Secondary Logo

Journal Logo

Original Article

End-of-life decision-making in the United States

Truog, R. D.*,†

Author Information
European Journal of Anaesthesiology: February 2008 - Volume 25 - Issue - p 43-50
doi: 10.1017/S0265021507003419
  • Free

Abstract

End-of-life decision-making in the United States

The beginnings of the ethical and legal debates about end-of-life decision-making can be traced to the late 1950s, with the emergence of mechanical ventilation as a life-sustaining therapy. Prior to that time, respiratory failure led to cardiac arrest and death, while severe neurological injury led to apnoea, followed by cardiac arrest and death. With mechanical ventilation, however, physicians were for the first time able to keep these patients alive.

French neurophysiologists were the first to recognize and discuss the ethical dilemma created by patients who had suffered severe and irreversible neurological injury but who could be kept alive for prolonged periods of time with mechanical ventilation. They labelled this condition as ‘coma dépasse,' or beyond coma [1]. Interest in this condition was mostly theoretical until December 1967, when Christian Barnard [2] performed the first heart transplant in Cape Town. Astute journalists who covered the event were quick to realize that the miraculous procedure raised at least one very thorny ethical issue. Was the donor ‘dead' at the time his beating heart was removed, and if not, did Barnard kill the patient by removing his heart?

Brain death

The import of these events was not lost on Henry Beecher, an anesthesiologist at Boston's Massachusetts General Hospital. Within months of Barnard's first transplant, he formed an ad hoc committee at Harvard Medical School to examine the ethical and legal status of patients with severe brain injury who are sustained on mechanical ventilation. Beecher identified two issues that needed to be addressed. First was the concern that hospitals would become filled with patients on ventilators who had no hope of recovery, and second was the realization that the emerging science of organ transplantation could not develop without an ethical and legal supply of transplantable organs [3].

The Harvard Committee proposed a set of criteria that (1) could be established at the bedside by clinical examination without complex or expensive testing, and that (2) seemed to be predictive of irreversible coma. The primary accomplishment of the committee was therefore to advance a scientific claim - criteria for the diagnosis of irreversible coma - and indeed these criteria have stood the test of time. No patient who has met these criteria has ever recovered consciousness [4].

Beecher realized, however, that the Harvard Committee could not stop there. In addition, it was necessary for the Committee to make the claim that ‘irreversible coma' was equivalent to ‘death.' Only by diagnosing these patients as dead could they (1) be removed from the ventilator, or (2) have their organs removed for transplantation, without the physician being liable for having ‘killed' the patient. In order to meet these legal and ethical needs, Beecher's criteria for establishing ‘irreversible coma' were morphed into criteria for establishing death on the basis of neurological criteria, or ‘brain death.' The failure of the Beecher committee (or any subsequent apologists) to provide a convincing defense of the equivalence of ‘irreversible coma' and ‘death' has plagued the concept to this day [5-7]. Despite this fundamental and unresolved problem, ‘brain death' has nevertheless survived to form the legal basis for organ procurement and transplantation in the United States and throughout much of the world [8,9].

Withdrawal of mechanical ventilation

Karen Ann Quinlan was a 19-yr-old female who had suffered severe hypoxic injury following a drug overdose. After several years, her parents recognized that she was never going to ‘wake up,' and asked her clinicians to withdraw her ventilator and allow her to die. In the years that followed the Harvard report, the concept of brain death had become increasingly accepted in both law and clinical practice, so that when patients were diagnosed as brain-dead, physicians were free to either discontinue mechanical ventilation or to offer the family the option of organ donation.

But Karen Ann Quinlan did not meet the criteria for brain death. Her physicians therefore refused to withdraw her ventilator, arguing in court that doing so would violate the Hippocratic injunction against killing. In weighing the various ethical and legal issues at stake, the judge in this case decided that the rights of patients to refuse treatments that they no longer wanted should be given higher priority than physicians' concerns about killing. As such, he ordered the physicians to withdraw the ventilator from her, despite their objections that this was unethical (as an historical footnote, the ventilator was indeed withdrawn, but Karen did not die. To the surprise of her physicians she was in fact able to breathe on her own, and she survived for another 10 yr, being fed through a gastrostomy tube) [10].

This 1976 case was a watershed in end-of-life decision-making in the United States, since it gave the autonomy-based right to refuse unwanted medical treatment a higher priority than the concerns of physicians about violating long-standing medical traditions against the killing of patients. Over time, this tension was resolved only by general agreement to change the terminology associated with ventilator withdrawal. Instead of saying that physicians kill patients when mechanical ventilation is withdrawn, today we speak of physicians ‘allowing patients to die.' This subtle but important shift in terminology has allowed ventilator withdrawal to proceed in cases like that of Karen Ann Quinlan, but with the moral burden of the patient's death placed squarely on the shoulders of the patient's disease rather than on those of the physician [11]. Today the majority of deaths in both paediatric and adult ICUs follow the withdrawal of life-sustaining treatment, a practice that was prohibited only a few years ago [12,13].

This linguistic ‘slight of hand' has not been entirely successful, however. American bioethicist Dan Brock [14] has argued that simply using different language does not necessarily change the essential features of the situation. In Brock's view, an act should be described as a killing if it is the proximate cause of the patient's death, and Quinlan's physicians were therefore correct in their view that when ventilator withdrawal leads directly to death, the physician has killed the patient. According to Brock [14], however, just because an act is a killing does not necessarily make it wrong. Even though ventilator withdrawal is a killing, it can be ethically justified when performed with the consent of the patient or surrogate in situations with a poor prognosis and when carried out by someone in a societally sanctioned role (e.g. a physician).

The Great American Consensus

The Quinlan case and many others like it over the following years have established a strong and consistent principle in American law and ethics: ‘Patients have a virtually unlimited right to refuse any unwanted medical treatment, even if necessary for life itself.'

This right, firmly rooted in Constitutional guarantees of liberty and privacy, has consistently taken precedence over other concerns of the medical profession, such as physicians' obligations of beneficence to patients. Perhaps the most dramatic example of this principle can be seen in the American approach to patients who are Jehovah's Witnesses [15]. Courts have consistently affirmed the right of competent adults to refuse blood transfusions, even when the transfusion would clearly be life-saving. While physicians who disagree may attempt to transfer these patients to other physicians, in an emergency, physicians are prohibited from abandoning the patient and are required to abide by the patient's wishes, regardless of the physician's personal beliefs. The only exceptions to this have been in the case of immature minors who cannot make autonomous choices for themselves, and in situations where the patient has parental obligations such that the death of the parent would leave one or more children without care and protection.

In this context it is interesting to note that France, which so closely followed the United States in developing a Constitution based on the principles of liberty and equality, has taken a different approach toward patients who are Jehovah's Witnesses. Although generally affirming the importance of informed consent in medical treatment, the French nevertheless frequently permit the transfusion of Jehovah's Witness patients against their will, citing the beneficence-based obligations of the medical profession [16].

Who decides: the question of surrogate decision-makers

An interesting and important difference in end-of-life decision-making between the United States and Europe has been in the role of surrogate decision-makers. The logic in the United States has run as follows: (1) patients have the right to decide for themselves; (2) when they cannot decide for themselves, someone must decide for them; and (3) this ‘someone' should ideally be the individual specifically designated by the patient to play this role, or in the absence of that, the person with the strongest genetic and/or emotional ties to the patient. In accord with this approach, federal guidelines require hospitals to ask all patients at the time of admission whether they would like to fill out an ‘advance directive,' naming the person whom they would like to make decisions on their behalf if they should become incapable of making decisions for themselves [17,18].

Until recently, this issue has been handled differently in Europe. I recall a conference several years ago where someone in the audience asked a prominent Italian researcher about how he had obtained informed consent for research from his patients, all of whom were intubated and on a ventilator. ‘Did you ask one of the family members for permission to enrol the patient in research,' the questioner asked. ‘Of course not,' was the reply. ‘I would never ask a family member for permission to enrol a patient in research. They do not understand the complexities of research and you cannot trust their motives. As the patient's physician, I was the one who decided whether it was best for a patient to be a part of the research study.'

This comment illustrates what I believe has been a common unwillingness among European physicians to involve families as surrogates for decision-making. In addition to concerns about their capacity for understanding and perhaps questions about their motives, European research has shown that family members of ICU patients often have symptoms of anxiety and depression that may impair their ability to act as competent decision-makers for their loved ones [19].

Despite these differences, practices in Europe and North America seem to be converging on the issue of surrogate decision-making. Recent guidelines from the European Union emphasize the role of surrogates in obtaining informed consent for the treatment of incompetent patients (particularly in the context of medical research) [20]. Furthermore, North American clinicians are increasingly recognizing that good end-of-life decision-making cannot depend upon simply following the commands of the surrogate, it rather requires firm guidance from the physician, advising the surrogate about the suitability of the medical options and developing agreement with the surrogate about decisions that are in the patient's best interest [21]. As such, the differences in practice between North America and Europe have probably been exaggerated beyond the reality of the situation, and over time even these differences are likely to become less distinct.

Does a right to refuse imply a right to demand?

As noted, through the 1970s and 1980s Americans acquired a virtually unlimited legal right to refuse any unwanted medical therapy. Given the emphasis that Americans place upon individualism and autonomy, it is not surprising that during the 1990s and the current decade this debate has turned to the question of whether Americans have the right to demand medical therapies that they desire but that their physicians believe are unwarranted.

These issues, which have not been resolved, were illustrated nicely by the case of Catherine Gilgunn, a 72-yr-old female who was admitted to Boston's Massachusetts's General Hospital after she fell and broke her hip [22,23]. Before her hip surgery could be performed she developed seizures that could not be controlled, eventually leaving her comatose, responsive only to painful stimuli. Her attending physician recommended a Do-Not-Resuscitate (DNR) order and comfort care, but Gilgunn's daughter disagreed, saying ‘Mother wanted everything possible to save her life regardless of cost.' The DNR order was not entered, and she remained with ‘full code' status.

European physicians may see this case as a striking example of ‘autonomy gone amuck,' and wonder how American physicians could ever be so misguided as to let a family member play such a dominant role in decision-making for a patient. In reality, however, most American physicians look at this case in exactly the same way, and recognize that this case is a horrible example of how our reliance on family members as surrogate decision-makers can lead to terrible outcomes for patients. In this case, the physicians involved finally did take a strong and legally courageous stand.

After 1 month, the ICU team requested an ethics consult. At a meeting with the daughter, the attending physician stated that Gilgunn's condition was ‘hopeless from a medical point of view' and that further interventions were futile. The daughter disagreed, and ended up walking out of the meeting. The ethics consultant wrote in the chart that cardiopulmonary resuscitation (CPR) was ‘contraindicated, inhumane and unethical. The order protecting (the patient) from this has been entered and should not be removed.' Although the family refused to participate in any further meetings with the medical team, shortly thereafter she had a cardiac arrest and was pronounced dead without an attempt at resuscitation.

Gilgunn's daughter sued the attending physician and the ethics consultant. Although the jury believed that, if Gilgunn had been competent, she probably would have wanted to have CPR performed, the jury concluded that this would have been ‘futile' and that the physicians had no obligation to provide it.

Unfortunately, cases like this are relatively rare. Most of the time, physicians find it easier to simply abide by the family's wishes and not to fight them. In Texas and a few other states, however, new legislation has been enacted that makes it easier for physicians to challenge the demands of families for care that the clinicians believe is inappropriate or futile [24,25]. Again, in this sense the clinical practice patterns of Europe and North America may be converging within the context of their respective cultural traditions.

Are tube feedings a ‘medical treatment'?

While the Great American Consensus has consistently affirmed the rights of patients to refuse treatments like mechanical ventilation, cardiac assist devices and haemodialysis, the ‘Consensus' is much more controversial when it comes to discontinuation of tube feedings. In American law, the landmark decision on this issue was the case of Nancy Cruzan, decided by the United States Supreme Court in 1990 [26,27]. Cruzan was a 25-yr-old female who was left in a persistent vegetative state (PVS) following an automobile accident. In essence, her clinical condition was parallel to that of Karen Ann Quinlan 14 yr earlier, with the legal question now being withdrawal of tube feedings instead of withdrawal of the ventilator.

To the surprise of many, the Supreme Court held that tube feedings are a medical treatment, and that decision-making for withdrawal of tube feedings should, in principle, be no different from that for withdrawal of other treatments, like mechanical ventilation. As Justice Sandra Day O'Connor wrote: ‘Requiring a competent adult to endure such procedures against her will burdens the patient's liberty, dignity and freedom to determine the course of her own treatment. Accordingly, the liberty guaranteed by the (constitution) must protect, if it protects anything, an individual's deeply personal decision to reject medical treatment, including the artificial delivery of food and water.'

Withdrawal of tube feedings therefore have the same legal status within the Great American Consensus as withdrawal of any other life-sustaining treatment [28]. In the clinical sphere, however, differences remain. For many clinicians, patients and families, tube feedings have enormous symbolic importance. Some hold that feeding of the sick and infirm is a fundamental human obligation, not the delivery of a ‘medical treatment.' Recently the Pope issued an opinion that, while open to interpretation, has been read as prohibiting the withdrawal of tube feedings [29].

Those who oppose the withdrawal of tube feedings often use the rhetoric that it is ‘starving the patient to death' and thereby just a slow form of euthanasia [30]. Actual clinical descriptions of the process show, however, that this is not the case [31-33]. Patients who undergo the withdrawal of tube feedings typically die over 2 or 3 weeks of dehydration, not starvation, and the process is associated with a slow build up of metabolites that contribute to progressive sedation, somnolence and finally a peaceful death. Indeed, some have argued that reduction of oral intake and dehydration may be a very common mode of death for those who do not die of acute organ failure, and so may be seen as one of the most ‘natural' ways to die [34].

What about oral feedings?

In my view, however, the question of tube feeds may not be quite as simple as this. In an article I wrote with neurologist Thos Cochrane, we argued that many patients like Nancy Cruzan are actually capable of coordinated swallowing and thus can be fed, at least theoretically, so long as one has a tremendous amount of patience and can spend the many hours each day required to keep patients like this nourished and hydrated [35-37]. In other words, many of these patients are fed through a gastrostomy tube more as a matter of convenience than because of a medical necessity.

In this regard, it is interesting that none of the judges who decided the Cruzan case ever asked whether she had the capacity to take nourishment and hydration orally. What if she had been capable, as seems to be the case for at least some patients in a PVS? Is it ethical, or legal, to simply stop the patient's tube feeds and not supply oral feeds? If so, then the justification cannot be based on the rights of patients and surrogates to refuse unwanted medical interventions, since oral feedings are clearly not medical interventions.

If we decide that it is permissible to withdraw tube feeds but not oral feeds, then perhaps every patient who undergoes withdrawal of tube feeds should be evaluated by a feeding therapist and have oral feeds continued if they are capable of swallowing. This evaluation is not typically performed, but what if it were and what if the patient were found to retain only a rudimentary capacity for swallowing that entailed a high risk of aspiration? Would it be ethical to feed such a patient knowing that the feedings would eventually cause the patient to develop pneumonia? These questions have, for the most part, been neither asked nor answered.

Current practice in the United States does not include evaluating patients for their capacity to take oral feeds before discontinuing tube feeds, but for those patients who could take oral feeds the justification cannot be the one given by the US Supreme Court, but must rest on another foundation. Cochrane and I argue that irrespective of whether feeds are administered artificially through a gastrostomy tube or naturally by the oral route, the same principles that govern other forms of life-sustaining treatment apply. Oral feeds may therefore also be withheld from patients, but only when a full and careful assessment shows that the patient is either incapable of taking oral feeds or that the burdens of the feeds would outweigh their benefits.

Why focus on withdrawal of treatment? Why not assist the dying process?

Throughout the history of this debate, the focus has always been upon which treatments may be withdrawn, and under what circumstances. As more than one person has observed, this is less compassionate than the way many people treat their pets. Once a decision is made to allow a pet to die, the veterinarian usually injects a lethal dose of a sedative, bringing about a quick and comfortable death. Why do we treat our human loved ones any differently?

A full discussion of the debates about euthanasia is clearly well beyond the scope of this article. If philosopher Dan Brock is correct that the commonly performed act of withdrawing mechanical ventilation is a justified form of killing the patient, then we need to think carefully about why we have been so resistant to adopting euthanasia [38].

For many, the biggest concerns about euthanasia relate to the ‘slippery slope,' that is, if we begin to permit euthanasia for cases that clearly seem justified, over time our practice will expand to include those who are disabled and those who have become a burden upon their families and society. Euthanasia has been practiced with varying degrees of legality in The Netherlands now for several decades, however, and for the most part these slippery slope concerns have not yet materialized [39,40].

Another strategy for blocking the slippery slope has been to permit only physician-assisted suicide (PAS), and not euthanasia. In PAS, physicians may write a prescription for a lethal medication, but it is up to the patient to actually take the drugs. Proponents of PAS argue that because the patient must personally make the final act, this practice is much less susceptible to abuse than euthanasia. In the United States, PAS has been legalized in one state (Oregon) for several years now and again concerns about abuse and overuse have not manifested themselves [41,42].

Just how bad does the neurological condition and/or prognosis need to be?

If I were to speculate about the next scientific and ethical frontier for these questions, I would bet that it will involve addressing the question, ‘Just how bad does the neurological condition and/or prognosis need to be before it is acceptable to allow a decision to withdraw life support?' Remember that in 1968, the neurological condition of the patient had to be equivalent to brain death for withdrawal of the ventilator to be ethically and legally acceptable. Through the Quinlan case and the series of cases that followed, the courts came to accept a more liberal threshold - the PVS - as a necessary degree of neurological damage to permit the withdrawal of life support.

PVS is defined as a state of permanent unconsciousness. Patients in a PVS have sleep-wake cycles and roving eye movements, but these are thought to originate from subcortical brain levels and are not associated with any awareness or responsiveness. Over the decades, both jurists and philosophers have found this to be a comfortable and defensible threshold for permitting the withdrawal of life support, since permanent unconsciousness seemed incompatible with a meaningful existence [43].

However, several issues have eroded upon this comfortable standard. First, there is no doubt that life support is withdrawn from many patients who do not meet the criteria for PVS. As long as no one challenges the decision-making of those at the bedside, the patient's family and the clinicians are usually the ones who decide whether the neurological status of the patient is sufficiently poor to justify withdrawal of treatment. While this issue has not been carefully studied, clinical experience suggests that the thresholds considered sufficient to justify withdrawal are highly variable.

The courts have begun to address current practice through a small number of cases that question whether life support can be withdrawn from patients who are in a ‘minimally conscious state' (MCS), a relatively new diagnostic category for patients who are in a ‘near-PVS.' Unlike PVS, which is clearly defined as the absence of consciousness, MCS has no clear conceptual boundaries, and is therefore ethically and legally problematic [44,45].

Finally, recently a well-documented case report demonstrated the use of functional-magnetic resonance imaging to show that a patient who met all of the diagnostic criteria for PVS was actually aware of her environment and in a ‘locked-in' state. While cases like this are widely regarded as exceptionally rare, they do raise questions about our ability to accurately diagnose states of consciousness that fall outside of the category of brain death [46].

Concerns for the future

The approach to end-of-life decision-making in the United States has been logically constructed over the past 50 yr, and while controversy persists at the margins, for the most part the Great American Consensus is firmly embedded in ethics and law and widely accepted by the public (Table 1).

Table 1
Table 1:
. Key concepts in understanding end-of-life decision-making in the United States.

Throughout the world, however, the last few decades have seen a rise in religious fundamentalism across many of the great religious traditions. These movements tend to share a vitalistic view about the sanctity of human life, and fundamentally challenge the Great American Consensus by seeing life as a gift from God and rejecting the claim that patients and their surrogates have a right to refuse life-sustaining treatments when they regard the burdens as disproportionate to the benefits.

In the United States, this tension became the dominant issue in the national debate over Terry Schiavo, a young female who, like Karen Ann Quinlan and Nancy Cruzan before her, was in a PVS. Her husband and legal surrogate requested withdrawal of her feeding tube. Carrying the banner of Christian fundamentalists in the United States, leaders in government and even the President of the United States tried to intervene to keep her alive [47]. The case was heard in a variety of legal settings, and jurists were unanimous in affirming and reaffirming the core principle of the Great American Consensus and in supporting the right of her husband, as her legally authorized surrogate, to authorize removal of her feeding tube. After all possible legal appeals were finally exhausted, her feeding tube was removed and she died peacefully.

Despite resolution of the Schiavo case in accord with the principles outlined above, it may serve as a bellwether of changes to come, as a trend away from the secular humanism that has characterized Western Civilization since the Enlightenment of the seventeenth century. Further developments in our approach to decision-making around end-of-life care will likely depend on the direction of these larger social, political and theological trends in the years to come.

Acknowledgement

Financial support from the Division of Medical Ethics and the Division of Critical Care Medicine.

References

1. Mollaret P, Goulon M. Le coma depasse. Rev Neurol 1959; 101: 3-15.
2. Barnard CN. Reflections on the first heart transplant. S Afr Med J 1987; 72: 19.
3. Ad Hoc Committee of the Harvard Medical School to Examine the Definition of Brain Death. A definition of irreversible coma. Report of the Ad Hoc Committee of the Harvard Medical School to Examine the Definition of Brain Death. JAMA 1968; 205: 337-340.
4. Plum F. Clinical standards and technological confirmatory tests in diagnosing brain death. In: Youngner SJ, Arnold RM, Schapiro R, eds. The Definition of Death: Contemporary Controversies. Baltimore, USA: The Johns Hopkins University Press, 1999: 34-65.
5. Truog RD. Is it time to abandon brain death? Hastings Cent Rep 1997; 27: 29-37.
6. Truog RD, Robinson WM. Role of brain death and the dead-donor rule in the ethics of organ transplantation. Crit Care Med 2003; 31: 2391-2396.
7. Truog RD. Organ donation without brain death? Hastings Cent Rep 2005; 35: 3.
8. Wijdicks EFM. Brain death worldwide - accepted fact but no global consensus in diagnostic criteria. Neurology 2002; 58: 20-25.
9. Wijdicks EFM. The diagnosis of brain death. N Engl J Med 2001; 344: 1215-1221.
10. McIntyre RL. The significance of the legacy of Karen Ann Quinlan. Trends Health Care Law and Ethics 1993; 8: 7-16.
11. Truog RD, Levine DZ, Hutchinson T et al.. Discontinuing immunosuppression in a child with a renal transplant: are there limits to withdrawing life support? (Discussion). Am J Kidney Dis 2001; 38: 901-915.
12. Prendergast TJ, Claessens MT, Luce JM. A national survey of end-of-life care for critically ill patients. Am J Respir Crit Care Med 1998; 158: 1163-1167.
13. Burns JP, Mitchell C, Outwater KM et al.. End-of-life care in the pediatric intensive cave unit after the forgoing of life-sustaining treatment. Crit Care Med 2000; 28: 3060-3066.
14. Brock DW. Death and dying. In: Veatch RM, ed. Medical Ethics. Boston, USA: Jones and Bartlett Publishers, 1989: 329-356.
15. Layon AJ, D'Amico R, Caton D, Mollet CJ. And the patient chose: medical ethics and the case of the Jehovah's Witness. Anesthesiology 1990; 73: 1258-1262.
16. Le Conseil d'État statuant au contentieux, sur le rapport de la 5èmesous-section N° 198546, Séance du 12 octobre 2001, lecture du 26 octobre 2001 Mme X. http://www.conseil-etat.fr/ce/actual/index_ac_lc0115.shtml. 1-15-2007.
17. Emanuel LL, Emanuel EJ. The Medical Directive: a new comprehensive advance care document. JAMA 1989; 261: 3288-3293.
18. Goodman MD, Tarnoff M, Slotman GJ. Effect of advance directives on the management of elderly critically ill patients. Crit Care Med 1998; 26: 701-704.
19. Pochard F, Azoulay E, Chevret S et al.. Symptoms of anxiety and depression in family members of intensive care unit patients: ethical hypothesis regarding decision-making capacity. Crit Care Med 2001; 29: 1893-1897.
20. Lemaire F. The European Directive 2001/20 for clinical research: friend or foe? Intensive Care Med 2006; 32: 1689-1690.
21. Emanuel EJ, Emanuel LL. Four models of the physician-patient relationship. JAMA 1992; 267: 2221-2226.
22. Paris JJ, Cassem EH, Dec W, Reardon FE. Use of a DNR order over family objections: The case of Gilgunn v. MGH. J Intensive Care Med 1999; 14: 41-45.
23. Capron AM. Abandoning a waning life. Hastings Cent Rep 1995; 25: 24-26.
24. Flamm AL. The Texas “futility” procedure: no such thing as a fairy-tale ending. Med Ethics 2004; 11: 4.
25. Fine RL, Mayo TW. Resolution of futility by due process: early experience with the Texas advance directives. Ann Intern Med 2003; 138: 743-746.
26. Annas GL, Arnold B, Aroskar M et al.. Bioethicists' statement on the US Supreme Court's Cruzan decision. N Engl J Med 1990; 323: 686-687.
27. Annas GJ. Nancy Cruzan and the right to die. N Engl J Med 1990; 323: 670-673.
28. Casarett D, Kapo J, Caplan A. Appropriate use of artificial nutrition and hydration fundamental principles and recommendations. N Engl J Med 2005; 353: 2607-2612.
29. Sulmasy DP. Terri Schiavo and the Roman Catholic tradition of forgoing extraordinary means of care. J Law Med Ethics 2005; 33: 359-362.
30. Ahronheim JC, Gasner MR. The sloganism of starvation. Lancet 1990; 335: 278-279.
31. Billings JA. Comfort measures for the terminally ill. Is dehydration painful? J Am Geriatr Soc 1985; 33: 808-810.
32. Ganzini L, Goy ER, Miller LL, Harvath TA, Jackson A, Delorit MA. Nurses' experiences with hospice patients who refuse food and fluids to hasten death. N Engl J Med 2003; 349: 359-365.
33. Eddy DM. A conversation with my mother. JAMA 1994; 272: 179-181.
34. McCue JD. The naturalness of dying. JAMA 1995; 273: 1039-1043.
35. Mitchell SL, Buchanan JL, Littlehale S, Hamel MB. Tube-feeding versus hand-feeding nursing home residents with advanced dementia: a cost comparison. J Am Med Dir Assoc 2003; 4: 27-33.
36. Mitchell SL. Financial incentives for placing feeding tubes in nursing home residents with advanced dementia. J Am Geriatr Soc 2003; 51: 129-131.
37. Truog RD, Cochrane TI. Refusal of hydration and nutrition: Irrelevance of the “artificial” versus “natural” distinction. Arch Intern Med 2005; 165: 2574-2576.
38. Brock D. Voluntary active euthanasia. Hastings Cent Rep 1992; 22: 10-22.
39. Van der Maas PJ, Van der Wal G, Haverkate I et al.. Euthanasia, physician-assisted suicide, and other medical practices involving the end of life in the Netherlands, 1990-1995. N Engl J Med 1996; 335: 1699-1705.
40. Van der Wal G, Van der Maas PJ, Bosma JM et al.. Evaluation of the notification procedure for physician-assisted death in the Netherlands. N Engl J Med 1996; 335: 1706-1711.
41. Hedberg K, Hopkins D, Kohn M. Five years of legal physician-assisted suicide in Oregon. N Engl J Med 2003; 348: 961-964.
42. Ganzini L, Harvath TA, Jackson A, Goy ER, Miller LL, Delorit MA. Experiences of Oregon nurses and social workers with hospice patients who requested assistance with suicide. N Engl J Med 2002; 347: 582-588.
43. Wade DT. Ethical issues in diagnosis and management of patients in the permanent vegetative state. BMJ 2001; 322: 352-354.
44. Lo B, Dornbrand L, Wolf LE, Groman M. The Wendland case - withdrawing life support from incompetent patients who are not terminally ill. N Engl J Med 2002; 346: 1489-1493.
45. Beaumont JG, Kenealy PM. Incidence and prevalence of the vegetative and minimally conscious states. Neuropsychol Rehabil 2005; 15: 184-189.
46. Owen AM, Coleman MR, Boly M, Davis MH, Laureys S, Pickard JD. Detecting awareness in the vegetative state. Science 2006; 313: 1402.
47. Annas GJ. “Culture of life” politics at the bedside - the case of Terri Schiavo. N Engl J Med 2005; 352: 1710-1715.
Keywords:

ETHICS; ORGAN DONATION; PROFESSIONAL-FAMILY RELATIONS; PERSISTENT VEGETATIVE STATE; DECISION-MAKING

© 2008 European Society of Anaesthesiology