Annals Heart Failure-In the Clinic

inthe

clinicin the clinic

Heart Failure

Prevention page ITC6-2

Diagnosis page ITC6-4

Treatment page ITC6-6

Practice Improvement page ITC6-14

CME Questions page ITC6-16

Section EditorBarbara Turner, MD, MSED Sankey Williams, MDDarren Taichman, MD, PhD

Physician WriterLee R. Goldberg, MD, MPH

The content of In the Clinic is drawn from the clinical information and education resources of the American College of Physicians (ACP), includingPIER (Physicians’ Information and Education Resource) and MKSAP (MedicalKnowledge and Self-Assessment Program). Annals of Internal Medicineeditors develop In the Clinic from these primary sources in collaborationwith the ACP’s Medical Education and Publishing Division and with the assistance of science writers and physician writers. Editorial consultants fromPIER and MKSAP provide expert review of the content. Readers who are interested in these primary resources for more detail can consult http://pier.acponline.org, http://www.acponline.org/products_services/mksap/15/?pr31, and other resources referenced in each issue of In the Clinic.

CME Objective: To review current evidence for the prevention, diagnosis, and treatment of heart failure.

The information contained herein should never be used as a substitute for clinical judgment.

© 2010 American College of Physicians

What are the risk factors forheart failure?Over the past decade, treatment ofheart failure has shifted from focus-ing on acute exacerbations whenthe patient is “in” heart failure totreating heart failure as a chronicand potentially preventable syn-drome. In the current model, thereare risk factors that lead to heartfailure, and modifying these riskfactors can prevent symptoms anddelay death. In addition, aggressivetreatment can improve both thequantity and quality of life. TheAmerican College of Cardiology(ACC) and American Heart Asso-ciation (AHA) have developed astaging system to help clinicians select therapies that improve out-comes for those at risk for or suf-fering from heart failure (3) (Box).

The incidence of heart failure ap-proaches 10 per 1000 persons olderthan 65 years. At age 40 years, thelifetime risk for heart failure forboth men and women is 1 in 5, andthe same lifetime risk exists at age80 years despite a much shorter lifeexpectancy (1).

African Americans also face an increased risk for heart failure.African Americans between age 45and 64 years are 2.5 times morelikely to die of heart failure thanwhite persons in the same agerange (4). Men have a higher rateof heart failure than women,

although this difference narrows aswomen get older.

Treat conditions and behaviors thatare known to increase the risk forheart failure (Box). In addition, anepidemiologic study has linked in-creased risk for heart failure tophysical inactivity, obesity, and low-er levels of education (5).

HypertensionOf persons presenting with heartfailure, 75% have hypertension (1).Long-standing untreated hyperten-sion is associated with both systolicand diastolic heart failure and is anindependent risk factor for coronaryartery disease (CAD). Clinical tri-als have shown that a reduction insystolic or diastolic blood pressurecan reduce the risk for heart failure(6). Even modest decreases in sys-tolic blood pressure reduce mortali-ty and risk for heart failure (7).

Type 2 diabetes mellitusDiabetes markedly increases therisk for heart failure and is an inde-pendent risk factor for CAD. Inthe population-based Reykjavikcross-sectional study of 19 381 participants, heart failure was diag-nosed in 3.2% of all persons com-pared with 6.0% and 11.8% of persons with abnormal glucose regulation and type 2 diabetes, respectively (8).

The HOPE (Heart Outcomes PreventionEvaluation) trial found that, in patients atleast 55 years old with either atherosclerosis

© 2010 American College of Physicians ITC6-2 In the Clinic Annals of Internal Medicine 1 June 2010

Approximately 5 million persons in the United States have heart fail-ure, and according to the National Heart Lung and Blood Institute,the number is increasing. Heart failure is the most frequent cause of

hospitalization in U.S. patients older than 65 years and leads to about 300 000deaths per year (1). Heart failure is a major problem in the rest of the worldas well, but few accurate data are available. The most common cause of heartfailure in industrialized countries is ischemic cardiomyopathy, whereas othercauses, such as infectious diseases, assume a larger role in underdevelopedcountries. Despite recent advances in management of heart failure, the 30-day, 1-year, and 5-year mortality rates after hospitalization for heart failure are 10%, 22%, and 42%, respectively (2).

Prevention

The ACC/AHA Guidelines forStages of Heart FailureStage A: At risk for heart failure

(coronary artery disease,hypertension, or diabetesmellitus) but has yet to showimpaired left ventricular functionor hypertrophy.

Stage B: Asymptomatic leftventricular dysfunction (has neverhad symptomatic heart failure).

Stage C: Current or past symptomsof heart failure associated withunderlying structural heartdisease.

Stage D: Has truly refractory heartfailure and may be eligible forspecialized, advanced treatmentstrategies, such as mechanicalcirculatory support, procedures tofacilitate fluid removal,continuous inotropic infusions, orcardiac transplantation or otherinnovative or experimentalsurgical procedures, or for end-of-life care, such as hospice.

Common Conditions andBehaviors that Increase the Riskfor Heart Failure• Hypertension• Diabetes• Cardiotoxic substance use• Hyperlipidemia• Thyroid disorders• Tachycardia• Coronary artery disease

© 2010 American College of PhysiciansITC6-3In the ClinicAnnals of Internal Medicine1 June 2010

or diabetes and at least 1 other risk factorbut no history of heart failure, the an-giotensin-converting enzyme (ACE) in-hibitor ramipril reduced the risk for stroke,myocardial infarction (MI), and death fromcardiovascular disease by 22% while alsosignificantly reducing heart failure (7).

Cardiotoxic substance useAlcohol is a direct myocardial toxinand can be the primary cause ofheart failure. Abstinence from alco-hol may reverse left ventricular dys-function (9). Despite the lack ofclinical trials, most clinicians rec-ommend abstinence from alcoholor limited alcohol intake for pa-tients with left ventricular dysfunc-tion. Tobacco and cocaine use significantly increase the risk forCAD, which can lead to heart fail-ure. Cocaine also has direct effectson the myocardium. Chemothera-peutic agents, such as anthracyclineand trastuzumab, can also exerttoxic effects on the myocardium.

HyperlipidemiaHyperlipidemia is strongly associat-ed with CAD, which may lead toheart failure. Large-scale clinical tri-als have shown the benefit of lipidlowering for primary and secondaryprevention of cardiovascular events.

The CARE (Cholesterol and RecurrentEvents) trial found that pravastatin signifi-cantly reduced the incidence of heart fail-ure, subsequent cardiovascular events,and mortality (10).

Thyroid disordersBoth hyperthyroidism and hypo-thyroidism are associated withheart failure, and restoration of aeuthyroid state can potentially re-turn ventricular function to normal(11, 12). Hyperthyroidism is associ-ated with atrial fibrillation andtachycardia, which may complicateor worsen heart failure.

TachycardiaStudies have shown that rapid pro-longed ventricular rates can lead tocardiomyopathy. Restoration of nor-mal rhythm or rate control in pa-tients with poorly controlled

atrial fibrillation and other supraven-tricular tachycardias can improvefunction and potentially prevent leftventricular dysfunction (13-15).

Coronary artery diseaseBecause coronary disease is a majorrisk factor for heart failure, aggres-sive risk-factor modification withcholesterol-lowering drugs and aspirin, ACE inhibitors, and β-blockers can significantly reducemortality and the risk for futurecardiovascular complications, in-cluding heart failure. Although hy-pertension is the most common riskfactor for heart failure, antecedentMI is a very close second (16).

The CAPRICORN (Carvedilol Post-InfarctSurvival Control in Left Ventricular Dys-function) trial showed that the β-blockercarvedilol significantly reduced mortalityin patients with left ventricular dysfunctionwith or without heart failure after MI andwho also received ACE inhibitors, revascu-larization, and aspirin (17).

What medications should be usedfor primary prevention of heartfailure?Several classes of medications havebeen shown to prevent heart failurein at-risk populations. These in-clude hydroxymethylglutaryl co-enzyme A reductase inhibitors inpatients with hyperlipidemia, ACEinhibitors in patients with diabetes,and nearly all antihypertensivemedications when used to lowerblood pressure to goal levels.

What nondrug interventionsshould be used for primaryprevention of heart failure?There is increasing evidence thatobesity and decreased physical activi-ty can increase the risk for heart fail-ure (5). Maintaining a healthyweight and exercising regularly canhelp lower blood pressure and lipids,both of which reduce the risk forcoronary disease. Obstructive sleepapnea is also associated with hyper-tension and a much higher incidenceof heart failure and cardiovasculardisease in general (18). Treatment of

1. Writing group mem-bers. Heart diseaseand stroke statis-tics—-2010 update: areport from theAmerican Heart Asso-ciation. Circulation.2010;121:e46-e215.[PMID: 20019324]

2. Loehr LR, RosamondWD, Chang PP, et al.Heart failure inci-dence and survival(from the Atheroscle-rosis Risk in Commu-nities study). Am JCardiol.2008;101:1016-22.[PMID: 18359324]

3. Hunt SA, AbrahamWT, Chin MH, et al.2009 focused updateincorporated into theACC/AHA 2005Guidelines for the Di-agnosis and Manage-ment of Heart Failurein Adults: a report ofthe American Collegeof Cardiology Foun-dation/AmericanHeart AssociationTask Force on Prac-tice Guidelines: de-veloped in collabora-tion with theInternational Societyfor Heart and LungTransplantation. Cir-culation.2009;119:e391-479.[PMID: 19324966]

4. Centers for DiseaseControl and Preven-tion (CDC). Mortalityfrom congestiveheart failure—-Unit-ed States, 1980-1990.MMWR Morb MortalWkly Rep. 1994;43:77-81. [PMID: 8295629]

5. He J, Ogden LG, Baz-zano LA, et al. Riskfactors for congestiveheart failure in USmen and women:NHANES I epidemio-logic follow-up study.Arch Intern Med.2001;161:996-1002.[PMID: 11295963]

6. National Heart, Lung,and Blood InstituteJoint National Com-mittee on Prevention,Detection, Evaluation,and Treatment ofHigh Blood Pressure.The Seventh Reportof the Joint NationalCommittee on Pre-vention, Detection,Evaluation, and Treat-ment of High BloodPressure: the JNC 7report. JAMA.2003;289:2560-72.[PMID: 12748199]

7. HOPE Investigators.Effects of ramipril oncoronary events inhigh-risk persons: re-sults of the HeartOutcomes PreventionEvaluation Study. Cir-culation.2001;104:522-6.[PMID: 11479247]

© 2010 American College of Physicians ITC6-4 In the Clinic Annals of Internal Medicine 1 June 2010

sleep apnea may reduce this risk (19,20). Smoking cessation can signifi-cantly reduce the risk for cardiovas-cular disease, including heart failure

(21). There is no evidence that theroutine use of nutritional supple-ments can prevent left ventriculardysfunction.

8. Thrainsdottir IS, As-pelund T, Thorgeirs-son G, et al. The asso-ciation betweenglucose abnormali-ties and heart failurein the population-based Reykjavikstudy. Diabetes Care.2005;28:612-6.[PMID: 15735197]

9. Walsh CR, Larson MG,Evans JC, et al. Alco-hol consumption andrisk for congestiveheart failure in theFramingham HeartStudy. Ann InternMed. 2002;136:181-91. [PMID: 11827493]

10. Sacks FM, Pfeffer MA,Moye LA, et al. Theeffect of pravastatinon coronary eventsafter myocardial in-farction in patientswith average choles-terol levels. Choles-terol and RecurrentEvents Trial investi-gators. N Engl J Med.1996;335:1001-9.[PMID: 8801446]

11. Klein I, Ojamaa K.Thyroid hormoneand the cardiovascu-lar system. N Engl JMed. 2001;344:501-9. [PMID: 11172193]

12. Fadel BM, EllahhamS, Ringel MD, et al.Hyperthyroid heartdisease. Clin Cardiol.2000;23:402-8.[PMID: 10875028]

13. Coleman HN III, Tay-lor RR, Pool PE, et al.Congestive heartfailure followingchronic tachycardia.Am Heart J.1971;81:790-8.[PMID: 5088355]

14. Peters KG, KienzleMG. Severe car-diomyopathy due tochronic rapidly con-ducted atrial fibrilla-tion: complete re-covery afterrestoration of sinusrhythm. Am J Med.1988;85:242-4.[PMID: 3400701]

15. Grogan M, SmithHC, Gersh BJ, et al.Left ventricular dys-function due to atri-al fibrillation in pa-tients initiallybelieved to have id-iopathic dilated car-diomyopathy. Am JCardiol.1992;69:1570-3.[PMID: 1598871]

16. Levy D, Larson MG,Vasan RS, et al. Theprogression from hy-pertension to con-gestive heart failure.JAMA.1996;275:1557-62.[PMID: 8622246]

them into dilated, hypertrophic,and restrictive types. Most causesof heart failure lead to cardiac dilatation. Hypertrophic cardio-myopathy is due to genetic abnor-malities or hypertension. Restrictiveheart failure is usually due to systemic infiltrative diseases.

More important is the functional distinc-tion between systolic heart failure andheart failure with preserved left ventricularfunction. In systolic heart failure, the heartis dilated, and the ejection fraction is below50%. In heart failure with preserved leftventricular function, which occurs moreoften in elderly patients with hypertension,there is less dilatation and a normal ejec-tion fraction. In patients with heart failure,those with preserved ejection fraction areresponsible for 50% of hospitalizations andhave a similar survival rate as those withsystolic heart failure (23, 24).

What is the role of B-typenatriuretic peptide in the evaluationand treatment of heart failure?Serum levels of B-type natriureticpeptide (BNP) and N-terminal-pro-B-type natriuretic peptide(NTproBNP) increase with in-creases in ventricular volume andpressure (25) and therefore can beused as markers for ventricular vol-ume and pressure overload. NT-proBNP has a longer half-life inthe serum than does BNP. The bestdata for the usefulness of bothBNP and NTproBNP are in thesetting of acute dyspnea to deter-mine the contribution of heart fail-ure, especially with concomitant

What symptoms and signs shouldprompt clinicians to consider thediagnosis of heart failure?Patients with underlying risk factors,including CAD, valvular heart dis-ease, and long-standing hypertensionmay be asymptomatic. Do not waitfor symptoms to develop before eval-uating and treating these patients forearly left ventricular dysfunction(Stage B). There is very strong evi-dence that treatment of asympto-matic left ventricular dysfunction willdelay the onset of symptomatic heartfailure and improve survival (22).

Once structural or functional heartdisease affects the ability of the my-ocardium to fill and pump blood nor-mally, patients may develop dyspnea,fatigue, exercise intolerance, and fluidretention that leads to pulmonarycongestion and edema. The breathingdifficulties and cough of heart failureare sometimes initially misdiagnosedas bronchitis, pneumonia, or asthma,especially in young patients. Physicalsigns of heart failure may reflect theunderlying cause, as shown by elevat-ed blood pressure or an abnormal car-diac murmur, or the resulting fluid retention, as shown by elevated jugu-lar venous pressure, pulmonary crack-les, a third heart sound, and lower-extremity edema.

What are the types of heartfailure, and how should cliniciansdistinguish them?Heart failure has many causes, andit is sometimes useful to divide

Prevention... Risk factors for heart failure include hypertension; diabetes; the useof cardiotoxic substances, such as alcohol, tobacco, and cocaine; hyperlipidemia;thyroid disorders; tachycardia; and coronary artery disease. Treatment for theserisk factors reduces the risk for heart failure.

CLINICAL BOTTOM LINE

Diagnosis

17. Dargie HJ. Effect ofcarvedilol on out-come after myocar-dial infarction in pa-tients withleft-ventricular dys-function: the CAPRI-CORN randomisedtrial. Lancet.2001;357:1385-90.[PMID: 11356434]

18. Bradley TD, Floras JS.Obstructive sleepapnoea and its car-diovascular conse-quences. Lancet.2009 Jan3;373(9657):82-93.Epub 2008 Dec 26.[PMID: 19101028]

19. Shahar E, WhitneyCW, Redline S, et al.Sleep-disorderedbreathing and car-diovascular disease:cross-sectional re-sults of the SleepHeart Health Study.Am J Respir Crit CareMed. 2001;163:19-25. [PMID: 11208620]

20. Tkacova R, Rankin F,Fitzgerald FS, et al.Effects of continu-ous positive airwaypressure on obstruc-tive sleep apnea andleft ventricular after-load in patients withheart failure. Circula-tion. 1998;98:2269-75. [PMID: 9826313]

21. Lung Health StudyResearch Group. Theeffects of a smokingcessation interven-tion on 14.5-yearmortality: a random-ized clinical trial. AnnIntern Med.2005;142:233-9.[PMID: 15710956]

22. Goldberg LR, JessupM. Stage B heart fail-ure: management ofasymptomatic leftventricular systolicdysfunction. Circula-tion 2006 Jun20;113(24):2851-60.[PMID 16785351]

23. Owan TE, HodgeDO, Herges RM, et al.Trends in prevalenceand outcome ofheart failure withpreserved ejectionfraction. N Engl JMed.2006;355(3):251-9.[PMID: 16855265]

24. Bhatia RS, Tu JV, LeeDS, et al. Outcomeof heart failure withpreserved ejectionfraction in a popula-tion-based study. NEngl J Med.2006;355(3):260-9.[PMID: 16855266]

© 2010 American College of PhysiciansITC6-5In the ClinicAnnals of Internal Medicine1 June 2010

ventricular dilatation is often pres-ent. The degrees of left ventricularsystolic and diastolic dysfunctionare important in predicting progno-sis, and the treatment of systolicheart failure may differ from thetreatment of heart failure with pre-served left ventricular function.

Stress testingUse a traditional exercise stress testto evaluate patients for coronary ischemia, to quantify functional capacity, and to identify exercise-induced arrhythmias. Use pharma-cologic stress testing with dipyri-damole, dobutamine, or adenosinewith nuclear imaging or echocar-diography to look for ischemia inpatients who cannot exercise. Usemetabolic stress testing with respi-ratory gas analysis to determine theextent of disability, to differentiatebetween cardiac or pulmonary limi-tation to exercise, and to determinefunctional class in patients who arecandidates for cardiac transplanta-tion or in whom the cause of exer-cise intolerance is unclear (30).

Cardiac catheterization andendomyocardial biopsyConsider cardiac catheterization inpatients with heart failure whenechocardiography is insufficient todefine the severity of valvular heartdisease and when ischemic heart dis-ease is present or suspected. In addi-tion, do a right heart catheterizationin patients who do not respond totraditional therapies or in whompulmonary hypertension may becontributing to their symptoms. Donot perform an endomyocardialbiopsy in most patients with suspect-ed myocarditis unless giant-cell myocarditis is being considered.

Other laboratory studiesTo rule out occult thyroid disease,consider obtaining serum levels ofthyroid-stimulating hormone in allpatients with new-onset heart fail-ure. Because anemia, renal insuffi-ciency, infection, and concurrentpulmonary disease can exacerbate

pulmonary disease (26). BNP levelsalso can be elevated in women, old-er patients, persons with renal dis-ease, obese patients, and in patientswith acute MI and some noncar-diac conditions, so interpret themin the context of all available clini-cal data and do not consider themas stand-alone tests (27).

In addition, BNP levels can behelpful in risk stratification andprognosis (28). Routine monitoringof BNP and NTproBNP remainscontroversial, and studies to resolvethe controversy are underway (29).

What other tests should cliniciansconsider in the evaluation ofpatients with suspected heartfailure?ElectrocardiographyThe ACC and AHA recommendelectrocardiography (ECG) in anypatient at risk for or with a historyof cardiac disease, including new-onset or exacerbated heart failure.Whenever possible, compare thetracing with a previous baselinetracing. Results can show the pres-ence of ventricular hypertrophy,atrial abnormality, arrhythmias,conduction abnormalities, previousMI, and active ischemia.

EchocardiographyPerform 2-dimensional echocardio-graphy with Doppler in all patientswith suspected heart failure. It is akey study for determining left ven-tricular cavity size and function,identifying wall motion abnormali-ties, measuring left ventricular ejec-tion fraction (LVEF) and rightventricular function, documentingthe presence of valvular abnormali-ties, and differentiating betweensystolic heart failure and heart fail-ure in the setting of preserved leftventricular function. In heart failurewith preserved left ventricularfunction, the ejection fraction isnormal (>50%), and there is evi-dence of ventricular hypertrophy. Insystolic dysfunction, the ejectionfraction is less than 50%, and left

© 2010 American College of Physicians ITC6-6 In the Clinic Annals of Internal Medicine 1 June 2010

heart failure, consider including acomplete leukocyte count, serumelectrolytes, blood urea nitrogen,

serum creatinine, chest radiography,pulmonary function studies, andappropriate cultures.

acute coronary syndromes andcoronary ischemia, severe hyperten-sion, atrial and ventricular arrhyth-mias, infections, pulmonary emboli,renal failure, and medical or dietarynonadherence. In addition, theguidelines emphasize early use ofloop diuretics to relieve volumeoverload and frequent assessmentsfor hypoperfusion and end-organdysfunction. Finally, there are rec-ommendations regarding importantprocesses, including medicationreconciliation, patient and familyeducation, initiation of guideline-mandated medications, and earlypostdischarge follow-up.

The updated guidelines also includestrengthened recommendations on 2medications, hydralazine and isosor-bide dinitrate, that are particularly ef-fective in African Americans whencombined with ACE inhibitors andβ-blockers. In addition, the guidelineshave been updated to reflect new in-formation on the use of implantablecardioverter-defibrillators (ICDs) andcardiac resynchronization (CRT) devices. The guidelines urge cliniciansto consider the functional capacityand overall prognosis of the patientbefore recommending an ICD. Theguidelines support the use of CRT toimprove symptoms, exercise capacity,quality of life, LVEF, and survival andto decrease hospitalizations in pa-tients with persistently symptomatic

How should clinicians evaluatefunctional capacity in patientswith suspected heart failure todetermine treatment?Clinicians should determine func-tional capacity by using the NewYork Heart Association (NYHA)classification system (Box). Trackingchanges in clinical NYHA class atevery visit may identify patients withprogressive heart failure who mayeventually benefit from specializedcare or cardiac transplantation.

Additional functional capacity teststhat can be followed over time in-clude the 6-minute walk test (Box)and formal exercise or pharmaco-logic stress testing. Measuring peakoxygen consumption ( V̇O2) at thetime of exercise testing is the mostpotent predictor of prognosis, butthe testing is not available at allcenters.

What are the key points of theupdated heart failure guidelines?The ACC/AHA guidelines for thediagnosis and management of heartfailure in adults were updated in2009. Several key changes weremade to reflect new informationfrom clinical trials from NorthAmerica and around the world.The one new section is on the hos-pitalized patient with acute heartfailure. This section emphasizes theimportance of identifying the causeof the decompensation, including

Diagnosis... Be alert for the development of heart failure in any patient with vas-cular disease; older persons; African Americans; men; patients with hypertension,hyperlipidemia, and diabetes; and patients who smoke, drink alcohol, or use illicitdrugs. Dyspnea and fatigue are the primary symptoms of heart failure. In additionto history and physical examination, use 2-dimensional Doppler echocardiographyto assess left ventricular function along with ECG and additional studies to deter-mine the cause of the heart failure and to identify exacerbating factors.

CLINICAL BOTTOM LINE

Treatment

New York Heart Association(NYHA) Classification System• NYHA class I (mild): Patient has

asymptomatic left ventriculardysfunction. Normal physicalactivity does not cause unduefatigue, palpitation, or shortnessof breath.

• NYHA class II (mild): Patient hasfatigue, palpitation, or shortnessof breath with normal physicalactivity.

• NYHA class III (moderate): Patienthas shortness of breath withminimal activity, including usualactivities of daily living.

• NYHA class IV (severe): Patienthas shortness of breath at restand is unable to perform anyphysical activity withoutdiscomfort. Physical activity ofany kind increases discomfort.

How to Perform the 6-Minute Walk TestAsk the patient to walk for 6

minutes in a straight line backand forth between 2 pointsseparated by 60 feet. Allow thepatient to stop and rest or evensit, if necessary. At either end ofthe course, place chairs that canquickly be moved if the patientneeds to sit. Note the totaldistance walked in 6 minutes,which correlates well with othermeasures of functional capacity.Sex-specific equations have beendeveloped that use age, height,and weight to calculate predicteddistances for healthy adults.

25. Morrison LK, Harri-son A, Krish-naswamy P, et al.Utility of a rapid B-natriuretic peptideassay in differentiat-ing congestive heartfailure from lung dis-ease in patients pre-senting with dysp-nea. J Am CollCardiol. 2002;39:202-9. [PMID: 11788208]

26. Maisel AS, Krish-naswamy P, NowakRM, et al. Rapidmeasurement of B-type natriuretic pep-tide in the emer-gency diagnosis ofheart failure. N EnglJ Med. 2002;347:161-7. [PMID: 12124404]

27. Wang TJ, Larson MG,Levy D, et al. Impactof obesity on plasmanatriuretic peptidelevels. Circulation.2004;109:594-600.[PMID: 14769680]

28. Dao Q, Krish-naswamy P, Kazane-gra R, et al. Utility ofB-type natriureticpeptide in the diag-nosis of congestiveheart failure in an ur-gent-care setting. JAm Coll Cardiol.2001;37:379-85.[PMID: 11216950]

29. de Lemos JA,McGuire DK, DraznerMH. B-type natriuret-ic peptide in cardio-vascular disease.Lancet. 2003;362:316-22.[PMID: 12892964]

30. Myers J, MadhavanR. Exercise testingwith gas exchangeanalysis. Cardiol Clin.2001;19:433-45.[PMID: 11570115]

31. The CONSENSUS Tri-al Study Group. Ef-fects of enalapril onmortality in severecongestive heart fail-ure. Results of theCooperative NorthScandinavianEnalapril SurvivalStudy (CONSENSUS).N Engl J Med.1987;316:1429-35.[PMID: 2883575]

32. The SOLVD Investi-gators. Effect ofenalapril on survivalin patients with re-duced left ventricu-lar ejection fractionsand congestiveheart failure. N EnglJ Med. 1991;325:293-302. [PMID: 2057034]

33. The SOLVD Investi-gators. Effect ofenalapril on mortali-ty and the develop-ment of heart failurein asymptomatic pa-tients with reducedleft ventricular ejec-tion fractions. N EnglJ Med. 1992;327:685-91. [PMID: 1463530]

© 2010 American College of PhysiciansITC6-7In the ClinicAnnals of Internal Medicine1 June 2010

heart failure who are undergoing op-timum medical therapy and have car-diac dyssynchrony (as evidenced by aprolonged QRS duration).

Finally, the guidelines clarify thetreatment goals in patients withboth heart failure and atrial fibrilla-tion. The primary goals are ventric-ular rate control using β-blockersand anticoagulation using warfarin.

When should clinicians beginfirst-line drug therapy with ACEinhibitors or angiotensin-receptorblockers? What are thealternatives for patients whocannot tolerate these drugs?ACE inhibitorsUse ACE inhibitors in all patientswith left ventricular dysfunction re-gardless of functional class (even inthe absence of symptoms) except inpatients with intolerance or a con-traindication, such as angioedema.These vasodilators alter the naturalhistory of the disease and improvesurvival and quality of life. Numerousrandomized, placebo-controlled clini-cal trials have shown that ACE in-hibitors reduce mortality in patientswith left ventricular dysfunction,even in those without symptoms.

CONSENSUS (Cooperative North Scandi-navian Enalapril Survival Study) evaluated253 patients with NYHA class I to IV heartfailure who were randomly assigned toenalapril or placebo in a blind study. Allpatients were also receiving diuretics, and93% received digitalis glycosides. The mor-tality rate was reduced by 27% (P < 0.001)in the patients receiving enalapril com-pared with placebo (31).

The SOLVD (Studies of Left Ventricular Dys-function) treatment trial randomly as-signed 2569 patients with NYHA class I toIV heart failure to enalapril or placebo. Inpatients with heart failure, those receivingenalapril had a 16% (P < 0.005) reductionin mortality rate, a 30% (P < 0.001) reduc-tion in heart failure hospitalizations, a 7%(P < 0.01) reduction in total hospitaliza-tions, a 44% (P < 0.01) reduction in wors-ening heart failure, and a 23% (P < 0.02) re-duction in MI (32) compared with thosereceiving placebo.

The SOLVD prevention trial enrolled 4228patients with NYHA class I heart failure andasymptomatic left ventricular dysfunctionand randomly assigned them to enalapril orplacebo. Patients receiving enalapril had an 8% reduction in mortality rate, a 31% (P < 0.001) reduction in heart failure hospi-talizations, a 50% (P < 0.01) reduction inepisodes of worsening heart failure, and a24% (P < 0.01) reduction in MI comparedwith those receiving placebo (33).

Initiate enalapril, captopril, lisino-pril, or ramipril at low doses andtitrate upward while monitoringblood pressure. The end point forblood pressure can be as low as 80 to 90 mm Hg systolic bloodpressure as long as the patient isasymptomatic. Important side ef-fects include cough, worsening re-nal insufficiency, and hyperkalemia.

Angiotensin-receptor blockersConsider using angiotensin-receptorblockers (ARBs) in patients whohave intolerable side effects fromACE inhibitors, such as cough.

The ELITE (Evaluation of Losartan in theElderly) I trial compared captopril withlosartan in elderly patients with heart fail-ure and showed a decrease in all-causemortality (8.7% vs 4.8%; risk reduction,46%; P = 0.035) in the losartan group. Ad-missions with heart failure were the samein both groups (5.7%), as was improve-ment in NYHA functional class from base-line (34). The ELITE II trial also comparedcaptopril with losartan, but all-cause mor-tality (11.7% vs. 10.4% average annualmortality rate), sudden death, and resusci-tated arrests (9.0% vs. 7.3%) did not signifi-cantly differ between the groups (hazardratios, 1.13 [95% CI, 0.95 to 1.35]; P = 0.16,and 1.25 [CI, 0.98 to 1.60]; P = 0.08) (35).

Val-HeFT (Valsartan-Heart Failure Trial)randomly assigned patients with heartfailure to valsartan or placebo in additionto standard heart failure medications.Mortality did not differ between groups,but the incidence of the combined endpoint of morbidity or mortality was 13.2%lower with valsartan than with placebo(relative risk, 0.87 [CI, 0.77 to 0.97]; P = 0.009) (36). In a subgroup analysis, pa-tients who were not receiving an ACE in-hibitor but who were randomly assignedto receive valsartan had a 33% reduction

34. Pitt B, Segal R, Mar-tinez FA, et al. Ran-domised trial oflosartan versus cap-topril in patientsover 65 with heartfailure (Evaluation ofLosartan in the Eld-erly Study, ELITE).Lancet.1997;349:747-52.[PMID: 9074572]

35. Pitt B, Poole-WilsonPA, Segal R, et al. Ef-fect of losartan com-pared with captoprilon mortality in pa-tients with sympto-matic heart failure:randomised trial-theLosartan Heart Fail-ure Survival StudyELITE II. Lancet.2000;355:1582-7.[PMID: 10821361]

36. Cohn JN, Tognoni G.A randomized trialof the angiotensin-receptor blocker val-sartan in chronicheart failure. N EnglJ Med.2001;345:1667-75.[PMID: 11759645]

37. Maggioni AP, AnandI, Gottlieb SO, et al.;Val-HeFT Investiga-tors (Valsartan HeartFailure Trial). Effectsof valsartan on mor-bidity and mortalityin patients withheart failure not re-ceiving angiotensin-converting enzymeinhibitors. J Am CollCardiol.2002;40:1414-21.[PMID: 12392830]

38. Granger CB, McMur-ray JJ, Yusuf S, et al.Effects of candesar-tan in patients withchronic heart failureand reduced left-ventricular systolicfunction intolerantto angiotensin-con-verting-enzyme in-hibitors: the CHARM-Alternative trial.Lancet.2003;362:772-6.[PMID: 13678870]

39. Loeb HS, Johnson G,Henrick A, et al. Ef-fect of enalapril, hy-dralazine plus isosor-bide dinitrate, andprazosin on hospital-ization in patientswith chronic con-gestive heart failure.The V-HeFT VA Co-operative StudiesGroup. Circulation.1993;87:VI78-87.[PMID: 8500244]

© 2010 American College of Physicians ITC6-8 In the Clinic Annals of Internal Medicine 1 June 2010

adding this therapy increased survival inthose who were already taking other neu-rohormonal blockers, including ACE in-hibitors and β-blockers (41).

When should clinicians add β-blockers, aldosteroneantagonists, and loop diuretics?β-Blockersβ-Blockers should be used in allNYHA classes of heart failure if thepatient is stable on ACE inhibitors orother vasodilators and is not volumeoverloaded. β-Blockers can reduceheart failure symptoms, improve clin-ical outcomes and ejection fraction,and significantly decrease mortalityrate. Patients with less severe heartfailure have the greatest long-termbenefit, including those with left ven-tricular dysfunction but no symp-toms. Many studies testing carvedilol,bisoprolol, and long-acting metopro-lol succinate have found reductions inhospitalizations, sudden death, andoverall mortality in patients withheart failure. Data on other β-block-ers are lacking; therefore, cliniciansshould select one of the agents forwhich mortality data are available.

The CAPRICORN trial randomly assignedpatients with left ventricular dysfunctionafter MI with or without heart failure to β-blockade with carvedilol. There was asignificant reduction in mortality that waseven more marked in the group withoutsymptomatic heart failure (17).

The U.S. carvedilol trial randomly assigned696 patients to the carvedilol group and 398to the placebo group. Patients were classi-fied with NYHA class I to IV heart failure. A65% (P < 0.001) reduction in mortality wasseen in the carvedilol group. Cardiovascularhospitalizations were reduced (42).

The CIBIS (Cardiac Insufficiency BisoprololStudy) I randomly assigned 320 patients tobisoprolol, 5 mg/d, or placebo. There was astatistically insignificant 20% reduction inmortality and a significant reduction inheart failure hospitalizations (43). TheCIBIS II randomly assigned patients withNYHA class III to IV heart failure to bisopro-lol, 5 mg/d, or placebo. A total of 3.6% ofpatients in the bisoprolol group had sud-den cardiac death versus 6.3% in the place-bo group (P < 0.01) (44).

in all-cause mortality. This result is similarto the magnitude of mortality reductionwith ACE inhibitors (37).

Evidence from the randomized, placebo-controlled CHARM (Candesartan Cilexitil[Atacand] in Heart Failure Assessment ofReduction in Mortality and Morbidity)-Alternative trial showed that the ARB can-desartan decreased a combined end pointof death from cardiovascular causes orhospitalization due to heart failure whencompared with placebo in patients withleft ventricular dysfunction who could nottolerate ACE inhibitors (38).

Some studies have suggested thatcombining ACE inhibitors andARBs may be beneficial in reduc-ing left ventricular size and de-creasing hospitalizations with anequivocal effect on mortality (34-38). Patients should be carefullymonitored for hyperkalemia and re-nal dysfunction. For patients withelevated blood pressure despitemaximum ACE inhibitor and β-blocker dosing, consider the ad-dition of an ARB in non-AfricanAmericans or the combination ofhydralazine and long-acting nitrates in African Americans.

Hydralazine and nitratesPatients who cannot tolerate eitherACE inhibitors or ARBs shouldreceive hydralazine and long-actingnitrates. This combination im-proves clinical outcomes and de-creases mortality in patients withheart failure and depressed ejectionfraction (39, 40). However, thecombination does not seem to haveas much effect on mortality rates asACE inhibitors. Hydralazine plusnitrates should also be consideredin addition to standard therapy, in-cluding an ACE inhibitor or ARB,in African-American patients withsymptomatic heart failure, becausethis combination may favorably affect myocardial remodeling andmortality in these patients.

A-HeFT (African American Heart Failure Tri-al), which compared isosorbide plus hy-dralazine with placebo in African Ameri-cans with heart failure, showed that

40. Johnson G, Carson P,Francis GS, et al. In-fluence of preran-domization (base-line) variables onmortality and on thereduction of mortali-ty by enalapril. Veter-ans Affairs Coopera-tive Study onVasodilator Therapyof Heart Failure (V-HeFT II). V-HeFT VACooperative StudiesGroup. Circulation.1993;87:VI32-9.[PMID: 8500237]

41. African-AmericanHeart Failure Trial In-vestigators. Combi-nation of isosorbidedinitrate and hy-dralazine in blackswith heart failure. NEngl J Med.2004;351:2049-57.[PMID: 15533851]

42. Packer M, BristowMR, Cohn JN, et al.The effect ofcarvedilol on mor-bidity and mortalityin patients withchronic heart failure.U.S. Carvedilol HeartFailure Study Group.N Engl J Med.1996;334:1349-55.[PMID: 8614419]

43. CIBIS Investigatorsand Committees. Arandomized trial ofbeta-blockade inheart failure. TheCardiac InsufficiencyBisoprolol Study(CIBIS). Circulation.1994;90:1765-73.[PMID: 7923660]

44. The Cardiac Insuffi-ciency BisoprololStudy II (CIBIS-II): arandomised trial.Lancet. 1999;353:9-13. [PMID: 10023943]

45. Effect of metoprololCR/XL in chronicheart failure: Meto-prolol CR/XL Ran-domised Interven-tion Trial inCongestive HeartFailure (MERIT-HF).Lancet.1999;353:2001-7.[PMID: 10376614]

46. Packer M, Coats AJ,Fowler MB, et al. Ef-fect of carvedilol onsurvival in severechronic heart failure.N Engl J Med.2001;344:1651-8.[PMID: 11386263]

47. Pitt B, Zannad F,Remme WJ, et al.The effect ofspironolactone onmorbidity and mor-tality in patientswith severe heartfailure. RandomizedAldactone Evalua-tion Study Investiga-tors. N Engl J Med.1999;341:709-17.[PMID: 10471456]

© 2010 American College of PhysiciansITC6-9In the ClinicAnnals of Internal Medicine1 June 2010

The MERIT-HF (Metoprolol CR/XL Random-ized Intervention Trial-Heart Failure) ran-domly assigned 3991 patients with NYHAclass II to IV heart failure to metoprololCR/XL, up to 200 mg/d, versus placebo. All-cause mortality was reduced 34% (P < 0.001), and sudden death was reduced59% (P < 0.001) for patients receivingmetoprolol versus placebo (45).

The COPERNICUS (Carvedilol ProspectiveRandomized Cumulative Survival) trialrandomly assigned patients with NYHAclass IV heart failure to carvedilol or place-bo. The combined risk for death or hospi-talization decreased 24% with carvedilol (P < 0.001) (46).

Initiate β-blockers at the lowestdose and slowly titrate upwardevery 2 to 4 weeks to the highesttherapeutic dose tolerated, as limit-ed by bradycardia, hypotension, orside effects. Instruct patients tocheck their body weight and watchfor worsening heart failure symp-toms during initiation and upwardtitration of β-blockade.

Aldosterone antagonistsIf patients continue to have NYHAclass III to IV symptoms despitetherapy with ACE inhibitors and β-blockers, consider treatment withlow doses of an aldosterone antago-nist. Spironolactone has been studiedthe most but can occasionally causepainful gynecomastia in men.

RALES (Randomized Aldosterone Evalua-tion Study), a large, randomized, placebo-controlled trial involving 1663 patientswith NYHA class III to IV heart failure on ap-propriate therapy with or without spirono-lactone, was halted 18 months early by theData Safety Monitoring Board becausethere were significantly fewer deaths in thespironolactone group than in the placebogroup (284 vs. 386 deaths; 35% reduction;P < 0.001) (47).

Eplerenone is a newer, more selec-tive aldosterone antagonist withfewer undesirable side effects andhas been shown to decrease all-cause mortality in patients with anejection fraction less than 40% afteracute MI (48).

Higher rates of hyperkalemia havebeen found in patients taking ACEinhibitors and spironolactone, ne-cessitating careful monitoring ofserum potassium levels (49). Week-ly serum electrolytes should be obtained until there is evidence ofstability in potassium levels. Thecombination of ACE inhibitors,ARBs, and spironolactone shouldbe avoided because of a significant-ly increased risk for hyperkalemia.

DiureticsDiuretics, which are the only thera-py that acutely produces sympto-matic benefits, can reduce pul-monary capillary wedge pressureand edema and improve exercisecapacity. No clinical trials have as-sessed their long-term safety or ef-fect on mortality in heart failure.

A single trial comparing furosemide withtorsemide found that torsemide had betteroral absorption and that patients receivingtorsemide were less likely to be readmittedfor heart failure (50).

In general, torsemide orbumetanide should be reserved for patients who do not respond toadequate doses of furosemide.

Loop diuretics should be used incombination with a low-sodiumdiet to control volume overload,maintain a stable weight, and improve the functional capacity ofpatients with NYHA class II to IVheart failure. Diuretics should neverbe used alone to treat heart failure,because they do not prevent theprogression of disease or maintainclinical stability over time.

For patients resistant to loop diuret-ics, thiazide diuretics may be addedto augment diuresis. Furthermore,the use of a thiazide diuretic in com-bination with a loop diuretic can bepart of an effective “sliding” diureticregimen based on a patient’s dailyweight and symptoms. A secondclass of diuretic may act synergisti-cally with the first by blocking theadaptive processes that limit diuretic

48. Eplerenone Post-Acute Myocardial In-farction Heart FailureEfficacy and SurvivalStudy Investigators.Eplerenone, a selec-tive aldosteroneblocker, in patientswith left ventriculardysfunction aftermyocardial infarc-tion. N Engl J Med.2003;348:1309-21.[PMID: 12668699]

49. Juurlink DN, Mam-dani MM, Lee DS, etal. Rates of hyper-kalemia after publi-cation of the Ran-domized AldactoneEvaluation Study. NEngl J Med.2004;351(6):543-51.[PMID: 15295047]

50. Murray MD, DeerMM, Ferguson JA, etal. Open-label ran-domized trial oftorsemide com-pared withfurosemide therapyfor patients withheart failure. Am JMed. 2001;111:513-20. [PMID: 11705426]

51. The Digitalis Investi-gation Group. Theeffect of digoxin onmortality and mor-bidity in patientswith heart failure. NEngl J Med.1997;336:525-33.[PMID: 9036306]

52. Khand AU, RankinAC, Kaye GC, ClelandJG. Systematic re-view of the manage-ment of atrial fibrilla-tion in patients withheart failure. EurHeart J. 2000;21:614-32. [PMID: 10731399]

53. Adams KF Jr, Gheo-rghiade M, UretskyBF, et al. Clinical ben-efits of low serumdigoxin concentra-tions in heart failure.J Am Coll Cardiol.2002;39:946-53.[PMID: 11897434]

54. Rathore SS, Wang Y,Krumholz HM. Sex-based differences inthe effect of digoxinfor the treatment ofheart failure. N EnglJ Med.2002;347:1403-11.[PMID: 12409542]

55. How to diagnose di-astolic heart failure.European StudyGroup on DiastolicHeart Failure. EurHeart J. 1998;19:990-1003.[PMID: 9717033]

© 2010 American College of Physicians ITC6-10 In the Clinic Annals of Internal Medicine 1 June 2010

effectiveness. With all diuretics, cli-nicians should frequently monitorpatient renal function and elec-trolytes, especially potassium levels.

What is the role of digoxin in thetreatment of heart failure?Digoxin can alleviate symptoms anddecrease hospitalizations in patientswith heart failure; however, it shouldbe reserved specifically for patientswith symptomatic NYHA class II toIV heart failure, because research in-dicates that it provides no survivaldifference compared with placebo(51). Furthermore, digoxin alone doesnot seem to be effective in rate con-trol for patients with atrial fibrillation,because it provides only rate controlat rest (52). The ACC/AHA guide-lines recommend the use of a β-blocker with digoxin for rate controlof atrial fibrillation.

Ensure that electrolytes and renalfunction are stable before startingdigoxin, and monitor serum levels,especially if renal function is chang-ing. Some controversy exists over theappropriate serum level of digoxin. Arecent study suggested that lowerserum levels of digoxin were as effi-cacious as “therapeutic” levels, with alower risk for side effects (53). Infact, in a post hoc subgroup analysisof 1 recent study, mortality rate wasincreased in women receiving digox-in compared with men, which mayhave been due to higher serumdigoxin levels (54).

What drug therapy is appropriatefor patients with heart failure andpreserved left ventricular function?The goals of heart failure treatmentin the setting of preserved left ven-tricular function are to control heartrate to allow for adequate filling ofthe ventricle; to maintain normal si-nus rhythm, if possible; to controlvolume status to decrease diastolicpressures; to control blood pressureor other stimuli predisposing to leftventricular hypertrophy; and to mini-mize myocardial ischemia in the set-ting of left ventricular hypertrophy,

even in the absence of epicardialcoronary disease.

There have been few randomizedtrials of the treatment of heart failurewith preserved left ventricular func-tion, and recommendations are basedon investigations in small groups ofpatients or on theoretical concepts.The publication of consensus guide-lines on the definition of heart fail-ure with preserved left ventricularfunction has allowed for the designof multicenter clinical trials (55).

ACC/AHA and other guidelinessuggest that patients with heartfailure with preserved left ventricu-lar function should be treated withdiuretics, β-blockers, ACE in-hibitors, ARBs, and nitrates. Calci-um-channel blockers, such as vera-pamil and diltiazem, may alsoalleviate symptoms and improve ex-ercise capacity. Avoid overdiuresis,because dehydration can lead tolightheadedness and syncope in pa-tients with diastolic dysfunction.

When should clinicians useinotropic agents in patients withheart failure?Inotropic agents, such as dobuta-mine and milrinone, can improvecardiac output in patients with lowcardiac output and decrease afterloadin patients with severe heart failureunresponsive to the traditional heartfailure medications. However, all in-otropic agents, except digoxin, havebeen associated with excess mortalityand should be reserved for patientsunresponsive to traditional oral med-ications for heart failure. Inotropicagents can be used short-term to sta-bilize cardiogenic shock and to allowfor other therapies, including revas-cularization, valve repair, or initiationof more traditional therapies. In ad-dition, inotropic agents can be usedto bridge a patient to cardiac trans-plantation or insertion of a left ven-tricular assist device. Finally, inotrop-ic agents can be used as palliationwhen heart failure becomes refractoryand the patient is not a candidate for

56. Dunkman WB, John-son GR, Carson PE, etal. Incidence ofthromboembolicevents in congestiveheart failure. The V-HeFT VA CooperativeStudies Group. Cir-culation.1993;87:VI94-101.[PMID: 8500246]

57. Al-Khadra AS, SalemDN, Rand WM, et al.Warfarin anticoagu-lation and survival: acohort analysis fromthe Studies of LeftVentricular Dysfunc-tion. J Am Coll Cardi-ol. 1998;31:749-53.[PMID: 9525542]

58. Sullivan MJ, CobbFR. The anaerobicthreshold in chronicheart failure. Rela-tion to blood lactate,ventilatory basis, re-producibility, and re-sponse to exercisetraining. Circulation.1990;81:II47-58.[PMID: 2295152]

59. Myers J, Gianrossi R,Schwitter J, et al. Ef-fect of exercise train-ing on postexerciseoxygen uptake ki-netics in patientswith reduced ven-tricular function.Chest.2001;120:1206-11.[PMID: 11591562]

60. Sullivan MJ, CobbFR. Central hemody-namic response toexercise in patientswith chronic heartfailure. Chest.1992;101:340S-346S.[PMID: 1576862]

61. O’Connor CM, Whel-lan DJ, Lee KL, et al.Efficacy and safety ofexercise training inpatients with chron-ic heart failure: HF-ACTION randomizedcontrolled trial.JAMA 2009;301:1439-1450.[PMID: 19351941]

62. Moss AJ, Zareba W,Hall WJ, et al. Pro-phylactic implanta-tion of a defibrillatorin patients with my-ocardial infarctionand reduced ejec-tion fraction. N EnglJ Med. 2002;346:877-83. [PMID: 11907286]

63. Bardy GH, Lee KL,Mark DB, et al. Amio-darone or an im-plantable cardiovert-er-defibrillator forcongestive heart fail-ure. N Engl J Med.2005;352:225-37.[PMID: 15659722]

© 2010 American College of PhysiciansITC6-11In the ClinicAnnals of Internal Medicine1 June 2010

either cardiac transplantation or aventricular assist device.

When should clinicians considerusing anticoagulants in patientswith heart failure?Embolic stroke is associated with di-lated cardiomyopathy with depressedejection fraction below 35%, valvularlesions (especially mitral stenosis),and atrial fibrillation. The incidenceof thromboembolic events was about2.7 per 100 patient-years in 1 largetrial of patients with heart failure(56). Many experts advocate anti-coagulation to reduce the risk forstroke in patients with heart failureand depressed ejection fraction be-low 35% who have no contraindica-tions, but anticoagulation remainscontroversial for such patients with-out atrial fibrillation, documentedclot, or valvular heart disease. In 1trial, the use of warfarin in such pa-tients was not associated with a re-duction in all-cause mortality (57).Therefore, it seems most appropriateto initiate anticoagulation with war-farin in patients with left ventricularclot documented on echocardiogramor ventriculogram, atrial fibrillation,or previous embolic event and to useaspirin or clopidogrel in patientswith coronary disease regardless ofejection fraction.

What should clinicians advisepatients about exercise? Doformal exercise programs providebenefit?Exercise improves physical and psychological well-being. In patientswith heart failure, it improves peak V̇O2 (58, 59) as well as metabol-ic and hemodynamic indices and de-lays the onset of the anaerobicthreshold (58, 60). A recent NationalInstitutes of Health-sponsored clini-cal trial found that 30 minutes of ex-ercise on a treadmill or stationary bicycle most days of the week led toa significant reduction in death orheart failure complications (61).

Enroll patients with medically stable NYHA class II, III, and

perhaps class IV heart failure in along-term aerobic exercise programtailored to the patient’s functionalcapacity. A structured cardiac rehabilitation program may be par-ticularly effective, because it canprovide supervised exercise as wellas support in making lifestyle mod-ifications. Patients with worseningheart failure should temporarilystop exercise until symptoms arestabilized. In addition, if patientsshow evidence of exercise-inducedischemia, stop exercise until furtherevaluation and therapy are initiated.

When should clinicians considerplacement of an intracardiacdevice in patients with heartfailure?Consider placement of an ICD tomonitor heart rate and rhythm andto correct arrhythmia when it occursfor patients with left ventricular dys-function and an ejection fraction lessthan 30% in NYHA class I, II, or IIIand an overall life expectancy ofmore than 6 months. Data suggestthat patients with class IV symptomsdo not benefit from ICDs and thatpatients with class II symptoms maybenefit most (1, 62, 63). Studiesshow a clear decrease in suddendeath and overall mortality. Considerthe patient’s functional capacity aswell as overall prognosis before rec-ommending an ICD.

The DEFINITE (Defibrillators in Non-ischemicCardiomyopathy Treatment Evaluation) tri-al randomly assigned 458 patients with di-lated nonischemic cardiomyopathy andLVEF less than 36% to standard medicaltherapy or standard medical therapy plus asingle-chamber ICD. Over 29 months of fol-low-up, 28 deaths occurred in the ICD groupcompared with 40 in the standard medicaltherapy group. Although overall mortalitywas not significantly lower, there were 3 sud-den deaths in the ICD group versus 14 in thestandard therapy group (P = 0.006) (64).

In MADIT (Multicenter Automatic Defibril-lator Implantation Trial) II, 1232 patientswith a previous MI and an ejection fractionless than 30% were randomly assigned (inthe absence of electrophysiologic testingor other risk stratification) to ICD

64. Kadish A, Dyer A,Daubert JP, et al. Pro-phylactic defibrilla-tor implantation inpatients with nonis-chemic dilated car-diomyopathy. N EnglJ Med.2004;350:2151-8.[PMID: 15152060]

65. Young JB, AbrahamWT, Smith AL, et al.Combined cardiacresynchronizationand implantable car-dioversion defibrilla-tion in advancedchronic heart failure:the MIRACLE ICD Tri-al. Multicenter In-Sync ICD Random-ized ClinicalEvaluation (MIRACLEICD) Trial Investiga-tors. JAMA.2003;289:2685-94.[PMID: 12771115]

66. Cleland JG, DaubertJC, Erdmann E, et al.The effect of cardiacresynchronizationon morbidity andmortality in heartfailure. N Engl J Med.2005;352:1539-49.[PMID: 15753115]

67. Moss AJ, Hall WJ,Cannom DS, et al.Cardiac-resynchro-nization therapy forthe prevention ofheart-failure events.N Engl J Med2009;361:1329-1338.[PMID: 19723701]

© 2010 American College of Physicians ITC6-12 In the Clinic Annals of Internal Medicine 1 June 2010

placement with conventional drug thera-py or conventional drug therapy alone. TheICD group had a 28% reduction in mortal-ity at 3 years (P = 0.007) (62).

SCD-HeFT (Sudden Cardiac Death inHeart Failure Trial) randomly assigned2521 patients with NYHA class II or IIIheart failure and an LVEF less than 35%to conventional therapy for heart failureplus placebo; conventional therapy plusamiodarone; or conventional therapyplus a conservatively programmed,shock-only, single-lead ICD. During a me-dian follow-up of 45.5 months, mortalitywas 29% in the placebo group, 28% in theamiodarone group, and 22% in the ICDgroup. The ICD therapy was associatedwith a 23% decreased risk for death (P = 0.007) compared with placebo (63).

Placement of a biventricular pace-maker can improve quality of lifeand decrease hospitalizations in pa-tients with heart failure, an ejectionfraction less than 35%, a QRS in-terval greater than 120 msec onECG, and symptoms despite maxi-mum medical therapy. In addition,those with NYHA class I and IIheart failure may benefit from de-creased hospitalizations and im-proved ejection fractions.

In the MIRACLE-ICD (Multicenter InSyncICD Randomized Clinical Evaluation) trial,369 patients with class III or IV heart failure,ejection fraction, and a QRS interval of 130msec or more received an ICD with resyn-chronization device. Those in whom thedevice was turned on had improved quali-ty of life, functional status, and exercise ca-pacity but no change in heart failure sta-tus, rates of hospitalization, or survival (65).

In the CARE-HF (Cardiac Resynchronizationin Heart Failure) study, 813 patients withNYHA class III or IV heart failure due to leftventricular systolic dysfunction and cardiacdyssynchrony who were receiving stan-dardized drug therapy were randomly as-signed to receive medical therapy alone orwith cardiac resynchronization. The studyconcluded that, in these patients, cardiacresynchronization improved symptomsand quality of life and reduced the risk fordeath (66).

In MADIT-CRT, 1820 patients with ischemicor nonischemic cardiomyopathy, an ejec-tion fraction of 30% or less, a QRS duration

of 130 msec or more, and NYHA class I or IIsymptoms were randomly assigned toCRT-ICD or ICD alone. There was a 41% re-duction in the risk for heart failure in theCRT-ICD group, a finding that was evidentprimarily in a prespecified subgroup of pa-tients with a QRS duration of 150 msec ormore. CRT was associated with a signifi-cant reduction in left ventricular volumesand improvement in the ejection fraction.The overall risk for death did not signifi-cantly differ between the 2 groups, with a3% annual mortality rate in each treat-ment group (67).

When should clinicians hospitalizepatients with heart failure?Hospitalize patients with severeNYHA class IV heart failure, char-acterized by dyspnea at rest, severefatigue, or volume overload unre-sponsive to oral diuretics. Also hos-pitalize patients with life-threateningventricular arrhythmias, atrial ar-rhythmias that worsen heart failuresymptoms or cause hypotension,syncope, or sudden cardiac death.Finally, hospitalize patients withheart failure and unstable angina ornew ECG changes to rule out MI.

When should clinicians consult acardiologist about management ofpatients with heart failure?If symptoms worsen despite opti-mum medical therapy, consult acardiologist about the need for hos-pitalization for parenteral inotropicdrug treatment; catheterization;placement of an ICD, biventricularpacemaker, or left ventricular assistdevice; or cardiac transplantation.Any signs or symptoms of new orworsening coronary ischemiashould also prompt cardiology con-sultation. In addition, patients withany of the ominous signs of ad-vanced heart failure, including in-tolerance to any afterload-reducingagent or β-blocker, hyponatremia,worsening renal function, or fre-quent hospitalizations, may benefitfrom evaluation by a cardiologist.Consider obtaining pulmonaryconsultation when primary lungdisease, such as chronic obstructivepulmonary disease or sleep apnea,

68. American College ofCardiology/Ameri-can Heart Associa-tion Task Force onPractice Guidelines(Committee to Re-vise the 1995 Guide-lines for the Evalua-tion andManagement ofHeart Failure).ACC/AHA Guidelinesfor the Evaluationand Management ofChronic Heart Failurein the Adult: Execu-tive Summary. Circu-lation.2001;104:2996-3007.[PMID: 11739319]

69. American College ofCardiology.ACC/AHA 2005Guideline Update forthe Diagnosis andManagement ofChronic Heart Failurein the Adult: a reportof the American Col-lege ofCardiology/Ameri-can Heart Associa-tion Task Force onPractice Guidelines(Writing Committeeto Update the 2001Guidelines for theEvaluation and Man-agement of HeartFailure): developedin collaboration withthe American Col-lege of Chest Physi-cians and the Inter-national Society forHeart and LungTransplantation: en-dorsed by the HeartRhythm Society. Cir-culation.2005;112:e154-235.[PMID: 16160202]

70. Heart Failure Societyof America. HFSA2006 Comprehen-sive Heart FailurePractice Guideline. JCard Fail. 2006;12:e1-2. [PMID: 16500560]

© 2010 American College of PhysiciansITC6-13In the ClinicAnnals of Internal Medicine1 June 2010

is believed to be contributing to thepatient’s symptoms.

What is the role of diet andmonitoring weight in themanagement of heart failure?Despite a lack of definitive evi-dence, ACC/AHA and otherguidelines recommend sodium re-striction in patients with sympto-matic heart failure and the avoid-ance of salt-retaining medications,such as nonsteroidal anti-inflam-matory drugs. Some cliniciansrecommend that patients withmore advanced heart failure limitintake to 2 g of sodium per dayand 2 qt of fluid per day to in-crease the effectiveness of diuretictherapy. Limitation of salt andfluid intake results in fewer hospi-talizations for decompensatedheart failure. Patients who havecardiovascular risk factors, such ashyper-lipidemia, obesity, or dia-betes, should also be encouragedto follow dietary recommenda-tions specific to these underlyingconditions.

Monitoring daily weight can pre-vent heart failure exacerbations byallowing for adjustments in diuret-ics. Instruct patients to weighthemselves daily and contact theirclinician for instructions on how toadjust their diuretics if their weightexceeds a predetermined threshold

(usually 2-lb increase overnight or5-lb increase over 3 days).

What other approaches shouldclinicians recommend to patientsto prevent exacerbations of heartfailure?Advise patients to adhere to theirfluid and salt restriction and med-ical regimen, to weigh themselvesdaily, and to report deviations fromtheir “dry weight” before they be-come symptomatic. Some patientscan learn to use a sliding dose ofdiuretic to maintain their weight.Support from nurses, dietitians,home health staff, and physicaltherapists can be invaluable inhelping patients prevent exacerba-tions. Patients should receive pneu-mococcal vaccine and annual in-fluenza immunization.

Patients with established CADshould begin aggressive risk-factormodification, including attention todiet, exercise, weight control, andsmoking cessation. Prescribe behav-ior modifications and pharmaco-logic therapy unless contraindicat-ed. Many studies have shown thatrisk-factor modification with cho-lesterol-lowering drugs and the useof aspirin or other antiplateletdrugs, ACE inhibitors, and β-blockers can significantly reducethe risk for future cardiovascularevents and can reduce mortality.

Treatment... Determine NYHA functional class to guide treatment in patients withheart failure. Limit salt and fluid intake in patients with symptomatic heart failure,and recommend regular exercise as tolerated. Regardless of symptoms, begin first-line drug therapy with ACE inhibitors or ARBs (or, if these are not tolerated, hy-dralazine and nitrates) as well as β-blockers in patients who are not volume over-loaded. Add loop diuretics and digoxin in patients with NYHA classes II, III, and IVheart failure and aldosterone antagonists in those with class III and IV heart failure.Monitor potassium and renal function. Consult a cardiologist in patients with severeheart failure who may require hospitalization for inotropic agents; placement of ICDdevices, pacemakers, or left ventricular assist devices; or cardiac transplantation.Recognize that anticoagulation for patients with depressed ejection fractions re-mains controversial. Teach patients to participate in their own care by encouragingthem to exercise and to monitor their diet, medical regimen, and weight.

CLINICAL BOTTOM LINE

Practice Improvement

71. Pharmacy BenefitsManagement Strate-gic HealthcareGroup and the Med-ical Advisory Panel;Department of Vet-erans Affairs, Veter-ans Health Adminis-tration. ThePharmacologic Man-agement of ChronicHeart Failure. Ac-cessed atwww.oqp.med.va.gov/cpg/CHF/CHF_Base.htm on 11 October2007.

inthe

c linicTool Kit

in the clinic

Heart Failure

PIER Moduleswww.pier.acponline.orgAccess the PIER modules on heart failure and percutaneous coronary intervention fromthe American College of Physicians. PIER modules provide an evidence-based, electronicresource for clinical recommendations and links to patient information material at thepoint of care.

Patient Informationwww.annals.org/intheclinic/toolkit-hf.htmlDownload copies of the Patient Information sheet that appears on the following page forduplication and distribution to your patients.

Quality Improvement Toolswww.ihi.org/ihi/search/searchresults.aspx?searchterm=heart+failure+tools&searchtype=basicLinks to a variety of helpful tools for managing various aspects of heart failure, compiledby the Institute for Healthcare Improvement.www.cardiologyinoregon.org/information/information.html#toolkitResources from the Oregon Heart Failure GAP Toolkit, part of an American College ofCardiology project in 3 states to improve heart failure care.

1 June 2010Annals of Internal MedicineIn the ClinicITC6-14© 2010 American College of Physicians

What do professionalorganizations recommend withregard to the care of patientswith heart failure?In 2001, the ACC/AHA publishedguidelines for the Evaluation andManagement of Chronic HeartFailure in the Adult (68), and thesewere updated in 2005 (69) andagain in 2009 (1).

In addition to the ACC/AHAguidelines, other important guide-lines include the Heart Failure Soci-ety of America 2006 ComprehensiveHeart Failure Practice Guideline(70) and the Department of Veter-ans Affairs/Veterans Health Admin-istration 2003 guidelines relating tothe pharmacologic management ofchronic heart failure (71).

What measures do stakeholdersuse to evaluate the quality of carefor patients with heart failure?The Centers for Medicare & Med-icaid Services (CMS) have started aPhysician Quality Reporting Initia-tive (PQRI) program, throughwhich clinicians can report a

designated set of quality measureson claims for services and earnbonus payments. Of the currentmeasures in the PQRI program, 2relate to heart failure. The first issimilar to the Ambulatory CareQuality Alliance measure on theuse of ACE inhibitors or ARBs,calling for use of these agents inpatients older than 18 years with adiagnosis of heart failure and leftventricular dysfunction. The secondmeasures use of β-blocker therapyin the same population.

In addition, the Agency forHealthcare Research and Quality isusing quality indicators to measurethe hospital admission rate forheart failure, and CMS has begunthe public reporting of hospital-level 30-day mortality for patientswith heart attack and heart failure.

Hospital readmission rates within30 days of discharge as well as doc-umentation of discharge teachingand instructions are also perform-ance measures that are evaluated byseveral regulatory agencies.

In the ClinicAnnals of Internal Medicine

Pati

ent

Info

rmat

ion

THINGS YOU SHOULDKNOW ABOUT HEARTFAILURE

• Heart failure, which is sometimes called congestiveheart failure, is when the heart can’t pump as wellas it should. Because the heart has a hard timegetting blood to the rest of the body, patients withheart failure can feel weak and tired.

• In some patients with heart failure, fluid (edema)builds up in the lungs and parts of the body, makingit hard to breathe and causing swelling in the legs.

• Heart failure can be caused by many differentconditions that directly or indirectly affect the heart.People with high blood pressure, diabetes, highcholesterol, and coronary artery disease can developheart failure. Treating these conditions may preventheart failure.

• Treating heart failure means working together withyour doctor to control salt in your diet, watchingyour weight, and taking all your medications everyday. It’s important to keep your regular doctorappointments.

• Heart failure affects nearly 5 million adults, and 550000 new cases are diagnosed each year. It is morecommon in older people but can occur at any age.Although there is no cure yet, heart failure is verytreatable, and millions of Americans lead a full lifeby managing their condition through medicationsand by making healthy changes in their lifestyles.

Heart Failure Symptoms• Breathlessness during activity, at rest, or while

sleeping

• Wheezing or coughing that may be dry or mayproduce white or pink blood-tinged phlegm

• Swelling in the feet, ankles, legs, or abdomen orunexplained weight gain

• A constant lack of energy and difficulty performingeveryday activities

• A sense of having a full or sick stomach

• A feeling like the heart is racing or pounding

• A feeling the heart is skipping beats or occasionallypounding very hard

For More Information

Web Sites With Good Information on Heart Failurewww.doctorsforadults.com/images/healthpdfs/heartfail.pdfAmerican College of Physicians

www.americanheart.org/presenter.jhtml?identifier=1486American Heart Association

www.nhlbi.nih.gov/health/dci/Diseases/Hf/HF_WhatIs.htmlNational Heart, Lung, and Blood Institute

4.

3.

CME Questions

1 June 2010Annals of Internal MedicineIn the ClinicITC6-16© 2010 American College of Physicians

A 40-year-old woman is evaluated for 2 months of progressive dyspnea onexertion, orthopnea, and lower-extremityedema. She has no other medicalproblems and takes no medications,including over-the-counter drugs, andshe does not use illicit drugs. She doesnot smoke cigarettes and rarely drinksalcohol. There is no family history ofheart disease.

On physical examination, she is afebrile.Blood pressure is 120/80 mm Hg, andpulse is 80/min. Estimated centralvenous pressure is 8 cm H

2O. Her lungs

are clear. Cardiac examination reveals aregular rhythm, an S

3, and no murmurs.

She has mild ankle edema. Chestradiograph shows mild vascularcongestion. Electrocardiogram showsnormal sinus rhythm. Initial laboratoryevaluation reveals a normal hemoglobinlevel and metabolic profile, includingthyroid studies.

Which is the most appropriate initialdiagnostic test?

A. B-type natriuretic peptide levelB. EchocardiographyC. Radionuclide ventriculographyD. Stress test

A 48-year-old man is evaluated in theemergency department for dyspnea onexertion and paroxysmal nocturnaldyspnea for 3 days. He has a history oftype 2 diabetes mellitus andhypertension but no other medicalproblems. He does not smoke cigarettes.He currently takes metformin, lisinopril,and low-dose aspirin.

On physical examination, he is afebrile.Blood pressure is 130/80 mm Hg, pulse is100/min, and respiration rate is 20breaths/min; BMI is 40 kg/m2. Jugularveins are distended. Cardiac examinationreveals a normal S

1and S

2, the presence

of an S3, and a regular rate and rhythm

with no murmurs. The point of maximumimpulse is not displaced, and there areno heaves. Pulmonary auscultation

discloses crackles at the bilateral lungbases. There is mild bilateral edema tothe shins. Laboratory studies reveal aserum creatinine level of 76.3 µmol/L(1.0 mg/dL) and a B-type natriureticpeptide level of 100 ng/L. Theelectrocardiogram shows only sinustachycardia, without pathologic Q wavesor suspicious ST changes. Chestradiograph is pending.

Which is the most likely diagnosis?

A. Acute heart failureB. Acute pulmonary embolismC. Cor pulmonaleD. Recent myocardial infarction

A 60-year-old woman is evaluated forfollow-up after hospitalization 2 weeksago for pulmonary edema and volumeoverload that readily resolved withintravenous diuretics. She is currentlyfeeling well without edema or shortnessof breath. Stress echocardiography donein the hospital had negative results forischemia and showed an ejectionfraction of 60% and no significantvalvular abnormalities. She has a historyof hypertension, hyperlipidemia, andchronic atrial fibrillation. She takesmetoprolol (75 mg twice daily),hydrochlorothiazide, warfarin, aspirin,and pravastatin.

On physical examination, she is afebrile.Blood pressure is 150/90 mm Hg andpulse is 50/min. Jugular veins are notdistended, and her lungs are clear.Cardiac examination shows an irregularlyirregular rhythm with variable intensityof the S

1with no murmurs. There is no

edema.

Which is the most appropriateadjustment to her treatment?

A. Add candesartanB. Add digoxinC. Change hydrochlorothiazide to

furosemideD. Increase metoprolol dose

A 70-year-old woman is evaluated for a1-month history of dyspnea on exertionand fatigue. She can still performactivities of daily living, includingvacuuming, grocery shopping, andascending 2 flights of stairs carryinglaundry. She has a history ofhypertension, mild chronic obstructivepulmonary disease, and smoking. Hermedications are lisinopril,hydrochlorothiazide, and albuterol asneeded.

On physical examination, she is afebrile.Blood pressure is 110/80 mm Hg andpulse is 70/min. Jugular veins are notdistended. There is a grade 2/6holosystolic murmur at the left sternalborder that radiates to the axilla, whichwas not noted during an examination 1year ago. Rate and rhythm are regular, S

1

and S2are normal, and there is no S

3. The

lung sounds are distant but clearwithout wheezing, and there is noedema. Laboratory studies show normalhemoglobin and thyroid-stimulatinghormone levels. Electrocardiogram showslow voltage and left-axis deviation.Echocardiogram shows an ejectionfraction of 35%, global hypokinesis, andmild mitral regurgitation. Chestradiograph shows flattening of thediaphragms but is otherwise normal.

Which is the most appropriatetreatment?

A. AmlodipineB. CarvedilolC. DigoxinD. LosartanE. Spironolactone

1.

2.

Questions are largely from the ACP’s Medical Knowledge Self-Assessment Program (MKSAP, accessed athttp://www.acponline.org/products_services/mksap/15/?pr31). Go to www.annals.org/intheclinic/

to obtain up to 1.5 CME credits, to view explanations for correct answers, or to purchase the complete MKSAP program.