diagnosis, evaluation, and treatment of hyponatremia ... e… · experts in hyponatremia convened...

CME INFORMATION

Diagnosis, Evaluation, and Treatment of Hyponatremia:Expert Panel Recommendations

Release Date: October 2013Expiration Date: October 2014Estimated time to complete the activity: 3 hoursJointly sponsored by Tufts University School of Medicine Office of Continuing Education, and In 2 MedEd, LLC

This activity is supported by an unrestricted educational grant from Otsuka America Pharmaceutical, Inc.

There is no fee to participate in this CME-certified activity.

Program OverviewHyponatremia is the most common disorder of electrolytes encountered in clinical practice. Although many cases are mild andrelatively asymptomatic, hyponatremia is nonetheless important clinically because of the potential for substantial morbidityand mortality. Despite knowledge of hyponatremia since the mid-20th century, this common disorder remains incompletelyunderstood in many basic areas because of its association with a plethora of underlying disease states, and its causation bymultiple etiologies with differing pathophysiological mechanisms. Optimal treatment strategies have not been well defined,both due to these reasons, and because of marked differences in symptomatology and clinical outcomes based on the acutenessor chronicity of the hyponatremia. The approval of the first vasopressin receptor antagonist (vaptan) in 2005 heralded thebeginning of a new era in the management of hyponatremic disorders. Since then the field has evolved considerably, includingnew data on previously unrecognized morbidities and mortalities associated with hyponatremia, the approval of a secondvaptan, and additional clinical experience with vaptans and other therapies for the treatment of patients with hyponatremia. Inview of this, a panel of experts in hyponatremia was convened in 2012 to update the panel’s earlier recommendations for theevaluation and treatment of hyponatremia.

Target AudienceThis activity was designed to meet the needs of endocrinologists, hepatologists, nephrologists, gastroenterologists, cardiolo-gists, internists, emergency room physicians, pharmacists, and any healthcare provider who is likely to encounter patients withhyponatremia.

FacultyJoseph Verbalis, MD (Consensus Panel Chairman)Professor of Medicine and PhysiologyChief of the Division of Endocrinology and MetabolismCo-Director of the Georgetown-Howard Universities Center for Clinical and Translational ScienceClinical Director of the Center for the Study of Sex Differences in Health, Aging, and Disease Georgetown UniversityWashington, DC

Cynthia Anne Korzelius, MD (CME Activity Director)Assistant Chief of Adult Inpatient MedicineNephrologist

Assistant Clinical ProfessorTufts University School of MedicineNewton-Wellesley HospitalNewton, MA

Steven Goldsmith, MDDirector of the Heart Failure ProgramHennepin County Medical Center inProfessor of MedicineUniversity of MinnesotaMinneapolis, MN

Arthur Greenberg, MDProfessor of MedicineDuke University Medical CenterDurham, NC

Robert Schrier, MDEmeritus Professor of MedicineUniversity of Colorado DenverAurora, CO

Richard Sterns, MDProfessor of MedicineUniversity of Rochester School of MedicineChief of MedicineDirector of the Rochester General Internal Medicine Residency ProgramRochester General HospitalRochester, NY

Christopher Thompson, MDDeputy Specialty Director, EndocrinologyBeaumont Hospital, Dublin, Ireland

Educational Activity GoalGiven new developments in the field of hyponatremia and its management—along with high interest by the multitudes ofclinicians who see hyponatremia in their practices and/or hospitals—the need is clear for current, evidence-based recom-mendations to fill the demonstrated educational and practice gaps of treating physicians. These evidence-based, expert-authored recommendations will reflect current scientific and treatment realities of hyponatremia management. By completingthis activity physicians should be better educated concerning: (1) risks for morbidity and mortality associated with hypona-tremia; (2) recognition, accurate diagnosis, and assessment of hyponatremia; and (3) strategies for managing the condition—incooperation with other specialists—based on clinical signs, biochemical measurements, risk factors, symptoms, rate of onset,and underlying causative factors.

After completing this activity, learners should be able to:� Identify and assess patients at risk for hyponatremia.� Achieve timely and effective diagnosis and management of patients with hyponatremia, taking into account the effects ofunderlying comorbid conditions and diuretic usage.

� Carefully monitor and control the rate of correction of serum sodium levels in patients with chronic hyponatremia to avoidpermanent and potentially fatal neurologic complications.

� Balance the potential interactions of one treatment with another to achieve optimal resolution of both the hyponatremia andthe underlying disease(s).

Core Competencies for Quality Patient CareThis educational activity primarily addresses Medical Knowledge core competency as defined by the Accreditation Council forGraduate Medical Education/American Board of Medical Specialties Competencies. Secondary competencies addressed bythis activity include Patient Care and Procedural Skills, as well as Practice-Based Learning and Improvement.

Accreditation Statement

PhysiciansThis activity has been planned and implemented in accordance with the Essential Areas and policies of the AccreditationCouncil for Continuing Medical Education through the joint sponsorship of Tufts University School of Medicine(TUSM) and In 2 MedEd, LLC. TUSM is accredited by the ACCME to provide continuing medical education forphysicians.

Credit DesignationTUSM designates this enduring material for a maximum of 3 AMA PRA Category 1 Credits�. Physicians should claim onlythe credit commensurate with the extent of their participation in the activity.

Requirements for Successful CompletionTo receive CE credit, participants must register, view the content, complete the evaluation, and successfully complete thepost-test with a minimum score of 80%. Certificates will be available electronically or in print format after successfulcompletion of the activity.

Disclosure of Conflict of InterestTufts University School of Medicine Office of Continuing Education (TUSM-OCE) assesses conflict of interest with itsinstructors, planners, managers and other individuals who are in a position to control the content of CME activities. All relevantconflicts of interest that are identified are thoroughly vetted by TUSM-OCE for fair balance, scientific objectivity of studiesutilized in this activity, and patient care recommendations. TUSM-OCE is committed to providing its learners with high qualityCME activities and related materials that promote improvements or quality in healthcare and not a specific proprietary businessinterest of a commercial interest.The faculty reported the following financial relationships or relationships to products or devices they or their spouse/lifepartnerhave with commercial interests related to the content of this CME activity:

Name of Faculty or Presenter
Reported Financial relationship
Cynthia Anne Korzelius, MD
Has no real or apparent conflicts of interest to report
Arthur Greenberg, MD
Grant/Research Support, Speakers’ Bureau, Advisory Committee,Consultant: Otsuka America Pharmaceutical, Inc.
Steven Goldsmith, MD
Grant/Research Support, Speakers’ Bureau: Otsuka America Pharmaceutical, Inc.Consultant: Otsuka America Pharmaceutical, Inc.; Medtronic
Robert Schrier, MD
Consultant: Otsuka America Pharmaceutical, Inc.; Janssen Pharmaceuticals, Inc.
Richard Sterns, MD
Has no real or apparent conflicts of interest to report
Christopher Thompson, MD
Consultant: Otsuka European Pharmaceuticals
Joseph Verbalis, MD
Grant/Research Support, Advisory Committee: Otsuka America Pharmaceutical, Inc.Consultant: Otsuka America Pharmaceutical, Inc., Cardiokine, Inc./Cornerstone Therapeutics, Inc.
The planners and managers reported the following financial relationships or relationships to products or devices they or their spouse/life partner have withcommercial interests related to the content of this CME activity:

Name of Planner or Manager
Reported Financial Relationship
In 2 MedEd editors and planners, Kim L. Farina, PhD, andJoan Weiss, BS, MS

Have no real or apparent conflicts of interest to report

TUSM OCE staff, Karin Pearson, Lara Shew, Mirosleidy Tejeda, andCarolyn S. Langer, MD, JD, MPH

Have no real or apparent conflicts of interest to report

Disclosure of Unlabeled UseThis educational activity may contain discussion of published and/or investigational uses of agents that are not indicated by the FDA. TuftsUniversity School of Medicine Office of Continuing Education and In 2 MedEd, LLC do not recommend the use of any agent outside of the labeledindications.The opinions expressed in the educational activity are those of the faculty and do not necessarily represent the views of Tufts University School of Medicine,Elsevier, and In 2 MedEd, LLC. Please refer to the official prescribing information for each product for discussion of approved indications, contraindications,and warnings.

Peer ReviewThis CME supplement has been peer-reviewed by The American Journal of Medicine.

Method of Participation:TUSM designates this enduring material for a maximum of 3 AMA PRA Category 1 Credits�. Physicians should claim only the credit commensurate with theextent of their participation in the activity.There is no fee for participating in this educational activity.

Online Participation:Online Activity URL: http://www.amjmed.com/issues?issue_key=S0002-9343(13)X0014-2To receive CE credit online, participants must register, view the content, go to https://hyponatremiacme.amjmed.com to complete the evaluation andsuccessfully complete the post-test with a minimum score of 80%. Certificates will be available electronically after successful completion of the activity.

Print Supplement Participation:After reviewing the print supplement, participants can complete the post-test and evaluation online, by mail, or by fax:Online URL: https://hyponatremiacme.amjmed.comMail: Educational Measures, 7373 S. Alton Way, Centennial, CO 80112Fax: 303-339-2439 (No cover sheet is necessary)The certificates will be e-mailed or mailed to the information provided on the evaluation.

MediaJournal supplement

DisclaimerParticipants have an implied responsibility to use the newly acquired information to enhance patient outcomes and their own professional development. Theinformation presented in this activity is not meant to serve as a guideline for patient management. Any procedures, medications, or other courses of diagnosisor treatment discussed or suggested in this activity should not be used by clinicians without evaluation of their patient’s conditions and possible contra-indications on dangers in use, review of any applicable manufacturer’s product information, and comparison with recommendations of other authorities.

ADA/OEO Nondiscrimination PolicyTufts University School of Medicine (TUSM) considers all applicants and activity participants without regard to race, color, national origin, age, religiouscreed, sex or sexual orientation. TUSM is an Equal Opportunity Employer. TUSM does not discriminate on the basis of a disability in admissions, access totreatment, or employment in their programs and activities as identified in the American with Disabilities Act.Policy on Privacy and Confidentiality: To view the privacy policy for TUSM OCE, please see: http://md.tufts.edu/Education/Continuing-Ed-Microsite/Privacy-and-Confidentiality-PolicyFor questions regarding certification, please contact Tufts University School of Medicine-Office of Continuing Education at [email protected] or617-636-6579.On the AJM journal website, visit the following URL for help: http://www.amjmed.com/webfiles/helpfiles/periodicals_help.htmFor accessing the AJM supplement via ScienceDirect, visit the following URL: http://help.sciencedirect.com/flare/sdhelp_Left.htm#CSHID¼home.htmjStartTopic¼Content%2Fhome.htmjSkinName¼svs_SD

http://www.amjmed.com/issues?issue_key=S0002-9343https://hyponatremiacme.amjmed.comhttps://hyponatremiacme.amjmed.comhttp://md.tufts.edu/Education/Continuing-Ed-Microsite/Privacy-and-Confidentiality-Policyhttp://md.tufts.edu/Education/Continuing-Ed-Microsite/Privacy-and-Confidentiality-Policymailto:[email protected]://www.amjmed.com/webfiles/helpfiles/periodicals_help.htmhttp://help.sciencedirect.com/flare/sdhelp_Left.htm#CSHID�home.htmjStartTopic�Content%2Fhome.htmjSkinName�svs_SDhttp://help.sciencedirect.com/flare/sdhelp_Left.htm#CSHID�home.htmjStartTopic�Content%2Fhome.htmjSkinName�svs_SD

SUPPLEMENT

Diagnosis, Evaluation, and Treatment of Hyponatremia:Expert Panel RecommendationsJoseph G. Verbalis, MD,a Steven R. Goldsmith, MD,b Arthur Greenberg, MD,c Cynthia Korzelius, MD,d

Robert W. Schrier, MD,e Richard H. Sterns, MD,f Christopher J. Thompson, MD, FRCPIgaGeorgetown University Medical Center, Washington, DC; bUniversity of Minnesota, Minneapolis, MN; cDuke University Medical Center,Durham, NC; dTufts University School of Medicine, Boston, MA; eUniversity of Colorado, Denver, CO; fUniversity of Rochester, Rochester,NY; gRoyal College of Surgeons in Ireland School of Medicine, Dublin, Ireland.

Funding: Thimeeting that wassponsored by the TEducation and In 2from Otsuka Ame

Conflict of Intand non-CME (CAmerica Pharmacsupport and non-CInc., and Cornersttial conflicts of ina consultant for

0002-9343/$ -seehttp://dx.doi.org/1

ABSTRACT

Hyponatremia is a serious, but often overlooked, electrolyte imbalance that has been independentlyassociated with a wide range of deleterious changes involving many different body systems. Untreatedacute hyponatremia can cause substantial morbidity and mortality as a result of osmotically induced ce-rebral edema, and excessively rapid correction of chronic hyponatremia can cause severe neurologicimpairment and death as a result of osmotic demyelination. The diverse etiologies and comorbiditiesassociated with hyponatremia pose substantial challenges in managing this disorder. In 2007, a panel ofexperts in hyponatremia convened to develop the Hyponatremia Treatment Guidelines 2007: Expert PanelRecommendations that defined strategies for clinicians caring for patients with hyponatremia. In the 6 yearssince the publication of that document, the field has seen several notable developments, including newevidence on morbidities and complications associated with hyponatremia, the importance of treating mild tomoderate hyponatremia, and the efficacy and safety of vasopressin receptor antagonist therapy for hypo-natremic patients. Therefore, additional guidance was deemed necessary and a panel of hyponatremiaexperts (which included all of the original panel members) was convened to update the previous recom-mendations for optimal current management of this disorder. The updated expert panel recommendations inthis document represent recommended approaches for multiple etiologies of hyponatremia that are based onboth consensus opinions of experts in hyponatremia and the most recent published data in this field.� 2013 Elsevier Inc. All rights reserved. � The American Journal of Medicine (2013) 126, S1-S42KEYWORDS: Antidiuretic hormone; Aquaretics; Hypo-osmolality; Natriuresis; Syndrome of inappropriate antidiuretic

sh

reoe

o

f0

hormone secretion; Vasopressin; Volume regulation

Hyponatremia is the most common disorder of electrolytesencountered in clinical practice, occurring in 15%-30% ofacutely or chronically hospitalized patients.1 Although manycases are mild and relatively asymptomatic, hyponatremia is

supplement is based, in part, on a closed roundtableeld in October 2012 in New York City and was jointlyufts University School of Medicine Office of ContinuingMedEd, LLC, through an unrestricted educational grantica Pharmaceutical, Inc.rest: Steven Goldsmith, MD has received grant supportntinuing Medical Education)-related fees from Otsukautical, Inc. Arthur Greenberg, MD has received grantME-related fees from Otsuka America Pharmaceutical,ne Therapeutics. Cynthia Korzelius, MD has no poten-terest to disclose. Robert Schrier, MD has served asOtsuka America Pharmaceutical, Inc., and Janssen

ront matter � 2013 Elsevier Inc. All rights reserved..1016/j.amjmed.2013.07.006

nonetheless important clinically because: 1) acute severehyponatremia can cause substantial morbidity and mortality;2) adverse outcomes, including mortality, are higher inhyponatremic patients with a wide range of underlying

Pharmaceutical. Richard Sterns, MD has no potential conflicts of interest todisclose. Joseph Verbalis, MD has received grant support from OtsukaAmerica Pharmaceutical, Inc, as well as non-CME-related fees from OtsukaAmerica Pharmaceutical, Inc., Cardiokine, and Cornerstone Therapeutics.Christopher Thompson, MD has served as a consultant for Otsuka Euro-pean Pharmaceuticals.

Authorship: All authors reviewed the data that were cited in themanuscript and wrote their respective sections of this manuscript.

Requests for reprints should be addressed to Joseph G. Verbalis, MD,Endocrinology and Metabolism, Georgetown University Medical Center,232 Building D, 4000 Reservoir Rd. NW, Washington, DC 20007.

E-mail address: [email protected]

mailto:[email protected]://dx.doi.org/10.1016/j.amjmed.2013.07.006

S6 The American Journal of Medicine, Vol 126, No 10A, October 2013

diseases; and 3) overly rapid correction of chronic hypo-natremia can cause severe neurological deficits and death.

Despite knowledge of hyponatremia since the mid-20thcentury, this common disorder remains incompletely un-derstood in many basic areas because of its association witha plethora of underlying disease states, its causation bymultiple etiologies with differing pathophysiologicalmechanisms, and marked differences in symptomatologyand clinical outcomes based on the acuteness or chronicityof the hyponatremia.2 For these reasons, optimal treatmentstrategies have not been well defined.

The approval of the first vasopressin receptor antagonist(drugs in this class are also referred to as vaptans, forvasopressin antagonists), conivaptan (Astellas Pharma, U.S.,Inc., Northbrook, IL), for clinical use by the US Food andDrug Administration (FDA) in 2005 heralded the beginningof a new era in the management of hyponatremic disorders.However, proper and effective use of these and othertherapies requires careful thought and guidance. In 2005, aconsensus panel of experts in hyponatremia was convenedto review existing therapies for hyponatremia and toevaluate the situations where aquaretic agents should beconsidered as alternatives or supplements to accepted cur-rent therapies; the initial review by this group was pub-lished in 2007.3 Since then, new data on previouslyunrecognized morbidities and mortalities associated withhyponatremia have emerged, additional clinical experiencewith vaptans and other therapies for patients with hypo-natremia has accumulated, and the FDA and the EuropeanMedicines Agency (EMA) approved a second vaptan, tol-vaptan (Otsuka Pharmaceutical Co., Ltd., Tokyo, Japan).4

In view of these developments, a similar panel of expertsin hyponatremia, including all of the participants of theinitial panel, was convened in 2012 to update the earlierrecommendations for the evaluation and treatment ofhyponatremia.

METHODS

Search StrategyThe PubMed database was electronically searched fromJanuary 1, 2006 through December 31, 2012. Additionalstudies were obtained from reference libraries of the panelmembers and from publication reference lists. Searcheswere limited to English-language studies on humans usingthe search terms hyponatremia and: hypervolemia, euvole-mia, hypovolemia, heart failure, liver cirrhosis, syndromeof inappropriate ADH (antidiuretic hormone secretion),chemically induced, congenital, ethnology, etiology, ge-netics, diagnosis, radiography, radionuclide, imaging, andultrasonography. For each topic section, study selection wasperformed by an assigned reviewer.

ConsensusTopic summaries and recommendations were reachedthrough consideration of the strength and quality of available

evidence (eg, trial, type of trial vs. case report, number oftrial subjects). Both negative and positive studies wereconsidered. When the literature did not support recommen-dations based solely on evidence, recommendations reflectconsensus opinion reached by the expert panel throughroundtable discussion and multiple draft reviews. Fortopics for which consensus was not reached, divergentopinions are noted. The quality of the data available tosupport each recommendation was noted and taken intoaccount for the final recommendations. Several steps weretaken to ensure that objectivity was maintained throughoutthe various steps of manuscript preparation. First, the TuftsUniversity Continuing Medical Education course director(C.K.), who has no potential conflicts of interest, wasspecifically tasked with identification and avoidance ofbias and with ensuring the objectivity of the manuscript.Second, a separate external review for objectivity wasconducted by the Tufts University Office of ContinuingMedical Education.

Level of EvidenceEach panel member critically evaluated relevant literatureto inform recommendations for each topic. The authorsacknowledge the paucity of available randomizedcontrolled trials with clinical outcome measures to supportevidence-based recommendations for most available ther-apies. As such, the use of a quality-of-evidence scoringsystem to grade the strength of supporting data for eachrecommendation was not feasible. The panel recognizesthe need for expert guidance in hyponatremia managementin the absence of such literature, and the need for futurestudies to evaluate management strategies not currentlysupported by evidence from high-quality randomizedcontrolled trials.

CLINICAL SIGNIFICANCE OF HYPONATREMIAThe rationale for these expert recommendations for thesafe and effective treatment of hyponatremia is the clinicalsignificance of this disorder. A wealth of evidence existsthat fully justifies hyponatremia as an important clinicaltreatment target. However, perhaps the most convincingrationale for these recommendations is the need for a betterunderstanding of the potential consequences of not effec-tively treating hyponatremia in the many patients with thisdisorder, both hospitalized and outpatients.

Incidence and PrevalenceHypo-osmolality is one of the most common disorders offluid and electrolyte balance encountered in hospitalizedpatients. The incidence and prevalence of hypo-osmolardisorders depend on the nature of the patient populationbeing studied, as well as on the laboratory methodsand diagnostic criteria used to ascertain hyponatremia.Most investigators have used the serum sodium concen-tration ([Naþ]) to determine the clinical incidence of

Verbalis et al Hyponatremia Treatment S7

hypo-osmolality. When hyponatremia is defined as a serum[Naþ] below 135 mmol/L (sodium is univalent, so 1 mmol/L ¼ 1 mEq/L), incidences as high as 15%-30% have beenobserved in studies of both acutely5 and chronically6 hos-pitalized patients. Similarly high incidences have been re-ported in patients with specific disease states, includingpatients with heart failure (HF) and cirrhosis; reports fromrecent trials and registries suggest that hyponatremia is seenin up to 27% of patients admitted with acute HF,7-10 andthat up to 50% of patients with cirrhosis and ascites arefound to be hyponatremic.11 A more recent study thatanalyzed adverse outcomes in a large number of hospital-ized hyponatremic patients proposed revising the definitionof hyponatremia to serum [Naþ]


adjustment.28 This study provides clear documentation of anincreased incidence of falls in so-called asymptomatichyponatremic patients.

The clinical significance of the gait instability and falldata were further evaluated in a study that compared 553patients with fractures to an equal number of matchedcontrols. Hyponatremia was found in 13% of the patientspresenting with fractures, when compared with only 4% ofthe controls.29 Similar findings were reported in studies of364 elderly patients with large-bone fractures in NewYork,30 in 1408 female patients with early chronic renalfailure in Ireland31 (Figure 1), and in 5208 elderly patientsfollowed for 12 years in the Rotterdam Longitudinal AgingStudy.32 More recently, published studies have shown thathyponatremia is associated with increased bone loss inexperimental animals and with a significantly increased ORfor osteoporosis of the femoral neck (OR 2.87; P

Figure 2 Mechanism of renal water reabsorption induced by arginine vasopressin (AVP)activation of the V2 receptor on renal collecting duct principal cells. AVP binds to theG-protein-linked V2 receptor on the basolateral membrane. G-protein-coupled receptorsignaling consists of three steps: a hepta-helical receptor that detects a ligand (in this case,AVP) in the extracellular milieu, a G-protein (Gas) that dissociates into α subunits bound toGTP and bg subunits after interaction with the ligand-bound receptor, and an effector(adenylyl cyclase) that interacts with the dissociated G-protein subunits to generate secondmessengers. AVP activates adenylyl cyclase, increasing the intracellular concentration ofcyclic adenosine monophosphate (cAMP). Protein kinase-A (PKA) is the target of thegenerated cAMP. The binding of cAMP to the regulatory subunits of PKA induces aconformational change, causing these subunits to dissociate from the catalytic subunits.These activated subunits (C) are anchored to an aquaporin-2 (AQP2)-containing endocyticvesicle via an A-kinase anchoring protein (AKAP). The local concentration and distributionof the cAMP gradient is limited by phosphodiesterases (PDE). Phosphorylation of the AQP2water channels in the endocytic vesicles leads to movement of the vesicles toward theluminal membrane via microtubules and actin filaments with eventual fusion into the luminalmembrane, thereby increasing the water permeability of this membrane. Water (H2O) is thenreabsorbed from the urine in the collecting duct into the principal cells along osmotic gra-dients. When AVP is not bound to the V2 receptor, AQP2 water channels are retrieved by anendocytic process, and water permeability returns to its original low rate. AQP3 and AQP4water channels are expressed constitutively at the basolateral membrane and allow intra-cellular water to exit into the blood of the vasa recta. In the presence of medullary hyper-osmolality, water therefore moves across the principal cell and returns into the circulation.This process results in urinary concentration, or antidiuresis. In the presence of nonosmoticAVP stimulation, water is retained and hyponatremia can occur.Adapted with the permission of the American Society of Nephrology, from Bichet DG.J Am Soc Nephrol. 2006;17:920-922; permission conveyed through Copyright ClearanceCenter, Inc.


AVP levels relative to osmolality are potential candidatesfor treatment with agents that block activation of the AVP-mediated antidiuretic effects in the kidneys. Exceptionsinclude patients with hypovolemic hyponatremia who aremore safely treated with solute and volume repletion andpatients with hyponatremias in which AVP is not an etio-logic factor, including acute water intoxication and renalfailure; therapies for all hyponatremic disorders will be

discussed in the appropriate sections based on the etiologyof the hyponatremia.

CLASSIFICATION AND DIFFERENTIAL DIAGNOSISOF HYPONATREMIAThe presence of significant hypo-osmolality indicates excesswater relative to solute in the extracellular fluid (ECF)


compartment. Because water moves freely between the ECFand the intracellular fluid (ICF) compartments, an excess oftotal bodywater relative to total body solute is present as well.

Differentiation of Hypotonic Hyponatremiafrom Other Causes of HyponatremiaThe osmolality of body fluid normally is maintained withinnarrow limits by osmotically regulated AVP secretion andthirst. Although basal plasma osmolality can vary amongindividuals, the range in the general population under con-ditions of normal hydration is between 280 and 295 mOsm/kg H2O. However, total osmolality is not always equivalentto effective osmolality, often termed plasma tonicity. Onlysolutes that are impermeable to the cell membrane andremain relatively compartmentalized within the ECF are"effective" solutes, because these are capable of creatingosmotic gradients across cell membranes and thereby effectosmotic movement of water between the ICF and the ECFcompartments. As such, the concentration of effective sol-utes in plasma should be used to determine whether clini-cally significant hypo-osmolality is present. Sodium and itsaccompanying anions are the major effective plasma solutes,so hyponatremia and hypo-osmolality are usually synony-mous. However, there are 2 situations where hyponatremiaand hypo-osmolality are discordant; these are important torecognize clinically because they represent situations wherehyponatremia does not need to be treated.

Pseudohyponatremia. Marked elevations of either lipids orproteins in plasma can cause artifactual decreases in serum[Naþ] because of the larger relative proportion of plasmavolume that is occupied by the excess lipids or proteins.Because the increased protein or lipid will not appreciablychange the total number of solute particles in solution, thedirectly measured plasma osmolality will be normal in suchcases and, therefore, the patient will be isotonic rather thanhypotonic45 (Table 1).

Isotonic or Hypertonic Hyponatremia. Hyponatremiawith normal or even increased osmolality occurs wheneffective solutes other than sodium are present in theplasma. The initial hyperosmolality produced by the addi-tional solute causes an osmotic shift of water from the ICFto the ECF compartment that, in turn, produces a dilutionaldecrease in the serum [Naþ]. Hyperglycemia is the mostcommon example of this phenomenon. Depending on the

Table 1 Classification of Hyponatremia by Plasma Tonicity

Serum SodiumConcentration (mmol/L)

Plasma Osmolality(mOsm/kg H2O) Typica

Hypotonic


status and urine electrolyte excretion permits sufficientcategorization of the underlying etiology to allow initiationof therapy and direct further diagnostic evaluation. Thefollowing sections will describe the diagnostic criteria,common etiologies, and pathophysiologies of the 3 majorclassifications of hypotonic hyponatremia based on the pa-tient’s ECF volume status: hypovolemic hyponatremia,euvolemic hyponatremia, and hypervolemic hyponatremia.It should be emphasized that these statements representconsensus-based clinical recommendations, not dogmaticdiagnostic pathways; clinical acumen must always informand temper their application. Nonetheless, studies haveclearly documented that inappropriate diagnosis of hypo-natremia often leads to illogical therapies and worse clinicaloutcomes.48 Furthermore, following a simple algorithm fordiagnosing and treating hyponatremia led to significantlyimproved management outcomes.49 Thus, the importance ofappropriate classification and diagnosis of hypotonic hypo-natremia should not be underestimated. In some cases,the ECF volume status may be ambiguous, making itdifficult to classify hyponatremia using standard criteria ofECF volume assessment; in other cases, multiple etiologiesare present. Nonetheless, application of a standard approachto diagnosis and therapy with regular reassessmentof responses to therapy will generally lead to improvedoutcomes.

Hypovolemic HyponatremiaThe presence of clinically detectable decreased ECF volumeis usually indicative of solute depletion. Hyponatremia withvolume depletion (hypovolemia) can arise in a variety ofsettings. Because intravascular volume cannot be easilymeasured directly, volume depletion is diagnosed clinicallyfrom the history, physical examination, and laboratory re-sults. Patients with clinical symptoms or signs of volumedepletion (eg, vomiting and diarrhea, orthostatic decreasesin blood pressure and increases in pulse rate, dry mucusmembranes, and decreased skin turgor) should be consid-ered to be hypovolemic, unless there are alternative expla-nations for these findings. When available, directhemodynamic measurements can provide corroboration ofthe clinical impression. Elevations of blood urea nitrogen(BUN), creatinine, the BUN-creatinine ratio, and uric acidlevel are helpful laboratory clues to the presence of volumedepletion. However, these findings are neither sensitive norspecific and can be affected by other factors (eg, dietaryprotein intake, use of glucocorticoids). Measuring theurine sodium excretion is usually more helpful. The spoturine [Naþ] should be


hypervolemic, unless there are alternative explanations forthese findings. When available, hemodynamic measurementscan corroborate the clinical impression. Elevation of plasmalevels of brain natriuretic peptide provides useful laboratorysupport for the presence of volume overload. The urine [Naþ]or fractional sodium excretion are usually low (spot urine[Naþ]


typically increase with volume depletion, were lower inpatients with thiazide-induced hyponatremia when com-pared with normonatremic patients taking thiazides.64 Takentogether with the frequent lack of clinical evidence forvolume depletion, these data suggest a role for abnormalthirst and water intake in individuals who develop thiazide-induced hyponatremia.

The urine [Naþ] concentration will be increased as aneffect of diuretic administration; a level >30-50 mmol/Lcannot be taken as evidence of euvolemia in a patientreceiving a diuretic. Despite the point made previouslyabout serum uric acid in thiazide-induced hyponatremia,a recent study reported that an increased fractional excretionof uric acid may be a more reliable indicator of the presenceof euvolemia and SIADH in patients receiving diuretics.52

However, we believe that the diagnosis of SIADH should bemade with caution in a patient who is receiving a diuretic,particularly a thiazide. If the clinical suspicion exists that apatient receiving a diuretic actually has SIADH, reevalua-tion after cessation of the diuretic and a saline challenge isrequired.

Cerebral Salt Wasting. Cerebral salt washing (CSW) is asyndrome that has been described following subarachnoidhemorrhage, head injury, or neurosurgical procedures, aswell as in other settings. The initiating event is loss of so-dium and chloride in the urine, which results in a decrease inintravascular volume, leading to water retention and hypo-natremia because of a baroreceptor-mediated stimulus toAVP secretion. The sodium resorptive defect has beenattributed to a proximal tubular defect that is accompaniedby increased uric acid and urea excretion, features that maymake the use of BUN and uric acid levels unreliable fordistinguishing between SIADH and CSW.67 The incidenceof CSW is unknown, but it is generally agreed to be un-common. In a series of 187 consecutive cases of hypona-tremia in neurosurgical patients, only 3.7% had CSW and2.7% had CSW with SIADH.68

Differentiation of CSW from SIADH hinges uponestablishing that a period of urinary sodium loss and volumedepletion preceded the development of hyponatremia.Because infusion of isotonic saline into a patient witheuvolemia and SIADH results in a rapid excretion of thesodium and fluid load to maintain balance, a high urine[Naþ] and urine flow rate alone do not establish that CSW ispresent. Physicians should review vital signs, weight, he-matocrit, and input/output records to determine what thepatient’s volume status and net fluid balance were justbefore and during the development of hyponatremia. Cur-rent physical findings and hemodynamic measures shouldalso be taken into account. Often, patients suspected to haveCSW actually have SIADH, with high sodium and urineoutputs driven by high sodium and fluid inputs. A cautiousreduction in fluid replacement in such patients will distin-guish them from individuals with CSW; only patients withthe latter diagnosis will develop signs of volume depletionas fluid replacement is tapered.

Mineralocorticoid Deficiency. Patients with isolatedglucocorticoid deficiency from adrenocorticotropic hormone(ACTH) suppression or deficiency do not have mineralo-corticoid deficiency, so they do not have inappropriaterenal sodium wasting or hyperkalemia. In these patients,hyponatremia results from a failure to fully suppressAVP release in response to hypo-osmolality. Such patientsare euvolemic (see the subsequent section: EuvolemicHyponatremia, Glucocorticoid Deficiency). The most com-mon form of isolated mineralocorticoid deficiency, hypo-reninemic hypoaldosteronism (type IV renal tubularacidosis), is associated with volume expansion and does notresult in significant hyponatremia. In contrast, in patientswith mineralocorticoid deficiency from primary adrenalinsufficiency caused by adrenal destruction or hereditaryenzyme deficiencies, renal sodium wasting leads to hypo-volemia and a secondary volume stimulus to AVP release.Ingestion of water or administration of hypotonic fluids maylead to water retention and hyponatremia, as with volumedepletion from other causes. Volume depletion with highurine [Naþ] and accompanying hyperkalemia should raisesuspicion of mineralocorticoid deficiency; low urine [Kþ]can provide additional confirmation. However, the absenceof hyperkalemia does not exclude consideration of adrenalinsufficiency, especially in children with volume depletion.In 18 children with proven mineralocorticoid deficiency,hyponatremia was observed in 88% but hyperkalemia inonly 50%.69 If combined mineralocorticoid and glucocorti-coid deficiency from adrenal destruction is suspected, cor-ticosteroids should be administered promptly even asdiagnostic confirmation by measurement of aldosterone andACTH levels, and cortisol response to cosyntropin (ACTH)stimulation, is undertaken.

Euvolemic HyponatremiaEuvolemic hyponatremia always occurs as a result of arelative or absolute excess of body water. Although theexcess body water can accumulate secondarily to over-drinking, the capability of the kidneys to excrete a largevolume of electrolyte-free water makes excessive wateringestion as a sole cause of euvolemic hyponatremia veryuncommon. The overwhelming majority of cases arise as aresult of reduced renal electrolyte-free water excretion dueto the antidiuretic actions of AVP at the kidney V2R. Veryrarely, hyponatremia may be caused by non-AVP-mediatedmechanisms. The major disorders associated with euvolemichyponatremia are described below.

SIADH. SIADH is the most common cause of euvolemichyponatremia, and is associated with many different disor-ders that can be divided into several major etiologic groups.2

The criteria necessary for its diagnosis were originallydefined by Bartter and Schwartz in 196770 and remainessentially unchanged (Table 2). These criteria haveattained widespread clinical acceptance, but a number ofinterpretive considerations apply. First, to diagnose SIADH

Table 2 Criteria for Diagnosing SIADH

Decreased effective osmolality of the extracellular fluid (Posm 100 mOsmol/kg H2O with normal renal function) at some level of plasma hypo-osmolality.Clinical euvolemia, as defined by the absence of signs of hypovolemia (orthostasis, tachycardia, decreased skin turgor, dry mucousmembranes) or hypervolemia (subcutaneous edema, ascites).Elevated urinary sodium excretion (>20-30 mmol/L) while on normal salt and water intake.Absence of other potential causes of euvolemic hypo-osmolality: severe hypothyroidism, hypocortisolism (glucocorticoid insufficiency).Normal renal function and absence of diuretic use, particularly thiazide diuretics.

H2O ¼ water; kg ¼ kilogram; mmol ¼ millimole; mOsmol ¼ milliosmole; Posm ¼ plasma osmolality; SIADH ¼ syndrome of inappropriate antidiuretichormone secretion; Uosm ¼ urine osmolality.


it is not necessary for urine osmolality to exceed plasmaosmolality. In the setting of hypo-osmolality, AVP secretionshould be physiologically suppressed to promote an aqua-resis; the urine should, therefore, be maximally dilute (ie,urine osmolality �100 mOsm/kg H2O in adults). A urineosmolality >100 mOsm/kg, therefore, reflects inappropriateantidiuresis and is compatible with the diagnosis of SIADH.In SIADH due to a reset osmostat, urine osmolality need notbe inappropriately elevated at all levels of plasma osmo-lality, but simply at some level under 275 mOsm/kg H2O, asAVP secretion can be suppressed at lower levels of plasmaosmolality, resulting in maximal urinary dilution.71 Clinicaleuvolemia must be present to diagnose SIADH; this diag-nosis cannot be made in a hypovolemic or edematous pa-tient. This does not mean that patients with SIADH cannotbecome hypovolemic for other reasons; but, in such cases,the diagnosis of SIADH cannot be made until the patient isrendered euvolemic. Urine sodium excretion helps distin-guish hypo-osmolality caused by a decreased effectivearterial blood volume, in which case renal sodium conser-vation occurs, from dilutional disorders, in which renal so-dium excretion is normal or increased due to ECF volumeexpansion. Elevated urine [Naþ] is not specific to SIADHand is also seen in renal causes of solute depletion such asdiuretic use or mineralocorticoid deficiency. Conversely,if patients with SIADH become hypovolemic or solutedepleted—for instance, during sodium and water restric-tion—urine [Naþ] may fall. Therefore, although the mea-surement of urine [Naþ] is central to the diagnosis ofSIADH, elevated levels are neither pathognomonic noressential. The final criterion emphasizes that SIADH re-mains a diagnosis of exclusion, and the absence of otherpotential causes of hypo-osmolality must always be verified,as discussed below. Of note, measurement of plasma AVPhas never been a criterion for diagnosing SIADH because:1) AVP levels are variably elevated in patients with SIADH,sometimes near assay detection limits; 2) AVP is difficult tomeasure accurately because it circulates at such low plasmalevels and because sampling handling, storing, and assayingare difficult; and 3) AVP levels are elevated in all classifi-cations of hyponatremia—hypovolemic, euvolemic, andhypervolemic—and thus would not help to differentiateamong these diagnostically. In addition to the classicalcriteria of Bartter and Schwartz,70 additional parametershave been suggested as valuable in the differential diagnosis

of SIADH from other causes of hyponatremia. Copeptin is alarge fragment of the AVP prohormone that is more stablethan AVP and easier to assay. One group has suggested thatthe measurement of the ratio of copeptin to urine [Naþ]reliably distinguishes between hypovolemic hyponatremiaand SIADH;72 the same group has also advocated that thefinding of increased fractional uric acid excretion is highlypredictive of SIADH, even in patients on diuretic therapy.52

However, neither of these measurements is widespread inclinical practice.

Nephrogenic Syndrome of Inappropriate Anti-diuresis. Recent studies of children with hyponatremia havediscovered 2 genetic mutations of the V2R leading to itsconstitutive activation and antidiuresis in the absence ofAVP-V2R ligand binding.73 These patients met all theclassic criteria for a diagnosis of SIADH, except that theplasma AVP levels were found to be below detection limitsby radioimmunoassay. At least 1 kindred has been describedin which several individuals bearing this mutation did notmanifest clinically recognized hyponatremia until late intoadulthood.74 The true incidence of these and similar V2Rmutations, as well as how often they are responsible for thepattern of euvolemic hyponatremia with low or unmeasur-able plasma AVP levels found in approximately 10% ofpatients with SIADH,75 remains to be determined. However,based on the low number of reported cases to date, neph-rogenic syndrome of inappropriate antidiuresis (NSIAD)appears to be rare as a cause of hyponatremia.

Glucocorticoid Deficiency. Isolated glucocorticoid defi-ciency occurs in association with pituitary disorders thatimpair normal ACTH secretion, leading to secondary adre-nal insufficiency. The cortisol deficiency leads to failure tosuppress AVP; thus, the biochemical abnormalities in iso-lated glucocorticoid deficiency more closely resembleSIADH rather than classical Addison’s disease. Aldosteronesecretion, which is under primary control of the renin-angiotensin system, remains intact; therefore, patients withhypopituitarism generally do not develop ECF volumecontraction. The clinical observation that anterior pituitaryinsufficiency ameliorates, and sometimes even completelymasks, the polyuria of patients with coexistent central dia-betes insipidus76 has led to a longstanding appreciation ofthe importance of glucocorticoids in water excretion.


Hyponatremia occurs frequently in ACTH-deficient patientswithout diabetes insipidus,77 and diabetes insipidus may notbecome manifest until glucocorticoid therapy is started andenables normal electrolyte-free water excretion.

Although an apparent hypovolemia-mediated stimulus toAVP secretion is lacking, nonosmotic AVP secretion hasnonetheless been strongly implicated in the impaired waterexcretion of glucocorticoid insufficiency, possibly second-ary to associated hypotension. Elevated plasma AVP levelshave clearly been documented in animals and patients78

with hypopituitarism. That these elevated AVP levels werecausally related to impaired water excretion was proven bystudies using an AVP-V2R antagonist, which demonstratednear normalization of urinary dilution in adrenalectomizedmineralocorticoid-replaced rats.79 Therefore, the hypona-tremia of glucocorticoid insufficiency is due to a combina-tion of impaired electrolyte-free water excretion in theabsence of normal glucocorticoid activity in the kidney, aswell as the antidiuretic action of nonosmotically stimulatedAVP secretion.

Glucocorticoid deficiency is primarily identified inhyponatremia by a high level of clinical suspicion, alongwith formal measurement of plasma cortisol concentrations.Ideally, this would be in response to dynamic stimulationwith synthetic ACTH, which is a simple screening test. Insituations where acute ACTH/cortisol deficiency developsand causes hyponatremia—for instance, following braintrauma or subarachnoid hemorrhage—adrenal atrophy willnot have occurred, and the cortisol response to cosyntropinmay give a false reassurance that cortisol dynamics arenormal. Diagnosis is more difficult in this situation, as testsof the entire hypothalamic-pituitary-adrenal axis (such as theinsulin tolerance test) may be contraindicated because of therisk of seizures. Simply measuring a 9 AM cortisol canprovide useful empirical evidence; a level below 10 mg/dL(300 nmol/L) is unphysiological in an acutely ill patient andcan serve as an indication for glucocorticoid therapy inconditions such as neurosurgical hyponatremia, where acuteACTH deficiency is a reasonable possibility.68

Hypothyroidism. Hyponatremia secondary to hypothy-roidism is so rare that some investigators have questionedwhether hypothyroidism is in fact causally related tohyponatremia.80 If hyponatremia occurs in patients with anelevated thyroid-stimulating hormone, the assumptionshould not be made that hypothyroidism is the cause of thelow serum [Naþ], as impaired water excretion leading tohyponatremia is seen only in patients with more severehypothyroidism who typically are elderly and meet criteriafor myxedema coma.81 Hyponatremia may result fromeither primary or secondary hypothyroidism, although whenhyponatremia accompanies hypopituitarism it is usually amanifestation of secondary glucocorticoid deficiency ratherthan hypothyroidism.

The major cause of impaired water excretion in hypo-thyroidism appears to be an alteration in renal perfusion anda reduced glomerular filtration rate (GFR) secondary to the

systemic effects of thyroid hormone deficiency on cardiacoutput and peripheral vascular resistance.82 In uncompli-cated hypothyroidism, there appears to be little elevation ofplasma AVP levels. However, as the hypothyroidism be-comes more severe, the effective arterial blood volume candecrease sufficiently to stimulate AVP secretion via baro-receptor mechanisms. Additionally, the impaired cardiacfunction that often occurs with advanced myxedema canlead to an elevation in plasma AVP levels. Whether hypo-natremia develops at any stage of disease progression de-pends on the relative balance between water intake andexcretory capacity; because maximal solute-free waterclearance decreases as these defects become more pro-nounced, the incidence of hyponatremia increases as theseverity of the underlying hypothyroidism worsens.

EAH. Detailed balance studies performed during the re-covery from an ultramarathon race show that runners withEAH excreted a large volume of dilute urine in contrast tonormonatremic finishers who excreted a small volume ofhighly concentrated urine; both groups had equivalent so-dium losses as reflected by positive sodium balances duringrecovery.83 The decrease in serum [Naþ] after enduranceexercise is directly proportional to the increase in bodyweight, and the athletes with EAH tended to gain weightduring the exercise.84 In marathon runners, low body massindex, race time exceeding 4 hours, consumption of fluidsevery mile, following advice to “drink as much aspossible” during the race, and greater frequency of urina-tion during the race have all been associated with EAH. Insome studies, female sex and the use of nonsteroidal anti-inflammatory drugs were also risk factors.56 Thus, whileathletes with normonatremia and hypernatremia are oftendehydrated, most runners with EAH are overhydrated asa result of excessive, and perhaps ill-advised, wateringestion over an extended race time during which waterexcretion is limited by nonosmotically stimulated AVPsecretion.57,85,86

Low Solute Intake. Some cases of euvolemic hypona-tremia do not fit particularly well into either a euvolemic orhypovolemic category. Among these is the hyponatremiathat sometimes occurs in patients who ingest large volumesof beer with little food intake for prolonged periods (beerpotomania). Even though the volume of fluid ingested maynot seem sufficiently excessive to overwhelm renal-dilutingmechanisms, in these cases, solute-free water excretion islimited by very low urine solute excretion as a result of thelow solute content of beer, because �50 mOsmol of urinarysolute excretion are required to excrete each liter of maxi-mally dilute urine. Because of this, water retention andhyponatremia will result when fluid intake exceeds themaximum volume of urine that can be excreted based on theavailable urine solute.87 Similar cases have been reported inpatients on very-low-protein diets88 or who restrict them-selves to “tea and toast” diets; both these diet types are alsolow in solute content. Because urine osmolality is typically


very low in such patients, there is no significant role forAVP in producing the hyponatremia.

Primary Polydipsia. Excessive water intake itself is rarelyof sufficient magnitude to produce hyponatremia in thepresence of normal renal function. However, it is often asignificant contributing factor to hyponatremia in patientswith polydipsia, particularly those with underlying defectsin electrolyte-free water excretion. The most dramatic casesof primary polydipsia are seen in psychiatric patients, par-ticularly those with acute psychosis secondary to schizo-phrenia.89 Studies of psychiatric patients with polydipsiahave shown a marked diurnal variation in serum [Naþ] (eg,from 141 mmol/L at 7:00 AM to 130 mmol/L at 4:00 PM),suggesting that many such patients drink excessively duringthe daytime but then correct themselves via a water diuresisat night.90 This and other considerations have led to definingthis disorder as the psychosis-intermittent hyponatremia-polydipsia syndrome. Polydipsia has been observed in up to20% of psychiatric inpatients, with incidences of intermit-tent hyponatremia ranging from 5%-10%.91 Hyponatremicpatients have often been prescribed medications, such asselective serotonin-reuptake inhibitors or phenothiazines,which can contribute to SIADH. Other conditions, such ascentral nervous system (CNS) sarcoidosis92 and cranio-pharyngioma,93 can also be associated with increased thirstand fluid ingestion. Consequently, polydipsic patientsshould be evaluated with a computed tomography or mag-netic resonance imaging scan of the brain before concludingthat excessive water intake is due to a psychological cause.

Sometimes excessive water intake alone will be sufficientto overwhelm renal excretory capacity and produce severehyponatremia. Although the water excretion rate of normaladults can exceed 20 L/day (d), maximum hourly ratesrarely exceed 800-1000 mL/hour (h). Studies of waterloading in nonexercising athletes have indicated a similarpeak urine excretion rate of 778 � 39 mL/h.94 Becausemany psychiatric patients drink predominantly during theday or during intense drinking binges, they can transientlyachieve symptomatic levels of hyponatremia with total dailyvolumes of water intake


water-impermeable thick ascending limb of the loop ofHenle. Thus, a GFR of 100 mL/min theoretically leads to adaily filtrate of 144 L, with 20% (ie, 28 L) reaching thedistal diluting segment.44 With normal renal function andmaximal AVP suppression, the renal capacity to excretesolute-free water is, therefore, enormous. Because mostHF patients become hyponatremic with only 2-3 L/d of fluidintake, the nonosmotic release of AVP, rather than intrarenalhemodynamic abnormalities, is likely to be the dominantfactor in the pathogenesis of hyponatremia in HF. Studieswith AVP antagonists demonstrating a prompt correction inthe serum [Naþ] in patients with HF support this notion (seethe subsequent section: Vasopressin Receptor Antagonists).

Cirrhosis. Hyponatremia occurs commonly in patients withadvanced cirrhosis, but rarely in the absence of ascites.106

The pathophysiology of hyponatremia in cirrhosis is asso-ciated with portal hypertension and a resultant arterialvasodilation of the splanchnic circulation.107 Among severalputative mediators, the most compelling evidence implicatesendothelial or inducible nitric oxide synthase with increasednitric oxide.108 As a result of the vasodilation, arterialstretch receptors in the carotids and aortic arch are unloadedwith a resultant decrease in the CNS tonic inhibition ofsympathetic efferent outflow. This arterial underfilling re-sults in activation of the sympathetic nervous system and therenin-angiotensin-aldosterone system, as well as the non-osmotic secretion of AVP. The net effect of this neurohu-moral activation is renal vasoconstriction with attenuation ofsystemic vasodilation and sodium and water retention,resulting in hyponatremia.109

Although the decrease in GFR and increased tubular fluidreabsorption diminish the kidneys’ maximal capacity toexcrete solute-free water in patients with cirrhosis, the majormediator of the hyponatremia appears to be nonosmoticAVP stimulation.51 Plasma AVP concentration has beenshown to be increased in patients with cirrhosis in thepresence of hyponatremia/hypo-osmolality that in normalsubjects would cause suppression to undetectable levels.110

Moreover, as with HF, V2R antagonists have been shown tocorrect the hyponatremia in patients with cirrhosis4 (see thesubsequent section: Vasopressin Receptor Antagonists).

Acute Kidney Injury, Chronic Kidney Disease, andNephrotic Syndrome. Even with complete suppression ofAVP release, hyponatremia may occur in acute kidneyinjury as a result of the diminished GFR. In oliguric ornonoliguric acute kidney injury, the urine output is rela-tively fixed, and water intake in excess of urine output andinsensible losses will cause hyponatremia. Patients withadvanced chronic kidney disease (CKD) are also more proneto develop hyponatremia than individuals with normalkidney function for the same reason. A study of 655,493patients with CKD and a mean estimated GFR of 50.2 �14.1 mL/min/1.73 m2 demonstrated a 13.6% prevalence ofhyponatremia at baseline; 26% of patients had �1 episodeof hyponatremia during a 5.5-year follow-up.111 The

decreased serum [Naþ] in these patients with CKD corre-lated with increased mortality independent of comorbidconditions. Whether nonosmotic stimulation of AVP isinvolved in patients with CKD who have hyponatremia isnot known, but certainly the decreased GFR mustcontribute. In patients with end-stage renal disease who areon dialysis, it has been shown that predialysis hyponatremiawas present in 29.3% of patients and correlated withincreased mortality.112 This relationship was independentof the mode of hemodialysis, ultrafiltration volume, HF, orvolume overload.

Hyponatremia with nephrotic syndrome has been lessfrequently reported, perhaps because many of these patientshave normal kidney function. Moreover, volume overload inpatients with nephrotic syndrome may suppress AVPsecretion.113 However, when serum albumin concentrationfalls below 2 g/dL, intravascular hypovolemia may causenonosmotic stimulation of AVP secretion and lead tohyponatremia.

RATE OF CORRECTION OF HYPONATREMIANo data suggest that the etiology of the hyponatremia or themethod used to correct hyponatremia influence susceptibil-ity to complications from overly rapid correction. Conse-quently, the rate of correction of hyponatremia must betaken into account before deciding the most appropriatetherapy for any hyponatremic patient.

Brain Adaptation to HyponatremiaTo understand the scientific rationale supporting guidelinesfor correcting hyponatremia, and why the consequences andtreatment of acute and chronic hyponatremia differ, it isessential to appreciate how the brain adapts to hyponatremiaand the time course over which this process occurs.

Acute versus Chronic Hyponatremia. Treatment regi-mens for hyponatremia should always respect the patho-physiology of the disease. Because intracellular andextracellular osmolality must be equal, cells either swellwith water or extrude solutes when the serum [Naþ] islow.114,115 Given the confines of the skull, cell swelling ismost important in the brain. When hyponatremia developsquickly over several hours, the ability of the brain to adaptis exceeded, and cerebral edema may result. Thus, patientswith acute (


first fatality from acute postoperative hyponatremia wasreported in 1936, and an account of the first successfultreatment was published by the same author 2 years later, inwhich a woman was rescued from her moribund conditionby the bolus infusion of 130 mL of 5% saline (NaCl),enough to increase the serum [Naþ] by about 4 mmol/L.119

Over the ensuing decades, few refinements were made tothis approach. To avoid confusion with 5% dextrose inwater, 5% NaCl has been largely replaced with 3% NaCl. In2005, a consensus conference convened to develop treat-ment guidelines for acute water intoxication from EAH incompetitive runners advocated treatment with a 100-mLbolus of 3% NaCl, enough to increase the serum [Naþ]approximately 2 mmol/L.86 A small, quick increase in theserum [Naþ] (2-4 mmol/L) is effective in treating acutehyponatremia because reducing brain swelling even slightlywill substantially decrease intracerebral pressure.120

Osmotic Demyelination. The adaptation that permitssurvival in chronic hyponatremia also makes the brainvulnerable to injury from overzealous therapy. When hypo-natremia is corrected too rapidly, the brain’s ability torecapture lost organic osmolytes can be outpaced, leadingto osmotic demyelination.121,122 Complications of rapidcorrection of chronic hyponatremia were first recognized inthe 1970s. Clinical observations in patients with centralpontine and extrapontine myelinolysis led to experimentalstudies showing that the human disorder could be reproducedin chronically hyponatremic dogs, rabbits, and rats. Animalswith severe, uncorrected chronic hyponatremia do notdevelop brain lesions, which confirms that myelinolysis is acomplication of the rapid correction of hyponatremia and notthe electrolyte disturbance itself. Similarly, demyelinationcan occur occasionally in patients who develop acutehypernatremia, and it can be induced by rapid induction ofsevere hypernatremia in normonatremic animals.123,124

The neurological complications of chronic hyponatremiapresent in a stereotypical biphasic pattern that has beencalled the osmotic demyelination syndrome (ODS).121 Pa-tients initially improve neurologically with correction ofhyponatremia, but then, one to several days later, new,progressive, and sometimes permanent neurological deficitsemerge. Most patients with ODS survive, and those withpersistent deficits can be diagnosed with magnetic resonanceimaging.125 Several lines of evidence have linked thepathogenesis of myelinolysis to the slow reuptake of organicosmolytes by the brain, which can predispose to disruptionof the bloodebrain barrier and influx of immune-competentproteins.126 In experimental models, brain regions that areslowest to recover osmolytes are the most severely affectedby myelinolysis.127 Uremia protects against myelinolysis,and brain osmolytes are recovered more rapidly duringcorrection of hyponatremia in uremic animals than in non-uremic animals.128 Finally, infusion of myo-inositol—amajor osmolyte lost in the adaptation to hyponatremia—protects against mortality and myelinolysis from rapidcorrection of hyponatremia in rats.129

The risk of ODS varies, depending on several factors. Itis highly unlikely to occur in patients who have beenhyponatremic for

Expert Panel Recommendation: Treatment of SymptomaticAcute Hyponatremia

- Indications:� Self-induced acute water intoxication (eg, psychiatricdiseases such as acute psychosis or schizophrenia,endurance exercise, “ecstasy” use);

� Known duration of hyponatremia 25 mmol/L during a 48-hour interval.

Current published estimates of the recommended 2-daylimit are actually quite similar—18 mmol/L versus 15-20 mmol/L within 48 hours.130,131,137 However, a 2-daylimit can also be confusing. These limits are sometimesexpressed as applying to the first 2 days of therapy, ignoringthe possibility that initial therapy in the first day might bedelayed or ineffective and might be followed by a largeincrease in serum [Naþ] on subsequent days. It is unlikelythat the adverse neurological events caused by a large os-motic insult will be lessened by such a delay. In fact, if theduration of severe hyponatremia were prolonged before alarge increase in serum [Naþ], chronicity would be expectedto enhance the brain’s vulnerability to injury.129,147 A 2-dayincrement is also difficult to implement in practice becauseclinicians instinctively base their treatments on changes thathave occurred since the previous day.

A 1-day increase of 12 mmol/L/d was initially proposedbased on a literature review and observational studies ofoutcomes in patients with severe hyponatremia.121 The samelimit was recently validated in a single-center observationalstudy of 255 patients with serum [Naþ] �120 mmol/L.141Four patients with typical ODS were identified, all pre-senting initially with serum [Naþ] 12 mmol/L/d; noneurological sequelae were observed among 118 patients(85%) corrected by �12 mmol/L/d. Other evidence, how-ever, suggests that this 1-day limit may be too high,particularly for patients with severe malnutrition, alco-holism, or advanced liver disease who may be especiallysusceptible to osmotic demyelination.148 Although notrigorously proven to increase the susceptibility to ODS,alcoholism, hypokalemia, malnutrition, and liver disease are

Table 3 Factors That Place Patients at High Risk of Developingthe Osmotic Demyelination Syndrome with Correction of ChronicHyponatremia

High Risk of Osmotic Demyelination Syndrome� Serum sodium concentration �105 mmol/L� Hypokalemia*� Alcoholism*� Malnutrition*� Advanced liver disease*

L ¼ liter; mmol ¼ millimole.*Unlike the rate of increase in serum sodium concentration, neither

the precise level of the serum potassium concentration nor the degree ofalcoholism, malnutrition, or liver disease that alters the brain’s toleranceto an acute osmotic stress have been rigorously defined.


present in a high percentage of patients who develop thesyndrome after correction of hyponatremia. Unlike the rateof increase in serum [Naþ], neither the precise level of theserum [Kþ] nor the degree of alcoholism, liver disease, ormalnutrition that alter the brain’s tolerance to an osmoticstress has been (or perhaps can be) defined. Cliniciansshould be extra cautious about raising the serum [Naþ]when it is known or suspected that a patient harbors theserisk factors to any significant degree. A prospectivecohort study of 184 consecutive patients with serum [Naþ]�120 mmol/L confirmed that sequelae were associated withmore rapid correction; but, of the 9 patients with sequelaewhose serum [Naþ] was measured during the first 24 hoursof correction, 3 had been corrected by 12 mmol/L, 2 by11 mmol/L, and 1 by 10 mmol/L.139 Similarly, case reportsand case series of patients with ODS have included a fewpatients corrected by

Expert Panel Recommendation:Managing Excessive Correc-tion of Chronic Hyponatremia

- Starting serum [Naþ] �120 mmol/L: Interventionprobably unnecessary.

- Starting serum [Naþ] 8 mmol/L;

� Consider therapeutic re-lowering of serum [Naþ] ifcorrection exceeds therapeutic limits;

� Consider administration of high-dose glucocorticoids(eg, dexamethasone, 4 mg every 6 hours) for 24-48hours following the excessive correction.

- Re-lowering serum [Naþ]:� Administer desmopressin to prevent further waterlosses: 2-4 mg every 8 hours parenterally;

� Replace water orally or as 5% dextrose in waterintravenously: 3 mL/kg/h;

� Recheck serum [Naþ] hourly and continue therapyinfusion until serum [Naþ] is reduced to goal(Figure 3).

Expert Panel Recommendation: Avoiding Osmotic Demye-lination Syndrome (ODS) in Patients with ChronicHyponatremia

- Population at risk: hyponatremia with serum [Naþ]�120 mmol/L of >48 hours’ duration; for example,outpatients drinking conventional volumes of water ortreated with thiazides and hospital-acquired hypona-tremia with a known duration of >48 hours.

- Increased vigilance in patients at heightened risk ofODS (see Table 3).

- Goal:� Minimum correction of serum [Naþ] by 4-8 mmol/Lper day, with a lower goal of 4-6 mmol/L per day ifthe risk of ODS is high.

- Limits not to exceed:� For high risk of ODS: 8 mmol/L in any 24-hourperiod;

� For normal risk of ODS: 10-12 mmol/L in any 24-hour period; 18 mmol/L in any 48-hour period.


self-induced water intoxication, careful monitoring andtherapeutic interventions to prevent and reverse over-correction are indicated for patients with a serum [Naþ]�120 mmol/L, particularly those with comorbidities thatincrease the risk of osmotic demyelination (Table 3,Figure 3). Serum [Naþ] measurements at 4- to 6-hour in-tervals and monitoring of urine volume are advisable untilmildly hyponatremic levels of �125 mmol/L have beenreached. In high-risk patients, correction by more than8 mmol/L/d should be actively avoided; whereas in patientswithout major risk factors for osmotic demyelination,correction by 8-12 mmol/L in the first day of therapy isgreater than necessary but is unlikely to cause harm as longas the 2-day increment does not exceed 18 mmol/L. If theday’s increase has exceeded 8 mmol/L, active therapies toraise the serum [Naþ] any further should be avoided for thenext 24 hours.

To prevent overcorrection, ongoing measures to in-crease the serum [Naþ] (eg, saline or vaptan therapy)should be temporarily withheld once the targeted dailyincrease has been achieved. For the rest of the day,further correction from urinary free water losses shouldbe prevented either by replacing losses with 5% dextrosein water or oral water or by terminating further urinarylosses by administering 2-4 mg of desmopressin paren-terally (Figure 3). Alternatively, rather than waiting foran unwelcome aquaresis in patients with potentiallyreversible causes of hyponatremia, one group has advo-cated preemptive administration of desmopressin every6-8 hours in combination with a slow infusion of 3%saline titrated to achieve a 6-mmol/L/d increase in serum[Naþ]. This strategy creates a state of iatrogenic SIADHand permits a controlled increase in the serum [Naþ]with a low risk of inadvertent overcorrection; desmo-pressin is stopped once the serum [Naþ] has been raisedto 128 mmol/L.155

Desmopressin is not a reliable therapeutic option forpatients corrected with vaptans, but urinary water lossesusually stop after the drug is metabolized. Correction by>12 mmol/L/d is uncommon with vaptan therapy, and nocases of osmotic demyelination have been reported aftervaptan therapy alone (without concurrent saline therapy).Nonetheless, it is prudent to withhold the next day’s doseafter a large increase in serum [Naþ], with resumption of thesame dose or a lower dose in subsequent days.

If overcorrection occurs, therapeutic re-lowering of theserum [Naþ] can be considered, but it has not been validatedin controlled trials. Re-lowering of the serum [Naþ] pre-vents osmotic demyelination in experimental animals156 andhas been shown to be well tolerated in a small series ofpatients.152 It can be achieved by administering 2-4 mg ofdesmopressin in combination with repeated 3-mL/kg in-fusions of 5% dextrose in water administered over 1 hour—measuring the serum [Naþ] after each infusion to determinethe need for more 5% dextrose in water—until the serum[Naþ] has been returned to a level below the therapeuticlimit for the patient. In the absence of risk factors forosmotic demyelination, a limit of 10-12 mmol/L in any24-hour period or 18 mmol/L in any 48-hour period ap-pears to be appropriate; in high-risk patients (Table 3),correction by >8 mmol/L in any 24-hour period may bejustification for therapeutic re-lowering (Figure 3). Studiesin experimental animals have also shown benefit fromadministration of high doses of glucocorticoids to stabilizeand prevent osmotic disruption of the bloodebrain barrier,but efficacy of this approach has not been verified inhuman patients.157


VASOPRESSIN RECEPTOR ANTAGONISTS(VAPTANS)Vaptans have long been anticipated as a more effectivemethod to treat hyponatremia by virtue of their unique effectto selectively increase solute-free water excretion by thekidneys.158 The recent approval by the FDA of 2 suchagents, conivaptan and tolvaptan, for clinical use and theapplication of a third drug, lixivaptan, mark the beginning ofa new era in the management of hyponatremic disorders.Intelligent use of vaptans for FDA-approved indications willneed to be based on existing knowledge of the pathophys-iology of hyponatremia and a physiologic understanding ofhow these agents work gleaned from the results of clinicaltrials and accumulated experience with clinical use.

Vasopressin ReceptorsAVP receptors (AVPR) are G-protein-coupled receptors.The 3 known subtypes differ in localization and in signaltransduction mechanisms.159 The AVP V1a (V1aR) andV1b (V1bR) receptors are Gq-coupled receptors that activatephospholipase C and increase cytosolic free calcium; thephysiological effects caused by activation depend primarilyon the localization of the receptors and include vasocon-striction, platelet aggregation, ionotropic stimulation, andmyocardial protein synthesis (all V1aR) and pituitary ACTHsecretion (V1bR). V2Rs are found on the principal cells ofthe renal collecting tubules and the vascular endothelium,where they mediate the antidiuretic effects of AVP andstimulate release of von Willebrand factor and factor 8,respectively. V2R-mediated vasodilatation has also beendescribed at high concentrations of AVP. Binding of AVPto its V2R activates the Gs-coupled adenylyl cyclase system,thereby increasing intracellular levels of cyclic adenosinemonophosphate. In the kidney, this activates protein kinaseA, which then phosphorylates preformed AQP2 waterchannels localized in intracellular vesicles. Phosphorylationstimulates trafficking of the vesicles to the apical membrane,followed by insertion of AQP2 into the membrane160

(Figure 2). Activation of this signal transduction cascade is

Table 4 AVP Receptor Antagonists Evaluated in Clinical Trials

Conivaptan Lixivap

Compound YM-087 VPA-98Receptor V1a/V2 V2Route of administration IV OralUrine volume [ [Urine osmolality Y YSodium excretion/24 hours 4 4 at

highCompany developing agent Astellas Pharma US, Inc. Corner

Status FDA-approved Phasecom

[ ¼ increased; Y ¼ decreased; 4 ¼ no change; AVP ¼ arginine vasopAdministration; IV ¼ intravenous; V1a ¼ vasopressin receptor 1a; V2 ¼ vasopr

necessary to render the collecting duct permeable to water.AQP2 membrane insertion and transcription, and hence,apical membrane water permeability, is reduced when AVPis absent or chronically suppressed.

Mechanism of ActionBinding of the antagonists to V2R blocks activation of thereceptor by endogenous AVP. The increased urine outputproduced by the V2R antagonists is quantitatively equiva-lent to that of diuretics such as furosemide; qualitatively it isdifferent in that only water excretion results and excretion ofurinary solutes is not augmented.161 Thus, V2R antagonistsproduce solute-sparing water excretion in contrast to classicdiuretic agents that block distal tubule sodium transporters,leading to simultaneous electrolyte and water losses. For thisreason, the renal effects produced by V2R antagonists havebeen termed aquaretic to distinguish them from the renaleffects produced by classical diuretic agents, which arenatriuretic and kaliuretic as well. This is not simply a se-mantic issue, because appreciating these important differ-ences in renal effects is crucial for the intelligent clinical useof AVP receptor antagonists. For example, the negativewater balance induced by aquaretic agents has less adverseeffect on neurohormonal activation and renal function thancomparable degrees of urine output induced by loop diureticagents, because only one third of the negative water balanceinduced by aquaretics derives from the ECF, whereas twothirds comes from intracellular water.162

Vaptans in Clinical Use and DevelopmentFour nonpeptide agents have been studied in clinical trials(Table 4). Conivaptan is a combined V1aR and V2Rantagonist, while all of the others are selective V2Rantagonists.

Conivaptan is FDA approved for euvolemic and hyper-volemic hyponatremia in hospitalized patients. It is availableonly as an intravenous preparation and is given as a 20-mgloading dose over 30 minutes, followed by a continuousinfusion of 20 or 40 mg/d.163 Generally, the 20-mg

tan Satavaptan Tolvaptan

5 SR-121463 OPC-41061V2 V2Oral Oral[ [Y Y

low dose [ atdose

4 4

stone Sanofi-Aventis Otsuka AmericaPharmaceutical, Inc.

3 studiespleted

Developmentsuspended

FDA- and EMA-approved

ressin; EMA ¼ European Medicines Agency; FDA ¼ US Food and Drugessin receptor 2.


continuous infusion is used for the first 24 hours to gaugethe initial response. If the correction of serum [Naþ] is felt tobe inadequate (eg, 12 mmol/L/d. The maximum correction limit should bereduced to 8 mmol/L over the first 24 hours in patients withrisk factors for osmotic demyelination, as stressed previ-ously (Table 3). The most common side effects of con-ivaptan include headache, thirst, and hypokalemia.164

Tolvaptan, an oral V2R antagonist, is also FDA ap-proved for the treatment of euvolemic and hypervolemichyponatremia. In contrast to conivaptan, the availability oftolvaptan in tablet form allows both short- and long-termuse.4 Similar to conivaptan, tolvaptan treatment must beinitiated in the hospital so that the rate of correction can bemonitored carefully. In the US, patients with a serum [Naþ]30 days such asSALTWATER (a multicenter, open-label extension ofSALT-1 and SALT-2) and EVEREST.40,167 Based largelyon the hepatic injury noted in the TEMPO trial, on April 30,2013 the FDA recommended that: “[tolvaptan] treatmentshould be stopped if the patient develops signs of liverdisease. Treatment duration should be limited to 30 days orless, and use should be avoided in patients with underlyingliver disease, including cirrhosis.”168 The EMA hasapproved the use of tolvaptan for SIADH but not forhyponatremia due to heart failure or cirrhosis. Based on theTEMPO trial results, the EMA has also issued a warningabout the possible occurrence of hepatic injury in patientstreated with tolvaptan, but it did not recommend any re-striction on the duration of treatment of SIADH patientswith tolvaptan.169 Accordingly, the authors believe thatappropriate caution should be exercised in patients treatedwith tolvaptan for hyponatremia for extended periods (eg,>30 days), but this decision should be based upon theclinical judgment of the treating physician. Patients who arerefractory to or unable to tolerate or obtain other therapiesfor hyponatremia, and in whom the benefit of tolvaptan


treatment outweighs the risks, remain candidates for long-term therapy with tolvaptan; but in such cases, liver functiontests should be monitored carefully and serially (ie, every 3months), and the drug discontinued in the event of signifi-cant changes in liver function tests (ie, 2� ULN of ALT).With rare exception, tolvaptan should not be used in patientswith underlying liver disease given the difficulty of attrib-uting causation to any observed deterioration of hepaticfunction. One such exception may be hyponatremic patientswith end-stage liver disease awaiting imminent liver trans-plantation, who are at little risk of added hepatic injury andwill benefit from correction of hyponatremia before surgeryto decrease the risk of ODS postoperatively.170

Expert Panel Recommendations: Hyponatremia fromGastro-intestinal Losses

- Because urine [Naþ] may be high if vomiting leads toobligate urinary bicarbonate loss, urine chloride shouldbe measured if vomiting is present to confirm thepresence of solute and volume depletion.

- This is typically a chronic hyponatremia, so currentlimits for rate of correction of chronic hyponatremiasshould be observed (see Current Recommendations forRate of Correction of Hyponatremia).

- After any urgent fluid resuscitation to stabilize bloodpressure, tailor the repletion fluid to correct accom-panying potassium or base deficits; use potassium-supplemented fluid or custom-formulated fluidincorporating sodium bicarbonate; remember thatpotassium administration will also increase the serum[Naþ].

- Monitor serum [Naþ] increase and urine volumefrequently and urine osmolality as needed; switch tohypotonic fluid to retard the rate of correction once thecorrection approaches goal (Figure 3).

THERAPY OF HYPONATREMIAS

Hypovolemic HyponatremiaThe first and key step in the successful treatment of hypo-volemic hyponatremia is to establish that volume depletionis present. Once this is done, treatment is straightforward—with correction of the volume deficit, the relative waterexcess will correct itself via a water diuresis. Indeed, withsevere hyponatremia due to volume depletion, the bulk ofthe treatment effort may be devoted to the prevention of anoverly rapid increase in serum [Naþ] due to the ensuingspontaneous aquaresis. Monitoring urine volume andosmolality will permit detection of the aquaresis and allowclinicians to anticipate and avoid an unduly rapid rate ofincrease in serum [Naþ] (see previous section: ManagingExcessive Correction of Chronic Hyponatremia).

When ECF volume depletion is obvious and potentiallylife threatening, resuscitation with isotonic fluid will likelyhave been begun empirically even before the results ofroutine laboratory tests have been returned. Volumeexpansion should be continued until blood pressure isrestored and the patient has clinical euvolemia. Not infre-quently, the initial volume estimate is equivocal, and bothvolume depletion and SIADH remain as diagnostic con-siderations. In that circumstance, a fluid challenge can beboth diagnostic and therapeutic. With volume depletion,administering isotonic saline leads to an increase in both theserum and urine [Naþ] once intravascular volume has beenrestored. If SIADH is present, administering saline also re-sults in an increase in urine [Naþ]. However, serum [Naþ]concentration may actually fall with isotonic saline admin-istration as the administered sodium is excreted in a smallvolume of concentrated urine and the water is retained.Maximum urine-concentrating ability is insufficient topermit net water retention if hypertonic saline is used. Incases where there is even a remote possibility that the pri-mary diagnosis is SIADH and either significant CNSsymptoms from hyponatremia are present or the startingserum [Naþ] is �120 mmol/L, hypertonic saline (eg, 3%NaCl) should be used for the initial diagnostic volumechallenge to avoid any risk of lowering the serum [Naþ]further.

Hypovolemic hyponatremia is nearly always chronicrather than acute, and the current limits for rate of correctionof chronic hyponatremias should be carefully observed (seeprevious section: Rate of Correction of Hyponatremia).Current recommendations for treating hyponatremia in pa-tients with specific disorders associated with hypovolemiaare given below.

Gastrointestinal Disease. Hyponatremia associated withgastrointestinal fluid loss is seldom acute or severe enoughin its own right to require hypertonic saline for urgentcorrection. Isotonic saline is the mainstay of treatment.Potassium chloride should be added if hypokalemia andmetabolic alkalosis are present due to vomiting, and anisonatric mixture of NaCl and sodium bicarbonate can beused when metabolic acidosis is present because of diarrhea.Potassium is exchangeable with intracellular sodium, sosupplementation to correct hypokalemia will raise serum[Naþ] to the same degree as sodium repletion. Potassiumdosing should be taken into account in predicting the rate atwhich the serum [Naþ] increases in response to treatment.Specific therapy for the underlying disorder should beinitiated, and antiemetic and antidiarrheal agents can be usedas appropriate.

Diuretic Therapy. Thiazides produce hyponatremia by atleast 3 separate mechanisms. They interfere with function ofthe distal tubule diluting site, produce volume depletion thatstimulates nonosmotic AVP release, and deplete potassiumleading to cellular uptake of sodium. As these disorders arereversed by withholding the diuretic and correcting sodiumand potassium deficits, the serum [Naþ] may increase veryrapidly in association with development of an aquaresis.ODS and an increased risk of death with rapid correction of

Expert Panel Recommendations: Diuretic-InducedHyponatremia

- Diuretic-induced hyponatremia is always a chronichyponatremia, so current limits for rate of correction ofchronic hyponatremias should be observed (see Cur-rent Recommendations for Rate of Correction ofHyponatremia).

- Thiazides interfere with urinary dilution. Discontinua-tion of thiazides and correction of volume deficits maybe followed by a rapid, spontaneous water diuresis thatcan raise serum [Naþ] very quickly; numerous cases ofODS have been reported after correction of severethiazide-induced hyponatremia.

- Serially follow changes in urine osmolality togetherwith urine volume to detect the development of anaquaresis with heightened risk of overly rapidcorrection.

- The focus of therapy for patients with serum [Naþ]125mmol/L; once the serum [Naþ] has reached 125 mmol/L, the risk of hyponatremia-related CNS complicationsis low; if the starting serum [Naþ]


intravascular volume, as stated above. Isotonic saline,hypertonic saline, and supplemental oral NaCl have beenadvised, together with fludrocortisone in some in-stances.176,177 Because the response to oral NaCl and flu-drocortisone is unpredictable, we favor the use ofhypertonic saline to correct hyponatremia (if its severitywarrants) and isotonic saline for maintenance of intravas-cular volume.

Mineralocorticoid Deficiency. Volume repletion withisotonic saline will be required initially in patients withprimary adrenal insufficiency. Fludrocortisone is usedchronically for mineralocorticoid replacement, which willprevent hypovolemia-induced hyponatremia from recurring.Hyporeninemic hypoaldosteronism (type IV renal tubularacidosis) is characterized by volume expansion and is not acause of hyponatremia; acquired mineralocorticoid defi-ciency severe enough to lead to volume depletion andhyponatremia occurs only with bilateral adrenal failure fromadrenal destruction or adrenalectomy. As such, patientspresenting with features of mineralocorticoid deficiencyshould be suspected to have glucocorticoid deficiency aswell. The latter deficiency should be treated urgently withglucocorticoid at stress doses (eg, 50-100 mg of hydrocor-tisone given parenterally every 8 hours) while definitivetesting results are awaited. Because stress doses of hydro-cortisone also activate the mineralocorticoid receptors,replacement with fludrocortisone is not required until

Expert Panel Recommendations: Hyponatremia from CSW

- The overwhelming majority of patients in the neuro-surgical setting with hyponatremia after subarachnoidhemorrhage, trauma, or surgery have SIADH, notCSW.

- Reduced BUN and uric acid values are features of bothCSW and SIADH and cannot be used to distinguishbetween these disorders.

- Diagnosing CSW requires demonstration of a period ofinappropriate renal sodium and fluid loss preceding thedevelopment of volume depletion and hyponatremia; ahigh urine output and urinary sodium content duringsodium infusion alone are insufficient evidence becausea patient with SIADH will excrete any administeredsodium and fluid to maintain balance.

- To distinguish between SIADH and CSW, the responseto a cautious reduction in fluid supplementation shouldbe observed; CSW patients will develop signs of vol-ume depletion, while SIADH patients will demonstratereduced urine output while remaining euvolemic.

- Intravenous sodium supplementation is preferred forpatients with unreliable oral intake; isonatric fluidssuffice, once volume depletion has been corrected inCSW; a high-sodium diet or NaCl tablets may be usefulin the rare patient with CSW who has satisfactory oralintake.

the patient is titrated to lower replacement doses of gluco-corticoids (see subsequent section: Therapy of Hypona-tremias, Glucocorticoid Deficiency).

Euvolemic HyponatremiaAs with other forms of hyponatremia, the treatment ofpatients with euvolemic hyponatremia will vary greatly,depending on 3 main aspects of their presentation:

1. The treatment of the underlying condition that hasprecipitated the hyponatremia.

diagnosis, evaluation, and treatment of hyponatremia ... e… · experts in hyponatremia convened...

Documents