Article In Brief
Findings from a survey of physicians from three academic tertiary centers reflect variability in how brain death is determined, suggesting the need for formalized training around a standard protocol for those who perform these assessments as part of their routine clinical practice. Of particular concern to experts in neurocritical care is the fact that 10 percent of respondents did not perform an apnea test.
There appears to be substantial variability in the way physicians approach brain death assessments and a need for more formalized training, according to the results of a survey published in the January 25 online issue of Neurology. The survey results extend previous research that has shown substantial variability in brain death policies from hospital to hospital.
Most striking was a finding indicating that a little more than 10 percent of survey respondents did not complete an apnea test. Moreover, nearly one in four of the respondents indicated they did not receive formal training in brain death assessments. Indeed, survey results regarding use of ancillary tests in brain death determinations suggest that some physicians are unfamiliar with their own hospital protocols. A position statement on brain death determinations approved this year by the AAN and published in the January 2 online edition of Neurology also called for uniform training and standardized protocols.
“Some physicians who indicate they perform brain death examinations in practice may have an incomplete understanding of the clinical examination, as well as indications for confirmatory testing,” lead author Sherri Braksick, MD, assistant professor of neurology at the University of Kansas Medical Center, told Neurology Today.
“Our results, in conjunction with the prior paper that demonstrated substantial variability in brain death policies from hospital to hospital (Greer, et al, 2016) and the new AAN position statement (Russell, et al, 2019) suggest that uniformity throughout all aspects of the brain death declaration process — state statutes, hospital policies, and the clinical examination — should be instituted.”
Survey Methodology, Findings
Dr. Braksick and colleagues invited physicians at three academic tertiary centers by email to participate in the survey. Invitations with a link to the survey were sent to all individual department or division chairs and administrative assistants for distribution to all attending physicians, fellows, residents, and interns.
Participants were asked if they performed brain death examinations on adults in clinical practice and what neurological examination components they assessed as part of the brain death examination. The survey also inquired about the use of ancillary tests and the rationale for when a test was ordered.
Participants were excluded if they reported that they did not perform adult brain death examinations as a part of their clinical practice, regardless of their reported specialty.
A total of 68 respondents from 23 departments reported completing brain death examinations in clinical practice and were included in the final data analysis. Attending physicians comprised the majority of the respondents (61.8 percent), followed by residents (29.4 percent), fellows (7.4 percent), and interns (1.5 percent). Most of the respondents (77 percent) were neurologists, neurosurgeons, or neurocritical care specialists.
One third of physicians surveyed about their practice of brain death determinations reported obtaining ancillary testing automatically, regardless of clinical examination findings.
“This behavior may result in situations where a clinical examination is consistent with brain death, but the ancillary test is inconclusive or falsely positive,” the authors of the study wrote.
Moreover, 10 of the 20 physicians who indicated they ordered ancillary testing as part of a standard evaluation said they did so because it was required by their hospital policy.
Dr. Braksick and colleagues reviewed policies at the three surveyed sites and found in fact that none require automatic ancillary testing as part of the brain death assessment in an adult, indicating that they were unfamiliar with local policies by some of the participating physicians.
Ten respondents reported that doing the ancillary testing made them more comfortable with the diagnosis; three indicated it was done to protect them from liability; and three indicated other reasons for doing so, including evidence of persistent metabolic abnormalities, family requests, and “other individual situations.”
Additionally, 13 physicians (28.3 percent) who perform ancillary testing on an as-needed basis indicated they order the test if a patient breathes during the apnea test. “By definition, a spontaneously breathing patient is not brain dead, and confirmatory tests should not be considered in this situation,” the authors of the paper wrote. “The overuse of ancillary testing may predispose clinicians to underestimate the importance of the clinical examination and rely excessively on imperfect radiographic or electrophysiologic tests.”
Survey participants were asked to indicate which components of a standard brain death examination, as outlined in the AAN guideline, they regularly perform. Slightly more than a quarter (26.5 percent) of physicians reported checking peripheral reflexes (patellar, etc.), and 23.5 percent reported evaluation of the plantar response/Babinski sign during a brain death examination, neither of which is indicated in AAN guidelines.
“Multiple physicians also evaluated peripheral or spinal reflexes, which do not assess brain function, and likely only serve to confuse the examiner and family members who may witness an elicited movement due to a spinally-mediated reflex in an otherwise dead patient,” Dr. Braksick and colleagues wrote. “Of note, the presence of triple flexion to a painful stimulus, a spinally-mediated reflex, may be present in a brain dead patient, however, any evidence of coordinated and typical decorticate or decerebrate (flexor or extensor) posturing implies brainstem function and is not consistent with brain death.”
Experts interviewed by Neurology Today agreed that the survey, although small, highlights the need for standardization of training and protocols.
“The findings are not surprising, but that doesn't mean they aren't important,” said David Y. Hwang, MD, FAAN, assistant professor in the division of neurocritical care and emergency neurology at Yale University. “There have been a lot of data showing that protocols differ even from hospital to hospital, so it follows logically that there would be a lot of variability among physicians.”
“Even though it is limited in scope in terms of response and the number of centers surveyed, the study highlights the wide variability on the ground in brain death assessment,” Dr. Hwang added.
Others registered greater alarm. “The findings are extremely concerning,” said Ariane Lewis, MD, associate professor of neurology at NYU Langone Health. “If determinations deviate from the 2010 AAN guidelines for determination of brain death in adults, patients may inappropriately be declared brain dead leading to stories of ‘recovery’ from brain death and loss of public trust in the use of neurologic criteria to declare death.”
The finding that 10 percent of respondents did not perform an apnea test, if it represents the actual state of practice with patients being assessed for brain death, is especially striking, independent experts said.
“The determination of apnea is a cardinal sign of brain death and therefore is a prerequisite for brain death determination,” said James L. Bernat, MD, FAAN, professor of neurology at Dartmouth Geisel School of Medicine and co-author of an editorial that accompanied the survey in Neurology. “The only justifiable reason to omit apnea testing is if doing it exposes a patient to harm. When that is the case, either brain death should not be determined, or an ancillary test should be performed, such as a test showing the absence of intracranial circulation, to confirm the presence of brain death.”
“Because apnea testing is not an elective part of brain death determination, its omission, when it can be performed safely, is a serious quality deficiency,” Dr. Bernat told Neurology Today.
Dr. Lewis offered a possible caveat. “Apnea testing is arguably the most important part of a brain death determination because it evaluates the lowest level of the brainstem, the medullary chemoreceptors,” she said. “I find the fact that apnea testing may not be performed routinely to be disturbing, but I'm not sure whether the 10 percent figure indicates that apnea testing is not performed or simply that the respondents themselves are not personally responsible for performing apnea testing.”
Dr. Lewis added, “The majority of respondents were neurologists and neurosurgeons, and at many institutions these individuals are involved in the clinical part of a brain death determination, but an intensivist performs apnea testing.”
Dr. Braksick acknowledged the limitation. “Our survey did not require the respondent to specify if they personally completed the apnea test or solicited assistance from a critical care physician,” she said in comments to Neurology Today. “When reviewing the raw data from our survey, one respondent who indicated they did not complete an apnea test added a comment that the apnea test is done by the critical care team. The remainder of the respondents that indicated they did not complete an apnea test provided no additional information to clarify this variable.”
What is clear is the need for training around a standard protocol at least among those physicians most likely to perform brain death assessments as part of routine clinical practice, experts agreed. “Brain death assessment may not be necessary for a lot of physicians, but for neuro-intensivists it's very common,” Dr. Hwang said. “The survey could help the AAN push for a creating a training course that would provide certification or credentialing of some kind.”
Dr. Lewis concurred. “It is imperative that all clinicians who perform brain death assessments be adequately trained, and that a competency assessment be performed at routine intervals, similar to the training and credentialing of clinicians for performance of the NIH Stroke Scale or the modified Rankin Scale for stroke.”
She added that the Neurocritical Care Society is in the process of creating a brain death training and assessment tool that will be available to all clinicians. “Ideally, credentialing to perform a brain death evaluation [will] ultimately require completion of this assessment.”