ARTICLE IN BRIEF
In a new paper, investigators identified four neurological variables linked to death within 60 minutes of withdrawal of life support that will help establish neurologic criteria for organ donation in neurocritical patients.
Removing life support from a patient with a devastating brain injury creates an opportunity to harvest healthy organs, but only if the patient dies within an hour or so. If the patient's heart continues to beat longer than that, poor oxygenation of tissues makes the organs unsuitable for donation, and the harvesting effort must be abandoned.
Identifying the most promising donors would facilitate and possibly increase organ donation, and researchers at the Mayo Clinic in Rochester, MN, have developed criteria they believe will help.
The study, published in the April 27 Neurology, is based on 149 severely brain-injured patients who were taken off life support from 2002 to 2008. The authors identify four variables linked to death within 60 minutes of withdrawal of life support – no corneal reflex, no cough reflex, the presence of an extensor response or the absence of motor response, and an oxygen index (OI) greater than 4.2. (The OI combines airway pressure and blood oxygen levels into a single calculation. A higher score indicates declining lung function.)
“The association with earlier death increased with the presence of each additional variable,” the authors wrote. “The presence of these abnormal neurological signs suggests potentially severe irreversible brainstem dysfunction regardless of etiology.”
Patients declared brain dead after a devastating injury or illness have long served as organ donors, but patients who don't quite meet the criteria for brain death may also qualify for donation after cardiac death (DCD) if they cannot survive the withdrawal of a ventilator and other life-sustaining measures.
Currently, determining if the patient will die within 60 minutes involves temporarily withdrawing life support and applying an algorithm developed at the University of Wisconsin, known as the Wisconsin Donation After Cardiac Death Evaluation Tool. The algorithm produces a predictive score based on a patient's age, body mass index, oxygen saturation in the blood, method of intubation (endotracheal vs. tracheostomy), level of spontaneous respiration, and the need for vasopressors. The United Network for Organ Sharing (UNOS) also developed a measure based on respiration, ventricular assistance and other criteria.
But these predictive tools incorporate almost no neurologic data, and focus primarily on the patient's respiratory and hemodynamic status after being temporarily disconnected from mechanical ventilatory support, according to Alan H. Yee, DO, a neurologist at the Mayo Clinic in Rochester, MN, and lead author of the Neurology paper.
“We found specific neurologic signs that may be helpful in guiding the neurologist who sees these patients,” Dr. Yee said. “Our study will help the clinician determine which patients with irreversible neurologic injury will expire within one hour of removing life support.”
Dr. Yee and his colleagues hope to validate their findings, and then develop a predictive algorithm that will help physicians predict with greater accuracy how long it will take a patient to expire after being removed from life support.
“If we can predict when a patient with a primary brain injury but healthy organs will expire after withdrawal of life support, then maybe that will make physicians more comfortable about contacting their organ procurement agencies, and that could increase organ donation,” said Dr. Yee. “The need for organs is very high.”
More than a decade ago, Stephan A. Mayer, MD, conducted a study published in Neurology in which he reported that about 25 percent of brain damaged patients died from cardiac arrest within 60 minutes after extubation, but nearly a third lived for 24 hours — far too long for organ donation under a DCD protocol.
Determining which patients would die quickly defied reliable prediction at the time, so he was impressed by the findings of Dr. Yee and colleagues.
“If you can identify someone who is likely to undergo cardiac arrest within a hour after the withdrawal of life support, that's a good potential donor, and that's exactly what this study addresses,” said Dr. Mayer, professor of neurology and neurological surgery at Columbia University College of Physicians and Surgeons, and director of the Neurological Intensive Care Unit at New York Presbyterian Hospital.
“If you think you have a potential DCD organ donor you have to get a surgeon, and OR nurses, and a transplant team, and perfusionists ready in advance,” said Dr. Mayer, current president of the Neurocritical Care Society. “It's like mobilizing a small army, and false alarms are a big waste of time.”
But predicting exactly when a patient will die can be difficult, according to Rob Kochik, executive director of the Finger Lakes Donor Recovery Network in Rochester, NY. “From 2007–2009 we coordinated 37 donation after cardiac death donors, and 13 of these patients did not die within the time frame required for organ donation,” said Kochik. “It would be very helpful for organ procurement organizations to have a tool like this to help evaluate patients as potential DCD candidates. This would be helpful not only to us, but also to the families of these patients.”
However, Dr. Mayer also worries that DCD protocols improving the ability to predict death might also arouse ethical concerns in families. “The potential misunderstanding in DCD organ donation involves the worry that society doesn't value me [the patient] as a human being,” Dr. Mayer said. “People may think, ‘Maybe you're doing this so you can take my loved one's organs to support the life of someone you value more.’ On the other hand, people on waiting lists are dying every day before they get an organ. We're trying to make some good out of tragic circumstances in which we're helpless to save someone with a massive brain injury. Maybe by doing this we can get some good out of it.”
In addition to facilitating organ donation, the ability to predict death after extubation would help physicians respond to families who want to know how long their family member will live after withdrawal of life support, according to Claude Hemphill III, MD, Kenneth Rainin Chair in Neurocritical Care and associate professor of clinical neurology and neurological surgery at the University of California-San Francisco.
“A lot of times we say to families, we don't know,” Dr. Hemphill said. “It could be a few minutes, a few hours, a few days. It's difficult to predict in individual patients. This particular article is valuable because it brings quantitative information to the process of withdrawal of life support. It helps us with a qualitative decision we've been making for a long time. It moves the field forward.”
While he believes that the article may result in more organ donation, Dr. Hemphill does not consider it the “missing link” in facilitating that process. “The missing link is getting consent from more families,” he said. “There are many more patients who would meet criteria for brain death and possibly for DCD who do not become organ donors because the family does not consent. That's the biggest problem.”
Still, being able to provide families with better predictions about when a patient will die after withdrawal of life support will certainly help to obtain consent, according to Bradford B. Worrall, MD, associate professor of neurology and public health sciences, and associate medical director of the Stroke Unit at the University of Virginia Health System.
“It is often very challenging in the critical care setting to help prepare a family for how quickly someone may die,” he said. “And it's challenging in both directions – predicting which patients are going to die the moment the ventilator is disconnected, and those who, despite a devastating neurological condition, may take days to die. This paper is a great start, and I think this information, once validated, could be helpful in discussions with families who have made this decision.”