Medical Management of Advanced Heart Failure

CLINICAL CARDIOLOGY CLINICIAN’S CORNER

Anju Nohria, MD Context Advanced heart failure, defined as persistence of limiting symptoms de- Eldrin Lewis, MD spite therapy with agents of proven efficacy, accounts for the majority of morbidity and mortality in heart failure. Lynne Warner Stevenson, MD Objective To review current medical therapy for advanced heart failure. EART FAILURE HAS EMERGED AS Data Sources We searched MEDLINE for all articles containing the term advanced a major health challenge, in- heart failure that were published between 1980 and 2001; EMBASE was searched from creasing in prevalence as age- 1987-1999, Best Evidence from 1991-1998, and Evidence-Based Medicine from 1995- adjusted rates of myocardial 1999. The Cochrane Library also was searched for critical reviews and meta-analyses Hinfarction and stroke decline.1 Affect- of congestive heart failure. ing 4 to 5 million people in the United Study Selection Randomized controlled trials of therapy for 150 patients or more States with more than 2 million hospi- were included if advanced heart failure was represented. Other common clinical situ- talizations each year, heart failure alone ations were addressed from smaller trials as available, trials of milder heart failure, con- accounts for 2% to 3% of the national sensus guidelines, and both published and personal clinical experience. health care budget. Developments her- Data Extraction Data quality was determined by publication in peer-reviewed lit- alded in the news media increase pub- erature or inclusion in professional society guidelines. lic expectations but focus on decreas- Data Synthesis A primary focus for care of advanced heart failure is ongoing iden- ing disease progression in mild to tification and treatment of the elevated filling pressures that cause disabling symp- moderate stages2,3 or supporting the cir- toms. While angiotensin-converting enzyme inhibitors and ␤-adrenergic agents can culation mechanically for limited peri- slow disease progression and prolong survival, titration and tolerability often present ods in end-stage disease.4 Most of the challenges. Most patients are not eligible for surgical intervention but do benefit from burden of this disease is borne between a medical regimen tailored to individual clinical and hemodynamic profiles and from heart failure management programs that reduce rehospitalization. Survival ranges from these 2 boundaries by patients with ad- 80% at 2 years for patients rendered free of congestion to less than 50% at 6 months vanced heart failure, a quarter of the di- for patients with refractory symptoms, in whom end-of-life options may include hos- agnosed heart failure population. pice care and inactivation of implantable defibrillators. Advanced heart failure is defined as Conclusions Current management of advanced heart failure is based more on con- symptoms limiting daily life (New York sensus than on randomized trials. Systematic investigation should address not only new Heart Association class III or IV) de- therapies but also strategies for selecting and optimizing therapies already available. spite previous therapy with angiotensin- JAMA. 2002;287:628-640 www.jama.com converting enzyme (ACE) inhibitors, diuretics, digoxin, and more recently ␤-adrenergic blocking agents when tol- dex term advanced heart failure. Litera- reviewed from multiple sources.6-10 For erated.5 As the syndrome of heart fail- ture searches prepared during the pro- study selection, randomized controlled ure with preserved left ventricular ejec- cess of establishing 2001 guidelines6 were trials of therapy for at least 150 patients tion fraction (LVEF) is still undergoing also reviewed: EMBASE from 1987- were included if advanced heart failure definition, advanced heart failure with 1999 for English articles on human heart was represented. Other common clini- LVEF of 25% or less is the focus of this failure with emphasis on controlled tri- cal situations were addressed from review (TABLE 1). als and meta-analyses; the Best Evi- dence database from the American Col- Author Affiliation: Cardiovascular Division, Brigham and Women’s Hospital, Boston, Mass. Literature Review lege of Physicians from 1991-1998 and Financial Disclosure: Dr Stevenson is a consultant for We conducted a MEDLINE search for all Evidence-Based Medicine from 1995- Scios Inc and Medtronic Inc. Corresponding Author and Reprints: Lynne Warner articles from 1980-2001 using the in- 1999; and the Cochrane Library for meta- Stevenson, MD, Cardiovascular Division, Brigham and analyses of strategies for heart failure,con- Women’s Hospital, 75 Francis St, Tower 3-A, Bos- gestive. Approaches to management ton, MA 02115 (e-mail: [email protected]). See also Patient Page. Clinical Cardiology Section Editor: Michael S. Lauer, specific for advanced heart failure were MD, Contributing Editor.

Downloaded From: https://jamanetwork.com/ on 09/25/2021 MEDICAL MANAGEMENT OF ADVANCED HEART FAILURE

Table 1. Characteristics of a Sample of Patients With Advanced Heart Failure Undergoing Evaluation for Heart Transplantation* Characteristics 1000 Patients With Advanced Heart Failure Age, y 51 (12) Men, % 78 Coronary artery disease, % 47 Left ventricular ejection fraction, % 22 (8) Left ventricular diastolic diameter, mm 73 (11)

Peak Vo2, mL/kg per min 13 (4) 754 Patients With Repeated Measures

Baseline Hemodynamics Hemodynamics After Redesign of Therapy† Right atrial pressure, mm Hg 12 (7) 7 (4) Pulmonary capillary wedge pressure, mm Hg 25 (9) 16 (6) Cardiac index,L·minϪ1 ·mϪ2 2.1 (0.7) 2.6 (0.6) Systemic vascular resistance, dynes/s per cmϪ5 1650 (600) 1150 (300) *Data from Drazner et al.25 Data are presented as mean (SD) unless otherwise indicated. †Hemodynamic improvement obtained primarily on intravenous vasodilators and diuretics with the final measurements made on a revised oral regimen, which included captopril and nitrates in most patients with the occasional substitution or addition of hydralazine.

smaller trials, trials of milder heart fail- tients receiving amiodarone. Common Elucidation of Symptoms. Effective ure, consensus guidelines, and clinical viral infections often aggravate heart fail- therapy requires that clinicians thor- experience both published and per- ure for several weeks even after resolu- oughly understand what patients personal. Data quality for extraction was tion of viral symptoms. ceive to be the most limiting aspects of determined by publication in peer- Coronary artery disease is present in their lives. Regular review of specific ac- reviewed literature or inclusion in pro- 50% to 70% of patients with advanced tivities including dressing, climbing fessional society guidelines. heart failure.11 In such patients, the quest stairs, getting the mail, and pushing a for reversible ischemia creates many grocery cart provide better information Evaluation of Advanced questions. Revascularization is sup- than the New York Heart Association Heart Failure ported by controlled data for LVEF rang- class definitions.22 Symptoms of ad- Search for Potentially Reversible Fac- ing from 35% to 50% and by registry data vanced heart failure are dominated by tors. Evaluation of primary cause has for LVEF that is less than 35% with domi- those related to congestion, a reflection been detailed elsewhere.6 Potentially re- nant ischemic symptoms.16 Although of elevated filling pressures.9 High left- versible factors should be sought repeat- noninvasive testing is frequently per- sided filling pressures cause shortness of edly. Atrial fibrillation is present in 25% formed to demonstrate ischemic re- breath or coughing when lying down (or- to 50% of patients with advanced heart gions for revascularization,17,18 there is thopnea) or immediate dyspnea on light failure.11 Left ventricular ejection frac- no controlled evidence for screening or exertion, such as dressing or walking tion and clinical status frequently im- intervention without angina. across a room. Dyspnea occurring after prove after therapy, with continuing con- Left ventriculectomy for nonisch- sustained exercise reflects multiple he- troversy regarding the benefits of emic cardiomyopathy (popularized by modynamic, pulmonary, and skeletal restoring sinus rhythm vs diligent rate Batista) is no longer actively per- muscle abnormalities. Elevated right- control.12-14 Although digoxin may con- formed because poor outcomes were sided filling pressures can cause uncom- trol resting rate, control of ambulatory found.19 For large dyskinetic regions af- fortable edema or ascites, anorexia, early rate generally requires ␤-adrenergic ter infarction, aneurysmectomy may be satiety, abdominal fullness, and discom- blocking agents or amiodarone . undertaken. A proposed modification fort when bending. Symptoms attribut- Heavy alcohol consumption can cause with endocardial patch placement, de- able to low resting cardiac output are less and aggravate heart failure although scribed by Beyersdorf et al,20 is under specific and usually include lack of en- modest use has been associated epide- clinical investigation. Mitral valve re- ergy and fatigue. Daytime sleepiness or miologically with a lower incidence.15 pair or replacement is sometimes con- difficulty concentrating may reflect dis- Obesity not only exacerbates but also can sidered for severe valvular regurgita- turbed nocturnal sleep, severely re- cause cardiomyopathy that may reverse tion secondary to dilated heart failure.21 duced cerebral perfusion, depression, or dramatically after major weight loss al- Surgical morbidity can be high for pa- boredom due to withdrawal from usual though weight loss is rarely achieved. tients debilitated from advanced heart activities. Anemia and pulmonary emboli exacer- failure, for which patient indications for Definition of Hemodynamic Profile. bate heart failure. Thyroid abnormali- valvular surgery have not been estab- The approach to advanced heart failure ties should be sought, particularly in pa- lished. has been simplified by the consider-

Downloaded From: https://jamanetwork.com/ on 09/25/2021 MEDICAL MANAGEMENT OF ADVANCED HEART FAILURE

ation of 4 hemodynamic profiles. Most caused by local factors unrelated to cen- or greater than 10 mm Hg to predict a patients can be classified into 1 of these tral venous volume. A third heart sound pulmonary wedge pressure of less than during a 2-minute bedside assess- is present all the time in some patients or greater than 22 mm.25 Abdomino- ment.22 The 2 fundamental hemody- with heart failure and is never detected jugular pressure,26 the Valsalva maneu- namic abnormalities relate to presence in others, but changing intensity in an ver,27 and intensity of the pulmonic com- or absence of elevated filling pressures individual suggests changing filling ponent of the second sound provide (wet or dry) and perfusion that is ad- pressures. additional information to practiced equate or critically limited (warm or Skill in assessing jugular venous pres- examiners. cold) (FIGURE 1). sures requires regular practice and is the The most accessible evidence of per- Identification of elevated filling pres- most important aspect of the physical as- fusion is blood pressure. Too often re- sures in chronic heart failure relies sessment of volume status both in hos- corded casually, the blood pressure re- heavily on the symptom of orthopnea pital and in clinic.24 Jugular venous pres- quires careful auscultation to determine and the finding of elevated jugular ve- sure in centimeters is conveniently the width of the pulse pressure. In pa- nous pressure.23 Rales are absent in more measured as vertical distance from the tients with age and disease severity typi- than 80% of patients with chronically el- top of the pulsations down to the ster- cal of transplantation candidates, pro- evated filling pressures due to compen- nal angle. Estimated right atrial pres- portional pulse pressure of less than 25% sation of the pulmonary lymphatics. sure in millimeters of mercury is 3⁄4ϫ [(systolic – diastolic blood pressure)/ Residual interstitial fluid rarely compro- (the height of the jugular venous systolic blood pressure] suggests a car- mises oxygenation but often causes pressure above the sternal angle in cen- diac index, which is calculated as car- marked distress from the sensation of re- timeters+5 cm between the sternal angle diac output in liters per minute divided stricted inspiration. Edema is relatively and right atrium). When using right- by body surface area in square meters insensitive to elevated filling pressures sided filling pressures to estimate the left, (L·min−1·m−2), below 2.2 L·min−1·m−2.23 in populations younger than 70 years, oc- concordance is anticipated in approxi- This estimate requires further valida- curring in only about 25% of patients,23 mately 80% of patients with heart fail- tion, particularly in elderly patients with while in older patients, edema is often ure, using right atrial pressure of less than less compliant vessels. Patients who fall asleep during questions or have pulsus alterations may have severely reduced Figure 1. Two-Minute Assessment of Hemodynamic Profile perfusion. Cool temperature of the forearms and legs may be more specific for Evidence for Congestion low cardiac output than cold hands and (Elevated Filling Pressure) feet. Low perfusion is suspected when pa- Orthopnea High Jugular Venous Pressure tients with volume overload develop Increasing S 3 symptomatic hypotension even with low- Loud P2 Edema dose ACE inhibitors. Ascites Rales (Uncommon) In a series of 486 patients admitted Abdominojugular Reflux to a heart failure service with low LVEF, Valsalva Square Wave 67% of those with clinical evidence of Congestion at Rest? decompensation appeared on initial evaluation as profile B (wet and warm), No Yes while 28% appeared as profile C (cold Evidence for Low Perfusion Warm and Dry Warm and Wet and wet), and only 5% appeared to be Narrow Pulse Pressure No cold and dry.28 The rates of death and Pulsus Alterations AB Cool Forearms and Legs of cardiac transplantation at 1 year were May Be Sleepy, Obtunded ACE Inhibitor–Related twice as high in patients presenting as Symptomatic Hypotension Cold and Dry Cold and Wet profile C compared with those with pre- Declining Serum Sodium Level Yes Worsening Renal Function LCsenting as profile B. Although the sim- Low Perfusion at Rest? plification into these profiles has helped guide therapy and prognosis, many pa- Diagram indicating 2ϫ2 table of hemodynamic profiles for patients presenting with heart failure. Most patients admitted with congestion may ac- tients can be classified in a 2-minute bedside assessment according to the signs and symptoms shown although in practice some patients may be on the border between the warm-and-wet and cold-and-wet pro- tually have borderline perfusion, a wet files. This classification helps guide initial therapy and prognosis for patients presenting with advanced heart and lukewarm profile. failure. Although most patients presenting with hypoperfusion also have elevated filling pressures (cold and wet profile), many patients present with elevated filling pressures without major reduction in perfusion (warm and wet profile). Patients presenting with symptoms of heart failure at rest or minimal exertion without clinical Design of Therapy evidence of elevated filling pressures or hypoperfusion (warm and dry profile) should be carefully evaluated to The first priority of treatment for ad- determine whether their symptoms result from heart failure. Reprinted with permission from Dr Stevenson. vanced heart failure is relief of symp-

Downloaded From: https://jamanetwork.com/ on 09/25/2021 MEDICAL MANAGEMENT OF ADVANCED HEART FAILURE

toms, without which enthusiasm is lim- ated natriuretic peptide levels are asso- Assuming that they are already receiv- ited for decreasing disease progression ciated with improved outcome.36-38 ing ACE inhibitors, the major change is and prolonging survival, the funda- enhancement of the diuretic regimen. Pa- mental goals for therapy of heart fail- Therapy for Hemodynamic Profiles tients with profile B can often be treated ure with few or no symptoms. It will Profile B: Wet and Warm. The esti- as outpatients when fluid overload is re- be assumed that patients presenting mated hemodynamic profile guides ini- cent, diuretic responses have been well- with advanced heart failure5 have al- tial therapy.9 For patients without evi- characterized, and good follow-up is ready undergone attempted therapy dence of elevated filling pressures or available. During hospitalization for more with diuretics, ACE inhibitors, in most hypoperfusion (profile A), further complex cases, symptom relief can of- cases digoxin, and may have previous therapy is adjusted for the longer-term ten be achieved by intravenous loop di- or current therapy with ␤-adrenergic goals of maintaining stable volume sta- uretics in boluses or infusions, some- blocking agents (FIGURE 2). tus and preventing disease progression. times supplemented with other diuretics Focus on Filling Pressures. In the For patients who are profile B (wet and such as metolazone.39 Initial improve- past, heart failure was often viewed as a warm), therapy needs to dry them out. ment in congestive symptoms can be ac- disorder of low cardiac output. Focus on acute maximization of cardiac output led to therapies that increased mortality. In- Figure 2. Stepped Therapy for Heart Failure creasingly, however, therapy for heart Disease Severity failure has focused on reducing the el- Asymptomatic Symptomatic Advanced Refractory evated filling pressures that can occur Transplantation/Mechanical with or without resting hypoperfu- Assist Devices sion.29,30 Because the dominant symp- Reevaluate Diagnosis and Therapy toms are those of congestion, relief of to Relieve Persistent Congestion: resting symptoms requires reduction of More Diuresis? Nitrates ± Hydralazine? elevated filling pressures. Heart Failure Disease Management Programs Hospice Benefits of reducing filling pressures extend, however, beyond initial symp- If Needed, Use Torsemide, Intermittent Metolazone tomatic improvement. Mitral regurgita- Add Spironolactone if Normal Potassium-Handling

tion usually takes more than 50% of total Diuretics to Treat Fluid Retention left ventricular stroke volume in patients symptomatic at rest and is most ef- Digoxin for Persistent Symptoms

fectively reduced and redistributed for- Medications ? β-Blockers ? ward by therapies that reduce left ACE Inhibitor or May Need ventricular filling pressures and the ef- Angiotensin II Receptor Blocker if Severe Cough or Angioedema With ACE Inhibitor to Withdraw fective regurgitant orifice.31 Elevated filling pressures not only carry a high cost Consider 2000 mL Fluid Restriction

for myocardial oxygen consumption but No Added Salt + Salt and 2 g Na compromise the gradient for myocar- Fluid Intake dial perfusion, often increasing angina in As Tolerated Exercise Training Activity heart failure due to coronary artery dis- Aerobic ease. Neurohormonal activation and its The step diagram demonstrates addition of therapies in relation to the clinical severity of heart failure with decrease are strongly related to levels of reduced left ventricular ejection fraction. Angiotensin-converting enzyme (ACE) inhibitors are prescribed at 32,33 every level of disease severity, but they may have to be withdrawn for symptomatic hypotension or progres- left ventricular filling pressures. Ex- sive renal dysfunction in 10% to 30% of patients approaching end-stage disease, as indicated by the asterisk. ercise capacity is improved with reduc- Angiotensin receptor–blocking agents (ARBs) are a reasonable alternative for patients who cannot tolerate tion of filling pressures below the levels ACE inhibitors due to angioedema or severe cough but they are not appropriate for patients intolerant to ACE inhibitors due to symptomatic hypotension, renal failure, or hyperkalemia. ␤-Adrenergic blocking agents are usually achieved for relief of resting prescribed for patients with mild to moderate symptoms of heart failure, but they are not initiated in patients symptoms.34 Left ventricular filling pres- with severe symptoms of heart failure unresponsive to stabilization with other therapies. Diuretics are prescribed to maintain fluid balance, with spironolactone added in patients with severely symptomatic disease sures are the major factor leading to el- when renal function and potassium handling are preserved. Fluid retention persisting despite high-dose loop evated pulmonary pressures and right diuretic therapy may be better managed with torsemide, a loop diuretic with better absorption. Metolazone 25 effectively potentiates loop diuretic effects, but regular use should be avoided due to severe electrolyte deple- ventricular dysfunction. Malnutrition tion. When severe symptoms persist, patients should be reevaluated to diagnose and treat persistent conges- in heart failure35 and circulating cyto- tion. Some patients may benefit from addition of nitrates with or without hydralazine. Transplantation and kine levels relate closely to elevated right- mechanical assist devices are relevant to only a very small population with advanced heart failure. Restriction of sodium and fluid intake is increasingly required as heart failure becomes more severe. Exercise is recom- sided filling pressures and hepato- mended for all patients except those with severe resting dyspnea. Heart failure management programs are splanchnic congestion. Reductions of most cost-effective in patients at high risk for repeated heart failure hospitalizations, but they may be useful at every stage of disease. elevated filling pressures and associ-

Downloaded From: https://jamanetwork.com/ on 09/25/2021 MEDICAL MANAGEMENT OF ADVANCED HEART FAILURE

celerated by intravenous vasodilators ing pressures are lowered rapidly through Patients who appear dependent on ino- such as nitroglycerin. Nesiritide, a hu- venous and arterial vasodilation, with in- tropic infusions should undergo care- man recombinant form of the endoge- creases in cardiac output, which in turn ful re-evaluation to make sure that fill- nous brain natriuretic peptide, has been improve response to intravenous diuret- ing pressures are optimally reduced. shown to relieve symptoms rapidly. This ics. Titration is usually monitored inva- When hypotension prevents weaning of newly released drug is used with a bo- sively using a pulmonary artery cath- inotropic agents, ACE inhibitors may lus and fixed dose infusion, with or with- eter. The optimal hemodynamic profile need to be stopped. In some patients, hy- out invasive hemodynamic monitor- achieved is then maintained by adjust- dralazine, particularly in combination ing. It acts primarily as a vasodilator ing oral vasodilator agents, usually com- with nitrates, may facilitate inotropic although it also may potentiate the effect binations of ACE inhibitors, nitrates, and weaning.44 of diuretics in some patients.40 As with sometimes hydralazine, as nitroprus- Inotropic infusions are frequently used other vasodilators, the major risk is hy- side is weaned.43 Monitored nitroprus- during hospitalization, in part because potension, which persists slightly longer side infusion rarely causes symptom- of initial convenience. Although the ben- than with nitroglycerin, due to the longer atic hypotension but is occasionally efits of routine inotropic infusions of- half-life of 18 minutes. It is not known complicated by suspected cyanide tox- ten may not justify the risks, they can be whether vasodilator infusions have ben- icity, which increases with dose, dura- life-saving while definitive procedures are efits beyond early symptom relief. tion, and hepatic dysfunction. Intrave- being arranged for patients with rapidly For patients with profile B, inotro- nous nitroglycerin causes both arterial progressive hemodynamic collapse. pic therapy is not often necessary and and venodilation in this population, usu- When hemodynamic status is initially in fact may be detrimental. In fact, in a ally at higher doses than nitroprusside. unclear in patients with severe symp- recent study that demonstrated no ben- Successful transition to an effective oral toms, intravenous inotropic infusion may efit from milrinone infusion, the base- vasodilator regimen is straightforward. provide temporary stability until a more line blood pressure was 120 mm Hg, so Intravenous nesiritide may also be effec- definitive profile can be defined. For many patients would likely have been tive for vasodilation, but experience with chronic decompensation, brief inotro- profile B.41 Patients whose outpatient this new agent is limited. pic therapy may be appropriate in those clinical status had been stable while re- Intravenous inotropic infusions are as- patients with high baseline blood urea ceiving a ␤-blocking agent may con- sociated with increased risk for ische- nitrogen levels who have not demon- tinue it if perfusion is adequate and vol- mic events and tachyarrhythmias. In a strated effective diuresis in response to ume balance can be easily restored, but controlled trial of hospitalized patients repeated high dose intravenous loop di- those whose doses had recently been in- without hypotension, milrinone caused uretics with combined thiazides. In gen- creased should resume the dose that more tachyarrhythmias and symptom- eral, use of current inotropic agents in was previously tolerated. atic hypotension than placebo,41 so it heart failure may be considered as “un- Profile C: Wet and Cold. For pa- should not be used for the indication of til” therapy: until diuresis, until resolu- tients with clinical hypoperfusion, it is hypotension. Dobutamine is less likely tion of transient conditions such as pneu- usually necessary to “warm up in order to cause hypotension and is much less monia, until transplantation, or until to dry out.” For these patients in whom expensive than milrinone but also in- death, as discussed below. reflex responses support the failing cir- creases heart rate risk and of arrhyth- Profile L: Cold and Dry. Patients culation, ␤-blockers and ACE inhibi- mias. Another major limitation of ino- with profile L, those with low cardiac tors may need to be withdrawn until sta- tropic therapy is the complexity of output without clinical evidence of el- bilization is achieved, particularly for adjusting oral regimens as infusions are evated filling pressure, may be surpris- patients with symptomatic hypoten- weaned. Prolonged physiologic effects ingly stable clinically and often do not sion. Although improvement of sys- of these infusions prior to discharge may present with urgent symptoms. Those temic perfusion may often require in- mask inadequacy of the diuretic regi- admitted as cold and dry often have un- travenous therapy, for most patients low men and intolerance to vasodilator appreciated congestion. Unless they cardiac output is associated with high doses, setting the stage for readmis- have subnormal filling pressures or ex- systemic vascular resistance with pre- sion. This may be more likely with mil- cessive vasodilation, they often do not dictable improvement with vasodilator rinone due to its long half-life, further improve acutely with adjustments of therapy alone (Table 1). There is con- prolonged in the presence of renal dys- oral therapy. Inotropic infusions may siderable controversy about the rela- function. For this reason, it is recom- lead to inotropic dependence and tachy- tive roles of vasodilators and inotropic- mended that patients receiving inotro- phylaxis. Gradual introduction of ␤-ad- vasodilator agents such as dobutamine, pic infusions remain hospitalized for at renergic blocking agents, if tolerated, low-dose dopamine, and milrinone.42 least 48 hours after inotropic discon- may be associated with later clinical im- The intravenous vasodilator best stud- tinuation.9 There is also concern that ad- provement,45 which has also been ob- ied in advanced heart failure is nitro- mission for inotropic infusions as “tune- served with amiodarone.46 Benefits may prusside, a direct nitrovasodilator. Fill- ups” may create inotropic dependence.42 be greater when initial heart rates are

Downloaded From: https://jamanetwork.com/ on 09/25/2021 MEDICAL MANAGEMENT OF ADVANCED HEART FAILURE

high.47 No trials have targeted this small travenous vasodilators, then with oral put from excessively reduced filling subgroup. therapies. Although not a primary tar- pressures during diuresis. When mea- get of therapy, systemic vascular resis- sured, filling pressures usually still ex- Clinical Assessment vs tance measurement may guide titra- ceed the optimal levels needed to main- Hemodynamic Monitoring tion of vasodilators vs diuretics to tain cardiac output.50 Cardiorenal to Guide Therapy reduce filling pressures. Invasive he- interactions may involve altered bal- The clinical assessment used to define modynamic measurement may allow ance of vasodilating and vasoconstrict- profiles (Figure 1) is frequently a suf- rapid achievement of lower filling pres- ing hormones triggered by changes in ficient guide for initial therapy al- sures, but it is not known whether this cardiopulmonary pressures. Higher cre- though bedside skill in estimating he- leads to better long-term outcomes. atinine and blood urea nitrogen levels modynamics has been challenged.48The The availability of newer techniques are associated with a lower likelihood usual goals of clinically guided therapy to estimate hemodynamics may en- of maintaining freedom from conges- are normal jugular venous pressure, hance individualization of therapy. The tion52 and a greater likelihood of re- resolution of orthopnea and edema, noninvasive bioimpedance device to as- hospitalization and death.50 systolic blood pressure of at least 80 sess filling pressures and cardiac out- Optimal clinical status may be main- mm Hg and proportional pulse pres- put and the implanted hemodynamic tained for many patients at a creatinine sure of at least 25%, stable renal func- monitor for home transmission of fill- and blood urea nitrogen level higher than tion, and the ability to walk the ward ing pressures are undergoing evalua- levels during volume overload. In some without dizziness or dyspnea. tion. Brain natriuretic peptide (BNP) patients, renal function slowly im- Invasive hemodynamic monitoring is levels, which are increased by left- proves during chronic maintenance of a often used when more precise mea- ventricular distention, provide an in- lower volume state. The cardiorenal syn- sures of filling pressures, perfusion, and dex of volume status38 somewhat as he- drome is 1 of the major factors leading systemic vascular resistance are de- moglobin A1C provides an index of blood to frequent inotropic infusions, which sired. Urgent hemodynamic monitor- glucose control. Titration of therapy to can relieve congestion temporarily, but ing guides initial intervention in criti- BNP levels has been proposed49 but not the impasse usually reappears once ino- cal situations during which therapies tested in a large advanced heart failure tropic therapy is weaned. Chronic dis- must be provided rapidly to avoid cir- population. It remains to be deter- continuation of ACE inhibitors with culatory collapse. Hemodynamic moni- mined as to what extent these monitor- transition to hydralazine-nitrate combi- toring may be particularly helpful when ing technologies will influence inter- nations may become necessary when se- another condition complicates man- ventions and alter outcomes. rum creatinine levels continue to rise agement (such as pulmonary disease), higher than 3 mg/dL (265.2 µmol/L) the initial profile is not clear from clini- Cardiorenal Syndrome and/or blood urea nitrogen levels rise cal assessment and response, multiple in Heart Failure higher than 80 to 100 mg/dL (28.6- therapies need to be adjusted simulta- Among patients experiencing the late 35.7 mmol/L). Under these conditions, neously, therapy guided by empiric as- stages of advanced heart failure, the most angiotensin receptor antagonists can- sessment is ineffective to maintain common reason that symptoms cannot not substitute for ACE inhibitors due to symptom relief, or intravenous inotro- be relieved, despite aggressive manage- similar renal effects. In rare cases with pic agents cannot be weaned. ment, is the recently recognized cardio- otherwise good hemodynamic stability, Pulmonary artery catheterization is renal syndrome. Some patients, often chronic peritoneal or hemodialysis may routinely performed to evaluate pul- those with baseline renal impairment, be considered. The solution to the car- monary vascular resistance in poten- demonstrate progressive decrements in diorenal syndrome awaits better under- tial transplantation candidates. Leav- renal function even as diuresis relieves standing of the mechanisms. ing the catheter in while redesigning symptoms. This may be most common therapy over a 24- to 72-hour period with long-standing volume overload, Discharge Medical Regimen has often been followed by prolonged right ventricular dysfunction, and high Diuretics.Identification and aggres- stabilization without transplanta- baseline diuretic requirements. Aggra- sive treatment of elevated filling pres- tion.36,43 This experience has been gen- vated renal dysfunction occurs in ap- sures in advanced heart failure is the eralized in some centers to other pa- proximately 25% of patients hospital- major difference between heart failure tients with severe symptoms. When ized with heart failure, at times forcing regimens before and after referral to ter- therapy is tailored using a strategy of choices between treating renal func- tiary care centers. The average patient hemodynamic monitoring with clini- tion and providing symptom relief.50,51 referred with severe congestive symp- cal assessments, the goals are to ap- The term prerenal does not clarify ei- toms undergoes approximately 4 L proach pulmonary wedge pressure of ther the mechanism or the solution of of net diuresis.52 The diuretic regimen 16 mm Hg or less and right atrial pres- the cardiorenal syndrome. It was at one to maintain optimal volume status is sure of 8 mm Hg or less, first with in- time assumed to reflect low cardiac out- lower than that required to initiate net

Downloaded From: https://jamanetwork.com/ on 09/25/2021 MEDICAL MANAGEMENT OF ADVANCED HEART FAILURE

diuresis. When weight gain suggests vol- a 13% severe hyperkalemia in the pilot of optimal volume status, when pa- ume retention, the usual dose is supple- trial, the larger trial excluded patients tients often become more sensitive to the mented with transient dose doubling or with creatinine levels of more than 2.0 hypotensive effects of vasodilators. For intermittent addition of a thiazide, in a mg/dL (176.8 µmol/L), stopped rou- ACE inhibitors, the complex actions in- flexible patient-guided program.22 Pa- tine potassium supplementation, and clude both vasodilation, which can be tients with recurrent volume retention monitored electrolyte levels with a vigi- acutely beneficial, and neurohormonal despite high maintenance doses of fu- lance hard to maintain in general prac- antagonism, which improves long- rosemide may benefit from more reli- tice. The usual patient with advanced term prognosis but may decrease car- able absorption of the more costly loop heart failure has labile renal function pre- diac output and blood pressure when cir- diuretic torsemide (TABLE 2).53 disposing to hyperkalemia and the risk culatory compensation stimulates the Spironolactone, the aldosterone in- of hyperkalemia is further increased by renin-angiotensin system. Captopril, a hibitor, was associated with a 35% de- diabetes, high ACE inhibitor doses, or shorter acting ACE inhibitor given 3 to crease in rehospitalization in the Ran- combination therapy with angiotensin re- 4 times daily, may be easier to initiate domized Aldactone Evaluation Study ceptor–blocking agents. General use when difficulty is anticipated due to hy- trial of advanced heart failure.54 Poten- should be limited to patients requiring potension. ACE inhibitor doses may oc- tial mechanisms include inhibition of al- loop diuretics who have serum creati- casionally need to be decreased or dosterone-related fibrosis and cardiac and nine levels that are stable and lower than stopped to allow weaning of intrave- vascular remodeling, and preservation of 2.0 mg/dL (176.8 µmol/L). nous inotropic agents. Between 10% electrolyte balance, with a modest di- ACE Inhibitors and Other Oral Va- and 30% of patients with advanced uretic effect apparent in some patients. sodilators. The vasodilator regimen mer- heart failure are unable to tolerate ACE After the 25-mg dose was associated with its reevaluation after the establishment inhibitors due to hypotension or renal

Table 2. Common Medications for Advanced Heart Failure* Medication Initial Dose Maximum Dose Loop diuretics Furosemide 20-40 mg 1-2 times daily Titrate up to 400 mg/d† Bumetanide 0.5-1.0 mg 1-2 times daily Titrate up to 10 mg/d Torsemide 50 mg 1-2 times daily‡ Titrate up to 200 mg/d Supplemental thiazides Metolazone 1.25 mg/d 1⁄2 hour before loop diuretic Intermittent use to restore stable weight, dose up to 5 mg twice daily Hydrochlorothiazide 25-100 mg/d before loop diuretic dose To use instead of metolazone if weaker effect is desired Spironolactone (only with loop 25 mg/d or every other day 25 mg twice daily, occasionally higher diuretics) for refractory hypokalemia Angiotensin-converting enzyme inhibitors Captopril 6.25 mg twice daily 50-100 mg 4 times daily Enalapril maleate 2.5 mg twice daily 10-20 mg twice daily Fosinopril sodium 5-10 mg/d 40 mg/d Lisinopril 2.5-5.0 mg/d 20-40 mg/d Quinapril hydrochloride 10 mg twice daily 40 mg twice daily Ramipril 1.25-2.5 mg/d 10 mg/d ␤-Blockers Bisoprolol 1.25 mg/d 10 mg/d Carvedilol 3.125 mg twice daily 25-50 mg twice daily Metoprolol tartrate 6.25 mg twice daily 75 mg twice daily Metoprolol CR/XL§ 12.5-25 mg/d 200 mg/d Digoxin 0.125 mg every other day to 0.25 mg/d No titration except to avoid toxic effects Other vasodilators Isosorbide dinitrate࿣ 10 mg 3 times daily 80 mg 3 times daily Sublingual isosorbide 2.5 mg as occasion requires or prior to exercise to decrease dyspnea Hydralazine 25 mg 3 times daily 150 mg 4 times daily *Data are adapted from Hunt et al.6 †Titrate to achieve patient dry weight. Optional volume status may occasionally be higher than dry weight in the setting of disproportionate right ventricular failure or limitation of renal function by the cardiorenal syndrome (see “Cardiorenal Syndrome in Heart Failure” section). ‡Usually substituted for furosemide after persistent or recurring fluid retention, so initial doses may be higher. §CR/XL indicates controlled-release/extended-release metoprolol succinate. ࿣Efficacy and doses of other nitrate preparations not well established.

Downloaded From: https://jamanetwork.com/ on 09/25/2021 MEDICAL MANAGEMENT OF ADVANCED HEART FAILURE

dysfunction.4,55 In such patients, angio- compounds are prescribed by many namics often occurs after several tensin-receptor blockers are also contra- heart failure consultants to improve months. Improvement correlates in part indicated, and a combination of hydrala- symptoms, with minimal adverse ef- with reductions in elevated baseline zine and nitrates is often recommended. fects except headaches. A small con- heart rates.48 A strategy of short-term For class III to IV heart failure, hydrala- trolled trial confirmed improved exer- inotropic therapy to facilitate initia- zine-nitrate therapy has shown the same cise capacity when nitrates are added tion of ␤-blocker therapy is under in- effect on heart failure end points as to ACE inhibitors in advanced heart fail- vestigation in patients otherwise too un- ACE inhibitors although ACE inhibi- ure and suggested decrease in left ven- stable. It is not known whether the tors reduced deaths occurring unex- tricular dimensions.61 benefits of ␤-blocker therapy would be pectedly.55,56 Digitalis Glycosides. Digitalis is rec- realized in this population and whether Once a patient is stabilized and has ommended for heart failure that re- their magnitude would outweigh the been released from the hospital, ACE mains symptomatic despite ACE inhibi- known risks of inotropic therapy. inhibitor doses can be increased as tol- tors, ␤-blockers, and diuretics. In the Target daily doses of 50 to 100 mg erated. Titration to the target doses used Digitalis Investigation Group trial, carvedilol or 200 mg controlled-release/ in clinical trials (150 mg captopril, 20 heart failure hospitalizations were de- extended-release metoprolol succinate mg enalapril maleate, and 20-40 mg li- creased, most notably in patients at high- (metoprolol CR/XL) achieved in se- sinopril)3,57,58 is recommended when est risk.62 Most patients with advanced lected trial populations are less often tolerated. After the original ACE in- heart failure receive digitalis unless con- achieved in general use.65 Gradual up- hibitor trials, however, the overall regi- traindicated due to conduction system titration is indicated when tolerated, but men has expanded to include ␤-adren- disease or renal failure. Although ino- even low doses can improve LVEF and ergic blocking agents. Although every tropic state can be increased acutely with reduce the rate of rehospitalization.66 attempt should be made to advance be- high doses, the major benefit of di- Clinical experience suggests increasing yond the starting doses, there is de- goxin is now attributed to autonomic ef- doses over longer intervals such as creasing emphasis on arbitrary target fects seen at low doses.63 monthly when tolerability is a concern. doses.6,10 Maintenance of lower ACE in- Neurohormonal Antagonism With Patients must receive detailed instruc- hibitor doses is often reasonable, par- ␤-Blockers. For patients who can be sta- tions regarding early signs of volume re- ticularly for elderly patients with bor- bilized without evidence of congestion, tention and the distinction between mild derline hypotension. Compared with the focus of further therapy is improve- lassitude and circulatory collapse. Oc- lower doses, high doses of ACE inhibi- ment of long-term outcome by decreas- casionally, severe depression can de- tors led to only a 1% reduction in the ing disease progression and death. The velop. Adjustment of other medica- absolute number of annual events in the 2 neurohormonal antagonist classes that tions is frequently necessary during Assessment of Treatment with Lisino- slow disease progression are ACE in- initiation of ␤-blocking agents. When pril And Survival trial,58 a difference that hibitors and ␤-blockers. ACE inhibi- blood pressure limits the recom- was not reproduced in another trial.59 tors are recommended for all stages of mended doses of ACE inhibitors and Hydralazine has been used effec- heart failure.6 Recent trials have ex- ␤-blocking agents, it is not known how tively to wean some patients appar- panded the more limited candidate popu- they should be balanced although the ently dependent on intravenous ino- lation for ␤-blocking agents, demon- general consensus is that the use of both tropic infusions.44 Hydralazine may also strating that even patients with LVEF of agents at moderate doses is preferable to be added to ACE inhibitors to in- lower than 25% and severe symptoms the use of either alone.6,10 The critical role crease vasodilation for hypertension or frequently tolerate and benefit from cau- of heart failure management programs severely elevated systemic vascular re- tious initiation of these agents.45,64 These is highlighted by the need for meticu- sistance. Although hydralazine and ni- patients were carefully selected, how- lous patient follow-up to achieve safe and trates are commonly combined, either ever, excluding those with obvious vol- effective titration of ␤-blocking agents. can be used alone or in combination ume overload or recent intravenous in- with ACE inhibitors. fusions. When started, ␤-blockers often Maintaining Stability Nitrates were the first vasodilators increase filling pressures and decrease Hospital Discharge. Rehospitaliza- shown to improve outcome in chronic cardiac output, so initiation is not rec- tion occurs in 30% to 50% of patients heart failure.60 For patients with ad- ommended for those who are hemody- within 3 to 6 months after heart fail- vanced heart failure, nitrates are often namically compromised. Either in or out ure hospitalization.67 Failure to meet added to ACE inhibitors for further re- of the hospital, patients on the edge of criteria for discharge may increase the duction of measured filling pressures decompensation may be pushed into pul- rate of rehospitalization. Although there or to decrease dyspnea that occurs early monary edema or cardiogenic shock by is strong pressure to discharge pa- with exertion despite good volume sta- any dose of ␤-blockers. tients quickly, the last hospital day con- tus. Although unsupported by com- In patients who tolerate ␤-blockers, sumes fewer resources than the first day mercial interest, the generic nitrate improvement in LVEF and hemody- of rehospitalization. Criteria for dis-

Downloaded From: https://jamanetwork.com/ on 09/25/2021 MEDICAL MANAGEMENT OF ADVANCED HEART FAILURE

impact exceeds that demonstrated for balance is improved.72 Formal exercise Box. Discharge Criteria any single drug, emphasizing the im- training is not currently reimbursed for After Hospitalization portance of individualized care. First ap- patients with heart failure. With Heart Failure preciated in transplantation programs Adjunctive Therapies Clinical Status Goals and clinical trials, multiple variations ex- ist on the theme of intensive follow-up for Advanced Heart Failure Achievement of dry weight Definition of optimal blood by specialized heart failure nurses who The scope of this review does not in- pressure range provide education and preemptive tele- clude detailed discussion of therapies Walking without dyspnea or phone contact to identify problems and not yet incorporated into guidelines. dizziness allow patients the opportunity to ask There is some consensus that evalua- Stability Goals questions. Benefit was not shown for tion and therapy for nocturnal respi- regular follow-up care with a primary ratory disturbances should be pur- Twenty-four hours without changes 73 in oral heart failure regimen physician and nurse team, in which pa- sued. Biventricular pacing to At least 48 hours off intravenous tient satisfaction improved but hospi- synchronize left and right ventricular 70 inotropic agents, if used talization rate increased. A recent meta- contraction has improved functional ca- Fluid balance even on oral diuretics analysis indicates that hospitalization pacity and quality of life in patients with Renal function stable or improving rates are reliably decreased by experi- QRS duration over 150 millisec- 74 Home Maintenance Plan enced heart failure physician-nurse onds, present in 20% to 30% of pa- Patient and family education about teams making decisions but not by cen- tients with advanced heart failure. On- Sodium restriction tralized nursing services providing going trials will help determine the role Fluid limitation patient contact but no intervention ex- that this therapy will ultimately play in Medication schedule cept through multiple primary provid- advanced heart failure. Insertion of im- Medication effects ers.71 Meticulous monitoring and re- plantable cardioverter defibrillators is Exercise prescription sponse to changing volume status is the recommended for patients who have Flexible diuretic plan most important aspect of ongoing care had sustained ventricular arrhythmias Scheduled call to patient within for advanced heart failure.6 or cardiac arrest, or inducible ventricu- 3 days Patient education includes frequent lar tachycardia75 if the prognosis for ex- Indications for when to call nurse, review of a flexible diuretic plan ad- tended good quality of life remains oth- physician, or 911 justed according to daily weights and erwise good. Risk stratification remains Clinic appointment within 5 76 to 10 days specific information on sodium restric- difficult in other patients. Applica- tion.22 Fluid restriction to 2 L daily may tion of this technology to a broader improve stability for patients with re- heart failure population is being tested peated fluid retention despite high- in ongoing trials. charge include at least 24 hours of stable dose diuretics, particularly those with Compromised nutrition represents fluid status, blood pressure, and renal unappreciated high fluid intake previ- complex interplays between gastrointes- function on the oral regimen planned ously. Although low serum sodium tinal tract symptoms that decrease food for home (BOX).9 Patients should be free does not necessarily indicate excess wa- intake and absorption, transition to a pro- of dyspnea or symptomatic hypoten- ter intake or respond to free water re- tein-wasting state associated with chronic sion while at rest, washing, and walk- striction in this population, it reflects inflammatory activation similar to many ing on the ward. Patients who have re- intense neurohormonal activation such chronic diseases, and, less commonly, in- ceived intravenous inotropic agents that patients may need to restrict fluid creased metabolic rate.35,77 Caloric should be observed for at least 48 hours intake to maintain fluid balance as di- supplements can be helpful when in- after weaning to avoid masking inad- uretics become less effective. take is low but rarely reverse malnutri- equacy or intolerability of the dis- Vaccination for influenza and pneu- tion unless underlying factors also charge regimen.9 Education should pro- monia are recommended. For patients improve. Supplementation with micro- ceed throughout hospitalization.22 without resting symptoms, regular ex- nutrients is under investigation.78 Eryth- ercise is strongly advised, such as a walk- ropoietin with iron has been described Heart Failure Disease Management ing program outside or in enclosed shop- to improve clinical status in patients with Although most therapy for advanced ping areas or stationary bicycle and heart failure and moderate anemia.79 heart failure reflects expert consensus, treadmill exercise. The randomized con- Stress reduction techniques should heart failure management has shown trolled trials of exercise training in pa- be considered for patients with intrin- consistent benefits in historically con- tients with heart failure have demon- sic and situational anxiety. Depres- trolled and prospective randomized strated remarkable improvements in sion is increasingly recognized to de- studies, with a 25% to 75% decrease in function and reduced hospitalization crease quality of life, functional hospitalizations.68,69 The magnitude of rates, with indication that autonomic capacity, and perhaps survival with

Downloaded From: https://jamanetwork.com/ on 09/25/2021 MEDICAL MANAGEMENT OF ADVANCED HEART FAILURE

heart failure,80 but pharmacotherapy has dysfunction and of heart failure as the 2-year survival of 87%, twice that of not been studied in the population car- cause of unexplained dyspnea. For pa- those with a score of 3 through 5.52 rying both diagnoses. tients with known heart failure, it is not Accurate prediction of survival is known how much prognostic informa- meaningful only for large populations. Prognosis of Advanced tion will be provided beyond that from Even then, most estimates are based on Heart Failure routine clinical assessment. populations with lower prevalence of im- Survival ranges from 80% at 2 years for Although often considered as one of plantable cardioverter defibrillators than patients rendered free of congestion to the neurohormonal markers, BNP is re- the current 12% to 25% in advanced less than 50% at 6 months for patients leased predominantly from the left ven- heart failure. Numerical estimates of sur- with refractory symptoms at rest. Clini- tricle under stress and closely reflects vival on medical therapy are most rel- cal class, although imprecise, consis- left ventricular filling pressures. Fill- evant to those few eligible for high-risk tently provides a general estimate of ing pressures have been shown to pre- intervention such as cardiac transplan- prognosis. For patients with low LVEF, dict outcome, particularly when mea- tation or trials of new assist devices.37 reported yearly mortality has been from sured after adjustment of therapy.36 The Most patients and families facing ad- 10% to 15% with stable class II symp- elevation of filling pressure is also a ma- vanced heart failure need to under- toms, 15% to 25% for class III, and 30% jor determinant of many other prog- stand that this disease is likely to be to 50% or higher for class IV.81 Prog- nostic factors, including echocardio- fatal and that the end may come unex- nosis is usually described as worse with graphic mitral inflow patterns, severity pectedly. Once that is understood, those coronary artery disease, but the risk of mitral and tricuspid regurgitation, individuals who still demand specific added by coronary artery disease may pulmonary artery systolic pressure, and numbers for life expectancy should re- come from patients who also have dia- right ventricular ejection fraction. ceive generous estimates and encour- betes.82 Once class III through IV symp- Measurements of functional capac- agement to plan ahead for activities and toms have developed and LVEF is lower ity such as peak oxygen consumption events that provide meaning to their lives. than 25%, its exact value is less impor- from formal exercise testing are useful tant than ventricular dilation, which when following up with individual pa- Quality of Life With predicts higher mortality at every stage tients for more aggressive interven- Advanced Heart Failure of disease.83 In a study of patients re- tion such as cardiac transplantation,87 For many patients with advanced heart ferred with class IV symptoms and for adjusting exercise prescriptions, and failure, quality of life is as important as LVEF lower than 25%, 2-year survival for gauging ability to perform physi- anticipated survival. Some of the frus- was 60% with left ventricular diastolic cal work. The 6-minute walk distance tration with multiple assessment tools dimension less than 70 mm, but only correlates generally with outcome and results from the limited impact of our 20% with dimension greater than 85 is easier to perform,88 but it is not pre- current drug therapies on quality of life. mm.37 The degree of preservation of cise enough for the preceding indica- Diuretic therapy, for which there is little right ventricular function also pre- tions. Most of the predictive power of trial data but universal acceptance of ef- dicts outcome in advanced disease.84 exercise tests comes from the highest ficacy, may yield the most immediate Prognosis at every stage is wors- and lowest ranges, which are fre- and dramatic symptomatic improve- ened by even modest degrees of renal quently evident from a careful history ment. Although ACE inhibitors have not insufficiency.52,85 Serum sodium con- and observation of the patient. consistently improved quality of life, centration is a robust predictor of sur- All parameters may be more predic- they allow better maintenance of cur- vival.86 Low serum–sodium levels in- tive when reassessed after rigorous op- rent quality of life, seldom perceived by dicate intense activation of reflex timization of therapies within a dis- patients as improvement.89 ␤-Adrener- systems to preserve perfusion, al- ease management program. Clinical gic blocking agents have been associ- though they can also be caused by status remains a useful, perhaps suffi- ated with better quality of life than has chronic thiazide diuretic use. Mul- cient, predictor for most advanced heart placebo in some studies, but they often tiple parameters of neurohormonal and failure. Patients presenting with class appear neutral despite marked improve- cytokine activation predict outcome, in- IV symptoms who can be stabilized ments in LVEF.2,90 From anecdotal excluding serum norepinephrine, endo- without evidence of congestion can be periences, it seems likely that symptom- thelin, and tumor necrosis factor ␣, but upgraded to a prognosis similar to that atic improvement for some patients these are not routinely measured. The of class III patients. Using a simple con- receiving long-term ␤-blocker therapy bedside assay for brain natriuretic pep- gestion score of 0 through 5 at 1 month may be partly counterbalanced by se- tide correlates with clinical severity of after hospitalization, with 1 point each vere fatigue or depression developing in heart failure and with prognosis.38 for orthopnea, jugular venous disten- a small proportion of patients. Broader use of BNP levels as a screen- tion, ankle edema, recent weight gain, Patients with heart failure can ex- ing test is likely to increase identifica- or need for diuretic increase, showed press preferences for perceived health tion of asymptomatic left ventricular patients with a score of 0 to have a vs length of survival.91 Patients with

Downloaded From: https://jamanetwork.com/ on 09/25/2021 MEDICAL MANAGEMENT OF ADVANCED HEART FAILURE

New York Heart Association class I or nous inotropic infusions, but serious me- ure will occasionally maintain outpa- II symptoms rarely wished to trade time chanical, physiologic, and financial limi- tients who are receiving continuous or take significant risks for better health tations remain. inotropic infusions as palliative end- although most patients with class IV A unique aspect of end-stage heart stage care,6 there is more concern about symptoms expressed willingness to disease as opposed to other chronic intermittent inotropic infusions admin- trade more than half their remaining conditions is the availability of implant- istered 1 to 3 times weekly in outpa- time or risk death by more than 50% able cardioverter defibrillators. Al- tient infusion clinics. Improvements de- so they could feel better. Preferences for though these devices decrease recur- scribed on this therapy have generally quality over length of life correlated rence of life-threatening arrhythmias,75 been ascribed to the increased fre- with jugular venous pressure eleva- the appearance of tachyarrhythmias for quency of clinical contact.97 Many pa- tion, dyspnea, and peak oxygen con- some patients heralds a steepening de- tients receiving this resource-intensive sumption.91 Some patients wanted sur- cline of cardiac and clinical function. therapy at one center may be main- vival at all costs despite poor quality of Increasing hospitalizations for decom- tained elsewhere on standard oral life. The responses suggest that an un- pensation increase the likelihood of ob- therapy alone. There are currently no ac- derstanding of patient preferences serving nonsustained ventricular tachy- cepted indications for this practice.6 should help guide selection of therapy cardias. Successful defibrillation can Some lessons learned from the care of for an individual and set parameters by prolong the end of life whose quality terminal malignancy can be translated which new therapies for advanced heart has diminished to intolerable levels. In into end-stage heart failure manage- failure are evaluated. the past and in countries where defi- ment. As we recognize the elements of Functional capacity, quality of life, or brillators are not often implanted, fewer discomfort, often perceived as actual both improved early during therapy patients survive to suffer from refrac- pain, and fearfulness during terminal cir- with some agents subsequently found tory circulatory failure. Before implant- culatory failure,98 we are learning to pre- to increase mortality, such as flose- ing a device that provides a detour at scribe anxiolytics and narcotics as well quinan, a vasodilator with some ino- the end of the road, careful reconnais- as diuretics. When there is alignment of tropic action, and the inotropic agents sance should be made of the outlook patient and family wishes and commu- pimobendan and vesnarinone.92-94 for the journey. Patients should under- nity resources, hospice is increasingly These medications were not devel- stand the option to deactivate the de- used for patients with end-stage heart oped further due to serious adverse fibrillation mode. failure after implantable defibrillators are events. Although there are currently no The appropriate role of inotropic in- deactivated. known agents for heart failure that pro- fusions remains unclear. They are often duce sustained improvement in qual- initiated with the expectation that brief Horizon for Advanced ity of life with a small increased risk of support during decompensation will en- Heart Failure death, such an agent, if developed, hance diuresis and accelerate discharge The landscape of end-stage heart fail- might be a valuable option for the small although this assumption has been chal- ure treatment is changing very rap- population of patients with otherwise lenged.42 In some patients, infusions can- idly, as though viewed from a train. refractory heart failure symptoms. not be weaned without clinical deterio- Bases of controlled clinical trial evi- ration, often with progressive renal dence fall far behind the clinical deci- End-Stage Heart Failure dysfunction. Patients are occasionally dis- sions required. Current management as Despite the multiple interventions shown charged on these infusions awaiting heart failure advances is based more on to improve survival, heart failure re- transplantation.95 Unfortunately, most consensus than on randomized trials. mains a progressive disease. Fewer than patients cannot expect transplantation. Systematic investigation should ad- 5% of the patients with advanced heart Reluctance to confront the impending dress not only new therapies but also failure in the United States are eligible end of life may lead to the complex pre- strategies for selecting and optimizing for cardiac transplantation, performed in scription of home inotropic infusions, re- therapies already available. approximately 2500 patients yearly. Im- quiring indwelling central catheters with The prolongation of mild to moder- plantable mechanical cardiac support de- high infection risks, and precious com- ate heart failure with ACE inhibitors and vices have reversed life-threatening hy- munity nursing resources. For most pa- ␤-blockers, and the proliferation of poperfusion when used as a bridge for tients truly refractory to other thera- implantable defibrillators have already patients awaiting transplantation. The re- pies, this approach complicates the end created a different population of end- cently released results of the Random- of life while prolonging it by only days stage heart failure, in which right heart ized Evaluation of Mechanical Assist or weeks. The expected survival on home failure and renal dysfunction often domi- Treatment for Congestive Heart failure inotropic infusions is less than 50% at 3 nate. Current mechanical support devices (REMATCH) trial4 show that these de- to 6 months.95,96 offer a limited future but increase expec- vices improve survival as destination Although it is anticipated that pro- tations of immortality.4 Careful resource therapy for patients requiring intrave- grams specializing in advanced heart fail- allocation will be required to establish the

Downloaded From: https://jamanetwork.com/ on 09/25/2021 MEDICAL MANAGEMENT OF ADVANCED HEART FAILURE

appropriate balance between advancing ville, Md: Agency for Health Care Policy and Re- S. Left ventricular dysfunction: bedside Valsalva ma- search; June 1994. AHCPR Publication No. 94-0612. noeuvre. Br Heart J. 1980;44:560-569. technology for a select population and 8. Francis G. Approach to the patient with severe heart 28. Nohria A, Tsang S, Dries DL, Fang JC, et al. Bed- effective disease management for many. failure. In: Rose EA, ed. Management of End-Stage Car- side assessment of hemodynamic profiles identifies diac Disease. Philadelphia, Pa: Lippincott-Raven, 1998: prognostic groups in patients admitted with heart fail- Even as the end approaches, heart fail- 38-53. ure. J Card Fail. 2000;6:64. ure is characterized by wide excursion 9. Stevenson LW, Massie BM, Francis GS. Optimiz- 29. Stevenson LW, Tillisch JH. Maintenance of car- between good and bad days, often pre- ing therapy for complex or refractory heart failure: a diac output with normal filling pressures in patients management algorithm. Am Heart J. 1998;135:S293- with dilated heart failure. Circulation. 1986;74:1303- serving hope for another good day until S309. 1308. the last. Uncertainty regarding time and 10. Consensus recommendations for the manage- 30. Steimle AE, Stevenson LW, Chelimsky-Fallick C, ment of chronic heart failure: on behalf of the mem- et al. Sustained hemodynamic efficacy of therapy tai- mode of death exceeds that for other ter- bership of the advisory council to improve outcomes lored to reduce filling pressures in survivors with ad- minal illnesses.99 Even our predictions nationwide in heart failure. Am J Cardiol. 1999;83: vanced heart failure. Circulation. 1997;96:1165- of imminent death are patches of fog, 1A-38A. 1172. 11. Stevenson WG, Stevenson LW, Middlekauff HR, 31. Rosario LB, Stevenson LW, Solomon SD, Lee RT, from which survivors can emerge unex- et al. Improving survival for patients with advanced Reimold SC. The mechanism of decrease in dynamic pectedly. Each patient travels a unique heart failure: a study of 737 consecutive patients. mitral regurgitation during heart failure treatment: im- J Am Coll Cardiol. 1995;26:1417-1423. portance of reduction in the regurgitant orifice size. journey, on which many share remark- 12. Middlekauff HR, Wiener I, Stevenson WG. Low- J Am Coll Cardiol. 1998;32:1819-1824. able determination to prevail. As we em- dose amiodarone for atrial fibrillation. Am J Cardiol. 32. Johnson W, Omland T, Collins CM, Stevenson LW. 1993;72:75F-81F. Neurohormonal activation rapidly decreases after in- bolden our patients to understand and 13. Twidale N, McDonald T, Nave K, Seal A. Com- travenous vasodilator and diuretic therapy for class IV influence their course, as we appreciate parison of the effects of AV nodal ablation versus AV heart failure [abstract]. Circulation. 1998;98:I-780. their individual preferences for quality nodal modification in patients with congestive heart 33. Azevedo ER, Newton GE, Floras JS, Parker JD. Re- failure and uncontrolled atrial fibrillation. Pacing Clin ducing cardiac filling pressure lowers norepinephrine and length of life, we will guide each Electrophysiol. 1998;21:641-651. spillover in patients with chronic heart failure. Circu- other through the changing manage- 14. Pardaens K, Van Cleemput J, Vanhaecke J, Fa- lation. 2000;101:2053-2059. gard RH. Atrial fibrillation is associated with a lower 34. Chomsky DB, Lang CC, Rayos G, Wilson JR. Treat- ment of advanced heart failure. exercise capacity in male chronic heart failure pa- ment of subclinical fluid retention in patients with symp- tients. Heart. 1997;78:564-568. tomatic heart failure: effect on exercise perfor- Author Contributions: Study concept and design: 15. Abramson JL, Williams SA, Krumholz HM, Vac- mance. J Heart Lung Transplant. 1997;16:846-853. Nohria, Lewis, Stevenson. carino V. Moderate alcohol consumption and risk of 35. Carr JG, Stevenson LW, Walden JA, Heber D. Acquisition of data: Lewis, Nohria. heart failure among older persons. JAMA. 2001;285: Prevalence and hemodynamic correlates of malnutri- Analysis and interpretation of data: Stevenson. 1971-1977. tion in severe congestive heart failure secondary to is- Drafting of the manuscript: Nohria, Lewis, Stevenson. 16. Alderman EL, Fisher LD, Litwin P, et al. Results chemic or idiopathic dilated cardiomyopathy. Am J Car- Critical revision of the manuscript for important of coronary artery surgery in patients with poor left diol. 1989;63:709-713. intellectual content: Nohria, Lewis, Stevenson. ventricular function (CASS). Circulation. 1983;68: 36. Stevenson LW, Tillisch JH, Hamilton M, et al. Im- Administrative, technical, or material support: Steven- 785-795. portance of hemodynamic response to therapy in pre- son. 17. Elefteriades JA, Kron IL. CABG in advanced left ven- dicting survival with ejection fraction less than or equal Study supervision: Stevenson. tricular dysfunction. Cardiol Clin. 1995;13:35-42. to 20% secondary to ischemic or nonischemic dilated Funding/Support: Dr Stevenson’s work was sup- 18. Louie HW, Laks H, Milgalter E, et al. Ischemic car- cardiomyopathy. Am J Cardiol. 1990;66:1348-1354. ported in part by the WT Young Corp, Lexington, Ky, diomyopathy: criteria for coronary revascularization 37. Stevenson LW, Couper G, Natterson B, et al. Tar- and the Fannie Rippel Foundation, Basking Ridge, NJ. and cardiac transplantation. Circulation. 1991;84 get heart failure populations for newer therapies. Cir- Dr Nohria is supported by the Pfizer grant for train- (suppl 5):III290-III295. culation. 1995;92:II174-II181. ing in clinical investigation. Dr Lewis is the recipient 19. McCarthy PM, Starling RC, Young JB, Smedira NG, 38. Cheng V, Kazanagra R, Garcia A, et al. A rapid bed- of a Minority Faculty Development Award. Goormastic M, Buda T. Left ventricular reduction sur- side test for B-type peptide predicts treatment outcomes Acknowledgment: We thank Jerry Cornish and Tom- gery with mitral valve repair. J Heart Lung Trans- in patients admitted for decompensated heart failure: mie Rae Fuller for their expert technical and secre- plant. 2000;19:S64-S67. a pilot study. J Am Coll Cardiol. 2001;37:386-391. tarial assistance. 20. Beyersdorf F, Doenst T, Athanasuleas C, Suma H, 39. Cody R. Clinical trials of diuretic therapy in heart Buckberg GD. The beating open heart for rebuilding failure: research directions and clinical consider- ventricular geometry during surgical anterior restora- ations. J Am Coll Cardiol. 1993;22(suppl a):65A- REFERENCES tion. Semin Thorac Cardiovasc Surg. 2001;13:42-51. 71A. 1. American Heart Association Heart and Stroke 2000 21. Bolling SF, Pagani FD, Deeb GM, Bach DS. Inter- 40. Colucci WS, Elkayam U, Horton DP, et al, for the Statistical Update. Washington, DC: American Heart mediate-term outcome of mitral reconstruction in car- Nesiritide Study Group. Intravenous nesiritide, a na- Association; 2001. diomyopathy. J Thorac Cardiovasc Surg. 1998;115: triuretic peptide, in the treatment of decompensated 2. Packer M, Bristow MR, Cohn JN, et al, for the US 381-388. congestive heart failure. N Engl J Med. 2000;343: Carvedilol Heart Failure Study Group. The effect of carve- 22. Grady KL, Dracup K, Kennedy G, et al. Team man- 246-253. dilol on morbidity and mortality in patients with chronic agement of patients with heart failure: a statement 41. Felker GM, Leimberger JD, Cuffe MS, Gheor- heart failure. N Engl J Med. 1996;334:1349-1355. for healthcare professionals from the Cardiovascular giade M. Combination of digoxin and milrinone is not 3. The SOLVD Investigators. Effect of enalapril on sur- Nursing Council of the American Heart Association. associated with increased risk of adverse outcomes in vival in patients with reduced left ventricular ejection Circulation. 2000;102:2443-2456. patients hospitalized with heart failure: results from fractions and congestive heart failure. N Engl J Med. 23. Stevenson LW, Perloff JK. The limited reliability the OPTIME-CHF trial [abstract]. Circulation. 2001; 1991;325:293-302. of physical signs for estimating hemodynamics in 104:II-571. 4. Rose EA, Gelijns AC, Moskowitz AJ, Heitjan DF, et chronic heart failure. JAMA. 1989;261:884-888. 42. Felker GM, O’Connor C M. Inotropic therapy al. Long-term use of a left ventricular assist device for 24. McGee SR. Physical examination of venous pres- for heart failure: an evidence-based approach. Am end-stage heart failure. N Engl J Med. 2001;345: sure: a critical review. Am Heart J. 1998;136:10-18. Heart J. 2001;142:393-401. 1435-1443. 25. Drazner MH, Hamilton MA, Fonarow G, Creaser 43. Pierpont GL, Cohn JN, Franciosa JA. Combined 5. Adams KF, Zannad F. Clinical definition and epi- J, Flavell C, Stevenson LW. Relationship between right oral hydralazine-nitrate therapy in left ventricular fail- demiology of advanced heart failure. Am Heart J. 1998; and left-sided filling pressures in 1000 patients with ure: hemodynamic equivalency to sodium nitroprus- 135:S204-S215. advanced heart failure. J Heart Lung Transplant. 1999; side. Chest. 1978;73:8-13. 6. Hunt SA, Baker DW, Chin MH, Cinquegrani MP, 18:1126-1132. 44. Binkley PF, Starling RC, Hammer DF, Leier CV. et al. ACC/AHA Guidelines for the Evaluation and Man- 26. Butman SM, Ewy GA, Standen JR, Kern KB, Hahn Usefulness of hydralazine to withdraw from dobuta- agement of Chronic Heart Failure in the Adult. JAm E. Bedside cardiovascular examination in patients with mine in severe congestive heart failure. Am J Car- Coll Cardiol. 2001;38:2101-2113. severe chronic heart failure: importance of rest or in- diol. 1991;68:1103-1106. 7. Konstam M, Dracup K, Baker D, et al. Evaluation ducible jugular venous distension. J Am Coll Cardiol. 45. Packer M, Coats AJ, Fowler MB, et al. Effect of and Care of Patients with Left-Ventricular Systolic Dys- 1993;22:968-974. carvedilol on survival in severe chronic heart failure. function, Clinical Practice Guideline No. 11. Rock- 27. Zema MJ, Restivo B, Sos T, Sniderman KW, Kline N Engl J Med. 2001;344:1651-1658.

Downloaded From: https://jamanetwork.com/ on 09/25/2021 MEDICAL MANAGEMENT OF ADVANCED HEART FAILURE

46. Doval HC, Nul DR, Grancelli HO, Perrone SV, Bort- 64. Goldstein S, Waagstein F, Hjalmarson. The effect in patients with heart failure according to the etiol- man GR, Curiel R. Randomised trial of low-dose amio- of metoprolol CR/XL in patients with severe heart fail- ogy of left ventricular systolic dysfunction. JAmColl darone in severe congestive heart failure. Grupo de ure in MERIT. J Am Coll Cardiol. 2001;37:183A. Cardiol. 2001;38:421-428. Estudio de la Sobrevida en la Insuficiencia Cardiaca en 65. Massie BM, Greenber B, Gilbert EM, et al. Carve- 83. Lee TH, Hamilton MA, Stevenson LW, et al. Im- Argentina (GESICA). Lancet. 1994;344:493-498. dilol initiation by community physicians in COHERE. pact of left ventricular cavity size on survival in ad- 47. Schleman KA, Lindenfeld JA, Lowes BD, et al. Pre- J Card Fail. 2001;7:60. vanced heart failure. Am J Cardiol. 1993;72:672-676. dicting response to carvedilol for the treatment of heart 66. Bristow MR, Gilbert EM, Abraham WT, et al, for 84. Di Salvo TG, Mathier M, Semigran MJ, Dec GW. failure: a multivariate retrospective analysis. J Card Fail. the MOCHA Investigators. Carvedilol produces dose- Preserved right ventricular ejection fraction predicts 2001;7:4-12. related improvements in left ventricular function and exercise capacity and survival in advanced heart fail- 48. Badget RG, Lucey CR, Mulrow CD. Can the clini- survival in subjects with chronic heart failure. Circu- ure. J Am Coll Cardiol. 1995;25:1143-1153. cal examination diagnose left-sided heart failure in lation. 1996;94:2807-2816. 85. Dries DL, Exner DV, Domanski MJ, Greenberg B, adults? JAMA. 1997;277:1712-1719. 67. Roglieri JL, Futterman R, McDonough KL, et al. Stevenson LW. The prognostic implications of renal 49. Troughton RW, Frampton CM, Yandle TG, Espiner Disease management interventions to improve out- insufficiency in asymptomatic and symptomatic pa- EA, Nicholls MG, Richards AM. Treatment of heart fail- comes in congestive heart failure. Am J Manag Care. tients with left ventricular systolic dysfunction. JAm ure guided by plasma aminoterminal brain natri- 1997;3:1831-1839. Coll Cardiol. 2000;35:681-689. uretic peptide (N-BNP) concentrations. Lancet. 2000; 68. Hanumanthu S, Butler J, Chomsky D, Davis S, Wil- 86. Lee WH, Packer M. Prognostic importance of se- 355:1126-1130. son JR. Effect of a heart failure program on hospital- rum sodium concentration and its modification by con- 50. Weinfeld MS, Chertow GM, Stevenson LW. Ag- ization frequency and exercise tolerance. Circula- verting-enzyme inhibition in patients with severe gravated renal dysfunction during intensive therapy tion. 1997;96:2842-2848. chronic heart failure. Circulation. 1986;73:257-267. for advanced chronic heart failure. Am Heart J. 1999; 69. Rich MW, Beckham V, Wittenberg C, Leven CL, 87. Mancini DM, Eisen H, Kussmaul W, Mull R, Ed- 138:285-290. Freedland KE, Carney RM. A multidisciplinary inter- munds LH Jr, Wilson JR. Value of peak exercise oxy- 51. Krumholz HM, Chen YT, Vaccarino V, et al. Cor- vention to prevent the readmission of elderly pa- gen consumption for optimal timing of cardiac trans- relates and impact on outcomes of worsening renal tients with congestive heart failure. N Engl J Med. 1995; plantation in ambulatory patients with heart failure. function in patients Ͼ or=65 years of age with heart 333:1190-1195. Circulation. 1991;83:778-786. failure. Am J Cardiol. 2000;85:1110-1113. 70. Weinberger M, Oddone EZ, Henderson WG. Does 88. Bittner V, Weiner DH, Yusuf S, et al, for the SOLVD 52. Lucas C, Johnson W, Hamilton MA, et al. Free- increased access to primary care reduce hospital re- Investigators. Prediction of mortality and morbidity with dom from congestion predicts good survival despite admissions: Veterans Affairs Cooperative Study Group a 6-minute walk test in patients with left ventricular previous class IV symptoms of heart failure. Am on Primary Care and Hospital Readmission. N Engl J dysfunction. JAMA. 1993;270:1702-1707. Heart J. 2000;140:840-847. Med. 1996;334:1441-1447. 89. Rector TS, Johnson G, Dunkman WB, et al. Evalu- 53. Murray MD, Forthhofer MM, Bennett SJ, Per- 71. McAlister FA, Lawson FM, Teo KK, Armstrong PW. ation by patients with heart failure of the effects of kins SM, Stroupe KT, Tierney WM, Brater C. Effec- A systematic review of randomized trials of disease enalapril compared with hydralizine plus isosorbide tiveness of torsemide and furosemide in the treat- management programs in heart failure. Am J Med. dinitrate on quality of life. Circulation. 1993;87:VI- ment of congestive heart failure: results of a 2001;110:378-384. 71-77. prospective, randomized trial. Circulation. 1999;100 72. Hambrecht R, Gielen S, Link A, et al. Efects of ex- 90. Metra M, Nardi M, Giubbini R, Dei Cas L. Effects (suppl 1):300. ercise training on left ventricular function and periph- of short- and long-term carvedilol administration on 54. Pitt B, Zannad F, Remme WJ, et al, for the Ran- eral resistance in patients with chronic heart failure: a rest and exercise hemodynamic variables, exercise ca- domized Aldactone Evaluation Study Investigators. The randomized trial. JAMA. 2000;283:3095-3101. pacity and clinical conditions in patients with idio- effect of spironolactone on morbidity and mortality 73. Tkacova R, Rankin F, Fitzgerald FS, Floras JS, Bra- pathic dilated cardiomyopathy. J Am Coll Cardiol. in patients with severe heart failure. N Engl J Med. dley TD. Effects of continuous positive airway pres- 1994;24:1678-1687. 1999;341:709-717. sure on obstructive sleep apnea and left ventricular 91. Lewis EF, Johnson PA, Johnson W, Collins C, Grif- 55. Fonarow GC, Chelimsky-Fallick C, Stevenson LW, afterload in patients with heart failure. Circulation. fin L, Stevenson LW. Preferences for quality of life or et al. Effect of direct vasodilation with hydralazine vs 1998;98:2269-2275. survival expressed by patients with heart failure. J Heart angiotensin-converting enzyme inhibition with cap- 74. Cazeau S, Leclercq C, Lavergne T, et al. Effects Lung Transplant. 2001;20:1016-1024. topril on mortality in advanced heart failure: the Hy-C of multisite biventricular pacing in patients with heart 92. Kubo SH. Effects of pimobendan on exercise tol- trial. J Am Coll Cardiol. 1992;19:842-850. failure and intraventricular conduction delay. N Engl erance and quality of life in patients with heart fail- 56. Cohn JN, Johnson G, Ziesche S, et al. A compari- J Med. 2001;344:873-880. ure. Cardiology. 1997;88(suppl 2):21-27. son of enalapril with hydralazine-isosorbide dinitrate 75. Moss AJ, Hall WJ, Cannom DS, et al, for the Mul- 93. Massie BM, Berk MR, Brozena SC, et al. Can fur- in the treatment of chronic congestive heart failure. ticenter Automatic Defibrillator Implantation Trial In- ther benefit be achieved by adding flosequinan to pa- N Engl J Med. 1991;325:303-310. vestigators. Improved survival with an implanted de- tients with congestive heart failure who remain symp- 57. Pfeffer MA, Lamas GA, Vaughan DE, Parisi AF, fibrillator in patients with coronary disease at high risk tomatic on diuretic, digoxin, and an angiotensin Braunwald E. Effect of captopril on progressive ven- for ventricular arrhythmia. N Engl J Med. 1996;335: converting enzyme inhibitor? results of the flosequinan- tricular dilatation after anterior myocardial infarc- 1933-1940. ACE inhibitor trial (FACET). Circulation. 1993;88:492- tion. N Engl J Med. 1988;319:80-86. 76. Stevenson WG, Stevenson LW, Middlekauff HR, 501. 58. Packer M, Poole-Wilson PA, Armstrong PW, et al. Saxon LA. Sudden death prevention in patients with 94. Cohn JN, Goldstein SO, Greenberg BH, et al, for Comparative effects of low and high doses of the an- advanced ventricular dysfunction. Circulation. 1993; the Vesnarinone Trial Investigators. A dose- giotensin-converting enzyme inhibitor, lisinopril, on mor- 88:2953-2961. dependent increase in mortality with vesnarinone bidity and mortality in chronic heart failure: ATLAS Study 77. Anker SD, Chua TP, Ponikowski P, et al. Hor- among patients with severe heart failure. N Engl J Med. Group. Circulation. 1999;100:2312-2318. monal changes and catabolic/anabolic imbalance in 1998;339:1810-1816. 59. The NETWORK Investigators. Clinical outcome chronic heart failure and their importance for cardiac 95. Miller LW. Outpatient dobutamine for refrac- with enalapril in symptomatic chronic heart failure; a cachexia. Circulation. 1997;96:526-534. tory congestive heart failure: advantages, tech- dose comparison. Eur Heart J. 1998;19:481-489. 78. Witte KK, Clark AL Cleland JG. Chronic heart fail- niques, and results. J Heart Lung Transplant. 1991; 60. Cohn JN, Mathew KJ, Franciosa JA, Snow JA. ure and micronutrients. J Am Coll Cardiol. 2001;37: 10:482-487. Chronic vasodilator therapy in the management of car- 1765-1774. 96. Sindone AP, Keogh AM, Macdonald PS, McCo- diogenic shock and intractable left ventricular failure. 79. Silverberg DS, Wexler D, Sheps D, et al. The effect sker CJ, Kaan AF. Continuous home ambulatory in- Ann Intern Med. 1974;81:777-780. of correction of mild anemia in severe, resistant con- travenous inotropic drug therapy in severe heart fail- 61. Elkayam U, Johnson JV, Shotan A, et al. Double- gestive heart failure using subcutaneous erythropo- ure: safety and cost efficacy. Am Heart J. 1997;134: blind, placebo-controlled study to evaluate the effect etin and intravenous iron: a randomized controlled 889-900. of organic nitrates in patients with chronic heart fail- study. J Am Coll Cardiol. 2001;37:1775-1780. 97. Oliva F, Latini R, Politi A, et al. Intermittent ure treated with antiotension-converting enzyme in- 80. Skotzko CE, Krichten C, Zietowski G, et al. De- 6-month low-dose dobutamine infusion in severe heart hibition. Circulation. 1999;99:2652-2657. pression is common and precludes accurate assess- failure: DICE multicenter trial. Am Heart J. 1999;138: 62. The effect of digoxin on mortality and morbidity ment of functional status in elderly patients with con- 247-253. in patients with heart failure. N Engl J Med. 1997; gestive heart failure. J Card Fail. 2000;6:300-305. 98. Krumholz HM, Phillips RS, Hamel MB, et al. Re- 336:525-533. 81. Uretsky BF, Sheahan RG. Primary prevention of suscitation preferences among patients with severe 63. Slatton ML, Irani WN, Hall SA, et al. Does di- sudden cardiac death in heart failure: will the solu- congestive heart failure: results from the SUPPORT goxin provide additional hemodynamic and auto- tion be shocking? J Am Coll Cardiol. 1997;30:1589- project. Circulation. 1998;98:648-655. nomic benefit at higher doses in patients with mild to 1597. 99. Stevenson LW. Rites and responsibility for resus- moderate heart failure and normal sinus rhythm? JAm 82. Dries DL, Sweitzer NK, Drazner MH, Stevenson citation in heart failure: tread gently on the thin places. Coll Cardiol. 1997;29:1206-1213. LW, Gersh BJ. Prognostic impact of diabetes mellitus Circulation. 1998;98:619-622.

Downloaded From: https://jamanetwork.com/ on 09/25/2021