Active Versus Passive Euthanasia

Once a decision has been made that treatment is no longer serving the patient’s interest, it would seem to follow that a quick and painless death would be preferred to a long dying process, with associated discomfort for the patient or his or her family. Active euthanasia can be defined as a physical act that causes the death of a patient and was intended to be in the patient’s interest. Passive euthanasia is generally understood to include withholding or withdrawing treatment, with the intent and expectation that death will occur sooner rather than later, based on the belief that an early death is in the patient’s interest. Both practices have the intent and usually the consequence that death will occur sooner rather than later. Active euthanasia is nearly universally prohibited, and passive euthanasia is widely tolerated. The major reason for the distinction is based primarily on concern for future patients, not the patient at hand. Indeed, active euthanasia often seems more merciful from the patient’s perspective, precisely because the suffering is reduced. This concern is less relevant when suffering can be relieved with sedation and analgesia.

Some have claimed that terminal sedation constitutes active euthanasia, but there is broad consensus, in ethics and in law, that if the intent is to relieve suffering, the fact that death may be accelerated does not constitute active euthanasia.^2,3

The major objection to active euthanasia is based on the concern that physicians and others will slide down a “slippery slope,” meaning that they will become progressively less sensitive and careful about who is a suitable candidate for euthanasia. The claim is made that if the traditional barrier against killing is lowered, doctors will give in to pressures and temptations to end suffering, including the suffering of family, nurses, and the physician, by taking advantage of the quick release that killing provides.

There is some evidence for this view in studies of the two modern societies in which active euthanasia has been widely practiced and tolerated by the state. Lifton’s interviews with physicians who worked in Germany in the 1930s and 1940s demonstrate a progression from killing patients with terminal or severely disabling conditions to the slaughter of healthy people in the death camps.⁴ Physicians who were directly involved and responsible for the executions reported that they believed at the time that there was no moral distinction between the two types of killings. They justified and defended their involvement in the death camps primarily as an example of their duty as physicians to relieve suffering.

In contemporary Holland, active euthanasia is widely practiced and tolerated by state policy. Public policy limits the practice to clearly competent patients, but cases of children incapable of consenting who were killed at parental request have been reported.⁵

The prohibition of active euthanasia based on “slippery slope” concerns implies that present patients’ interests may be sacrificed to those of future patients. However, if suffering can be eliminated with medications, there should be few, if any, patients whose interests will be harmed by prohibiting active killing. In fact, the vast majority of patients in the contemporary ICU depend on technologies whose discontinuation usually will not result in a long dying process. The major exception is patients who depend only on nutrition and hydration, whose deaths can take 1 week or longer but for whom sedation can reduce suffering.

Withholding Food and Water

Once the judgment has been made that continued existence no longer serves the patient’s interest, it should not matter, from the patient’s perspective, what treatment is being used to keep him or her alive. The only exception would be treatments that serve to keep the patient comfortable during the dying process. However, if the patient is incapable of experiencing discomfort, whether because of his or her condition or treatments that render him or her unconscious, then any treatment that keeps the patient alive contrary to his or her interests not only may be discontinued but also should be discontinued. It is not only permissible to discontinue treatments that are not serving the patient’s interest, it is obligatory.

For these reasons, treatments that provide nutrition and hydration have come to be seen as analogous to other treatments that keep patients alive. In addition to the President’s Commission,² the American Medical Association, numerous state courts, and the United States Supreme Court have ruled that food and water can be discontinued when other requirements for withdrawing life-sustaining treatment have been met.

It is understandable that discontinuing food and water is psychologically more distressing, particularly to nursing personnel, than withdrawing other forms of treatment, in part because feeding is such an innate human instinct, unlike transfusing, resuscitating, or providing oxygen. Therefore decisions to discontinue food and water often require more discussion and staff support.

One aspect of withholding or withdrawing food and water that distinguishes it from withholding other medical treatments is the extent of the opportunity for abuse. Because every patient requires nutrition and hydration, the power to withhold it gives the physician a power that is almost indistinguishable from active euthanasia.⁶ This is not to say opportunities to discontinue other forms of technology, such as mechanical ventilators or surgery, are not subject to abuse. The concern about abuse arises not so much in the ICU, where almost all patients are utterly dependent on some technology, but elsewhere in or out of the hospital where food and water may be the only “treatment” keeping a seriously ill or handicapped patient alive.

Competence, Incompetence, and Baby Doe

Autonomy is a central principle in American medical ethics. It implies that a competent person has a nearly absolute right to decide what shall be done to his or her body. Courts have consistently upheld the right of competent patients to refuse lifesaving treatment, even if it appears foolish or unwise to others, whether for religious or secular reasons or for no reason at all.

The definition of competence is itself controversial,⁷ but many adolescent patients meet the standard. The most common definition relies on the ability of the patient to understand the consequences of his or her decision. Physicians therefore should be sensitive to distinctions between the wants and the interests of pediatric patients.

The vast majority of PICU patients, regardless of age, are incompetent because of developmental status, disease, or medication. Traditionally, parents had nearly complete discretion to make decisions on behalf of their children. This practice came under intense criticism in the 1970s and 1980s upon disclosure of many cases in which handicapped and critically ill infants who appeared to have good prospects for long, meaningful life were allowed to die.⁸ These cases most commonly involved infants with Down syndrome or spina bifida. In some instances, infants with Down syndrome and duodenal atresia were allowed to die of dehydration. In one center, more than 50% of infants with spina bifida had standard treatment withheld with the intention that they would die.⁹ This pattern of inappropriate undertreatment has been called the Baby Doe problem, named after a celebrated case involving a newborn with Down syndrome and esophageal atresia who was allowed to die without surgery.¹⁰

The response of pediatricians and state courts was generally to respect the wishes of parents, even when decisions appeared to be contrary to the interests of the child.^11,12 In 1982, the federal government promulgated regulations requiring reports to a hotline and investigations by “Baby Doe squads” of alleged withholding of medically beneficial treatment based on handicap. These regulations were found unconstitutional but were reinstated pursuant to amendments to a child abuse statute that provided funds to the states for implementation of child abuse and neglect programs.^12a The most controversial aspect of these regulations is their apparent prohibition of withholding or withdrawing medically beneficial treatment from any infant based on handicap or prognosis for quality of life. The three exceptions allowed are (1) if the infant is permanently comatose, (2) if the infant is imminently dying, or (3) if the treatment would be “inhumane.” The legal significance of these regulations and the role of the law in these cases in general have been the sources of continuing confusion and controversy.

Legal Implications of Withholding or Withdrawing Life Support

Involvement of state and federal legislatures and courts in decisions involving treatment of dying patients is rapidly evolving, and varies among jurisdictions.¹³ Generalizations are difficult, and the law varies among states. Nonetheless, some legal aspects are reasonably clear.

The most widespread source of confusion is the failure to recognize the difference between the law in theory and the law in practice. Although a variety of statutes, court decisions, and regulations have been interpreted, in many cases, as prohibiting discontinuation of treatment, the empiric fact is that, in the United States, it is uncertain whether any physician has ever been found to have civil or criminal liability for withholding or withdrawing life-sustaining treatment from any patient unless the physician’s conduct was “outrageous.” The law in practice has been remarkably deferential to physician discretion.¹⁴

Even on theoretical grounds, there are reasons to believe that the Baby Doe regulations do not regulate decisions by physicians, parents, or hospitals. Although the substantive standards appear very restrictive, the implementation standards and sanctions defer to the states. All that is necessary for states to be in compliance with the law is to provide assurances to the federal government that reports of alleged medical neglect will be handled in a prescribed way. States are not required to initiate investigations but are permitted to rely on reports, in accordance with traditional regulations involving child abuse and neglect. A study by the U.S. Office of the Inspector General found all states were in compliance with the Baby Doe regulations, although the federal Civil Rights Commission argued that more restrictive legislation is needed.¹⁵

In 1990, the U.S. Supreme Court issued its first opinion on the specific issue of discontinuing treatment of an incompetent patient. The case involved Nancy Cruzan, a young adult who had been in a persistent vegetative state for 8 years and was being kept alive with nasogastric feedings. Her parents had requested that the feeding be discontinued and that she be allowed to die. The Court upheld a Missouri statute that required “clear and convincing” evidence that a patient had expressed her wish that she would want to die in that specific circumstance. It is unclear whether the decision has any relevance for children, even in Missouri, and it is important to realize that the decision has little or no relevance for decisions in the majority of states that do not have statutes as restrictive as those in Missouri. The central finding in the Court’s opinion was that states are free to make their own laws in this area. The majority of states at present defer to physician judgment in theory and virtually all defer in practice. Even in Wisconsin, where an appellate court decision held that parents may never consent to withholding of lifesaving treatment of a child unless he/she is in a persistent vegetative state,^16,17 physicians and hospitals routinely ignore the decision, and there have been no prosecutions.

The difficulty of sustaining a legal challenge against decisions to discontinue treatment was dramatized in an Illinois case involving a 2-year-old child, Sammy Linares, who was in a persistent vegetative state and ventilator dependent for 8 months. The hospital refused the parents’ request to discontinue the ventilator, based on the hospital attorney’s claim that to do so would be illegal. The father held the medical staff at gunpoint while he discontinued the ventilator until his son died. Despite committing at least two prima facie felonies involving illegal possession and use of a deadly weapon, the district attorney could not convince a grand jury to indict the father. Despite the clear illegality of his action in theory, in practice the legal system was typically sympathetic and supportive of the decision because it appeared to violate no interests of the child.¹⁸ In a similar case, a federal appellate court upheld the right of a hospital to discontinue life support, over parental objection, for a child with profound brain injury.¹⁹

Despite the nearly universal legal deference to physician judgment, many physicians overtreat infants because they either misunderstand the law or have an exaggerated fear of liability. A survey of neonatologists found that 30% to 50% would continue treatment even when they thought it was not indicated, based on their belief that the Baby Doe regulations required such treatment.²⁰ Another survey of a group of California neonatologists before and after the Baby Doe regulations also found a trend toward overtreatment.²¹

In a Virginia case, an anencephalic infant known as Baby K was kept alive for 2.5 years, including intermittent ventilator support, at the request of the mother. A federal court ruled that resuscitation was required under a federal statute known as the EMTALA Act, a law that was intended to prevent dumping of patients from emergency rooms. That ruling is generally not followed in other courts, and the Center for Medicaid and Medicare Services has ruled that EMTALA does not apply to inpatients.²² It is also important to note that the courts generally rule only when asked. Nor does the ruling prove that there would have been liability if the doctors had refused to treat the infant. In a related case, a jury acquitted a doctor at the Massachusetts General Hospital who discontinued life-sustaining treatment for a woman in a persistent vegetative state, over the objection of the family.²³ The contrast with the Baby K case is that court opinion was not sought.

Futility

The increasing success of technology in maintaining vital functions has led to an increase in survival of patients with little prospects for long or meaningful life, resulting in conflict between parents who want “everything” done, and medical staff, who believe continued treatment is not in the interests of the child. The debate over withholding or withdrawing treatment against family objection frequently focuses on the claim that treatment is futile, and therefore need not be offered. When treatment is truly medically ineffective—such as mechanical ventilation in a child with extreme pulmonary hypoplasia—a claim of medical futility is warranted. More commonly, however, those who wish to make a unilateral decision to limit treatment, over family objection, are relying on a social notion of futility; that is, a claim that even if the treatment is medically effective, it will not result in a life worth living, or is not worth the cost. These judgments may be defensible, but they are not medical judgments, and physicians have little basis for assuming authority to make such judgments unilaterally. These difficult cases usually need to be resolved through negotiation and persuasion, with the help of institutional ethics committees.^24,25

Hospital Ethics Committees

With the difficulty in reaching a consensus on substantive standards for decisions to discontinue life support, interest grew in procedural guidelines. The conceptual basis for the guidelines is the general notion that because one can never know whether a decision actually is right, the question of whether or not a decision is morally defensible depends on the process by which the decision is made. The process is similar to other areas of decision making, including the judicial process, in which the legal correctness of a decision resides almost completely in the process by which it is made; in the scientific method, which concedes the impossibility of ultimately knowing the truth but accepts successive approximations if they withstand scrutiny of the process by which they are made; and judgments about the quality of medical care, which ultimately depend less on outcomes than on process.

One procedural theory of ethical decision making that influenced the formation of ethics committees is called ideal observer theory.²⁶ The theory argued that a decision is morally right if it could receive approval from an ideal ethical observer with the following characteristics: 1) omniscient, meaning access to the relevant and available facts; 2) omnipercipient, meaning the ability to empathize, to vividly imagine how others feel, or to put one’s self in another’s shoes; 3) disinterested, having no vested interest in the outcome; 4) dispassionate, not being overwhelmed with emotion at the time critical decisions must be made; and 5) consistent, meaning that similar cases will be decided similarly.

In retrospect, decisions that appeared to be indefensible could be analyzed in this framework. In most cases, the failure to appreciate readily available facts appeared to be at the center of controversial decisions. Erroneous assumptions about prognosis, the availability of alternative care arrangements, or misunderstandings about the law are common examples. Because no single person can have the godlike qualities of an ideal ethical observer, interest arose in a process that might better emulate the model than traditional doctor–parent decisions. The federal Baby Doe regulations recommended but did not require such committees, which they referred to as “Infant Care Review Committees.” Since 1994, the Joint Commission on Accreditation of Hospitals has required ethics committees or some other mechanism for resolving disputes about terminal care, and they have become a standard method for resolving controversial decisions.²⁷ These committees typically are interdisciplinary groups of 10 to 20 people, including hospital-based professionals and community members, which advise and recommend but do not typically decide on treatment plans.²⁸ Some hospitals have ethics consultants who work independently of an institutional ethics committee. Studies of their activities are limited,²⁹ but their growth has been accompanied by an apparent disappearance of the problem of undertreatment. That problem has been replaced by an apparent growth in overtreatment, which may be aggravated by the influence of hospital attorneys or risk managers on committees.³⁰ Consultation with committees usually is voluntary, though some hospitals require consultation for specified complex cases, such as “futility” cases.

Ethics committees often serve other functions besides ethical consensus development. They have been used to protect decision makers from liability. Malpractice charges require a finding of negligence, which means the physician failed to take adequate care in how a decision was made. Review and approval by an ethics committee may not only be helpful in defending against such a charge but may deter initiation of a suit. Similarly, a district attorney or judge considering legal charges may be less likely to pursue a case in which there was a special effort to obtain consultation and in which there was broad consensus beyond the family and attending physician. Committees have also served a therapeutic role, helping to manage tensions among medical staff, particularly nurses, when they do not necessarily disagree with a decision but appreciate the opportunity to review it in a broader forum than can be found within the ICU.

Caring for the Terminally Ill

Once a decision has been made that continued treatment is no longer in the patient’s interest, attention can turn to implementing the decision in a way that minimizes suffering for the patient, family, medical staff, and other patients.

Decisions to terminate care should be well documented in the chart and in the orders. With regard to legal liability, inadequate documentation is more likely to create vulnerability than the opposite. Because the decision to terminate care should have the support of all involved, there should be no reason to conceal the decision or the reasons for it. Consultation with an ethics committee should be documented, either by the attending physician or by a representative of the committee.

If a judgment has been made that the patient would be better off dead, there is rarely a reason to continue some treatments while stopping others. The common practice of discontinuing one life-prolonging measure—resuscitation—while maintaining other life-supporting treatments may be illogical. These seemingly contradictory measures are sometimes motivated by concerns for the feelings of family or staff.³¹ As long as the patient is not suffering from continued treatment, this practice may be justified for a brief period, but prolonged treatment while waiting and hoping that a patient will die despite the treatment at some point becomes a misuse of limited resources, whether of personnel or an ICU bed.

Similarly, “limited codes” or “partial codes” generally should not be tolerated if their only purpose is to treat the sensitivities of medical and nursing staff. One hazard of partial resuscitation is that it may succeed in keeping the patient alive but result in additional damage to the patient because of inadequate oxygenation or perfusion. In summary, if a decision has been properly made that a patient is better off dead, it usually is best to effect that decision as soon as possible by discontinuing all measures that might prolong the patient’s life.

Brain Death

All states now support brain death as a legal standard for determination of death, although there is growing controversy about the moral and clinical relevance of the concept. This standard requires irreversible cessation of all brain activity, including the brainstem. The criteria for determining that brain death has occurred are medical criteria not established by statutes or court cases. They change as new technology and research evolves; therefore it is incumbent on the physician to keep abreast of developments in this area. One study showed considerable ignorance and confusion on the subject.³² The clinical criteria promulgated by a Harvard committee 20 years ago have come under increasing criticism, as it is clear that many patients declared to be brain dead by standard criteria have not lost all brain function, particularly the hypothalamus, evidenced by normal regulation of antidiuretic hormone activity.³³ Guidelines have been developed for applying clinical standards to children, including neonates.³⁴ Infants born alive with anencephaly typically have brainstem function and therefore do not meet any current definition of brain death.³⁵

Organ Retrieval

There is a wide gap between the number of patients who could benefit from organ transplantation and the supply of organs. There also is a gap between the number of people who say they are willing to donate their own organs or those of family members when death has occurred and the number who actually do donate. These observations led to a national “required request” law, which requires hospitals to inform relatives in appropriate cases that organ donation is an option. The law has not resulted in the anticipated increase in organ supply, for a variety of reasons. Nonetheless, the ICU staff not only has a legal duty to ask but also should realize that organ donation is generally perceived as a benefit to those who consent to it, not as a burden.

Questions may arise regarding who should ask and when. It is a longstanding principle that those involved with the care of potential recipients, such as members of the transplant team, should not be in contact with the family of the potential source of the organ because of the obvious conflicts of interest. It is important to coordinate such decisions to maximize the efficiency and effectiveness of the transplantation; therefore it is appropriate that those caring for the “donor” communicate with the transplant team. It also may be appropriate for members of the transplant team to meet with the family of the “donor” after the decision has been made, to answer questions.

The shortage of organs has also stimulated new efforts to remove organs more efficiently from patients who are not brain dead and who die by traditional criteria of cessation of heart and lung function. Protocols have been established for discontinuing ventilator support in the operating room so that organs can be removed promptly after death is declared—donation after cardiac death and before organ viability is threatened by warm ischemia (see Chapter 13).³⁶

References are available online at http://www.expertconsult.com.