The use of partial code status in pediatric medicine presents clinicians with unique ethical challenges. The clinical vignette describes the presentation of a pulseless infant with a limited life expectancy. The infant’s parents instruct the emergency medicine providers to resuscitate but not to intubate. In an emergency, without a clear understanding of parents’ goals, complying with their request risks an ineffective resuscitation. The first commentary focuses on parental grief and how, in certain circumstances, a partial code best serves their needs. Its authors argue that providers are sometimes obligated to endure moral distress. The second commentary focuses on the healthcare team’s moral distress and highlights the implications of a relational ethics framework for the case. The commentators emphasize the importance of honest communication and pain management. The final commentary explores the systems-level and how the design of hospital code status orders may contribute to requests for partial codes. They argue systems should discourage partial codes and prohibit resuscitation without intubation.
In most resuscitation situations, children are assumed to be “full code.” Code status is not routinely discussed upon admission unless the child has a medical condition that is potentially life-threatening or limiting.1 Emergency department and critical care providers are experts at delivering high-quality resuscitation tailored to the needs of the individual patient. They understand that resuscitation is less likely to be successful unless all components of cardiopulmonary resuscitation (CPR) are provided – oxygen delivery with intubation (when a skilled airway provider is available), chest compressions, and delivery of electricity and/or cardiac medications. In the setting of cardiac arrest, (1) cardiac medications without chest compressions and (2) chest compressions and/or medications without continuous oxygen delivery will generally, but not always, prove ineffective.2 Although bag-mask ventilation (BMV) or noninvasive ventilation strategies may prove sufficient for administration of continuous oxygenation and support of ventilation in some cases, advanced airways are often considered the gold standard for cardiac arrests. Advanced airways, including intubation, provide improved ventilation, reduce the risk of aspiration, and allow for uninterrupted chest compressions.3 Consequently, “partial code” situations may appear illogical and inappropriate to medical providers and may result in questioning caregiver decisions surrounding code limitations.
The Case
Emma, a 6-month-old infant with type 2 Gaucher disease presents to an emergency department in full arrest. The team ascertains from the paramedics that she suddenly stopped breathing at home and turned blue. Type 2 Gaucher disease is the least common and most severe form of Gaucher disease. It involves progressive neurodegeneration, typically leading to death in infancy or early childhood. Most patients experience respiratory failure as a result of seizure, apnea, laryngospasm, or pneumonia.4 Emma’s parents, Susan and James, called 911. When the paramedics arrived, they initiated chest compressions and BMV. They continued CPR en route without return of spontaneous circulation (ROSC). Susan and James tell the medical team that they know Emma will likely die of respiratory failure and that mechanical ventilation will not change the long-term outcome. They ask that she not be intubated. What should the medical team do? Should they perform a limited resuscitation?
Emily Berkman, an Intensivist and Ethicist, and Doug Diekema, an Emergency Medicine Physician and Ethicist, Comment
In situations like this case, where a child with a life-limiting condition has experienced cardiac arrest, allowing natural death would be a reasonable choice. Therefore, if allowing the child to die is ethically permissible, then allowing the family to have some control over what is and is not provided in a limited resuscitation, even if not physiologically likely to accomplish the goal of resuscitation, is also ethically permissible. The absolute ethical limit would be a request to perform procedures that would cause harm or suffering to the child while offering no benefit.5 Those should be refused. However, partial codes are not always ineffective, and gold standard resuscitation may not always be required for ROSC.6 As pediatric providers, we have a primary duty to seek our patient’s benefit – the child. However, because children live within the context of a family unit, the practice of pediatrics encompasses care for the whole family. As a result, we also have a secondary duty to care for families. Respecting and honoring partial code status, assuming no additional harm to the child, can allow us to simultaneously discharge these duties that may appear to be in conflict at first glance.
A limited code status may signal that parents and caregivers recognize that their child may die, yet may not be prepared for that to happen today. A request for limits that fall short of a do not attempt resuscitation (DNAR) order may reflect a desire to make some effort while limiting interventions that seem too drastic or too invasive. This parental and caregiver assessment of risks and benefits is a personal one that occurs in the worst of circumstances. Families attempt to make decisions that are best for their child, but also need to make decisions they can live with.7 As survivors of the trauma of their child’s death, they risk being haunted by regret.8 As providers, especially those working in the emergency department, we often do not know the ongoing narrative being written by a particular family of a child with serious illness. In this case, the medical team may not have known the parents’ reasons for forgoing intubation. Perhaps there was a traumatic intubation history or, alternatively, the family may wish to avoid the difficulty they would have making future decisions to remove the breathing tube to allow her death.
Proceeding with resuscitation without intubation may allow the parents to witness the medical team’s tremendous efforts and provide time for parents to make a mutual decision to allow natural death. Although it may not change the ultimate outcome for this child, it may change her parents’ experience of her death – hopefully for the better.
We would be remiss if we failed to acknowledge the feelings of members of the resuscitation team. In these cases, “partial code” status can lead to significant moral distress among team members who think their efforts were futile given the constraints placed by the child’s parents or that they would have made a different decision had the child been their own.9 Although these feelings are valid and require emotional support, the personal views of healthcare providers should not override parental decision-making unless those decisions place the child at significant risk of serious harm.10 Moreover, in emergency departments and critical care settings, there is often no time to process these differences in opinion in real-time. Support must be offered to all involved team members following challenging codes, regardless of outcome.11 As medical providers, we will go on to help the next child in need, and the memory of this baby will likely fade. Healthcare providers who are parents will go home to their own children. However, these parents will go home without their baby. Sometimes, our duty to ease the journey of those who lose a child in our care requires that we suffer some moral distress.
Andrew Silverman, a Pediatric Neurology Resident, and Heather Fitzgerald and Beth Epstein, Nurse Ethicists, Comment
Whether the parents’ do not intubate (DNI) preference reflects a partial understanding of their child’s likely need for intubation in the event of arrest, insufficient planning with primary providers, or an evolving situation through which the parents grieve and gradually comprehend what a resuscitation entails, the responding team lacks in depth knowledge of the parents’ reasoning. While recognizing that resuscitation decisions are deeply personal and that a partial code may help the family navigate this terrible loss or mitigate the enduring trauma of returning home from the hospital without their child, the team remains constrained from providing care consistent with the child’s presentation.12 As Rousseau notes, “…if the idea is to save a life and restore the person to the living, why would we not perform a full code, even if brief and time-limited?”13
Faced with requests to limit interventions perceived as insufficiently beneficial or sufficiently harmful, providers are at risk for moral distress. Moral distress occurs when clinicians are constrained from taking ethical actions or are forced to take ethically inappropriate actions based on professional obligations, resulting in a sense of complicity in wrongdoing.14–16 Unresolved moral distress becomes moral residue if providers feel persistently misaligned with core professional values and standards, which, over time, contributes to depression, anxiety, and burnout.9,17–19 In a recent international study of almost 6000 emergency providers, most (66.7%) were often unsure of the appropriateness of an attempted resuscitation.20 Further, 58% reported moral distress when the burdens of resuscitation were felt to outweigh the benefits.
The morally distressing aspect of the partial code in this case may not, in fact, be about the possible ineffectiveness of it (especially in light of a desire to respect the parents’ conceivable need for a bit more time to emotionally navigate the death of their child). Instead, the suffering and harm that is inflicted on a dying child may constitute the greater, morally distressing trigger. It may be impossible for emergency providers to avoid performing a partial code, at least initially, but they could act to minimize their moral distress and mitigate harm to the infant, parents, and family unit using a relational ethics approach.
Relational ethics recognizes the connectedness and interdependence of children and their parents.21 Acknowledging the embeddedness of the critical parent-child relationship brings forward the notion that the correct ethical action may be the one that preserves that relationship – not necessarily ensuring continued survival of the baby but ensuring that the parents, regardless of the outcome, are able to see themselves as “good” parents who advocated for their baby. Even in the high-speed context of the emergency department, this framework activates moral agency, may abbreviate participation in a partial code, and thus mitigate associated moral distress for the resuscitation team. For example, the attending can be honest with the parents early on, even as the baby is taken into the trauma bay, and explain that resuscitation with bag-mask ventilation could achieve ROSC but that someone would have to provide that ventilation indefinitely, standing at the bedside around the clock.22 Janvier suggests too that providing pain medication and other measures to address suffering can be put into place.23 These early conversations with the parents, together with attentive management of the infant’s potential pain and suffering, address the constraints keeping the team from meeting their obligations to the infant, while concomitantly respecting the suffering of the parents. Additional conversations with parents can ascertain their readiness to let go, their desire to hold their infant if he or she dies, and their interest in calling family members to the bedside to say goodbye.
Partial codes can harm both patients and clinicians but may be unavoidable in emergency settings. However, a relational ethics approach privileges the connection between infant and parents and allows the team to focus on how to preserve that relationship. Supporting parents by providing additional time, while being honest about what’s really happening, and treating the baby’s pain and suffering preserves professional integrity. If their baby dies, knowing she was not in pain at the end might help the parents in the long run. A relational ethics framework can help establish a patient-centered, family- and team-oriented model that optimizes compassionate care for all those involved in emergency department resuscitations.
Jason Batten, an Anesthesiology Resident, and Emily Shearer, an Emergency Medicine Resident, Comment
In the case, it is not immediately clear how Susan and James arrived at their preference for a partial code and what they hope to achieve by it. However, providers must realize that decisions such as these are not made in a vacuum. Rather, they are formed within the complex milieu of innumerable interactions between the parents and the healthcare system. This includes conversations with healthcare providers throughout the entire trajectory of their child’s illness and exposure to other forms of information (pamphlets regarding code status, Physician Orders for Life-Sustaining Treatment forms, etc). Taken together, we can refer to this as the choice environment created by the healthcare system.
The primary way the choice environments affect parental decision-making is through framing, in which the manner of choice presentation impacts what decision is made. Most people are familiar with framing in everyday contexts, such as salesmanship, in which the seller frames available choices to increase sales. Even when not intentional, framing can impact decisions about critical matters, such as resuscitation and life-sustaining treatments.24,25 Framing can occur via top-down mechanisms (eg, on a policy level, a hospital leader designs code status options that include partial code as an option) or via bottom-up mechanisms (eg, in an individual conversation, a provider describes a partial code as a reasonable option).26
Indeed, some hospitals provide a partial code order as part of their code status orders, whereas others do not.27,28 Enshrining partial codes in hospital policy legitimates them as a viable treatment choice, prompting providers to consider and potentially offer them.27 Thus, the design of code status orders can impact whether a partial code is framed as a reasonable option and whether it is offered at all.
It is also possible that the parents were exposed to the idea of a partial code as an unintentional byproduct of a prior conversation.29 Unfortunately, code status discussions are sometimes reduced to a menu, in which parents are sequentially asked whether they would “want” each intervention, devoid of meaningful context. Suppose DNAR and DNI orders were discussed in this manner: in that case, the parents may not have recognized that a DNI order – often meant to apply to prearrest respiratory failure – does not necessarily have clear implications for treatment during a cardiac arrest.30–32 Hospitals that provide DNAR and DNI orders as checkboxes on a form may be prone to propagating these sorts of misunderstandings.
Considering these issues, how should hospitals approach the design of code status orders? One question raised by the case is whether hospitals should include partial codes as part of their routinely available code status orders.6,33,34 Several essential considerations argue against their inclusion. Available data suggest that outcomes of partial codes are worse than full codes.35–37 Since some providers discuss code status options as a menu, the inclusion of a partial code order may make it more likely that partial codes are offered even when they do not make clinical sense for a particular patient. Thus, the risks of providing a partial code order on a policy level may outweigh the benefits.
Additionally, given that parental code status preferences may be driven by misunderstandings, how can hospitals promote an accurate understanding of the relationship between DNAR and DNI orders? Some experts recommend increasing clarity by providing 3 code status orders: (1) full code for cardiac arrest, (2) DNAR but ok to intubate, and (3) DNAR or DNI.38,39 Under this approach, the preference for resuscitation during cardiac arrest (Full Code versus DNAR) is separated from the selection for intubation for prearrest respiratory failure (ok to intubate versus DNI). Of note, being Full Code but DNI is not an option in this formulation. In addition to shaping the choice environment as described above, this systems-level intervention also gives providers a policy-level justification for not offering an ongoing partial code to parents.
Because of the complexity of clinical medicine, we acknowledge that there may be cases where a partial code is the most sensible option (for example, cardiac medications and/or cardioversion for ventricular tachycardia).6 Thus, in addition to the 3 code status orders described above, hospitals can provide a designation for exceptional cases (eg, an “Other” option). This would allow providers the leeway to offer a partial code when it is both sensible and desired but would otherwise exclude partial codes from the list of routinely provided options.
Outcome of the Case
The providers comply with the parents’ request not to intubate. After approximately 30 additional minutes of CPR, the team achieves ROSC. Emma’s breathing, however, remains dependent on BMV provided by the respiratory therapist. The attending physician presents her parents with the options of intubating her or stopping the BMV and letting her pass if her heart stops again. Susan and James decide it is time to let Emma go. She does not breathe on her own, and her parents hold her one last time.
Armand H. Matheny Antommaria, Section Editor, Comments
In an emergency, it may not be possible to ascertain immediately, in the absence of an out-of-hospital DNAR order, whether a request for a partial code is adequately informed. Presumed consent may, therefore, provide an adequate ethical justification to provide full resuscitation while seeking to clarify the parents’ preferences and understanding. It may be reasonable for parents to desire CPR for readily reversible causes of cardiac arrest but not desire prolonged mechanical ventilation. Providers should agree to such requests if they are adequately informed. An explanation of the parents’ reasons to the resuscitation team members will hopefully clarify the logic and appropriateness of the request and alleviate the providers’ moral distress. Providers should, however, decline requests for partial codes that lack a coherent physiologic rationale (eg, medications without compressions).
Dr Silverman conceptualized, critically reviewed, and revised the manuscript; Drs Silverman, Batten, Berkman, Fitzgerald, Epstein, Shearer, and Diekema provided ethical commentaries on the presented clinical vignette; Dr Burgart supervised and critically reviewed the manuscript for important intellectual content; and all authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.
FUNDING: No external funding.
CONFLICT OF INTEREST DISCLOSURES: The authors have indicated they have no conflicts of interest relevant to this article to disclose.
Comments