|Home | About | Journals | Submit | Contact Us | Français|
Heart failure remains a common diagnosis, especially in older individuals. It continues to be associated with significant morbidity and mortality, but major advances in both diagnosis and management have occurred and will continue to improve symptoms and other outcomes in patients. The Canadian Cardiovascular Society published its first consensus conference recommendations on the diagnosis and management of heart failure in 1994, followed by two brief updates, and reconvened this consensus conference to provide a comprehensive review of current knowledge and management strategies.
New clinical trial evidence and meta-analyses were critically reviewed by a multidisciplinary primary panel who developed both recommendations and practical tips, which were reviewed by a secondary panel. The resulting document is intended to provide practical advice for specialists, family physicians, nurses, pharmacists and others who are involved in the care of heart failure patients.
Management of heart failure begins with an accurate diagnosis, and requires rational combination drug therapy, individualization of care for each patient (based on their symptoms, clinical presentation and disease severity), appropriate mechanical interventions including revascularization and devices, collaborative efforts among health care professionals, and education and cooperation of the patient and their immediate caregivers. The goal is to translate best evidence-based therapies into clinical practice with a measureable impact on the health of heart failure patients in Canada.
L’insuffisance cardiaque demeure un diagnostic répandu, surtout chez les sujets âgés. Elle continue d’être associée à une morbidité et à une mortalité importantes, mais de grands progrès ont été accomplis sur les plans du diagnostic et de la prise en charge de la maladie et ils amélioreront encore les symptômes et le pronostic des patients. C’est en 1994 que la Société canadienne de cardiologie a publié son premier rapport consensuel sur le diagnostic et la prise en charge de l’insuffisance cardiaque, suivi de brèves mises à jour; et elle a choisi de répéter l’exercice afin de faire le point de manière globale sur les connaissances et les stratégies thérapeutiques actuelles.
Les conclusions et méta-analyses d’essais cliniques récents ont été analysées par un comité principal pluridisciplinaire qui a formulé ses recommandations et ses conseils pratiques avant de les soumettre à un second comité. Le document qui en résulte vise à guider de manière concrète les spécialistes, médecins de famille, infirmières, pharmaciens et autres intervenants appelés à soigner les insuffisants cardiaques.
La prise en charge de l’insuffisance cardiaque commence par un diagnostic juste et repose sur l’administration de traitements pharmacologiques associatifs, sur une individualisation de l’approche thérapeutique (selon les symptômes, le tableau clinique et la gravité de chaque cas), sur des interventions de type mécanique appropriées, telles que l’angioplastie ou d’autres dispositifs, sur une approche pluridisciplinaire concertée et sur l’enseignement au patient et à ses aidants naturels, pour une meilleure coopération. L’objectif est de transposer dans la pratique clinique les meilleurs traitements issus de la recherche, éprouvés et susceptibles d’exercer un impact véritable sur la santé des insuffisants cardiaques au Canada.
Many definitions of heart failure have been used, reflecting the existing understanding of the pathophysiological condition at that time. Heart failure is a complex syndrome in which abnormal heart function results in, or increases the subsequent risk of, clinical symptoms and signs of low cardiac output and/or pulmonary or systemic congestion. Because most evidence-based recommendations are based on clinical trials where significant left ventricular systolic dysfunction is present, the term ‘heart failure’ is used in this document to refer to predominant left ventricular systolic dysfunction unless otherwise stated. Diastolic heart failure (or heart failure with preserved systolic function [PSF]), right heart failure, left or right ventricular failure, biventricular heart failure, congestive heart failure (CHF), acute or chronic heart failure, cardiomyopathy, dilated cardiomyopathy, restrictive cardiomyopathy, ischemic cardiomyopathy and nonischemic cardiomyopathy are examples of other terms often used in clinical practice and research to describe specific presentations and underlying pathologies.
Heart failure is common, especially in older patients, and its incidence is predicted to increase (1). It reduces quality of life, exercise tolerance and survival. Depending on the severity of symptoms, heart dysfunction, age and other factors, heart failure can be associated with an annual mortality of 5% to 50%. A better understanding of the underlying pathophysiological mechanisms, combined with many new treatments developed over the past 20 years, has greatly improved the prognosis; many patients can now hope for long periods of stable, improved symptoms and improved heart function. Nonetheless, an inexorable course can also occur, and many new approaches to treatment continue to be developed.
This consensus conference was convened by the Canadian Cardiovascular Society (CCS) to review new evidence and update previous consensus conferences (2–4) to provide a set of evidence-informed recommendations that would provide clinicians, and other health care professionals involved in the management of heart failure patients, with clear directions and options to optimize care of individual patients. Furthermore, a concurrent plan for knowledge translation was developed. Through increased use of these evidence-based proven therapies and other recommendations based on the consensus of heart failure experts where adequate clinical trial evidence was not available, the purpose is to improve health outcomes and quality of life across the broad spectrum of heart failure patients in Canada and to measure that impact. Specific patient subgroups are identified in individual recommendations when appropriate. The present document does not repeat the reviews of data presented in the previous consensus conferences, but aims to highlight new data while updating previous recommendations where appropriate. Readers are referred to the previous publications for additional background information and rationale. New or expanded sections cover diagnosis and investigation, acute heart failure (AHF), multidisciplinary care and heart failure clinics, polypharmacy, device therapy, surgical approaches, heart failure in the elderly and issues related to end-of-life care. Following a review of the literature and a critical appraisal of the evidence, recommendations were arrived at by informed consensus through face-to-face meetings, conference calls, e-mail correspondence, and final review by all members of both the primary and the secondary panel. The primary panelists were principally responsible for the document, but the secondary panelists reviewed the recommendations and provided feedback, and some were involved in section working groups.
The class of recommendation and the grade of evidence were determined as follows:
Class I: Evidence or general agreement that a given procedure or treatment is beneficial, useful and effective.
Class II: Conflicting evidence or a divergence of opinion about the usefulness or efficacy of the procedure or treatment.
Class IIa: Weight of evidence is in favour of usefulness or efficacy.
Class IIb: Usefulness or efficacy is less well established by evidence or opinion.
Class III: Evidence or general agreement that the procedure or treatment is not useful or effective and in some cases may be harmful.
Level of Evidence A: Data derived from multiple randomized clinical trials or meta-analyses.
Level of Evidence B: Data derived from a single randomized clinical trial or nonrandomized studies.
Level of Evidence C: Consensus of opinion of experts and/or small studies.
The diagnosis of clinical heart failure is made when symptoms and signs of impaired cardiac output and/or volume overload are documented in the setting of abnormal systolic and/or diastolic cardiac function. The cardinal triad of edema, fatigue and dyspnea is neither a sensitive nor a specific manifestation of heart failure, and atypical presentations of heart failure should be recognized (Table 1), particularly when evaluating women, obese patients and the elderly. A relevant clinical history and physical examination should be performed in all patients, and initial investigations should be targeted to confirm or exclude heart failure as the diagnosis and to identify systemic disorders (eg, thyroid dysfunction) that may affect its development or progression (Figure 1). Measurement of plasma natriuretic peptides, such as BNP, is likely to become more widely available and is helpful because low concentrations are useful in excluding heart failure and high concentrations can confirm heart failure in patients presenting with dyspnea when the clinical diagnosis remains uncertain (5).
Two-dimensional and Doppler transthoracic echocardiography are the initial imaging modalities of choice in patients suspected to have heart failure because they assess systolic and diastolic ventricular function, wall thickness, chamber sizes, valvular function and pericardial disease. Radionuclide angiography is useful in patients where echocardiographic images are poor (eg, obese patients and patients with emphysema). Cardiac catheterization with hemodynamic measurements and contrast ventriculography, or magnetic resonance imaging when available, may be used in specific cases in which initial noninvasive tests are inconclusive.
Once heart failure is diagnosed, functional capacity should be assessed to document the degree of physical limitations, and the NYHA functional classification (6) is recommended as a simple, validated measure of heart failure clinical severity (Table 2). A six-minute walk test may help assess exercise limitations and prognosis. Cardiopulmonary exercise testing is infrequently necessary but may be used to determine the extent to which heart failure contributes to exercise intolerance, particularly in patients in whom there is disparity between the reported symptoms and the clinical assessment. When coronary artery disease is suspected, noninvasive testing, such as radionuclide perfusion imaging or stress echocardiography, is useful to ascertain the presence or extent of myocardial infarction, ischemia or viability that may warrant further evaluation. Coronary angiography should also be considered, especially in those who have angina or positive noninvasive tests and are candidates for revascularization. Endomyocardial biopsy is not recommended in the routine evaluation of heart failure; it has limited diagnostic value except in suspected rare disorders, such as infiltrative or inflammatory myocardial diseases.
The role of exercise training in the management of heart failure was last reviewed in the 2001 CCS consensus guideline update for the management and prevention of heart failure (3). Exercise intolerance is recognized as a hallmark of heart failure. Until the late 1980s, heart failure patients were advised to avoid physical activity in the hope that it might minimize symptoms and protect the already damaged heart. It is now understood that exercise intolerance in heart failure has a multifactorial etiology and that parameters such as intracardiac filling pressures and LVEF may not be reliable predictors of exercise capacity. Changes in the periphery and left ventricular function are both important determinants of exercise capacity. Exercise training programs in selected patients have been shown to be safe, but they also can reverse many of these peripheral abnormalities that are believed to play a role in exercise intolerance and improve overall exercise capacity (7,8). As a result, there has been a gradual move from reluctance to consider exercise programs for patients with heart failure and left ventricular dysfunction toward referral of selected patients.
Numerous clinical and mechanistic studies and some randomized studies have shown that regular exercise performed by either interval training (eg, biking and treadmill training) or steady state exercise can safely increase physical capacity by 15% to 25% and improve symptoms and quality of life in patients with NYHA II to III heart failure (7,8). However, the studies have been small and have used mainly physiological end points. The Exercise training meta-analysis of trials in patients with chronic heart failure (ExTraMATCH) Collaborative (9) addressed the question of whether exercise training reduces morbidity and mortality in heart failure patients by using individual patient data from nine relatively small studies published since 1990 involving a total of 801 patients. The ExTraMATCH review provided further support for the safety of exercise training in stabilized NYHA I to III heart failure patients, and reported relative risk reductions of 32% for death and 23% for the combined end point (death or hospital admission) for exercise training versus usual care. Several detailed discussions are available on exercise and CHF (10–13).
The data stimulate continued enthusiasm for ongoing research into how exercise training may affect outcomes and for conducting a more definitive trial – recognizing the difficulties of performing such a trial in this patient population. A 3000-patient, multicentre trial, Heart Failure and A Controlled Trial Investigating Outcomes of Exercise TraiNing (HF-ACTION), sponsored by the National Heart, Lung, and Blood Institute is ongoing in the United States, Canada and Europe.
Forced fluid intake beyond normal requirements to prevent thirst is not recommended (class III, level C).
Despite the clear survival benefits supporting the use of pharmacological therapies in the management of heart failure patients, prognosis associated with recurrent and prolonged hospitalizations remains poor. In recent years, many small, randomized clinical trials evaluating different multidisciplinary strategies have shown benefits on recurrent hospitalizations and duration of hospital stay. In these studies, many of the interventions were similar, including patient education, telemanagement, and home and hospital-based clinic visits with health care professionals specialized in heart failure care. Based on these trials, systematic reviews and meta-analyses have been published evaluating the effectiveness of multidisciplinary heart failure management programs (15–17). Strategies incorporating postdischarge follow-up by a multidisciplinary team of specially trained staff and/or access to specialized heart failure clinics reduced mortality and all-cause hospitalizations. Although there were conflicting results between earlier systematic overviews on the survival benefit of these interventions, a recent review (15) found a significant reduction in all-cause mortality.
Patients with recent or recurrent hospital admissions for heart failure appear to benefit the most from multidisciplinary heart failure or function clinics. Despite an improvement in short-term clinical outcomes, the persistence of long-term benefits and cost-effectiveness of these strategies remain to be determined after patients have stabilized. Heart failure or function clinics characterized by specialized multidisciplinary care can provide evidence-based medical therapy and referral to appropriate electrophysiological and surgical interventions. Patient education is a common key principle to improve patients’ recognition of early warning symptoms and signs and to provide the patient with strategies they can use to intervene early and prevent further acute deterioration.
All patients with heart failure and an LVEF less than 40% should be treated with an angiotensin-converting enzyme(ACE) inhibitor in combination with a beta-blocker unless a specific contraindication exists (class I, level A).
There have been many landmark clinical trials and meta-analyses of the use of ACE inhibitors (19–25) and beta-blockers (26–29) in heart failure, as well as other meta-analyses (30–32), such that ACE inhibitors and beta-blockers have become standard therapy and should be considered in all patients diagnosed with heart failure. The timing of introduction should be individualized to maximize tolerability and long-term persistence with therapy. In general, acute symptoms should be relieved, but an ACE inhibitor or a beta-blocker should be introduced as early as the patient’s condition allows. Heart rate and blood pressure abnormalities may dictate which drug class should be used first or preferentially uptitrated. Because most of the clinical trials studied ACE inhibitors first, most physicians would start with an ACE inhibitor and add a beta-blocker but not necessarily delay introduction until the target ACE inhibitor dose was reached. The Cardiac Insufficiency Bisoprolol Study(CIBIS) III (33), a recent open-label trial of 1010 patients with mild to moderate heart failure and an LVEF of 35% or less, showed that both strategies of ACE inhibitor or beta-blocker for the first six months, followed by the combination for six to 24 months, were similar, with some small, nonsignificant differences in tolerability and outcome. Heart rate, blood pressure and comorbidities may dictate which drug class should be used first or preferentially uptitrated. If an ACE inhibitor is not tolerated, there is good evidence that an angiotensin receptor blocker (ARB) can be substituted (34,35), and this may also apply if a beta-blocker is not tolerated, although those data are not as strong. In patients who are already on combination ACE inhibitor plus beta-blocker, but continue to have heart failure symptoms or hospitalizations, an ARB should be added (36–38). Aldosterone antagonists (spironolactone is the only agent available in Canada) are effective in patients with severe heart failure postmyocardial infarction or in chronic follow-up, especially if recently hospitalized for heart failure (39,40). Symptoms, blood pressure sitting and standing, heart rate, renal function and electrolytes should be followed closely when combinations of drugs affecting the reninangiotensin-aldosterone system are used. A previous study (22) compared an ACE inhibitor with isosorbide dinitrate and hydralazine combination and found that the ACE inhibitor enalapril reduced mortality at two years. The recent African-American Heart Failure Trial (A-HeFt) (41) of self-identified African-American patients with systolic heart failure showed that adding a fixed-dose combination of isosorbide dinitrate plus hydralazine to standard therapy reduced mortality as well as first hospitalization for heart failure and improved quality of life.
Patients with heart failure are generally elderly and have multiple comorbidities; therefore, the addition of multidrug therapy for heart failure adds to an already complex pharmacological regimen. As such, drug interactions, additive adverse effects (such as hypotension) and poor medication adherence occur commonly (Table 4).
Patients with heart failure are clinically fragile (42,43) and are especially susceptible to drugs that worsen heart failure symptoms (either by reducing contractility or by causing fluid retention). Drug-induced heart failure has been recently reviewed (44,45). Medications implicated in exacerbation of heart failure include calcium channel blockers (nifedipine, diltiazem and verapamil), thiazolidinediones (pioglitazone, rosiglitazone), antiarrhythmic agents, doxorubicin, NSAIDs, celocoxib and beta-blockers (Table 4). Isolated reports (46) also implicate corticosteroids, tricyclic antidepressants, penicillins, clozapine, venlafaxine, zidovudine, licorice-containing products and anti-cancer agents. Because patients may receive these agents from various health care providers, good communication is necessary to avoid iatrogenic heart failure exacerbations.
Heart failure with PSF (LVEF greater than 40%) is also known as diastolic heart failure, but because the commonly available measurement techniques are imperfect in detecting diastolic dysfunction even if it is present, we use the former as the preferred term in the present recommendations. Although it has been recognized for a number of years that patients can have heart failure with PSF, only more recently has there been a greater appreciation of the prevalence of this condition (48,49). Depending on the study examined, approximately 40% to 50% of patients with clinical heart failure have PSF (50). Clinically, it is difficult to distinguish this condition from heart failure with a reduced LVEF (less than 40%). However, heart failure with PSF is more prevalent in the elderly, in women and in patients with a history of hypertension or, less often, ischemic heart disease. In practice, the diagnosis is generally based on the finding of typical symptoms and signs of heart failure in a patient who is shown to have a normal LVEF and no valvular abnormalities on echocardiography (51). The mortality associated with this condition may be somewhat better than that found with heart failure with a reduced LVEF, although some studies have suggested it may be the same. However, studies have generally shown that morbidity, especially heart failure hospitalizations, is similar to that found with heart failure and a reduced LVEF. Recommendations for treatment of this condition remain speculative because of the limited data available on various therapies (52). The treatment is based on control of physiological factors known to exert important effects on ventricular diastolic function. ACE inhibitors may improve relaxation and cardiac distensibility directly. Diuretics are useful to treat fluid overload but should be used cautiously to avoid producing a significant reduction in preload that could adversely affect cardiac filling. Beta-blockers may be useful to improve symptoms by decreasing heart rate and increasing diastolic filling time. Calcium channel blockers, such as verapamil, may also be useful for improving symptoms by decreasing heart rate and increasing diastolic filling time. A recent study (53) showed that ARBs may be useful for reducing heart failure hospitalizations.
Atrial fibrillation is a relatively common problem for heart failure patients. The presence of atrial fibrillation can potentially cause an adverse effect in several different ways (54–57). Loss of atrial enhancement of ventricular filling may compromise cardiac output. The increase in ventricular rate in those who are not controlled may increase myocardial oxygen demand and, because of a decrease in diastolic time, produce a decrease in coronary perfusion. Also, a poorly controlled ventricular rate may cause impairment of both cardiac contraction and cardiac relaxation. Atrial fibrillation may also cause atrial thrombosis with the increased risk of embolization from the atria. Interestingly, there are no current data to support that aggressive rhythm control improves mortality or morbidity, although a large clinical trial, Atrial Fibrillation and Congestive Heart Failure (58), is in progress. However, there are data to support that a rate-control strategy is associated with fewer hospitalizations and fewer side effects from drug therapy. Thus, the treatment for these patients should be individualized.
AHF can best be defined as the rapid onset of symptoms and signs secondary to any abnormalities in cardiac function that may be life threatening and require urgent treatment. AHF can present de novo in a patient with no known cardiac dysfunction, but more commonly it presents as an acute worsening of chronic heart failure, sometimes referred to as AdHF. AHF has emerged as a major public health problem leading to an increase in hospitalizations (59,60). Early readmission for heart failure is common (61,62) and, based on 1996/97 Canadian data (63), once a hospitalized patient is discharged with heart failure, the readmission rate for heart failure is 16% in one month and 53% in one year. This underscores potential gaps in care and a need to develop and implement consensus guidelines for the management of AHF. Current heart failure guidelines have focused exclusively on chronic heart failure (64,65), at least in part because there are very few large randomized controlled trials in AHF. At present, there is only one national or international consensus guideline for the management of AHF (66).
An accurate and rapid diagnosis of AHF is important to the timely institution of appropriate therapy and to improve clinical outcomes (67–69), and is based on a careful evaluation of symptoms and clinical findings, supported by appropriate investigations such as electrocardigraphy, chest radiography and, if available, echocardiography and biomarkers. It is important to classify patients based on the presence or absence of congestion and signs of low cardiac output and impaired tissue perfusion to administer the appropriate therapy (70,71). Detailed information on demographics and common clinical presentations of AdHF have been reported from the Acute Decompensated Heart Failure National Registry (ADHERE) (72). Clinical parameters that are predictive of high risk include impaired renal function, low SBP, high respiratory rate, low serum sodium and the presence of comorbid conditions (73,74). Several trials have now clearly established the utility of BNP and NT-proBNP as diagnostic and prognostic biomarkers in patients with AHF (75–77). These assays are most useful in patients in whom the diagnosis is not clinically obvious. They should not, however, replace a careful clinical evaluation. A BNP concentration less than 100 pg/mL or an NT-proBNP concentration less than 300 pg/mL indicates low probability for AdHF. Conversely, a BNP concentration greater than 500 pg/mL or an NT-proBNP concentration greater than 1800 pg/mL indicates a very high probability of AHF.
Among the allied heath care professionals who are involved with the management of patients with AHF, nurses probably play the most important role. In addition to assessment of patients, nursing actions that are crucial to AHF patients’ outcomes are administration of medications, evaluation of treatment effectiveness, and education and ongoing communications with patients, patients’ families and the health care team (78).
A proposed treatment algorithm is illustrated in Figure 4.
Diuretics provide symptomatic relief and should be the first-line treatment for patients with pulmonary or systemic congestion, but there is no trial evidence that diuretics improve outcomes after the acute presentation. Diuretics may cause neurohormonal activation and aggravate systemic vasoconstriction. Consequently, hemodynamic improvements with diuretics may be attenuated and relief of symptoms may be incomplete (diuretic resistance), increasing the subsequent risk of rehospitalization. Although diuretic resistance can be overcome by continuous IV infusion, or combining a loop and thiazide diuretic (79), patients with AHF may derive incremental benefit from the addition of IV vasoactive therapy (vasodilator or inotrope) (80).
Vasodilators can rapidly reduce ventricular filling pressures and myocardial oxygen consumption. They can also decrease systemic vascular resistance, decrease ventricular workload, increase stroke volume and improve cardiac output (81). Nitroglycerin is a vasodilator commonly used to relieve pulmonary congestion in patients with AHF. While it is an effective vasodilator, frequent dose titration of IV nitroglycerin is often necessary to produce the desired hemodynamic effects and symptom relief. Doses greater than 140 μg/min to 160 μg/min may be necessary to sufficiently decrease filling pressures and alleviate symptoms (82). Because IV nitroglycerin requires frequent dose titration and may cause dose-dependent hypotension, patients with AHF treated with IV nitroglycerin should be monitored in an intensive care unit and may require invasive hemodynamic monitoring while being treated. Nesiritide, a peptide identical to human BNP, has been studied in clinical trials but is not available in Canada. Emerging data suggest that early initiation of IV vasoactive therapy reduces the subsequent length of hospital stay (67,68).
Historically, positive inotropes have been the mainstay for adjuvant therapy for AHF because they improve cardiac output. However, large-scale clinical trials evaluating dobutamine and milrinone for AHF are lacking. The use of dobutamine is supported by small studies documenting improved hemodynamics in AHF patients (83). However, evidence from outcome-driven trials (84) indicates a lack of efficacy in many AHF patients and has revealed safety concerns. Given the lack of compelling evidence supporting the use of positive inotropes and the increased incidence of adverse effects, positive inotropic support for patients with AHF should be reserved for patients with signs of low or very low cardiac output.
Patients with impending respiratory failure from pulmonary edema require rapid initiation of supported ventilation. Endotracheal intubation is often required. However, recent data (85,86) suggest that judicious use of noninvasive ventilation, including continuous positive airway pressure and bilevel positive airway pressure, may obviate the need for intubations in up to 75% of cases.
The CCS and the Heart Rhythm Society recently published a position paper on ICD use (88). Given that no new, relevant randomized controlled trials have been published since then, the recommendations for ICD implantation in heart failure patients in the present document are similar.
Three large randomized studies (89–91) (and a subsequent meta-analysis ) have compared the use of an ICD with antiarrhythmic drug therapy (primarily amiodarone) in patients with a history of life-threatening ventricular arrhythmias. Most of the patients in these three trials had left ventricular dysfunction, and many had symptomatic heart failure. Although heart failure per se was not a specific inclusion criterion in any of the trials, the majority of patients had coronary artery disease with prior myocardial infarction or noncoronary congestive cardiomyopathy, with mean ejection fractions in the range of 30% to 35%. As a primary end point, all-cause mortality was reduced in all studies in the defibrillator-treated patients compared with in the antiarrhythmic drug-treated patients (significantly lower in the Antiarrhythmics Versus Implantable Defibrillators [AVID] study  and in the meta-analysis ); in the secondary analyses of the studies and the meta-analysis, patients with lower ejection fractions (less than 35%), higher NYHA class (classes III or IV) and older age had a higher risk of death and received greater relative and absolute benefits from ICD therapy than did patients without these risk factors. ICDs are the therapy of choice for the prevention of sudden death and all-cause mortality in patients with a history of sustained ventricular tachycardia or ventricular fibrillation, cardiac arrest or unexplained syncope in the presence of left ventricular dysfunction. Patients with symptomatic heart failure, especially with ejection fractions less than 35%, are at particularly high risk of death and stand to receive at least as much benefit as patients not meeting these clinical criteria.
All of the ‘primary prevention’ multicentre trials, which assessed the usefulness of implanted defibrillators to reduce all-cause mortality, selected patients with low LVEF; the most common LVEF cut-off was 35%, although a large study, the Multicenter Automatic Defibrillator Implantation Trial II (MADIT II ), had a cut-off of 30%. Most studies did not specifically select patients with symptomatic CHF, although the largest study, the Sudden Cardiac Death in Heart Failure Trial (SCD-HeFT ), did select patients with current symptoms, NYHA class II or III, and a history of heart failure for more than three months.
When considering the risk of sudden death and potential benefit from an ICD, the contribution of systolic dysfunction per se versus heart failure symptoms has not been fully defined. Secondary analyses of most studies have indicated that the absolute risk of sudden death, as well as the relative and absolute mortality benefits of an ICD, was greater for patients with lower LVEFs (eg, MADIT II suggested that the majority of the benefit was found in the patients with LVEFs below the median of 27%). In the Defibrillators in Non-Ischemic Cardiomyopathy Treatment Evaluation (DEFINITE) study (95), which comprised patients with nonischemic cardiomyopathy, mortality was significantly reduced in patients in NYHA class III with an LVEF less than 30% compared with patients with an LVEF of 30% to 35%. In SCD-HeFT, all-cause mortality was greater in the subgroup with LVEFs less than 30%, and the relative benefit from the ICD was greater (albeit not statistically significantly) than in those patients with LVEFs between 30% and 35%.
It is important to note that in one randomized clinical trial, the Defibrillator in Acute Myocardial Infarction Trial (DINAMIT ), which specifically selected patients soon (less than 40 days) after a myocardial infarction, with an average LVEF of 28% and 52% of the patients having had symptomatic heart failure, there was no significant benefit from the ICD compared with control therapy. Therefore, ICDs are not recommended within the first month after myocardial infarction, and the data suggest that some time needs to elapse after a myocardial infarction before patients are sufficiently stable to derive benefit from prophylactic ICDs. There are fewer data for patients with nonischemic cardiomyopathy than for patients with coronary artery disease, and the absolute mortality for patients with nonischemic cardiomyopathy, at any given LVEF, is less than for patients with coronary artery disease and ischemic cardiomyopathy.
The contribution of heart failure symptoms (as distinct from LVEF) to the absolute and relative benefit of an ICD is less clear. In MADIT II, in which patients with NYHA class I, II or III could be enrolled, patients with greater symptoms of heart failure appeared to derive relatively greater benefit from an ICD. On the other hand, patients in SCD-HeFT with class III heart failure appeared to derive less benefit (smaller relative risk reduction) than those with NYHA class II symptoms. It is plausible to assume that heart failure patients with class III symptoms have higher all-cause mortality than patients with class II symptoms but that they may receive less absolute benefit from an ICD if nonarrhythmic death rates are increased in class III patients or, conversely, that certain patient subsets may receive more absolute benefit if the relative benefit is similar but absolute death rates are increased. Given the uncertainties in the secondary analyses from the clinical trials, there is no clear evidence that NYHA functional class within I to III should be used as a selection criterion for the implantation of an ICD.
Despite patient education, lifestyle modification and improved pharmacological therapy available for heart failure, many patients have persistent severe symptoms. Commonly, these patients have intra- and interventricular conduction delays that are associated with cardiac mechanical dyssynchrony. This compromises ventricular function and is frequently associated with severe symptoms and poor prognosis. CRT uses biventricular pacing to attempt to synchronize the activation of the septum and left ventricular free wall, and to improve the overall left ventricular function. The left ventricular free wall can be paced percutaneously through the coronary sinus in the majority of patients. Failing that, a minithoracotomy can be performed, with the placement of an epicardial lead on the left ventricle.
Since the last CCS heart failure consensus conference, two major studies (97,98) and many smaller studies have been published on CRT in heart failure patients. The large-scale Comparison of Medical Therapy, Pacing, and Defibrillation in Heart Failure (COMPANION) study (97) assessed the role of CRT, with or without ICD, in patients with NYHA III to IV symptoms on optimal medical therapy, a QRS duration greater than or equal to 120 ms, and an LVEF of 35% or less. Patients were randomly assigned to medical therapy alone or to medical therapy in combination with CRT or CRT/ICD therapy. At 12 months, compared with medical therapy alone, CRT significantly reduced the rate of death or hospitalization from any cause. CRT therapy alone nonsignificantly reduced the risk of death from any cause by 24% compared with medical therapy alone (P=0.059). Of interest, in the group treated with CRT/ICD, the risk of death was reduced significantly by 36% (P=0.003). Finally, the NYHA class, six-minute walked distance and quality of life scores were significantly better in the CRT group than in the medical therapy-only group.
The second large-scale study, Cardiac Resynchronisation in Heart Failure (CARE-HF ), assessed patients with a history of heart failure of at least six weeks who were in NYHA class III or IV despite optimal medical therapy, and who had an LVEF of 35% or less and a QRS interval of at least 120 ms on an electrocardiogram. Patients with a QRS interval of 120 ms to 149 ms were required to meet two of three additional criteria for dyssynchrony: an aortic pre-ejection delay of more than 140 ms, an interventricular mechanical delay of more than 40 ms or delayed activation of the posterolateral left ventricular wall. Patients were randomly assigned to CRT plus optimal medical therapy or optimal medical therapy alone. The CRT group, compared with the medical therapy-only group, had significantly fewer deaths from any cause and fewer unplanned hospitalizations for a major cardiovascular event. The CRT group also had significantly fewer deaths from any cause than the medical therapy group. As well, the CRT group had better improvement in ejection fraction, overall symptoms and quality of life scores than the medical therapy-only group.
In the past three years, two CRT meta-analysis have been published (99,100). The first meta-analysis (99) showed that CRT reduced the number of deaths from progressive CHF by 51% and heart failure hospitalizations by 29%. In this meta-analysis, CRT was not associated with a significant reduction in all-cause mortality. The second meta-analysis (100) combined data from nine clinical trials to look at the efficacy of CRT. It showed that CRT reduced heart failure hospitalizations but its benefit was seen mainly in patients with NYHA III to IV symptoms. This meta-analysis did not examine the effect of CRT alone in improving survival. With respect to safety issues, the second meta-analysis pooled data from 18 trials to show that CRT was associated with a 0.4% death rate (associated with implantation) and had a 90% implantation success rate.
Unanswered questions remain about exactly who benefits from CRT therapy. Why do all severely symptomatic patients with a wide QRS not benefit from this form of therapy? Does the QRS duration itself matter as much as the finding of cardiac dyssynchrony on echocardiography? What is the best way to evaluate cardiac dyssynchrony? What is the role of CRT in patients with chronic atrial fibrillation? What about patients with right bundle branch block? Should CRT be used in less symptomatic patients to prevent the progression of symptoms? Is there a better way to optimize CRT function by using certain echocardiographic parameters? Several smaller studies have attempted to answer these questions, but the results are not conclusive.
Heart failure remains a disease primarily addressed with medical therapy, and surgical therapy has traditionally been limited to a small minority of patients. Cardiac transplantation remains the preferred treatment for the fortunate few who are eligible and receive a suitable donor organ. Unfortunately for the vast majority of patients, orthotopic heart transplantation is not an option, and they must rely on alternative forms of medical and/or surgical therapy for their debilitating disease.
Because ischemia can depress myocardial function and may progress to further myocardial damage, heart failure patients with coronary artery disease should have all atherosclerotic risk factors aggressively treated; should be investigated for evidence of ischemia or viability; and should be evaluated for revascularization in the presence of persistent angina or documented large areas of ischemia or viability when appropriate. Percutaneous revascularization can generally be performed safely on most heart failure patients with a suitable coronary anatomy. To date, no trial has prospectively evaluated surgical revascularization as a therapy for heart failure. The ‘deemed’ indications for revascularization in patients with triple-vessel disease and impaired left ventricular function (derived from the Coronary Artery Surgery Study [CASS]  results) do not apply in patients with heart failure symptoms. The Surgical Therapy for Ischemic Congestive Heart Failure (STICH) trial is a large, National Institutes of Health-sponsored, multinational trial evaluating surgical therapy, including revascularization, for ischemic cardiomyopathy (102). Similarly, no prospective study has ever shown that left ventricular reconstructive surgery for anterior hypokinesia or akinesia results in either prognostic or symptomatic benefit (103,104). The second primary objective of STICH is to define the role of left ventricular reconstruction in patients undergoing coronary revascularization. The trial has three arms, which are medical therapy, conventional revascularization and revascularization with left ventricular remodelling. Highly symptomatic patients who require surgical revascularization are still candidates for this trial if they meet the minimum criteria for remodelling surgery. In addition, the predictive role of ‘viability’ assessments will be evaluated in long-term follow-up. The trial has not yet completed enrolment or follow-up.
Recent interest has focused on mitral valve repair for both ischemic and dilated forms of cardiomyopathy. Bolling et al (105), Wu et al (106) and Badhwar et al (107) have shown that the mitral valve can be repaired with an undersized annuloplasty ring in selected patients with reasonable morbidity and mortality. However, two-year survival following surgery is estimated at 70%, and a recent retrospective analysis determined no survival benefit of mitral valve repair in patients with systolic left ventricular dysfunction. Again, no prospective data are available to determine the relative benefits of mitral valve repair in this patient population.
The advent of mechanical circulatory support in the past few years has again dramatically altered the landscape for heart transplantation (108–110). To date, nine active programs in Canada provide mechanical circulatory support, with three centres providing destination therapy. Advances in device technology will predictably allow for a smaller, more efficient and reliable system. It is foreseeable that in the future, patients listed for heart transplantation will have the option of receiving a biological or a mechanical heart, similar to the choice offered to them today with respect to heart valves. Furthermore, patients who are considered to be unsuitable for cardiac transplantation due to fixed pulmonary hypertension may be ideal candidates for destination therapy and long-term mechanical circulatory support. The Randomized Evaluation of Mechanical Assistance for the Treatment of Congestive Heart Failure (REMATCH) trial (110) showed that mechanical support provided significantly better short-term survival and quality of life over optimal medical therapy. Almost 1500 patients in Canada are denied cardiac transplantation annually due to our stringent screening process and may potentially benefit from mechanical circulatory support. Other centrifugal and coaxial pumps are under investigation.
Novel surgical therapies include left ventricular remodelling, mechanical circulatory assistance and, more recently, isolated cell transplantation or gene therapy (111-116). The advent of cell transplantation provides great promise for the future because it may be a useful adjunct to several of the previously mentioned therapies. Furthermore, some of the early failures seen with left ventricular remodelling may be prevented with adjuvant cell therapy, and in the case of mechanical circulatory assistance, adjuvant cell transplantation may increase the proportion of patients who are successfully bridged to recovery. However, the era of cell transplantation is only dawning. Several key issues remain unanswered, including the potential side effects of cell transplant therapy. Ongoing prospective clinical trials of angiogenesis by either gene or cell therapy should provide important new data in the next three years.
Although most clinical trials of therapy have studied patients with an average age in the mid-60s, heart failure is common among the elderly, who bear a greater burden of comorbidity and polypharmacy (117), as well as psychogeriatric comorbidities, caregiver burden, health service use, functional decline and frailty (118–120). Frailty characterizes elderly persons with a progressively eroding ability to independently perform activities of daily living, such as bathing, toileting, dressing, grooming and feeding (121). These associations have ramifications on the diagnosis and prognosis of heart failure (122,123) (Table 5). A comprehensive understanding by clinicians of the interaction between heart failure and age-associated syndromes is essential to deal effectively with this growing epidemic.
Heart failure is associated with cognitive impairment in the domains of attention, short-term memory and executive functions (insight, judgement, problem solving and decision-making), and has been associated with nonadherence to treatment, accelerated functional decline and mortality (124–127). Acute and fluctuating cognitive impairment, or delirium, can be precipitated by decompensated heart failure (128). Generally under-recognized by health care providers, delirium is usually reversible, although it may persist well beyond hospital discharge (129). The Confusion Assessment Method is an effective screening instrument for delirium (sensitivity and specificity of 94% to 100% and 90% to 95%, respectively) (130). Chronic cognitive impairment can occur in patients with stable heart failure and is known as dementia if it impinges on independent function (such as adherence to prescribed therapy). A number of screening instruments for dementia exist; although none has clearly been shown to be superior to the widely used and studied Mini-Mental State Examination, the Mini-Cog is briefer (131,132). Cognitive impairment reported by a caregiver should not be overlooked (131).
Symptoms of depression are common in heart failure patients (133). Depression reduces quality of life; increases the risk of functional impairment, rehospitalization and mortality; and may reduce adherence to prescribed therapy (134). Depression and heart failure share common clinical features in elderly patients, including weight gain, sleep disturbances, fatigue, poor energy and cognitive disturbances (134). A number of instruments exist to screen for depression in the elderly, and while none is clearly superior, the Geriatric Depression Scale may be advantageous in identifying patients with mild depression (135). Short versions of the Geriatric Depression Scale have been validated and have acceptable psychometric properties (136).
A few randomized trials of therapies conducted specifically in elderly populations, in conjunction with a multitude of data from observational data sets, suggest that most recommendations on heart failure therapies are applicable to elderly patients. Observational data suggest that ACE inhibitor use in elderly heart failure patients may preserve cognition, slow functional decline, and reduce hospitalizations and perhaps even mortality, even in patients with relative contraindications, such as mild to moderate renal impairment (137–139). The beta-blocker nebivolol has been studied in 2128 patients 70 years of age or older with clinical evidence of heart failure regardless of ejection fraction (140). After a follow-up of less than two years, a significant benefit for nebivolol was seen with reduction of the combined primary end point of mortality and cardiovascular hospitalization. The Japanese Diastolic Heart Failure Study (141) will further evaluate the effects of the beta-blocker carvedilol in 800 elderly Japanese patients with heart failure and a documented ejection fraction greater than 40%.
Elderly patients are vulnerable to adverse drug events (ADEs) due to the growing complexity of medication regimens, age-related physiological changes and a higher burden of comorbid illnesses (142). Cardiovascular medications are frequently associated with ADEs in the elderly (143). Digoxin toxicity can occur at therapeutic serum concentrations (144). Falls are common presentations of ADEs in the elderly, often from postural hypotension. In randomized trials of medications for heart failure, titration to target doses is less frequently successful in older patients due to higher side effect rates. As such, care must be taken with titration of medications to target doses to avoid ADEs. In particular, orthostatic hypotension is a frequent side effect in elderly patients, but if recognized, it can be managed to allow for use of evidence-based therapies (Table 6).
Cardiovascular medications in general, and heart failure medications in specific, are underprescribed to older patients, despite the observation that, as a result of a higher baseline incidence of cardiovascular events, the absolute benefit of evidenced-based therapies may be greater in the elderly population (145,146).
Systematic reviews (147) support the role of heart failure management programs in elderly heart failure populations. While active involvement of caregivers in patient monitoring and medication adjustment is common to studies showing benefit, the optimal way of providing heart failure management remains an ongoing subject of debate. The precise design of such care delivery systems depends, in part, on local resources and infrastructure. Comprehensive geriatric assessment, shown to improve function, prevent hospitalization and institutionalization, reduce the risk of adverse drug reactions and improve suboptimal prescribing, may have a role in the management of frail elderly patients with heart failure (148–150).
The occurrence of diabetes mellitus and renal insufficiency in older heart failure patients carries a significantly worse prognosis and a greater likelihood of ADEs. The potential for contradictory recommendations may arise when these comorbidities are managed in separate settings. Conflicting advice from multiple care providers can result in patient confusion, nonadherence and adverse outcomes. Recommendations to limit diuretic use to maintain renal function or dietary advice to control blood glucose that results in increased sodium intake may lead to worsening heart failure symptoms. An integrative approach to care is required, based on shared therapeutic goals and involving all care providers, including the primary care physician and the patient.
Death from heart failure is due to sudden cardiac death, brady- or tachyarrhythmias, or progressive heart failure. Although there are multiple prognostic markers in heart failure, including ejection fraction (151,152), predicting time of death is notoriously challenging, especially given the cyclical nature of the disease. Recent technological advances have led to increased complexity of care and decision-making at the end of life. As a result, advanced care planning for patients with CHF must be addressed earlier in the course of the disease, allowing patients the opportunity to review the issues surrounding death from heart failure before the development of an acute exacerbation.
NYHA class II patients have a better prognosis; however, they are at a proportionally higher risk of sudden cardiac death (‘drop’). Patients with class IV symptoms have a one-year mortality as high as 75% and a significantly higher risk of dying of progressive heart failure characterized by increasing shortness of breath, orthopnea, and decreasing blood pressure and level of consciousness (‘drown’) (153–155). Patients in NYHA class II to III who receive an ICD may progress to class IV symptoms, with the likely mode of death changing from drop to drown.
Patient preferences suggest that critically ill patients living with heart failure want treatment at the end of life (156); however, these preferences are not stable over time. This may reflect the cyclical nature of CHF, with decision-making being dependent on symptomatic status. In fact, patients with CHF assign a higher degree of importance to symptom management than to survival and are more likely to decline treatment when the likelihood of an adverse event increases (157,158).
Physician preferences for treatment also influence decision-making at end of life. In the Study to Understand Prognoses and Preferences for Outcomes and Risks of Treatments (SUPPORT) (156), 24% of physicians did not correctly perceive their patient’s resuscitation preference. Physician preference was based on the physician’s resuscitation preference if he or she were in the patient’s condition. Although these patients were ill and many physicians expected them to die within two months, only 25% of patients reported that they had discussed preferences for resuscitation with their physicians. Open and honest dialogue with patients and their families regarding treatment at end of life is necessary to guarantee that care reflects patient preferences and avoids unnecessary conflict among the patient, family and clinicians during a very difficult time. Suggested domains of care that should be considered at the end of life are given in Table 7 (159).
It is critical with the current technologically advanced therapy that programs proactively include a comprehensive process for potential device withdrawal that includes detailed informed consent and advance care planning. As increasing numbers of ICDs are implanted, there will be a greater number of patients with terminal CHF with active ICDs. Turning off the defibrillator function of an ICD in patients with advanced heart failure may change the mode of death from progressive heart failure to sudden death, often a more preferable and less symptomatic form of death (160,161).
Advanced directives, or living wills, enable competent persons to maintain control over their medical care should they lose their decision-making capacity (162). Advanced directives determine what decisions are to be made and who makes the decisions (162,163). Incorporating this early in the care of the heart failure patient allows patient wishes to be upheld throughout the disease course. Advanced directives require communication among patients, families and health care providers (163). Given the episodic nature of heart failure, issues regarding therapy should be revisited throughout the course of the disease. It is critical that physicians help capable patients clarify and regularly update their views about these issues with both complicated (eg, ventilation) and simple (eg, IV fluids) therapies (162).
Quality end-of-life care has three crucial elements: support of dying patients and their families; control of pain and other symptoms; and decisions on the use of life-sustaining therapies (164). End-of-life care incorporates features of truth-telling, consent, capacity, substitute decision-making, advance care planning and appropriate use of life-sustaining treatment (164). Symptom control is especially important for patients with progressive heart failure because many patients will feel increasing dyspnea and a sensation of drowning at the end of life. Although heart failure physicians are comfortable with the use of diuretics for symptom control, many patients will require short- or long-acting narcotics for dyspnea.
Caregivers for people with chronic illnesses, such as heart failure, experience an increased morbidity and mortality. Caregiver burden is an independent risk factor for mortality, as well as emotional distress and loneliness (165). Although it is important for caregivers to be involved in all aspects of care, as the number of care tasks increase, as well as the perceived degree of difficulty in performing these tasks, so does the degree of depression among caregivers (166). Younger caregivers experienced a greater degree of distress than did their older counterparts (167). Female spouses providing more than 9 h of caregiving activities per week had a twofold increased risk of developing coronary artery disease (168).
The provision of optimal care to patients with heart failure presents many challenges to the patient, their family or care-givers, the physician, other health care providers and the health care system. An accurate and timely diagnosis is critical to initiate treatment that will relieve symptoms, improve quality of life, reduce hospitalizations and prolong survival. The past 20 years have seen dramatic changes in our understanding of heart failure and the introduction of many new treatment modalities. These consensus recommendations should provide an evidence-based road map to translate knowledge into practice and allow health care practitioners to make the best clinical judgments and decisions for their individual patient. Practical tools to improve implementation are being developed by a Clinical Practice and Health Outcomes Impact Working Group of the CCS, which is also identifying potential organizational barriers to implementation and specific measurable outcome audit criteria. Because new evidence will continue to be published, these recommendations will be updated in 12 months. Our goal is that this will improve the delivery of best care and practices to heart failure patients in Canada.
This consensus conference was supported by the Canadian Cardiovascular Society. The authors are indebted to John H Parker and Lise Hodgson of the Canadian Cardiovascular Society, Gordon Marchiori PhD and Kim Harrison for logistic and administrative support
SECONDARY PANELISTS: Tom Ashton MD FRCPC, Penticton, British Columbia; Victor Huckell MD FRCPC, University of British Columbia, Vancouver, British Columbia; Debra Isaac MD FRCPC, University of Calgary, Calgary, Alberta; Marie-Helene Leblanc MD FRCPC, Hopital Laval, Sainte-Foy, Quebec; Gary E Newton MD FRCPC, Mount Sinai Hospital, Toronto, Ontario; Joel Niznick MD FRCPC, The Ottawa Hospital, General Campus, Ottawa, Ontario; Sherryn N Roth MD FRCPC, Scarborough General Hospital, Toronto, Ontario; Denis Roy MD FRCPC, Institut de Cardiologie de Montreal, Montreal, Quebec; Stuart Smith MD FRCPC, St Mary’s Hospital, Kitchener, Ontario; Bruce A Sussex MD FRCPC, Health Sciences Centre, St John’s Newfoundland; Salim Yusuf MD FRCPC, McMaster University, Hamilton, Ontario.
The following primary panel members also represented their respective societies:
Ross Tsuyuki, Canadian Pharmacists Association; Anna Svendsen, Canadian Nurses Association; George Heckman, Canadian Geriatrics Society; Errol J Sequeira, College of Family Physicians of Canada.
CONFLICT OF INTEREST: The panelists had complete editorial independence in the development and writing of this manuscript, and functioned on a pro bono basis. A full description of the planning of this consensus conference and the ongoing process (including the needs assessment, the methods of searching for and selecting the evidence for review, and the conflict of interest statements of panel members) is available at <www.ccs.ca>.