Initiating Ventilator Support

Deciding to initiate ventilator support and other life-sustaining therapies in the neonate presents unique challenges unparalleled elsewhere in medicine. Often, the discussion begins and prognostication occurs before the health care provider can even lay eyes on the patient. If the mother presents in advanced labor, there may not be the luxury of time before a decision must be made. Additionally, the emotions, pain, and exhaustion of childbirth can cloud logic and reasoning for all parties involved. Sometimes the decision is clear cut and easy; a full-term infant with meconium aspiration will clearly benefit from assisted ventilation, should he or she develop respiratory failure. Other situations quickly challenge the principles of beneficence and nonmaleficence. Are we doing more harm or good by intubating an extremely preterm infant born at 23 weeks? Will we cause undue suffering by initiating assisted ventilation in an infant with trisomy 13 and multiple congenital anomalies? Would comfort care be more appropriate in these situations? As John Paris and colleagues state, “society’s role is to assure that the patient is not undertreated by the omission of beneficial therapies or overtreated with unwanted or unwarranted interventions.”

Whenever possible, discussions should occur with the family prior to the time of delivery. This is more easily accomplished in the case of prenatally diagnosed congenital anomalies and genetic disorders than with an unanticipated preterm delivery. It may be beneficial for all the parties who will be involved in the care of the child to meet with the family and on multiple occasions, when possible. For example, in the case of a complex congenital heart lesion, meeting with cardiology, cardiovascular surgery, neonatology, genetics, feeding specialists, etc., may provide the parents with a more complete picture and help them to make an educated and informed choice about their wishes for their child. In all prenatal counseling the caveat must be included that all information will not be available until after the infant is born and even the best laid out plans may change.

For the extremely preterm infant, right or wrong, decisions and discussions about initiating ventilator support usually begin with gestational age. But it should be noted that even under the best circumstances, dating via ultrasound is accurate only at ±7 to 10 days. Nonetheless, above 25 weeks’ gestation, most clinicians will provide resuscitation and life-sustaining therapies. Below 22 weeks, few would intervene aggressively and comfort care would be provided. The “gray zone” of 22 to 24 weeks is where ethical dilemmas typically occur. Other factors, such as antenatal corticosteroid exposure, sex, weight, and multiples, have also been shown to significantly influence outcomes for extremely preterm infants in addition to gestational age. The National Institute of Child Health and Human Development has developed a website with an online calculator to estimate mortality/morbidity risks for extremely preterm infants and assist providers in counseling parents. While this tool provides a useful starting point, it is only a report of population statistics derived from major academic centers and does not estimate the outcome for an individual infant. Outcome data also vary widely from country to country and from hospital to hospital and may be affected by publication time lag.

If the decision is made to initiate ventilator support, it is important to recognize that the discussion does not stop there. Communication between the medical team and the family must be ongoing. The infant’s condition and best interests must be constantly reassessed. If assisted ventilation is deemed to be no longer in the infant’s best interest, withdrawal of ventilator support is an ethically justifiable option.

Withdrawal (Nonescalation) of Ventilator Support

Withdrawal of care, specifically removal of assisted ventilation, has become the most frequent mode of death for neonates in the neonatal intensive care unit (NICU). But that was not always the case. In a landmark article in the New England Journal of Medicine , Duff and Campbell first described the rare practice of withdrawal of care in 1973. Since that time the practice has become a common occurrence that health care providers face in the NICU. At the time, the courts had not addressed the issue of whether it was legally permissible to withhold or withdraw life-sustaining treatment. Nevertheless, Duff and Campbell recommended that such decisions were permissible and that they should be made privately by parents and doctors together. That recommendation has been adopted by most bioethicists and professional societies, within limits. Two publications, the Nuffield Council on Bioethics’ extensive report in Britain on “Critical Care Decisions in Fetal and Neonatal Medicine” and the 2007 policy statement of the American Academy of Pediatrics (AAP) on “Noninitiation or withdrawal of intensive care for high-risk newborns,” take the position that there are times when withdrawal of care is appropriate. Generally, two driving themes guide decisions regarding withdrawal of care. One is the anticipated quality of life of potential survivors. The other is the perception that treatment is medically futile. The same principles apply to nonescalation of ventilator support as to withdrawal. Nonescalation is generally applied to situations in which parents do not want to withdraw life support but they recognize that further escalation of treatment may simply prolong the dying process. In such situations, doctors may continue ventilation but not monitor blood gases, adjust ventilator settings, or change to more aggressive approaches such as high-frequency or jet ventilation.

Quality of Life

There are two issues in any discussion about quality of life (QoL) in neonates. The first is about prognostication. Generally, we make decisions based not upon the infants’ current QoL but based on our best prediction of what their QoL is likely to be in the future. Such predictions have well-described margins of error. For example, in a study, many infants who had an Apgar score of 0 at 10 min not only survived, but when they were of school age, they had no disability.

The second issue related to QoL focuses on the ways in which we define and assess an individual’s QoL. Again, the issues for young children are quite different from those for adults. For adults, we can simply ask them. For young children, others must make a determination based on objectively measurable or observable factors. Payot and Barrington reviewed the sorts of scales that are available to measure QoL. They write, “The scales that are available to measure quality of life make comparisons between different health states, evaluate the effectiveness of medical interventions, and describe the life trajectories of individuals or groups.” They go on to note that any assessment of QoL necessarily requires many different types of data, including objective facts, social relationships, family structure, physical health, mental abilities, and many other factors. In some cases, a divergence between objective measures of disability or impairment and a subjective measure of self-assessed well-being has, in adults, led to what has been described as “the disability paradox.” By this paradox, a life with severe functional limitations may be considered of high quality by the individual him- or herself but of very limited quality to external observers. Of this phenomenon, Payot and Barrington write,

The fact that even impaired survivors can view their lives in a positive light has been dismissed as “making lemonade” by some commenters. We feel, in contrast, that it is a cause for celebration that individuals with impairments (lemons) can nevertheless find value and meaning in their lives (and make lemonade). Such resilience should not be dismissed, but should not at the same time impede our efforts to investigate methods to reduce such impairments …. We can illustrate the gulf which sometimes may exist between subjective and objective evaluations of QoL with studies regarding the QoL of infants with meningomyelocele, who have multiple functional problems in the long term, but have a QoL which they themselves find to be good or very good, including those who are wheelchair bound and incontinent.

Medical Futility

The concept of futile care is hard to define and is controversial. Efforts to prolong life, once considered an outcome of healing, may now be viewed by some as harmful acts of prolonging suffering. The early literature on medical futility challenged the “life at all cost” stance. In the now classic 1974 JAMA article by Richard McCormick, a renowned Jesuit moral theologian, entitled “To Save or Let Die: The Dilemma of Modern Medicine,” McCormick argued there was no moral obligation to force treatment on a patient who was dying or who was totally dependent on intrusive measures to sustain life, and there was no obligation to do so for a patient who had no possibility for meaningful relationships. That stance has continued to prevail and has influenced nearly every group that has attempted to set standards and guidelines directed at the care of the dying neonate.

An increasingly difficult conundrum in medicine has arisen owing to a shift to a consumer-driven model: what does a physician do when a request for treatment is believed to be futile, ineffective, or inappropriate. This issue first appeared in the literature when Paris and colleagues wrote of a case in which the physicians at Boston Children’s Hospital who had cared for a profoundly compromised baby for some 23 months refused a mother’s request to put her child once again on a ventilator. This case was one of the first cases in the literature to highlight a growing issue in which physicians were being requested to provide medical therapies believed to be futile. Prior to this the demand for certain treatments had been viewed as simply requests, such as antibiotics for viral infections or a computed tomography scan for routine headaches. Most physicians agree that such treatment ought not to be given, but with increasing practice constraints and parental demands, many found it easier to just go along with the requests that were perceived as harmless. However, that approach is flawed. The role of the physician was changed from one with knowledge and expertise to one that responded to whatever the patient preferred. Over time, the requests escalated into demands for more and more exotic and inappropriate treatments. With the shift to a consumer-driven model for health care came the belief that informed consent and patient autonomy implied not only that patients had the right to accept or reject certain therapies, but also that patients had the right to choose whatever treatment they desire. In the minds of some bioethicists, such as Veatch and Spicer, patient autonomy, or the right of self-determination, dictates that families have a right to demand whatever life-prolonging therapies they want, and that physicians are obligated to provide it, even if the request “deviates intolerably” from established standards or is, from the physician’s perspective, “grossly inappropriate.” Veatch and Spicer’s position would require the physician to unrelentingly inflict aggressive procedures and treatments on infants if requested to do so by the parents despite the overwhelming evidence that the interventions cannot relieve or change the child’s devastating condition. Robert Truog, the director of clinical ethics at Harvard Medical School, also supports this position. In an essay, he states that doctors should honor the family’s choices on end-of-life care “even when we believe their decisions are wrong.” However, these approaches create great moral distress, when families seek treatments that are not in the best interest of the child or are seen as prolonging suffering.

Since 2005, physicians have lashed out against such stances and the notion that patients have the right to demand inappropriate care. Even for simple requests, the medical community has pushed for and educated against antibiotic use for viral infections, with success. This has led some physicians to ask if they are, in fact, obligated to do what they believe to be futile or ineffective. The Nuffield Council on Bioethics and the AAP make it very clear that newborns are to be treated as any other patient—on the basis of their best interest. This position implies that although parents may and should continue to be involved in decision making for their children, they do not have the exclusive right to refuse—or to demand—medical treatment for the child. To avoid clashes between parents and physicians on treatment choices, it is advisable, whenever possible, that there be a joint discussion that involves the parents and the medical care team with regard to survival rates, outcomes, and severity of potential disabilities. It is important that the family receive consistent information from all providers.

The “futility” debate has mainly centered around end-of-life treatments, specifically cardiopulmonary resuscitation (CPR) and the need for “do not resuscitate” orders. It is now well understood that for certain patient populations there is near 100% mortality. Blackhall states that, even if the family requests CPR, the physician should decline to provide it in these situations. Futility is generally applied to an acute setting, when the medical condition has deteriorated to a point at which further increase in aggressive medical treatment, such as CPR, would not change the impending death. In cases in which death is imminent, there has been an increased emphasis on providing a “peaceful death.” A study by Wall and Partridge reveals that in the face of imminent death, most neonatologists recommend the withdrawal or withholding of treatment. A study found that, even though withdrawal of care at the end of life had remained constant over 10 years, withholding treatment had increased while the use of CPR in an unstable, dying neonate had decreased. Most of the change seen in management of death seemed likely to be due to a change in physicians’ practice. In 2000 and 2003, statements from the AAP and the Institute of Medicine called for improvement in care for the dying neonate, creating greater awareness among caregivers regarding the end-of-life process. More recently, the AAP Committee on Fetus and Newborn again noted there is no ethical difference between withdrawal and withholding of an intervention. When death is imminent or medical care is considered futile, the goal is to provide a peaceful, controlled setting for the infant and family.

Are These Debates Resolvable?

The medical literature regarding “medical futility” and “quality of life” continues to grow. But disagreements persist about what these terms mean or what implications they convey. For medical futility, Younger has asked: Does it signify absolute impossibility? Is it purely physiologic? Does it include the ability to achieve only a heartbeat but not to achieve discharge from the hospital? For QoL, Younger and others colleagues questioned whether it is ever possible to assess another person’s QoL. All assessments seem to incorporate problematic considerations about social value and cultural values. Lantos et al. noted that even among physicians there is no consensus on these terms. Physicians disagree on the chances of success, the goals of therapies, and the prediction of long-term outcomes. Some declare a treatment futile only if the chance of success is 0%, whereas others view a treatment as futile if the success rate is <10%, and even others view it so if the success rate is as high as 18%. Others consider a treatment futile if all it can hope to provide is a couple of days in an intensive care unit. On the other hand, “[s]uch a goal can be of supreme value to a dying patient or the patient’s family.”

There are similar disagreements about QoL. Many doctors, for example, consider the QoL that is associated with trisomy 18 to be so terrible that death is preferable. Many parents disagree. The lack of agreement on the meaning of medical futility and QoL, along with the varied use of both terms, makes it increasingly difficult to develop guidelines to assist the clinician in deciding to remove assisted ventilation in a particular neonate. Hackler and Hiller, in their article about a devastated child who was forced to undergo repeated resuscitation attempts because of parental requests pleaded for a better approach and compassion for treating the dying child. It is not the aggressive intervention at all costs approach but the patient’s response to treatment and continued best interests of the newborn that should direct our care.

Chronic Ventilation

As technology and survival for some of the most fragile neonatal patients improve, so do the ability and need for chronic ventilation. The prevalence of home mechanical ventilation for children is 5-6.3/100,000. The 2005 pediatric census in Massachusetts showed a threefold increase in the number of children needing chronic respiratory ventilation compared to the previous 15-year interval census. The number needing long-term ventilation can only be expected to rise in the future as technologies improve and are applied to a broader patient population. In the neonate, it is no longer only infants with complications related to prematurity, such as severe bronchopulmonary dysplasia (BPD), needing tracheostomy and prolonged mechanical ventilation, but those with a host of neonatal conditions that can lead to the final common pathway of ventilator dependence. Children with congenital anomalies, neuromuscular diseases, congenital or perinatal acquired neurologic disorders, and craniofacial anomalies now compose the majority of those with tracheostomy and long-term ventilation needs. Chronic lung disease due to prematurity represents only 7% to 29% of children needing long-term ventilation in recent population studies. However, a study showed that, in ∼58% of the infants who needed tracheostomy for prolonged mechanical ventilation in the NICU, the underlying disease was BPD. Long-term ventilation consumes a large allocation of resources both for hospitals and for communities. These fragile infants require prolonged hospitalization, home equipment, home nursing, and frequent readmissions, and up to one-third of patients live in acute or long-term care facilities.

Infants needing long-term ventilation are at a high risk for morbidities and mortality. There is uncertainty in the prognosis, length of therapy, and outcomes of these infants. The 5-year survival rate for infants with tracheostomy and chronic ventilation is approximately 80%. Reports show that half of the deaths are unexpected, acute events, such as dislodged or plugged tracheostomy tube, and were not related to an underlying disease process. The 5-year decannulation rate reported is ∼25%. Children with chronic lung disease had the highest rate of successful decannulation. Of the children able to be decannulated, the mean time needing ventilation was 24 to 39 months. However, reports suggest that if a child is not off of positive-pressure ventilation by 5 years and decannulated by 6 years of age, there is little chance of tolerating removal of the ventilator after that age (<10%). These children also face developmental impairments, with 14% having extremely delayed neurocognition.

Uncertain prognosis, unknown length of treatment, and high risk of death for infants needing long-term ventilation can pose many ethical dilemmas for both providers and families. Performing a tracheostomy and providing long-term ventilation from infancy, with the potential for lifelong ventilator dependency, can cause great moral distress. Other ethical dilemmas surround which patient population to offer chronic ventilation and should family dynamics factor in the decision. It may seem more acceptable to offer long-term ventilation to patients who have a chance of decannulation versus patients with life-limiting conditions, such as trisomy 13. However, it is hard to state that diagnosis alone should drive the decision on which patients receive chronic ventilation. Family wishes should also be factored into the decision, but how does that work when there is a difficult family structure with limited resources? Does that mean the child is placed in medical foster care when families are unable to provide care for their medically complex child or a child is placed in a long-term care facility? These are difficult decisions that families and medical care providers face.

Appropriate Care after Withdrawal of Life Support

One of the other moral issues facing physicians involved in the withdrawal of medical treatments is to assure the patient and family that the withdrawal will not produce suffering for the patient and ensuring the medical team will not abandon them. As Civet notes, pain during the end of life is of particular concern. With regard to the withdrawal of mechanical ventilation, care should be directed toward avoiding air hunger or dyspnea. This may present as gasping or struggle in the patient. To prevent these events physicians should be willing to provide analgesic and/or antianxiety medication. The goal of these medications is symptom relief, not death, and the physician should provide the appropriate dose to relieve the symptoms but not to cause apnea.

Families have described the feeling of abandonment by the health care team once the decision was made to withdraw care. The death of a child is the most difficult thing a family may experience; the health care team can have an impact on the lasting memories of their child’s final moments, both positively and negatively. It should always be the goal to provide the support the family needs during this difficult time. Chaplains, palliative care teams, and bereavement programs can help provide comfort to the families. In making a decision, what is called for is not rigid rules but a realistic assessment of the infant’s physical prospects based on current data from the medical literature, as well as outcome data. Acknowledging different cultural norms and family values is also important. As in all medical decision making, the primary consideration in treatment decisions for newborns is and continues to be a commitment to act in the best interests of the patient.

Finally, doctors should always be aware of the possibility that a child will not die after “life support” has been withdrawn. This was famously true of Karen Ann Quinlan, a case that led to a landmark legal decision. In that case, the doctors testified in court that the withdrawal of mechanical ventilation would lead to Ms. Quinlan’s death. After the ventilator was withdrawn, she survived for 7 years. This has happened in pediatric cases as well and can lead to moral distress, distrust, and communication problems.