By NEIL MACDONALD
Science creates new ethical issues, but it also leads to a re-evaluation of existing standards. Offering food and drink to the dying is intimately tied with compassionate care. Today, we can extend this offer through the employment of technical means of sustenance. Under commonly encountered circumstances, we now know that forcing nutrition on patients with advanced cancer may actually harm them, while controversy swirls around the hydration of patients in their final days.
One should distinguish matters of principle from matters of fact, for both are involved and related in the ethics of, hydration and nutrition. Controversies relating to matters of fact require evidence for their resolution, and that evidence comes from observation and research. The principles governing clinical ' decisions about hydration and nutrition relate to the rights of patients to make their own informed choices and refusals about medical treatment. In the case of Robert Corbeil, the judge of the Quebec Superior Court declared that Mr. Corbeil had the right to refuse to eat and to be fed, even if he died as a result. The critical issue was whether the patient was sufficiently lucid and balanced of mind to make a decision on his own behalf. That others may have found his decision unreasonable was irrelevant.
It is also a matter of principle that doctors are not obliged by law or by ethics to begin or continue with treatments that are therapeutically useless and not in the patient's best interests. This principle comes particularly into play when patients cannot make decisions on their own behalf. Whether and in what conditions hydration and nutrition are therapeutically useless, or even possibly harmful, is largely a matter of fact. a matter to be settled by observation. So clinical ethics requires a judicious combination of both principles and facts to arrive at clinical decisions that respect both the needs of a patient's body and the choices of the patient's mind.
In this chapter we explore ethical issues related to hydration and nutrition in patients with advanced chronic illness. The model considered is the patient with advanced cancer, but many of the principles are relevant to patients with other advanced chronic, ultimately fatal disorders. The central theme of the chapter is the impact of scientific studies on ethics, with consequent need to maintain basic principles of morality and to constantly review the interpretation of these principles in light of one's current understanding of biology.
Table 8.1 1 outlines the ethical issues that arise at the end of life; a number of them encompass questions on nutrition and hydration:
1. Resource allocation. The patient who cannot eat requires relatively expensive enteral or parenteral therapy in order to maintain calorie intake. Is this a justified expense?
2. Prolonging futile therapy. in the final scene of King Lear, one of his attendants says about the dying Lear: "Wex not his ghost-Oh let him pass; he hates him that would upon the rack of this tough world stretch him out longer." 2 Meaningful prolongation of life is unlikely to occur with enteral or parenteral nutrition and hydration in patients with advanced cancer. If this is the case, is it morally acceptable to maintain these therapies?
3. Withholding or withdrawing or therapy. Perhaps the prolongation of life achieved with nutrition and hydration is trivial, but will we not die if these are removed? What is the difference between these acts and active euthanasia?
Current ethical thinking categorizes technical sustaining of hydration or nutrition as medical procedures. As such, decisions to forego these interventions should carry the same ethical weight as decisions to withdraw transfusion therapy or dialysis. Ethics and law are not always bedfellows, but the modem views of ethicists and judges on this point appear to be consonant.
Although ethicists, palliative care physicians, and nurses may share this view, it is not widespread among broader groups of North American physicians. A sizable minority of physicians regard maintenance of hydration and nutrition as essential features of medical care and as somehow distinct from other technical processes such as transfusion or dialysis.3,4
Part of this dilemma may relate to the maintenance of the concept of -ordinary" and -extraordinary' care as factors in decision making. Rather than establishing a -cut-off linerelated to the complexity of therapy or its weight as a cultural totem, the emphasis should be on patient-family choice or the rational assessment of therapeutic benefit. The terms "ordinaryand "extraordinary" were frequently used in Catholic ethical discussions; they are now commonly replaced by the more apt terms -proportionate" and "disproportionate."
Against this background of current thought, I hold the view that decisions. to maintain hydration and nutrition will depend upon the informed decision of the patient or the patient's surrogates. Presumably, their decisions must be"' influenced by the advice they receive from medical attendants, advice which stems from knowledge of pathophysiologic abnormalities and their effects on patient well-being.
Table 8.1. Ethical issues in palliative care
Resource allocation
Can we afford to maintain our values? Is there evidence that the increasing lack of resources may cause selection of second-rate health care options with consequent compromise of patient care?
Life-prolonging treatment
Issues related to the costs (human and fiscal) of using aggressive therapies in situations in which the risk of adverse effects is high and beneficial outcome problematic. Do Not Resuscitate (DNR) regulations. Are they often irrelevant? Is there a risk that health professionals may conclude that DNR discussions satisfy the requirement for discussing death and dying issues with the patient and family?
Problem priority
Do we often ignore pain or other aspects of suffering while addressing more trivial elements of illness?
Pain and symptom control
Is not relief of relievable suffering a fundamental patient-family right?
Communication, consent, and competence
Do we ensure that all patients have access to the most competent care? Do patients understand relative levels of health care competence and are they in a position to make a decision based on this information? Do patients have the necessary information for making logical choices? Do they fully comprehend the information we provide? Do we sometimes use relentless therapeutic exercises as a substitute for conducting difficult supportive conversations?
Confidentiality
How do we exchange privileged information within a team while preserving patient-family rights and dignity?
Research
In situations in which cognitive failure is common, do we ensure that patients are competent to give informed consent? Does participation in a research study alter the covenant of the physician to look after a patient? What are the responsibilities for continuing care when the study is over? Are some studies by their very nature impossible to carry out in a fragile population of dying patients and their families? Are principles relevant to informed consent in research studies applicable to obtaining informed consent in the nonresearch setting? Sponsorship of research: does private research support drive the public research agenda, and subsequently influence resource allocation, fueling futile efforts with increased cost in suffering and waste of resources?
Education
The principles of palliative care are fundamental components of medical practice. Does our current medical educational system recognize and reflect this tenet?
Issues related to withholding or withdrawal of therapy
Do we, and do our patients and families understand the difference between these acts and physician-assisted suicide or euthanasia?
Palliative care standards
Should palliative care programs be expected to demonstrate a basic standard of care and competence so as not to delude the community, damage the credibility of palliative care, and waste scarce public and private support?
Media influences
Public views of medical care are shaped by media priorities and the accuracy and balance of reportage. Is the public properly informed on medical issues?
Cultural issues
Truth-telling is only one example of cultural differences in the management of dying patients. How do we respect cultures yet maintain a consistent ethical approach?
Dual ethical standards
Many patients are in their current situation because they are influenced by the blandishments of advertisements for lethal products. Why should business people hold different ethical standards than health professionals? Why should one group regard it as ethical to do whatever is legally allowed for profit when health professional groups regard the promotion of illness for gain as a fundamentally amoral activity?
Responsibilities of health professionals to:
Patient
Patient and family
Institution
Society
Own family
Ethical resolution of conflicting demands
Euthanasia and physician-assisted suicide
Source: Adapted and reprinted with permission from MacDonald N. From the front lines. J Palliat Care 1994; 10(3):44--47.
Table 8.2. Arguments for and against hydration
Arguments against
1. Comatose patients do not experience pain, thirst, etc.
2. Fluid may prolong the dying process.
3. Less urine output means less need for bed pan, urinal, commode, or catheter.
4. Less gastrointestinal fluid and less vomiting.
5. Less pulmonary secretions and less cough, choking, and congestion.
6. Minimize edema and ascites.
7. Decreased fluids and electrolyte imbalance act as natural anesthetics for the central nervous system with decreased levels of consciousness and decreased suffering.
Arguments for
1. Dying patients are more comfortable if they receive adequate hydration.
2. There is no evidence that fluids alone prolong life to any meaningful decree.
3. Dehydration and electrolyte imbalance can cause confusion, restlessness, and neuromuscular irritability.
4. Water is administered to dying people who complain of thirst, so why not give parenteral hydration.
5. Arguments regarding poor quality of life detract from efforts to find ways to improve comfort and life quality.
6. Parenteral hydration is the minimum standard of care and discontinuing this treatment is to break a bond with the patient.
7. Withholding fluid to the dying patient sets a precedent for withholding therapies to other compromised patient groups.
Decisions on Hydration
As stated in another forum,5 the arguments for and against hydration can be summarized as presented in Table 8.2. In looking at the scientific aspects of these arguments, one is struck by the paucity of evidence to sustain sometimes vigorously espoused views. Indeed, dehydration will result in decreased urine output, but the beneficial effects of fluid restriction on vomiting, pulmonary symptoms, and, particularly, edema and ascites, are not well established. Evidence relating fluid and electrolyte imbalance to improved pain control rests on animal work.
Conversely, the correlation between control of thirst and hydration status in patients at the end of life is modest.6 Well-hydrated patients can experience severe thirst, presumably because of the influence of oral factors such as mouth breathing and infection. Thirst can be controlled with small amounts of oral fluids, as can be attested by any of us who, thirsty after physical exertion, note rapid relief after our first few liquid gulps-relief occurring long before our fluid and electrolyte status could be appreciably corrected. This observation in healthy individuals is mirrored by the consensus views of most palliative care physicians and nurses who similarly note control of thirst with small amounts of fluids and maintenance of a clean mouth.
In the author's opinion, the most critical physiologic argument relating to the hydration controversy concerns mentation and the relationship between polypharmacy, so common in dying patients, and their hydration status. Fainsinger and Bruera believe that adoption of a more vigorous hydration posture in their unit in part is responsible for the diminished incidence of delirium that they have noted in recent years. 7 They are concerned that the benefits of reduced urine output in dehydrated patients is more than balanced by the risk of adverse drug reactions secondary to failure to clear the primary agent or toxic drug metabolites. Aligned to their view is the concern that the touted benefits of dehydration in the patient in the last days of life may be a specious argument if the patient's semicomatose or sedated status (secondary to the need to control delirium), now used to justify withholding fluids, was caused by earlier diminished fluid intake.
Changes in mental status are among the most devastating symptoms experienced by dying patients. While the degree of anguish felt by the delirious patient is hard to define in terms of suffering (patients who recover from delirious states usually do not remember the incident),' their behavior is a source of great distress to family members and may generate conflict between the family and similarly concerned health professionals. The waves of conflict often spread as nurses in immediate contact with the patient may feel that physicians are not taking sufficient steps to control the delirium. Is a laissez-faire attitude towards hydration correlated with an increased incidence of delirium in dying patients? The provocative observations of Bruera and Fainsinger 7,8 should be followed up with more definitive studies.
Cancer control programs must include four preventive strategies, starting with prevention of the disease, if possible; prevention of invasive disease (through early diagnostic programs); prevention of mortality in patients with invasive disease (anticancer therapy); and prevention of suffering.9 We now believe that we can control much of the suffering associated with pain through early identification and management of the problem. Similarly, if delirium and other central nervous system effects can be prevented through earlier consideration of hydration problems in association with drug therapy, given new scientific considerations, our current view on the ethics of hydration in patients with advanced illness would shift.
Pending further information, the author believes that the following guidelines for hydration are reasonable.
Patient choice*
1 . Competent patients can accept or refuse hydration when given pertinent information.
2. For incompetent patients, a balancing of advantages and disadvantages should be undertaken. The patient's interests are paramount in decisions. Incompetent patients may be represented by a legal guardian or health care agent.
*Modeled upon guidelines on enteral feeding published by the Bioethics Centre of the University of Alberta.
Decision-making process
1. Patients, families and friends, and caregivers should know that hydration can be ethically withheld and withdrawn.
2. Hydration may be initiated on a time-limited basis to allow an assessment of the advantages and disadvantages.
3. A time-limited trial of hydration is always recommended if there is doubt about the advantages or disadvantages of hydration.
4. After a decision has been made to initiate or stop hydration, the emotional impact of the decision on caregivers, family members, and others must be recognized.
5. The advantages and disadvantages of hydration for family members, close friends, caregivers, and society may influence the decisions made by competent patients or made for incompetent patients.
6. Unless there is a contraindication, patients who are not otherwise cognitively impaired and who are receiving pharmacologic agents that could cause delirium should receive adequate hydration to ensure reasonable drug metabolism and excretion.
Hypodermoclysis
Most articles discussing drawbacks of hydration in dying patients refer to the difficulties of maintaining hydration using intravenous techniques. Here is an example where recent clinical research skews the hydration argument somewhat in the direction of rehydration. Primarily because of the work of Schen and Singer-Edelstein,10 Pluck,11 Hays,12 and subsequently, Fainsinger, Bruera, and their colleagues in Edmonton,13 hypodermoclysis (subcutaneous infusion of fluids) is now widely used in palliative care programs. Clysis is not a new technique. Although subcutaneous clysis was once accepted, it fell into disuse because of the injudicious use of nonelectrolyte solutions (which are poorly absorbed when administered subcutaneously) and the improvement in intravenous techniques and fluid sources. We now know that, with the assistance of modest doses of hyaluronidase, solutions containing normal saline can be infused into patients in doses up to 120 ml/hour (small studies report patients tolerating boluses of even larger amounts) 14
The use of hypodermoclysis to hydrate patients has many advantages over the intravenous route. 13 These may be stated as follows:
1. The problem of difficult and painful intravenous access is obviated.
2. Physicians or specially skilled nurses are not required to commence a subcutaneous infusion.
3. Hospitalization is not necessary and clysis, either continuous or intermittent, can be maintained in a home setting for long periods of time.
4. Subcutaneous sites can normally be sustained for many days without change. Thrombosis is not a problem whereas local cellulitis is normally not a substantial problem in the absence of associated administration of drugs. Even in these situations, careful observation of the subcutaneous site and rotation of site will prevent major complications.
Nutrition"Dis-moi ce que tu manges, je te dirai ce que tu es.Brillat-Savarin, Physiologie du Gout
Poor appetite, weight loss, and alterations in bowel function (e.g., chronic nausea and constipation) are common problems in patients with advanced cancer. Ultimately, malnutrition may actually result in a patient's demise, possible accounting for 25% of patient deaths.15 While the defining qualities of the cachexia-anorexia syndrome remain imprecise, clinicians recognize that most weight loss in cancer patients, and associated metabolic abnormalities, is caused by a cascade of cytokines and/or tumor products generated as products of tumor-host interaction. Successful control of this problem will depend upon reversal of the aberrant metabolic pattern. In the interim, some patients may have temporary reversal of weight loss, and improvement in appetite following therapy with pharmacologic agents such as megestrol. Commonly employed enteral and parenteral nutritional aids may temporarily improve weight loss, but they do not significantly influence the overall course of illness. Unlike hydration, the use of enteral and parenteral feeding is associated with significant cost and the risk of adverse effects.
In contrast with hydration, less controversy surrounds decisions to withhold enteral or parenteral nutrition in the last days of life. At least this is the case among palliative care physicians who may hold varying views on the importance of hydration but tend to think as one with respect to technical approaches to feeding dying patients. These views are not uniform among physicians, as patients on general hospital wards may still die with an enteral tube in place or with parenteral feeding only recently discontinued. As discussed earlier in the section on hydration, physicians appear to rank order the technical means of delaying death and, like the community from which they arise, find great difficulty in withdrawing sustenance. Decisions on artificial nutrition can be a focal point for family disputes. Family members do not appreciate the biology of the cachexia-anorexia syndrome; they only know that a loved one is losing weight and strength, cannot eat, and appears to be starving to death.
Nevertheless, ethical issues concerning decisions on nutrition are more straightforward today than those involving hydration. There is no evidence that current techniques prevent asthenia or prolong life for cancer patients with the cachexia-anorexia syndrome. Three factors must, however, be kept in mind:
1. As easier techniques for hydration, such as the use of hypodermoclysis, influence the arguments for and against hydration, it may be that a clearer understanding of the cachexia-anorexia syndrome and the introduction of more specific nutritional aids will change current approaches.
2. Not all cancer patients who are losing weight and cannot eat have the cachexia-anorexia syndrome. Some are, indeed, starving to death and would benefit from enteral or parenteral nutrition. Often included in this group are patients with head and neck cancer, or cancer of the esophagus who cannot swallow, patients with malabsorption disorders, and patients with ovarian cancer and intermittent bowel obstruction. As the definition of the cachexia-anorexia syndrome is not exact, absolute criteria for identifying patients who will benefit from parenteral nutrition are not in place. When in doubt, it is reasonable to offer a time-limited trial of enteral or parenteral nutrition, keeping in mind that there is no ethical distinction between offering a therapy, with subsequent withdrawal, if it appears to be futile, and not starting a therapy in the first place.
3. Asthenia and cachexia are the major problems limiting patient mobility and creating dependence upon families and institutions. This symptom constellation should be selected for a task force research initiative. The basic science foundation must be expanded and integrated, but sufficient information is at hand to sustain a multidisciplinary research drive.16
Earlier in this chapter a decision process for hydration in patients with terminal illness was outlined (see Table 8.2). The first five points also apply to enteral and parenteral nutritional decisions. Adoption of a purely clinical analysis of nutritional benefit, while scientifically valid, may not fully satisfy our ethical obligations to patients and families. In every culture in the world through recorded time rituals surrounding food and drink are powerful components of community and family life. To be near a loved one in their last hours and to offer them sustenance remains a commonly held view of the obligation of family and friends. Like Brillat-Savarin, they may link a loved one's food intake with their fundamental human qualities.
It may be difficult for family and friends to address the existential distress of a loved one or to communicate fully and clearly at the end of life when patterns of communication were not well established earlier in relationships. Family members will need a gentle, careful explanation of why assisted nutrition is no longer necessary. The maintenance of technical links, if this substitutes for speaking to family members, is ethically questionable.
Few patients with advanced cancer complain of hunger towards the end of their lives.17 When present, it is normally readily alleviated with modest amounts of oral food intake. Therefore, anguish associated with unrelieved hungers is a rare event in dying patients. When present, it clearly must be relieved with whatever means are available, natural or artificial.
Although a technically difficult, potentially dangerous, and expensive artificial feeding strategy may be abjured, health professionals remain obligated to further, not to discourage, the symbolic links that patients, families. and friends share. Physical symbolic links are important in such circumstances; the benefits of a drink of water extend far beyond its effect on osmolarity, and the value of a morsel of food, beyond its caloric count. Therefore. decisions to withhold technical approaches to nourishing a patient must be accompanied by a set of practical approaches to maintain, as much as possible, natural means of intake of food and drink. These include the following.
1. Patients who may have difficulty with oral feeding because of neurologic impairment or partial obstruction may be able to take certain types of food frequently and in small amounts, when nursing support is available and when the feedings are skillfully administered. Withdrawal of artificial feeding, except in the very last days of life, creates an obligation to assess the possibility of small oral feedings and to-teach proper techniques for maintaining oral nutrition to responsible family members.
2. As with hydration, if the patient is sentient, the patient's desires, hold primacy. Food should never be forced upon patients who do not wish to be fed. Exceptions to this principle require justification.
3. Oral feeding is sometime difficult because of associated correctable gastrointestinal problems. Unresolved nausea, severe constipation, abdominal cramping, heartburn, depression, or a dirty mouth should not contribute to the inability to take oral nourishment. The skillful use of appetite
4. Hospital environments are often not conducive to communal repasts. Correctable factors include serving appetizing food that meets the patient's desires, served at the right temperature, in the right amount, in the most attractive way; removal of adverse cues, such as unpleasant odors; and addition of positive reinforcements. These may include music, a restful setting, and, if it was previously the patient's custom, the provision of small amounts of alcohol.
In conclusion, the use of assisted nutrition and hydration has to be judged in light of the current understanding of pathophysiology and the governing realizable clinical goals for this patient now. These goals will change as the patient's condition changes. It makes little clinical or ethical sense to go all the way with a treatment that made sense when started, but no longer has clinical meaning now.
Acknowledgments
1 am grateful to Dr. David J. Roy for his review of this chapter and his helpful comments. Sections of this chapter are adapted, with permission, from Indications and ethical considerations in the hydration of patients a4th advanced cancer. In: Bruera E. Higginson 1, eds. Cachexia-Anorexia Syndrome in Cancer Patients. New York: Oxford University Press. 1996.
References
1. MacDonald N. From the front lines. J Palliat Care 1994; 10(3):44-47.
2. Shakespeare W. King Lear Speech of Kent, Act V, Scene 111.
3. Solomon MZ, O'Donnell L, Jennings B, et al. Decisions near the end of life: professional views on life-sustaining treatments. Am J Public Health 1993; 83(l):14-23.
4. Christakis NA, Asch DA. Biases in how physicians choose to withdraw life support. Lancet 1993; 342:642-646.
5. MacDonald N, Fainsinger R. Indications and ethical considerations in the hydration of patients with advanced cancer. In: Bruera E, Higginson 1, eds. Cachexia-Anorexia Syndrome in Cancer Patients. New York: Oxford University Press, 1996.
6. Musgrave C, Bartal N, Opstad J. The sensation of thirst in dying patients receiving IV hydration. J Pallia Care 1995; 11(4):17-2 1.
7. Bruera E, Franco jj, Maltoni M, Watanabe S, Suarez-Almazor M. Changing pattern of agitated impaired mental status in patients with advanced cancer: association with cognitive monitoring, hydration, and opioid rotation. J Pain Symptom Manage 1995; 10(4):287-291.
8. Bruera E, Fainsinger RL, Miller MJ, Kuehn N. The assessment of pain intensity in patients with cognitive failure: a preliminary report. J Pain Symptom Manage 1992; 7(5)267-270.
9. MacDonald N. Palliative care-the fourth phase of cancer prevention. Cancer Detect Prevent 199; 15(3):253-255.
10. Schen RJ, Singer-Edelstein M. Subcutaneous infusions in the elderly. J Am Geriatr Soc 1981; 24:583-585.
11. Gluck SM. Advantages of hypodermocysis. J Am Geriatr Soc 1984; 32:691-692.
12. Hays H. Hypodermoclysis for symptom control in terminal cancer. Can Family Phys 1985; 31:1253-1256.
13. Fainsinger R, MacEachern T, Miller MJ, et al- The use of hypodermoclysis for rehydration in terminally ill cancer patients. J Pain Symptom Manage 1994; 9(5)298-302.
14. Bruera E, de Stoutz ND, Fainsinger RL, Spachynski K, Suarez-Almazor M, Hanson J. Comparison of two different concentrations of hyaluronidase in patients receiving one-hour infusions of hypodermoclysis. J Pain Symptom Manage 1995; 10(7):505-509.
15. Warren RS. The immediate causes of death in cancer. Am J Med Sci 1932: 184:610.
16. MacDonald N, Alexander R, Bruera E. Cachexia-anorexia-asthenia. J Pain Symptom Manage 1995; 10(2):151-155.
17. McCann RAM, Hall WJ, Groth-Juncker A. Comfort care for terminally ill patients. The appropriate use of nutrition and hydration. JAMA 1994; 272(16):1263-1266.
