18.1 Management of the actively dying patient

Just as each phase of the illness journey has its unique challenges, so too does the care of the actively dying patient. The period leading to death is characterized by increasing prevalence and intensity of physical, psychological, existential, and social concerns, and it is often a challenging time for patients, their families, and health-care providers.

Addressing the whole-person needs of the patient at this time with excellent clinical skills, compassion, and understanding can greatly reduce the suffering experienced as well as support and maintain patients’ dignity and autonomy (Cassem, 1991; Thompson and Chochinov, 2010). The goal of care during this phase is to minimize distress for patients and their families by meticulous symptom control and supportive care, while maintaining the well-being and moral integrity of all.

The premises that underlie end-of-life care are:

The idea of a ‘good death’ is a concept that medicine is trying to identify and quantify, although the construct is itself also controversial (see Chapter 2.4) (Miyashita et al., 2007; Hughes et al., 2008). Whilst ensuring that goals of care shift to allow for those priorities that contribute to a ‘good death’—this transition must also be characterized by non-abandonment.

Recognizing that a patient is imminently dying is the prerequisite for providing appropriate end-of-life care. Indeed one of the common problems in the care of the dying is failure to appreciate that the patient is indeed close to death. Prognostication and recognizing that death is near is, therefore, a cardinal clinical skill (Glare and Sinclair, 2008).

Establishing a short-term prognosis is part of the therapeutic alliance and yet is one of the most challenging aspects of care for many. Using and communicating this knowledge requires an individualized, patient-centred, and culturally appropriate approach (see Chapter 2.3).

The terminal phase is defined as a period of irreversible decline in functional status prior to death. The terminal phase can last from a few hours to days and very occasionally to weeks. Worsening of certain symptoms is indicative of entering the terminal phase (Pitorak, 2003):

Recognizing this phase allows for the pre-emptive planning of care for the patient, preparation of family, patient, and staff, and may involve discussions around preferences for end-of-life care, place of death, and anticipated symptom control.

Whilst there are situations where it is possible to identify the terminal phase with some accuracy, many diseases have a natural history of progression and exacerbations that make the terminal phase difficult to identify (Lunney et al., 2003).

End-of-life care discussions make a difference. Well-conducted discussions about end-of-life care foster trust and non-abandonment for patients and carers and improve end-of-life outcomes for both the patient and the bereaved (Steinhauser et al., 2000; Detering et al., 2010). Discussing the approaching terminal phase and end of life with patients and families reduced the number of aggressive medical interventions near death, and, compared to those who had not had these conversations, reduced distress and improved quality of life of both patient and family/caregiver. Importantly, recent studies have demonstrated that end-of-life discussions were not associated with increased patient emotional distress or psychiatric disorders (Emanuel et al., 2004; Wright et al., 2008; Mack et al., 2010) and Mack and colleagues found a reduction in family/caregiver bereavement morbidity when the dying patient had the benefit of end-of-life discussions (Mack et al., 2010).

When the terminal phase is recognized, this needs to be communicated to the family and patient with empathy and equanimity. This is a task requiring sensitivity and skill, which can be improved with specific training. It is important for health-care professionals to be aware of their own feelings of grief, fear, anxiety, and guilt, and to acknowledge the impact this can have on the way they approach this conversation and ongoing communication (Curtis et al., 2004; Hancock et al., 2007).

Cultural awareness is extremely important and influences how and what information is communicated. Using interpreters can be vital (Crawley, 2005; Searight and Gafford, 2005; Volker, 2005). Frequently patients and families fear uncontrolled pain and other distressing symptoms, and should be confidently reassured about the management of these. When discussing future care planning and possible symptoms to anticipate, the family and the patient may have different information needs. Ideally the aim is to maintain a balance between supporting the patient’s and family’s sense of control and dignity, reducing their fear of suffering, and maintaining the hope for good quality of living, whilst preparing for death (Steinhauser et al., 2001).

Care of the family is integral to caring for a dying patient. Surveys of families of patients dying from advanced cancer have identified a series of critical needs (Table 18.1.1). These can be used as a checklist for assessing family needs, preparing for terminal care at home, and promoting family well-being. Families with dependent children have complex needs and make choices around end of life that have been found to differ from others (Nilsson et al., 2009).

Among patients with chronic disease, life-threatening events and death occur in two distinct situations:

In patients with terminal illness, cardiopulmonary resuscitation (CPR) may prolong, but cannot reverse, the dying process; however, the patient, family, and indeed some health-care professionals may not fully comprehend the ineffectiveness of this intervention for the patient. It is important to discuss with patients, family, and team members the role of CPR and not for resuscitation/do not resuscitate orders as failure to discuss these may result in patients receiving unwanted interventions (Hofmann et al., 1997; Loggers et al., 2009). This conversation is particularly difficult because it highlights the reality that the person’s death is imminent. Such discussions are frequently avoided until the days just before death. However, these discussions have not been found to change patient or surrogate satisfaction with the clinician (Anderson et al., 2011). Despite being extremely unwell and aware of their disease process, dying patients differ in their capacity to acknowledge prognosis, with the majority only partially aware of the proximity of death (Chochinov et al., 2000; Mack, 2010).

The conversation should focus on the positives of what can and will be done for the person at the end of their life (Harlos, 2009)—with the emphasis on their dignity, comfort, and the prevention of distress. This should be done with sensitivity and empathy, allowing time to deal with the questions, concerns, and sometimes strong emotions associated with this issue (Lacey and Sanderson, 2010).

The effect of the illness on the body and, sometimes, its treatment can cause sudden life-threatening medical crises for patients with advanced chronic illness. Some of these events may be reversible, some may have a very small chance of being reversed, and some may not be reversible at all.

Given their state of chronic ill health and ultimate incurability, patients vary in their approach as to how they would want to be treated in this situation. Opinions are often influenced by any one of many factors, such as religious or cultural beliefs, fear of suffering, fear of death, unfinished life plans, or the advanced stage of the disease:

The issues regarding advance care planning for patients with anticipated dying from vital organ failure as a consequence of their disease are different from the issues of those patients with unanticipated, acute life-threatening events. It is inappropriate to conflate these two issues in ascertaining patient preferences for end-of-life care.

It is important that this discussion and any subsequent decision is clearly documented in the patient’s medical records and communicated to other health-care providers.

As a person approaches death, their ability to make decisions may become impaired. Encouraging advance care planning, directives, or discussions with the patient prior to the terminal phase greatly reduces the burden on the substitute decision-maker and ensures the patient’s choices are respected (Detering et al., 2010; Keating et al., 2010).

At the end of life, studies clearly suggest that existential and psychosocial issues greatly concern dying patients (Boudreau and Cassell, 2010). Through understanding and addressing factors that may potentially overwhelm the patient, family, and/or health-care providers, the necessary preconditions for coping can be established (Chochinov et al., 2008; Cherny, 2009; Cassell, 2010; Cassell and Rich, 2010). This is a vital and integral part of the care of the actively dying person and is discussed in detail in Section 17 of this textbook.

Problems of weakness, fatigue, and loss of appetite become more evident as death approaches. In the terminal phase, these problems usually start to assume relatively less significance for patients—although not necessarily for families. Comfort-related symptoms become increasingly important—particularly pain, dyspnoea, delirium, fear/anxiety, and respiratory secretions. Active assessment and pre-emptive management of these problems is crucial, especially as the patient becomes less able to communicate. Whilst the overall approach to managing these problems is presented elsewhere, the following discussion relates to the end-of-life setting, which creates some specific challenges.

Only essential medications should be given and they should be administered by the least invasive route that can provide adequate relief. An appropriate rule is to identify and cease any drug which does not contribute to the patient’s immediate comfort. Likewise observations and investigations that do not contribute to comfort should be ceased. Families will need reassurance that meticulous care continues with comfort being the goal of care (Abrahm, 2005). If persistent reservations about the withdrawal of a medication that is not causing the patient harm (such as an antibiotic) are expressed by the patient and/or family, the medication may be continued. It is often helpful to set a time limit on such treatment, whilst identifying specific improvements that would justify further treatment beyond that point.

Medications for pain and symptom control should be continued and the route converted from oral to transdermal and/or parenteral (the subcutaneous route is preferred wherever possible as it is less traumatic for the patient). Loss of the ability to swallow in the terminal phase should be anticipated and planned for by provision of as-needed (PRN) medications so that patients are not left without adequate symptom control, especially for a crisis (e.g. seizure, airway obstruction, bleeding, and so on). As a person’s physical condition changes so too can their renal function and metabolism of medications. Careful monitoring and adjustment of medications is frequently required (see Table 18.1.2).

Pain management continues to be crucial. Based on a precautionary principle, it is important not to discontinue analgesics in the dying patient. Ensure availability of medications to be given by a non-oral route (preferably subcutaneous) and review the clinical need for any medications for which there is no parenteral form available, for example, some of the adjuvant analgesics. Whilst decreased levels of physical activity may mean that pain control is adequate despite ceasing oral adjuvants, if this is not the case, opioids may need to be increased to cover additional requirements. Alternative adjuvant medications, for example, anti-inflammatory or antineuropathic agents that can be delivered subcutaneously may be required (see Section 9 of this textbook for details of pain management).

Up to 70% of patients experience dyspnoea in the last 6 weeks of life, and its prevalence, severity, and associated distress increase as death approaches (Ben-Aharon et al., 2008; Kamal et al., 2012) (see also Chapter 8.2). It is usually multifactorial. Though it may be clinically appropriate to manage reversible causes of dyspnoea, in the final days of life dyspnoea can be more refractory to interventions, as the dominant causes become less amenable to modification (e.g. obstructing tumour; general weakness and frailty; restrictive conditions, including chest wall disease; lymphangitis carcinomatosis; pleural infiltrate or refractory effusions; diffuse extensive lung pathology or infection.) In this setting, pharmacological management with oral or parenteral opioids to reduce the distress of dyspnoea has been shown to be of benefit (Allard et al., 1999; Ben-Aharon et al., 2008) (see Box 18.1.1). Benzodiazepines may be useful for reducing panic and anxiety frequently associated with dyspnoea but have not been shown to improve dyspnoea itself. Oxygen therapy has been found to improve dyspnoea in the hypoxic patient (Booth and Wade, 2004; Gallagher and Roberts, 2004; Cranston et al., 2008; Uronis et al., 2008). Whilst non-pharmacological measures such as a fan and positioning are simple techniques that can improve the patient’s comfort, in contrast, relaxation and breathing techniques take considerable concentration and are difficult to initiate late in the terminal phase (Zhao and Yates, 2008). For the patient with refractory dyspnoea and severe distress in the terminal phase, sedation is an option to consider for maintaining adequate symptom control and reducing suffering (see Chapter 8.2).

In the terminal phase, delirium has been found to be present in up to 88% of patients when carefully screened for (Lawlor et al., 2000a, 2000b) (see also Chapter 17.5). Delirium may be hypoactive or hyperactive (de Rooij et al., 2005). Although witnessing agitated delirium is upsetting for family, caregivers, and health-care professionals alike, the degree of patient distress associated with the thought-disordered states of hypoactive delirium should also not be underestimated (Leonard et al., 2008; Chochinov and Breitbart, 2009). Refractory delirium is a poor prognostic factor and often heralds the terminal phase (Maltoni et al., 2012).

In a patient with hours to days to live, the key issue is deciding whether to symptomatically manage delirium rather than burdening the patient and family with a search for causes or trial of management for reversibility. The aim is to maintain the patient’s dignity, reduce any agitation and associated risks to the patient and caregivers, and to ease distressing hallucinations, delusions, and restlessness (Caraceni and Simonetti, 2009). The symptoms of delirium at the end of life need to be palliated confidently, with compassion, reassurance, and clear explanation of the impact of pharmacological and non-pharmacological strategies to improve symptom control (Breitbart and Alici, 2008). Medications for treating agitation and delirium in the terminal phase include antipsychotics able to be given parenterally—such as haloperidol, levomepromazine, risperidone, or olanzapine (by sublingual wafers) and chlorpromazine (which can be given rectally). Benzodiazepines may be required for more rapid control of agitation and anxiety, but can potentially worsen delirium if used alone. The doses used should be the lowest dose that is effective for the symptoms of concern (see Chapter 17.5).

The condition identified as ‘terminal restlessness’ is a frequently described though little-studied state of agitation and reduced conscious level which appears to be closely associated with the onset of the terminal phase (Brajtman, 2005). Diagnostically it is most consistent with a refractory agitated delirium in an actively dying patient, possibly triggered by the onset of the terminal phase itself. However, before making this diagnosis it is worthwhile to consider, and exclude if clinically appropriate, other physical, medication-related, or medical causes of agitation. The latter includes status epilepticus, akathisia, serotonergic syndrome, myoclonus, hypoxia, physical pain or discomfort in an obtunded patient (e.g. from itch, urinary retention, or faecal impaction), medication toxicity, or paradoxical reaction (White et al., 2007). If no other cause is found and the clinical context is consistent with the onset of the terminal phase, palliation should be prompt and effective using sedating antipsychotics with or without benzodiazepines to reduce patient awareness and motor agitation.

During the last hours to days of life, terminal secretions (the ‘death rattle’) may become evident. They have been found to occur in around 44% of dying patients (Morita and Hyodo, 2004). The mechanism remains uncertain, but it may be due to pooling of saliva and secretions in the upper airways along with reduced swallowing reflexes, dysphagia, and inability to expectorate. This has been called type I death rattle or ‘true’ death rattle (Bennett et al., 2002; Wildiers and Menten, 2002) in contrast to type II rattle, which is due to pooling of bronchial secretions. Type II or ‘pseudo’ death rattle can appear over days and may not respond as well to antimuscarinic/anticholinergic medications. Included in type II are other causes of retained secretions, such as bronchorrhoea from primary lung tumour, infection, dysphagia, bleeding, airway obstruction, pulmonary oedema, or tracheo-oesophageal fistula (Bennett et al., 2002; Mercadante et al., 2011). No significant association has been found between hydration status and the development of death rattle (Ellershaw et al., 1995; Morita and Hyodo, 2004). The true death rattle is a strong predictor of approaching death—in one study 76% of patients died within 48 hours of its onset (Wildiers and Menten, 2002).

Antimuscarinic agents can be used to reduce the salivary secretions. Although most studies of the efficacy of antimuscarinic agents on reducing salivary secretions are mostly based on single-dose studies, a randomized trial of three anticholinergic agents in 333 patients (Wildiers et al., 2009) showed beneficial responses in up to 76% of patients by salivary secretion reduction. This was less marked in secretions thought to be of bronchial origin or ‘pseudo death rattle’. The evidence supporting this practice is not robust but therapeutic trial is reasonable, paying attention to the adverse effects profiles (Wee and Hillier, 2008; Mercadante et al., 2011,). It is important to reassure families that terminal secretions are rarely distressing for the patient, although they may be distressing for the family.

With progression of illness and reduced overall function there is an associated reduced ability to carry out independent activities of daily living. Transferring may require assistance and the environment of care may often require modification to improve access, safety, and comfort. If available, the input of an occupational therapist and physiotherapist is very useful preferably prior to the onset of the terminal phase.

Meticulous nursing and active monitoring for symptom distress is essential in the care of dying patients. Common problems include immobility, pressure area care, care of bladder and bowel, secretions, xerostomia, and odours. For the imminently dying patient, measuring and recording patient data not directly related to comfort (such as pulse and blood pressure) is superfluous and can be discontinued unless there are religious, cultural, or other concerns (Cherny et al., 1996).

Incontinence of urine, reduced mobility impacting on toileting options, and urinary retention are all potential problems. These can impact on a person’s sense of dignity and should be dealt with sensitively. Catheterization or, if no retention, incontinence pads, are often required. Ultrasound bladder scanning is a valuable investigation to minimize inappropriate catheterization. Judicious rectal intervention may be appropriate to maintain regular bowel motions if the patient is uncomfortable. Faecal impaction and urinary retention can cause severe distress and should always be excluded as a cause of agitation in the obtunded patient.

Uncontrolled suffering in a dying patient is considered a medical emergency (Cherny, 2009), but suffering cannot be treated unless it is recognized and diagnosed (Cassell, 1999) (see also Chapter 18.2). Occasionally in the terminal phase, symptoms and suffering are refractory to all standard palliative interventions. Suffering is considered refractory if, despite use of aggressive and concerted efforts to determine and treat the cause, suffering persists. Refractory suffering can relate to intractable symptoms including complex pain, severe dyspnoea, nausea and vomiting, neuropsychiatric symptoms including agitated delirium, fitting, and very occasionally severe anxiety or depression, or other irreversible crises. All available and appropriate diagnostic, therapeutic, and consultative processes need to be exhausted (including consultation with pain, palliative care, psychiatry, and other relevant experts if available) before classifying suffering as refractory (Krakauer and Quinn, 2009) (Box 18.1.2).

Palliative sedation is an option in the terminal phase if refractory suffering is present. The importance of using the minimum level of sedation required to adequately manage refractory symptoms and suffering is stressed. Existential distress and suffering can be viewed as intrinsically linked to the impact of terminal illness on the whole (Cassell and Rich, 2010) (see Chapter 18.2 for further discussion on sedation).

Identification of patients at high risk for uncontrolled pain, dyspnoea, nausea and vomiting, agitated delirium, acute airway obstruction, or massive haemorrhage is crucial. It allows contingency planning for a crisis, and occasionally prophylactic management. This usually enables a family to often care for their dying relative at home or in the desired setting with minimal angst and distress.

The specific drugs, doses, and routes of administration ordered as part of a crisis plan should be appropriate to the clinical situation. If severe symptoms requiring rapid relief are anticipated, maintenance of parenteral access is recommended. This might include a subcutaneous butterfly. In the home care situation, whenever possible, family members/carers must be instructed in preparation and administering medications for an emergency (see Table 18.1.2).

In the last days to weeks of life, anorexia, weight loss, and swallowing difficulties are common. In patients with disease-related dysphagia or bowel obstruction, the ability to maintain adequate nutritional intake is reduced. Weakness, cognitive deterioration, or other factors increase aspiration risk and further compromise a person’s ability to eat.

Reduced oral intake is often distressing for the family. Requests for artificial nutrition via enteral or parenteral routes are common. As the patient’s conscious level deteriorates, aspiration risk is greater and the risk versus benefit of oral intake with regard to aspiration needs to be explored with the family. Often compromises are reached at this stage. However, concerns about cessation or non-initiation of enteral or parenteral feeding often poignantly relate to the expectations and hopes that patient and family still hold (McClement et al., 2003, 2004).

The discussion needs to be centred on the ineffectiveness and burden of artificial nutrition versus any likely benefit. If there are religious or culturally based reservations about discontinuing nutritional support, it may be maintained, unless there is evidence of patient harm from the intervention (Cherny et al., 1996).

In the actively dying patient who is comatose or poorly responsive, the decision to provide artificial hydration is based on clinical judgement, and assessment of the burden of the intervention. Where families are concerned about the need for fluids, provision of gentle hydration is reasonable. Excessive hydration, particularly in the hypoalbuminaemic patient, may exacerbate oedema and discomfort, and if a trial of therapy is offered, the family should be warned that worsening oedema or other signs of fluid overload would be a reason to cease the treatment (Hanks et al., 2009). Reassurance can be given, particularly in the last days, that maintaining good mouth hygiene alleviates much of the sensation of thirst and discomfort. The majority of trials conducted at the very end of life show no significant burden nor benefit from hydration. If it is decided to administer fluids in the last days or weeks of life, the subcutaneous route is a well-tolerated and minimally invasive option. 1–1.5 L over 24 hours of normal saline can be delivered in this manner (Dalal et al., 2009).

The clinician at the bedside of the dying patient should ensure adequate relief of patient distress, and provide support and guidance to the grieving family. This often includes responding to family members who ask what to expect, and how to recognize the dying process. The physician and nurse’s role is to evaluate and treat the patient for pain, dyspnoea, agitation, or any other distressing symptoms. Once adequate comfort is achieved, the clinician can reassure the family that the patient is not distressed.

Families vary in the degree of professional support necessary at this time. Increased involvement is valuable when family supports are limited or coping is frail. In other situations, it may be enough for the clinician to acknowledge the primacy of family and friends in this moment, taking a background role, but available if necessary.

If it is clear that breaths are agonal and the patient is about to die, explain this to the family with the reassurance, again, that the patient appears comfortable. Adequate provision for family privacy and time with the patient is crucial. Awareness and respect for cultural and/or religious practices related to death is important. If the death occurs in an institution, provision for necessary cultural practices should be made and the family should be reassured of this. At the time of death, families often talk and cry over the dead person, and kiss and hug them. An offer of condolences from clinicians may be very comforting for a distressed family.

Most palliative care services provide bereavement support, and can refer those affected by the patient’s illness and death to an associated bereavement service after the patient has died. Adults need to recognize when a dependent, grieving child needs help. When a parent of dependent children is dying, offering their children age-appropriate grief and bereavement counselling and education may be very reassuring for the dying parent (Auman, 2007; Sandler et al., 2010).

It is important to prepare for some of the more difficult and unusual end-of-life care scenarios. These include unanticipated sudden events, such as trauma, postoperative overwhelming sepsis, or failure of treatment for acute potentially reversible disease as well as other treatment-related deaths. Young parents and other young people (adolescents), children, the isolated, incarcerated, or mentally ill, those in care isolation wards, and those in an intensive care unit are other unique groups whose end-of-life care may require a unique and thoughtful, planned approach. The unanticipated sudden event requires action to be taken with care to alleviate suffering and provide appropriate support to the family and carers. For those with advanced care plans, communication and respect for the plans is key to provision of care. ‘Crisis medications’ should be available on the hospital wards and in the home for adequate symptom control for those at risk of an acute event. Preparing a patient, family, and team may not always be possible and the grief and bereavement support become a very important part of care.

Young people and young children and their families may request more active management even at the end of life. Here, care for the family and early psychosocial support is vital with many wishing to return home (or remain at home) for end-of-life care in the final days. The isolated or incarcerated person may have little access to adequate multidisciplinary care and modifying limited resources and mobilizing adequate support may prove the greatest challenge. Provision of the minimal essential drugs is imperative and may require thoughtful proactive intervention.

It is the role of the treating team to ensure that all of their colleagues are ‘on the same page’, and that at this advanced stage of illness the patient and their family are receiving the same message regarding stage of disease, prognosis, and goals of care. The nursing and allied health professionals involved in the care of the patient should also have a clear understanding of goals of care and approach to the patient entering the terminal stage.

Despite sounding straightforward, much conflict and angst between staff members stems from lack of such communication. Furthermore, for staff faced with a dying patient on their ward, particularly nursing staff who do not often care for dying patients, the entire process may be distressing and confronting. It can be worthwhile to acknowledge the impact the patient’s dying and the family’s distress could be having on the team. For many treating team members, there are long associations with their patients and close relationships formed. Grieving needs to be recognized and space provided for this to occur.

As a member of a team caring for patients at the end of life, the anticipated and non-anticipated death can have a significant impact on staff and team members. Junior staff are often more vulnerable. It is important to recognize this and for appropriate support to be offered in the form of structured professional supervision, a time and space for debriefing, discussions, and willingness of senior team members to be available to respond to the distressed staff member. Many resources are available that may assist and these should be chosen based on suitability to staff and service needs.

Care of patients in the final stages of a life-limiting illness requires a high level of clinical vigilance and skill in order to ensure that the passage from life to death is as free from suffering as possible. Patients who are dying have a right to adequate relief of physical and psychological symptoms, and they and their families have a right to adequate support. The care of patients and their families requires (a) interdisciplinary cooperation of a health-care team incorporating physicians, nurses, social workers, and other allied health staff, and (b) a high level of clinical flexibility to address the evolving needs of the patient and family. Contingency planning and communicating expected symptoms and signs to family and carers, coupled with provision of adequate access to medications and non-pharmacological resources, can allow a person to die peacefully at home or in the desired setting with minimal distress. Participation in this process challenges the clinician’s emotional resources and medical skills. There is, however, much professional satisfaction in helping to ensure a ‘good death’ since relief of suffering is at the very heart of medicine. Familiarity with guidelines in the care of the dying can reduce the potential for distress in this important clinical endeavour.

Chapter adapted from Judith Lacey and Christine Sanderson, The Oncologist’s Role in Care of the Dying Cancer Patient, Cancer Journal (ISSN: 1528-9117), Volume 16, Issue 5, pp. 532–541, Copyright © 2010, with permission from Lippincott Williams & Wilkins, Inc.

Thank you to Dr Joanne Doran and Dr Jan Maree Davis for review and comments.

Complete references for this chapter are available online at <http://www.oxfordmedicine.com>.