CHAPTER      34

End-of-Life Care

SUSAN ZWEIZIG SARAH HUGHES LYNETTE DENNY JOANNA CAIN

The goal of end-of-life care is to provide maximal palliation of symptoms and maximal psychological support at the end of life, as part of the continuum of oncology care. Ensuring that our patients get the best care as the goal of medicine shifts from curing disease to relief of suffering is one of the obligations of health professionals. To do so effectively requires engaging not only the physicians and nurses who have been part of the oncology treatment, but also the skilled assistance of palliative care and hospice professionals with the patient and their caregiving network of family and friends. It is a particular challenge in low-resource settings, where many of the medications and support structures are limited or absent, making reliance on family and friends with guidance from mid-level or trained health assistants who may not have participated in whatever level of prior cancer care the woman received the best path to palliative care.

THE END-OF-LIFE TRANSITION

Discussions about the nature of disease, the status of treatment, and the expectation and side effects of treatment are an ongoing conversation in oncology care. With the lengthening of survival for even the most deadly gynecologic cancer, ovarian cancer, and the multiplicity of options for ongoing chemotherapy, the decision to limit chemotherapy or other interventions involves multiple conversations over time aimed at balancing the potential side effects of new chemotherapies or other treatments against the expected benefit and impact on quality of life for the patient. In the setting of continuing conversation between the health care team and the patient, with or without her significant support individuals, there is often a point at which the ongoing therapy aimed at disease control or cure has a harm to quality of life that is greater than any benefit—a point at which the transition to comfort-focused end-of-life care is more clear, although palliative care may well have been initiated for some symptoms prior to this point (Table 34.1).

The balance can be difficult, as Callahan points out: “The most fundamental problem with technological medicine is twofold: that it can give us a longer life and a slower dying and that it can keep us alive when we might be better off dead” (1). Patient perspectives on what constitutes high-quality end-of-life care reinforces the importance of interactions most physicians would identify, such as communication, accessibility, continuity, and emotional support, but surprisingly, oncology patients compared to terminal patients with other diagnoses more often hold on to a desire to maintain hope (2). As a health professional, keeping this door open while realistically portraying prognosis and options and pursuing palliative care can be the hardest balance of all.

The timing and expectation of death for cancer patients continues to change. Thirty years ago, survival from advanced stage ovarian cancer of more than 2 years was very unlikely, and now is routine. As a backdrop to this advance, the age of death in the developed world has steadily increased, as the major cause of death has changed over the past hundred years from sudden death from infection, trauma, and childbirth in the young to a slower decline from chronic illnesses, usually in advanced age. Cancers that previously caused a rapid death now follow a more chronic recurring course, which changes not only patient and family perceptions about the length of time until they need to deal with the issue of death, but offers a longer time to work through the phases of dying and the opportunity to bring to closure key goals for an individual. Even the gender and role of caregivers at the end of life are changing, and the role of men as caregivers for women with gynecologic cancers deserves special attention (3). The development of palliative care teams, usually in hospital settings, has added significant expertise to symptom management, discharge planning, and accomplishing improved end-of-life care (4). Engagement of gynecologic oncologists into these teams further improves the breadth of knowledge and team perspective as end-of-life plans are made. Given the prolonged course of many of these cancers, the advance planning for end-of-life care has both more time for consideration and can evolve as the patient’s prognosis changes. The frequency and timing of hospice referral has been studied in older women with ovarian cancer, and significant disparities in care have been identified. Although more older women received hospice care during the time period studied, the proportion of late referrals to hospice stayed the same. Similar to other studies, significant disparities in the use of hospice were seen depending on race and socioeconomic status. Timely hospice referral was identified as a significant area for improvement in the care of older women dying of ovarian cancer (5). A recent study showed that even in the absence of a prolonged course of illness, the treating health care team can significantly improve end-of-life care by informing the patient of imminent death, thereby increasing the likelihood of patients and families getting the resources needed at the time of death, with no increase in anxiety seen (6).

ADVANCE PLANNING FOR END-OF-LIFE CARE: REDUCING THE EMOTIONAL BURDEN ON SURROGATE DECISION MAKERS

Along with improved symptom management at end of life have come improved vehicles for ensuring that patients have their wishes represented in their care, in terms of legal documents, but also ensuring that surrogate decision makers adequately represent patient wishes.

Serving as a surrogate for a woman with advanced cancer who can no longer make her own treatment decisions places a profound burden on the surrogate. Advance planning can make the burden lighter. Both the public and health care professionals often underestimate the burden of serving as a decision maker. Formal measures of stress document that months after serving as a surrogate in a decision to withdraw life-sustaining treatment, the surrogate’s emotional stress levels remain remarkably high. One study found that the stress of serving as a health care decision maker was twice as high as the stress levels on households who had lost their homes to fire in the Berkeley-Oakland firestorm (7). Decision makers commonly report waking up at two in the morning and wondering if they made the right decision. They make statements such as, “It was the hardest thing I have ever done in my life.”

When patients have not previously documented their wishes about life-sustaining treatment, the stress levels of their surrogate decision makers tend to be even higher. When surrogates are guided by the patient’s written advance directive, their stress levels are about 30% lower when measured 2 months after a decision to stop life support (7). Surrogates who “stand by” and affirm the patient’s prior written instructions worry less about whether they made the right decision. Those who had never talked with the patient ahead of time felt the anguish of “deciding for” a woman they love. An important message to women with cancer who are reluctant to engage in advance care planning is that they are continuing to take care of the ones they love by completing an advance directive and lessening their burden.

THE ROLE OF TRADITIONAL ADVANCE DIRECTIVES

Advance directives provide a woman with a means to legally document her preferences about medical treatments in the event she is no longer able to make her own decisions. The legal details and breadth of clinical context vary substantially by state within the United States, and by country of the world, with some legal systems not supporting this type of advance direction for their health care (8). In many jurisdictions, these documents allow the formal appointment of a surrogate who becomes the woman’s legally authorized medical decision maker in the event of future incapacity. Being asked to document one’s preferences for treatment in conditions like “imminent death regardless of treatment” and “persistent vegetative state” can be anxiety provoking for anyone. The context of a new diagnosis of cancer often makes completing an advance directive a statement of prognosis rather than planning for something that may or may not be needed, and perhaps overwhelming. It is difficult to combine conversations about advance care planning with initial discussions of curative treatment and creates a temptation to put off these discussions until they have no further treatment options to offer or the patient declines further curative treatment.

Waiting until the patient is gravely ill or near death to have these important conversations risks that the patient may become too ill to be able to make her wishes known. Without prior planning, the patient will not have appointed the surrogate of her choice. Without her guidance, family and friends may disagree on treatment decisions, which add further to family distress. In addition, patients and their loved ones often benefit from having advance planning be more of a process than an event. Having discussions with the patient about who should and should not be given medical information can naturally lead into who would be the most appropriate decision maker should the patient temporarily be unable to make decisions for herself. When the patient reveals strongly held value differences within her social network, or when she indicates that her most trusted surrogate is not her legal next of kin, the patient should be strongly encouraged to complete a written advance directive if that legal construct is available.

Advance directives are remarkably effective in allowing the patient to designate her choice of a surrogate decision maker. These legal instruments are also effective in silencing those whose wishes and values diverge from the patient’s. Advance directives are also helpful to focus end-of-life treatment decisions on what the patient would have wanted. Advance directives are effective in designating a surrogate, but are less effective in turning values into action, which require medical orders.

Advance directives alone may not be enough to ensure that patients’ wishes regarding life-sustaining treatment will be followed (9). In technologically developed countries, decisions about whether to use cardiopulmonary resuscitation, ventilators, antibiotics, and feeding tubes are an integral part of the care of women with advanced gynecologic cancers. Most persons with advanced stages of cancer and other chronic diseases ultimately make a decision to forego some or all forms of life-sustaining treatment (10).

When a thoughtful decision has been made to forego attempts at resuscitation, it is most distressing to family members to have the patient receive unwanted cardiopulmonary resuscitation (CPR). Emergency medical personnel who might inadvertently be called need written medical orders indicating not to provide specific treatments like CPR. Without such medical orders, emergency medical services (EMS) in most circumstances must attempt all possible life-sustaining treatments. As patients move from one place to another, a uniform set of medical orders is needed to ensure the patient’s wishes will be followed no matter what their setting of care.

One of the best-studied systems of medical orders to document advance care planning is the Physician Orders for Life-Sustaining Treatment (POLST) paradigm. The POLST document converts patient treatment preferences into written medical orders based on a conversation among health care professionals, the patient, and/or surrogates about treatment goals. The form transfers with the patient across care settings to ensure that her wishes are honored no matter whether she is cared for at home, in an ambulance, in a nursing home, or in an acute care hospital. A 1-year prospective study of nursing home residents with advanced chronic illness followed residents whose POLST forms included a Do Not Resuscitate (DNR) order and an order for comfort measures only. None of these patients received unwanted intensive care, ventilator support, or CPR (11). If such laws do not exist, health professionals taking care of women at the end of life should make sure that a patient’s wishes, and the support needed through orders, are present to avoid the harm that unwanted resuscitation and increased pain can have not only for the woman but for her family. In some legal systems, assuring that a DNR order is in place is solely the responsibility of health care professionals (9).

In summary, women beginning cancer treatment often benefit from the first stage of advance care planning, which is formally appointing a surrogate decision maker who shares their wishes and values. Women with advanced cancer who wish to forego one or more forms of life-sustaining treatment need to have carefully constructed guidance to all other health care personnel who currently are, or in the future may, provide care in life’s last chapter.

ROLE OF FAMILY IN DECISION MAKING

Given the sequential and repetitive conversations described above, the inclusion of family and other support persons is almost universal and clearly preferable. They are the ones with whom they have shared their lives and their dearest and deepest emotions. They are also the ones by whom the patient wants to be surrounded during the last phase of life. However, patients have the right to exclude or include whomever they wish in conversations. Also, medical events can be so acute that they trigger discussions with family members who struggle to consider what the patient might have wanted in that circumstance and are not clear about who should be the surrogate decision maker. Expert communication with the family then becomes the most critical means of ensuring that the patient’s best interests are served and there is a broad understanding of issues. Unfortunately, research suggests that communication is not as effective as it needs to be in these circumstances. Fifty-four percent of families did not understand basic information (diagnosis, prognosis, treatments) after physician–family meetings in an intensive care setting (12). Well-structured discussions that focus on the disease process, the goals of palliative therapy, and understanding the concerns of family members may help the family formulate decisions that support the dignity and choices that the patient herself would have chosen rather than create conflict that might impede an appropriate transition to a palliative care setting.

To prevent problems, it is wise to respect several rules:

   Always ask the patient if she has preferences in her contacts with friends and relatives. Often, she wants some to be more involved than others.

   Ask the patient for one or two contact persons to organize the communication with family and friends. Communicate essential information always in the presence of them or of other relatives or friends chosen by the patient. In this way, the family and friends are well informed and they share the same information with the patient. They can also help the patient with the evaluation of the information afterward.

   Never talk about the patient “behind her back.” If the physician does, and the patient finds out, the relationship of trust may be damaged. This also prevents the patient from thinking that the family knows more than she does, which can also strain family relationships.

   In countries where health care is a financial burden to the family, it is the responsibility of the physician to ensure that decision making concerning continuation of treatment or starting a treatment is based on the ethical principles of autonomy, beneficence, and nonmaleficence. This decision making cannot be driven by financial pressures alone.

At the end of life, we should expect emotional reactions by patients and their significant others, including family, to the inevitable lifestyle changes and impending death. It is the task of the physician to try to understand the thoughts and behavior of the patient regarding death and dying as well as her worries and unfinished emotional and personal business. Information on the patient’s values and beliefs should be sought—even information about very personal matters like relationships and sexuality might be needed for a proper assessment of her situation. The patient has to know that she has the choice to what extent she wants to participate in this assessment and what information she is comfortable to share both with the physician or care team and with her family and friends. In normal life, family and friends often meet the spiritual and emotional needs of a patient, and the role of the health team is to support these relationships, not replace them.

UNIQUE CIRCUMSTANCES FOR DECISION MAKING: PREGNANT PATIENTS

A rare but particularly poignant context of end-of-life decision making is that of the pregnant woman with terminal, metastatic cancer in whom survival even to the end of a normal pregnancy is unlikely. The choices for palliative therapy (e.g., pelvic radiation for control of local bleeding and pressure) may directly impact the likelihood of fetal survival. The possibility of delivery of a newborn—particularly prematurely, with all the known consequences—without surviving to mother the child make the choices even more complex for the woman and for her family. The primary objective of care for the mother is alleviation of symptoms, and choices to decrease the level of palliative care in order to facilitate fetal development are considered only if strongly desired and pursued by the patient herself. Maternal autonomy in making such decisions has been supported by both ethical and legal reviews, and these confirm that the patient’s wishes must guide decisions regarding fetal well-being, including accepting or not accepting caesarian section in the terminal phase of disease (13). There is no role for withholding appropriate and maximal pain and symptom management simply because the woman has a concurrent pregnancy (14). The health care team carries a special burden in this circumstance to ensure that the patient has all relevant and unbiased information regarding harms and benefits of all potential options for her and for her pregnancy to maximize her ability to make choices.

In the absence of the ability of the patient to speak for herself, the surrogate decision makers must weigh not only her wishes regarding a premature timing of delivery of the child and invasion with a surgical procedure such as a caesarian section, but also the likelihood of viability of the fetus and the supporting family’s ability and choices about caring for a significantly premature child. End-of-life directives for pregnant patients that specifically address these issues may be helpful in discussion and documentation of patient wishes (15). Some states and countries have special restrictions regarding decision making about limiting life-sustaining treatment in terminally ill pregnant women. In fact, 27 states have statutes that prohibit an out-of-hospital DNR order being written on a pregnant woman or otherwise have a “pregnancy limitation.” The role of the woman in fixing the parameters for balancing fetal well-being against her own needs for alleviation of pain and suffering is central, both if she can speak for herself or in representing those wishes by surrogate decision makers, and it is her health care team’s obligation to make sure that these are well defined and described, and that the patient’s wishes about her health care (autonomy) are respected in the terminal phase of her disease.

MAXIMIZING PALLIATION: MANAGING SYMPTOMS AT THE END OF LIFE

Pain Control

The mainstay of pain management at any time, and even more critically at the end of life, is a careful assessment of the type of pain (nociceptive [somatic or visceral] or neuropathic), the level of pain, the location and referral of pain, and the actions that increase or decrease the pain. This assessment will need multiple repetitions, as the kind of pain and level are likely to shift with disease progression. The management strategy is to focus on not only the type and level of pain but also the level of consciousness and the level of activity that a woman would like to achieve, and to vary the strategies accordingly (16).

The World Health Organization’s three-tier pain ladder starts with low to increasing levels of pain and increasing strategies from acetaminophen and nonsteroidal anti-inflammatory drugs to opioid analgesics for more severe pain. Usually using a 1 to 10 (10 being the worst) pain scale with patients fixes their position on the tier and suggests management strategy, although combinations from other levels or even within a level may be required for maximal comfort. Given that many gynecologic oncology patients are in the severe category (7 to 10) at the end of life, this might even include combining two opioids, particularly during pain syndrome escalation, to control the pain (17). In the course of adequate pain or symptom management, the anticipated side effects may shorten the time to death (principle of double effect), which is a well-recognized consequence of ensuring maximal palliative care. The ethical obligation is to alleviate the pain and suffering first and foremost, and the side effects that cannot be avoided (which rarely impact time to death simply because most patients already have substantial tolerance for narcotic pain medications) are accepted in this setting.

Pharmacologic measures to address severe nociceptive pain usually require narcotic medications. The options for administration are broad and can be tailored to the patient’s circumstances. As a principle that supports patient and family control, a means such as dermal, oral (pill, sublingual, and liquid), buccal, or rectal can make administration easier. With long-term sustained-release formulations, the goal is to establish adequate base pain coverage with every 8- to 12-hour dosing and adequate breakthrough pain coverage with an immediate-release form of the agents. Having the patient or caregiver track the amount and timing of breakthrough dosing can help adjust both the base and rescue dosing and timing. Sustained-release formulations are available for morphine, oxycodone, and hydromorphone (18). Drug dose escalation may be necessary to provide adequate pain relief. A modified release narcotic may take 2 to 3 days to reach a therapeutic level, whereas a transdermal patch may take 3 to 6 days, and a long acting medicine, like methadone, may take 5 to 6 days or longer (19). While titrating a long-acting narcotic, an intravenous (IV) formulation may be required to control pain. Once a patient’s pain is more adequately controlled with a sustained-release formulation, the short-acting agent should account for only 5% to 15% of the total drug dose.

Methadone is another option that has been utilized and can be particularly helpful when morphine is not adequate or is problematic for individual patients. It has a variable conversion ratio and pharmacokinetics in individuals and is much more difficult to titrate and manage with other medications. Methadone may be particularly helpful in patients in whom neuropathic pain resistant to opiods is a significant part of the pain complex. The final option to avoid IV pain management is the transdermal fentanyl patch. This is particularly helpful, when stabilized dosing is reached, for individuals who require continuous baseline coverage, and it is useful in patients who are unable to tolerate oral medications. The downside is the long time to peak concentrations, which renders initial evaluation and titration of dosages difficult and requires significant use of short-term release opioids to cover (20).

The use of IV forms of opioids with patient-controlled analgesia pumps can be accomplished in the home setting if needed. The use of a basal IV infusion with patient-controlled boluses is the preferred method. Subcutaneous infusion of opioids can also be used, with near equivalent efficacy, and it can be more easily managed in some home settings (21). For all methods, titration to comfort is the goal, with management of secondary symptoms such as nausea or sleepiness as needed by the patient and within the context of her goals. The last few hours of life can be marked by both increased or decreased pain, and a strategy to address this needs to be in place as part of the pain control program.

The multiplicity of tumor-related and therapy-related pain syndromes that result in neuropathies from invasion, radiation, or other therapy often require additional therapies (21). Anticonvulsants are helpful adjuncts. The most commonly used would be gabapentin, which is routinely used for postherpetic, chemotherapy-related, and diabetic neuropathies. Clonazepam and lamotrigine also may be effective. Additionally, tricyclic antidepressants (e.g., amitriptyline, desipramine, nortriptyline) may be helpful for neuropathic pain as they also treat underlying depression, have their own direct analgesic effect, and potentiate opioid analgesia (23). α 2-Adrenergic agonists, usually clonidine, have also been effective adjuvant analgesics for neuropathic pain as they work through a different receptor from opioids and decrease sympathetic outflow related to neuropathic pain. Corticosteroids such as dexamethasone can provide additional relief, particularly for inflammation irritating nerve endings. For the extensive pelvic tumor invading the nerves in the pelvic floor, the use of sacral or regional nerve blocks, epidural pumps, or ablation of nerve roots may play an important role (24). Implantated programmable pumps for neuroaxial infusion have been shown to increase analgesia and decrease the side effects that systemic opiods cause (43, 44). Topical lidocaine patches may also provide local relief.

Mind–body and psychological approaches, such as guided imagery, have been explored for dealing with cancer pain. These have been shown to decrease pain and positively affect symptoms (25,26).

Pain control in some settings, usually low-resource settings, may be restricted because of laws that do not allow patients access to adequate narcotic pain medicine at home or lack of access throughout the country. For these settings, use of any available pain medications along with local therapy (heat, cold, massage) may help add comfort. The fact that many women globally still die in severe pain argues for significant advocacy by the gynecologic oncology community, both in the U.S. and globally, with ministries of health and other relevant settings to address end-of-life pain control.

Nausea, Vomiting, and Bowel Obstruction

The etiology of nausea and vomiting at the end of life can be related to medications, constipation, or brain metastases, but more often for gynecologic cancer patients, it is related to malignant bowel obstruction. Ovarian and endometrial cancers have a higher likelihood of developing obstruction with disease progression, and therefore for effective symptoms management, it is critical to distinguish the obstructed bowel from constipation/obstipation.

Since the majority of patients are already at risk for constipation from medications (opioids, serotonin antagonist antiemetics) as well as low motility of the bowel, initiation and continued reevaluation of a bowel regimen is an important part of effective palliation. If patients can achieve adequate oral intake, the addition of fiber in the diet and fluid intake can assist in management. Patients maintained on an opiate regimen should take a baseline bowel regimen including a stimulant, such as senna, with or without a stool softener (27,28). If constipation persists, rectal laxatives or osmotic agents such as the polyphenolic compounds can be added as needed, and higher doses of these may be required to resolve constipation (29). There is insufficient evidence to recommend one laxative over another, and therefore, choice of laxative must be based on interpretation of individual contributing factors and side effect profiles. Metoclopromide is FDA-approved for treatment of cancer-related and opioid-induced constipation and may be helpful as a prokinetic agent in patients with gastroparesis or carcinomatous ileus. Other strategies that have been investigated with varying success include opioid rotation and administration of naloxone (30,31)

If obstruction is diagnosed, the consideration for surgery should be limited to women in good overall condition with a single site of obstruction. Various studies have shown that advanced age, poor general health or nutritional status, diffuse peritoneal carcinomatosis, ascites, palpable masses, prior radiation, and multiple obstructed sites carry a higher mortality and morbidity (32,33). Careful selection of patients has been shown to not only achieve symptom relief but also extend survival. Preoperative evaluation to ascertain the likelihood of benefit for individual patients should include imaging procedures (Gastro-grafin enemas, carefully chosen use of computed tomograph, and limited upper gastrointestinal evaluations) that target the identification of multiple sites of obstruction. Failure to identify the presence of multiple sites of obstruction can lead to a failed procedure, with consequent increase in the pain and discomfort of the patient during the dying phase.

Symptom management is usually the path taken in the treatment of bowel obstruction at the end of life. Nasogastric tubes provide quick relief, but are uncomfortable, and some reports suggest that they are ineffective in longer-term management of symptoms. Gastrostomy tubes are generally more comfortable, but require surgical or radiology intervention that may not be desired by patients. For individuals with partial and complete bowel obstruction and carcinomatous ileus, gastrostomy tubes can provide significant symptom management with just dependent drainage or suction (34). Left-sided omental caking may present an obstacle to successful percutaneous gastrostomy tube placement, and intraluminal stenting for relief of gastroduodenal obstruction can be useful in advanced ovarian cancer patients with significant upper abdominal disease (35). Distal rectal obstruction can sometimes be relieved through the use of stent placement (36,37), which is worth consideration. At the end of life, when the major objectives of treatment are quality of life and relief from the symptoms of obstruction, successful stent placement can avoid the need for laparotomy, with the associated significant morbidity and mortality that is seen in this group of patients. Stents are easiest to use in left-sided colon lesions and should be placed proximal to the dentate line for comfort (38). For many patients at the end of life, pharmacologic agents will be the mainstay of treatment for nausea and vomiting associated with bowel obstruction and carcinomatous ileus. Dopamine antagonists such as haloperidol are the drugs of choice for antiemetics in the setting of bowel obstruction (39). These agents can be associated with significant extrapyramidal side effects, which can be lessened or prevented with antihistamines, such as diphenhydramine, which also has antiemetic effects in patients with bowel obstruction. Additionally, hysocine hydrobromide and hysocine butyl bromide are helpful with patients experiencing colicky pain and nausea. Somatostatin analogues such as octreotide are well tolerated and can limit the secretions and motility and diminish the symptoms of bowel obstruction overall (40,41). Octreotide has the option of a long-acting formulation, which only requires monthly injections, and has been shown to assist in the setting of advanced ovarian cancer (42). Hospice Pharmacia formulates a suppository with Lorazepam, diphenhydramine, and Haloperidol that can be inserted rectally for easy drug delivery in outpatients with vomiting (43).

Finally, corticosteroids have been used to reduce inflammation, and potentially reduce the obstruction in some patients, and are worth a trial as part of conservative management. Dexamethasone in a dose of 4 to 8 mg per day for 3 to 5 days is generally used for this and additionally can potentiate the effect of pain medications and antiemetics (44). Improvement should be noted within 3 to 5 days if these agents are effective.

Several alternative treatments are worthy of mention, although not much data on their use exist in palliative care. Acupuncture, which has been found to be helpful in alleviating chemotherapy-related nausea, was associated only with transient improvement in symptoms of nausea in a small pilot of patients receiving palliative care for cancer (45). Medicinal cannabis has also been found to help with chemotherapy-related nausea. Dronabinol is a purified synthetic cannabinoid (delta 9 tetrahydrocannabinol) that is approved for use in the treatment of nausea and vomiting associated with chemotherapy; however, its side effect profile, including vertigo, xerostomia, dysphoria, and hypotension, is unacceptable to many patients (46).

Nutritional Management at End of Life

The value of end-of-life nutritional maintenance must be firmly set in the context of what the goal of the intervention is. Giving artificial nutrition in the dying phase of life has generally not been shown to prolong survival, adds risk, and may not enhance comfort (47). Families and patients can have cultural- and religious-based concerns about not providing “nutrition” in the dying phase, often framed as “starving” the patient, which need to be addressed. It is very important to explain to patients and families that hunger and thirst are symptoms that are experienced differently in healthy people from those who are seriously ill. Those who are near death are often comfortable even though they are no longer eating.

Stay updated, free articles. Join our Telegram channel

Join