How to withdraw ventilation compassionately

An expert offers advice on tackling the psychological and medical challenges of end-of-life extubation.


Compassionate extubation is a frequent occurrence in ICUs, but still not a simple one, Elaine Chen, MD, assistant professor of pulmonary and critical care and palliative medicine at Rush University Medical Center in Chicago, told attendees of CHEST 2020, held virtually in October.

There's relatively little recent data on the topic, but older studies showed that the percentage of ICU deaths that involved withdrawal or withholding of therapy increased from about half in a 1990 study to 90% in a 1997 report. The rate is thought to have held steady or increased since then, Dr. Chen added. “About 10% to 20% of all deaths nationwide now involve withdrawal or withholding [in the ICU],” she said.

Although these two concepts are often categorized together, extubation may feel more severe than choosing not to provide a medication or therapy. “Withholding and withdrawing—ethically, they're the same. Psychologically, it may feel very different. Some people may feel that withdrawing is very active, so they feel a significant moral responsibility when they're taking something away,” said Dr. Chen.

On the other hand, some family members and clinicians could find it easier to discontinue ventilation than to decline it, or any treatment, outright. “Having tried it first and then taking away something that hasn't worked may feel less psychologically challenging. So it really varies and depends on the case,” she said.

Photo courtesy of Dr Chen
Photo courtesy of Dr. Chen

Dr. Chen offered her advice on tackling psychological and medical challenges to make end-of-life extubation as painless as possible for patients, families, and clinicians.

Planning for extubation

Before an extubation can take place, there are several things clinicians and families should discuss.

“We give a range of prognosis, minutes to hours to days to weeks, and reassure the family that our job as the bedside clinicians is to ensure the patient's comfort, whether it is for a short period of time or a longer period of time,” Dr. Chen said.

To facilitate that reassurance, she recommends choosing one's language carefully. “We do not like the phrase ‘withdrawal of care.’ We are providing care up until the very end, even postmortem,” she said. “Use ‘withdrawal of life support’ or ‘compassionate extubation.’”

Clinicians should raise the question of autopsy and talk family members through the process before extubation. “Their job as a family is to focus on the patient as a person and talk to them, with the expectation that they hear and understand everything that they say,” said Dr. Chen. “Counsel them on what to expect, what they might see, what is normal . . . changes with heart rate, dyspnea, respiratory rate, skin changes, urine changes.”

To avoid too much unnecessary attention to these changes, she generally favors turning off monitors. “Family members can focus on looking at their loved one instead of staring at the monitor that has been their lifeline for so many days or weeks,” she said. “Some families may prefer the monitors to be left on. If that's the case, it's OK.”

Families should also be encouraged to plan any rituals or spiritual activities they might find helpful. Dr. Chen likes to have a chaplain present for an extubation, not just to support the family, but also the staff involved. She recommends that a nurse take charge of planning who is going to be present at the patient's bedside.

Of course, this has been complicated by COVID-19. “How many people can we allow into the unit at that time? . . . Obviously much more challenging with a COVID patient—in our institution, we do not allow families into the room,” said Dr. Chen.

Family members often ask how long the patient will live after extubation, a tough question to answer. “A lot of times I as the intensivist think that I have an idea of how long it might take, but I've been wrong so many times, so I emphasize the uncertainty here,” she said.

The speed with which ventilatory support is withdrawn is one way that the process has changed over the years. “There's the concept of a terminal wean, where there's a stepwise reduction of ventilatory support and then the ventilatory support is finally removed,” said Dr. Chen. “Historically, the pace can vary from several minutes to several days. And several days, today we would consider probably unacceptably long, a prolongation of the process.”

It's possible to err by going too fast though, too. A terminal extubation involves turning off the ventilator and immediately removing devices from the airway. “There is some concern that a terminal extubation may lead to unnecessary respiratory distress and dyspnea, so we really focus on this newer concept of compassionate extubation. We want to remove the ventilatory support but provide the most comfort, alleviate suffering as best that we can. It is a compassionate process,” she said.

The process

This process includes a number of steps for clinicians to take before actual extubation. “Make sure that do not resuscitate/do not intubate order has been written. Document the findings, the discussion, the goals,” said Dr. Chen.

Nutrition therapy should also be reviewed. “Make sure that they are NPO, so they don't have emesis at the time of extubation or aspirate the gastric contents,” she advised.

There's some debate about whether other interventions should be stopped in advance for these patients, Dr. Chen noted. “Our current recommendations are the abrupt discontinuation of life-sustaining treatments other than mechanical ventilation,” she said. “However, a stuttering withdrawal of varying types of life support can be associated with family satisfaction.”

Stuttering withdrawal, as described in the 2008 study in the American Journal of Respiratory and Critical Care Medicine Dr. Chen cited, entails sequentially removing life-sustaining treatments. “Renal replacement therapy, artificial hydration, artificial nutrition are removed earlier, and ventilation tends to be the last life support that is removed prior to death,” she explained.

One intervention that should definitely be stopped before extubation is paralytics, or neuromuscular blockers. “Patients look super comfortable when they're on a paralytic because their face is unable to show distress. So families have sometimes said to me, ‘Well, I don't want to take it off because it makes them look so comfortable.’ But that's the important thing: They look comfortable but you can't assess them.”

The drugs can also prevent patients from breathing on their own after extubation. Paralytics should be stopped early enough for their effects to wear off beforehand—four half-lives in the case of an infusion, Dr. Chen advised.

While most medications are being stopped, others, that alleviate symptoms, should be started or increased. “Opioids are of paramount importance here for managing dyspnea and discomfort or pain,” said Dr. Chen. Spontaneous breathing trials can be used to predict the appropriate dose of opioids after extubation, she noted.

Typically, the goal will be to titrate the opioids to maintain a respiratory rate of less than 20 or 30 breaths per minute. “Some patients may have a higher respiratory rate, just physiologically, and that's OK,” she said. Patients should have relaxed-looking facial muscles, and no accessory muscle use, flaring, grimacing, or agitation.

Opioid side effects generally aren't a major concern, even in patients who are opioid naive. “I'm not so much worried about their renal function or anything else in this phase because it's unlikely that the side effects . . . will happen in this short period.”

Dosing does depend on whether the patient is already on opioids. For those who aren't, Dr. Chen recommended 2 to 4 mg of morphine, 25 to 100 μg of fentanyl, or 0.5 to 1 mg of hydromorphone, all IV, with dosing every 15 minutes (or even every five minutes for fentanyl). “If a dose is ineffective, it is appropriate to escalate by 50% to 100%,” she said. Patients already on opioid infusion should get 50% to 100% of the current hourly dose, up to about 10% of the total daily dose.

Benzodiazepines may also be added for patients who aren't already on them. Dr. Chen recommends 1 to 2 mg of midazolam or lorazepam by IV every 15 minutes as needed for anxiety or agitation. “Some people will recommend a single dose about two minutes prior to the extubation just to ensure comfort at that time,” she noted.

Anticholinergics can also be used. “These may be a bit controversial, because it is not confirmed evidence base-wise whether they actually change symptoms; however, it is known that they do decrease secretions,” said Dr. Chen. Her preferred regimen is 200 μg of glycopyrrolate 15 minutes prior to extubation and then ongoing as needed.

The use of these medications to alleviate discomfort should not be confused with euthanasia. “Euthanasia is not legal,” she said. “Our goal is to ensure comfort while allowing the disease process to take its natural course. . . . The principle of double effect provides the ethical rationale for providing the relief of pain and other symptoms with sedating medications when this may have the foreseen but unintended consequence of hastening death. So the distinction here lies in your intent, the clinician's intent.”

Additionally, research has shown that sedatives and analgesics do not necessarily speed death in this situation and may actually slow it, “the theory being that they take away the work of breathing,” Dr. Chen said. “When you take away that work and that distress, the patient is much calmer and may survive a little longer.”

One more dilemma is whether to remove the endotracheal tube when the ventilator is turned off. “If there are upper airway issues, edema, or stridor, we can improve comfort by leaving the airway in,” she said. On the other hand, survivors of critical illness have reported distress from the endotracheal tube and suctioning. “At the end of the day, this is an individualized decision,” Dr. Chen said.

All these aspects of the process should be planned in advance of extubation, she noted. “I'd recommend that the multidisciplinary team meet outside and anticipate any other needs. Have medications available, if they're not on infusion already, [and discuss] any last-minute questions or concerns for counseling,” Dr. Chen said. If the family has agreed to turn monitors off, they can be set on comfort mode, so that clinicians will still be able to observe the patient's vitals from outside the room.

The family may be present in the room during extubation, or it might be better not to have them there. “Sometimes we ask them to step outside and come back, especially if we anticipate a lot of distress,” said Dr. Chen. “After everything is comfortable, we allow the family to have a few minutes, five to 10 minutes of quiet uninterrupted time, with their loved ones.”

Of course, it's hard to know how long their time together will be. “Some patients may survive for hours, two days, or even several days. At which time, it may be appropriate to transfer them out of the ICU,” said Dr. Chen.

After the patient's death, she recommends one additional, nonclinical step. “A sympathy card is an important thing that we can do to help the family have closure and appreciation, as well as help us have closure,” Dr. Chen said.