Infants’ Best Interests in End-of-life Care for Newborns
BACKGROUND AND OBJECTIVES: Pediatric bioethics presumes that decisions should be taken in the child’s best interest. If it’s ambiguous whether a decision is in the child's interest, we defer to parents. Should parents be permitted to consider their own interests in making decisions for their child? In the Netherlands, where neonatal euthanasia is legal, such questions sometimes arise in deciding whether to hasten the death of a critically ill, suffering child. We describe the recommendations of a national Dutch committee. Our objectives were to analyze the role of competing child and family interests and to provide guidance on end-of-life decisions for doctors caring for severely ill newborns.
METHODS: We undertook literature review, 7 consensus meetings in a multidisciplinary expert commission, and invited comments on draft report by specialists’ associations.
RESULTS: Initial treatment is mandatory for most ill newborns, to clarify the prognosis. Continuation of treatment is conditional on further diagnostic and prognostic data. Muscle relaxants can sometimes be continued after withdrawal of artificial respiration without aiming to shorten the child's life. When gasping causes suffering, or protracted dying is unbearable for the parents, muscle relaxants may be used to end a newborn's life. Whenever muscle relaxants are used, cases should be reported to the national review committee.
CONCLUSIONS: New national recommendations in the Netherlands for end-of-life decisions in newborns suggest that treatment should generally be seen as conditional. If treatment fails, it should be abandoned. In those cases, palliative care should be directed at both infant and parental suffering. Sometimes, this may permit interventions that hasten death.
- DELN —
- the decision to deliberately end the life of newborns with lethal drugs
- EoL —
- KNMG —
- Royal Dutch Medical Association
- NVK —
- Dutch Pediatric Society
The care of severely ill newborns can be extremely burdensome for both parents and caregivers. When prospects are hopeless, end-of-life (EoL) decisions have to be taken with the utmost care. In this article, EoL decisions are medical decisions with the effect, or the probable effect, that death is hastened. They include the decision to withhold or withdraw life-sustaining treatment, the decision to administer medication with a potentially life-shortening effect to alleviate pain and suffering, and the decision to deliberately end the life of newborns with lethal drugs (DELN). The Dutch terminology for the latter decision differs from that in North America and many other countries: in the Netherlands, the term “euthanasia” is used for situations in which the patient can voluntarily request death. Because newborns and children cannot make this voluntary request, we use DELN for what others would call neonatal euthanasia.
Building on previous reports, a commission of the Royal Dutch Medical Association (KNMG) has issued new recommendations for EoL decisions in newborns, with the aim of reaching a consensus that may be considered normative within the Dutch medicolegal context.1 We defined newborns as children up to age 1 year. In practice, however, the report concerns mostly newborn infants up to age 28 days. This article describes the recommendations developed by the Commission and pays attention to some of the contentious elements. The authors were members of the Commission that wrote the recommendations.
In 1992, the Dutch Pediatric Society (NVK) proposed a distinction between newborns who have no chance of survival, and those who may survive, but with a very poor quality of life.2 Quality of life was defined as a life of severe suffering, dependent on medical technology, with extremely restricted possibilities for communication and self-subsistence. In both situations, the NVK considered withdrawing life support to be justified. Five years later, the Ministry of Health convened a multidisciplinary commission to discuss the reporting of cases of DELN. Among others, this commission introduced the idea that continuation of treatment is conditional upon improvement (“life on approval”).3 This would justify, or even demand, withdrawing life support when the prognosis is shown to be grim. In 2005, the NVK adopted the so-called Groningen Protocol concerning the ending of life in severely handicapped newborns, who survive in severe suffering without any prospect for improvement.4,5 If these infants do not die after the withdrawal of life support, the active ending of life may be justified under the same conditions as stated in the Dutch euthanasia law, obviously apart from the impossible consent of the patient. Soon after, in 2006, the government decided to establish a commission of experts to assess cases of DELN.
This commission, however, in 6 years received just 1 report of DELN, whereas previous studies had led to the expectation of ∼20 cases per year.6 This discrepancy can be partly explained by the implementation of a 20-week ultrasound, which would lead to abortion in severe congenital anomalies, but also it was suggested that pediatricians may have felt that the criteria of the assessment committee were too vague to give them the confidence to report cases of DELN.
This undesirable situation was an important reason the KNMG asked a commission to provide guidance on EoL decisions for doctors caring for severely ill newborns.
On the basis of a thorough study of the existing literature, the Commission, consisting of experts in pediatrics, nursing, ethics, and law, held 8 sessions, drafting a report. This draft was then commented on by associations of obstetricians, pediatricians, pediatric neurologists, and pediatric surgeons, by 3 associations of parents, and by the Inspectorate of Health. Both the NVK and the KNMG decreed a final version (after a further round of comments) in May 2013.
We will summarize the main findings and advice of the report, following the sequence of events as they usually occur in time after the birth of a child whose prognosis is deemed very poor. In addition, we will define the areas of agreement and difference with standards for EoL care in North America.
Decisions about withdrawing or withholding life-supporting treatments are usually made by a team of professionals and the parents of the severely ill newborn. Within the professional team, views may differ on the chances of survival and, more often, on the quality of life if the child survives.7,8 The decision that further treatment is becoming futile is therefore not a 1-step decision: it takes time and repeated team discussions to realize that this is the case. In almost all situations, professional teams will be able to reach a consensus despite possible differences. However, if a consensus is unattainable, that should not prevent a decision from being made; the treating physician remains primarily responsible, even within a team.
Whether a child has a chance of survival is essentially a medical judgment. The views of parents regarding (dis)continuation of treatment should be taken into account, but cannot be decisive. Obviously, parents should be allowed the time to adapt, as far as that is ever possible, to a judgment that further treatment will not be beneficial, but continuing a futile treatment will be harmful for the child whose well-being is the physicians' primary responsibility. It is important to realize how difficult it may be for parents to grasp what they are told about the hopeless situation of their child from a medical point of view. Their need for time can justify postponing the decision to withdraw or withhold treatment, but only if this does not increase or prolong the suffering of the child.
It is crucial to inform parents carefully about the possible events after withdrawing or withholding life-supporting treatment and about the palliative care that is available to assure as much comfort as possible for their child.9–11 This often means that the child may be touched by the parents, and that tubes and lines will be removed as much as possible.
If parents and professionals have different opinions about the best way to care for the child who eventually will die, clear and repeated exchanges of mutual views (not only information) usually leads to a consensus, however painful. According to the KNMG commission, it is crucial that professionals are aware of the parents’ views on death and dying, as these may have a profound impact on their views and emotions.
Usually one second opinion or more will have been obtained before it is accepted that a consensus cannot be reached. If so, parents cannot force professionals to perform treatments that are clearly medically futile. Conversely, parents sometimes withhold permission for treatments that physicians consider mandatory, such as morphine for pain, or sedation for intractable suffering. Physicians may then need to disregard the parents' refusal and start treatment they view as beneficial and necessary.
Withholding or Withdrawing Life-Supporting Treatment and Provision of Terminal Care
In severely ill newborns, treatment is always started in a situation of extreme uncertainty. The first treatment decision to be taken is whether to start life support immediately after birth. Even when the chance of survival appears poor from the outset, physicians have no choice but to start life support, to prevent deterioration and obtain more certainty about the diagnosis and the chance of survival of the newborn (benefit of the doubt). Continuation of life support, in those cases, is conditional on subsequent developments in diagnosis and prognosis.
If it becomes clear, therefore, that the child will not survive, or that survival will only be possible with a severely damaged quality of life, then not only the withdrawal, but also the continuation of life-supporting treatment demands justification. The chance of survival, but also the prospects of independence, communication, the burden of a mostly lifelong treatment, and the relations between these factors must be taken into account. If the continuation of life support is no longer in the child’s best interest, it has no further justification. Then life support should be withdrawn.
Two types of life-supporting treatment demand specific attention in this context: artificial nutrition and hydration, and respiration support.
The administration of artificial nutrition and hydration must be regarded as a medical form of life support. As a consequence, a judgment that further treatment has become futile also applies to nutrition and hydration and implies their discontinuation.
This is not without emotional difficulties for those involved. Sometimes the argument is made that giving artificial nutrition and hydration contributes to the comfort of the child, because it prevents feelings of hunger and thirst (this would render the withdrawal inhumane). This, however, is uncertain; continuation, on the other hand, may have complications and may be the cause of a protracted dying process. Withdrawal of artificial nutrition and hydration, according to this KNMG commission, will not be inhumane if state-of-the-art palliative care is provided.
Some severely ill newborns do not need to be fed and hydrated artificially, but may be breast- or bottle-fed. This is normal basic care that needs to be stopped only if it causes discomfort through choking or dyspnea.
After the withdrawal of treatment, including withdrawal of artificial nutrition and hydration, the child will certainly die. Clinical experience indicates that the duration of the dying process may be several hours and sometimes as long as several days; however, literature on the subject is very scarce.12–14 There is also not sufficient knowledge about possible suffering as a consequence of withdrawing nutrition and hydration. Because it cannot be excluded, physicians need to discuss comfort care, including sedation, with parents.
The parents must be thoroughly prepared for the withdrawal of respiratory support because they may interpret any change in their child as a sign of suffering. It may be necessary to increase the dosages of analgesics and sedatives to prevent postwithdrawal suffering.15–17 Parents need to be prepared and know that the color of their child may change and that breathing may become very irregular and may appear to be difficult (gasping). It is unclear whether gasping causes suffering in the newborn, but it cannot be excluded. Sometimes infants show signs of suffering, such as clutching their fists or grimacing, making appropriate palliative measures necessary. Cases have been reported, however, in which optimal palliative care did not provide sufficient relief.15,16,18–20
Two questions may then come up about muscle relaxants: can or should they be continued after withdrawal of respiratory support, and, secondly, can or should they be started in children who did not receive them?
The KNMG commission distinguishes 3 possible scenarios: firstly, continuation after withdrawal of respiratory support. This may sometimes be justifiable in a scenario that was best described by Truog et al in 2000.21 This involves patients who are receiving neuromuscular blocking agents as part of the medical treatment when the decision to withdraw life support is made, but neuromuscular function cannot be restored for several hours, days, or even weeks. In these circumstances, insistence on the restoration of neuromuscular function by withdrawing blocking agents before the withdrawal of ventilation will delay the termination of life support well beyond the point at which the patient and family have determined that the burdens of such treatments outweigh the probable benefits.
The second scenario would be to administer relaxants de novo, in addition to high doses of analgesics and sedatives, with the aim of stopping prolonged gasping and potential suffering in newborns who are weaned from the respirator. According to the Commission, this could be acceptable, but it would mean DELN and should, in the Netherlands, be reported to the commission of experts.
Thirdly, relaxants are sometimes given to end a protracted dying process, which only adds to the suffering of the parents who have been through a lot of pain already. Even though careful preparation of the parents is often sufficient for them to support this, in extreme cases the Commission considers this reason acceptable for DELN. Again, this should be reported.
Treatment of Symptoms at the EoL
Often, severely ill newborns need medication for pain and other symptoms. Even though a proportionate use of opioids and other drugs will almost never shorten the life of the patient, physicians should always take into account this possibility and prepare the parents for it. It is completely understandable that sometimes physicians and parents may even hope that this will happen. However, in the view of the KNMG commission, this should never lead to disproportionate dosages as a means of ending the suffering of the patient.
Palliative sedation, the intentional and proportional lowering of consciousness in a patient who has otherwise intractable symptoms, may be an appropriate way to relieve the suffering of a dying newborn. It is a solution of the last resort. The drug of choice is midazolam, sometimes combined with propofol or esketamin. Dosages for very young children are unknown, but should be increased until the symptoms subside.
The last and most contentious decision at the EoL in newborns is deliberately ending it.4 In the Netherlands, each case of DELN has to be reported for review to a national commission of experts that, in view of the general prosecuting office’s decision to prosecute or not, assesses retrospectively whether the case followed a set of conditions deduced from the euthanasia law. DELN, in our commission’s view, is acceptable in cases of severe suffering that cannot be relieved by excellent palliative care including sedation or, as mentioned above, in exceptional cases of a protracted dying process that becomes unbearable for the parents. Examples could be asphyxiated newborns who remain restless under sedation or newborns who have epidermiolysis bullosa.
There is a small but particularly difficult group of newborns who have serious illnesses who remain stable after life support has been withdrawn or withheld to let the child die. These children do not appear to be suffering and are not dying. DELN would not be justifiable under Dutch regulations. These children have to wait for a complication such as a pneumonia to be allowed to die; sometimes this may take a very long time. For the moment, we feel that these situations are too difficult to solve, and that they may become very rare if the recommendations issued by our Commission are acted on.
This report about EoL decision-making in newborns is largely compatible with currently accepted practices in North America and many other countries in the industrialized world. American ethics and law recognize the rights of all patients or surrogates to refuse any medical treatment, even if life sustaining, if the patient or surrogate believes that the burdens of treatment exceed the benefits. This is true for the 2 treatments that were highlighted in our summary (mechanical ventilation and medical nutrition/hydration), but is also true for a variety of other life-sustaining interventions, such as dialysis, transplantation, pacemakers, and others.11,22,23 In addition, US law and ethics also support the practice of palliative sedation for patients of all ages, meaning that patients whose pain and suffering cannot be controlled with standard palliative care treatments may be sedated to the point of unconsciousness.24,25 The ethical justification for that approach, commonly used by doctors in the United States, is the doctrine of double effect, which permits actions that may hasten death, so long as the intention of the caregiver is to relieve the patient’s pain and suffering, not to cause death. This principle also defines the general legal stance in North America.
Guidelines in the United States permit withdrawal of life support in paralyzed patients, similar to the Dutch report, in the situation in which a patient receives muscle relaxation for therapeutic reasons and waiting for this medication to wear off may take hours or days.25
The first area in which the guidelines from the report differ from North American practice is the authority of parents and surrogates in the Netherlands. The views of parents should be taken into account, but cannot be decisive. Withdrawing or withholding life-supporting treatment is considered a medical decision, fully with regard to survival and at least partly with regard to quality of life. Criteria to value the expected quality of life have been formulated by multidisciplinary working groups to help physicians and parents decide.2,3,26 Physicians’ stewardship of the scarce resources of NICU beds and technology does not play a role. In contrast, in North America, physicians’ roles in determining what constitutes an acceptable quality of life is very limited and physicians do not have the authority to make these decisions unilaterally. Court interventions are sometimes sought in these matters and parents are given wide latitude in deciding what quality of life is acceptable for their child.27
The second way that the Dutch developments differ from the North American EoL practices is that they permit the use of interventions even when the intention is clearly to cause the death of the patient, not just to relieve pain and suffering. The Dutch endorse the use of muscle relaxants de novo with the aim of stopping prolonged gasping during ventilator withdrawal and to treat parental suffering by putting an end to a prolonged dying process. This situation may occur occasionally, when even state-of-the-art palliative sedation is insufficient to relieve pain and suffering. Such situations may justify, if the parents request it, active ending of the life of the dying newborn by administering muscle relaxants in addition to the analgesics and sedatives that are already being administered. All cases must be reported for review, to maintain full transparency and accountability. The North American view is that muscle relaxants should never be administered de novo to a terminally ill patient, because they do not have any sedative or analgesic properties and it is impossible for the clinician to be intending to relieve pain and suffering if that is not a pharmacological property of the drug.25
The significance of this document is twofold. First, the document has considerable legal standing because NVK and the KNMG have endorsed it, which means that this document expresses the opinion of the medical profession. In the absence of formal legislation, judicial decisions will be primarily based on the medical profession’s opinion. Second, the retrospective legal assessment imposes the risk that doctors are being accused of murder or manslaughter if they provide DELN, which may drive them to under-report. This may be a less serious impediment to the reporting if the conditions that justify DELN are described in great detail and publicly acknowledged by the medical profession.
- Accepted July 24, 2014.
- Address correspondence to A.A. Eduard Verhagen, MD, PhD, JD, Department of Pediatrics, University Medical Center Groningen, University of Groningen, PO Box 30.001, 9700 RB Groningen, the Netherlands. E-mail:
Drs Willems and Verhagen conceptualized and designed the study, carried out the initial analysis, coordinated and supervised the data collection, drafted the initial manuscript, and critically reviewed the manuscript; Mr van Wijlick carried out the initial analyses, coordinated and supervised the data collection, drafted the study, and critically reviewed and revised the manuscript; and all authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.
FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.
FUNDING: No external funding.
POTENTIAL CONFLICT OF INTEREST: The authors have indicated they have no potential conflicts of interest to disclose.
- Royal Dutch Medical Association (KNMG)
- Nederlandse Vereniging voor Kindergeneeskunde
- ↵Overleggroep toetsing zorgvuldig medisch handelen rond het levenseinde bij pasgeborenen. Toetsing als spiegel van de medische praktijk [Assessment as mirror of medical practice]. Rijswijk, Netherlands: Ministerie van Volksgezondheid Welzijn en Sport; 1997
- Sheldon T
- Verhagen AA
- Batton DG,
- Barrington KJ,
- Wallman C,
- American Academy of Pediatrics Committee on Fetus and Newborn,
- American Academy of Pediatrics Section on Surgery,
- Canadian Paediatric Society Fetus and Newborn Committee
- De Lisle-Porter M, Podruchny AM. The dying neonate: family-centered end-of-life care. Neonatal Netw. 2009;28(2):75–83
- Bell EF,
- American Academy of Pediatrics Committee on Fetus and Newborn
- Hellmann J,
- Williams C,
- Ives-Baine L,
- Shah PS
- Diekema DS,
- Botkin JR,
- Committee on Bioethics
- Moratti S
- Perkin RM,
- Resnik DB
- Tripp J,
- McGregor D
- ↵Staatssecretaris van Volksgezondheid Welzijn en Sport en de minister van Justitie. Brief van July 29, 2005 aan de Tweede Kamer. Kamerstukken. Vergaderjaar 2005–2006. 30300XVI. Nr 90. Den Haag: Ministerie van Volksgezondheid, Welzijn en Sport en de minister van Justitie
- Truog RD,
- Campbell ML,
- Curtis JR,
- et al.,
- American Academy of Critical Care Medicine
- ↵KNMG Commissie Aanvaardbaarheid Levensbeeindigend handelen. Medisch handelen rond het levenseinde bij wilsonbekwame patiënten. Houten: Bohn Stafleu Van Loghem; 1997
- American Academy of Pediatrics Committee on Bioethics
- Copyright © 2014 by the American Academy of Pediatrics