Recent Cases Add to Complexity of Treatment Withdrawal Decision-Making

By Stacey Kusterbeck

Many ethics consults involve conflicts over withdrawal of life-sustaining treatment. If patients’ families are aware of recent cases demonstrating recovery potential in patients with traumatic brain injury who were thought to have a low chance of survival, it can make the decision-making process even more challenging.

Clinicians may ask families to decide whether to withdraw life support within 72 hours of a brain injury. A recent study suggests that, at least for some patients, delaying the decision may be beneficial.¹ Researchers calculated the likelihood of withdrawal of life-sustaining treatment for traumatic brain injury patients in intensive care units (ICUs) at 18 trauma centers. The researchers paired 80 individuals for whom treatment was not withdrawn to 90 patients with similar model scores who did have life support withdrawn. For the patients who did not have life support withdrawal, 55% died; but of survivors, more than 30% recovered at least some independence in daily activities. Researchers found that six months after injury, remaining in a vegetative state was an unlikely outcome. The findings suggest that clinicians assume that patients will do poorly based on outcomes data and recommend withdrawal of life support. In turn, that increases poor outcome rates, resulting in more decisions to withdraw life support. Although death or severe disability is a common outcome for patients where withdrawal of life-sustaining treatment is being considered, the study findings support a more cautious approach, according to the authors.

“For years, there have been high-profile media coverage of cases where patients recovered from all sorts of diseases even though those patients were not expected to recover,” observes Thaddeus Mason Pope, JD, PhD, HEC-C, professor of law at Mitchell Hamline School of Law in Saint Paul, MN.

A significant percentage of patients subject to non-beneficial treatment determinations do survive to discharge. “But the core underlying reality is the same: prognostication is inexact and imprecise,” says Pope.

This means that surrogates must make decisions about withdrawing treatment under conditions of uncertainty. “Even patients determined and declared dead were later found not to be dead. In short, the limits of prognostication periodically manifest in colorful, concrete, and conspicuous ways,” says Pope.

Although these cases are statistical outliers, they get lots of attention. “We know from the mountains of recent work in behavioral economics that salience drives decisions more than data and evidence,” notes Pope.

Families may be less likely to accept a dim prognosis if they know about a case where a patient who survived the very same condition. Families sometimes use a highly publicized case to show that patients can beat the odds and recover from brain injuries judged to be irreversible. For instance, some family members bring up the case of Jahi McMath, which gained international attention, involving an alleged false-positive diagnosis of brain death.²

“First, ethicists should ensure that families have an accurate understanding of the case. Second, they should, if possible, distinguish the case if it is not clinically analogous,” says Pope. With brain death, for example, families might be thinking of media-reported cases of unresponsive wakefulness syndrome (formerly referred to as persistent vegetative state) where the patient recovered to a minimal conscious state, or to even more alertness. Although patients with unresponsive wakefulness syndrome are sometimes referred to as “brain dead,” that is materially different from death by neurological criteria.

Good communication on these medically complex points can help to resolve conflicts over treatment withdrawal. “Fortunately, this typically does not make the disputes refractory and intractable to resolution. But they may take more time and more family meetings,” says Pope.

Surrogates often get conflicting information from different specialists and from different rotating staff, however. For example, one clinician might state something that suggests recovery is possible because they are narrowly focused on only one organ. Another, making a more global assessment, might state that recovery is highly unlikely. “Ethicists can often identify such communication gaps and errors,” says Pope.

Sometimes the true problem is that families mistrust the clinical team’s recommendation to withhold or withdraw treatment because they are hearing different things from different people. “Surrogates repose more trust in the diagnosis and prognosis, and in the withholding or withdrawal recommendation, when they are confirmed by multiple clinicians,” says Pope.

REFERENCES

1. Sanders WR, Barber JK, Temkin NR, et al. Recovery potential in patients who died after withdrawal of life-sustaining treatment: A TRACK-TBI propensity score analysis. J Neurotrauma 2024; May 13. doi: 10.1089/neu.2024.0014. [Online ahead of print].
2. Shewmon DA, Salamon N. The extraordinary case of Jahi McMath. Perspect Biol Med 2021;64:457-478.