Treatment of the Low-Salt Syndrome inCongestive Heart Failure by the Controlled

Use of Mercurial DiureticsBy ALBERT L. RUBIN, M.D., AND WARREN S. BRAVEMAN, M\I.D.

The production of the low-salt syndrome in cardiac patients with refractory congestive heart failureis generally attributed either to fluid retention with electrolyte dilution or to excessive salt loss,induced by the frequent administration of mercurial diuretics. The former is often the important eti-ologic factor. In a study of a group of 25 refractory cardiac patients with normal plasma electrolytepatterns, a hyperchloremic acidosis was produced to restore a responsiveness to mercurial diuretics.During the mercurial-induced diuresis that followed, the urinary sodium concentration was sig-nificantlv lower than the plasma sodium concentration in all instances. From these observations, a

method of treatment of the low-salt syndrome, utilizing mercurial diuretics seemed feasible. Thisregimen was successfully carried out in two hyponatremic patients with return of plasma sodiumto normal and striking clinical improvement.

IN cardiac patients with fluid retention,two types of electrolyte imbalance havebeen associated with a transient refrac-

toriness to mercurial diuretics. One of these ishypochloremic alkalosis, and the other ishyponatremia in association with hypo-chloremia, the "low"-salt syndrome."The former is a well-defined clinical entity'

that sometimes occurs in the course of vigoroustreatment of the edematous cardiac patientwith salt restriction and protracted daily in-jections of a mercurial diuretic agent. Hypo-chloremic alkalosis may occur prior to theattainment of optimal body weight. Respon-siveness to mercurial diuretics is restored inthis situation when the hypochloremia iscorrected by the administration of ammoniumchloride.The latter electrolyte imbalance, the "low-

salt syndrome," can be present in a number of

From the Second (Cornell) Medical Division,Bellevue Hospital Center, and the Department ofMedicine, Cornell University Medical College,New York, N. Y.

This work was supported in part by grant H-2054from the United States Public Health Service, NewYork Heart Association, and Lederle LaboratoriesDivision of the American Cyanamid Company.

Presented as a Morning Lecture at the 28th Scien-tific Sessions of the American Heart Association,New Orleans, La., Oct. 23, 1955.A preliminary abstract of this paper appeared in

Circulation 12: 766, 1955.

655

disease states. These have been discussedrecently by Danowski and his associates2 whocall them the "low-salt syndromes" and classifythem according to total stores of extracellularsodium decreased, intact, and increased. Thepatients discussed in this article would fallinto that group demonstrating the "low-saltsyndrome" with increased sodium stores andcongestive heart failure.The "low-salt syndrome," as originally de-

scribed by Schroeder3 consisted of (1) drowsi-ness, weakness, and lethargy, (2) loss ofappetite, (3) nausea and vomiting and oc-casionally abdominal or muscular cramps.These symptoms occurred in a setting of (1)a successive depression of urinary volume, (2)decreased urinary excretion of chloride ion,not increasing after mercurial diuretic ad-ministration, (3) progressive gain in bodyweight, (4) rising nonprotein nitrogen contentof the blood, and (5) hyponatremia and hypo-chloremia. The author reported two mecha-nisms by which this syndrome could beproduced: by retention of water and dilutionof electrolytes, and by excessive depletion ofthe body salt, usually associated with the useof mercurial diuretics. The former mechanismis well recognized, having been seen manytimes in postoperative surgical patients givenparenteral glucose in water,4 and its significanceas al etiologic factor in the production

Circulation, Volume XIII, May, 1956

by guest on July 4, 2018http://circ.ahajournals.org/

Dow

nloaded from

TREATMENT OF LOW-SALT SYNDROTME IN CONGESTIVEi HEART FAILUREI

of the "low-salt syndrome" has receivedrecognition.' 6With regard to the latter mechanism, the

question may be raised whether the data estab-lish that mercurial diuretics contribute to thedevelopment of the clinical state described.In the original paper on the "low-salt syn-dome," only urinary chloride ion concentrationwas measured. Sodium excretion was notdetermined. Measuring urinary chloride con-centration alone gives a false conception ofsodium losses. A greater proportion of thechloride ion than of the sodium ion in theextracellular fluid is excreted in the urine afterthe administration of mercurials, apparentlybecause such agents act primarily on thechloride ion.7 All the subjects showed sig-nificantly positive daily water balance, and,in all cases but one, mercurial diuretics wereadministered very infrequently and inter-mittently. Only one of the cases reportedshowed a significant diuresis as judged byweight loss, after the administration of hyper-tonic saline solution to correct the electrolyteimbalance. Many did not lose as much weightafter therapy as they had gained during theirperiod of oliguria. For these reasons all thecases could represent examples of extracellularfluid expansion with electrolyte dilution, andno direct evidence is presented to iiicriminiatemercurial diuretics as etiologic agents. Tostate unequivocally that hyponatremia resultsfrom the administration of mercurial diuretics,it should be necessary to show that urinarysodium (concentrations in the diuresis followingdrug administration exceeds the plasma con-centration of sodium, in a setting where thefluid intake is restricted to approximate thepatient's insensible loss (e.g., 1 kilogram ofweight loss is equivalent to 1 L. of urine out-put). To our knowledge, this has unever beendemonstrated.

In our experience, most patients with re-fractory edema due to heart failure have normalplasma electrolyte concentrations.8 At thisstage in the natural history of their heartdisease, these patients are unable to handlewater as well as salt. This is not generallyappreciated, and while salt intake is rigidlyrestricted, fluid intake is not. The result is

fluid retention, further expansion of extra-cellular fluid volume, and consequentlyelectrolyte dilution.

Overhydration with resultant expansion ofthe extracellular fluid volume and dilution ofelectrolytes would appear to be the most im-portant etiologic factor in the production ofthis syndrome, which might better be termedthe "dilution syndrome."

In 25 refractory cardiac patients with normalplasma sodium and chloride concentrationsin whom a hyperchloremic acidosis was pro-duced by means of Diamox (acetazokamide)and ammonium chloride,8 a responsiveness tomercurial diuretics was restored. In all thesepatients the urinary sodium concentrationduring the mercurial-induced diuresis was

TABLE 1.-Plasma an(l ('rinaor Sodium anld Chlo-ride Concentrations and TWeight Loss in a 24-HourPeriod of Responsiveness to the Atdtministr ation of aiMercurial Diuretic in a Setting of Hyperch loretnicAcidosis. In Every Instance the Sodium (Concentrationin the Urine Is Significantlyt Lower Tharn the PlasmaConcentration

Plas-ma

Pt. Na-

E. G. 137140140

I. R. 131131

TL. B. 137137137

W. N. 136136136135138

E. B. 134135136136

T. J. 136135135

S. B. 142137

G. B. 137138

B. E. 137

Urine Plas-Na-, ma*a Cl-

mEq L

112.5 11899 12398 123115 11695 11545 12091 11677 11698 12081 11882 11790 12072 11682 11991 11957 11566 11590 11682 11168 11085 11793 10778 11887 116113 129

UrineCl- Wt. Loss

(Ibs.),/24 hours

158151150154175154151155165158162160167146161116140149136123133135129142132.8

7.94.06.510.55.510.06.05.011.05.57.08.03.010.04.55.04.510.03.54.57.55.05.04.56.0

656

by guest on July 4, 2018http://circ.ahajournals.org/

Dow

nloaded from

RUI3IN AND) BRAVEIMAN7

significantly lower than the plasma sodiumconcentration. Table 1 lists the plasma andurine sodium and chloride concentration andthe weight loss in pounds that occurred ill a24 hour period after the administration of 2 ml.of Mercuhydrin (meralluride) in the setting ofa hyperchloremic acidosis, on 40 separatepatient days. Note that the urine is alwayshypotoiiic for sodium with respect to theplasma, and that the concentration of chloridein the urine markedly exceeds that of sodium.

In these patients, plasma sodium concentra-tion remained within normal limits, probablyreflecting ion shifts between intracellular andextracellular fluid compartments. In patientswith hyponatremia, it was theorized that in asimilar situation, plasma sodium concentrationmight rise toward normal, if homeostaticmechanisms tending to regulate body sodiumwere functioning. For, if 1 L. of edema fluidis removed from the body with a sodiumconcentration less than that in the plasma,and is not replaced by fluid administration (i.e.,the patient loses 1 Kg. in weight), extracellularfluid sodium concentration should rise.

It appeared to us, developing the premisethat this syndrome represented a primary dilu-tion phenomenon, that the aim in therapyshould be the mobilization of excess fluid ill theexpanded extracellular fluid space, rather thanelevation of the extracellular fluid sodium ( on-tent. Accordingly, a group of five patients withheart disease, congestive failure, and refractoryfluid retentioni, who showed azotemia, oliguria,anorexia, amid nausea, and who had a markedhyponatremia and hypochloremia were treatedwith a regimen that had been successful inrestoring responsiveness to mercurial diureticsin cardiac patients with refractory fluid re-tention who had normal plasma electrolytepatterns. This regimen, which utilizes Diamoxand aqueous ammonium chloride, has pre-viously been described in detail.8 No attemptwas made to alter plasma sodium concentra-tions. In the setting of a hyponatremia, anelevated plasma chloride concentration,lowered blood pH, and lowered carbon dioxidecombining power, the daily administration ofMercuhydriml resulted in a marked and sus-tainied diuresis in all five cases, during the

course of which plasma sodium concentrationreturned to normal.Two cases from this group have been selected

for this report on the basis of the completenessof their data.

AMETHODS1. Each patient selected was carefully evaluated

to insure adequate digitalization and to rule outcomplicating disease, such as thyrotoxicosis, infec-tion, or pulmonary infarction.

2. Fluid intake was restricted to less than 1500ml., and salt intake to 2 grams or less daily.

3. Plasma pH, carbon dioxide combining power,chloride, sodium and potassium, urine output, andurine sodium, potassium, and chloride concentra-tions were determined daily.

4. Aqueous ammonium chloride was given dailyin 5 divided doses of 2 grams each, and Diamox in asingle daily dose of 750 mng. by mouth. If the pa-tient's initial carbon dioxide combining power waselevated (as in case 2), I)iamox was given withoutammonium chloride until its diuretic effect hadwaned, and then ammonium chloride w-as added.

5. When the chloride concentration in the urinehad risen over 40 milliequivalents per liter, Diamoxwas discontinued and 2.0 ml. of Mlercuhvdriin weregiven intramuscularly (laily. Ammonium chlorideadministration was continued throughout theperiod during which M\Iercuhy-diin was given.

6. Laboratory methods used for determinationsof plasma pH and carbon (lioxide combining power,plasma and urinary sodium, potassium, and chloridehave been previously described.8

CASE REPORTSCase 1. S. S., a 51-vear-old white luncheonette

owner, was admitted to the New York Hospitalfor the fifth time in severe congestive heart failure.He had first been hospitalized 2 years before witha n)yocardial infarction. Subsequent admissionswere for control of cardiac decompensation whichhad its onset 6 months after the occurrence of themvocaidial infarction. He was maintained on(digi-talis, a salt-poor diet, and restricted activity.

For 3 weeks prior to his present admission, hisdyspnea and orthopnea had become progressivelymore severe, despite continued use of digitalis,frequent mercurial injections, ammonium chloride,bed rest, and salt restriction.

Physical examination on admission revealed alethargic, severely dvspneic and orthopneic, cyanot-ic, acutely ill man with a blood pressure of 180/100,pulse 100 (regular), respiratory rate 30 per minute,temperature 39.2 C. He showed dlistention of theneck veins, bilateral pleural effusion, rales in bothlung fields, cardiomegaly with a gallop rhythm,hepatomegaly, and 4 plus pitting ankle and sacral

( 5 7

by guest on July 4, 2018http://circ.ahajournals.org/

Dow

nloaded from

TREATMENT OF LOW-SALT SYNDROME IN CONGESTIVE HEART FAILURE

TABiE 2.-Medications, Daily Weight Change, Plasma and Urine Electrolyte Concentrations, and Fluid Intake andOutput in Case 1

20-67 142-169

Plasma

PhNa+mEq

/L

131-112

168-16817.43 120

C1-mEq/L

98-8794

Co2-mM

/L

25-2122.6

K+mEq

4.6-6.65.8

168-16817.39120 99 19.615.2

168-16917.371122 106 115.215.1

169-168.5

168.5-153

153-134

Urine

Na+

Conc. TotalmEq mEq//L 24 hr.

1.8

1.3

1.7

7.34 120 1107 114.814.4 7.17.27 120 114 8.415.0 187.51622

7.34 128 112 112.213.0 198 840

134-126 7.451132 106 124.813.3 180.8j209

126-120 132 102 128 3.5

CoTml

Cl

nc. TotalEq mEq/L 24 hr.

0.61 12.8

0.91 17.4

1.61 38.6

5.11 43

118.41

136

139

K+

Conc. TotalmEq mEq/L /24 hr.

4.2178 25.3

11.4176.51 49.81 1100

36.1 71.5 66.9

30.1150.8j 35.6

842 26.5

1163 19

358 32

188

165

83

900

700

760

1000

1000

edema. Initial laboratory findings included 2 plusalbuminuria, hematocrit value 55 per cent, venous

pressure (antecubital vein) 300 millimeters of saline,circulation time (decholin, arm to tongue) 40 sec-

onds, blood urea nitrogen 41 mg. per cent, sodium131 mEq., chloride 98 mEq., and carbon dioxide25 mMI per liter.The patient was initially treated with complete

bed rest, 1 Gin. salt diet, oxygen, phlebotomy, andwas given additional digitalis until minor toxicsymptoms appeared. Despite these measures, heshowed no improvement. Table 2 is a tabulation ofthe daily weight, intake and output, plasma andurine electrolyte concentration, and medicationsgiven in this case. After 10 days his weight hadrisen, and his urine output had fallen below 1000ml. per day. The blood urea nitrogen was 67 mg.

per cent. The plasma sodium had fallen to 122mEq. and the plasma chloride to 87 mEq. per liter.

During the next 2 weeks, 5 per cent hypertonicsaline was administered intravenously in amountsof 100 to 200 ml. on 6 different occasions in an at-tempt to correct the hyponatremia and hypochlo-remia. At the end of this period the patient hadgained 13 pounds, showed anasarca, and was

symptomatically worse. Plasma sodium concentra-tion was 112 mEq. and plasma chloride concentra-tion was 98 mEq. per liter.

At this time, a review of the patient's clinicalcourse showed that in this 20-day period, a signifi-cantly positive daily water balance was present.Fluid intake had averaged 2300 ml. daily, while his

urine output averaged 800 ml. daily. During thisperiod, the patient had been afebrile and had notperspired excessively. The major factor in his hypo-natremia and hypochloremia was felt to be dilution,and it was decided that therapy should be directedprimarily toward mobilization of this excess totalbody water.

Accordingly, the patient was given 750 mg. ofDiamox in a single dose by mouth and 10 Gm. ofaqueous ammonium chloride in 5 divided dosesdaily for 4 days. Fluid intake was restricted to ap-proximate his urine output plus estimated insensibleloss. During this 4-day period of Diamox and am-

monium chloride administration, his weight didnot change. A hyperchloremic acidosis was pro-duced. The plasma pH fell from 7.43 to 7.27 andplasma chloride rose from 94 to 114 mEq. per liter.Carbon dioxide combining power fell from 22.6 to8.4 mai per liter. No symptoms of acidosis de-veloped. Plasma sodium remained low at 120 mEq.per liter, and plasma potassium stayed within nor-

mal limits. Urine output averaged 850 ml. dailywith very low-sodium content. Urine chloride con-

tent, initially low, rose as the plasma chloride levelincreased (table 2 and fig. 1).

In this setting of a hyponatremia and hyper-chloremic acidosis, Mercuhydrin (2 ml. intramuscu-larly) was administered daily for 5 days. Ammo-nium chloride administration was continued, andpotassium chloride was given on the second andthird days. A striking diuretic response was achieved,with a peak urine output of 812 liters occurring on

BloodUreaNitro-genmg.%o

wt.Change

Ibs.Day

1-20

21

22

23

24

25

26

27

28-35

Medication

iDiamox,NH4C1

DiamoxNH4C1

1)iamox,NH4C1

Diamox,NH4C1

MeIere,NH4C1

Mlerc, KC1,NH4C1

Alerc, KC1,NH4C1

Mlere, 2x

Intakecc./24 hr.

2300 0.1).50027

19

18

Outputcc./24 hr.

800 O.D.325

650

935

700

7100

8600

2600

658

by guest on July 4, 2018http://circ.ahajournals.org/

Dow

nloaded from

I{UBIN ANI) B3RAVIIAN

WEIGHtT 170 @ _ -

'60K

Lbs. 150 L

140

130

PLASMA 114Chloride mEq/L

94_

28-

C02 mMol/L **_ __8

745

-725

135rSodium mEq/L r

115 /

Potossium mEq/L 6 F3

-~~~~~~~~~~

*-- WO-

1000 10UR/NE

800 -8

Electrolyte

Excretion

400mEq/24 hr.

200

Diamox, Diamox, Diomox, Diomox, Merc, Merc, Merc, MercNH4CI N H401 NH4CI NH4CI NH4CI NH4CI, KCI

KriMEDICA TIONS

DAYS OF FV5l^_ADMNINISTRATI7ON

FI(;. 1. Case 1. A graphic rel)resentation of the course of observation. On the ordinates are liste(l thebody weight, Ilasma electrolyte concentrations and pH, and urine electrolyte excretions and outp1utfor each 24-hour period. On the abscissas are listed the medicaitions given aIn1d the number of (lays

each medication, or combinat ioni of medications, was given.

the second day of mnercurial administration. In this3-day period, the patient lost 42 pounds in wveight.Total(daily urinary excretion of sodium an(l chloridewas high, with chloride content exceeding sodiumon each (ay. However, it is. miost signifcant to notethat Witil respect to the plasma, the urine each daywas hypotonic so far as sodiumi was concerned. Inthis 3-day perio(l, the hyponatremia was correcte(l,the plasma sodiunm rising froom 120 to 132 miEq.pel liter. The patient's electrolyte pattern at theend of this time had returned almost to normal(sodiumi 132 miEq., potassium 3.4 mql'., clhloride102 mIiq., ca.bon (lioxidle 28 mnM per liter). The1)lood urea nitrogen had fallen to 18 mig. per cent.Clinicall lhe was dlraiatically imnproved and sub-

sequently he could be allowed out of bed. Figure 1

is a graphic rel)resentation of the patient's responseto this regimen.

This imiprovement continued in the next 2 weekswith a further 6-pound weight loss on mercuhliy-(drinalone.

In his sixth hospital week, while arrangemnentsfor further (.cae at home were being matde, the pa-

tient (liel quietly while asleep, floparently from a

mnyocardial iinfarction l)ulmonary embolismll.

Case 2. L. T., a 53-year-old white male clerk wa.sadmitted to the New York Hospital in l)uliiioiai'ryedlema. He hald a 20-yea.rt history of hypertensionancl chronic lung disease. When alnmittecl to a hos-pital 1 year before, he -as in chronic( heart failureand. respon(le(l well to a regimen of digitalis, saltrestri(tion, an(l mercurial diuretics. At that timethe laboratory findings included a hemaitocrit valueof 37, a 1)loo(1 urea nitrogenIof 15 Ilg. peI (cent, and

carlbon dioxi(le combining power of 36 mM\[ perliter. The electrocardiogram showed alplttern ofleft ventricular hypertrophv. Roentgenogram of the(hest was comp)atible with pulmonary emphysema.Gastrointestinal and hemnatologic work-up) revealedno explanation for the low- hematocrit valmue. Thepatient did m-ell on an ambulatory regimen of digi-talis, mercurials, and a salt-poor (liet until 2 monthsprior to his lresent allmission, at which time lhe ladan upper respiiratorv infection, and clysspnea andankle edema recurred. During the 2 weeks prior toa.lmlllissioO, shortness of breath and ankle swelling

pH

= KM NaOCl- H20

- 6;59

by guest on July 4, 2018http://circ.ahajournals.org/

Dow

nloaded from

TREATMENT OF LOW-SALT SYNDROME IN CONGESTIVE HEART FAILURE

increased despite mercurial injections twice a week.During this period his fluid intake was high andsalt intake was drastically curtailed.

Physical examination on admission revealed ablood pressure of 180/110, pulse 100 (regular), rate30 per minute, temperature 37.2 C. He seemed bothacutely and chronically ill, and was disoriented,dyspneic, and cyanotic. There was grade III retinop-athy. Neck veins were distended. The chest hadan increased anteroposterior diameter, was hyper-resonant to percussion, had diminished tactilefremitus, and fine, medium, and coarse rales through-out. His heart was enlarged to the anterior axillaryline, and an apical systolic murmur was present.The liver was felt 5 fingerbreadths below the rightcostal margin. Ascites and 4 plus ankle and sacraledema were present.

Laboratory findings on admission included urinespecific gravity of 1.012 with 2 plus albuminuria, a

hematocrit value of 37 per cent, and a blood ureanitrogen of 29 mg. per cent. The venous pressurewas 200 mm. of saline in the antecubital vein, andthe circulation time was 18 seconds (arm to tongue,decholin). Plasma sodium concentration was 120mEq. per liter. Roentgenogram of the chest re-vealed cardiomegaly and was compatible with pul-monary emphysema. The electrocardiogram showeda pattern suggestive of left ventricular hypertrophyin the precordial leads, with no axis deviation andclockwise rotation.With the history of salt restriction, oliguria,

weight gain, and high fluid intake, it was felt thatthe hyponatremia was the result of dilution ratherthan inordinate urinary losses of salt. Table 3 liststhe daily weight, intake and output, plasma andurine electrolyte concentrations, and medicationsgiven in this case.

The patient was placed on bronchodilators, bed

TABLE 3. Medications, Daily Weight Change, Plasma and Urine Electrolyte Concentrations, and Fluid Intake andOutput in Case 2

Medication

MercD)iamox 750mg.

Diamox

1)iamox

I)iamox

1)iamox, KC1

Diamox, KCl

D)iamox,NH4C1, KCIDiamox,NHW, KCI

D)iamox, NHW,KCl

Diamox, NH3,KC1

Merc, NH4Cl,KCI

Mere, NH.IC1,KC1

Mere, NH4C1,KCI

Merc, NH4C1,KCI

Merc, NH4C1,KCI

6

a.ID EiL. E

0X

29

17

19

u:D

b£

U

156-154154-148.3148.3-147147-145.4145.4144.9144.9-144

144-144144-144144-1431431134

134-129129-118.3

Plasma

PhNa+ C C02- K+mEq mEq mm mEq/L /L /L

120

7.47 120 81

7.44 120 83

37

33

4.6

3.5

32.913.0

7.41 122 87.6130.413.4

7.955 125 93.8129.5 3.4

7.341 125 1102 121.314.2

7.361 131 1100

[138 104

23 3.4

25 4.9]

Urine

Nat C

Conc. Total Conc.mEq mEq, mEq/L 24 hr. /.

9.3 14.0

6.0 21.0

22.0 50.6

16.5 51.2

12.5 38.8

6.0 17.4

4.0 7.6

6.0 10.2

2.8

TotalmEq/24 hr.

K+

Conc. TotalmEq mEq,/L 24 hr

4.2146.3 69.5

3.61 12.6131.511106.0 13.8116.51 38

6.0 18.6118.8 58.3

5.41 16.7123.8 73.8

4.21 12.2123.5 68.2

6.01 11.4131 58.9

!1 30.4 61.9157.5 97.8

11.51 19.51 78.41133.3166

31.51 60 121 230

112.2

59.5 113

65.31274.31114.8482.2145.31190.351 193.8 130.'2494.8 54.8 208.2

Day

12

3

4

5

6

7

8

9

10

11

12

13

14

15

15-17

Intakecc./24 hr.

High

2.0 gm.salt

Fluids re-

stricted to1500 cc.

Outputcc./

24 hr.

2650

3500

2300

3100

3100

2900

1900

1700

1700

1900

4200

3800

4500

660

by guest on July 4, 2018http://circ.ahajournals.org/

Dow

nloaded from

RUBIN ANI) BRAVEIMIAN

rest, salt restriction, and was given a mercurialdiuretic the first (ay. He lost no weight during thefirst 24 hours, although his pulmonary status im-

proved. He was then given 750 mg. of Diamox bymouth dailv for 7 days with 5 Gm. of potassiumchloride on the fifth and sixth days. During thistime, due to a misunderstanding, fluid intake was

not restricted. The patient responded with a diuresis,losing 7.2 Kg. in weight. The major weight loss, 6Kg., occurre(l in the first 4 days, during which timethe plasma carbon dioxide combining power fellfrom 37 to 29.5 mM\1 per liter. This initial response

to Diamox alone is attributable to the high plasmabicarbonate content. During this period of diuresisthe urine contained very little chloride, sodium con-

tent was low, and potassium high. At the end of this6-day period, the plasma sodlium was 125 mEq.,chloride 93.8 mEq., carbon dioxide combining power

29.5 m\IL, and p)otassium 3.4 imEq. per liter. Theplasma pH was 7.35, falling from pre-treatmentlevel of 7.47 (fig. 2).

For the next 4 days Diamox and potassium chlo-ride were continued, ancl 8 graims of aqueous am-

WEIGHT

Lbs.

monium chloride in 4 diviided dailv doses were added.During this period the plasma sodium remained at125 mEq. and the carbon dioxide combining power

fell to 21.3 mrM per liter. The plasma pH was 7.34at the end of this period. No significant weightchange occurred. Through error, fluid restrictionhad not yet been instituted. Urinary output averaged1800 ml. dailv.

Urinary chloride excretion, initially low, increased11 fold in daily steplwise fashion as the plasma chlo-ride concentration returned to a normal level. In

this setting of hyponatremia and a normal plasmachloride concentration, Diamox administration was

stopped, fluid intake was restricted to 1500 ml. aday and Mlei(uhydrin, 2.0 ml. intramus(ularly, was

administered (lailv for 7 davs. Ammonium chlorideandlpotassium chloride supplements were continuedthroughout this perio(l. A sustained diuresis ensuedand the patient lost 25.5 pounds in weight. Duringthis period the urine (ontained a high concentrationof chloride ion and a moderately high concentrationof sodium ion (table 3). floweter, each liter of urinewas signifijaitly hypotonic for sodium, writh/ respect to

158

148

138

128

- 1IR

PLASMA 102_Chloride mEq/L

82

C02 mMol/L 3OF20

a~

-- -~~~w

Potassium mEq/L 5F _ _ _

~*

URINE 600 -6

Electrolyte FL

Excretion I:

mEq/24hr. 1 Jb

= KNa

= Cl

- H20

MEDICATION

DAYS OFADMINISTRATION

\ \ .... .

Diamox Diamox Diamox, Merc, Merc, Merc,+ KOI NH4CI, NH4CI, NH4CI, NH4CI,

KCI KCI KCI KCI

3 2 4 2

FIG. 2. Case 2. A graphic representation of the course of observation dluring which the regimendcscribed in the text w^as used. Where the days of drug administration are more than one, the urine

electrolyte excretions and urine outputs graphed represent averages of the number of days recorded ineach instance.

pH

Sodium mEq/L _130 r

77301C m : -

-2m

661

by guest on July 4, 2018http://circ.ahajournals.org/

Dow

nloaded from

62TREATXIENT OF LOW-SALT SYND)1ROIIE IN CONGEjSTIVE, HE',ART FAILUREI

th1e plasma. Thus, at the end of this mercurial diureticI)eriod, the plasma sodium concentration had risenfrom 125 to 138 mEq. per liter. Figure 2 is a graphicrepresentation of the patient's response to thisregimen.

The psatient was symptomatically strikinglyT imll-proved, was edema free, and had a normal plasmaelectrolyte pattern. He was discharge(l from thehospital, and is at present being maintained edema-free on a regimen of salt and fluid restriction, dail-ammonium chloride and bi-weekly mercurial injec-tions.

The other 3 patients treated by this regimenshowed similar clinical improvement, and rises inllasma sodium concentrations from 118-135 mEq.,119-128 mEq. and 122-136 mEq. per liter, respec-tively. These changes occurred in the course of asustainedl diuresis resulting in each instance in amarked weight loss (14, 16, and 24 l)ouncls respec-tivelv).

DISCUSSIO.N

The 2 cases reported above typify thegroup of cardiac patients that we have seenwith fluid retention, hyponatremia, and hypo-chloreinia. The positive water balance thatwe have found to be a consistent feature inthese cases is clearly demonstrated. Treatmentin this situation poses a difficult problem. In

most instances, as in case 1, the administrationof hypertonic saline solution ini an attemipt tocorrect the hyponatremia does not, improvethe clinical status of these patients. It oftenresults in a further weight gain and a progres-sion of already distressing symptoms.A responsiveness to mercurial diuretics was

restored by using Diamox and amimnioniuchloride to produce that rise in plasma chlorideconcentration necessary to provide for thepresentation of all adequate chloride load tothe renal tubules. Our emnpiric observatiollhas beeni that this state is achieved when theurinary chloride concentration has risen toover 40 mE(q. per liter. The mechanism ofthe action of Dianiox in producing a rise inplasma chloride concentration has beenstudied.'The significant characteristic of the iner-

curial-induced diuresis that occurred, once thenecessary hyperchloremia has been achieved,was that each liter of urine had a (considerlablylower sodium content than the extra(ce(llularfluid. As a result, with fluid intake restricted,the sodium concentration of the extiracelltular

Urine Vol.

Na=90 mEq/LFIG. 3. l)iagrammatic representation of the changes in extracellular fluid volume and sodium c(o1-

centration that occur as a result of the diuresis achieved in the setting of a hyperchloremic acidosisby the application of the regimen described in the text,.

Sodium 'gain' by body as a result of diuresis of a urine that is hypotonic for sodium with respectto the E.C.F. can be calculated as: [(120 (E.C.F. cone./l,.) - 90 (urine cone./L.)l X 7 (liters ofurine) = 210 mEq. of sodium available for redistribution ini the 1,.C .F. remaining after diuresis.

Initial E.C.F.27 Liters

Final E.C.F. Vol.20 Liters

+ Mercuhydrin

Na =120 mEq/L

(;162

by guest on July 4, 2018http://circ.ahajournals.org/

Dow

nloaded from

RUBIN AND BRAVEMAN

fluid rose toward normal (fig. 3). Strikingclinical improvement occurred in associationwith the diuresis and the return to normal ofthe plasma sodium concentration.Producing an acidosis in an already azotemic

patient is potentially hazardous, and the neces-sity for continued close clinical and laboratoryobservations cannot be overemphasized. How-ever, with such observations, the acidosis hasbeen without ill effect. The degree of acidosisand rise in plasma chloride concentration neces-sary to restore a responsiveness to mercurialdiuretics varies with each indivridual patient.The essential indication that the necessaryacidotic state has been achieved is a rise inurinary chloride concentration. It is to beemphasized that this is unrelated to a specificlevel of plasma pH or plasma chloride con-centration. This principle is clearly demon-strated in the 2 cases reported above. In case1, the urinary chloride concentration, initially12.8 mEq. per liter, rose to 43 mEq. per literwhen the plasma pH was 7.27 and the plasmachloride concentration was 114 mEq. per liter.By contrast, in case 2 the urine chloride con-centration initially 2.8 mEq. per liter rose to78.4 mnEq. per liter (indicative of an adequatesetting for mercurial administration) when theplasma pH was 7.34 and the plasma chlorideconcentration was only 102 mEq. per liter.The effectiveness of this therapeutic ap-

proach, directed primarily toward reducingthe expanded extracellular fluid voluime, andnot primarily toward increasing the sodiumcontent of the extracellular fluid, supports thepostulation that the mechanism for develop-ment of this syndrome is primarily fluid reten-tion with extracellular fluid expansion, andnot salt depletion. In keeping with this view isthe role that mercurial diuretics play in thesuccess of this regimen, when heretofore theyhave beei purported to be a significant causa-tive factor in producing the "low-salt"syndlrome.

C: ONCLLUSIONS

]. An effective regimen for treatment ofthe "loN-salt syndrome" utilizing Diamox,ammoinium chloride, aild mercurial diureticshas been presented.

2. It is suggested that extracellular fluid cx-

pansion with electrolyte dilution is the majorfactor in the production of the "low-saltsyndrome," which might more aptly be termedthe "dilution syndrome."

ACKNOWLEDGMENTThe authors express applreciation for the technical

assistance of Miss Naomi Schechter ancl Mrs. RuthAronson, and thank Dr. Henry Cari and Dr. ElliottHochstein for the opportunity to see patients re-ported in this paper. We are indebted to Dr. ThomasP. Almy for his mnany helpful suggestions in carryingout this work.

SUMMAIRG IN- INTERLINGUA

1. Es presentate un efficace regime pro letractamento del "syndrome de basse nivellosde sal," utilisante Diamox, chlorido de am-mollium, e diureticos mercurial.

2. Nos opina (cue le expansion extracellularde fluido occurrente in le presentia de dilutionelectrolytic es le major factor in le productiondel "'synidrome a basse nivellos de sal." Ilesserea plus appropriate designar iste syn-dromre como '"syiidrome de dilution."

REFERENCES1 SCHWARTZ, WV. B., AND WVALLACE, A. M4.: Elec-

trolyte equilibrium (luring mercurial (liuresis..J. Clin. Invest. 30: 1089, 1051.

2 DANOWSKI, T. S., F:1RGUS, E. B., AND MATEER,F. .M.: The low salt syndromes. Ann. Int. Med.43: 643, 1955.

3 SCHROEDER, H. A.: Renal failure associated withlow extracellular sodium chloride. J.A.-M.A.141: 117, 1949.

' SCHROEDEIR, H. A.: Studies on congestive heartfailure. I. The importance of restriction of saltas eompared to water. Am. Heart .J. 22: 141,1941.

5 LE ITTER, L., WESTON, R. E., AND GROSSMAN, J.:The low salt syndrome; its origins and varieties.Bull. N. Y. Aca(l. Med. 29: 833, 1953.

6 NEWIMAN, E. V.: Hyponatremia syndrome. Arch.Int. Med. 95: 374, 1955.

AXELROD, D. R., CAPPS, .J. N., AND PITTS, R. F.:Potentiaition of diutetic action of S.lvigan byammonium chloride. Fed. Proc. 9: 6, 1950.

8 RUBIN, A. L., THOMIPSON, H. G., BRAVEMAN,AW. S., AND LUCK E-, E. H.: The management ofrefractorv edemia, in heart failure. Ann. Int.Med. 42: 358, 1955.

9 BRAVEMAN, AV. S., RUiIN, A. L., AND M\IEAD, A.WV.: `Mechanisms of (levelopment of hvperchlor-emia following blockage of bicarbonate ion reab-sorlption in the renial tubule. Clinical ResearchProc. 3: 136, 195l95.

663

by guest on July 4, 2018http://circ.ahajournals.org/

Dow

nloaded from

ALBERT L. RUBIN and WARREN S. BRAVEMANControlled Use of Mercurial Diuretics

Treatment of the Low-Salt Syndrome in Congestive Heart Failure by the

Print ISSN: 0009-7322. Online ISSN: 1524-4539 Copyright © 1956 American Heart Association, Inc. All rights reserved.

75231is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TXCirculation

doi: 10.1161/01.CIR.13.5.6551956;13:655-663Circulation. 

http://circ.ahajournals.org/content/13/5/655located on the World Wide Web at:

The online version of this article, along with updated information and services, is

  http://circ.ahajournals.org//subscriptions/

is online at: Circulation Information about subscribing to Subscriptions: 

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

  document. Permissions and Rights Question and Answer

of the Web page under Services. Further information about this process is available in thewhich permission is being requested is located, click Request Permissions in the middle columnClearance Center, not the Editorial Office. Once the online version of the published article for

can be obtained via RightsLink, a service of the CopyrightCirculationoriginally published in Requests for permissions to reproduce figures, tables, or portions of articlesPermissions:

by guest on July 4, 2018http://circ.ahajournals.org/

Dow

nloaded from