Fig. 8.1
Pathways leading to heart failure in older adults
Prognostic Models
A first step in providing quality end-of-life care is identification of prognosis for residents in SNF [16]. However, heart failure is frequently associated with an unpredictable trajectory of exacerbations, plateaus or periods of stability, and functional decline [17]. Compared with other terminal illnesses such as cancer where time to end-of-life is predicted in months, predicting end-of-life for patients with heart failure may be measured in years. As a result, heart failure presents prognostic challenges to the SNF health care team. Understanding the variable nature of the course of heart failure is necessary to appreciate the difficulty in applying a palliative care model to care of heart failure patients. Several prognostic models using both single and multiple variables have been studied, but issues with over or under estimation of mortality risk have been reported (Table 8.1) [3, 18, 19].
Table 8.1
Heart failure prognostic models
Model | Prediction | Features | Considerations |
|---|---|---|---|
Seattle Heart Failure Model [20] | 1, 2, 3 year survival | Uses clinical variables, therapeutic interventions, and devices: | May not apply to SNF populations (e.g. patients with dementia) |
1. Age | Developed in patients with predominantly left ventricular systolic HF | ||
2. Sex | Does not include comorbid illnesses | ||
3. Weight | Renal function not included in the model as it was tested in patients without significant renal dysfunction | ||
4. Ejection fraction | |||
5. NYHA class | |||
6. Systolic blood pressure | |||
7. Daily diuretic dose | |||
8. Serum sodium | |||
9. Hemoglobin | |||
10. % lymphocytes | |||
11. Uric acid | |||
12. Total cholesterol | |||
13. Medications (e.g. angiotensin receptor blockers, angiotensin receptor blockers, β-blockers, statins, aldosterone blockers) | |||
14. Ischemic etiology | |||
15. Devices | |||
Graphic impact of adding or removing interventions | |||
Validated with large outpatient populations | |||
Gold Standard Framework [21] | Indicator for nearing end-of-life and need for palliative/hospice care | Uses three components to predict declining HF status: | May overestimate number of patients nearing end-of-life [18] |
1. Decreasing activity tolerance | |||
2. Increasing symptom burden | |||
3. HF specific indicators include: | |||
(a) NYHA III/IV | |||
(b) admissions to inpatient settings | |||
(c) Level of symptom burden |
The Seattle Heart Failure Model includes the greatest number of variables and was validated in large outpatient populations. Levy and colleagues developed and validated the Seattle model to predict 1-, 2-, and 3-year survival in heart failure patients with the use of clinical status, therapy and laboratory parameters [20]. The model allows estimation of the benefit of adding medications or devices to an individual patient’s therapeutic regimen. An online calculator is available at http://www.seattleheartfailuremodel.org. The model may not be generalizable to patients with dementia which is common in the SNF population. Similarly, other major comorbid illnesses such as cancer or renal failure are not accounted for in the model [20].
The Gold Standard Framework was developed to identify people nearing the end-of-life to facilitate appropriate palliative and end-of-life care. The model has three components beginning with the surprise question: “Would you be surprised if the patient were to die in the next few months, weeks or days” [21]. General indicators of decline include decreasing activity tolerance and increasing symptom burden. Heart failure specific indicators are the third component which includes poor functional status, repeated admissions, and high symptom burden despite optimized therapies.
In a comparison of the Seattle and Gold Standard models, neither was found to be accurate in predicting mortality at 12 months [18]. However, renal function measured using serum creatinine was the most sensitive and specific factor for predicting death at 12 months. Lower serum creatinine predicted lower mortality at 1 year (<140 μmol/l).
An accurate prognostic model specifically for use in the SNF population is currently not known. Furthermore, some models incorporate variables that may not be readily available in SNF. As difficulty in prognosticating mortality delays initiation of palliative and hospice care, it may be more prudent to use uncontrollable symptoms as a trigger for end-of-life care [22, 23].
End-of-Life Care Components
To provide quality end-of-life care for HF patients in SNF, programs that address the goals of this care for residents and their care giving families must be developed. Components of high quality end-of-life care programs include:
Focusing on needs of the resident and family
Encouraging and supporting shared decision-making
Documenting advanced directives to support quality of life and a good death
A pain-free and aware death and acknowledgement of the imminence of death [24]
A framework acknowledging cultural implications about illness and dying and providing for seamless transitions of care
Shared Decision-Making
Effective communication and shared decision-making among the health care team, residents and families is integral to achieving quality end-of-life care for residents with life-limiting cardiovascular disease including heart failure. Shared decision-making differs from patient centered care in which communication focuses on the needs and emotions of the patient [27]. In shared decision-making, the resident (and families) and provider(s) collaborate to make treatment decisions. Information sharing moves in both directions with the goal of equal empowerment to make decisions. Participation by the healthcare team (medicine, nursing and other personnel involved in care) may make the process easier for the resident and family [28, 29]. Training of healthcare providers in shared decision-making is important. Lack of skill and confidence in shared decision-making among providers is a barrier to meaningful advance care planning [30, 31]. (See also Chap. 2 about communication and decision-making.)
Shared decision-making preferably continues from admission to SNF care until end-of-life. To enable active participation in decision-making at end-of-life, it is necessary that the resident, family, and healthcare providers recognize and accept that death is impending. For residents and families, knowledge and acceptance that heart failure is a life-limiting illness is not always clear or desired [24, 32]. Although prognostication is challenging, clinical signs such as progressive renal dysfunction, cachexia, escalating diuretic doses, intractable symptoms and declining functional status are potential indicators of impending death [33].
Advance Directives
Advance directives including identification of a healthcare proxy and contingency planning are important considerations at time of admission to a SNF [34]. Do Not Resuscitate (DNR) orders are important but insufficient for outlining resident wishes. Contingency planning specific to heart failure includes preferences related to acute decompensation, hospitalization, implanted cardiac rhythm devices and cardiac arrest [25, 35, 36]. Discussions regarding implantable defibrillators are particularly important as over 20 % of patients receive shocks at end-of-life and shocks are associated with decreased quality of life [35, 37]. In the event a resident does not have a contingency plan or in the absence of readily accessible onsite electrophysiological deactivation equipment, implantable defibrillators can be deactivated by placing a donut magnet over the device. Accordingly, donut magnets should be readily available in SNF. The donut magnet should be left in place until electrophysiological equipment is available to permanently deactivate the device [35].
Although advance directives including DNR documentation are an important consideration for SNF residents, 66 % were reported not to have DNR or Do Not Hospitalize (DNH) directives in a sample of 1,926,742 Medicare beneficiaries in 2001 [38]. The frequency of directives varies widely by state and race [34, 38]. African-Americans are the least likely to have either a DNR (odds ratio 0.35) or DNH (odds ratio 0.26) directive. Factors associated with an increased likelihood of having either a DNR or DNH advanced directive are outlined in Table 8.2.
Table 8.2
Factors associated with increased likelihood of DNR or DNH directives
Age 95 years and older |
Caucasian |
Functional dependence |
Impaired cognition |
Neoplasm |
Heart failure |
Chronic obstructive lung disease |
Dementia |
Arrhythmia |
Coma |
Hypertension |
Quality of Life
As heart failure progresses, the focus shifts away from life-extending therapeutics to a focus on supporting and maintaining quality of life [3]. However, knowledge deficits about palliative care exist among patients with heart failure, their care-giving families, and healthcare providers [39]. It is important to explain to residents and families that medical therapies to manage symptoms of heart failure will continue including therapies for symptoms arising from comorbid illnesses. Early initiation of palliative care supports higher satisfaction among residents and their families at end-of-life [40].
Communication patterns used by heart failure patients with their healthcare providers were explored by Green and colleagues in a narrative systematic review of 18 publications [41]. A majority of patients with heart failure wished to have a discussion of both prognostic indicators related to their illness and their feelings about end-of-life issues. Exchange of information between patients and providers may be impeded if the patient has cognitive deficits or fatigue [42]. Therefore, it is prudent to include designated family members or healthcare proxies in these discussions.
Symptom Management
Symptoms of advanced heart failure can be distressing and negatively affect quality of life for patients and their families. The most common symptoms and sources of distress among patients with advanced heart failure are dyspnea, pain, fatigue, anxiety, and depression [13, 17, 25, 43]. In a study of dying residents in long-term care in the last three months of life (n = 674), half had pain and shortness of breath. Difficulty with food and fluid intake was problematic for the majority (75 %) and one-third had dehydration and weight loss [44]. Barriers to symptom management in SNF care include resident factors (e.g. inability to report symptoms) and system factors (e.g. lack of knowledge and or procedures for symptom assessment and management). For residents unable to self-report symptoms, evidence supports that staff and family assessment of quality of care of symptoms are similar [44].
As a first step in symptom management, heart failure medical therapies should be optimized as per published guidelines [9, 45]. However, residents with hypotension or renal failure may not tolerate maximal guideline therapies. The symptom experience encompasses more than simply a physical sensation. Symptoms are generally multiple in numbers and incur physical sensations, cognitive interpretations, and affective reactions to symptoms [46, 47]. It follows that interventions to manage symptoms are therefore multidimensional in nature. (See also Chap. 3.)
Dyspnea
Dyspnea is one of the most common symptoms in advanced heart failure. Interventions for dyspnea are directed at reducing ventilatory demand, ventilatory impedance, and improving central perceptions of dyspnea and inspiratory muscle function [48]. For residents with refractory dyspnea, low dose opiates can be used to reduce ventilatory demand and improve the resident’s perception of dyspnea [49, 50]. To assess benefit and observe for adverse effects of opiates, it is appropriate to begin with short-acting opiates. For example, oral morphine starting at 2.5–5 mg every 4 h can be given and slowly titrated upward as needed. Respiratory depression secondary to opiates is generally avoided with slow up-titration [25]. It is appropriate to treat continuous symptoms with long-acting opioids. Among patients with renal dysfunction, metabolites of morphine can accumulate resulting in signs of toxicity such as myoclonus, anxiety and delirium. In such cases, alternate choices for managing dyspnea include hydromorphone or methadone [25].
Although commonly perceived by patients and clinicians to be a useful treatment for dyspnea, oxygen is appropriate only for patients with hypoxia [51, 52]. Nonpharmacological strategies for dyspnea include maintaining a calm environment to prevent sympathetic stimulation, sitting the resident up leaning forward and using a fan directed at the resident’s face [53–55]. For residents with cardiomegaly, avoiding sleeping in the left lateral decubitus position is suggested [56, 57].
Pain
Pain of both cardiac and non-cardiac origin is prevalent among patients with heart failure [13, 58, 59]. The prevalence of pain also increases with worsening functional class [60]. Although pain is not a typical symptom of heart failure per se, approximately 75 % of patients experience various types of pain. Consequently, regular assessment of pain at end-of-life is warranted including documentation of nonverbal signs (grimacing, decreased functional status) among residents with cognitive dysfunction. Importantly, pain may present atypically in older adults manifested by changes such as newly developing confusion, fatigue, or depression [61].
Treatment varies with the cause of the pain. Treating underlying comorbid illness such as degenerative joint disease, arthritis, and angina is generally effective for most patients [13]. However, non-steroidal anti-inflammatory drugs should be avoided due to risk of fluid retention, negative effects on renal function and the potential for gastrointestinal bleeding [5]. For moderate to severe pain, short acting opiates initially can be used as described for dyspnea. Long-acting opiates can be used after the short-acting opiates are up titrated to a satisfactory level to manage pain [3].
Fatigue
Fatigue is one of the most common symptoms and one of the most distressing to patients with heart failure [13, 14]. The subjective sensation of fatigue ranges from lack of energy to persistent tiredness to profound exhaustion [62]. Fatigue is non-specific in that this symptom is associated with chronic illness (e.g. cancer, chronic obstructive lung disease), benign causes and more serious acute illness (e.g. acute coronary syndrome) [63]. Further, it can result from physical as well as mental illness (e.g. depression). The symptom of fatigue has implications beyond quality of life and functional capacity. Among patients with heart failure and reduced ejection fraction, both severe exertional fatigue (HR = 2.59, 95 % CI: 1.09–6.16, p = 0.03) and severe general fatigue (HR = 3.20, 95 % CI: 1.62–6.31, p = 0.001) predicted increased risk of mortality [64]. Interventions for fatigue begin with assessment and identification of any underlying cause whether physiological, psychological, or situational. Treatable causes of fatigue include but are not limited to anemia, infection, dehydration, electrolyte imbalance, thyroid dysfunction, depression, polypharmacy and pain [65]. Obtaining palliative care consultation is reasonable if a likely source of fatigue remains elusive [66].
Depression
Depression is common among patients with heart failure with approximately 20 % of patients having a major depressive disorder [67]. Quality of life is negatively affected as depression also is associated with increased hospitalization rates and higher symptom burden [43, 68, 69]. Identification of depression in the SNF population is challenging. Diagnosis is complicated by comorbid illness in some instances and cognitive deficits in others. Screening using mood is recommended over physical symptoms in this population as physical symptoms such as loss of appetite or insomnia are common in advanced heart failure [25, 70].
Considering the relationship between symptoms of heart failure and depression, careful guideline-based management of heart failure is the first step in mitigating depressive symptoms. Other interventions commonly used for depression or anxiety (e.g. psychotherapy, antidepressant medication, exercise) may be considered. Sertraline, a selective serotonin reuptake inhibitor tested in a randomized placebo-control trial of a population generally younger than found in long-term care, was safe although investigators initially reported no differences in efficacy between sertraline and placebo [71, 72]. In a subsequent analysis using growth mixture modeling, sertraline was more effective than placebo in 20 % of the responders [71]. Tricyclic antidepressants are avoided in patients with heart failure due to quinidine-like effects [3, 73]. Interventions such as exercise and psychotherapy should be considered when feasible at end-of-life.
Grief and Bereavement Management
Palliative care interventions are aimed at improving quality of life for residents by relieving distressing symptoms experienced over the disease trajectory to end-of-life. In addition to medical and nursing management of symptoms, the components of palliative care for residents and their caregivers include: (1) coordination of care using interdisciplinary care teams, (2) psychosocial and spiritual support along the disease continuum to end-of-life, (3) use of a palliative care specialist alongside medical care (integrated care delivery), and (4) inclusion of families and caregivers in all plans of care including transitioning to hospice care, grief and bereavement. Transitioning to hospice care is a mechanism for improving end-of-life care, however evidence indicates that patients with heart failure residing in long term care are less likely to enter hospice (OR = 0.613, CI: 0.477–0.787) [74].
Palliative care interventions help residents’ families cope with the resident’s illness, their personal bereavement, and sense of loss [75]. Often, the terms grief and bereavement are used interchangeably, but are conceptually different. Grief is a process and describes a person’s physical, emotional, cognitive, social and spiritual response to loss. Bereavement is the time period after the loss during which grief is experienced [76]. Bereavement outcomes vary among individuals. A good bereavement outcome is exemplified by lessening of grief over time and a return to usual life activities. Poor bereavement outcomes are described as little or no decrease in the intensity of grief after a 2 year grieving period. The chronic and progressive nature of heart failure may increase risk of a poor bereavement outcome if there is prolonged anticipatory grieving. Risk is increased particularly if their loved one also has cognitive dysfunction or dementia [77]. Other risk factors for poor bereavement outcomes that may relate to families of nursing home residents are outlined in Table 8.3. Referrals to additional resources such as grief counseling are appropriate for those at risk for poor bereavement outcomes.
Table 8.3
Risk factors for poor bereavement outcomes
Ambivalent relationship with deceased person |
Poor family coping (lack of cohesion, high conflict) |
History of mental illness of the bereaved person |
Prolonged care giving |
Lack of social support |
Bereavement interventions generally are provided by front-line caregivers (Registered Nurses, nurse’s aides, social workers, chaplains, volunteers) trained in the various aspects of bereavement. Bereavement support includes clear information about the dying process, attention and active listening by the staff, and assistance obtaining end-of-life services. Long-term care residents without dementia and their families may benefit from dignity therapy during the course of palliative care [78, 79]. Dignity therapy is a brief psychotherapeutic technique which uses reminiscence to record and document what is important to the individual. (See also Chap. 10.) After the document is edited to the satisfaction of the resident, it can be shared with loved ones if desired. The document potentially may be a useful bereavement tool for the family, but should be used based on individual preference. Negative response to the dignity document has been reported by those with higher caregiver burden or negative relationships with the patient [78].
< div class='tao-gold-member'>
Only gold members can continue reading. Log In or Register to continue
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
