Optimizing Clinical Outcomes in Acute Decompensated Heart Failure

Login or register to view PDF.
Abstract

Acute decompensated heart failure (ADHF) is a syndrome defined by worsening fatigue, dyspnea, or edema that results from deteriorating heart function and usually leads to hospital admission or unscheduled medical intervention. It is not a homogenous syndrome, but has many faces and varying presentations. Although outcomes in stable heart failure have seen improvements in morbidity and mortality, ADHF continues to portend poor outcomes. Prompt diagnosis and identification of ADHF and its causes are critical to optimising management. Diagnostic modalities include thorough clinical evaluation, natriuretic peptide assays and, in some cases, invasive hemodynamic measurements. Optimal evidence-based treatments remain poorly defined despite the high prevalence of this condition and its associated morbidity and mortality. Complex treatment algorithms have developed, but randomised controlled trials in this area are quite sparse. Therefore, specific treatment protocols are largely a product of expert consensus and are experiential at best. Pharmacological agents such as diuretics, vasodilators, natriuretic peptides, and positive inotropes can improve symptoms. Bedside ultrafiltration, non-invasive ventilation and a host of mechanical circulatory devices have also improved the outlook for heart failure, but their optimal role in ADHF has yet to be defined. As the number of patients with ADHF increases, further investigation is needed to develop innovative, safer, and more effective therapies.

Disclosure
Sabu Thomas, MD, has no conflicts of interest to declare. Andrew Boyle, MD, is an advisor to Thoratec and Medtronic, and receives honoraria from CHF Solutions, Medtronic, and St Jude Medical. Gary Francis, MD, is an adviser to Merck, ARCA, Gilead, Forest Laboratories, sanofi-aventis, and Boston Scientific.
Correspondence
Gary S Francis, MD, Professor of Medicine, University of Minnesota, Division of Cardiology MMC 508, Minneapolis, MN. E: Franc354@umn.edu
Received date
27 February 2009
Accepted date
10 March 2009
Citation
US Cardiology - Volume 6 Issue 1;2009:6(1):50-54
Correspondence
Gary S Francis, MD, Professor of Medicine, University of Minnesota, Division of Cardiology MMC 508, Minneapolis, MN. E: Franc354@umn.edu

Acute decompensated heart failure (ADHF) is a syndrome defined by worsening fatigue, dyspnea, or edema that results from deteriorating heart function and usually leads to hospital admission or unscheduled medical intervention.1 Among patients over 65 years of age it remains the leading cause of hospital admission (>1 million admissions per year in the US alone), has an exceptionally high rate of readmission, and represents the most costly cardiovascular disorder in developed countries.2 Despite the many advances in cardiovascular disease, this syndrome continues to portend a dismal prognosis, with 60-day mortality rates approaching 20%.3,4 Furthermore, despite the development of a number of pharmacological and device-based therapies that have improved the morbidity and mortality for stable congestive heart failure, these improvements have not translated into better survival for ADHF. According to data from the Framingham heart study, between 1950 and 1969, 30-day and one-year mortality rates for ADHF were 12 and 28%, respectively, compared with almost identical 30-day and one-year mortality rates of 11 and 28%, respectively, between 1990 and 1999.5 In contrast, the current average stay (four to six days) and in-hospital mortality rates (4–5%) for ADHF have improved significantly.6

The majority of patients with ADHF deteriorate days to weeks after a period of symptom stability. Most have previously established heart failure. A multitude of causes exist for acute decompensation of stable heart failure patients. Medication and dietary non-compliance as well as diuretic resistance are common causes of acute decompensation. Important cardiac causes for acute decompensation of stable heart failure patients include new or worsening myocardial ischemia, new infarction, atrial fibrillation , and other rhythm disorders. Other non-cardiac causes include exacerbation of hypertension, anemia, thyroid disease, ingestion of toxins (cocaine, alcohol), development of fever, and new/worsening infections. ADHF of both systolic and diastolic heart failure appear indistinguishable by history and physical examination. Diastolic heart failure—also referred to as ‘heart failure with preserved ejection fraction’—represents nearly 50% of cases.7 Once patients with diastolic heart failure are admitted for ADHF, their subsequent prognosis is similar or worse than that of patients with ADHF who have a low ejection fraction.8 The absence of targeted therapies for diastolic dysfunction combined with a narrow therapeutic window for optimizing volume status makes this subset of patients a particular challenge.

Given the scope of ADHF, significant effort has been directed toward the management of this syndrome. The purpose of this article is to summarize the available treatments and outline the evidence for their efficacy. Particular attention will be devoted to informing clinicians of the optimal utilization of existing strategies. The reader is encouraged to use this article in combination with the more comprehensive practice guidelines provided by the Heart Failure Society of America (HFSA)9 and the European Society of Cardiology (ESC),10 from which the majority of the recommendations of this article are derived.

Diagnosis

The diagnosis of ADHF is always a clinical diagnosis. The initial assessment should include a focused history and physical examination. Patients most commonly present with cough, dyspnea, and fatigue, which rapidly become more severe and may be associated with chest pain or pressure. However, the signs and symptoms of ADHF can overlap with those of many other medical conditions; common mimickers of ADHF include asthma and exacerbations of chronic obstructive pulmonary disease (COPD). ADHF can develop gradually or precipitously (acute pulmonary edema). Laboratory testing is undertaken to verify the diagnosis when uncertainty exists. Echocardiography is performed to reveal specific features of cardiac geometry, valvular function, and chamber size and function. Nevertheless, the onus lies with the clinician to rapidly identify patients with ADHF in order to avoid a need for ventilatory support, a delay in hospital discharge, a need for hospital readmission, and an increase in overall resource utilization.11,12

On presentation, patients are typically tachypneic and may be using accessory respiratory muscles to breathe. Chest auscultation usually reveals crackles from pulmonary edema and up to one-third of patients have wheezing referred to as ‘cardiac asthma.’ If present, hypotension may indicate cardiogenic shock from severe ventricular dysfunction. Examination of the heart may reveal the presence of an S3 or S4 gallop. Of the many symptoms that occur during ADHF, dyspnea on minimal exertion is the most sensitive, whereas paroxysmal nocturnal dyspnea remains the most specific.13 Elevated internal jugular vein pressure is the most important physical finding of ADHF.13 In terms of diagnostic tests, interstitial edema and pulmonary venous congestion on chest radiography significantly increase the likelihood of ADHF.13

Assays for B-type natriuretic peptide (BNP) or N-terminal BNP are useful in ruling out ADHF when normal, but are not required in cases of ADHF that are obvious based on clinical criteria alone. In patients recruited to the Breathing Not-Properly study, BNP values >100pg/ml were 90% sensitive and 76% specific for the diagnosis of ADHF.14 More recently N-terminal B-type natriuretic peptide has been used. Its values are roughly five- to eight-fold higher than standard BNP assays. Natriuretic peptides are also useful in long-term risk stratification in patients with ADHF.15

On admission to hospital, clinicians should routinely obtain an electrocardiogram, a transthoracic echocardiogram (with Doppler), a complete blood count, a basic metabolic profile, arterial blood gases, cardiac enzymes (if ongoing ischemia is suspected), and thyroid function tests. Finally, although central venous lines and pulmonary artery catheterization can provide useful information regarding intracardiac hemodynamics in selected patients, their routine use is not recommended.16

Treatment

There are many options based on clinical trials for treating patients with stable, chronic heart failure. These treatments include angiotensin-converting enzyme (ACE) inhibitors,17,18 angiotensin-II receptor blockers (ARBs),19,20 β-blockers21,22 aldosterone antagonists,23 and nitrates/ hydralazine.18 Device therapies include implantable cardioverter defibrillators,24 cardiac resynchronization therapy,25 and ventricular assist devices (VADs).26 However, the existing therapies for ADHF have not been subjected to randomized controlled trials to the same extent and are therefore subject to bias and the imprecision of non-scientific experiential observation. Large clinical trials in patients with ADHF are challenging to perform because the short-term hospital mortality rate is low (4–5%), and symptom improvement and non-fatal clinical events have proved difficult to measure in a robustly quantitative manner. The short-term goals of treatment for ADHF include hemodynamic stabilization, support of oxygenation and ventilation, and relief of symptoms.

Pharmacological therapies that have been used to treat ADHF patients include intravenous (IV) diuretics, IV nitroglycerin, nitroprusside, nesiritide, vasodilating inotropes (dobutamine, milrinone), and, occasionally, vasopressor-type inotropes (dopamine and norepinephrine). Morphine sulfate, tourniquets, and phlebotomy are no longer routinely used. Non-pharmacological modalities used in patients with ADHF include ultrafiltration, positive pressure ventilation, and mechanical circulatory support devices.

Diuretic Therapy

Under most circumstances, the relief of tissue congestion is the primary goal of ADHF therapy, as volume overload is central to its pathophysiology. The clinical goal is to relieve dyspnea. Most patients who present with severe ADHF have pulmonary capillary wedge pressures that exceed 25mmHg.16 Patients whose filling pressures are reduced by diuretics will nearly always experience improvement in symptoms.27 Diuretic-induced reductions in BNP levels between admission and discharge, when observed, is comforting to physicians and can mean that such patients are less likely to fall victim to quick repeat hospitalizations—an important quality control measure in heart failure therapy.28 However, sequential measurements of plasma natriuretic pepide levels are not routinely recommended, as day-to-day variance can be substantial.29

Clinicians rely heavily on diuretic therapy with loop diuretics, in part because thiazide diuretics alone are ineffective when the glomerular filtration rate (GFR) is reduced.30 There is some evidence that continuous infusions of loop diuretics are superior to bolus infusions because of reduced activation of the renin–angiotensin–aldosterone system and more consistent tissue levels of diuretic in the kidney.31 When blood pressure is stable, loop diuretics can be combined with vasodilator therapy such as nitroprusside to quickly reduce filling pressure and improve cardiac output.
Non-loop diuretics such as metolazone and aldosterone antagonists can be used together with loop diuretics in patients who fail to respond to IV loop diuretics alone. More natriuresis is achieved when metolazone is combined with furosemide than with either drug used alone. Spironolactone can be added to reduce edema and preserve potassium, but larger doses in the range of 100mg/day may be required. Aldosterone antagonists should be avoided in patients with renal insufficiency (Cr ≥2mg/dl) and hyperkalemia.

In a meta-analysis of small randomized controlled trials, mortality rates are lower among stable heart failure patients receiving diuretics compared with placebo; however, the mortality benefits of loop diuretics in ADHF are uncertain.32 Thus, there is some degree of caution regarding their routine use in ADHF, as some studies, albeit mostly observational, report worsening mortality with increasing diuretic dose.28,33,34 Loop diuretics can directly worsen outcomes by further reducing GFR, exacerbating electrolyte abnormalities, and activating the renin–angiotensin–aldosterone system. In a small trial of 104 patients presenting to the emergency department with ADHF and respiratory distress, patients were randomized to receive either high-dose furosemide (80mg IV every 15 minutes) combined with low-dose IV isosorbide dinitrate (1mg/hour doubled every 10 minutes) or high-dose isosorbide dinitrate (3mg IV every five minutes) alone. Mechanical ventilation was required more in patients in the high-dose furosemide group compared with the isosorbide group alone (40 versus 13%; p<0.004).35 No one doubts the value of IV loop diuretics in the setting of ADHF, but it should be noted that their use can further reduce GFR, and some patients remain essentially unresponsive. Adding IV thiazide (Diuril) 500–1,000mg or oral metolazone can sometimes prompt a diuresis when loop diuretics alone are ineffective. Occasionally, ultrafiltration or even hemodialysis is necessary. Some investigators believe that ultrafiltration should be more widely used as an adjunct or even a substitute for high-dose continuous-loop diuretic drip.
Studies have prompted investigators to evaluate alternative strategies, such as adenosine-1 receptor blockers, for reducing congestion.34 Neveretheless, until such therapies are widely tested, loop diuretics will likely continue to be a cornerstone of ADHF therapy.

Ultrafiltration

Bedside ultrafiltration performed by cardiologists is a relatively novel approach for the treatment of ADHF. Although this therapy is costly, if patients can be kept out of the hospital for longer, it may prove cost-effective. In the Ultrafiltration versus IV Diuretics for Patients Hospitalized for Acute Decompensated Heart Failure Trial (UNLOAD), 200 ADHF patients were randomized to ultrafiltration versus diruetics alone. The UNLOAD trial demonstrated that ultrafiltration improved weight loss and dyspnea, decreased the need for vasoactive drugs, and reduced readmission rates with serum creatinine levels that were similar to levels seen in the diuretic group.36

Smaller clinical trials comparing ultrafiltration with IV diuretics also showed greater volume removal, more BNP reduction, shorter hospital stays, and less need for rehospitalization.34,37,38 The UNLOAD trial indicated that ultrafiltration could keep patients out of hospital longer. Ultrafiltration and adenosine-1 receptor blockers are undergoing additional study.

Vasodilators

IV vasodilators reduce congestion and improve cardiac index without increasing myocardial oxygen consumption. Randomized controlled trials in ADHF have demonstrated that IV nitrate therapy resulted in acute improvement of dyspnea.33,39 Although its use has waned in recent years, IV morphine sulfate 2–4mg acts centrally to inhibit sympathetic drive to the periphery, thus reducing symptoms of dyspnea and alleviating anxiety and air-hunger. It is most commonly used as first-line therapy for acute pulmonary edema in the emergency department setting.

As the majority of patients with ADHF have increased or normal blood pressure, reducing impedance to ejection remains important to improving forward flow.14 Short-term nitroprusside is particularly useful in patients with ADHF when blood pressure is normal or increased, particularly 24 hours after an acute myocardial infarction.40 ACE inhibitors, ARBs, and β-adrenergic blockers should be continued if tolerated and blood pressure remains satisfactory. Low blood pressures (~90–100mHg systolic) may not necessarily preclude the use of these agents if there are no symptoms and careful monitoring is ensured.41 The mean aortic pressure should be kept at 65–70mmHg. As with stable heart failure patients, renal function and potassium should be carefully monitored. The combined use of oral isosorbide dinitrate and hydralazine is an under-utilized treatment method and should be used when switching from nitroprusside to oral agents.42

Nesiritide

The atrial natriuretic peptides (ANPs) and BNPs are endogenously released in response to increased stretching of the atria and ventricle, respectively. They act to counteract increased neurohormonal activity by promoting natriuresis and vasodilation. The Vasodilation in the Management of Acute Congestive Heart Failure Trial (VMAC) randomized 498 patients with New York Heart Association (NYHA) class IV heart failure to nesiritide, IV nitoglycerin, or placebo.

The results demonstrated that pulmonary capillary wedge pressure was marginally reduced at three hours (-5.8mmHg for nesiritide versus placebo; p<0.001) and dyspnea was reduced at three hours. The 30-day readmission rate was 20% among patients in the nesiritide group compared with 23% in the nitroglycerin group. Although nesiritide was rapidly accepted as a therapy for ADHF after US Food and Drug Administraton (FDA) approval, it has recently come under scrutiny because of fears regarding potential adverse effects.43 Furthermore, a meta-analysis of randomized controlled trials of nesiritide versus placebo that looked at 30-day survival revealed a non-significant trend toward increased mortality in the nesiritide group (7.2 versus 4%; p=0.059).44 However, more recent data from the Follow-Up Serial Infusions of Nesiritide in Advanced Heart Failure (FUSION-II) study suggests that nesiritide is safe.45 It is a reasonable, albeit expensive, IV vasodilator to be used in the setting of ADHF, but lower doses may be safer and more efficacious. A large randomized controlled mortality trial of nesiritide is currently under way.

Positive Inotropes

Although positive inotropes are necessary for some patients with ADHF presenting with shock or inadequate cardiac output, their routine use for symptomatic and hemodynamic optimization is not recommended. These agents can be arrythmogenic and are associated with increased mortality.3

Dobutamine with its β-1 and mild β-2 activity is considered a positive inotrope with mild vasodilator activity. Its role in ADHF has been primarily to increase cardiac index in patients with inadequate forward flow. Older studies looking at infusions of dobutamine for up to five days in ADHF patients has shown improved symptoms for up to one month.46,47 However, more contemporary data in the ambulatory setting indicate that intermittent dobutamine therapy is associated with worse outcomes.48 The Acute Decompensated Heart Failure National Registry (ADHERE) registry revealed a higher mortality rate for IV positive inotropes, including dobutamine, compared with nesiritide or nitroglycerin.49 Dobutamine can also instigate myocardial ischemia and can worsen mitral regurgitation. However, in low-output states dobutamine can be life-saving and should be tried prior to the insertion of an intra-aortic balloon pump (IABP). Norepinephrine, vasopressin, and/or epinephrine are used to restore blood pressure if cardiogenic shock is present, usually prior to the insertion of an intra-aortic balloon pump.

Milrinone is a phosphodiesterase III inhibitor and increases myocardial contractility and peripheral vasodilation via elevated levels of cyclic adenosine monophosphate (cAMP). Unlike dobutamine, it can be added to β-antagonists. The Outcomes of a Prospective Trial of IV Milrinone for Exacerbations of Chronic Heart Failure (OPTIME-CHF) studied 951 patients admitted with ADHF with stable blood pressure. Patients were randomized to either milrinone infusion or placebo. At 60 days post-infusion, no difference was found for the primary end-point of total hospital days. There was a trend toward increased mortality in the milrinone group. There were more atrial arrhythmias and hypotensive episodes in the milrinone group.3 Milrinone is now used mainly when ADHF is associated with severe pulmonary hypertension and/or right-sided heart failure as it is a pulmonary artery pressure-lowering agent.

Despite the weight of evidence in favor of using positive inotropic agents sparingly, these drugs continue to be widely used in ADHF. In our opinion, these agents should be reserved for short-term or palliative settings, especially when there is a very low cardiac index (we favor dobutamine) or severe pulmonary hypertension (we favor milrinone).

Respiratory Therapies

In addition to supplemental oxygen therapy for hypoxia and pulmonary edema, short-term positive pressure ventilation may be considered for ADHF patients. Although previous studies have shown that continuous positive airway pressure (CPAP) improves mortality and reduces the need for mechanical ventilation,50–53 a more recent study did not demonstrate a mortality benefit.54 We do not routinely use CPAP in ADHF patients. Although not rigorously compared in randomized controlled trials, bi-level positive airway pressure (BIPAP) is probably similar in terms of these outcomes, but it does carry the additional risk of aspiration pneumonia. Among patients with central sleep apnea, CPAP improves many secondary outcomes, but the primary study was probably not sufficiently powered to demonstrate a mortality benefit.52 In ADHF associated with acute myocardial infarction and/or cardiogenic shock, mechanical ventilation with intubation is recommended over non-invasive ventilation, although the complete indications for endotracheal intubation are beyond the scope of this article.

Mechanical Circulatory Support

Patients with ADHF who remain in cardiogenic shock should be considered for mechanically-assisted circulatory support with one of the following devices: intra-aortic balloon pump (IABP), cardiopulmonary assist device, or implantable left VADs. These patients typically are severely ill, intubated, on a ventilator, and have poor urine output. They often have low cardiac indices (<2l/minute per m2), systolic blood pressures less than 90mmHg, and significantly elevated pulmonary capillary wedge pressures and hypoxemia with metabolic acidosis despite adequate medical therapy. The mortality rate is exceptionally high despite any therapy.

IABPs have two major hemodynamic mechanisms: displacement of blood to the proximal aorta during diastole with improved coronary blood flow and afterload reduction during systole via rapid balloon deflation. The expected hemodynamic effects seen with IABPs include a decrease in systolic pressure with an increase in mean and diastolic pressure, a reduction of the heart rate, a decrease in the mean pulmonary capillary wedge pressure, and an elevation in the cardiac output.53 By reducing left ventricular (LV) wall tension, an IABP also reduces myocardial oxygen demand. IABP alone is sometimes occasionally hemodynamically effective in patients with cardiogenic shock resulting from an acute myocardial infarction54 A large non-randomized multicenter trial demonstrated that IABP reversed end-organ hypoperfusion in acute myocardial infarction and cardiogenic shock when vasopressor therapy was ineffective.51 The mortality rate nevertheless is dismal and exceeds 80% in cardiogenic shock when IABP is not combined with coronary reperfusion or revascularization strategies.55,56

Cardiopulmonary assist devices are rapidly emerging as life-saving temporary treatment in patients with severe ADHF. They have important roles to play in acute cardiogenic shock, cardiopulmonary arrest, high-risk percutaneous coronary interventions (PCIs), and fulminant myocarditis presenting with shock.57 Cannulae are placed in the aorta and right atrium. Blood from the venous catheter is fed into an oxygenator and heat exchanger and returned to the arterial circuit. There role is now rapidly expanding, but they require a team approach that includes cardiac surgeons, cardiologists, and dedicated nurses.

There are three broad categories of short-term VADs used today for ADHF: centrifugal, axial flow, and left atrial-to-femoral-arterial pumps. Long-term VADs used as a bridge-to-transplant or as destination therapy are beyond the scope of this article and will not be discussed here.

Centrifugal pumps use rotating impellers that serve as a short-term bypass pump. The Bio-Medicus™ (Bio-Medicus Inc, Minneapolis, MN) and Sarns™ device (Sarns/3M, Ann Arbor, MI) are currently available for clinical use. The pumps can also be used as a right ventricular assist device (RVAD) or provide bi-ventricular (BiVAD) support in addition to its traditional left VAD role. These devices can be placed percutaneously or via median sternotomy. Despite their versatility, centrifugal pumps do have several drawbacks that include non-pulsatile flow that can lead to poor end-organ perfusion and, depending on pump design, can contribute to significant hemolysis.

Axial flow pumps are placed in a retrograde fashion across the aortic valve into the LV. Here they draw blood out of the left ventricle into the aorta in a non-pulsatile fashion with relatively low rates of hemolysis. The Impella™ device (Abiomed Inc., Danvers, MA) was implanted in 24 patients with post-cardiotomy heart failure. The mortality rate of 54%, which was non-inferior compared with patients on IABP support alone.58

Finally, left-atrial-to-femoral-arterial pumps such as the TandemHeart™ (Cardiac Assist Inc., Pittsburgh) have a venous catheter placed in the left atrium across the interatrial septum and an arterial line placed in the iliac artery. These can be placed in the catheterization laboratory in under 30 minutes and will allow stabilization of the patient in ADHF until more definitive therapies are available. The role of this device for short-term stabilization was evaluated in 18 patients with cardiogenic shock due to a myocardial infarction. After a mean of four days, the cardiac index improved from 1.7 to 2.4l/minutes/m2 and there was a significant increase in mean blood pressure and reduction in pulmonary artery, pulmonary capillary wedge, and central venous pressures.59 However, compared with IABP there was no difference in mortality.60

Conclusions

ADHF is a complex medical problem that is growing in magnitude and scope. The treatment options are increasingly complex and this has resulted in some consolidation of heart failure specialists in large tertiary referral centers. Older treatments such as diuretics and vasodilation are still widely used, but newer approaches including ultrafiltration and various circulatory mechanical assist devices are rapidly emerging. Complex treatment algorithms have developed, but randomized controlled trials are quite sparse. Judgment regarding specific treatment protocols is largely a product of expert consensus or experiential at best. As the population of patients with ADHF continues to grow, it will drive the development of more innovative and, hopefully, safer and more effective therapies.

References
  1. Felker GM, Adams KF Jr, Konstam MA, et al., Am Heart J, 2003;145(Suppl. 2):S18–25.
    Crossref | PubMed
  2. McCullough PA, Philbin EF, Spertus JA, et al., J Am Coll Cardiol, 2002;39:60–69.
    Crossref | PubMed
  3. Cuffe MS, Califf RM, Adams KF Jr, et al., JAMA, 2002;287: 1541–3.
    Crossref | PubMed
  4. Binanay C, Califf RM, Hasselblad V, et al., JAMA, 2005;294: 1625–33.
    Crossref | PubMed
  5. Levy D, Kenchaiah S, Larson MG, et al., N Engl J Med, 2002;347:1397–1402.
    Crossref | PubMed
  6. Aranda JM, Johnson JM, Conti JB, Clin Cardiol, 2009;31: 47–52.
    Crossref | PubMed
  7. Adams KF Jr, Fonarow GC, Emerman CL, et al., Am Heart J, 2005;149:209–16.
    Crossref | PubMed
  8. Redfield MM, Jacobsen SJ, Burnett JC, et al., JAMA, 2003;289:194–202.
    Crossref | PubMed
  9. Heart Failure Society of America, J Card Fail, 2006;12: e86–103.
    Crossref | PubMed
  10. Nieminen MS, Bohm M, Cowie MR, et al., Eur Heart J, 2005;26:384–416.
    Crossref | PubMed
  11. Bales AC, Sorrentino MJ, Postgrad Med, 1997; 101: 44–6.
    Crossref | PubMed
  12. Wuerz RC, Meador SA, Ann Emerg Med, 1992;21: 669–74.
    Crossref | PubMed
  13. Wang CS, Fitzgerald JM, Schulzer M, et al., JAMA, 2005;294:1944–56.
    Crossref | PubMed
  14. Maisel AS, Krishnaswamy P, Nowak RM, et al., N Engl J Med, 2002;347: 161–7.
    Crossref | PubMed
  15. Fonarow GC, Horwich TB, Rev Cardiovasc Med, 2003;(Suppl. 4): S20–28.
    PubMed
  16. Binanay C, Califf RM, Hasselblad V, et al., JAMA, 2005;294: 1625–33.
    Crossref | PubMed
  17. SOLVD Investigators, N Engl J Med, 1991;325: 293–302.
    Crossref | PubMed
  18. Cohn JN, Johnson G, Ziesche S, et al., N Engl J Med, 1991;325:303–10.
    Crossref | PubMed
  19. Cohn JN, Tognoni G, N Engl J Med, 2001;345: 1667–75.
    Crossref | PubMed
  20. McMurray JJ, Ostergren J, Swedberg K, et al., Lancet, 2003;362:767–71.
    Crossref | PubMed
  21. Hjalmarson A, Goldstein S, Fagerberg B, et al., JAMA, 2000;283:1295–1302.
    Crossref | PubMed
  22. Packer M, Bristow MR, Cohn JN, et al., N Engl J Med, 1996;334:1349–55.
    Crossref | PubMed
  23. Pitt B, Zannad F, Remme WJ, et al., N Engl J Med, 1999;341:709–17.
    Crossref | PubMed
  24. Bardy GH, Lee KL, Mark DB, et al., N Engl J Med, 2005;352:225–37.
    Crossref | PubMed
  25. Cleland JG, Daubert JC, Erdmann E, et al., N Engl J Med, 2005;352:1539–49.
    Crossref | PubMed
  26. Rose EA, Gelijins AC, Moskowitz AJ, N Engl J Med, 2001;345(2); 1435–43.
    Crossref | PubMed
  27. Nohria A, Mielniczuk LM, Stevenson LW, Am J Cardiol, 2005;96(6A):32–40G.
    Crossref | PubMed
  28. Silver MA, Maisel A, Yancy CW, et al., Congest Heart Fail, 2004;10(Suppl. 3):1–30.
    Crossref | PubMed
  29. O’Hanlon R, O’Shea P, Ledwidge M, et al., J Cardiac Fail, 2007;13(1):50–55.
    Crossref | PubMed
  30. Emerman CL, Marco TD, Costanzo MR, et al., J Card Fail, 2004;10:S116.
    Crossref
  31. Salvador DR, Rey NR, Ramos GC, et al., Cochrane Database Syst Rev, 2005(3):CD003178.
    Crossref | PubMed
  32. Faris R, Flather M, Purcell H, et al., Int J Cardiol, 2002;82: 149–58.
    Crossref | PubMed
  33. Neuberg GW, Miller AB, O’Connor CM, et al., Am Heart J, 2002;144: 31–8.
    Crossref | PubMed
  34. Domanski M, Norman J, Pitt B, et al., N Engl J Med, 1982;306: 1129–35.
  35.  
  36.  
  37.  
  38.  
  39.  
  40.  
  41. Anand IS, Rector TS, Kuskowski M, et al., Circulation, 2008;34–42.
    Crossref | PubMed
  42. Mullens W, Abrahams Z, Francis GS, et al., JACC, 2008;52(3): 200–207.
    Crossref | PubMed
  43. Sackner-Bernstein JD, Skopicki HA, Aaronson KD, Circulation, 2005;111:1487–91.
    Crossref | PubMed
  44. Sackner-Bernstein JD, Kowalski M, Fox M, et al., JAMA, 2005;293:1900–1905.
    Crossref | PubMed
  45. Yancy CW, Krum H, Massie BM, et al., Circ Heart Fail, 2008;1:9–16.
    Crossref | PubMed
  46. Unverferth DV, Magorien RD, Lewis RP, et al., Am Heart J, 1980;100:622–30.
    Crossref | PubMed
  47. Liang CS, Sherman LG, Doherty JU, et al., Circulation, 1984;69:113–19.
    Crossref | PubMed
  48. Dies F, Krell MJ, Whitlow P, Circulation, 1986;74(Suppl. II):38.
  49. Abraham WT, Adams KF, Fonarow GC, et al., J Am Coll Cardiol, 2005;46:57–64.
    Crossref | PubMed
  50. Peter JV, Moran JL, Phillips-Hughes J, et al., Lancet, 2006;367:1155–63.
    Crossref | PubMed
  51. Winck JC, Azevedo LF, Costa-Pereira A, et al., Crit Care, 2006;10:R69.
    Crossref | PubMed
  52. Bradley TD, Logan AG, Kimoff JR, et al., N Engl J Med, 2005;353:2025–33.
    Crossref | PubMed
  53. Scheidt S, Wilner G, Mueller H, et al., N Engl J Med, 1973;288:979.
    Crossref | PubMed
  54. Flaherty JT, Becker LC, Weiss JL, et al., J Am Coll Cardiol, 1985;6(2): 434–46.
    Crossref | PubMed
  55. Kern MJ, Aguirre F, Bach R, et al., Circulation, 1993;87(2):500–511.
    Crossref | PubMed
  56. Ishihara M, Sato H, Tateishi H, et al., Am Heart J, 1992;124(5):1133–8.
    Crossref | PubMed
  57. Kato S, Morimoto S, Hiramitsu S, et al., Am J Cardiol, 1999;83(4):623–5.
    Crossref | PubMed
  58. Siegenthaler MP, Brehm K, Strecker T, et al., J Thorac Cardiovasc Surg, 2004;127(3):812–22.
    Crossref | PubMed
  59. Thiele H, Lauer B, Hambrecht R, et al., Circulation, 2001;104(24):2917–22.
    Crossref | PubMed
  60. Thiele H, Sick P, Boudriot E, et al., Eur Heart J, 2005;26(13):1276–83.
    Crossref | PubMed