J O U R N A L O F T H E A M E R I C A N C O L L E G E O F C A R D I O L O G Y VO L . 6 5 , N O . 5 , 2 0 1 5

ª 2 0 1 5 B Y T H E A M E R I C A N C O L L E G E O F C A R D I O L O G Y F O U N D A T I O N I S S N 0 7 3 5 - 1 0 9 7 / $ 3 6 . 0 0

P U B L I S H E D B Y E L S E V I E R I N C . h t t p : / / d x . d o i . o r g / 1 0 . 1 0 1 6 / j . j a c c . 2 0 1 4 . 1 2 . 0 1 0

THE PRESENT AND FUTURE

STATE-OF-THE-ART REVIEW

Hyponatremia in Acute DecompensatedHeart FailureDepletion Versus Dilution

Frederik H. Verbrugge, MD,*y Paul Steels, MD,z Lars Grieten, PHD, MSC,*z Petra Nijst, MD,*y W.H. Wilson Tang, MD,xWilfried Mullens, MD, PHD*z

ABSTRACT

Fro

Ha

ve

Cle

Gr

fou

rep

Ma

Hyponatremia frequently poses a therapeutic challenge in acute decompensated heart failure (ADHF). Treating physicians

should differentiate between depletional versus dilutional hyponatremia. The former is caused by diuretic agents, which

enhance sodium excretion, often with concomitant potassium/magnesium losses. This can be treated with isotonic saline,

whereas potassium/magnesium administration may be helpful if plasma concentrations are low. In contrast, as impaired

water excretion, rather than sodium deficiency, is the culprit in dilutional hyponatremia, isotonic saline administration

may further depress the serum sodium concentration. Because free water excretion is achieved by continuous sodium

reabsorption in distal nephron segments with low water permeability, diuretic agents that impair this mechanism (e.g.,

thiazide-type diuretic agents and mineralocorticoid receptor antagonists) should be avoided, and proximally acting

agents (e.g., acetazolamide and loop diuretic agents) are preferred. Vasopressin antagonists, which promote low water

permeability in the collecting ducts and, hence, free water excretion, remain under investigation for dilutional hypo-

natremia in ADHF. (J Am Coll Cardiol 2015;65:480–92) © 2015 by the American College of Cardiology Foundation.

H yponatremia, defined as a serum sodium(Naþ) concentration <135 mEq/l, is themost common electrolyte disorder in hos-

pitalized patients (1). Both admission and hospital-acquired hyponatremia are associated with anincreased risk for adverse outcomes, including pro-longed hospital stay, need for discharge to a short-or long-term care facility, and all-cause mortality(2). In a subanalysis from the OPTIMIZE-HF (Orga-nized Program to Initiate Lifesaving Treatment inHospitalized Patients with Heart Failure) registry,including 47,647 patients with acute decompensatedheart failure (ADHF), hyponatremia was present inw20% upon admission (3). In addition, the incidence

m the *Department of Cardiology, Ziekenhuis Oost-Limburg, Genk, Belgiu

sselt University, Diepenbeek, Belgium; zBiomedical Research Institute, F

rsity, Diepenbeek, Belgium; and the xDepartment of Cardiovascular Medi

veland, Ohio. Dr. Verbrugge is supported by a PhD fellowship from t

ieten, Nijst, and Mullens are researchers for the Limburg Clinical Resea

ndation Limburg Sterk Merk, Hasselt University, Ziekenhuis Oost-Limb

orted that they have no relationships relevant to the contents of this pap

nuscript received July 30, 2014; revised manuscript received November 3

of hospital-acquired hyponatremia during deconges-tive treatment for ADHF is probably w15% to 25%(4,5). Hyponatremia frequently poses an importanttherapeutic challenge in ADHF because simpleadministration of the depleted ion (as with other de-ficiencies) cannot be easily performed, and there isan obvious concern for harmful fluid overload. More-over, the pathophysiology of hyponatremia in ADHFis often dilutional, rather than depletional (6,7).Importantly, ubiquitous use of powerful Naþ-wastingdiuretic agents in this context hampers differentia-tion between the 2 conditions, each of which requiresa totally different approach. This review, therefore,aims to provide, on the basis of the currently

m; yDoctoral School for Medicine and Life Sciences,

aculty of Medicine and Life Sciences, Hasselt Uni-

cine, Heart and Vascular Institute, Cleveland Clinic,

he Research Foundation–Flanders. Drs. Verbrugge,

rch Program UHasselt-ZOL-Jessa, supported by the

urg, and Jessa Hospital. Drs. Steels and Tang have

er to disclose.

0, 2014, accepted December 2, 2014.

TABLE 1 Pathophysiology of Hyponatremia in

Acute Decompensated Heart Failure

Mechanism of Action

Dilutional hyponatremia

Increased sensitivity of osmotic AVPrelease / Lower osmo-checkpoint*

Baroreceptor activation/angiotensin II

Increased nonosmotic AVP release Baroreceptor activation/angiotensin II

Impaired AVP degradation Liver and/or kidney dysfunction

Increased thirst Baroreceptor activation/angiotensin II

Decreased distal nephron flow Impaired glomerular filtration/Increasedproximal tubular reabsorption

Depletional hyponatremia

Low sodium intake Salt-restricted diet

Exaggerated nonurinary sodium losses Diarrhea, ascites

Exaggerated natriuresis Diuretics, osmotic diuresis

Sodium shift toward the intracellularcompartment

Potassium and/or magnesiumdeficiency

*This is the level of plasma osmolality that is pursued by the homeostatic mechanisms of thebody.

AVP ¼ arginine vasopressin.

AB BR E V I A T I O N S

AND ACRONYM S

ADHF = acute decompensated

heart failure

AVP = arginine vasopressin

ENaC = epithelial sodium

channel

GFR = glomerular filtration

rate

MRA = mineralocorticoid

receptor antagonist

Naþ = sodium

TAL = thick ascending

limb of Henle’s loop

J A C C V O L . 6 5 , N O . 5 , 2 0 1 5 Verbrugge et al.F E B R U A R Y 1 0 , 2 0 1 5 : 4 8 0 – 9 2 Hyponatremia in Decompensated Heart Failure

481

available evidence, a pathophysiology-based assess-ment and management strategy for the importantclinical challenge of hyponatremia in ADHF.

PATHOPHYSIOLOGY OF

HYPONATREMIA IN ADHF

Table 1 provides a summary of the pathophysiology ofhyponatremia in ADHF.

DILUTIONAL HYPONATREMIA. It is generally assumedthat hyponatremia in ADHF is more often a problemof impaired water excretion than Naþ depletion (6,7).Two key events driving increased water retentionand progression of hyponatremia in ADHF are in-creased nonosmotic release of arginine vasopressin(AVP) and insufficient tubular flow through diluting(distal) segments of the nephron. In many ways, thisis the opposite of the situation in diabetes insipidus,where deficient AVP production and/or functionin the presence of normal tubular flow results invery diluted urine, exaggerated water losses, and,ultimately, in hypernatremia, unless compensationoccurs through increased water intake.AVP and regulat ion of p lasma osmola l i ty . AVP, acyclic octapeptide (1,099 D) with a 3-amino acid tail,is a key regulator of water homeostasis. AVP is syn-thesized in large-diameter neurons in the supraopticand paraventricular nuclei of the anterior hypothal-amus. After axonal transport into nerve terminalswithin the posterior lobe of the pituitary, AVP isreleased into the bloodstream in response to plasmahypertonicity (8). In healthy, normally hydratedpersons, circulating AVP levels are very low(w1 pg/ml) because of its rapid degradation andexcretion by the liver and kidneys (t1/2 ¼ 15 to 20 min)(9). Consequently, hepatic dysfunction and renalinsufficiency may contribute to increased plasmalevels of AVP in ADHF. AVP exerts its water-retainingeffects primarily through stimulation of high-affinityV2 receptors in the collecting ducts of the nephron(Figure 1) (9). V2 stimulation increases both the syn-thesis and availability of aquaporin-2 channels on theluminal side of these collecting ducts, which other-wise have low water permeability (10). The hypertonicenvironment of the renal interstitium subsequentlyprovides a strong impetus for water reabsorption,resulting in decreased water excretion and, conse-quently, less diuresis. Importantly, this system is verysensitive to small changes in plasma AVP levels, whichare difficult to detect, even by modern assays (11). Athigher concentrations, the low-affinity V1a receptor,which is expressed in the liver, collection ducts,and vasa recta of the nephron, is stimulated (Figure 1).This further enhances hypertonicity in the renal

interstitium—hence antidiuresis—through in-creased hepatic urea production, improvedmedullary urea reabsorption in the collectingducts, and reduced blood flow through thevasa recta (Figure 1) (12,13). The vasa recta arestraight capillaries in the renal medulla with ahairpin loop. They receive a lower proportionof blood flow compared with the renal cortex,especially during antidiuresis. This is impor-tant, as high blood flow through the vasa rectawashes out the tonicity gradient from therenal cortex toward the medulla, which isneeded to concentrate the urine. Finally,another effect of V1a receptor stimulation is

promotion of prostaglandin synthesis in the collectingducts. This counteracts V2 effects on aquaporin-2, thusstimulating water excretion (14). This latter effect mayexplain why some heart failure patients demonstrate anormal response to water loading, with production ofadequately diluted urine, despite having elevatedplasma AVP levels (15).Nonosmot ic arg in ine vasopress in re lease inheart fa i lure . Several studies have demonstratedthat AVP levels are generally elevated in heart failure(16–18). Moreover, aquaporin-2 expression wasmarkedly up-regulated in a rat model of congestiveheart failure (19), which would be expected to in-crease water permeability and reabsorption in thecollecting ducts of the nephron. It is tempting tospeculate that both events are causally related andpromoting excessive free water reabsorption, leadingto hyponatremia in ADHF. Certainly, baroreceptoractivity, sympathetic overdrive, and angiotensin II,all stimulated by decreased effective circulatory

FIGURE 1 Effects of AVP in the Nephron

LOW WATER PERMEABILITY

INTERLOBARARTERY

VASA

RECTA

AP2 EXPRESSIONV2

V1a:

V1aV1a: VASOCONSTRICTION

UREA REABSORPTIONINCREASED HEPATIC UREAPRODUCTION

BLOOD FLOW

OSMOTIC GRADIENTOSMOTIC GRADIENT

INCREASED UT-A1 TRANSPORTER

UreaUT-A1 transporterAquaporine-2 (AP2)StimulatedInhibited

H2O

H2OH2O

H2O

Macula densa

Distal tubules Collecting ducts

TAL

V2-receptor effects: The primary effect of AVP on water homeostasis is mediated through

stimulation of high-affinity V2 receptors, already with small increases in plasma levels (9).

V2-receptor stimulation in collecting duct cells results in increased production of AP2

channels and facilitates recruitment of these channels on the luminal membrane, which

otherwise has low water permeability (10). Subsequently, water reabsorption is promoted

as the renal interstitium is hypertonic and the basolateral membrane of collecting duct

cells is highly permeable to water. Hypertonicity in the renal interstitium is accomplished

through solute reabsorption in parts of the nephron that have relatively low water

permeability, that is, sodium and chloride transport in the TAL and urea reabsorption in the

medullary part of the collecting ducts through UT-A1 transporters.

V1a-receptor effects: At high plasma levels, AVP further stimulates hypertonicity in the

renal interstitium, and hence water reabsorption, through activation of the low-affinity V1a

receptor. First, this promotes both hepatic urea production and UT-A1 transport in the

medullary part of the collecting ducts, resulting in higher urea concentrations and a

stronger osmotic gradient in the renal interstitium. Furthermore, V1a receptor stimulation

causes vasoconstriction in the vasa recta, preventing wash-out of solutes from the renal

medulla (12,13). On the other hand, V1a receptor stimulation in collecting duct cells also

promotes prostaglandin synthesis. Prostaglandins counteract the effects of V2 by impeding

AP2 recruitment in these collecting duct cells (14). This somewhat preserves the diluting

capacity of the distal nephron, even when AVP levels are high because of neurohumoral

activation and nonosmotic AVP release. AP2 ¼ aquaporin-2; AVP ¼ arginine vasopressin;

TAL ¼ thick ascending limb of Henle’s loop.

Verbrugge et al. J A C C V O L . 6 5 , N O . 5 , 2 0 1 5

Hyponatremia in Decompensated Heart Failure F E B R U A R Y 1 0 , 2 0 1 5 : 4 8 0 – 9 2

482

volume in ADHF, fuel nonosmotic AVP release, as wellas thirst and thus free water intake (20). Importantly,they also increase the sensitivity of osmotic AVPrelease (Figure 2A). Indeed, compared with healthyvolunteers, ADHF patients show a more pronouncedincrease in plasma AVP levels with osmotic loading(21). In contrast to this osmotic AVP release, whichincreases linearly and with very small changes inserum osmolality, nonosmotic AVP release is expo-nential (Figure 2B). Thus, isotonic plasma volumereductions of 5% to 10% will have little effect on AVPsecretion, whereas reductions of 20% to 30%,

resulting in baroreceptor activation and angiotensinII stimulation, may cause plasma levels many timeshigher than necessary to produce maximal anti-diuresis (50 to 100 pg/ml) (22). Importantly, at thesehigh levels, AVP has potent V1a -mediated vasocon-strictor effects that may contribute to hemodynamicdeterioration in ADHF (23).Insufficient tubular flow through diluting segmentsof the distal nephron. Healthy kidneys are ableto dilute urine to an osmolality as low as 30 to 60mOsm/l, w5- to 10-fold lower than serum osmolality(24). However, micropuncture studies have demon-strated that tubular fluid osmolality is quite similar(w150 mOsm/l) at the level of the macula densa dur-ing antidiuresis versus water diuresis (25). Thus,urine dilution depends heavily on the tubular func-tion of more distal parts of the nephron (i.e., distalconvoluted tubules and collecting ducts). Free waterexcretion is achieved by these segments throughcontinued solute reabsorption, primarily through thethiazide-sensitive Naþ/chloride cotransporter andaldosterone-sensitive epithelial sodium channels(ENaCs), in the presence of low water permeability(24,26). Importantly, less sodium is reabsorbed byENaCs in conditions of low tubular flow, irrespectiveof aldosterone levels (27). Consequently, if more so-dium is retained in the tubular lumen, free waterexcretion is restricted. The ability of the kidneys toexcrete free water is, therefore, directly dependenton: 1) the amount of tubular fluid offered to the distalnephron; and 2) the relatively low water permeabilityof this segment, which can only be overcomeby insertion of aquaporin-2 channels through AVPstimulation. For instance, a healthy person with aglomerular filtration rate (GFR) of 180 l/day(125 ml/min), of which 10% reaches the distal convo-luted tubules and with total AVP suppression, wouldbe able to drink w18 l (10% of 180 l) over the course ofa day without developing hyponatremia (Figure 3A).Yet, with the GFR lowered to 43.2 l/day (30 ml/min),and only 5% distal tubular flow because of increasedproximal reabsorption in ADHF, maximal waterintake without hyponatremia development wouldbe decreased tremendously to 2.16 l/day (5%of 43.2 l/day) (28,29). The latter is even an over-estimate, as it presumes that water permeability inthe distal nephron is as low as with total AVP sup-pression, which is clearly not the case in ADHF(Figure 3B). An intriguing study by Bell et al. (30)further supports the concept of decreased distalnephron flow in AHDF. In this study, free waterexcretion was increased in ADHF patients withhyponatremia after infusion of 5% mannitol, an os-motic diuretic that enhances tubular flow through the

FIGURE 2 Determinants of AVP Plasma Levels

12

8

4

50

40

30

20

10

05 10 15 20

260 270 280 290 300 310Plasma Osmolality (mOsm/l)

% Plasma osmolality increase% Iso-osmotic plasma volume decrease

Normal

Osmotic Non-osmotic

Angiotensin IIbaroreceptor activity

Osmotic AVP Release

Osmotic Versus Non-osmotic AVP release

Plas

ma

AVP

(pg/

ml)

Plas

ma

AVP

(pg/

ml)

B

A

(A) Osmotic AVP release: AVP is near total suppression in normal

circumstances, but increases linearly with plasma osmolality (9).

Neurohumoral activation through angiotensin II and baroreceptor

activity increases the amount of AVP release for any given plasma

osmolality (20). In addition, it moves the set-point of total AVP

suppression to a lower plasma osmolality, hence promoting hy-

potonicity. (B) Osmotic versus nonosmotic AVP release: osmotic

AVP release is very sensitive and increases linearly, already with

very small changes in plasma osmolality (blue curve) (9). Non-

osmotic AVP release makes little contribution to overall AVP

plasma levels until a 10% to 15% decrease in isotonic plasma

volume, after which AVP release increases markedly, resulting in

plasma levels many times higher than needed to achieve maximal

antidiuresis (red curve) (22). Abbreviations as in Figure 1.

J A C C V O L . 6 5 , N O . 5 , 2 0 1 5 Verbrugge et al.F E B R U A R Y 1 0 , 2 0 1 5 : 4 8 0 – 9 2 Hyponatremia in Decompensated Heart Failure

483

nephron. Mannitol exerts its diuretic effect throughsolvent drag and water retention in the tubularlumen, thereby also increasing distal delivery. Similarfindings have been demonstrated in patients withcirrhosis, where the effective circulatory volume andrenal perfusion are also impaired (31). However, itremains possible that correction of hyponatremiawith mannitol in ADHF and cirrhosis is partly due toincreased intravascular volume after infusion, sup-pressing baroreceptor tonus and, hence, nonosmoticAVP release (32).

DEPLETIONAL HYPONATREMIA. Because increasedNaþ avidity characterizes heart failure from the very

beginning and is exacerbated by unrestrainedneurohumoral up-regulation, Naþ depletion is rela-tively rare in ADHF not treated with diuretic agents(33,34). Still, osmotic diuresis because of hypergly-cemia in the case of severe uncontrolled diabetes,gastrointestinal losses, and third-space losses may allcontribute to a negative Naþ balance, especially ifpatients adhere scrupulously to the salt-restricteddiets recommended by the guidelines (35–37). Onegroup called these diets into question, demonstratingthat heart failure patients adhering to such dietsexperienced increased hospitalizations and a highermortality rate (38–40). However, it should be notedthat these patients also received a high dose ofdiuretic agents (125 to 500 mg furosemide equivalentstwice daily), increasing the risk of a negative Naþ

balance. Indeed, the combination of very low dietarysodium intake and exaggerated losses might leadto progressive depletion of whole-body sodium stores.Loop diuret i c - induced hyponatremia . The useof powerful Naþ-wasting loop diuretic agents isnearly ubiquitous in ADHF, with 88% of patients inADHERE (Acute Decompensated Heart Failure Regis-try) receiving them (41). Moreover, 70% continue toreceive daily maintenance therapy with loop diureticagents, although many are probably not persistentlyvolume-overloaded. Loop diuretic agents block theNaþ/potassium/chloride cotransporter in the thickascending limb of Henle’s loop (TAL). Naþ and chlo-ride transport in the TAL, which has a relatively lowwater permeability, makes a crucial contribution tothe hypertonicity of the renal interstitium, in additionto the other important factor, medullary urea trans-port in the collecting ducts (42,43). This hypertonicity,washed out by blood flow through the vasa recta withinhibition of TAL transport by loop diuretic agents, isthe major driver of water reabsorption in the distalnephron. Because of this interference of loop diureticagents with the renal capacity to concentrate urine,less free water is reabsorbed, resulting in productionof hypotonic urine and, therefore, relative protectionagainst hyponatremia (Figure 4A) (44). However, inconditions of profound volume depletion with strongneurohumoral activation and compromised renalblood flow, loop diuretic agents will fail to elicitmeaningful water diuresis. Eventually, GFR and distalnephron flow will become depressed in the presenceof strongly up-regulated AVP, creating the necessaryenvironment for hyponatremia development (29).Other d iuret i c agents and hyponatremia . Miner-alocorticoid receptor antagonists (MRAs) are acornerstone in the treatment of heart failure withreduced ejection fraction and are promising inselected patients with preserved ejection fraction

FIGURE 3 Maximal Free Water Excretion in Healthy Subjects Versus Patients With Heart Failure and an Impaired GFR

Healthy personMAXIMAL WATER DIURESIS

Heart failure patientMAXIMAL WATER DIURESIS

GlomerulusGFR = 125 ml/min

= 180 l/day285 - 295 mOsm/l

Macula Densa18 l/day

150 mOsm/l

Distal Tubules(low water permeability)

Distal Tubules(low water permeability)

Collecting ducts(low water permeability)

Collecting ducts(low water permeability)

AVP suppressed:LOW WATER PERMEABILITYCONTINUED SOLUTEREABSORPTION

MAXIMALFREE WATER EXCRETION

~ 18l/day

Urine: 30-60 mOsm/l

MAXIMALFREE WATER EXCRETION

<1 l/day

Urine: 100-150 mOsm/l

HIGH BLOOD FLOW WASHOUT LOW GRADIENT

TAL25% reabsorption

= 45 l/day

TAL25% reabsorption

= 10.8 l/day

GlomerulusGFR = 30 ml/min

= 43.2 l/day285 - 295 mOsm/l

Macula Densa2.16 l/day

150 mOsm/l

Proximal Tubules65% reabsorption

=117 l/day285 - 295 mOsm/l

Proximal Tubules70% reabsorption

=30.24 l/day285 - 295 mOsm/l

VASOCONSTRICTIONLOW BLOOD FLOW NO WASHOUT HIGH GRADIENT

500 mOsm/l

700 mOsm/l

300 mOsm/l

300 mOsm/l

OSMOTIC

OSMOTIC

GRADIENT

GRADIENT

GRADIENT

GRADIENT

VASA

RECTA

VASA

RECTA

CONTINUED WATERREABSORPTION

AVP not suppressed:HIGH WATER PERMEABILITYSTRONG OSMOTIC GRADIENT

A

B

(A) Through glomerular filtration, a healthy person forms 180 l of tubular fluid each day. As the proximal tubules, thin descending limb, and

ascending limb of Henle’s loop all have high water permeability, reabsorption in these nephron segments is isotonic, and tonicity of the tubular

fluid remains at 285 to 295 mOsm/l, equal to plasma. In contrast, the TAL has low water permeability. Thus, solutes in general and sodium/

chloride in particular are reabsorbed without water, contributing to the hypertonicity of the renal interstitium. Consequently, osmolality is

relatively stable (w150 mOsm/l) in the tubular lumen at the level of the macula densa, at which point 90% of reabsorption has occurred, with

10%, or 18 l, remaining on a daily basis (25). In the case of total AVP suppression, the distal nephron has very low water permeability, while

sodium and chloride are continuously reabsorbed; therefore, urine can be diluted up to 30 to 60mOsm/l. Consequently, almost 18 l of free water

excretion can be achieved, which is themaximum a person can drink without developing plasma hypotonicity and hyponatremia. (B)Maximal free

water excretion in a patient with heart failure and a GFR of 30ml/min is considerably lower. In this case, 43.2 l of tubular fluid is formed each day.

As proximal tubular reabsorption increases in heart failure, only 5% or 2.16 l of tubular fluid remains at the level of the macula densa, again with

an osmolality of w150 mOsm/l (25). As AVP is typically increased in heart failure and the minimal medullar tonicity is greater because of poor

flow through the vasa recta, the osmotic gradient for water reabsorption is increased in the presence of a more leaky distal nephron. Thus, water

is continuously reabsorbed, limiting maximal free water excretion to <1 l. This realistic example illustrates how difficult it can be for a patient to

prevent hyponatremia progression with fluid restriction only. GFR ¼ glomerular filtration rate; other abbreviations as in Figure 1.

Verbrugge et al. J A C C V O L . 6 5 , N O . 5 , 2 0 1 5

Hyponatremia in Decompensated Heart Failure F E B R U A R Y 1 0 , 2 0 1 5 : 4 8 0 – 9 2

484

FIGURE 4 Effects of Diuretics on the Kidney’s Concentrating and Diluting Capacity

Effect of loop Diureticson the Concentrating Capacity of the Kidneys

MAXIMALANTIDIURESIS

MAXIMAL WITHLOOP DIURETICS

MAXIMALANTIDIURESIS

MAXIMAL WITHTHIAZIDE-TYPE

DIURETICS

DISTAL CONVOLUTED TUBES& COLLECTING DUCTS

IMPAIRED SODIUM REABSORPTION

UNCHANGED WATER REABSORPTIONThiazide-type diuretics & MRA

DISTAL CONVOLUTED TUBES& COLLECTING DUCTS

CONTINUED SODIUM REABSORPTIONWATER REABSORPTIONImpaired osmotic gradient

Na/Cl Co-transporter

Na/K/2CI Co-transporter

ENaC

Na/Cl Co-transporterNa/K/2CI Co-transporter

ENaC

LOOP DIURETICS

1200 mOsm/l

1200 mOsm/l 1200 mOsm/l

300 mOsm/l

300 mOsm/l 300 mOsm/l

300 mOsm/l

800 mOsm/l

THIAZIDE-TYPEDIURETICS

MRA

Effect of THIAZIDE-TYPE Diuretics and MRAon the Concentrating Capacity of the Kidneys

OSMOTIC

OSMOTIC

GRADIENT

GRADIENT

OSMOTIC

OSMOTIC

GRADIENT

GRADIENT

OSMOTIC

OSMOTIC

GRADIENT

GRADIENT

OSMOTIC

OSMOTIC

GRADIENT

GRADIENT

A

B

(A) Loop diuretic agents block the sodium/chloride/potassium cotransporter in the TAL, interfering with the generation of hypertonicity in the

renal interstitium and decreasing the osmotic gradient that promotes water reabsorption. Loop diuretic agents have no effect on sodium or

chloride transport in the distal nephron, which has low water permeability, and they therefore do not interfere with the kidneys’ capacity to

produce hypotonic urine. (B) Thiazide-type diuretic agents block the sodium/chloride symporter, whereas MRAs inhibit ENaCs in the distal

nephron. As both make a negligible contribution to the hypertonicity achieved in the renal interstitium, thiazide-type diuretic agents and MRAs

do not interfere with the water reabsorption gradient. However, as sodium and chloride reabsorption in the relatively water-impermeable distal

nephron is the mechanism of urinary dilution, this process is impaired, resulting in a higher urinary tonicity and, hence, lower free water

excretion compared with loop diuretic agents. ENaC ¼ epithelial sodium channel; MRA ¼ mineralocorticoid receptor antagonists; other

abbreviations as in Figure 1.

J A C C V O L . 6 5 , N O . 5 , 2 0 1 5 Verbrugge et al.F E B R U A R Y 1 0 , 2 0 1 5 : 4 8 0 – 9 2 Hyponatremia in Decompensated Heart Failure

485

Verbrugge et al. J A C C V O L . 6 5 , N O . 5 , 2 0 1 5

Hyponatremia in Decompensated Heart Failure F E B R U A R Y 1 0 , 2 0 1 5 : 4 8 0 – 9 2

486

(45,46). Together with thiazide-type diuretic agents,often recommended as first-line therapy in case ofloop diuretic resistance, they interfere with sodiumreabsorption by ENaCs and Naþ/chloride cotrans-porters, respectively, in the distal convoluted tubulesand collecting ducts (Figure 4B) (47). Thus, theyhave a direct effect on the diluting tubular segmentsof the kidneys and may cause hyponatremia, evenwithout pronounced hypovolemia, as they generateless hypotonic urine (47–49). From a pathophysio-logical perspective, it is, therefore, prudent toavoid their use in any patient with hyponatremia—atleast temporarily, until serum sodium levels arecorrected—and to prefer a proximally workingdiuretic (e.g., acetazolamide) in such cases with vol-ume overload and diuretic resistance (29).

Potass ium and magnes ium deplet ion . Both loop-and thiazide-type diuretic agents are a major cause ofpotassium andmagnesiumwasting in ADHF. This mayhave important implications, such as a higher risk ofarrhythmic death (50). In addition, potassium deple-tion shifts Naþ towards the intracellular compartmentto preserve cellular volume homeostasis, which maycontribute to development of hyponatremia (51).Indeed, intracellular Naþ concentrations are higher inmuscle cells of ADHF patients with hypokalemia (52).Furthermore, magnesium is critical for functioning ofthe Naþ/potassium ATPase that pumps Naþ out ofcells; thus, hypomagnesemia may also exacerbateextracellular Naþ depletion.

INITIAL APPROACH TO HYPONATREMIA IN

DECOMPENSATED HEART FAILURE

The Central Illustration presents a stepwise diagnosticand therapeutic approach to hyponatremia in de-compensated heart failure.

CONFIRM PLASMA HYPOTONICITY. In patients withhyponatremia, the first step is generally to assesswhether plasma hypotonicity is present by measuringthe plasma osmolality (reference value: 285 to295 mOsm/l). This may not be necessary whenhyponatremia is mild (serum Naþ concentration130 to 134 mEq), asymptomatic, and easily correctedby empiric treatment. Elevated triglyceride orcholesterol levels, immunoglobulins, and monoclonalgammopathies—through laboratory artifacts—may allcause falsely low serum Naþ concentrations withoutaffecting plasma osmolality (53–55). Additionally,the presence of effective osmoles raising serumosmolality—most notably glucose in uncontrolleddiabetes and/or hyperosmolar radiocontrast media—may result in hyponatremia with normal, or evenhigh, plasma tonicity (56,57). Indeed, the serum Naþ

concentration decreases by approximately 2.4 mEq/lper 100 mg/dl increase in serum glucose (56). If theseunderlying causes are managed appropriately, hypo-natremia is probably not associated with worseoutcome (58).

AVOID AND TREAT DEPLETIONAL HYPONATREMIA.

In any patient with hypotonic hyponatremia, it is bestto avoid thiazide-type diuretic agents, MRAs, andENaC blockers (e.g., amiloride), as these medicationsinterfere directly with the kidneys’ capacity to pro-duce hypotonic urine. However, it should be stressedthat this advice is purely on the basis of the patho-physiological rationale, as high-quality evidence islacking. Further, potassium and magnesium storesshould be replenished if serum levels are low. Wepropose aiming for serum potassium levels $4 mEq/land magnesium levels $1.7 mEq/l ($2 mg/dl) in ADHFpatients presenting with hyponatremia, but this cut-off is arguably somewhat arbitrary.

DIFFERENTIATE BETWEEN DEPLETIONAL AND

DILUTIONAL HYPONATREMIA. In ADHF, cliniciansshould differentiate between depletional hypo-natremia, requiring the administration of saline,versus dilutional hyponatremia, where free waterexcretion should be promoted. Acute gastrointestinalor third-space losses, clinical signs of hypovolemia,and recent use of diuretic agents—especially at highdoses or in combination therapy—increase the likeli-hood of depletional hyponatremia. In this case, pro-duction of hypotonic urine is promoted by theadministration of isotonic saline, with subsequentnormalization of serum sodium levels. In contrast, itwould be impossible to correct dilutional hypona-tremia with isotonic saline, as free water excretion iscompromised and hypotonic urine production isimpaired. One might consider a fluid challenge with 1of isotonic saline over 24 h to differentiate betweenthese conditions, measuring the effect on serum so-dium levels. However, this should be avoided in caseof clear fluid overload and/or severe hyponatremia(serum sodium <125 mEq/l). Indeed, signs of volumeoverload indicate a component of dilutional hypona-tremia, in which case an improvement is unlikely,and the risk of further deteriorating congestion issubstantial. Further worsening of hyponatremiashould certainly be avoided in patients with severehyponatremia. Alternatively, one could measureurine osmolality, which should be adequately sup-pressed (<100 mOsm/l) in patients with depletionalhyponatremia, but not in patients with dilutionalhyponatremia. If urine osmolality is >150 mOsm/l,isotonic solutions should certainly be avoided, asadministration would result in further worsening of

CENTRAL ILLUSTRATION Approach to Hyponatremia in Decompensated Heart Failure

Assess plasma osmolality (can be skipped in cases of mild hyponatremia that is easily corrected with initiation of therapy)

Differentiate between hyponatremia conditions Water Excess (dilutional)Na+ deficit (depletional)

Pseudohyponatremia(Plasma osmolality ≥285 mOsm/L)Falsely low serum Na+ concentrations caused by laboratory artefacts (elevated triglyceride/cholesterol levels, immunoglobulins and monoclonal gammopathies)Increased plasma osmolality caused by hyperglycemia, hyperosmolar radio-contrast media, use of mannitol, ethanol, methanol, ethylene glycol

Hypotonic hyponatremia(Plasma osmolality <285 mOsm/L)

Depletional hyponatremia• Negative Na+ balance• Potassium (K+) and magnesium (Mg++) depletion

Clinical presentation:• Hypovolemia

Caused by:• Use of powerful Na+-wasting loop diuretics• Use of thiazide-type or combinational diuretics• Salt-restricted diets recommended by the guidelines• Acute gastro-intestinal or third-space losses

Urinary analysis:• Urinary osmolality adequately suppressed (<100 mOsm/L)• Urinary sodium depleted (<50 mEq/L)

Dilutional hyponatremia• Impaired water excretion• Excessive water reabsorption

Clinical presentation:• Hypervolemia (Edema, ascites, pleural effusion)

Caused by:• Non-osmotic release of arginine vasopressin (AVP)• Insufficient tubular flow through diluting segments

of the distal nephron

Urinary analysis:• Urinary osmolality inadequately suppressed (≥100 mOsm/L)

Depletional hyponatremia• Administer saline. One liter of infusate is expected

to change the serum sodium concentration with:

Dilutional hyponatremia• Limit water intake• Promote free water excretion

- AVP- Improve distal nephron flow (loop diuretics with

or without hypertonic saline, acetazolamide, renin-angiotensin system blockers, inotropesand vasodilator therapy)

GENE

RAL

RECO

MM

ENDA

TION

S

Hyponatremia: Serum sodium (Na+) ≤135 mEq/L

For both depletional and dilutional hyponatremia• Stop distally working diuretics (thiazide-type, amiloride, mineralocorticoid receptor antagonists)• Correction of K+ and Mg++ deficiencies (aiming for serum K+ levels ≥4 mEq/L and Mg++ levels ≥1.7 mEq/L)

DIAG

NOSI

SDE

FINI

TION

(Na+)infusate + (K+)infusate – (Na+)serum

(TBW + 1)

TBW = α x body weight (kg); α = 0.6 in children and non-elderly men,0.5 in non-elderly women and elderly men, 0.45 in elderly women

Verbrugge, F.H. et al. J Am Coll Cardiol. 2015; 65(5):480–92.

A flowchart with a stepwise diagnostic approach to hyponatremia in decompensated heart failure is presented. Therapeutic options are

proposed according to the underlying pathophysiological culprit. AVP ¼ arginine vasopressin.

J A C C V O L . 6 5 , N O . 5 , 2 0 1 5 Verbrugge et al.F E B R U A R Y 1 0 , 2 0 1 5 : 4 8 0 – 9 2 Hyponatremia in Decompensated Heart Failure

487

Verbrugge et al. J A C C V O L . 6 5 , N O . 5 , 2 0 1 5

Hyponatremia in Decompensated Heart Failure F E B R U A R Y 1 0 , 2 0 1 5 : 4 8 0 – 9 2

488

hyponatremia. Finally, very low urinary Naþ and/orchloride concentrations (<50 mEq/l) are a relativelystrong argument for electrolyte depletion.

TREATMENT OF DEPLETIONAL

HYPONATREMIA

As with other deficiencies, pure depletional hypo-natremia is easily treated by administration of saline.Hypertonic saline will correct hyponatremia faster andwith a lower water load than isotonic saline, whichmight be preferred in patients who are already nor-movolemic on clinical examination. However, if nosevere hyponatremia symptoms are present, it is rec-ommended to correct the serum Naþ concentrationslowly, at a maximal rate of 5 mEq/l per day. If hypo-natremia is profound (<125 mEq/l), correction up to10 mEq/l per day is acceptable (59). At any time,increasing the serum Naþ concentration >10 mEq/l in24 h should be avoided because of the risk of centralpontinemyelinolysis (60). Importantly, replenishmentof potassium and magnesium stores through adminis-tration of supplements contributes to hyponatremiacorrection, and this should be taken into account whendetermining the correction rate (52,61). Althoughcertainly not a substitute for frequent monitoring, thechange in serum Naþ concentration with 1 l of infusatemight be approximated by the following formula:

��Naþ

�INFUSATE þ �

Kþ�INFUSATE � �

Naþ�SERUM

� ðTBW þ 1Þ�

with TBW ¼ a � body weight (kg), and a ¼ 0.6 in chil-dren and nonelderly men, 0.5 in nonelderly womenand elderly men, and 0.45 in elderly women (62). Takefor example a female patient, 50 years of age andweighing 62 kg, presents with a serum Naþ concen-tration of 130mEq/l and a serumKþof 3.5mEq/l. This ispresumably because of depletion, as she is on a highmaintenance dose of furosemide (120 mg twice daily)and has no overt signs of volume overload. Infusion of1 normal saline ([Naþ] ¼ 154 mEq/l) with addition of40 mEq Kþ supplements would be expected to raisethe serum Naþ concentration by 2 mEq/l ([154 mEq/l þ40 mEq/l � 130 mEq/l]/[(0.5 � 62) þ 1]). However,without Kþ supplements, the increase would only be0.75 mEq/l ([154 mEq/l � 130 mEq/l]/[(0.5 � 62) þ 1]).

TREATMENT OF HYPOTONIC

DILUTIONAL HYPONATREMIA

Acute treatment of hypotonic dilutional hypo-natremia is focused on promoting free water excretionto restore normal serum sodium levels. However, forlong-term treatment success, it is important to subse-quently prevent a positive free water balance. This

may be important, as the question whether hypona-tremia in ADHF reflects a therapeutic target or ismerely a marker of advanced disease has not beenfully established. Indeed, studies looking at theprognostic effect of hyponatremia correction haveyielded conflicting results (63,64). Madan et al. (63)demonstrated in a cohort of 322 ADHF patients withhyponatremia that long-term changes in serum Naþ

concentration (assessed 60 to 270 days after hospitaldischarge) are significant predictors of mortality, withimproving hyponatremia associated with betteroutcome. In contrast, after correction for diseaseseverity markers, Lee et al. (64) found that normali-zation of hyponatremia at hospital discharge was notassociated with improved survival (64). Obviously,confounding by different treatment strategies in bothstudies cannot be excluded, but a conservativeinterpretation is that only durable hyponatremiacorrection is potentially associated with improvedoutcome in ADHF. To achieve this, good heart failuremanagement is often fundamental.

ACUTE TREATMENT OF DILUTIONAL HYPONATREMIA.

To promote free water excretion in a patient withADHF and hypotonic dilutional hyponatremia, distalnephron flow should be increased, AVP levels shouldbe lowered, or AVP effects should be antagonized.Loop diuret i c agents with or without hypertonicsa l ine . The powerful inhibitory effect of loopdiuretic agents on Naþ transport in the TAL increasesthe amount of tubular fluid presented to the distalnephron. Loop diuretic agents also reduce hyperto-nicity of the renal interstitium, further facilitatingwater excretion (Figure 4A) (44). Because loopdiuretic agents are cheap and readily available, theyremain the first-line therapy in ADHF with dilutionalhyponatremia and volume overload. The addition ofhypertonic saline to improve loop diuretic efficacy inADHF is a controversial issue. Although counterin-tuitive from a pathophysiological point of view, somesmall studies have suggested more efficient decon-gestion and better renal preservation when loopdiuretic agents are combined with hypertonic saline(Table 2) (65–70). Importantly, decreases in plasmarenin activity, inflammatory markers, and evennatriuretic peptide levels have been demonstratedwith hypertonic saline administration in ADHFpatients who receive loop diuretic agents (71,72). Still,it remains difficult to draw any firm conclusions,because the use of high doses of loop diuretic agentsthat might have induced these alterations mighthave confounded these improvements with sodiumloading. As serum Naþ levels are more easily cor-rected, patients with hyponatremia might benefitmore from the addition of hypertonic saline to loop

TABLE 2 Studies on Hypertonic Saline in Patients With Acute Decompensated

Heart Failure

First Author (Ref. #) n Treatment Outcome

Paterna et al. (65) 60 IV furosemide 500–1,000 mgwith vs. without 150 ml1.4%–4.6% hypertonicsaline BID

Increase in diuresis, natriuresis,and serum sodium levels;decrease in serum creatinine;and shorter hospitalizationtime with hypertonic saline

Licata et al. (66) 107 IV furosemide 500–1,000 mgwith vs. without 150 ml1.4%–4.6% hypertonicsaline BID

Increase in diuresis, natriuresis,and serum sodium levels;decrease in serum creatinine;and improved survival withhypertonic saline

Paterna et al. (71) 94 IV furosemide 500–1,000 mgwith vs. without 150 ml1.4%–4.6% hypertonicsaline BID

Increase in diuresis andnatriuresis; decrease in BNPlevels; shorter hospitalizationtime; and lower 30-dayreadmission rate withhypertonic saline

Parrinello et al. (67) 133 IV furosemide 250 mg plus150 ml 3% hypertonicsaline BID vs. IV furosemide250 mg BID plus lowsodium diet (<80 mmol)

Increase in diuresis, natriuresis,and serum sodium levels;improved renal function; andfaster reduction of echo-PCWP with hypertonic saline

Paterna et al. (68) 1,771 IV furosemide 250 mg plus150 ml 3% hypertonicsaline BID vs. IV furosemide250 mg BID plus lowsodium diet (<80 mmol)

Increase in diuresis, natriuresis,and serum sodium levels;decrease in serum creatinine;shorter hospitalization time;lower readmission rate; andimproved survival withhypertonic saline

Issa et al. (69) 34 100 ml 7.5% hypertonic salineBID vs. placebo for 3 days

Improved in glomerular andtubular biomarkers withhypertonic saline

Okuhara et al. (70) 44 500 ml 1.7% hypertonic salinevs. glucose 5% with 40 mgfurosemide

Improved GFR and better diuresiswith hypertonic saline

BID ¼ twice daily; BNP ¼ B-type natriuretic peptide; GFR ¼ glomerular filtration rate; IV ¼ intravenous;PCWP ¼ pulmonary capillary wedge pressure.

J A C C V O L . 6 5 , N O . 5 , 2 0 1 5 Verbrugge et al.F E B R U A R Y 1 0 , 2 0 1 5 : 4 8 0 – 9 2 Hyponatremia in Decompensated Heart Failure

489

diuretic agents; however, this remains highlyspeculative.Acetazolamide . Alternatively, combinational diu-retic treatment with loop diuretic agents and acet-azolamide ensures minimal tubular Naþ reabsorptionproximal from the macula densa and maximal flowthrough the distal nephron (29). For this reason, ifcombination therapy is needed to overcome loopdiuretic resistance in ADHF, acetazolamide should bepreferred over thiazide-type diuretic agents, MRAs, orENaC blockers.AVP antagon is ts . AVP antagonists are the onlymedication class to directly promote free waterexcretion by prevention of aquaporin-2 channelavailability in the collecting ducts of the nephron(73,74). Three oral V2 receptor antagonists (tolvaptan,satavaptan, and lixivaptan) have been tested forADHF, with the former 2 shown to be efficacious inrestoring serum Naþ levels in patients with volumeoverload and hyponatremia (75–78). Similar data areavailable for conivaptan, an intravenous agent thatantagonizes both the V2 and V1a receptors (74,79,80).Table 3 summarizes the currently available evidenceon AVP antagonists in patients with ADHF and hypo-natremia. Importantly, EVEREST (Efficacy of Vaso-pressin Antagonism in Heart Failure Outcome Studywith Tolvaptan), including 4,133 patients with ADHFbut without hyponatremia as an inclusion criterion,compared tolvaptan with placebo and was poweredfor clinical endpoint analysis (81). The overall trial didnot show a significant reduction in all-cause mortalityor readmission rates, but, interestingly, a subanalysisin patients presenting with pronounced hypona-tremia (<130 mmol/l) suggested improved survivalfree from cardiovascular death or readmission (76,81).This promising finding warrants further study in anadequately powered, randomized clinical trial.

LONG-TERM MANAGEMENT OF HYPOTONIC DILU-

TIONAL HYPONATREMIA. Water rest r i c t ion . Waterrestriction is uniformly recommended by the guide-lines in ADHF patients with hyponatremia as it is inothers with dilutional hypotonic hyponatremia (35–37,59). Whilst limiting free water intake certainlyaids to prevent a positive free water balance, few datasupport its long-term efficacy, and adherence mightbe a problem. Indeed, thirst is a frequent, under-recognized, and distressing problem in heart failure,and many patients may find it difficult to comply withstrict water restriction (<1 l) (82). Nevertheless, arecent study found that ADHF patients with mildhyponatremia had improvements in quality of lifewith such a strategy (83).AVP antagonists . Although AVP antagonists arehighly efficacious for short-term correction of

hyponatremia, less is known about their long-termefficacy in providing a sustained correction of serumNaþ levels. In EVEREST, serum Naþ concentrationincreased more with tolvaptan compared with pla-cebo during in-hospital stay (81). However, patientsalso reported feeling thirsty significantly morefrequently, and the difference with serum Naþ levelsin the placebo group disappeared after instillation ofoutpatient management, which suggests that patientsmay have drank more to compensate for increasedwater losses.Renin-angiotensin system blockers. Renin-angiotensinsystem blockers are known to increase renal bloodflow and decrease proximal tubular sodium reab-sorption. Therefore, it is not surprising that they wereamong the first agents demonstrated to be effective inhypotonic dilutional hyponatremia in heart failure(84,85). Up-titration of renin-angiotensin systemblockers should always be promoted for this indica-tion if there are no contraindications, such as coex-isting renal dysfunction.

TABLE 3 Studies on AVP Antagonists in Patients With Acute Decompensated Heart

Failure and Hyponatremia

First Author (Ref. #) n/N Treatment Outcome

Gheorghiade et al. (77) 71/254(subgroup

analysis)

Tolvaptan 30, 45, or60 mg vs. placebo

Normalization of serumsodium after 24 h,greater decrease in bodyweight and edema,increased urine outputwith tolvaptan

Gheorghiade et al. (78) 51/319(subgroup

analysis)

Tolvaptan 30, 60, or90 mg vs. placebo

Normalization of serumsodium with tolvaptan

Ghali et al. (79) 19/74(subgroup

analysis)

Conivaptan 40 or80 mg vs. placebo

Normalization of serumsodium with conivaptan

Zeltser et al. (80) 28/84(subgroup

analysis)

Conivaptan 40 or80 mg vs. placebo

Increase in serum sodiumconcentration withconivaptan

Konstam et al. (81) 1,157/4,133(subgroup

analysis)

Tolvaptan 30 mg vs.placebo

No effect on mortality orrehospitalization,significant increase inserum sodium withtolvaptan

Aronson et al. (75) 90/118(subgroup

analysis)

Satavaptan 25 or 50 mgvs. placebo

Increase in serum sodiumconcentration withsatavaptan

Verbrugge et al. J A C C V O L . 6 5 , N O . 5 , 2 0 1 5

Hyponatremia in Decompensated Heart Failure F E B R U A R Y 1 0 , 2 0 1 5 : 4 8 0 – 9 2

490

Inotropes and vasodi la tor therapy . Increasing theeffective circulatory volume in heart failure is ex-pected to result in less nonosmotic AVP releaseand better renal blood flow, which, from a patho-physiological perspective, might help to correct dilu-tional hyponatremia. Because cardiac output in ADHFis rather insensitive to changes in cardiac pre-load—asthe Frank-Starling mechanism is depleted—improve-ments can only be obtained through direct stimula-tion of contractility with inotropes or reducingafterload with vasodilator therapy. As the formertreatment strategy is associated with increased

mortality, afterload reduction probably remains thebest option to increase the effective circulatory vol-ume; however, its use is limited by low arterial bloodpressure (86,87). Observational data have demon-strated that nitroprusside, titrated on arterial bloodpressure and with conversion to oral hydralazine andnitrates, is feasible and potentially associated withbetter outcomes in ADHF (88,89). Alternatively, ser-elaxin, a vasodilator agent under investigation, mightbe of particular interest because of its specific renalvasodilator properties (90,91). The RELAX-AHF-EU(Effect of Serelaxin Versus Standard of Care in AcuteHeart Failure Patients) trial is currently enrollingADHF patients and is powered for clinical endpointevaluation (92). Nevertheless, the specific effects ofboth inotropes and vasodilator therapy on hypo-natremia remain insufficiently elucidated.

CONCLUSIONS

The pathophysiology of hyponatremia in ADHF iscomplex, and a 1-size-fits-all approach is thereforelikely to fail. Appropriate differentiation betweendilutional and depletional hyponatremia is crucialand depends on good history taking, clinical exami-nation, and correct interpretation of laboratory re-sults. A targeted, pathophysiology-based approachshould help to treat this challenging condition effi-ciently, thereby minimizing adverse events.

REPRINT REQUESTS AND CORRESPONDENCE: Dr.Wilfried Mullens, Department of Cardiology, ZiekenhuisOost-Limburg, Schiepse Bos 6, 3600 Genk, Belgium.E-mail: wilfried.mullens@zol.be.

RE F E RENCE S

1. Mohan S, Gu S, Parikh A, et al. Prevalence ofhyponatremia and association with mortality:results from NHANES. Am J Med 2013;126:1127–37.e1.

2. Wald R, Jaber BL, Price LL, et al. Impact ofhospital-associated hyponatremia on selectedoutcomes. Arch Intern Med 2010;170:294–302.

3. Gheorghiade M, Abraham WT, Albert NM, et al.,OPTIMIZE-HF Investigators and Coordinators.Relationship between admission serum sodiumconcentration and clinical outcomes in patientshospitalized for heart failure: an analysis from theOPTIMIZE-HF registry. EurHeart J 2007;28:980–8.

4. Konishi M, Haraguchi G, Ohigashi H, et al.Progression of hyponatremia is associated withincreased cardiac mortality in patients hospitalizedfor acute decompensated heart failure. J Card Fail2012;18:620–5.

5. Shchekochikhin DY, Schrier RW, Lindenfeld J,et al. Outcome differences in community- versus

hospital-acquired hyponatremia in patients with adiagnosis of heart failure. Circ Heart Fail 2013;6:379–86.

6. Adrogue HJ, Madias NE. Hyponatremia. N EnglJ Med 2000;342:1581–9.

7. Oren RM. Hyponatremia in congestive heartfailure. Am J Cardiol 2005;95:2B–7B.

8. Oliet SH, Bourque CW. Mechanosensitivechannels transduce osmosensitivity in supraopticneurons. Nature 1993;364:341–3.

9. Bankir L. Antidiuretic action of vasopressin:quantitative aspects and interaction between V1aand V2 receptor-mediated effects. Cardiovasc Res2001;51:372–90.

10. Knepper MA. Molecular physiology of urinaryconcentrating mechanism: regulation of aquaporinwater channels by vasopressin. Am J Physiol 1997;272:F3–12.

11. Andersen LJ, Andersen JL, Schutten HJ, et al.Antidiuretic effect of subnormal levels of arginine

vasopressin in normal humans. Am J Physiol 1990;259 1 Pt 2:R53–60.

12. Star RA, Nonoguchi H, Balaban R, et al.Calcium and cyclic adenosine monophosphate assecond messengers for vasopressin in the rat innermedullary collecting duct. J Clin Invest 1988;81:1879–88.

13. Park F, Mattson DL, Skelton MM, et al. Local-ization of the vasopressin V1a and V2 receptorswithin the renal cortical and medullary circulation.Am J Physiol 1997;273 1 Pt 2:R243–51.

14. Breyer MD, Ando Y. Hormonal signaling andregulation of salt and water transport in the col-lecting duct. Annu Rev Physiol 1994;56:711–39.

15. Goldsmith SR, Francis GS, Cowley AW Jr.Arginine vasopressin and the renal response towater loading in congestive heart failure. Am JCardiol 1986;58:295–9.

16. Szatalowicz VL, Arnold PE, Chaimovitz C, et al.Radioimmunoassay of plasma arginine vasopressin

J A C C V O L . 6 5 , N O . 5 , 2 0 1 5 Verbrugge et al.F E B R U A R Y 1 0 , 2 0 1 5 : 4 8 0 – 9 2 Hyponatremia in Decompensated Heart Failure

491

in hyponatremic patients with congestive heartfailure. N Engl J Med 1981;305:263–6.

17. Goldsmith SR, Francis GS, Cowley AW Jr., et al.Increased plasma arginine vasopressin levels inpatients with congestive heart failure. J Am CollCardiol 1983;1:1385–90.

18. Francis GS, Benedict C, Johnstone DE, et al.Comparison of neuroendocrine activation inpatients with left ventricular dysfunction with andwithout congestive heart failure. A substudy of theStudies of Left Ventricular Dysfunction (SOLVD).Circulation 1990;82:1724–9.

19. Xu DL, Martin PY, Ohara M, et al. Upregulationof aquaporin-2 water channel expression inchronic heart failure rat. J Clin Investigation 1997;99:1500–5.

20. Robertson GL, Aycinena P, Zerbe RL. Neuro-genic disorders of osmoregulation. Am J Med1982;72:339–53.

21. Uretsky BF, Verbalis JG, Generalovich T, et al.Plasma vasopressin response to osmotic andhemodynamic stimuli in heart failure. Am J Physiol1985;248 3 Pt 2:H396–402.

22. Dunn FL, Brennan TJ, Nelson AE, et al. Therole of blood osmolality and volume in regulatingvasopressin secretion in the rat. Journal Clin Invest1973;52:3212–9.

23. Creager MA, Faxon DP, Cutler SS, et al.Contribution of vasopressin to vasoconstriction inpatients with congestive heart failure: comparisonwith the renin-angiotensin system and the sym-pathetic nervous system. J Am Coll Cardiol 1986;7:758–65.

24. Jamison RL, Oliver RE. Disorders of urinaryconcentration and dilution. Am J Med 1982;72:308–22.

25. Gottschalk CW. Micropuncture studies oftubular function in the mammalian kidney. FoliaMed Neerl 1962;5:11–30.

26. Leviel F, Hubner CA, Houillier P, et al.The Naþ-dependent chloride-bicarbonate ex-changer SLC4A8 mediates an electroneutralNaþ reabsorption process in the renal corticalcollecting ducts of mice. J Clin Invest 2010;120:1627–35.

27. Satlin LM, Carattino MD, Liu W, et al. Regula-tion of cation transport in the distal nephron bymechanical forces. Am J Physiol Renal Physiol2006;291:F923–31.

28. Gibson DG, Marshall JC, Lockey E. Assessmentof proximal tubular sodium reabsorption duringwater diuresis in patients with heart disease. BrHeart J 1970;32:399–405.

29. Verbrugge FH, Dupont M, Steels P, et al. Thekidney in congestive heart failure: ’are natriuresis,sodium, and diuretics really the good, the bad andthe ugly?’. Eur J Heart Fail 2014;16:133–42.

30. Bell NH, Schedl HP, Bartter FC. An explanationfor abnormal water retention and hypoosmolalityin congestive heart failure. Am J Med 1964;36:351–60.

31. Schedl HP, Bartter FC. An explanation forand experimental correction of the abnormalwater diuresis in cirrhosis. J Clin Invest 1960;39:248–61.

32. Schrier RW, Berl T, Anderson RJ. Osmotic andnonosmotic control of vasopressin release. Am JPhysiol 1979;236:F321–32.

33. Schrier RW, Abraham WT. Hormones andhemodynamics in heart failure. N Engl J Med1999;341:577–85.

34. McKie PM, Schirger JA, Costello-Boerrigter LC,et al. Impaired natriuretic and renal endocrineresponse to acute volume expansion in pre-clinicalsystolic and diastolic dysfunction. J Am Coll Car-diol 2011;58:2095–103.

35. Lindenfeld J, Albert NM, Boehmer JP, et al.,Heart Failure Society of America. HFSA 2010comprehensive heart failure practice guideline.J Card Fail 2010;16:e1–194.

36. McMurray JJ, Adamopoulos S, Anker SD, et al.,ESC Committee for Practice Guidelines. ESCguidelines for the diagnosis and treatment of acuteand chronic heart failure 2012: The Task Force forthe Diagnosis and Treatment of Acute and ChronicHeart Failure 2012 of the European Society of Car-diology. Eur J Heart Fail 2012;14:803–69.

37. Yancy CW, Jessup M, Bozkurt B, et al. 2013ACCF/AHA guideline for the management of heartfailure: a report of the American College of Car-diology Foundation/American Heart AssociationTask Force on Practice Guidelines. J Am CollCardiol 2013;62:e147–239.

38. Paterna S, Gaspare P, Fasullo S, Sarullo FM, DiPasquale P. Normal-sodium diet compared withlow-sodium diet in compensated congestive heartfailure: is sodium an old enemy or a new friend?Clin Sci (Lond) 2008;114:221–30.

39. Paterna S, Parrinello G, Cannizzaro S, et al.Medium term effects of different dosage ofdiuretic, sodium, and fluid administration onneurohormonal and clinical outcome in patientswith recently compensated heart failure. Am JCardiol 2009;103:93–102.

40. Parrinello G, Di Pasquale P, Licata G, et al.Long-term effects of dietary sodium intake oncytokines and neurohormonal activation in pa-tients with recently compensated congestive heartfailure. J Card Fail 2009;15:864–73.

41. Fonarow GC, Corday E, ADHERE ScientificAdvisory Committee. Overview of acutely decom-pensated congestive heart failure (ADHF): a reportfrom the ADHERE registry. Heart Fail Rev 2004;9:179–85.

42. Lassiter WE, Gottschalk CW, Mylle M. Micro-puncture study of net transtubular movementof water and urea in nondiuretic mammaliankidney. Am J Physiol 1961;200:1139–47.

43. Gottschalk CW, Lassiter WE, Mylle M, et al.Micropuncture study of composition of loopof Henle fluid in desert rodents. Am J Physiol1963;204:532–5.

44. Ali SS, Olinger CC, Sobotka PA, et al., Ran-domized Aldactone Evaluation Study In-vestigators. Loop diuretics can cause clinicalnatriuretic failure: a prescription for volumeexpansion. Congest Heart Fail 2009;15:1–4.

45. Pitt B, Zannad F, Remme WJ, et al. The effectof spironolactone on morbidity and mortality inpatients with severe heart failure. N Engl J Med1999;341:709–17.

46. Zannad F, McMurray JJ, Krum H, et al.,EMPHASIS-HF Study Group. Eplerenone in pa-tients with systolic heart failure and mild symp-toms. N Engl J Med 2011;364:11–21.

47. Jentzer JC, DeWald TA, Hernandez AF.Combination of loop diuretics with thiazide-typediuretics in heart failure. J Am Coll Cardiol 2010;56:1527–34.

48. Bozkurt B, Agoston I, Knowlton AA. Com-plications of inappropriate use of spironolactonein heart failure: when an old medicine spirals outof new guidelines. J Am Coll Cardiol 2003;41:211–4.

49. Goland S, Naugolny V, Korbut Z, et al.Appropriateness and complications of the use ofspironolactone in patients treated in a heart failureclinic. Eur J Intern Med 2011;22:424–7.

50. Cooper HA, Dries DL, Davis CE, et al. Diureticsand risk of arrhythmic death in patients with leftventricular dysfunction. Circulation 1999;100:1311–5.

51. Cooke RE, Segar WE, Cheek DB, et al. Theextrarenal correction of alkalosis associated withpotassium deficiency. J Clin Invest 1952;31:798–805.

52. Dyckner T, Wester PO. Effects of magnesiuminfusions in diuretic induced hyponatraemia.Lancet 1981;1:585–6.

53. Vaswani SK, Sprague R. Pseudohyponatremiain multiple myeloma. South Med J 1993;86:251–2.

54. Lawn N, Wijdicks EF, Burritt MF. Intravenousimmune globulin and pseudohyponatremia. N EnglJ Med 1998;339:632.

55. Lai MY, Lin CC, Chung SL, et al. Milky plasma,diabetes, and severe hyponatremia. Kidney Int2009;75:996.

56. Hillier TA, Abbott RD, Barrett EJ. Hypona-tremia: evaluating the correction factor for hy-perglycemia. Am J Med 1999;106:399–403.

57. Sirken G, Raja R, Garces J, et al. Contrast-induced translocational hyponatremia and hyper-kalemia in advanced kidney disease. Am J KidneyDis 2004;43:e31–5.

58. Milo-Cotter O, Cotter G, Weatherley BD, et al.Hyponatraemia in acute heart failure is a marker ofincreased mortality but not when associated withhyperglycaemia. Eur J Heart Fail 2008;10:196–200.

59. Spasovski G, Vanholder R, Allolio B, et al.Clinical practice guideline on diagnosis and treat-ment of hyponatraemia. Intensive Care Med 2014;40:320–31.

60. Pirzada NA, Ali II. Central pontine myelinol-ysis. Mayo Clin Proc 2001;76:559–62.

61. Kingston M, Bhuta S. Hazards of potassiumand multiple sources of sodium in causing osmoticdemyelination. Intern Med J 2012;42:730–1.

62. Adrogue HJ, Madias NE. Hypernatremia. NEngl J Med 2000;342:1493–9.

63. Madan VD, Novak E, Rich MW. Impact ofchange in serum sodium concentration on mor-tality in patients hospitalized with heart failureand hyponatremia. Circ Heart Fail 2011;4:637–43.

64. Lee SE, Choi DJ, Yoon CH, et al., KorHF Reg-istry. Improvement of hyponatraemia during

Verbrugge et al. J A C C V O L . 6 5 , N O . 5 , 2 0 1 5

Hyponatremia in Decompensated Heart Failure F E B R U A R Y 1 0 , 2 0 1 5 : 4 8 0 – 9 2

492

hospitalisation for acute heart failure is not asso-ciated with improvement of prognosis: an analysisfrom the Korean Heart Failure (KorHF) registry.Heart 2012;98:1798–804.

65. Paterna S, Di Pasquale P, Parrinello G, et al.Effects of high-dose furosemide and small-volumehypertonic saline solution infusion in comparisonwith a high dose of furosemide as a bolus, in re-fractory congestive heart failure. Eur J Heart Fail2000;2:305–13.

66. Licata G, Di Pasquale P, Parrinello G, et al.Effects of high-dose furosemide and small-volumehypertonic saline solution infusion in comparisonwith a high dose of furosemide as bolus in re-fractory congestive heart failure: long-term ef-fects. Am Heart J 2003;145:459–66.

67. Parrinello G, Paterna S, Di Pasquale P, et al.Changes in estimating echocardiography pulmonarycapillary wedge pressure after hypersaline plusfurosemide versus furosemide alone in decom-pensated heart failure. J Card Fail 2011;17:331–9.

68. Paterna S, Fasullo S, Parrinello G, et al. Short-term effects of hypertonic saline solution in acuteheart failure and long-term effects of a moderatesodium restriction in patients with compensatedheart failure with New York Heart Associationclass III (Class C) (SMAC-HF Study). Am J Med Sci2011;342:27–37.

69. Issa VS, Andrade L, Ayub-Ferreira SM, et al.Hypertonic saline solution for prevention of renaldysfunction in patients with decompensated heartfailure. Int J Cardiol 2013;167:34–40.

70. Okuhara Y, Hirotani S, Naito Y, et al. Intrave-nous salt supplementation with low-dose furose-mide for treatment of acute decompensated heartfailure. J Card Fail 2014;20:295–301.

71. Paterna S, Di Pasquale P, Parrinello G, et al.Changes in brain natriuretic peptide levels andbioelectrical impedance measurements aftertreatment with high-dose furosemide and hyper-tonic saline solution versus high-dose furosemidealone in refractory congestive heart failure: adouble-blind study. J Am Coll Cardiol 2005;45:1997–2003.

72. Tuttolomondo A, Pinto A, Di Raimondo D,et al. Changes in natriuretic peptide and cytokineplasma levels in patients with heart failure, aftertreatment with high dose of furosemide plushypertonic saline solution (HSS) and after a salineloading. Nutr Metab Cardiovasc Dis 2011;21:372–9.

73. Finley JJ 4th, Konstam MA, Udelson JE. Argi-nine vasopressin antagonists for the treatment of

heart failure and hyponatremia. Circulation 2008;118:410–21.

74. Chen SY, Abrahams Z, Sokos GG, et al. Con-ivaptan: potential therapeutic implications in heartfailure. Recent Pat Cardiovasc Drug Discov 2008;3:137–40.

75. Aronson D, Verbalis JG, Mueller M, et al.,DILIPO Investigators. Short- and long-term treat-ment of dilutional hyponatraemia with satavaptan,a selective arginine vasopressin V2-receptorantagonist: the DILIPO study. Eur J Heart Fail2011;13:327–36.

76. Hauptman PJ, Burnett J, Gheorghiade M,et al., EVEREST Investigators. Clinical course ofpatients with hyponatremia and decompensatedsystolic heart failure and the effect of vasopressinreceptor antagonism with tolvaptan. J Card Fail2013;19:390–7.

77. Gheorghiade M, Niazi I, Ouyang J, et al.,Tolvaptan Investigators. Vasopressin V2-receptorblockade with tolvaptan in patients with chronicheart failure: results from a double-blind, ran-domized trial. Circulation 2003;107:2690–6.

78. GheorghiadeM, GattisWA, O’Connor CM, et al.,Acute and Chronic Therapeutic Impact of a Vaso-pressin Antagonist in Congestive Heart Failure(ACTIV in CHF) Investigators. Effects of tolvaptan, avasopressin antagonist, in patients hospitalizedwith worsening heart failure: a randomizedcontrolled trial. JAMA 2004;291:1963–71.

79. Ghali JK, Koren MJ, Taylor JR, et al. Efficacyand safety of oral conivaptan: a V1A/V2 vaso-pressin receptor antagonist, assessed in a ran-domized, placebo-controlled trial in patients witheuvolemic or hypervolemic hyponatremia. J ClinEndocrinol Metab 2006;91:2145–52.

80. Zeltser D, Rosansky S, van Rensburg H, et al.,Conivaptan Study Group. Assessment of the effi-cacy and safety of intravenous conivaptan ineuvolemic and hypervolemic hyponatremia. Am JNephrol 2007;27:447–57.

81. Konstam MA, Gheorghiade M, Burnett JC Jr.,et al., Efficacy of Vasopressin Antagonism in HeartFailure Outcome Study With Tolvaptan (EVEREST)Investigators. Effects of oral tolvaptan in patientshospitalized for worsening heart failure: theEVEREST Outcome Trial. JAMA 2007;297:1319–31.

82. Waldreus N, Hahn RG, Jaarsma T. Thirst inheart failure: a systematic literature review. Eur JHeart Fail 2013;15:141–9.

83. Albert NM, Nutter B, Forney J, et al.A randomized controlled pilot study of outcomes

of strict allowance of fluid therapy in hypona-tremic heart failure (SALT-HF). J Card Fail 2013;19:1–9.

84. Dzau VJ, Hollenberg NK. Renal response tocaptopril in severe heart failure: role of furosemidein natriuresis and reversal of hyponatremia. AnnIntern Med 1984;100:777–82.

85. Packer M, Medina N, Yushak M. Correction ofdilutional hyponatremia in severe chronic heartfailure by converting-enzyme inhibition. AnnIntern Med 1984;100:782–9.

86. Packer M, Carver JR, Rodeheffer RJ, et al., ThePROMISE Study Research Group. Effect of oralmilrinone on mortality in severe chronic heartfailure. N Engl J Med 1991;325:1468–75.

87. O’Connor CM, Gattis WA, Uretsky BF, et al.Continuous intravenous dobutamine is associatedwith an increased risk of death in patients withadvanced heart failure: insights from the FlolanInternational Randomized Survival Trial (FIRST).Am Heart J 1999;138:78–86.

88. Mullens W, Abrahams Z, Francis GS, et al.Sodium nitroprusside for advanced low-outputheart failure. J Am Coll Cardiol 2008;52:200–7.

89. Mullens W, Abrahams Z, Francis GS, et al.Usefulness of Isosorbide Dinitrate and Hydralazineas add-on therapy in patients discharged foradvanced decompensated heart failure. Am JCardiol 2009;103:1113–9.

90. Metra M, Cotter G, Davison BA, et al., RELAX-AHF Investigators. Effect of serelaxin on cardiac,renal, and hepatic biomarkers in the Relaxin inAcute Heart Failure (RELAX-AHF) developmentprogram: correlation with outcomes. J Am CollCardiol 2013;61:196–206.

91. Ponikowski P, Mitrovic V, Ruda M, et al.A randomized, double-blind, placebo-controlled,multicentre study to assess haemodynamic effectsof serelaxin in patients with acute heart failure.Eur Heart J 2014;35:431–41.

92. ClinicalTrials.gov. Effect of serelaxin versusstandard of care in acute heart failure (AHF) patients(RELAX-AHF-EU). February 14, 2014. UpdatedAugust 26, 2014. Available at: http://clinicaltrials.gov/ct2/show/NCT02064868. Accessed December2, 2014.

KEY WORDS arginine vasopressin,distal, diuretics, kidney tubules,physiopathology, sodium