There is strong consensus in ethics and law that there is no meaningful distinction between withholding and withdrawing life-sustaining treatment. The President’s Commission on Ethical Problems in Medicine concluded that contrary to widespread feelings on the matter, withdrawing treatment was preferable to withholding for two reasons.¹ First, withdrawing treatment implies that there has been more time for assessment of prognosis and the likelihood of a successful outcome. It comes after a clinical trial. Good ethics starts with good facts, and a decision to withhold treatment generally is made with less data than the decision to withdraw after treatment has been tried. Second, a tradition of reluctance to withdraw treatment often led to inappropriate withholding of support from patients who, in retrospect, may have had a good chance for meaningful long-term survival. One common setting was the delivery room, where high-risk infants sometimes were not resuscitated because of the fear that treatment would succeed in the medical sense, but that the infant’s prognosis for meaningful survival would vanish without any means of reversing the decision to treat. Allowing treatment to be stopped allows all patients to enjoy the potential benefits of initial intervention and the possibility of success. The central ethical issue is whether treatment serves the interests of the patient. If it does not, whether it is being considered or has already been tried is irrelevant.

Two types of errors can occur. A type I error is one in which a patient lives who, in retrospect, would have been better off had he or she died much earlier. A type II error is one in which a patient dies who, in retrospect, probably would have enjoyed life. There is consensus that the latter is generally more serious than the former for two reasons: first, because life is valued so highly and most people would prefer even a handicapped life or existence with suffering over death; and second, because a type I error usually is reversible whereas a type II error by definition is not. Particularly in the intensive care setting, prolonged survivors who are better off dead usually are dependent on some technology that can be discontinued to allow a natural death to occur. The increasing ethical and legal acceptance of discontinuing nutrition and hydration makes it theoretically possible to reverse any decision that, in retrospect, was not in the patient’s interest.

A corollary of these observations is that uncertainty generally should be resolved by maintaining the patient’s life, because resuscitation that on reflection may not have been indicated usually can be reversed, whereas failure to keep the patient alive cannot be corrected.

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

Buy Membership for Pediatrics Category to continue reading. Learn more here