- Email: [email protected]
Ethics, Data, and Policy in Newborn Intensive Care
C H A P T E R
3
Ethics, Data, and Policy in Newborn Intensive Care William L. Meadow and John D. Lantos
PHILOSOPHY Ethics in the neonatal intensive care unit (NICU), as in all clinical contexts, starts with the traditional triangular framework of autonomy (do what the patient, or in this case the parent, thinks is right), paternalism (do what the doctor thinks is right), and beneficence and nonmaleficence (do the right thing). These concepts, independent of context or data, are timeless. The problem with timeless concepts, of course, comes in knowing what to do in real time. What exactly is the right thing? What facts should be brought to bear in the decision? What weight should each fact be given? And whose opinion should count in the end? Nonetheless, some applications of traditional ethical concepts in the NICU are already universally adopted (e.g., avoid futility, do not torture, and intervene when the data provide compelling evidence to do so). Unfortunately, much of NICU care falls between not resuscitating 21-week births, obligatory support of 28-week births, and not performing cardiopulmonary resuscitation on infants with lethal anomalies. The traditional ethical solution to medical dilemmas is to ground concerns in context, take data into account, and be sympathetic to patient preferences when the balance of benefits and burdens is not clear. The precise problem in the NICU is that the burdens are real, immediate, long term, and significant—months of painful procedures such as intubation, ventilation, intravenous catheterization, and any permanent sequelae that might ensue—whereas the benefits of NICU interventions are distant, statistical, and unpredictable. Moreover, NICU success is often viewed as “all or none”—that is, in most of the NICU follow-up literature a Bayley mental developmental index MDI or psychomotor developmental index PDI greater than 70 is classified as normal, whereas an MDI or PDI less than 70 is classified as an adverse outcome. Faced with a difficult case, it is rare that simply applying principles will help to devise a solution. Difficult cases are usually ones in which the principles themselves are in conflict, or their application to the case is ambiguous.
DATA What kind of information would parents, physicians, or judges want to know about the NICU? The essential truth at the intersection of NICU epidemiology and ethics is that survival depends sharply on gestational age (GA), within relatively precise boundaries. In the United States, as in virtually all the industrialized world, infants born after 27 weeks’ gestation have, from any ethical perspective, no increased mortality over infants born at term. Consequently, for these infants, the ethical principle of best interests requires their resuscitation, in the same way that sick children born at term deserve resuscitation. 18
Conversely, for infants born before 22 weeks’ gestation, survival is essentially zero. Consequently, these infants and their parents deserve our compassion, but not our interventions, on the ethical grounds of strict futility. In between, spanning roughly one gestational month, from 23 to 26 weeks, we will require not just data, but interpretation. First, there is the intriguing finding that GA-specific mortality for infants resuscitated in this gestational range has more or less reached a plateau. Within 10 to 20 years ago, the “border of viability” was steadily decreasing, thus improving outcomes for infants in the gray zone. Not so much, recently, and there is little reason to think that things will change for infants born in this gestational range in the foreseeable future. Nonetheless, epidemiologic progress has been made. Tyson et al (2008), using the vast data base of the National Institute of Child Health and Human Development network, attempted to go “beyond gestational age” and quantify additional risk factors for both mortality and neurologic morbidity in infants born on the cusp of viability. The analysis revealed that singleton status, appropriate in utero growth, antenatal steroids, and female gender all improve the likelihood of survival and intact neurologic outcome, independent of GA. Tyson’s algorithm, using information available at the time of birth, significantly improved predictive value and sensitivity of prognostication over GA alone. However, two problems remain. First, for many infants the predictive value of the Tyson algorithm is still not very good—that is, many of the lower-risk patients will still die, and many of the higher-risk patients will survive. Second, the Tyson algorithm, like GA, ignores a potentially important feature of NICU care—time. The algorithm stops accumulating data at the time of birth, and it does not account for prognostic features that might become available as the infant’s course unfolds in the NICU. In theory, making decisions over the course of time offers two advantages over prognostication in the delivery room. First, parents often appreciate the opportunity to get to know their baby as an individual, as opposed to evaluating anonymous population-based prognostications at the time of birth. Second, there is valuable information to learn while the baby is in the NICU. Two time-sensitive prognostic features have been evaluated in the context of infants born at the border of viability—serial illness severity algorithms (Score for Neonatal Acute Physiology [SNAP] scores) and serial that intuitions that the patient would “die before NICU discharge” (Meadow et al, 2008). Unfortunately, although SNAP scores on the first day of life have good prognostic power for death or survival, their power diminishes over time. Intriguingly, serial intuitions that an individual baby will die before discharge—offered by medical caretakers for
CHAPTER 3 Ethics, Data, and Policy in Newborn Intensive Care
patients who require mechanical ventilation and for whom there is an ethical alternative to continued ventilation, namely extubation and palliative care—are remarkably accurate in predicting a combined outcome of either death or survival with neurologic impairment (MDI or PDI <70). Children with abnormal results from a cranial ultrasound examination and corroborated predictions of death have a less than 5% chance of surviving with both MDI and PDI greater than 70 at 2 years, independent of their gestational age. The predictive power of these data, acquired over time during an individual infant’s NICU course, though not perfect, is greater than any algorithm available at the time of birth. What do prospective parents or medical caretakers consider when they are asked to decide whether or not to resuscitate their micro-premie? A fascinating insight has been offered by Janvier et al, (2008), who have done extensive surveys comparing responses to requests for resuscitation of sick micro-premies with resuscitation of comparably sick patients at other ages (from term infants to 80-year-olds). Consistently, it appears that micropremies are devalued—that is, for comparable likelihood of survival and comparable likelihood of neurologic morbidity in survivors, more people would let a micro-premie die first, or at least offer to resuscitate them last. There is no theory to account for these findings. Indeed, for male infants born at 23 weeks’ gestation, the likelihood of intact survival (i.e., neither death in the NICU nor permanent neurologic morbidity) is less than 10%. Given that the likelihood of burdensome therapy is 100%, how can resuscitation be justified? Intriguingly, combining the outcomes of death in the NICU with abnormal neurologic function in later life may not reflect the emotional reality of many NICU parents. For these parents, death in the NICU is not necessarily the worst outcome, because acknowledging that the baby was too small might be better consolation than not ever giving the baby a chance at life. If trying and failing is seen as a positive process, then the long-term burden of NICU care for parents should perhaps be calculated not as a function of all live births (the current practice), but as a function of infants who survive to discharge. Numerous studies analyzing various populations in several countries have converged on the same surprising observation: the incidence of neurologic morbidity in NICU survivors is not very different when comparing infants at 23, 24, 25, and 26 to 27 weeks’ gestation. The essential epidemiologic difference for infants born in this gestational range appears to be whether the baby will survive at all. Once the baby leaves the NICU, the risk of severe morbidity is largely the same; this is true in singlecenter and multicenter studies, in the United Kingdom, Canada, Europe, and the United States (Johnson et al, 2009; Tyson et al, 2008). Paradoxically, if avoiding survival with permanent crippling neurologic injury is the driving force behind resuscitation decisions, it appears that we should not be worrying about 23- and 24-weekers–rather we should not be resuscitating 26- or 27-weekers, since many more of them will survive, and survive with disability. But that seems very odd. Finally, there is epidemiologic difficulty in assigning value to morbidity in surviving infants in the NICU. Verrips et al (2008) have attempted to assess the effects of permanent
19
residual disability for NICU survivors and their immediate families; they have demonstrated consistently that children with disabilities and their parents place a much higher value on their lives, and the quality of those lives, than do either physicians or NICU nurses. The vast majority of infants who survive the NICU, even those with significant permanent neurologic compromise, have “lives worth living,” as judged by the people most affected by those lives. GA-specific mortality seems to preclude resuscitation for infants born before 22 weeks’ gestation and require resuscitation for infants born after 27 weeks’ gestation. In between, the outcomes are murky, prognostic indices are imperfect, and sociologic analyses of human behaviors (of parents and physicians) appear inadequate to develop any uniform approach that is satisfactory.
POLICY Neonatologists in the United States are sued for medical negligence approximately once every 10 years, and the average U.S. neonatologist will be sued more than once during his or her career (Meadow et al, 1997). Other than case-specific failures or oversights, are there any overarching themes that have arisen from medical malpractice cases in neonatology? In the United States most malpractice allegations are state based, as opposed to federal cases. Consequently, the decisions of lower state courts, or even state supreme courts, are not binding in any other state, but they can be informative. The first important legal case in neonatology in the United States was Miller v HCA (2003). In 1990, Mrs. Miller came to the Hospital Corporation of America (HCA) in Texas in labor at 23 weeks’ gestation. The fetus was estimated to weigh 500 to 600 g. No baby born that size had ever survived at that hospital. Mrs. Miller, her husband, and the attending physicians agreed that the baby was previable and that no intervention was indicated. The baby was born, but a different physician performed resuscitation, and the infant survived with brain damage. As a result, the Millers sued the hospital for a breach of informed consent, and they were awarded $50 million by a trial jury. The case wound its way to the Texas Supreme Court, which dismissed the verdict and articulated an “emergency exception” for physicians—that is, if a Texas physician finds himself or herself in the emergency position of needing to resuscitate a patient to prevent immediate death, the physician can try to perform resuscitation without being obligated to obtain consent from anyone. Whether it would be acceptable for a physician not to perform resuscitation in an emergency was left unarticulated by the Texas court. In Wisconsin, the case of Montalvo v Borkovec (2002) took the legal obligations of neonatologists and parents to a different place. The case derived from the resuscitation of a male infant born between 234⁄7 and 242⁄7 weeks’ gestation, weighing 679 g. The parents claimed a violation of informed consent, arguing that the decision to use “extraordinary measures” should have been relegated to the parents. The Wisconsin Appellate Court disagreed, holding that “in the absence of a persistent vegetative state, the right of a parent to withhold life-sustaining treatment from a child does not exist.” Because virtually no infant is
20
PART I Overview
born in a persistent vegetative state, this decision would apparently eliminate the ethical possibility in Wisconsin of a “gray zone” of parental discretion. No other jurisdiction in the United States has adopted this position. The Wisconsin Appellate Court, like the Texas Supreme Court, was silent on whether physicians have discretion not to resuscitate. However, in Texas and Wisconsin, physicians are apparently not liable if they choose to do so. A number of other state courts have addressed issues of treatment or nontreatment. In general, the courts are permissive of physicians who resuscitate infants. If courts are asked to sanction decisions to allow infants to die, most will do so only if there is consensus among physicians and parents, and occasionally ethics committees. Courts are not eager to punish physicians who treat infants over parental objections or to empower physicians to stop treatment when parents want it to continue.
BABY DOE REGULATIONS Of course, state-by-state civil malpractice cases are not the only administrative means for redrawing the interface between neonatal medicine and the law. The federal government has contributed as well. The Baby Doe Regulations were one of the first attempts to codify and impose a federal vision of appropriate ethical behavior in the NICU. In 1982, a baby with Down syndrome and esophageal atresia was born in Bloomington, Indiana. At that time, the standard of care for babies with Down syndrome was shifting. A decade or two earlier, most babies with Down syndrome who survived infancy were institutionalized. If babies needed surgery for an anomaly like Baby Doe’s atresia, the parents were given the option. Approximately half chose surgery and half chose palliative care (Shaw, 1973). Similarly, half of pediatricians thought that palliative care, rather than surgery, was the better option (Todres et al, 1977). By the early 1980s, this practice had started to change. More parents opted for active intervention to save their babies (Shepperdson, 1983), and more physicians believed that withholding treatment was medical neglect (Todres et al, 1988). Baby Doe was born into this shifting cultural milieu. Baby Doe’s parents refused to consent to surgery and chose palliative care instead. The pediatrician alerted the state child protection agency, which investigated the case. At a court hearing, the parents claimed that they were following the advice of their obstetrician and not their pediatrician. The court found that parents only had to follow the advice of a licensed physician and that, since they were doing so, they were not neglectful. The court did not take protective custody. The doctor and hospital appealed. The Indiana Supreme Court refused to hear the appeal, and the baby died after 8 days (Lantos, 1987). These facts became publicly known and led to a national controversy that eventually reached the Oval Office. President Reagan demanded federal action. It was difficult to decide which action should be taken, because the federal government has no jurisdiction over child abuse and neglect; it is the domain of the states. The federal government oversees civil rights enforcement, however, and the Reagan administration devised a legal strategy that defined not treating babies with Down syndrome or other congenital anomalies as discrimination against people with disabilities, rather than
medical neglect. This strategy gave the federal government a justification for oversight. The new regulations were implemented, and bright red signs containing federal hotline telephone numbers were posted in NICUs across the country. These signs proclaimed that withholding treatment on the basis of disability was a federal civil rights violation (Annas, 1984) and that federal investigative squads could review medical records to determine whether discrimination had taken place. These regulations were challenged and eventually struck down by the U.S. Supreme Court (Annas, 1984). A diluted version of the original Baby Doe guidelines was eventually incorporated into the federal Child Abuse and Treatment Act (Annas, 1986; Kopelman, 1988). That law, however, is primarily a funding mechanism to channel federal funds to state child protection agencies; it is not a regulation that can be enforced for physicians or hospitals. The Baby Doe regulations do not exist today and have not existed since 1984; however, they still hold symbolic power. The mention of Baby Doe strikes fear in the hearts of pediatricians who lived through the events, in part because pediatricians had made pediatricians into villains in the societal battle over child protection. There is a certain irony in the controversy over Baby Doe regulations. The original goal—to decrease the range of cases in which withholding treatment of newborns is permissible—did not need federal input. Progress was already being made. Many diseases that used to be considered incompatible with life or that were seen as leading to an unacceptable quality of life were being treated routinely. Metabolic therapies for genetic diseases (phenylketonuria [PKU], hypothyroidism, Gaucher disease) and surgical therapies for organ dysgenesis (congenital heart corrections, congenital diaphragmatic hernia (CDH) repair, extra corporeal membrane oxygenation (ECMO), dialysis) have been forcefully advanced by medical subspecialists in their corresponding fields (e.g., genetics, cardiac surgery, general surgery). With few exceptions (hypoplastic left ventricle being one of the best recognized), these advances have been relatively uncontroversial. Once sufficient data are gathered to demonstrate moderate efficacy, the innovations are widely and rapidly adopted. There is still controversy when treatments enable survival but have a high likelihood, or certainty, that survival will be accompanied by severe neurologic impairment. As a result, two questions must be asked. First, how severe will the neurologic impairment be? Second, what is the likelihood that the child will have the most severe possible impairment? The prognostic spectrum of Down syndrome is broad, but few infants with trisomy 21 are severely impaired. In contrast, almost all infants with trisomy 13 or 18 either die in infancy or are left with profound neurologic impairment. The outcomes for these chromosomal anomalies can be used to define the spectrum within which clinical decisions are made. For babies whose outcomes are likely to be similar to those seen in trisomy 21, it is no longer permissible to withhold life-sustaining treatment. For babies whose outcomes are likely to be similar to babies with trisomy 13, it is permissible to withhold or withdraw life-sustaining treatment and offer palliative care instead. The calculus becomes more complex in conditions associated with a wider range of outcomes, such as extreme prematurity or high myelomenigocele with hydrocephalus.
CHAPTER 3 Ethics, Data, and Policy in Newborn Intensive Care
The shift in moral standards regarding babies with Down syndrome was not related to technology, but rather sociology. The capacity to repair Arnold–Chiari malformation and duodenal atresia existed long before it was applied to children with myelomeningocele and Down syndrome. What has changed the mood of the country is a growing recognition that disability is as much a social construct as a medical construct, although it is always both and not one or the other.
BORN ALIVE INFANT PROTECTION ACT In 2002, the U.S. Congress passed a law called the Born Alive Infant Protection Act (BAIPA). BAIPA, like the discredited Baby Doe regulations, was an attempt to insert federal values into medical deliberations. There are some interesting similarities and distinctions between Baby Doe and BAIPA. The Baby Doe regulations addressed disability, whereas BAIPA applies to abortion. BAIPA declares that “for the purposes of Federal law, the words ‘person’, ‘human being’, ‘child’, and ‘individual’ shall include every infant who is born alive at any stage of development.” As Sayeed (2005) noted, “The agency arguably substitutes a nonprofessional’s presumed sagacious assessment of survivability for reasonable medical judgment.” It is unclear what the implications of this law have been. On the one hand, all infants born alive before BAIPA were also treated as human beings; however, that did not necessarily mean that they received all available life support and resuscitation. After all, patients can have do-not-resuscitate orders or receive palliative care rather than intensive care. Still, the purpose of BAIPA seemed to be less about the treatment of babies and more about restrictions on abortion. Although some authors have expressed concern that the influence of BAIPA may transform neonatal care of infants born at, or even below, the threshold of viability, it appears to have had little measurable effect to date. Partridge et al (2009) surveyed neonatologists in California and found that they were concerned about the implications of BAIPA. They write, “If this legislation were enforced, respondents predicted more aggressive resuscitation potentially increasing risks of disability or delayed death.” There have been no cases to date in which BAIPA has been invoked or in which physicians and hospitals have been found to be in violation of its requirements.
FUTURE DIRECTIONS There is no new technology in development that appears likely to affect outcomes significantly. Consequently, for infants who receive resuscitation in the delivery room, birthweight-specific mortality and morbidity are unlikely to change much in the near future. Nonetheless, three developments may change the way we think about newborns, and consequently shift the terrain of neonatal bioethics.
HIGH RISK MATERNAL-FETAL MEDICINE CENTERS Many children’s hospitals are now developing high-risk, maternal-fetal medicine centers. The goal of these centers is to identify fetuses at risk—particularly those with
21
congenital anomalies—and to care for those fetuses and their mothers in centers where there is expertise in fetal diagnosis, therapy, and neonatal care. The hope is that such centers will allow more timely, and therefore more effective, intervention for babies with congenital heart disease, congenital diaphragmatic hernia, or other anomalies. The medical effectiveness of fetal centers will depend on two distinct developments. First, on a population basis, these centers will only be as effective as fetal screening and diagnosis. The existence of these centers will almost certainly create an expectation and a demand for better fetal screening. Such screening is likely to include both better imaging and better screening tests that can be performed on maternal blood; both will lead to earlier diagnosis of fetal anomalies. These diagnoses will create more complex dilemmas for perinatologists and parents who will need to decide, in any particular case, whether to terminate the pregnancy, offer fetal therapy, or offer either palliative care or interventions after birth. Ironically, better fetal diagnosis may increase the likelihood of pregnancy termination, even when postnatal treatment is possible, such as in hypoplastic left heart syndrome. Second, the effectiveness of fetal centers will depend on the effectiveness of fetal interventions. Perhaps surprisingly, other than in utero transfusion for Rhesus disease or vascular ablation for twin-twin transfusion syndromes— neither of which are particularly new and neither of which is performed by pediatric surgeons or pediatricians—there is little evidence that any fetal intervention has had any effect on any neonatal outcome. This lack of demonstrated effectiveness has, thus far, not suppressed the proliferation of fetal intervention centers. There may be other factors, including institutional prestige, finances, and recruitment of “desirable” patients.
EXPANDED NEWBORN SCREENING In recent years, the number of diseases and conditions that can be diagnosed through newborn screening has expanded dramatically. Such screening is under the purview of states, rather than the federal government, and there is wide variation in the number of tests that are performed. In 1995 the average number of tests per state was five (range: zero to eight disorders). Between 1995 and 2005 most states added tests, so that the average number of screening tests done by 2005 was 24 (Tarini et al, 2006). The expansion of newborn screening raises three problems. First, even the most accurate test has false positives. For rare conditions, the percentage of positive tests that are false positives is increased. Thus, the more rare conditions that are added to a newborn screening panel, the more false positives there will be. False positives are associated with considerable parental anxiety and can lead to potentially dangerous and unnecessary diagnostic procedures or treatments. Second, expanded newborn screening costs money. Interestingly, the tests themselves are astoundingly inexpensive, which is why policy makers are tempted to add more to the panels. However, the follow-up counseling and testing after positive tests are expensive, and without such followup the screening programs will not work. The Centers for Disease Control and Prevention has recently expressed concern about these costs (Centers for Disease Control
22
PART I Overview
and Prevention, 2008). Finally, there is the potential for discrimination against patients for whom documented heterozygous carrier status conveys no recognized medical infirmity, but social or psychological stigma may be real. There is little funding available to assist or counsel these patients.
FINANCIAL CONSTRAINTS The American Academy of Pediatrics guidelines on neonatal resuscitation suggest that resuscitation should be obligatory at 25 weeks’ gestation or greater, optional at 24 weeks’ gestation, and unusual at 23 weeks’ gestation. These recommendations are thought to reflect the best understanding of both the ethical discussion and the epidemiologic facts surrounding NICU outcomes. The recommendations purport to reflect the traditional paradigm that data drive policy. However, particularly in the context of NICU survival for infants at 23 to 25 weeks’ gestation, there is good evidence that causation can work in the reverse direction—that is, policy drives data. In the Netherlands, Canada, and some parts of Oregon, survival at 25 weeks’ gestation is comparable and comparably good; more than 50% of infants born at 25 weeks’ gestation will survive to discharge. However, in the Netherlands, virtually no infant survives birth at 24 weeks’ gestation, whereas in Canada and Oregon the survival rate is 40%. In addition, in Oregon and some parts of Canada, survival at 23 weeks’ gestation is close to zero, whereas in other parts of Canada and the United States the survival rate is 20% or greater. Why? Certainly not because the Dutch, Canadians, or Oregonians have forgotten how to resuscitate small infants. Rather, they have chosen not to. Why not? Perhaps it is non-maleficence–the fear that survival with permanent neurologic morbidity may be cruel to the child, the family, or society at large. However, the incidence of neurologic morbidity in survivors is not different when comparing infants at 23, 24, 25, and 26 to 27 weeks’ gestation. If the fear of a permanent crippling neurologic injury is the driving force, we should not be resuscitating 26 or 27 weekers, since many more of them will survive, and survive with disability. But that seems odd. The approach in the Netherlands is consistent; there is a limited budget and a communitarian ethic. There is a certain rationale behind spending money on all pregnant women, instead of 1% of micro-premies. The United States appears ambivalent–we value individuals over community, are fascinated with high-technology, and claim to prize our children. On the other hand, we will not spend money to prevent unwanted teen pregnancy or to provide visiting nurses for new mothers. Finally the concept of generational conflict must be considered. We appear quite comfortable calling
delivery-room resuscitation of 24 weekers “optional,” based on gestational age alone. It is difficult to imagine the AMA recommending that resuscitation for 85-year-olds who come to the emergency department is “optional,” based on age alone. NICU care is often criticized as “too expensive.” We ask, “Compared to what?” Fewer than 5% of infants will be in the NICU for more than a short stay. The vast majority of these patients will survive. Only 24,000 infants of 4 million births die each year in the United States, and half of these will die within fewer than 7 days. In contrast, nearly 3 million adults will die in the United States each year and one third of these will have been admitted to an medical intensive care unit (MICU) in the 6 months before dying. MICU costs outpace NICU costs by at least 100 to 1.
SUMMARY Ethical philosophy is a place to start, not a place to finish. Data are relatively easy to acquire and agree on. Policy is intriguingly insensitive to data, but that may reflect social and political realities that exist beyond the NICU— perceptions of disability, abortion politics, individual versus communitarian emphasis, fascination with technology, discrimination, publicity, financial constraints—so that an ethical course of action in one country, one city, or one family might seem perverse elsewhere.
SUGGESTED READINGS Annas GJ: The Baby Doe regulations: governmental intervention in neonatal rescue medicine, Am J Public Health 74:618-620, 1984. Born-Alive Infants Protection Act of 2001: Report together with additional and dissenting views of the House Committee on the Judiciary, 107th Congress, 1st Session, August 2, 2001. 1-38, 3. (Purpose and Summary). Centers for Disease Control and Prevention: Impact of expanded newborn screening: United States, 2006, MMWR Morb Mortal Wkly Rep 57:1012-1015, 2008. Janvier A, Leblanc I, Barrington KJ: The best-interest standard is not applied for neonatal resuscitation decisions, Pediatrics 121:963-969, 2008. Johnson S, Fawke J, Hennessy E, et al: Neurodevelopmental disability through 11 years of age in children born before 26 weeks of gestation, Pediatrics 124:E249-E257. Lantos J: Baby Doe five years later: implications for child health, N Engl J Med 317:444-447, 1987. Meadow W, Lagatta J, Andrews B, et al: Just in time: ethical implications of serial predictions of death and morbidity for ventilated premature infants, Pediatrics 121:732-740, 2008. Miller v HCA, Inc, 118 S.W. 3d 758, 771 (Texas 2003). Montalvo v Borkovec, 647 N.W. 2d 413 (Wis. App. 2002). Shepperdson B: Abortion and euthanasia of Down’s syndrome children: the parents’ view, J Med Ethics 9:152-157, 1983. Todres ID, Guillemin J, Grodin MA, et al: Life-saving therapy for newborns: a questionnaire survey in the state of Massachusetts, Pediatrics 81:643-649, 1988. Tyson JE, Parikh NA, Langer J, et al: National Institute of Child Health and Human Development Neonatal Research Network. Intensive care for extreme prematurity: moving beyond gestational age, N Engl J Med 358:1672-1681, 2008. Complete references used in this text can be found online at www.expertconsult.com
3
Ethics, Data, and Policy in Newborn Intensive Care William L. Meadow and John D. Lantos
PHILOSOPHY Ethics in the neonatal intensive care unit (NICU), as in all clinical contexts, starts with the traditional triangular framework of autonomy (do what the patient, or in this case the parent, thinks is right), paternalism (do what the doctor thinks is right), and beneficence and nonmaleficence (do the right thing). These concepts, independent of context or data, are timeless. The problem with timeless concepts, of course, comes in knowing what to do in real time. What exactly is the right thing? What facts should be brought to bear in the decision? What weight should each fact be given? And whose opinion should count in the end? Nonetheless, some applications of traditional ethical concepts in the NICU are already universally adopted (e.g., avoid futility, do not torture, and intervene when the data provide compelling evidence to do so). Unfortunately, much of NICU care falls between not resuscitating 21-week births, obligatory support of 28-week births, and not performing cardiopulmonary resuscitation on infants with lethal anomalies. The traditional ethical solution to medical dilemmas is to ground concerns in context, take data into account, and be sympathetic to patient preferences when the balance of benefits and burdens is not clear. The precise problem in the NICU is that the burdens are real, immediate, long term, and significant—months of painful procedures such as intubation, ventilation, intravenous catheterization, and any permanent sequelae that might ensue—whereas the benefits of NICU interventions are distant, statistical, and unpredictable. Moreover, NICU success is often viewed as “all or none”—that is, in most of the NICU follow-up literature a Bayley mental developmental index MDI or psychomotor developmental index PDI greater than 70 is classified as normal, whereas an MDI or PDI less than 70 is classified as an adverse outcome. Faced with a difficult case, it is rare that simply applying principles will help to devise a solution. Difficult cases are usually ones in which the principles themselves are in conflict, or their application to the case is ambiguous.
DATA What kind of information would parents, physicians, or judges want to know about the NICU? The essential truth at the intersection of NICU epidemiology and ethics is that survival depends sharply on gestational age (GA), within relatively precise boundaries. In the United States, as in virtually all the industrialized world, infants born after 27 weeks’ gestation have, from any ethical perspective, no increased mortality over infants born at term. Consequently, for these infants, the ethical principle of best interests requires their resuscitation, in the same way that sick children born at term deserve resuscitation. 18
Conversely, for infants born before 22 weeks’ gestation, survival is essentially zero. Consequently, these infants and their parents deserve our compassion, but not our interventions, on the ethical grounds of strict futility. In between, spanning roughly one gestational month, from 23 to 26 weeks, we will require not just data, but interpretation. First, there is the intriguing finding that GA-specific mortality for infants resuscitated in this gestational range has more or less reached a plateau. Within 10 to 20 years ago, the “border of viability” was steadily decreasing, thus improving outcomes for infants in the gray zone. Not so much, recently, and there is little reason to think that things will change for infants born in this gestational range in the foreseeable future. Nonetheless, epidemiologic progress has been made. Tyson et al (2008), using the vast data base of the National Institute of Child Health and Human Development network, attempted to go “beyond gestational age” and quantify additional risk factors for both mortality and neurologic morbidity in infants born on the cusp of viability. The analysis revealed that singleton status, appropriate in utero growth, antenatal steroids, and female gender all improve the likelihood of survival and intact neurologic outcome, independent of GA. Tyson’s algorithm, using information available at the time of birth, significantly improved predictive value and sensitivity of prognostication over GA alone. However, two problems remain. First, for many infants the predictive value of the Tyson algorithm is still not very good—that is, many of the lower-risk patients will still die, and many of the higher-risk patients will survive. Second, the Tyson algorithm, like GA, ignores a potentially important feature of NICU care—time. The algorithm stops accumulating data at the time of birth, and it does not account for prognostic features that might become available as the infant’s course unfolds in the NICU. In theory, making decisions over the course of time offers two advantages over prognostication in the delivery room. First, parents often appreciate the opportunity to get to know their baby as an individual, as opposed to evaluating anonymous population-based prognostications at the time of birth. Second, there is valuable information to learn while the baby is in the NICU. Two time-sensitive prognostic features have been evaluated in the context of infants born at the border of viability—serial illness severity algorithms (Score for Neonatal Acute Physiology [SNAP] scores) and serial that intuitions that the patient would “die before NICU discharge” (Meadow et al, 2008). Unfortunately, although SNAP scores on the first day of life have good prognostic power for death or survival, their power diminishes over time. Intriguingly, serial intuitions that an individual baby will die before discharge—offered by medical caretakers for
CHAPTER 3 Ethics, Data, and Policy in Newborn Intensive Care
patients who require mechanical ventilation and for whom there is an ethical alternative to continued ventilation, namely extubation and palliative care—are remarkably accurate in predicting a combined outcome of either death or survival with neurologic impairment (MDI or PDI <70). Children with abnormal results from a cranial ultrasound examination and corroborated predictions of death have a less than 5% chance of surviving with both MDI and PDI greater than 70 at 2 years, independent of their gestational age. The predictive power of these data, acquired over time during an individual infant’s NICU course, though not perfect, is greater than any algorithm available at the time of birth. What do prospective parents or medical caretakers consider when they are asked to decide whether or not to resuscitate their micro-premie? A fascinating insight has been offered by Janvier et al, (2008), who have done extensive surveys comparing responses to requests for resuscitation of sick micro-premies with resuscitation of comparably sick patients at other ages (from term infants to 80-year-olds). Consistently, it appears that micropremies are devalued—that is, for comparable likelihood of survival and comparable likelihood of neurologic morbidity in survivors, more people would let a micro-premie die first, or at least offer to resuscitate them last. There is no theory to account for these findings. Indeed, for male infants born at 23 weeks’ gestation, the likelihood of intact survival (i.e., neither death in the NICU nor permanent neurologic morbidity) is less than 10%. Given that the likelihood of burdensome therapy is 100%, how can resuscitation be justified? Intriguingly, combining the outcomes of death in the NICU with abnormal neurologic function in later life may not reflect the emotional reality of many NICU parents. For these parents, death in the NICU is not necessarily the worst outcome, because acknowledging that the baby was too small might be better consolation than not ever giving the baby a chance at life. If trying and failing is seen as a positive process, then the long-term burden of NICU care for parents should perhaps be calculated not as a function of all live births (the current practice), but as a function of infants who survive to discharge. Numerous studies analyzing various populations in several countries have converged on the same surprising observation: the incidence of neurologic morbidity in NICU survivors is not very different when comparing infants at 23, 24, 25, and 26 to 27 weeks’ gestation. The essential epidemiologic difference for infants born in this gestational range appears to be whether the baby will survive at all. Once the baby leaves the NICU, the risk of severe morbidity is largely the same; this is true in singlecenter and multicenter studies, in the United Kingdom, Canada, Europe, and the United States (Johnson et al, 2009; Tyson et al, 2008). Paradoxically, if avoiding survival with permanent crippling neurologic injury is the driving force behind resuscitation decisions, it appears that we should not be worrying about 23- and 24-weekers–rather we should not be resuscitating 26- or 27-weekers, since many more of them will survive, and survive with disability. But that seems very odd. Finally, there is epidemiologic difficulty in assigning value to morbidity in surviving infants in the NICU. Verrips et al (2008) have attempted to assess the effects of permanent
19
residual disability for NICU survivors and their immediate families; they have demonstrated consistently that children with disabilities and their parents place a much higher value on their lives, and the quality of those lives, than do either physicians or NICU nurses. The vast majority of infants who survive the NICU, even those with significant permanent neurologic compromise, have “lives worth living,” as judged by the people most affected by those lives. GA-specific mortality seems to preclude resuscitation for infants born before 22 weeks’ gestation and require resuscitation for infants born after 27 weeks’ gestation. In between, the outcomes are murky, prognostic indices are imperfect, and sociologic analyses of human behaviors (of parents and physicians) appear inadequate to develop any uniform approach that is satisfactory.
POLICY Neonatologists in the United States are sued for medical negligence approximately once every 10 years, and the average U.S. neonatologist will be sued more than once during his or her career (Meadow et al, 1997). Other than case-specific failures or oversights, are there any overarching themes that have arisen from medical malpractice cases in neonatology? In the United States most malpractice allegations are state based, as opposed to federal cases. Consequently, the decisions of lower state courts, or even state supreme courts, are not binding in any other state, but they can be informative. The first important legal case in neonatology in the United States was Miller v HCA (2003). In 1990, Mrs. Miller came to the Hospital Corporation of America (HCA) in Texas in labor at 23 weeks’ gestation. The fetus was estimated to weigh 500 to 600 g. No baby born that size had ever survived at that hospital. Mrs. Miller, her husband, and the attending physicians agreed that the baby was previable and that no intervention was indicated. The baby was born, but a different physician performed resuscitation, and the infant survived with brain damage. As a result, the Millers sued the hospital for a breach of informed consent, and they were awarded $50 million by a trial jury. The case wound its way to the Texas Supreme Court, which dismissed the verdict and articulated an “emergency exception” for physicians—that is, if a Texas physician finds himself or herself in the emergency position of needing to resuscitate a patient to prevent immediate death, the physician can try to perform resuscitation without being obligated to obtain consent from anyone. Whether it would be acceptable for a physician not to perform resuscitation in an emergency was left unarticulated by the Texas court. In Wisconsin, the case of Montalvo v Borkovec (2002) took the legal obligations of neonatologists and parents to a different place. The case derived from the resuscitation of a male infant born between 234⁄7 and 242⁄7 weeks’ gestation, weighing 679 g. The parents claimed a violation of informed consent, arguing that the decision to use “extraordinary measures” should have been relegated to the parents. The Wisconsin Appellate Court disagreed, holding that “in the absence of a persistent vegetative state, the right of a parent to withhold life-sustaining treatment from a child does not exist.” Because virtually no infant is
20
PART I Overview
born in a persistent vegetative state, this decision would apparently eliminate the ethical possibility in Wisconsin of a “gray zone” of parental discretion. No other jurisdiction in the United States has adopted this position. The Wisconsin Appellate Court, like the Texas Supreme Court, was silent on whether physicians have discretion not to resuscitate. However, in Texas and Wisconsin, physicians are apparently not liable if they choose to do so. A number of other state courts have addressed issues of treatment or nontreatment. In general, the courts are permissive of physicians who resuscitate infants. If courts are asked to sanction decisions to allow infants to die, most will do so only if there is consensus among physicians and parents, and occasionally ethics committees. Courts are not eager to punish physicians who treat infants over parental objections or to empower physicians to stop treatment when parents want it to continue.
BABY DOE REGULATIONS Of course, state-by-state civil malpractice cases are not the only administrative means for redrawing the interface between neonatal medicine and the law. The federal government has contributed as well. The Baby Doe Regulations were one of the first attempts to codify and impose a federal vision of appropriate ethical behavior in the NICU. In 1982, a baby with Down syndrome and esophageal atresia was born in Bloomington, Indiana. At that time, the standard of care for babies with Down syndrome was shifting. A decade or two earlier, most babies with Down syndrome who survived infancy were institutionalized. If babies needed surgery for an anomaly like Baby Doe’s atresia, the parents were given the option. Approximately half chose surgery and half chose palliative care (Shaw, 1973). Similarly, half of pediatricians thought that palliative care, rather than surgery, was the better option (Todres et al, 1977). By the early 1980s, this practice had started to change. More parents opted for active intervention to save their babies (Shepperdson, 1983), and more physicians believed that withholding treatment was medical neglect (Todres et al, 1988). Baby Doe was born into this shifting cultural milieu. Baby Doe’s parents refused to consent to surgery and chose palliative care instead. The pediatrician alerted the state child protection agency, which investigated the case. At a court hearing, the parents claimed that they were following the advice of their obstetrician and not their pediatrician. The court found that parents only had to follow the advice of a licensed physician and that, since they were doing so, they were not neglectful. The court did not take protective custody. The doctor and hospital appealed. The Indiana Supreme Court refused to hear the appeal, and the baby died after 8 days (Lantos, 1987). These facts became publicly known and led to a national controversy that eventually reached the Oval Office. President Reagan demanded federal action. It was difficult to decide which action should be taken, because the federal government has no jurisdiction over child abuse and neglect; it is the domain of the states. The federal government oversees civil rights enforcement, however, and the Reagan administration devised a legal strategy that defined not treating babies with Down syndrome or other congenital anomalies as discrimination against people with disabilities, rather than
medical neglect. This strategy gave the federal government a justification for oversight. The new regulations were implemented, and bright red signs containing federal hotline telephone numbers were posted in NICUs across the country. These signs proclaimed that withholding treatment on the basis of disability was a federal civil rights violation (Annas, 1984) and that federal investigative squads could review medical records to determine whether discrimination had taken place. These regulations were challenged and eventually struck down by the U.S. Supreme Court (Annas, 1984). A diluted version of the original Baby Doe guidelines was eventually incorporated into the federal Child Abuse and Treatment Act (Annas, 1986; Kopelman, 1988). That law, however, is primarily a funding mechanism to channel federal funds to state child protection agencies; it is not a regulation that can be enforced for physicians or hospitals. The Baby Doe regulations do not exist today and have not existed since 1984; however, they still hold symbolic power. The mention of Baby Doe strikes fear in the hearts of pediatricians who lived through the events, in part because pediatricians had made pediatricians into villains in the societal battle over child protection. There is a certain irony in the controversy over Baby Doe regulations. The original goal—to decrease the range of cases in which withholding treatment of newborns is permissible—did not need federal input. Progress was already being made. Many diseases that used to be considered incompatible with life or that were seen as leading to an unacceptable quality of life were being treated routinely. Metabolic therapies for genetic diseases (phenylketonuria [PKU], hypothyroidism, Gaucher disease) and surgical therapies for organ dysgenesis (congenital heart corrections, congenital diaphragmatic hernia (CDH) repair, extra corporeal membrane oxygenation (ECMO), dialysis) have been forcefully advanced by medical subspecialists in their corresponding fields (e.g., genetics, cardiac surgery, general surgery). With few exceptions (hypoplastic left ventricle being one of the best recognized), these advances have been relatively uncontroversial. Once sufficient data are gathered to demonstrate moderate efficacy, the innovations are widely and rapidly adopted. There is still controversy when treatments enable survival but have a high likelihood, or certainty, that survival will be accompanied by severe neurologic impairment. As a result, two questions must be asked. First, how severe will the neurologic impairment be? Second, what is the likelihood that the child will have the most severe possible impairment? The prognostic spectrum of Down syndrome is broad, but few infants with trisomy 21 are severely impaired. In contrast, almost all infants with trisomy 13 or 18 either die in infancy or are left with profound neurologic impairment. The outcomes for these chromosomal anomalies can be used to define the spectrum within which clinical decisions are made. For babies whose outcomes are likely to be similar to those seen in trisomy 21, it is no longer permissible to withhold life-sustaining treatment. For babies whose outcomes are likely to be similar to babies with trisomy 13, it is permissible to withhold or withdraw life-sustaining treatment and offer palliative care instead. The calculus becomes more complex in conditions associated with a wider range of outcomes, such as extreme prematurity or high myelomenigocele with hydrocephalus.
CHAPTER 3 Ethics, Data, and Policy in Newborn Intensive Care
The shift in moral standards regarding babies with Down syndrome was not related to technology, but rather sociology. The capacity to repair Arnold–Chiari malformation and duodenal atresia existed long before it was applied to children with myelomeningocele and Down syndrome. What has changed the mood of the country is a growing recognition that disability is as much a social construct as a medical construct, although it is always both and not one or the other.
BORN ALIVE INFANT PROTECTION ACT In 2002, the U.S. Congress passed a law called the Born Alive Infant Protection Act (BAIPA). BAIPA, like the discredited Baby Doe regulations, was an attempt to insert federal values into medical deliberations. There are some interesting similarities and distinctions between Baby Doe and BAIPA. The Baby Doe regulations addressed disability, whereas BAIPA applies to abortion. BAIPA declares that “for the purposes of Federal law, the words ‘person’, ‘human being’, ‘child’, and ‘individual’ shall include every infant who is born alive at any stage of development.” As Sayeed (2005) noted, “The agency arguably substitutes a nonprofessional’s presumed sagacious assessment of survivability for reasonable medical judgment.” It is unclear what the implications of this law have been. On the one hand, all infants born alive before BAIPA were also treated as human beings; however, that did not necessarily mean that they received all available life support and resuscitation. After all, patients can have do-not-resuscitate orders or receive palliative care rather than intensive care. Still, the purpose of BAIPA seemed to be less about the treatment of babies and more about restrictions on abortion. Although some authors have expressed concern that the influence of BAIPA may transform neonatal care of infants born at, or even below, the threshold of viability, it appears to have had little measurable effect to date. Partridge et al (2009) surveyed neonatologists in California and found that they were concerned about the implications of BAIPA. They write, “If this legislation were enforced, respondents predicted more aggressive resuscitation potentially increasing risks of disability or delayed death.” There have been no cases to date in which BAIPA has been invoked or in which physicians and hospitals have been found to be in violation of its requirements.
FUTURE DIRECTIONS There is no new technology in development that appears likely to affect outcomes significantly. Consequently, for infants who receive resuscitation in the delivery room, birthweight-specific mortality and morbidity are unlikely to change much in the near future. Nonetheless, three developments may change the way we think about newborns, and consequently shift the terrain of neonatal bioethics.
HIGH RISK MATERNAL-FETAL MEDICINE CENTERS Many children’s hospitals are now developing high-risk, maternal-fetal medicine centers. The goal of these centers is to identify fetuses at risk—particularly those with
21
congenital anomalies—and to care for those fetuses and their mothers in centers where there is expertise in fetal diagnosis, therapy, and neonatal care. The hope is that such centers will allow more timely, and therefore more effective, intervention for babies with congenital heart disease, congenital diaphragmatic hernia, or other anomalies. The medical effectiveness of fetal centers will depend on two distinct developments. First, on a population basis, these centers will only be as effective as fetal screening and diagnosis. The existence of these centers will almost certainly create an expectation and a demand for better fetal screening. Such screening is likely to include both better imaging and better screening tests that can be performed on maternal blood; both will lead to earlier diagnosis of fetal anomalies. These diagnoses will create more complex dilemmas for perinatologists and parents who will need to decide, in any particular case, whether to terminate the pregnancy, offer fetal therapy, or offer either palliative care or interventions after birth. Ironically, better fetal diagnosis may increase the likelihood of pregnancy termination, even when postnatal treatment is possible, such as in hypoplastic left heart syndrome. Second, the effectiveness of fetal centers will depend on the effectiveness of fetal interventions. Perhaps surprisingly, other than in utero transfusion for Rhesus disease or vascular ablation for twin-twin transfusion syndromes— neither of which are particularly new and neither of which is performed by pediatric surgeons or pediatricians—there is little evidence that any fetal intervention has had any effect on any neonatal outcome. This lack of demonstrated effectiveness has, thus far, not suppressed the proliferation of fetal intervention centers. There may be other factors, including institutional prestige, finances, and recruitment of “desirable” patients.
EXPANDED NEWBORN SCREENING In recent years, the number of diseases and conditions that can be diagnosed through newborn screening has expanded dramatically. Such screening is under the purview of states, rather than the federal government, and there is wide variation in the number of tests that are performed. In 1995 the average number of tests per state was five (range: zero to eight disorders). Between 1995 and 2005 most states added tests, so that the average number of screening tests done by 2005 was 24 (Tarini et al, 2006). The expansion of newborn screening raises three problems. First, even the most accurate test has false positives. For rare conditions, the percentage of positive tests that are false positives is increased. Thus, the more rare conditions that are added to a newborn screening panel, the more false positives there will be. False positives are associated with considerable parental anxiety and can lead to potentially dangerous and unnecessary diagnostic procedures or treatments. Second, expanded newborn screening costs money. Interestingly, the tests themselves are astoundingly inexpensive, which is why policy makers are tempted to add more to the panels. However, the follow-up counseling and testing after positive tests are expensive, and without such followup the screening programs will not work. The Centers for Disease Control and Prevention has recently expressed concern about these costs (Centers for Disease Control
22
PART I Overview
and Prevention, 2008). Finally, there is the potential for discrimination against patients for whom documented heterozygous carrier status conveys no recognized medical infirmity, but social or psychological stigma may be real. There is little funding available to assist or counsel these patients.
FINANCIAL CONSTRAINTS The American Academy of Pediatrics guidelines on neonatal resuscitation suggest that resuscitation should be obligatory at 25 weeks’ gestation or greater, optional at 24 weeks’ gestation, and unusual at 23 weeks’ gestation. These recommendations are thought to reflect the best understanding of both the ethical discussion and the epidemiologic facts surrounding NICU outcomes. The recommendations purport to reflect the traditional paradigm that data drive policy. However, particularly in the context of NICU survival for infants at 23 to 25 weeks’ gestation, there is good evidence that causation can work in the reverse direction—that is, policy drives data. In the Netherlands, Canada, and some parts of Oregon, survival at 25 weeks’ gestation is comparable and comparably good; more than 50% of infants born at 25 weeks’ gestation will survive to discharge. However, in the Netherlands, virtually no infant survives birth at 24 weeks’ gestation, whereas in Canada and Oregon the survival rate is 40%. In addition, in Oregon and some parts of Canada, survival at 23 weeks’ gestation is close to zero, whereas in other parts of Canada and the United States the survival rate is 20% or greater. Why? Certainly not because the Dutch, Canadians, or Oregonians have forgotten how to resuscitate small infants. Rather, they have chosen not to. Why not? Perhaps it is non-maleficence–the fear that survival with permanent neurologic morbidity may be cruel to the child, the family, or society at large. However, the incidence of neurologic morbidity in survivors is not different when comparing infants at 23, 24, 25, and 26 to 27 weeks’ gestation. If the fear of a permanent crippling neurologic injury is the driving force, we should not be resuscitating 26 or 27 weekers, since many more of them will survive, and survive with disability. But that seems odd. The approach in the Netherlands is consistent; there is a limited budget and a communitarian ethic. There is a certain rationale behind spending money on all pregnant women, instead of 1% of micro-premies. The United States appears ambivalent–we value individuals over community, are fascinated with high-technology, and claim to prize our children. On the other hand, we will not spend money to prevent unwanted teen pregnancy or to provide visiting nurses for new mothers. Finally the concept of generational conflict must be considered. We appear quite comfortable calling
delivery-room resuscitation of 24 weekers “optional,” based on gestational age alone. It is difficult to imagine the AMA recommending that resuscitation for 85-year-olds who come to the emergency department is “optional,” based on age alone. NICU care is often criticized as “too expensive.” We ask, “Compared to what?” Fewer than 5% of infants will be in the NICU for more than a short stay. The vast majority of these patients will survive. Only 24,000 infants of 4 million births die each year in the United States, and half of these will die within fewer than 7 days. In contrast, nearly 3 million adults will die in the United States each year and one third of these will have been admitted to an medical intensive care unit (MICU) in the 6 months before dying. MICU costs outpace NICU costs by at least 100 to 1.
SUMMARY Ethical philosophy is a place to start, not a place to finish. Data are relatively easy to acquire and agree on. Policy is intriguingly insensitive to data, but that may reflect social and political realities that exist beyond the NICU— perceptions of disability, abortion politics, individual versus communitarian emphasis, fascination with technology, discrimination, publicity, financial constraints—so that an ethical course of action in one country, one city, or one family might seem perverse elsewhere.
SUGGESTED READINGS Annas GJ: The Baby Doe regulations: governmental intervention in neonatal rescue medicine, Am J Public Health 74:618-620, 1984. Born-Alive Infants Protection Act of 2001: Report together with additional and dissenting views of the House Committee on the Judiciary, 107th Congress, 1st Session, August 2, 2001. 1-38, 3. (Purpose and Summary). Centers for Disease Control and Prevention: Impact of expanded newborn screening: United States, 2006, MMWR Morb Mortal Wkly Rep 57:1012-1015, 2008. Janvier A, Leblanc I, Barrington KJ: The best-interest standard is not applied for neonatal resuscitation decisions, Pediatrics 121:963-969, 2008. Johnson S, Fawke J, Hennessy E, et al: Neurodevelopmental disability through 11 years of age in children born before 26 weeks of gestation, Pediatrics 124:E249-E257. Lantos J: Baby Doe five years later: implications for child health, N Engl J Med 317:444-447, 1987. Meadow W, Lagatta J, Andrews B, et al: Just in time: ethical implications of serial predictions of death and morbidity for ventilated premature infants, Pediatrics 121:732-740, 2008. Miller v HCA, Inc, 118 S.W. 3d 758, 771 (Texas 2003). Montalvo v Borkovec, 647 N.W. 2d 413 (Wis. App. 2002). Shepperdson B: Abortion and euthanasia of Down’s syndrome children: the parents’ view, J Med Ethics 9:152-157, 1983. Todres ID, Guillemin J, Grodin MA, et al: Life-saving therapy for newborns: a questionnaire survey in the state of Massachusetts, Pediatrics 81:643-649, 1988. Tyson JE, Parikh NA, Langer J, et al: National Institute of Child Health and Human Development Neonatal Research Network. Intensive care for extreme prematurity: moving beyond gestational age, N Engl J Med 358:1672-1681, 2008. Complete references used in this text can be found online at www.expertconsult.com