Ethicists are seeing increasing numbers of consults involving extracorporeal membrane oxygenation (ECMO), the most aggressive life-sustaining technology available.1 With ECMO, which is currently offered by about 250 U.S. hospitals, some patients are saved who would otherwise die.

“ECMO is great technology. We can buy time and keep them alive,” says Shunichi Nakagawa, MD, director of inpatient palliative care services at Columbia University Irving Medical Center. However, for some of those patients, Nakagawa says “the ethical dilemma is when to stop ECMO.”

The authors of one study analyzed ethical issues involving ECMO in a cardiothoracic surgery ICU at a large academic hospital between 2013 and 2015.1

“The biomedical ethics literature was primarily focused on ‘novel’ ethical dilemmas in these cases,” says Andrew Courtwright, MD, PhD, one of the study’s authors and an assistant professor of clinical medicine at University of Pennsylvania’s Perelman School of Medicine. For example, case studies reported on awake patients stranded on ECMO without a chance for recovery or transplantation, or appropriate code status for a patient on venoarterial ECMO without intrinsic cardiac rhythm.

“But it was unclear to what extent these types of cases were actually occurring,” Courtwright notes.

Researchers also wanted to know how decision-making authority was determined if there was a conflict about whether to withdraw ECMO. “We also wanted to highlight our experience in integrating routine ethics consultation into the care of ECMO patients,” Courtwright adds.

During the study, 113 patients were placed on ECMO. Of that group, 45 were seen by the ethics committee, but the percentage increased over time. In 2013, 21.7% of ECMO patients went through ethics consults; by 2015, almost all patients (93%) did. Not surprisingly, the most common ethical issue was disagreement on whether ECMO should be continued. There were conflicts between the healthcare team and surrogates, between surrogates, and within the healthcare team.

“We found that the vast majority of cases were similar to more ‘traditional’ ethics consult cases that involve disagreement over life-sustaining treatments such as tracheostomy, medical nutrition and hydration, and dialysis,” Courtwright reports.

Most conflicts were about whether sufficient time had passed to decide whether a reasonable trial of ECMO had been attempted. “Ethicists seemed to be most helpful in mediating consensus around time-limited trials of ongoing support, with clear criteria for what would constitute meaningful physiologic improvements,” Courtwright says.

In another study, researchers reviewed medical records of adults treated with ECMO at the Mayo Clinic from 2010 to 2014, specifically adults from whom ECMO was withdrawn.2

“We have been finding that an increasing fraction of ethics consults now involve questions about the withdrawal of advanced cardiac technologies,” says Daniel P. Sulmasy, MD, PhD, MACP, professor of biomedical ethics and acting director of the Kennedy Institute of Ethics at Georgetown University.

Investigators were surprised by how frequently ECMO is used as a “bridge to decision,” which often translates into a “bridge to nowhere.”

“The ethical consequence is that fully one-quarter of all ECMO treatments started were eventually withdrawn with the expectation of patient death,” Sulmasy says.

Of 235 ECMO-supported patients, 62 requested withdrawal. No patient had decisional capacity. For most patients (82%), ECMO had been initiated as a “bridge to decision” as opposed to a bridge to transplant or mechanical circulatory support. “It was also surprising that so many patients had DNR [do-not-resuscitate] orders, when ECMO amounts to a continuous form of resuscitation,” Sulmasy says.

Of the “bridge to decision” group of patients, 29% had a DNR order in place. “We are uncertain what DNR means in such circumstances,” Sulmasy adds.

Ideally, ethicists are part of the discussion about trying ECMO, and can discuss with the patient and family the possibility that it might not work. But the decision to withdraw ECMO should not rest solely on the shoulders of the family. “Above all, do not abandon patients or families to their own autonomy,” Sulmasy cautions. “Urge the clinicians to make recommendations to withdraw ECMO if that seems best.”

There are several scenarios that arise during ethics consults:

For some patients, ECMO becomes a “bridge to nowhere.” ECMO may be temporary for some. It either leads to recovery, transplant, or another device.

For others, none of that is possible. At that point, patients can be kept alive only on ECMO and only in the ICU. “They can’t get out of the ICU, out of the hospital, and cannot get home. For those patients, we are able to buy time, but only in the ICU,” says Nakagawa, who co-authored a paper on this topic.3

The ethical issue becomes: How long should the patient be kept in this condition when there is no exit? Columbia University Irving Medical Center’s clinical team calls on ethics to help resolve conflicts regarding when to stop ECMO. Some patients ask to stop treatment and die naturally, but the medical team thinks it is still too early. “Some patients are sick now, but can still get better,” Nakagawa observes.

Some patients still say they cannot continue. In one such case, the patient stated many times that he did not want to continue life on ECMO. Ethicists mediated several discussions among the patient, family, and medical team. The team discussed the possibility of the patient undergoing left ventricular assist device (LVAD) surgery, which would mean months of recovery.

“We emphasized that even in the best case scenario, medically, his quality of life would be significantly worse,” Nakagawa recalls.

The patient would end up on dialysis and a feeding tube. “In order to survive, that was the only choice. He did not want that surgery. He did not want to go through it,” Nakagawa says.

The man’s wife disagreed and insisted this was unlike the person she knew. “We told her that while that might be true, he does have decision-making capacity and is clearly expressing he is suffering, and we need to respect his wish,” Nakagawa says.

The patient continued to request withdrawing ECMO. Several days later, ethicists discussed withdrawal at a family meeting. Ultimately, ECMO was withdrawn, and the patient died in a few hours. “Thorough and extensive discussions among the patient, family, and medical team, with the help of the ethics team, helped the family fully understand his suffering,” Nakagawa adds.

Clinicians struggle to decide whether a patient is a candidate for ECMO. “A multitude of factors are at play,” says Robert D. Truog, MD, director of the Center for Bioethics and professor of medical ethics, anaesthesia, and pediatrics at Harvard Medical School.

These include: Is the disease reversible? Even if it is, what are the chances of recovery? What are the patient’s baseline comorbidities, and how should they be weighed in the decision? Does the patient have the physical and cognitive resources that prolonged recovery from ECMO will require?

At Boston Children’s Hospital, a small group of surgeons and intensivists have agreed to be on a WhatsApp group chat. Whenever there is a question about whether a patient is an ECMO candidate, the relevant clinical details are sent to the group.

“Whoever is available to respond weighs in with their thoughts, questions, and opinions,” Truog says. “The response rate tends to be high, even on nights and weekends.”

In this manner, the group seeks to come to a consensus on whether ECMO should be offered. It often stimulates discussion about factors that might otherwise have been overlooked. “While the system is not perfect, it does remove the decisional burden from just one person,” Truog says.

If patients cannot participate in decision-making, the family has to decide to stop ECMO. For some families, this is not a difficult decision. Members say they know the patient would not want to live this way, that it does not make sense to prolong the suffering, and ask that the clinical team let the patient go.

“Some families have difficulty accepting that reality. They cannot give up, and are hoping for a miracle,” Nakagawa notes.

In certain instances, this is the case even when the patient obviously is dying. The patient starts to develop pressure sores and needs to be turned regularly by nurses. “Sometimes, the digits and toes become ischemic and necrotic. It becomes really miserable for the patient and for the people who provide care. That creates an ethical dilemma,” Nakagawa says.

If there is a conflict between the patient, family, and medical team, clinicians first try to resolve it on their own or with a palliative care team. “But for very difficult cases, we ask for the ethics team’s help,” Nakagawa explains.

Conflicts arise in part because ECMO changes the perception of death. “Before that, if the heart stopped people died. But now we are able to keep them alive,” Nakagawa says.

This makes it harder for the patient and family to accept death. “It is very ironic. The more medical technology advances and we can make people live longer, the more difficult the end of life is going to be,” Nakagawa adds.


  1. Courtwright AM, Robinson EM, Feins K, et al. Ethics committee consultation and extracorporeal membrane oxygenation. Ann Am Thorac Soc 2016;13:1553–1558.
  2. DeMartino ES, Braus NA, Sulmasy DP, et al. Decisions to withdraw extracorporeal membrane oxygenation support: Patient characteristics and ethical considerations. Mayo Clin Proc 2019;94:620-627.
  3. Mulaikal TA, Nakagawa S, Prager KM. Extracorporeal membrane oxygenation bridge to no recovery. Circulation 2019;139:428-430.