodiumwere searched for ins tances in which the hyponatremiawas not definitely secondary to kidney disease, adrenalinsufficiency, or over-hydration during therapy. Threesuch instances were found, but there was no t sufficientevidence to includethem definitely in th e present group.The course, follow up, and subsidiarydiagnoses of theten tuberculous patients are indicated in Table I. Someor all of the following investigativeprocedures were car-ried ou t on each patient depending upon th e clinical

conditionand severityof the illness.Subjects studied over an extended period were pro-vided with a basic diet adequate in calories and protein,but containingnot over 0.5 gm. of sodium chloride perday. This was supplemented in some instances by milkand cream containing0.3 gm. of salt. To this, weighedquantities of table salt were added daily. Consecutive24-hour collectionsof urine were made during the entireperiod of study and were analyzed fo r chloride and so-dium. Five of the patients were treated with 1-2 gm.of streptomycin and one with 10 gm. of para-aminosalicylicacid per day as well, but no essential change in theirregime was made during the period of investigation.

Renal clearancesof endogenous creatininewere meas-

ured in seven patients. These patients were not cathe-terized, but a moderate diuresis was induced to minimizeth e error from residual urine. In one patient th e clear-ances of mannitol and of urea were determined.

In five patients acute balances of sodium and ofchloride were measured on one or more occasions overperiods of 24 hours at th e beginning of which an in-fusioncontaining15 gm. of sodium chlorideas a 39%solu-tion was given. Additional salt was given by vein orper os in all but one patient.

Five or 20 mgm. of desoxycorticosteroneacetate inoil 5 were administered to four patients in on e or twointramuscular doses daily. Te n cc . of aqueous adrenalcortical extract (Wilson) or 1 cc. of Lipo-adrenal ex-tract (Upjohn) were given in intramuscular doses at in-tervals of 12 hours.

Both procedures of the "water test" fo r adrenal insuffi-ciency were carried ou t as describedby Robinson, Power,and Kepler (8). The decrease in circulatingeosinophiles

following th e infus ion of1.5 mgm. of epinephrinein nor-mal saline given over a periodof one hour was studiedin three patients according to the technique of Recant,Forsham, and Thorn (9). In the remaining patients0.3 mgm. of epinephrine was given subcutaneously asdescribed by Thorn and his col leag ues (10 ). Totaleosinophilecounts were made four hours after the epi-nephrine was given,and in th e majority of pat ientsat twohours as well. Urine for determination of 17-ketosteroidswas preserved with 1 cc. of concentrated hydrochloricacid per liter and refrigerated until analyzed.

5 Kindlyfurnished by Dr. Kenneth Thompson of Or-ganon,

Inc., Orange, N. J.

TABLE II

Effectof variationsin the intakeof sodium chloride;analysesof serum and urine

Free intake of sodium chloride Maximum intakeof Minimum intake of sodium chloridesodium chloride

Subject Serum Urinary excretionFia

_ _ _ _ _

-_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ -Intake

Total srmAverage Total Final Foinalime sodium intake time sodium excretionNa Cl HCOa Sodium C h l o r i d e s o i m o d u exrtn

mEq./L mEq.IL mEq./L mEq./L mEq./day mEq./L mEq./day mEq./day days mEq./L mEq. Iday days mEq./L mEq. /dayI 126 88.8 26.4 40 64 220 12 134.0 8 7 128.0 8II 127 90.9 25.8 85 72 130 110 85 2 134.0 43 3 125.0 29III 124 80.8 32.6* 44 67 108 166 260t 6 132.0IV 131 87.5 28.3 92 93V 128 93.4 25.0 35 114 97 3 134.0VI 122 95.0 26 20 43 23 51 13 1 118.0 12

2 112.0 41VII 123 90.0 25.8t 103 170 268 11 135.0 13 3 132.0 11

6 126.0 5VIII 128 95.2 15 51 14 47 185 8 135.0 13 4 126.0 27IX 127 89.2 18.0 345 198 230 5 124.0 8 3 119.0 6.7X 129 92.8 33 88 44 117 97 13 133.0

* On e day previouslyt Desoxycorticosterone acetate in oil 5 mgm. per da y given over this period

Two dayspreviously 24.8on admission seven days previously As chloride

1547.


7/27/2019 asimp Na 1950

4/13

E. A. H. SIMS, LOUIS G. WELT, JACK ORLOFF, AN D JAMES W. N EED H A M

TABLE II I

Alterationsassociatedwith administrationof largeamounts of NaCL

Concentrations of Na in extracellularwaterSubject Sodium rSetaind Wt . Plasma a

Initial Final Anticipated* volume

mEq. % mEq./L mEq./L mEq./L Kgm. % mEq.I-A 427 93 133 134 145 1.5 18 - 87I-B 427 20 137 137 139 -0.4 1 - 42II 256 61 128 .136 133 7 - 92VI 856 98.5 114 135 137 35 -132VII 427 58 129 137 136 0.2 8 - 22X 324 41 131 136 134 -0.3 - 45

EV t 739 95.3 129 136 147 - 27 +104

* Estimated from positivebalance of Na an d assuming no change in total body water.t Patient with adrenal cortical insufficiencyin crisis.

Th e urinary excretionof sod ium and chlorideas in-fluencedby the gradual reductionof dietary salt wasmeasured.

METHODS

Serum proteinfractions were determined by the Milnemodificationof th e techniqueof Majoor (11), urea bythe gasometric method of Van Slykeusingurease or hy-pobromite (12), and mannitol by Elkinton'smodification-of the Smith, Finkelstein, and Smith procedure (13).Creatininewas determined by the method of Peters (14).Readings were made exactly 20 minutes after the ad-dition of alkaline picrate, since filtrates from serumdarken progressivelyon standing. 17-Ketosteroidsinurinewere estimatedcolorimetricallyeither by the method

of Cahen and Salter (15) or of Holtorffand Koch (16).Easinophileswere stained by th e techniqueof Randolph(17), using phloxineB in a 1: 1 aqueous dilution ofpropyleneglycol. Th e average of from two to fourcounts in a Levy chamber 0.2 mm. in depthwa s taken.Other methods used in this laboratory have been re-

TABLE IV

Effectof variationsof the concentrationof sodium in theserum on renalckarances in patientswith

asymptomatic hyponatremia

Renal clearance*

Subject Wt . Serums o d iu m

EndogenousMannitol UreacreatinineKgm. mEq./L cc./min.

I 44 134 111, 108126 101, 102133 69, 66127 82, 76

IV 43 131 61VI 54 112 73VII 45 135 67, 69

126 64,69ViII 54 .1-3 62,62

128 73, 75X 49 125 80, 73

* Calculationsforconsecutiveclearanceperiodsare given.

ported in previous publica tions (18). Th e methods fo rcalculating changes in plasma volume from changes inrelative re d cell volume and hemoglobin and for changesin t he dis tributi on o fwater and electrolytes have beendescribedelsewhere (19, 20).

RESULTS

The analytical and derived data are presentedin Tables I-VI and Figures 1-4.

I. General Clinical ObservationsAll the members of this group of patients in

whom hyponatremia was found were chronically

ill and malnourished, and four (II, III, VI, an dVII) were acutelyill when first studied. Sevenof the patientsdied within on e to five months fol-lowing admission,an d th e remaining three have apoor prognosis for ultimate recovery. The essen-tial clinical data on thesepatientsare summarized

TABLE V

Robinson-Power-Keplerwater tests in threeptientswith asymptomatichyponatremia

Procedure I Procedure II

Urine volume

Subject Date SedriumNiiht Maximum valueperiod day.(9 hrs.) vr o

mEq.IL cc. cc.I 11- 8-48 126.0 400 240 40

11-27-48 127.0 650 205 3012- 7-48 129.7 330 195 39

VII 3-17-49 135.0 920 355 10*X 8-30-49 130.0 615 265 43

* The intakeof sodium chloridewas 15.8gm. for sevendays prior to the test.

1548


7/27/2019 asimp Na 1950

5/13


TABLE VI

The effectof administrationof epinephrineon circulatingcosinophilesin six patients with asymp-

tomatichyponatremia

Circulating eosinophiles

Epineph-Subject rine % fall

admin. Initialcount

2 hrs. 4 hrs.

mgm. per cu. mm. % NI (11-18) 1.5 300 56 38

(12-4) 1.5 300 31 33III

7/27/2019 asimp Na 1950

6/13

E. A. H. SIMS,LOUIS G. WELT, JACK ORLOFF, AND JAMES W. NEEDHAM

nitrogenin the blood was normal in al l patientsandin none was there gross evidenceof renal orcardiovasculardisease, with the exceptionof one(I), who had calcification of th e abdominal aortaby X-ray, and another who gave a history of hy-

pertensionof unknown degree nine years previ-ously. Only one patient (III) wa s observed tohave significantcarbon dioxide retention, in spiteof the severity of th e pulmonary involvement inthe majorityof th e other patients. Concentrationsof potassium in th e serum were within normallimits. When these patientswere eat ing thewarddiet with a free intake of salt, which wa s usuallythe casewhen they were first studied,they contin-ued to excrete from 40 to 90 mEq. of sodium perday in spite of the fact that the concentrationsofsodium in the serum ranged from 122 to 131mEq./L (Table II).

In two patients th e tendency toward hypona-tremia was found to be reversible. It is shown inFigure 2 that when this patient (VII) was againgivena diet low in sodium 75 days after the first

periodof deprivation,he maintained his concen-tration of sodium in the serum between 138 and136 mEq./L. During th e interval his weight hadincreased2 Kgm. on a regime of bed rest andstreptomycin,his sputum production had de-

creased by a third, and his daily temperature riseshad decreased from an average of 102 to 100 de-grees F. However, the X-ray of his chest wasunchanged. Nine months after his first period ofdeprivationof salt, a second patient (I ) was placedon a diet for seven days which wa s as low in so-dium as could be arranged at th e sanitorium towhich he had been sent. Over this periodhis con-centrationof sodium in the serum remained con-stant at 133 to 134 mEq./L. During this intervala pneumoperitoneum had been induced and he hadgainedconsiderableweight.

II . Th e effects of variationin salt intakeFive patients (I, III, VII, VIII, and IX) were

given a moderate or high intake of salt for a pe-riod of five to 12 days. In one who sufferedfrom

CHLORIDE'

UE /0AY 0 o _I 0 le it to ts so 3 40 4 o 59 o t 75 0 es 900 e lo, #O * liO*,

FIG.2. CHANGESIN BLOODPRESSURE, SERUM CONCENTRATIONS OF SODIUM, AND URINARY EXCRETION OFSODIUMAND CHLORIDEIN RELATIONTO VARIATIONSIN INTAKE OF SALT AN DTO ADMINISTRATION OF DESOXY-CORTICOSTERONEACETATEAN D OF ADRENAL CORTICALEXTRACT IN A PATIENT (VII) WITH ASYMPTOMATICHYPONATREMIA

1550s


7/27/2019 asimp Na 1950

7/13


24

iIW

010

'ZIa10

a11

10

w

IUVWA _ VILVvs q / _ 5 mo/d _

..... . ..............I a . .. . .. .a .

5 l0 15 20 0 5 20TIME IN DAYS

FIG.3. EFFECT OF DESOXYCORTICOSTERONEACETATE UPON THE URINARYEXCRETIONOF SODIUMAND CHLORIDEIN FOURPATIENTS WITH ASYMPTOMATICHYPONATREMIA

advanced tuberculosisand carcinoma of the phar-ynx treated with irradiation, daily infusionscon-taining 13.5 gm. of sodium chlorideover a five-day period failed to raise the concentration of so-dium in the serum above 125 mEq./L. Therewas no evidence of congestivefailure and noperipheraledema in this case. Three of the pa-tients given more than 185 mM. of salt per dayhad

final concentrationsof sodium in the serum inthe low normal range, 132-135 mEq./L. In twoother patients (I I and X) final concentrations ofsodium in the serum of 133-135 mEq./L wereobserved followinga somewhat lower intake ofsalt.

The sodium in the diet of five patients (I, VI,VII, VIII, an d IX ) was reduced to 8-13 mEq./day for two days or more. Since the quantityofsalt added to the basic diet ha d been variouslyre -stricted during the preceding periods,this did no t

constitute an abrupt deprivationof salt. In allbut one instancethe renalexcretionof sodium wa sreduced to 8 mEq./day or. less, and in all instancesthe concentrationof sodium in the serum fell be-fore the minimal urinaryexcretionof this ion wasreached. This is illustrated in Figure 1, days 2-8and Figure 2, days 40-42. A fifth subject (II)who was acutely ill with tuberculouspneumonia

had an excretionof 29 mEq./day of sodium and aconcentrationin the serum of 125 mEq./L whenthe intakeof salt was reduced to 43 mEq./day.However, during these studies of deprivationtheexcretionof sodium, while often sharplyreduced,did not reach th e extremelylow valuesusuallyas-sociated with hyponatremia due to depletionofsalt in other conditions. It was strikingto observethroughout the studiesthat thesepatientswere freeof symptoms attributableto depletionof salt andsometimes had an indifferenceto their dietarysalt

1551


7/27/2019 asimp Na 1950

8/13


20

4-

lS

6

Sn

a

1:z;

z

L.B. A- 004T KG 45 45 445SERUMNA 136 135 136 132 135 135SERUMCL 92 97 96 95

.i0.| l l L A ill

.

M a C 13759WT KG 493 495 49.6SERUMNA 131 127 133 13 2 3SERUMCL 93

993 9 3

DC_

l o

hpoAreo

"M1. C 2960 s W.W.9 93703S E R U M A 32 132 129 132 133

SERU CL 96 9X 93 9f 91

DO

soe .... z1e 0 ^ .. ._ . _ .

Time in Dole

FIG.4. EFFECT OF ADRENALCORTICALEXTRACTUPON THE URINARYExCRE-TION OF SODIUM AND CHLORIDEIN FOUR PATIENTS WITH ASYMPTOMATICHYPONATREMIA

when the concentrationsof sodium in the serumwere at levels generallyassociatedwith collapseinother situations involvingdepletionof salt. Onlyone patient (X ) exhibited craving for salt whenon a sodium restricted regime. Muscle crampswere not seen.

One patient (VI) demonstrated that this groupis not immune to the effects of true depletionofsalt. This patient was acutely ill, febrile, andsweating considerably. It was later learnedthathe had rejected a considerable portion of the 85mEq. of sodium that were added to hi s diet for athree-day period. He excreted only 12 mEq. ofsodium the first day that hi s intake of this ionwas reduced to 13 mEq. During the followingdaya bout of mild diarrhea added to the loss of sodiumin sweat and sputum, and the concentration of

sodium in his serum fell from 118 to 112 mEq./L.At this pointhe developed diaphoresis,acute anx-iety, and borderline vascularcollapse,al l of whichwere rapidly improved by an infusionof hyper-tonic saline. It is notable that these signs andsymptoms developed when the concentration ofsodium in the serum was only 6 mEq. below thatat which the renal excretion of sodium had beensharplyreduced. In contrast to the crisis in Ad -dison's disease, the non-protein nitrogen concen-tration of the blood did not rise above 33 mgm. %o,nor th e potassium above 5.1 mEq./L. Therewas no elevationof temperature.

When there were alterations in the quantityofsalt ingestedas described above, there were varia-tions in the concentrations of sodium in the se -rum of 2. to 12. mEq./L. An additionalpatient

1552

25

20

IN

O0

2c

2 4 6 a 10 12 14 Z 4 6 a so 12 14


7/27/2019 asimp Na 1950

9/13


(X) whose intake of salt wa s not pushed to ahigh level had a variationof only 3.5 mEq./L inhis concentrationof sodium in th e serum. Theminimum concentrationsof sodium in Table IIoccurred before orapproximately at the time whenthe excretionof sodium had first reached a lowlevel, so they reflect a loss of sodium via the urine.

Infusionsof 256 mM. of sodium chlor ide asa3% solution were administeredto five patients.In addition, al l but one receivedadditionalsaltin the course of a 24-hour periodeither as normalsaline intravenouslyor as table salt by mouth(Table III). Fo r comparison, data ar e also pre-sented from a patient with Addison's disease(EV) in crisis during the first 18 hours of therapy.A ll th epatientsexcept VI and EV had free access

to water during the balanceperiod. When the ini-tial concentrationof sodium was low, from 41 %to 98.5%oof the administered salt was retained(althoughin this calculation no correctionwa smade for the variable loss of sodium in sweat).There were no significantshifts of sodium betweenthe intracellular and extracellular compartments.

It is assumed that where an antecedent deple-tion of both water and sodium exist, water will beretained followingthe administrationof salt andwill be manifestby increasesin body weight andplasma volume. In addition, the observed con-centrationof sodium in th e extracellularwater atthe end of the balanceperiodshould be less thanthe concentrationof sodium as estimatedfromthe positive balanceof this ion with the assump-tion tha t there had been no increasein total bodywater. This "anticipated"concentrationof so-dium in the extracellular water is presentedinTable III.

Th e insignificantchanges in weightand in theper cent increasein plasma volume and the goodagreement between th e observed and anticipatedconcentrationsof sodium in the extracellular fluidsuggest that in patients I-B, II, VII, and X thevolume of water was not previouslycontracted.In contrast, in patients I-A and VI, who hadsuffered antecedentextrarenal losses of salt andwater, there were retentionsof salt and increasesin plasma volume similar to those of th e patient(EV) with Addison's disease. Moreover, therewa s considerabledisagreementbetween th e ob-served and anticipatedconcentrationsof sodium inthe extracellularwater in patients I- A and EV.

Th e agreement of this functionin patient VI maybe explainedin part by a marked diaphoresisthatoccurred early in th e balanceperiod.

Endogenous creatinineclearancesranged from61 to 80 cc./min. (Table IV). Figures for theheightsof these patients ar e not available,but inview of their low weights,averaging 49 Kgm., itis probable that correctedclearanceswould fallwithinthe low normal range. In two of the pa-tients (VII and VIII) creatinine clearancesweremeasured when the concentrationof sodium inth e serum had been forced to a maximum by ahigh intake of salt and again when the subjecthad been deprived of salt. Under these varyingconditionsthe creatinineclearanceeither did notchange (VII) (Figure 2, days 13 and 43) or was

actuallyhigherin the hyponatremic state (VIII).In an additional patient(I) mannitol clearanceswere similarly measured both when the concen-tration of sodium in the serum was low and whenit was normal. These did not differ significantly.Urea clearanceswere somewhat higher at thelower concentrationof sodium in the serum inthis same patient.

III. The response to adrenal steroidsIn two patients (I and VII) 20 mgm. of

desoxycorticosteroneacetate (DOCA) in oil ad-ministeredfor four days produced a sharp re-ductionin sodium excretion, followedby an in-creasedexcretionon discontinuingthe injections(Figure 3). In patient I a gain in weight of 2.6Kgm. and the development of pitting edema of th eextremities wa s noted during the period whenDOCA was administered,while patient VIIgained 2.1 Kgm. over the same period. Althoughan exact balance cannot be computed from thedata .available, both patients appeared to retainmore water than could be accounted for on thebasis of the quantity of sodium retained. Fivemgm. of DOCA per day produced a similar butless marked reductionin the excretionof sodiumin patients I, VII, and X which wa s also followedby a secondary increase in excretionabove thebasal level after cessationof th e injections. Th eeffect of injectionsat 12-hour intervals of 10 cc.of adrenal cortical extract (Wilson) or of anequivalentamount of lipo-adrenal extract (Up-john) was studied in four patients. The resultsare shown in Figure 4. There was clearlyno effect

1553


7/27/2019 asimp Na 1950

10/13


when lipo-adrenalextractwa s given to patientI orwhen aqueous extractwas given to patientVIII. Areductionin the excret ionof sodium occurred dur-ing a portion of the period in which aqueous ex-tract was administered to patient VI I on tw o oc-

casions. However, th e reduction preceded theinjection on th e first occasionand was not sus-tained during the entire period of administrationon either occasion. In the caseof pat ientX th eexcretionof sodium varied spontaneously duringa number of weeks both beforeand after adrenalextract wa s given and reached a low value oneand three days prior to the injections. There wasno secondary increasein the excretionof sodiumabove the basal level on discontinuingthe injec-tions in any of the patients.

IV. Est imateof adrenalcortical functionAt one time or another during their course in

the hospitala diagnosisof adrenal insufficcywas seriouslyentertainedin four of the patients(I, III, VII, and X) on the basis of clinical orlaboratoryfindings. It wa s of interest, therefore,to study their response to the current tests ofadrenocorticalfunction. The response to depriva-tion of salt has been describedabove.

In three of the patients (I, VII, and X) theRobinson-Power-Kepler water test was done.Th e results are givenin Table V. Procedure I ofthe test wa s invariablypositivein that th e normaldiuretic response to ingestedwater was not ob -tained. In onlyone patient (VII) was ProcedureII positive, with an "A" value of 10. This test,however, was done when the patient had taken15.5 gm. of salt in his diet the precedingday,which is in excess of the "usual diet without salt"specifiedby the authorsof the test. In on e pa-tient (I) th e "A " value was not significantlychanged when the test was repeatedafter he hadbeen given 23 gm. of sodium chloride two daysprior to the test.

Th e eosinopeniceffect of epinephrinewas stud-ied in six patients. In al l of these a decreaseincirculating eosinophilesof 50%o or greater wasfound on at least on e test. The depressionofeosinophileswas, however, apparentlynot as sus-stained as in normal subjectsgiventhese quanti-ties of epinephrine,for when counts were made att wo hours as well as at four hours, in four out

of five instancesthe two-hour count wa s lower.Thus, patient X had no eosinopenic response atfour hours in the initial test; a second test demon-strated a 70% fall in eosinophilesat two hours.Th e initial eosinophilecount in one acutely ill

patient (III) wa s less than 3 per cu.mm., so thatepinephrine was not given.

Th e excretionof 17-ketosteroidsper 24 hourswa s low, 3.3 mgm., in one patient (VII) on a dayshortly after his admission with fever and ad-vanced tuberculosis. Five otherpatients had low-normal or normal values, the actual excretionap-pearingto be correlatedwith the state of nutri-tion of the patient.

A five-hourintravenousglucose tolerancetestwas normal in three patients tested (VII, VIII,

and X).Six patients came to autopsy (II, III, IV, V,IX, and partial in VI) and al l were found to havenormal adrenals. Two patients (I I an d III.) hadrenal lesions as described in Table II.

DISCUSSION

The ability to excrete an almost sodium-freeurine, to retain a large amount of intravenouslyadministeredsodium without benefitof exogenoussalt-retaininghormone, to respond with an eosino-peniato epinephrine,and the absence of gross orhistologicevidenceof adrenalcortical disease atpost-mortem al l militate against adrenal corticalhypofunctionas the primary defect responsiblefor this clinical picture, althoughthis cannot becompletelyexcluded. Nor does this syndromeappear to be a primary abnormalityof the kidneys.Examination of the urine in thesepatientsusuallyrevealed no abnormalities,and there were noevidencesof renal insufficiency. The renal tu -bular cells responded by increasingthe reabsorp-tion of sodium when DOCAwa s administeredindoses of 20 and 5 mgm. per day.This abnormalityseems not to be a specificderivative of tuberculosissince it has been re -portedin one case of non-tuberculous pulmonarydisease(4). Moreover, althoughthe observationsare not completelyconvincingat this time, thereare suggestionsthat it may be found among pa-tients with non-pulmonary disease. The commondenominator may not be an organ specific dis-order, but a secondaryeffect in some instances

1554


7/27/2019 asimp Na 1950

11/13


of severe disease associated with debility andmalnutrition. There are reports in the literaturethat lend some support to such a suggestion.Sunderman and Rose (21) studieda patient with

an inadequate dietary intakewho had a subnormalconcentrationof total base in the serum. Gollan(22) reported that the average concentrationofsodium in the sera of marasmic Italian childrenwa s lower than normal. This must be interpretedwith some caution, as these infants had a historyof diarrhealdisease.

The fact that this syndrome has none of thesymptoms of true sodium depletionis most im-portant. In th e development of the usual typesof salt depletionthere is a negativewater balancewith a consequent contractionof the volume ofbody water as well as a reductionin it s tonicity.It is probable that the contractionof the plasmavolume which sharesin the decreaseof the extra-cellular water and the dilution of th e cellular con-tents as a result of th e hypotonicityof the extra-cellular water both contributeto the physiologicabnormalitiesnoted in salt depletion. These ab-normalitiesare characterizedby a persistenthypo-tension, asthenia, disorderedrenal hemodynamics(as manifested by a depressed rate of glomerularfiltration and urea clearance)and by nitrogen re-tention; and, finally, peripheralvascularcollapse.Th e data from the present investigationsdo notsuggest such a state. Although endogenous cre-atinine clearancesmay have been somewhat re-duced , theywere no lower when the concentra-tion of the sodium in the serum wa s depressed.There wa s no nitrogen retention, and peripheralvascularcollapsedeveloped only when there hadbeen extrarenalloss of salt and water leadingtothe development of true salt depletion. These pa-tients do not appear to be suffering from a con-

traction of the volume of the extracellular fluidand plasma.A hypothesisis offered to explain the mecha-

nism of this syndrome which is admittedly specu-lative and unproven. It is that the primary de-fect in this abnormality is a chronic reduction ofcellular osmolarity. The overall effects of this al-terationwould be an abnormal excretion of sodiumin an effort to reduce the tonic ity ofthe extracel-lular f luids in conformity with the primarily re-duced tonicity of the intracellular fluids, but with

the maintenance of normal volumes of body wa-ter. Although there is as yet no primary and in-dependent evidence that such a state of chronicallyaltered cellular osmolaritydoes occur, acute al-

terationsin cellularosmolarityhave been observed.This phenomenon has been suggested by severalgroups to explain certain resul ts intheir investiga-tions (23-27).

The sequence of events that might followsucha primary decrease in cellular tonicity is as fol-lows: the first event is a movement of water fromthe cells to the extracellularspace in the interestsof osmotic uniformity. At this point al l the bodyfluids are relativelyhypertonicand th e intracellularvolume is contracted. The hypertonicityand cel-lular dehydration provoke thirst and stimulateanincreasedsecretionof th e posterior pituitaryanti-diuretic hormone. More water is ingested andthere is an increasein the reabsorptionof waterin the distal renal tubule. This tends to expandthe body fluids, but reduces the tonicity of thesefluids toward the "new" normal level. The cellu-lar dehydration,the hypertonicityof the bodyfluids and the expanded extracellular volumemay all be stimuli for t hedecreased reabsorptionof sodium in the renal tubules, thus increasingthe excretionof th is ion. As sodium is excreted,water followsuntil the "new" level of tonicity andoriginal volumes of th e body fluids are achieved.

This mechanism providesfo r the maintenanceof the volume of extracellular fluid and plasma,and , hence,avoids the undesirableeffects of theusual type of salt depletion. With this mecha-nism in mind, it is clear why the calculationof theanticipatedrise in concentrationof sodium in theextracellular water agrees so closely with the ob-served when there is a net positive balance ofsodium and the calctilation is made with the prem-

ise that there is no change in total body water.Although this hypothesis seems plausible andis not at variance with the observed data, it isadmittedly unproven an d awaits a critical test.

The differentiationof this syndrome from Ad-dison's disease is an important clinical problem.The tw o procedures that seem to offer the mosthelp in this regard are the demonstration of aneosinopeniafollowingadministration of epineph-rine and the excretion of an almost sodium-freeurine on a regime of minimal salt intake. The

1555


7/27/2019 asimp Na 1950

12/13

E. A. H. SIMS. LOUIS G. WELT, JACK ORLOFF, AND JAMES W. NEEDHAM

latter procedure is hazardous, but the risk can beminimized by careful and frequent clinical andchemical observations.

SUMMARY AND CONCLUSIONS

Tenlpatients with hyponatremia and pulmonarytuberculosiswere studied. Nine had far-advancedpulmonary tuberculosisand one miliary tubercu-losis with minimal pulmiionaryinvolvement. Nofrank evidence of renal or adrenal insufficiencywa s found. Al l showed advanced malnutrition,hypotension,and hypoalbuminem-iia.Abnormalretentioni of carbon dioxide wa s found in onlyone patient. When given free accessto salt, theywere found to excrete40-90 miiEq.of sodium perday in spite of concenitrationsin the serum rang-in g from 122 to 131 mnEq./L.Five patients ex-creted as little as 5-29 mEq. of sodium per daywhen the dietarysalt wa s severelyrestricted. Theexcretionof sodiumn,however, did not reach thelow values usually associatedwith hyponatremiadue to depletionof salt in other conditions. Thesepatients did not p)resentth e clinical signs orsymptoms or other evidence of sodium depletion.There was no evidentdehydration,peripheralcir-culatory failure, or hyperkalemia. Renal clear-ances of creatinine, mannitol, or urea were essen-

tially normal and were not lower when measuredin the hyponatremic as opposed to the normalstate. Five patients retained 41%oto 98% of alarge amount of salt given in a 24-hour period.Calculationsfrom data obtained during acute bal-ance periodsfollowingthese infusions supportedthe hypothesisthat these patients did not have acontractedbody water. The concentrationof so-dium in the serum remained in the subnormal orlow normal range despite high daily intakes ofsalt. This syndrome was reversedin two patients

associatedwith clinical improvement. There wa smore lability in the con-centrationof sodium inth e serum with varying intakesof salt than is usu-ally observed in normal subjects. The concentra-tions varied during relatively short periods overa range of 2 to 12 mnEq./L.

In three patients procedure I of the Robinson-Kepler-Power water test w-as positive. In onlyone was the secondIpart also positive. Six patientshad an essentially niormiialeosinopenicresponse toepinelhrine.

Four patients retained increasedquantities ofsodium when given desoxycorticosterone.Therewa s no definite response to smalldoses of adrenalcortical extract. The adrenal glands were normal

in the six patients whocame to

autopsy.It is suggested that a primary reduction in cel-lular osmolaritymay be th e initiating factor in thedevelopment of this syndrome. The increasedex-cretion of sodium at subnormal concentrationsofthis ion in the serum miightrepresentan attemptto reduce extracellular tonicity to conform witha reduced cellular osmolarity.

An eosinopenicresponse to epinephrine and theability to reduce th e excretionof sodium on aregimefree of salt aids in differentiating thissyndrome from Addison'sdisease.

ACKNOWLEDGM ENTS

We are grateful to Miss Alice Cattanaclhand the staffof th e Diet Department fo r making possiblethe diets lowin salt used in this study and to Miss Patricia Robinson,R.N., fo r her invaluablecooperation. The authors areindebted to Dr. John P. Peters for his guidance an(dcriticism throughout the course of this study.

BIBLIOGRAPHY

1. McCance, R. A., Medical problems in mineral metab-olism. Lancet,1936,1; 643, 704,765, and 823.

2. Gamble, J. L., Ross, S. G., and Tisdall, F. F. , Themetabolism of fixed base during fasting. J. Biol.Chem., 1923, 57, 633.

3. Black, D. A. K., Th e renal excretion of water andsalt. Irish J. M. Sc., 1949,715.

4. Winkler, A. W., and Crankshaw, 0. F. , Chloridede-pletion in conditionsother than Addison's disease.J. Clin. Invest., 1938,17, 1.

5. Thorn, G. W., Howard, R. P., and Dayman, H., Elec-trolyte changes in pulmonary tuberculosis, withspecial referenceto adrenalcortical function. Bull.Johns Hopkins Hosp.,1940,67, 345.

6. Westwater, J. O., Stiven, D., and Garry, R. C. , A

note on th e serumsodium level in

patientssuffering

from tuberculosis. Clin. Sci., 1939,4, 73.7. Kolmer, H. S. , Ellis, D., Smith,T., Collins, P., and

Greisheimer, E. M. , Acid-base conditionin tuber-culosis. Am. Rev. Tuberc., 1948,57, 400.

8. Robinson, F. J. , Power, M. H., and Kepler, E. J.,Two new proceduresto assist in th e recognitionand exclusion of Addison'sdisease; preliminaryreport. Proc. Staff Meet., Mayo Clinic, 1941,16,577.

9. Recant, L., Forsham, P. H., and Thorn, G. W., Theeffect of epinephrineon circulating eosinophils.Federation Proc., 1948,7, 99.

1556


7/27/2019 asimp Na 1950

13/13

1557SYM I'TOMATIC HYPONATREMIA IN PULMONARY TUBERCULOSIS

10. Thorn, G. W., Bayles, T. B., Massell,B. F., Forsham,P. H. , Hill, S. R. , Jr., Smith, S., III, and Warren,J. E., Studies on the relation of pituitary-adrenalfunction to rheumatic disease. New England J.Med., 1949, 241, 529.

11. Milne, J. , Serum protein fractionation: a comparisonof sodium sulfate precipitationand electrophoresis.J. Biol. Chem., 1947, 169, 595.

12. Peters, J. P., and Van Slyke, D. D., QuantitativeClinical Chemistry. Vol. II. Methods. Williams& Wilkins, Baltimore, 1932.

13. Elkinton, J. R. , The volume of distribution of man-nitol as a measure of the volume of extracellularfluid, with a study of the mannitol method. J.Clin. Invest., 1947,26 , 1088.

14. Peters, J. H. , The determination of creatinine andcreatine in blood and urine with the photoelectriccolorimeter. J. Biol. Chem., 1942, 146, 179.

15. Cahen, R. L., and Salter, W. T., Urinary 17-keto-steroids in metabolism; standardized chemical esti-mation. J. Biol. Chem., 1944, 152, 489.

16. Holtorff, A. F., and Koch, F. C., The colorimetricestimation of 17-ketosteroidsand their applicationto urine extracts. J. Biol. Chem., 1940,135, 377.

17. Randolph, T. G., Blood studies in allergy. I. The di-rect counting chamber determination of eosinophilsby propylene glycol aqueous stains. J. Allergy,1944,15, 89.

18. Needham, J. W., The metabolism of potassium indiabetic acidosis. Yale J. Biol. & Med., 1949,22 ,39.

19. Elkinton,J. R., Danowski, T. S., and Winkler, A. W.,Hemodynamic changes in salt depletionand in de-hydration. J. Clin. Invest., 1946,25, 120.

20. Elkinton, J. R., and Winkler, A. W., Transfers of in-tracellular potassium in experimental dehydration.J. Clin. Invest., 1944, 23, 93.

21. Sunderman, F. W., and Rose, E., Studies in serumelectrolytes. XVI. Changes in th e serum and bodyfluids in anorexia nervosa. J. Clin. Endocrinol.,1948,8, 209.

22. Gollan, F., Blood and extracellular fluid studies inchronic malnutrition in infancy. J. Clin. Invest.,1948,27, 352.

23. Yannet, H., and Darrow, D. C., The effects of de-pletion of extracellular electrolytes on the chemi-cal composition of skeletal muscle, liver, and car-diac muscle. J. Biol. Chem., 1940, 134, 721.

24. Elkinton, J. R., Winkler, A. W., and Danowski, T. S.,Inactivecell base and th e measurement of changes

in cell water. Yale J. Biol. & Med., 1944, 17,383.

25. Seldin, D. W., and Tarail, R., Effect of hypertonicsolutions on metabolism and excretionof e lect ro -lytes. Am. J. Physiol., 1949, 159, 160.

26. Gaudino, M., and Levitt, M. F., Influence of theadrenalcortex on body water distributionand renalfunction. J. Clin. Invest., 1949,28, 1487.

27. Welt, L. G., Orloff, J. , Kydd, D. M., and Oltman ,J. E., An example of cellular hyperosmolarity.J. Clin. Invest., 1950,29, 935.


Documents

asimp Na 1950