by Tracy K. Koogler,
Benjamin S. Wilfond, and Lainie Friedman Ross
Neonatal and pediatric critical care have dramatically
improved survival for children over the past twenty years.
Neonatologists continue to reduce the gestational age and
birth weight that a newborn must have achieved to survive,
and advances in pediatric surgery and pediatric organ
transplantation mean that complex congenital heart disease
and biliary atresia no longer always result in death. They
are no longer “lethal,” because with treatment, an infant
can survive these conditions and even enjoy a good quality
of life.1
Yet obstetricians, pediatricians, geneticists, and
neonatologists continue to regard a category of congenital
syndromes as “lethal anomalies.”2 Examples are trisomy
13, trisomy 18, and anencephaly. Although most professional
discussions do not provide a specific definition of “lethal
anomaly,” the term is generally used to refer to a child
with (1) severe neurological compromise and (2) structural
anomalies and/or functional disabilities that, if untreated,
would cause death within a few months. The structural
anomalies in these children include airway malformations,
congenital heart disease, and gastrointestinal defects; the
functional disabilities include swallowing dysfunction,
aspiration, and apnea.
These anomalies and disabilities are usually treatable, and
invariably are treated in an otherwise healthy infant. What
makes them sometimes “lethal,” then, is the decision not to
repair the anomalies or treat the disabilities in light of
the child’s poor neurological prognosis. In effect, as we
will argue, “lethal anomaly” is not an accurate clinical
description; instead, it serves to convey an implicit
normative view about quality of life.
Although many of the congenital syndromes that used to be
lethal no longer are, they are still routinely referred to
as “lethal anomalies.” But the label is not only
inaccurate, it is also dangerous: by portraying as a medical
determination what is in fact a judgment about the child’s
quality of life, it wrests from the parents a decision that
only the parents can make.
The hidden significance of the phrase is problematic because
many parents may not rate the child’s quality of life as
dismally as their child’s health care providers do. Data
show that health care providers rank severely disabled
health states lower than parents do,3 which can partly
explain why parents typically prefer more aggressive
treatment than health care providers.4 Data also show that
what health care providers tell parents about a child’s
prognosis is influenced by the providers’ attitude toward
neurological impairment, with the effect that what is told
to the parents depends on the provider. 5 Since this
information influences family decisionmaking, 6 the phrase
“lethal anomaly” obscures the normative nature of the
decision and interferes with authentic parental autonomy.
An Historical Overview
The practice of recommending that invasive treatments be
withheld or withdrawn from infants with “lethal anomalies”
first received critical attention in 1973, when Raymond Duff
and AGM Campbell challenged the prevailing “social taboo” by
making public and explicit their reasons for not providing
invasive treatment to children with neurological
disabilities.7 They reviewed 299 consecutive deaths in the
Yale Special Care Nursery and found that 43 (14 percent)
were related to withholding treatment. They explained that
the decisions were made by the physicians and parents
because the child’s prognosis “for meaningful life was
extremely poor or hopeless”—a quality-of-life judgment that
rarely had been publicly stated previously.8
Once the Pandora’s box was opened, medical ethicists,
theologians, and clinicians came out in support of and
against this practice.9 The dispute came to a head in the
1980s, when disability advocates and rightto-life groups
publicized the issues surrounding the withholding of
treatment, including nutrition, in handicapped newborns.
They focused public attention on the fact that hospitals and
physicians were allowing children with congenital causes of
developmental delay, such as Trisomy 21, to die from
correctable birth defects. One such case to reach national
attention was Baby Doe, who was born in 1982 with Trisomy 21
and a correctable gastrointestinal defect.10 As a result of
this and other cases, regulations were proposed that would
have made withholding treatment for correctable birth
defects illegal, regardless of the child’s potential quality
of life.11 These regulations were successfully challenged on
the ground that they were too restrictive.12 Eventually,
more modest regulations were passed as amendments to the
1984 Child Abuse Prevention and Treatment Act (often known
as the Child Abuse Amendments or the CAA).13
Although the CAA did not dictate how physicians must
practice, requiring instead only that the states develop a
process by which to respond to cases of non-treatment, they
were at least partly responsible for changing the standard
of practice for treating mentally and physically handicapped
newborns.14 The CAA allowed for non-treatment in only a
narrow range of cases: when the infant is chronically and
irreversibly comatose; when the treatment would (1) merely
prolong dying, (2) not effectively ameliorate or correct all
of the infant’s life-threatening conditions, or (3)
otherwise be futile in terms of the survival of the infant;
or if the provision of such treatment would be virtually
futile or inhumane.15 While the Federal Regulations
explicitly stated that an anencephalic newborn would fit in
these categories, 16 whether children with congenital
anomalies like trisomy 18 and 13 met the criteria was
unclear. Withholding treatment in children with trisomy 13
and 18 is still a common practice.
In the decades since Campbell and Duff, however, some
families have begun to demand invasive treatment for these
children despite medical recommendations for comfort
measures only. Court cases have ensued when physicians and
parents have disagreed about the “futility” of treatment. A
well-known example of such a case is Baby K of Virginia.17
Baby K had anencephaly and was being cared for in a nursing
home. On several occasions, she was transferred to the
emergency room in respiratory failure, where her mother
demanded mechanical ventilation. The physicians sought
judicial relief from providing repeated mechanical
ventilation that they deemed “futile.” The court found for
the baby based on the Emergency Medical Treatment and Active
Labor Act (EMTALA), which does not allow a hospital to
refuse life-saving treatment to anyone in extremis. Baby K
lived for two and a half years, the longest that a child
with anencephaly has ever survived. In fact, before Baby K,
physicians had assumed an anencephalic child could not
survive beyond one year. Clearly, the use of
mechanical ventilation was not physiologically futile in
this case; it successfully prolonged Baby K’s life.
Part of the problem, then, is
how to define futility. Mechanically ventilating a child
with anencephaly is not physiologically futile because it
will restore ventilation and prolongs the child’s life. But
it is futile if one uses a qualitative notion of futility;
the child will never be able to interact with his or her
environment, even if the child is weaned from the
ventilator.18
Baby K is just one example of children with so-called lethal
anomalies who have survived beyond infancy. Although these
children will have severe neurological handicaps, they can
and do achieve some minimal number of milestones.19 If
institutions aggressively treated all of these children who
were born alive, the three- to five-year survival rate might
even improve drastically, although survival into the teens
would likely remain rare.20
The Ethical Arguments
There are three main arguments against providing medical
treatment for life threatening problems to all infants with
“lethal anomalies.” The first is just that treatment is
“futile,” but as we’ve noted, this claim depends on a
conception of futility based on quality-of-life
considerations and not on physiology. In many cases,
treatment will not only address the clinical problem, but
may result in prolonged life.
The second argument is that the burdens of treatment
outweigh the benefits in these children. This claim is a
value judgment and includes both the medical and
psychosocial burdens and benefits. In general, parents are
presumed to be the judge of whether the benefits of a
treatment plan outweigh the burdens, all things considered.
If the parents judge the treatments as more burdensome than
beneficial, then they have the right to refuse further
treatment. On the other hand, if the parents judge the
treatments as beneficial, then their requests
should be honored unless there is clear evidence that the
judgment is mistaken. If the child appears to be
experiencing significant unremitting pain or suffering, for
example, then it is wrong to continue treatments that only
prolong his survival and suffering. But rarely is this the
case.
The third argument against treatment for life-threatening
problems focuses on resource allocation and whether it makes
economic sense to treat children with such a grim
neurological prognosis when the money could be used for
greater benefit elsewhere. The economic argument fails
because the congenital anomalies classified as lethal are
rare, so even if these children were given maximal
treatment, their care would account for only a small
percentage of the total U.S. health care budget. For
example, Ramesh Sachdeva and colleagues found that only 1.6
percent of patient days in the PICU were spent on futile
treatment of children who had either a “lethal anomaly” or a
very poor quality of life secondary to PVS.21
But rejecting the resource question because it is not a true
economic threat ignores the ethical issue in the resource
allocation debate. The question is whether a quality-of-life
criterion can be used as a threshold to determine whether
treatment can be morally withheld. Although Duff and
Campbell thought yes, the public outcry leading to the
passage of the Baby Doe regulations showed that some
disagreed vehemently.22 In the absence of a social consensus
to limit care, the practice has evolved to empower families,
and not providers at the bedside, to make treatment
decisions based on quality of life.
If these three arguments to withhold life-saving medical
treatment from all infants with “lethal anomalies” fail,
then what arguments would lead to providing the treatment?
One is that physicians treat children with disabilities
unjustly if they do not correct all life-threatening
problems. This stance calls on physicians to treat all
correctable medical conditions regardless of quality-of-life
considerations and parental preferences—even to take legal
custody if parents do not consent.
This argument makes two assumptions: that it is possible to
make medical decisions without incorporating quality-of-life
judgments, and that it is appropriate to ignore
quality-of-life judgments when a child’s neurological
prognosis is dismal.
We reject these assumptions. First, it is not possible to
ignore quality-of-life judgments when making medical
decisions to withhold treatment without reverting to a
vitalistic approach to medicine that we believe is untenable
with modern technology.23 Nancy Rhoden has shown that even
the CAA exemptions incorporate quality-of-life judgments,
although narrowly circumscribed as exemplified by the
permissibility of nontreatment of comatose infants who, with
invasive treatment, might be able to live for years or
decades.24 Second, quality-of-life judgments are not only
necessary; they are appropriate. As Rhoden has explained,
consciousness is not valued as an end in itself, but as a
condition for meaningful relationships. Meaningful
relationships do not require “normal” or “near-normal”
intelligence but only some ability to interact with others
and with one’s surroundings. Without these abilities, many
families decide their loved one would not want to continue
to live.25
A second argument in favor of treatment is the belief that
any life is better than death. Yet this argument also fails.
While some health care providers, ethicists, and families
believe that all life is sacred and should never be
shortened,26 their belief is but one of several possible
ways of ordering the benefits and burdens; it should not be
imposed on those who hold divergent beliefs.
If the arguments to provide treatment for all children with
“lethal anomalies” fail, like the three arguments to
withhold treatment, then the substantive question has to be
regarded as an open question. The appropriate policy issue,
then, may not be substantive but procedural: who should
decide what is appropriate medical care for these children?
Throughout history, parents have been the primary
decisionmakers for their children. Third-party intervention
is restricted to cases in which parents are either abusive
or neglectful of their responsibility. In caring for
children with severe neurological compromise, there is no
clear social consensus on whether treatment should be
required or prohibited. Given this ambiguity, parental
decisions to withhold or request treatment based on
benefit-burden calculations should be respected.
Two caveats: First, some physicians, depending on their own
religious or philosophical perspective, may find it
difficult to comply with a decision to withhold treatment,
and others may be uneasy about providing it. Yet respect for
parental decisionmaking in such socially ambiguous
circumstances should persuade physicians to put aside their
own views when caring for such families. Ideally, then, the
physician would accede to parental authority, provided that
the decision is legally permissible. Alternatively, clinical
responsibility could be transferred if the physician cannot
come to terms with the family’s decision.
Second, our arguments apply to treatment decisions for
children with so-called lethal anomalies. What limits to
family autonomy are appropriate when the neurological
prognosis is less severe, or if a grim neurological
prognosis is not associated with any other life-threatening
conditions, are questions that go beyond the scope of this
paper.
Hidden Meaning
Consider the case of Baby B, an eighteen-month-old infant
with trisomy 18 admitted to the University of Chicago
Pediatric Intensive Care Unit. Baby B had severe
gastroesophogeal reflux resulting in multiple episodes of
aspiration pneumonia. He had not been gaining weight well,
and his development was at the level of a two-month-old
child. Because of the child’s “lethal anomaly,” the child’s
primary physician believed that aggressive treatment of the
pneumonias was inappropriate and wanted to place a do not
resuscitate (DNR) order in the chart. The surgeons refused
to perform a Nissen fundoplication and gastrostomy tube
(which would reduce gastroesophageal reflux and so help
avert aspiration pneumonia) because they believed it was
“futile” to do so. But the mother wanted all medical
treatment provided, including the surgery and mechanical
ventilation, and refused to authorize a DNR order.
The normative nature of the decision not to treat a child
like Baby B is evidenced by the tendency to talk about the
treatment of pneumonia as “aggressive”27 or to describe a
fundoplication as “futile.”28 These procedures would be
considered routine in an otherwise healthy infant. A
gastrostomy tube and Nissen fundoplication is “medically
indicated”29 in a neurologically normal child with severe
reflux; it becomes “aggressive” when the child has trisomy
18. Likewise, mechanical ventilation is “appropriate” for
respiratory failure in an otherwise healthy child; it was
“futile” for both Baby B and Baby K. The problem with such
terms, then, is that they hide the quality-of-life judgments
imbedded in the decision to withhold treatment, and thereby
interfere with the parents’ authority to define what is best
for their child.
It has been almost three decades since Duff and Campbell
made public that quality-of-life factors are integral to the
decision to withhold or withdraw treatment in neonatal
intensive care units. During this time, neonatal and
pediatric intensive care practices have changed
dramatically, both in their technical capabilities30 and in
their acceptance of greater parental participation in
decisionmaking. 31 It is now time for the language to catch
up. “Lethal anomalies” are no longer necessarily lethal, and
should not be described as such. Frank discussions about
quality of life are important, even if challenging, and can
facilitate parental decisionmaking.
Acknowledgement
We thank Walter Glannon for his advice and reflection.
Disclaimer
The opinions expressed in this article are those of the
authors and do not reflect the opinions or policies of the
National Human Genome Research Institute, the National
Institutes of Health, or the Department of Health and Human
Services.
Article References:
1. J.D. Horbar et al., “Trends in Mortality and Morbidity
for Very Low Birth Weight Infants, 1991-1999,” Pediatrics
110 (2002): 143-51; W.T. Mahle, “Neurologic and Cognitive
Outcomes in Children with Congenital Heart Disease,” Current
Opinion in Pediatrics 13 (2001): 482-86; D.B. McElhinney and
G. Wernovsky, “Outcomes of Neonates with Congenital Heart
Disease,” Current Opinion in Pediatrics 13 (2001): 104-110;
and F.C. Ryckman et al., “Biliary Atresia-Surgical
Management and Treatment Options as They Relate to Outcome,”
Liver Transplantation and Surgery 4, suppl. 1 (1998):
S24-S33.
2. R.E. Behrman et al., eds., Nelson’s Textbook of
Pediatrics. 14th ed. (Philadelphia: WB Saunders, 1992),
284-86; K.L. Jones, ed., Smith’s Recognizable Patterns of
Human Malformation, 5th ed. (Philadelphia: W.B. Saunders,
1997), 20-21, 659-61; S.J. Hassed et al., “Perinatal Lethal
Conditions: The Effect of Diagnosis on Decision Making,”
Obstetrics and Gynecology 82 (1993): 37-42; A.P. Bos et al.,
“Avoidance of Emergency Surgery in Newborns with Trisomy
18,” Lancet 339 (1992): 913-15; and D.C. Wertz and M.B.
Knoppers, “Serious Genetic Disorders: Can or Should They Be
Defined?” American Journal of Medical Genetics 108 (2002):
29-35.
3. S. Saigal et al., “Differences in Preferences for
Neonatal Outcomes among Health Care Professionals, Parents,
and Adolescents,” JAMA 281 (1999): 1991-97.
4. M.W. Doron et al., “Delivery Room Resuscitation Decisions
for Extremely Premature Infants,” Pediatrics 102 (1998):
574-82.
5. A.G. Randolph et al., “Factors Explaining Variability
among Caregivers in the Intent to Restrict Life-Support
Interventions in a Pediatric Intensive Care Unit,” Critical
Care Medicine 25 (1997): 435-9; and A.G. Randolph et al.
“Variability in Physician Opinion on Limiting Pediatric Life
Support,” Pediatrics 103, no. 4 (1999): e46.
6. Randolph, “Variability,” e46.
7. R.S. Duff and A.G.M. Campbell, “Moral and Ethical
Dilemmas in the Special Care Nursery,” NEJM 289 (1973):
890-94.
8. Ibid., p. 893.
9. The major viewpoints are discussed in R. Weir,
“Pediatricians and Selective Nontreatment” and “Options
Among Ethicists,” Selective Nontreatment of Handicapped
Newborns (New York: Oxford University Press, 1984).
50-59,143-87.
10. In re Infant Doe, No. GU8204-004A (Ind. Ct. App. 12 Apr
1982), writ of mandamus dismissed bus nom. State ex rel
Infant Doe v Baker, No. 482 $ 140 (Ind. Sup. Ct. 27 May
1982), cert. Denied, 104 S. Ct. 394 (1983).
11. DHHS Notice to Health Care Providers, “Discriminating
Against Handicapped by Withholding Treatment or
Nourishment,” Federal Register 47, no. 116 (June 16, 1982)
26027; DHHS Interim Final Rule, “Nondiscrimination on the
Basis of Handicap Relating to Healthcare for Handicapped
Infants,” Federal Register 48, no. 45 (5 July 1983),
30846-52; and DHHS Final Rule, “Nondiscrimination on the
Basis of Handicap: Procedures and Guidelines Relating to
Health Care for Handicapped Infants,” Federal Register 49,
no. 8 (12 January 1984), 1622-54.
12. American Academy of Pediatrics, National Association of
Children’s Hospitals and Related Institutions, Children’s
Hospital National Medical Center, Plaintiffs v. Margaret M.
Heckler. 561 F. Supp. 395; 1983 U.S. Dist. LEXIS 17713
(1983).
13. Child Abuse Amendments of 1984 42 USCA 5101-5103 (1984).
14. See, for example, L.M. Kopelman, T.G. Irons, and A.E.
Kopelman, “Neonatologists Judge the “Baby Doe” Regulations,”
NEJM 318 (1988): 677-83; R.L. Ariagno, “Government
Regulations in the United States” in Ethics and Perinatology,
eds. A. Goldworth et al. (New York: Oxford University Press,
1995), 279-95; and I.D. Todres et al., “Life-Saving Therapy
for Newborns: A Questionnaire Survey in the State of
Massachusetts,” Pediatrics 81 (1988): 643-49. The change in
neonatal practice can also be explained in part by the
disability rights activists who successfully argued that
nontreatment was discriminatory and by research that showed
that many families with disabled children rated their lives
quite positively. See P.M. Ferguson, A. Gartner, and D. K.
Lipsky, “The Experience of Disability in Families: A
Synthesis of Research and Parent Narratives” in Prenatal
Testing and Disability Rights, eds. E. Parens and A. Asch
(Washington DC: Georgetown University Press, 2000) 72-94.
15. See, child abuse amendments, 1984.
16. Ibid., p. 5102.
17. In the matter of Baby K. 16F3d 590 (4th Cir. 1994). J.J.
Paris, R. K. Crone, and F. Reardon, “Physicians’ Refusal of
Requested Treatment: The Case of Baby L,” NEJM 322 (1990):
1012-1013.
18. S.J. Youngner, “Who defines futility?” JAMA 260 (1988):
2094-2095. 19. B.J. Baty, “Natural History of Trisomy 18 and
Trisomy 13: II. Psychomotor Development,” American Journal
of Medical Genetics 49, no. 2 (1994): 189-94.
20. B.J. Baty, “Natural History of Trisomy 18 and Trisomy
13: I. Growth, Physical Assessment, Medical Histories,
Survival, and Recurrence Risk,” American Journal of Medical
Genetics 49, no. 2 (1994): 175-88.
21. R.C. Sachdeva et al., “Resource Consumption and the
Extent of Futile Care among Patients in a Pediatric
Intensive Care Unit Setting,” Pediatrics 128 (1996): 742-7.
22. See Duff and Campbell, “Moral and Ethical Dilemmas,”
1973.
23. N.K. Rhoden, “Treatment Dilemmas for Imperiled Newborns:
Why Quality of Life Counts,” Southern California Law Review
58 (1986) 1283-1347.
24. Ibid., pp.1312-18.
25. Cruzan v. Director Missouri Dept. of Health, 58 U.S. L.W.
4916 (1990). In re Quinlan, 70 N.J. 10, 355 A.2d 647 (1976)
26. C.E. Koop, “The Sanctity of Life,” Journal of the
Medical Society of New Jersey 75 (1978): 62-9. See Weir,
“Options among Ethicists.”
27. J.C. Weeks et al, “Relationship between Cancer Patients’
Predictions of Prognosis and Their Treatment Preferences.
JAMA 279 (1998): 1709-14; and C. Hammerman et al.,
“Decisionmaking in the Critically Ill Neonate: Cultural
Background v. Individual Life Experiences,” Journal of
Medical Ethics 23 (1997): 164-69.
28. R.S. Stein et al., “CPR-Not-Indicated and Futility,”
Annuals of Internal Medicine 124 (1996): 75-7; E.M. Levine,
“A New Predicament for Physicians: The Concept of Medical
Futility, the Physician’s Obligation to Render Inappropriate
Treatment, and the Interplay of the Medical Standard of
Care,” Journal of Law & Health 9 (1994-5) 69-108; and L.J.
Schneiderman, N.S. Jecker, and A.R. Jonsen, “Medical
Futility: Its Meaning and Ethical Implications,” Annuals of
Internal Medicine 112 (1990): 949-54.
29. The subjectivity of the concept “medically indicated” is
discussed by Veatch in “The Concept of ‘Medical
Indications’” in R.M. Veatch The Patient-Physician Relation:
The Patient as Partner, Part 2 (Bloomington, Indiana
University Press, 1991) 54-62.
30. See Horbar, “Trends in Mortality and Morbidity for Very
Low Birth Weight Infants, 1991-1999,”2002. See also A.C.
Chang, “Pediatric Cardiac Intensive Care: Current State of
the Art and Beyond the Millennium,” Current Opinion in
Pediatrics 12, no. 3 (2000): 238-46; and J.M. Tilford et
al., “Differences in Pediatric ICU Mortality Risk over
Time,” Critical Care Medicine 26 (1998): 1737-43.
31. H. Harrison, “The Principles for Family-Centered
Neonatal Care,” Pediatrics 92 (1993): 643-50; and N.E. Page
and N.M. Boeing NM, “Visitation in the Intensive Care Unit:
Controversy and Compromise,” AACN Clinical Issues in
Critical Care Nursing 5, no. 3 (1994): 289-95.March-April
2003 HASTINGS CENTER REPORT 41
RESOURCE SECTION:
© The Hastings Center. Reprinted with permission. This
article originally appeared in the Hastings Center Report,
vol. 33, no. 2 (2003),” with the link from “The Hastings
Center” leading to our home page (http://www.thehastingscenter.org).
Any correspondence generated from this (if there is any) can
be directed to our general editorial email address
(editorial@thehastingscenter.org), or to the authors
themselves.