| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Editorial |
1 From the Departments of Radiology (R.B.G.) and Pediatrics (W.A.E.), Indiana University School of Medicine, 702 Barnhill Dr, Room 1053, Indianapolis, IN 46202-5200. Received October 14, 2004; accepted October 18. Address correspondence to R.B.G. (e-mail: rbgunder{at}iupui.edu).
Neonatal medicine has made it possible to preserve the lives of many newborns who, had they been born several decades ago, would undoubtedly have perished. Infants whose lungs are not sufficiently developed to permit them to breathe on their own can now be saved with increased use of antenatal steroid therapy and innovations such as high-frequency ventilators and exogenous surfactants. Along with the expansion in our abilities to save lives, however, have come new and sometimes difficult ethical questions that parents, physicians, and policy makers of the past never needed to confront (1). While radiologists are typically not directly responsible for making these decisions, those of us who provide clinical services to neonatologists and their patients and families benefit by understanding the medical and ethical issues involved.
One of the core clinical issues is this: Should we, meaning parents and medical caregivers, make every effort to save each and every premature neonate, no matter how premature? In some cases, the odds of survival may be so low, or the probability of permanent and severe disabilities so great, that reasonable people question whether it is appropriate to "do everything possible" (2). Underlying the appropriateness of such an action is a fundamental moral question: Under what circumstances, if any, should we refrain from interventions designed to prolong the life of a premature neonate? If "can" does not always imply "should," how can we distinguish between the neonates we should attempt to save and those we should not (3)?
The question is an important one. Of the 4 million babies born each year in the United States, approximately one-eighth, or 500 000, are born prematurely (ie, before 37 completed weeks of pregnancy) (4). Of all live births, approximately 12%–14%, or 560 000, are admitted to a newborn intensive care unit (5,6).
While survival rates vary from center to center and according to the patient population, infants born at 30 weeks gestation have a better than 90% probability of leaving the newborn intensive care unit alive; however, infants born at 22 or fewer weeks gestation have virtually no chance of survival (7). For this reason, some centers do not ask a neonatologist to be present at delivery if the pregnancy has lasted fewer than 23 weeks (8). For all extremely low-birth-weight infants (<1000 g), the rate of disabilities such as cerebral palsy, developmental delay, and vision and hearing impairment is about 20%. This rises to 50% for infants at 23 weeks, when the survival rate is only 10%–20% (7).
It should be emphasized that the capability to preserve the lives of extremely small and premature infants was developed only recently, and this development reflects extraordinary progress in the young medical field of neonatology. Between 1960 and 2000, the survival rate of infants weighing between 501 and 1000 g increased from approximately 5% to 70%, while that for infants weighing between 1001 and 1500 g increased from approximately 50% to 95% (9,10). It should be noted that such capabilities are not distributed evenly around the world. Premature infants who routinely survive in the United States would inevitably expire in poorer regions, simply because these regions lack the expertise and technologies found in newborn intensive care units in developed countries.
Those who advocate withholding treatment generally base their argument on one of three grounds: (a) the low probability of survival; (b) the high probability of severe disability; or (c) the high projected costs of medical care, both in the neonatal period and throughout the patient's life (11). Our accuracy in predicting what will happen to a particular infant, however, is less than 100%. Because neonatologists cannot confidently assess an infant's prospects for survival in the first few critical minutes after birth, some may recommend resuscitation of all infants born after 22 weeks gestation.
It also is difficult or impossible to predict at birth which extremely low-birth-weight infants will develop severe disabilities and which will leave the newborn intensive care unit disability free. The situation is different for infants with readily apparent severe malformations. Except for the most extreme cases, many neonatologists regard prematurity itself as an insufficiently reliable indicator.
What probability of death or severe disability is sufficient to warrant the withholding of treatment? It is doubtful that anyone would place the tipping point at a 50% chance of survival, particularly if that chance is associated with full functional capabilities. What about a 20% chance? What about a 5% chance? What about a 1% chance? Let us say that we are prepared to withhold treatment any time the probability of survival is less than one in 50. What if the patient in question differs in some unrecognized but important respect from the patient population whose probabilities we are quoting? Birth weight and gestational age are clearly important, but they are not the only salient factors. Some observers would question whether our data are robust enough to ensure that every feature of every patient has been taken into account in formulating a prognosis.
In coming to grips with this issue, we need to address two fundamental questions. First, what harm is done if an infant is treated aggressively but eventually expires in spite of our best efforts? Further, what if the infant survives but suffers from severe disabilities? Second, what harm is done if treatment is withheld from an infant on the basis of the high probability of death or severe disability, yet had treatment been provided the infant would have survived with full functional capacity? What if the infant would have survived but with a disability? Should we operate with a presumption in favor of preserving life? Some would argue that, if we do not know with a very high level of confidence whether recovery is possible, we should err on the side of life (11).
A presumption in favor of life that leads parents and physicians to choose initial aggressive treatment for a potentially viable infant does not mean that a less aggressive strategy or change to palliative care may not be adopted at a later time. This principle is manifested in practice by the fact that, in many neonatal intensive care units, a majority of infant deaths are preceded by a decision to withdraw treatment (12,13). An example would be an extremely premature infant with multiple–organ system failure from overwhelming sepsis. The neonatologist's role is to function as an advisor to parents or guardians, with whom ultimate decision-making authority rests. It is vital that all who are involved with the care of the patient, including both the physicians and nurses, do not object strongly to changes in the course of treatment.
In what circumstances do decisions to withhold treatment take place? One circumstance in which there is relatively little dispute about the decision is that in which an infant is in the process of dying and death cannot be prevented (13). This situation occurs in neonates born before 22 weeks gestation, and perhaps in older infants of a birth weight less than 500 g. In such situations, aggressive treatment could be deemed not only medically futile but an infliction of unnecessary suffering. The difficulty with such judgments, of course, is that situations that were once regarded as medically hopeless did not remain so, and circumstances that today prompt an attitude of resignation may not do so in the future. Are we prepared to risk sacrificing lives that need not necessarily be lost?
Another situation in which treatment is sometimes withheld is that in which the infant is not dying but the probability of severe disability is very high. For example, consider an extremely premature infant who suffers a high-grade intracranial hemorrhage and develops diffuse cystic encephalomalacia. Such an infant's prospects for severe mental and physical disability, including a lack of self-awareness and of the ability to act with purpose, may be judged so high that physicians and parents agree to withdraw life support or decline to resuscitate the infant if a clinical downturn occurs. Of course, judgments about what constitutes an acceptable quality of life may vary from physician to physician and family to family. One family may judge even relatively mild sensory, cognitive, and motor impairments unacceptable, while another may eagerly welcome a child that others would regard as neurologically devastated.
A third situation in which the possibility of withholding treatment arises is that in which the infant is likely to require extensive medical care throughout life, including frequent hospitalizations and, possibly, surgical operations, which will impose significant burdens not only on the child but on the family. These burdens may take psychological, social, and economic forms. Advocates for withholding treatment may couch their arguments in terms of the best interests of the child or in terms of what is best for the family. Some may even argue that the community as a whole cannot afford to spend hundreds of thousands of dollars saving the life of a premature neonate whose subsequent disabilities will only impose additional hundreds of thousands of dollars in medical costs over a lifetime.
Again, however, one difficulty with this line of argument is that judgments about what constitutes an acceptable quality of life, and an acceptable level of expenditure in order to achieve it, vary from person to person. To say that an infant cannot benefit from treatment is not strictly a medical judgment. The setting of that threshold involves moral judgments as well, including how much suffering a person can tolerate and the level of impairment a person can endure and still find life worth living. For example, sighted people generally overestimate the burden of blindness, as compared with the assessments of people living with visual impairment. A good deal of the suffering from a disability may, in fact, be remediable. If a disabled person lives with a loving and supportive family, it may be possible to lead a relatively rich and happy life even in the face of severe disability.
To say that any particular medical outcome is worse than death is fraught with difficulty. First, as we have seen, it is difficult to ascertain that an adequate level of confidence has been reached concerning the prognosis for extremely premature babies. The accuracy and precision of such predictions will never reach 100%, thus raising the possibility that some viable infants might die if not aggressively treated. Second, the very definition of unacceptable disability and suffering is difficult to specify. When inconsistencies and even frank disagreements arise, who will adjudicate the disputes? Should the parents always decide? Would neonatologists be defaulting on their fiduciary responsibility as advocates for the infant if they failed to object when parents decide that the prospect of relatively minor disabilities compels them to withhold treatment? What if other potential caregivers were on hand who would welcome the opportunity to devote their lives to the care of such a child?
It is important to recognize that even the most desperately ill premature infant is still a human being who deserves our respect and compassion. We manifest our awareness of this by the care we give infants for whom we have reached a decision to do nothing further. No one simply discards such infants as though they had already died. Instead, we continue to provide basic nursing care, with special emphasis on comfort for the infant and the family. Infants are kept warm and free of pain. Families are encouraged to be with their infants, and where possible, to hold them. When problems such as respiratory distress supervene, we provide symptomatic relief, such as suctioning secretions and administering morphine. The goal in such situations is not to forestall the death of the infant but to support the patient and family as death takes place. The death of a premature infant is not merely a biological outcome. It is also one of the most profound and poignant chapters in the life story of the family involved, and the entire health care team needs to acknowledge and respect this.
The attitude that hearts should be kept beating until the last possible moment is inappropriate. Death is clearly as natural a part of life as birth. The mere fact that patients are very young does not imply that they should not be allowed to die when their time has come. In general, there are many reasons to support a presumption on behalf of preserving life, at least until the longer-term prospects for survival and disability can be assessed. In some cases, unexpected cures may be achieved. Even where no cure is possible, however, physicians can and should seek to relieve suffering. When death is inevitable, physicians should step aside and allow it to happen, preserving dignity and helping the experience to be as meaningful as possible for families.
 |
References |
|---|
 TOP References |
|---|
This article has been cited by other articles:
|
|
 |
|
  E. D. Kinney The corporate transformation of medical specialty care: the exemplary case of neonatology. J. Law Med. Ethics, December 1, 2008; 36(4): 790 - 802. [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
