An 82-year-old woman who had not been to a physician in over 30 years was admitted to the hospital with fatigue and dyspnea. At home, she had been bedbound for the past two months and only got up to use the bathroom, which caused her dyspnea and panic. During this time, she had also lost 20 to 30 pounds of weight. On physical examination, she had rales half-way up both lung fields and a prominent systolic ejection murmur. An echocardiogram revealed severe systolic dysfunction.
She decided, with her son, that she did not want further intervention or investigation of either her cardiac status or of her weight loss. Upon hearing this, her physician called a palliative care consultation: “I’m not sure why, but I think it will help—she’s not really in pain or anything” he said when the palliative care physician talked with him. At the end of the conversation, he added: “And I don’t want her to give up.”
This case is not an unusual one and it highlights the uncertain role of palliative care for patients with heart failure. Does palliative care have something to offer in this case? Is she giving up? Is her doctor giving up? What are the appropriate treatments for her dyspnea? What other symptoms might she be expected to have? This chapter addresses these questions.
Despite the fact that heart disease is the leading cause of death in the United States, it accounts for a minority of patients in palliative care and hospice programs. According to the latest figures available, cardiovascular disease was the leading cause of death in 2006: it accounted for over 631,00 deaths in the United States, or more than 25.6% of all deaths. Two other diseases which are closely linked to heart disease, cerebrovascular disease and diabetes contribute to another 137,119 and 72,449 deaths annually. In the same year, cancer accounted for over 559,000 deaths or 23.0% of all deaths. By contrast, in 2008 while 11.7% of patients enrolled in hospice had a primary diagnosis of heart disease, 38.3% had a diagnosis of cancer. Compared to the burden in the general US population, cancer is over-represented among patients receiving hospice care and heart disease is under-represented. Given this incongruence, it is likely that many patients who might benefit from palliative care are not receiving it. Although it is likely that there are patients with heart disease who receive palliative care outside of formally being enrolled in hospice, these numbers serve as some objective data concerning the volume of patients formally enrolled in one type of palliative care program.
This chapter discusses the epidemiology of heart failure, the prevalence and treatment of common symptoms in patients with heart failure and ends with a consideration of some of this issues that are particularly challenging for patients with heart failure considering palliative care.
How is Heart Failure Different from Cancer?
Both heart failure and cancer are increasingly viewed as chronic conditions. This view implies that there is usually no specific point at which the illness switches from “treatable” to “terminal” but rather a transition exists and its nature depends on features that include illness severity and patient and family goals of care. The ideas denoted by the terms palliative care, survivorship and rehabilitation are all part of a recognition that optimizing quality of life and functional status can and should go hand-in-hand with disease-directed treatments. Both the Centers for Disease Control (CDC) and the American Cancer Society, for example, have devoted significant resources to educating the public around issues of long-term survival in cancer. Multiple stages exist for both heart failure and cancer during which a combination of disease-modifying and symptom-directed or palliative therapies are appropriate. Data from Lunney and colleagues has lent empirical evidence to the chronic and stuttering trajectory of symptoms and care that heart failure patients experience in the last year of life. It is a trajectory that frequently involves multiple exacerbations of symptoms, multiple changes in medications and management strategies and multiple emergency department visits and hospitalizations.
This view of heart failure as a chronic illness may make the transition to palliative care more challenging: Cardiologists and their patients may perceive that continuing advances in drug therapy and cardiac devices will prolong length of life and quality of life, and thus may be less inclined to consider the possibility of treatment failure than oncologists and their patients. As an example, Schoevaerdts and colleagues identified a number of differences between cancer and heart failure that may influence decisions about palliative care: The illness course is often perceived of as fluctuating as opposed to progressive; prognosis is perceived as unpredictable as opposed to predictable; the “terminal phase” of the illness is less clearly delineated; and sudden death is more frequent in the case of heart failure compared to cancer.
In this area of disease perception, there are data to suggest that physicians may view palliative care and heart failure as a challenging combination. In a qualitative study of British physicians including cardiologists, general practitioners and palliative medicine specialists, Hanratty and colleagues found that barriers to developing approaches that integrated palliative care and heart failure fell into three main groups: the organization of health care, the unpredictable course of heart failure, and physicians’ understanding of the roles of different specialists and disciplines caring for patients. It is likely that each of these may be a barrier for the optimum delivery of palliative care to patients with heart failure. The authors interpreted their findings to mean that these barriers can lead to late referrals and a lack of complete palliative care for patients with heart failure at multiple stages of their illness.
These differences in disease perception between cancer and heart failure have implications for how physicians care for patients with heart failure. A study by McKinley and colleagues compared care delivered by general practitioners in the last year of life to people who died with cancer and those who died with cardiovascular and respiratory disease. They found that when compared with people who died with cardiovascular and respiratory disease, those who died with cancers were more likely to have had a “terminal phase” of their illness identified by their health care providers and to have been prescribed more palliative (symptom control) medications. At the same time, there were some encouraging similarities in the care of both groups of patients with cancer and with heart disease: both groups had a similar number of specialist consultations, experienced similar continuity of care, had similar levels of comorbid conditions, and were equally likely to die at home.
Data also exist that suggest that patients view the experience of living with (and dying of) heart failure and cancer differently. Murray and colleagues conducted longitudinal in-depth interviews with both groups of patients and found that patients with heart failure generally had less information about and poorer understanding of their condition and prognosis and were less involved in decision making than those with heart failure. While those with cancer more explicitly talked about dying, they found that “Frustration, progressive losses, social isolation, and the stress of balancing and monitoring a complex medication regimen dominated the lives of patients with cardiac failure”.
Taken as a group, these studies suggest that both physician and patient perception may present challenges in introducing palliative care into the setting of heart failure.
Epidemiology and Mortality
There are approximately 5.8 million Americans living with heart failure and there are 670,000 new (or incident) cases per year. The number of patients discharged from hospitals with a diagnosis of heart failure was more than 1.1 million in 2006. From an economic standpoint, estimates for the direct and indirect costs of heart failure range from $39.2–58 billion a year.
In terms of mortality, newly diagnosed heart failure in the community carries a mortality of 24%, 37%, and 75% at 1, 2, and 6 years. For patients with advanced (New York Heart Association [NYHA] class IV) heart failure, the rate of rehospitalization or death is 81% at 1 year. In patients with a left ventricular ejection fraction of less than 25%, NYHA class IV symptoms (for example, dyspnea and angina at rest) for more than 90 days, oxygen consumption of 12 ml/kg or inotropic dependence, mortality is 50% at 6 months. For patients on continuous inotropic support, mortality is 75% at 6 months. Although these numbers parallel the mortality from a number of advanced stage cancers, another key difference for patients with heart failure is the frequency of sudden death: unlike cancer, where deaths may be perceived as following a predictable course, up to one-half of all deaths are sudden.
Guidelines
In the face of the high disease burden that heart failure confers on the population, the consideration of palliative care for these patients has taken on some urgency. In general, the integration of palliative care principles into disease-specific guidelines has been an important step in increasing the presence of palliative care in mainstream medicine for both cancer and non-cancer illnesses. The American Hospice Foundation, for example, has developed such guidelines for cancer and non-cancer illnesses and the National Comprehensive Cancer Network (NCCN) has done so for a range of cancer diagnoses. More recently, in cardiology and pulmonary medicine, guidelines issued by both the American College of Chest Physicians and the American College of Cardiology/American Heart Association (ACC/AHA) have addressed palliative strategies within their overall management guidelines. Finally, the Heart Rhythm Society (HRS) in collaboration with the American College of Cardiology, the American Geriatrics Society, the American Academy of Hospice and Palliative Medicine, the American Heart Association, and the European Heart Rhythm Association have developed a consensus statement on the management of ICDs and other implantable devices in patients nearing the end of life. These guidelines lay out clear ethical principles regarding the right of patients and families to withdraw implantable devices and also suggest communication strategies to discuss these decisions.
The newest (2009) ACC/AHA guidelines lay out clear assessment and management strategies for every stage of heart failure. Importantly, these guidelines include a section on “end-of-life considerations in the setting of heart failure.” Among the recommendations for patients with end-stage heart failure are areas such as ongoing education for patients and their families concerning prognosis, functional capacity, the role of palliative care, discussion of implanted cardiac defibrillators (ICD) deactivation, coordination of care, and the use of opioids for symptomatic relief. Importantly, the ACC/AHA guidelines also suggest that the use of inotropes and intravenous diuretics may be appropriate for some patients with end-stage heart failure who are receiving palliative care. The issue of inotropes in patients with heart failure in palliative care is discussed later in this chapter.
These recommendations parallel an overall palliative approach to any illness in their attention to symptoms, comprehensive assessment, whole-person care, goals of care, and coordination of care. In addition to these general principles, the ACC/AHA guidelines advise that “aggressive procedures (including intubation and AICDs) within the final days are not appropriate.” Although this presents a challenge to our recognized inaccuracies of prognostication, there are generally agreed upon criteria that include decreased to minimal responsiveness, severely compromised functional status and minimal urine output that palliative care physicians routinely use to assess the last days of life. Although some of these signs and symptoms have been shown to be valid and reliable predictors of prognosis in the setting of patients in palliative care with cancer, they have not been subjected to empirical scrutiny in the setting of heart failure. The ACC/AHA guidelines take a broad view of the question of determining when the “end of life is nearing” to include multiple subjective and objective factors, including functional status, oxygen dependence, and echocardiographic evidence of heart failure. In the end, they recommend an individualized approach to care: “Ultimately the decisions regarding when the end of life is nearing reflect a complex interaction between objective and subjective information, emotions and patient and family readiness.”
The ACC/AHA guidelines are a comprehensive attempt to gather state-of-the-art evidence for recommendations to clinicians. For many of the recommendations concerning disease-specific treatments, such as the use of angiotensin converting enzyme (ACE) inhibitors and beta blockers, evidence reaches the level of “class A” (data derived from multiple randomized clinical trials or meta-analyses). For all of the recommendations concerning palliative care, however, it is revealing that the level of evidence is “class C” (only consensus, case studies, and standard of care as opposed to data from clinical trials). This suggests that the field may be early in its integration of palliative care. At the same time, it is likely that the types of evidence needed to support some of the palliative interventions may differ from traditional clinical trials for disease-modifying medications.
Another set of guidelines specifically concerning palliative care and heart failure has been issued by a group of cardiologists and palliative care physicians. These guidelines were issued from a consensus conference “convened to define the current state and important gaps in knowledge and needed research on ‘Palliative and Supportive Care in Advanced Heart Failure.’ ” Although they did not formally grade their evidence in the same manner as the ACC/AHA guidelines, they derive their guidelines from expert opinion and review of the medical literature and other guidelines. They identified the need for research on symptom clusters, prognostication, and coordination of care approaches to heart failure as priorities as the fields of palliative care and heart failure become more tightly linked and move forward.
Symptoms
Despite some of the limitations of the empirical evidence behind these guidelines, we do have some data concerning symptoms in patients with heart failure. Two of the largest trials that examined symptoms for patients near the end-of-life include the SUPPORT trial (The Study to Understand Prognoses and Preferences for Outcomes and Risks of Treatments) and the United Kingdom Regional Study of Care of the Dying (RSCD). Each of these studies had a component that assessed symptoms of patients facing the end-of-life and each included patients with heart failure.
In one analysis of the SUPPORT trial that examined elderly patients with heart failure, Levenson and colleagues reported that 60% of patients had dyspnea and 78% had pain. Among those in the hospital, the number in pain was 41%. Notably, these symptoms persisted through the final days of life: patient surrogates reported that 41% of patients were in severe pain and 63% were severely short of breath at some point during the final three days before death. Psychological symptoms, including depression, anxiety, and confusion were all between 10 and 20% in this sample from SUPPORT. In the subsample of patients in the RSCD with heart failure, pain was experienced by 50%, dyspnea by 43%, and low mood and anxiety in 59 and 45% of patients respectively.
Smaller studies of the prevalence of symptoms in patients with heart failure have reported similar results. In a study of patients in one hospital, Jiang and colleagues found that 35% of patients screened positively on the Beck Depression Inventory and 13.9% of patients had a diagnosable major depressive disorder. These numbers have prognostic implications for patients: Jiang’s study found that untreated depression in the setting of heart failure is a risk factor for increased mortality. The high symptom burden on patients with heart failure is not just for those who are hospitalized: A study of patients with heart failure in a community-based hospice found a prevalence of pain in 25%, dyspnea in 60%, and confusion in 48%. If anything, these numbers may underestimate the numbers in a general population of home-bound patients with heart failure, as it is likely that these patients were receiving some symptomatic care through their hospice program.
Treatment
Standard Treatment for Heart Failure
Heart failure is unlike some illnesses in palliative care because a number of its treatments have quality of life benefits as well as disease-modifying effects. This contrasts with many types of cancer in which the disease-modifying treatments may have deleterious effects on quality of life while also having the potential to increase length of life. For example, two of the cornerstones of treatment for heart failure, laid out in the most recent ACC/AHA guidelines and in recent review articles include the use of ACE-inhibitors or angiotensin receptor blockers (ARBs) and beta blockers, both types of medications which have been shown to both improve survival and decrease repeated hospitalizations: i.e., mortality as well as quality of life benefits. Since they have documented effects on quality of life, these medicines should be continued in palliative care. Their use is an important example of how to implement the philosophy that palliative care might include the continuation of disease-modifying therapies as well as comfort-oriented therapies concurrently.
Two other medications for heart failure, diuretics and digoxin, also have effects on quality of life with less clear effects on mortality. Furosemide is the main diuretic for use for patients with fluid overload in the setting of heart failure. Its effects can be potentiated by spironolactone which, when added to a loop diuretic such as furosemide, has been shown to have both mortality and symptomatic benefits and also result in reduced hospitalizations in the setting of heart failure. Although digoxin has been shown to have no effect on mortality, it has been shown in a Cochrane review to have symptomatic benefits in terms of decreased dyspnea and result in fewer repeat hospitalizations. Therefore, these medications should be continued as well.
Intravenous inotropic medications (such as dobutamine or milrinone) may represent a dilemma for many patients, families, and providers. Although these medications have been shown to result in excess mortality in some randomized trials, they have also been shown to decrease hospitalizations and symptoms among patients with end-stage heart failure. Recommendations concerning these medications in advanced heart failure management generally includes their use as a “bridge” to cardiac transplantation and as a palliative measure in patients with end-stage heart failure whose symptoms are refractory to other treatments. In the setting of hospice and palliative care, the use of intravenous inotropes may represent a challenges to philosophy of care (Is this treatment “aggressive” or “palliative”?), education (Do we have the resources among our nurses and/or physicians to administer these safely in the home or nursing home setting?), and finances (In the face of the capitated reimbursement that hospices receive, are we able to afford this medication?). Although there are no clear guidelines at this point in time for the use of these medications in the palliative care or hospice context, there are case studies of patients benefiting from their use. As the number of patients with heart failure who elect to be cared for by hospices increases, more and more individual programs and physicians will face this dilemma.
Symptomatic Treatment
The treatments described above directed at the underlying pathophysiology heart failure will frequently have symptomatic benefits. However, in addition to such “disease-modifying” treatments, additional symptomatic treatment may be needed for every patient with heart failure.
Dyspnea
The key medications for the medical treatment of dyspnea in the setting of heart failure include diuretics and opioids. Wherever possible, treatment directed at the underlying etiology of the dyspnea will help to avoid potential side effects of opioids which can include confusion, somnolence, nausea, and constipation. For example, if dyspnea is due to a pleural effusion, drainage by thoracentesis may be effective temporarily, but these effusions frequently recur. If dyspnea is due to pulmonary edema, furosemide, or another loop diuretic may be effective.
Patients with heart failure may have other causes of dyspnea: when it is due to cardiac ischemia (angina), a nitrate should be used. This can include immediate release sub-lingual nitroglycerin and extended oral or transdermal nitrates when the angina is recurrent and predictable. Patients who get anginal symptoms and dyspnea with a predictable amount of exertion can be counseled to take a nitrate before activity. When dyspnea is due to rapid atrial fibrillation, rate control with digoxin, a beta-blocker, or other agent is appropriate to decrease cardiac workload.
Clinicians also need to be alert to co-morbid non-cardiac conditions in patients with heart failure before ascribing all symptoms to their primary diagnosis. When dyspnea is due to co-existing asthma or chronic obstructive pulmonary disease (COPD), an appropriate symptomatic approach includes inhaled beta-agonists (e.g., albuterol), inhaled steroids (e.g., beclamethasone), or inhaled anti-cholinergic medications (e.g., ipratropium bromide) as well as oral steroids (e.g., prednisone) in the setting of sever illness. In the setting where it is difficult to differentiate between heart failure and COPD as the cause for dyspnea, the measurement of brain natriuretic peptide (BNP) can help to distinguish between the two etiologies and help to direct therapy.
Pneumonia is also common in patients who are debilitated due to heart failure, and its treatment will depend on the patient’s overall prognosis and goals of care. There may, for example, be situations in which patients who appear to have a life expectancy of weeks to months develop a pneumonia, and they and their physicians make a decision to treat with either oral or intravenous antibiotics. There may be other times when a patient with heart failure is bedbound and minimally responsive and whose prognosis is likely days, where a candid and sensitive discussion with the patient and his or her family may result in a decision not to use antibiotics, but to emphasize treatments for the symptom of dyspnea alone.
Finally, sleep apnea has been shown to be associated to heart failure and its treatment may help to improve symptoms of fatigue and dyspnea. The formal diagnosis of sleep apnea requires an overnight sleep study in a sleep laboratory. In patients who are homebound, overnight pulse oximetry in the home can sometimes be an alternative. The standard treatment for sleep apnea is the nighttime use of continuous positive airway pressure (CPAP). More recently, a randomized trial has shown some efficacy using acetazolamide as treatment for sleep apnea. Although the use of CPAP is initially difficult for many patients, it is frequently something that they can adapt to and tolerate after continued use. Medical centers with experienced respiratory care programs will have expertise in the adjustment of specific ventilatory settings as well as suggestions to increase patient comfort and tolerance of CPAP.
Although oxygen is commonly used in hospice and hospital contexts, a review found little data to support its use in relieving dyspnea secondary to heart failure, especially when used as the sole treatment for dyspnea. As an adjunct to other symptomatic treatments as described above, its efficacy is less clear. It may hold symbolic value as an example of continued care and attention for the patient and family and therefore, its use, including the possibility that it may contribute to prolonging life, can be discussed openly with patient and family. Patients and families may feel that it is a requirement when they are in the hospital and not be aware that like any other intervention, its use can be considered in terms of overall goals for care for a given patient. Conversely, patients and families may perceive discontinuing oxygen as a sign that the physicians are “giving up” on a patient. Along with the administration of oxygen, patients in the hospital context as well as those being discharged, may be accustomed to having their pulse oximetry measured at regular intervals and oxygen titrated accordingly. As the patient’s clinical condition and goals for care change and as symptom relief becomes the primary focus for patients, both the administration of oxygen and the monitoring of it should be re-addressed with patients and their families.
Pain
Pain is common in patients with heart failure. Most of this pain is actually non-cardiac in origin and stems from co-existing arthritic or musculoskeletal disease. Anginal pain can be treated with nitrates. Other pain in patients with heart failure can be treated with standard doses of opioids according to World Health Organization (WHO) criteria. One should be cautious in patients with renal insufficiency, which is common in patients with advanced heart failure. First, non-steroidal anti-inflammatory (NSAIDs) medications can worsen renal insufficiency. Second, both morphine and hydromorphone (Dilaudid) have metabolites that can accumulate and lead to adverse neuro-excitatory effects including tremors, myoclonus, and more rarely, seizures. Generally, doses should start low and be increased slowly, especially in elderly patients. When neuro-excitatory side effects do occur, one can reduce the opioid dose, rotate opioids, or add a low dose of a benzodiazepine such as clonazepam. Since patients with heart failure are frequently on multiple medications and at risk for polypharmacy, this last approach should be reserved for cases when opioid dose-reduction or opioid rotation are unsuccessful. When the pain has a neuropathic component, treatment with an adjuvant medication, such as a tricyclic antidepressant (e.g., nortriptyline which has fewer anticholinergic side-effects than amitriptyline) or an anti-convulsant (e.g., gabapentin or pre-gabelin) are both effective.
Nonsteroidal anti-inflammatory drugs should generally be avoided because of the risk of renal dysfunction and increased sodium and fluid retention. In patients with co-existing arthritis whose pain has been amenable to NSAIDs, one needs to balance the risks and the benefits of NSAID therapy. A recent review in this area strongly recommended the use of acetaminophen instead of NSAIDs for these reasons. However, in patients whose prognosis may be weeks and who has unacceptable side effects to opioids or to steroids, the use of NSAIDs should be considered.
Depression
The prevalence of depression ranges from 24 to 42% in patients with heart failure. Predictors of depression in patients with heart failure include living alone, alcohol abuse, perception of medical care being a substantial economic burden, and poor health status. Its presence is associated with increased mortality in patients with heart failure. In addition to being highly prevalent, it is often overlooked in the treatment of heart failure—up to 50% of patients with depression had it go undetected in one study.
An approach to treatment includes counseling interventions and the use of selective serotonin reuptake inhibitors (SSRIs). Tricyclic antidepressants should generally be avoided due to potential cardiotoxic effects. The data concerning counseling and psychological interventions are mixed. A recent review from the Cochrane collaboration concerning depression in the setting of heart failure found that, although there is randomized clinical trial evidence for the effectiveness of counseling in the setting of depression following myocardial infarction, there is no such evidence for patients with heart failure. This review did, however, identify three observational trials suggesting some benefit to counseling. The overall message of these studies is that depression is common, should be screened for, and requires both medications as well as counseling approaches.
We now turn to three topics that of particular importance in heart failure in the palliative care context: resuscitation and automated implantable defibrillators (AICDs), prognosis, and disease management programs.
Related posts:
Full access? Get Clinical Tree
