Consensus practice recommendations: Pediatric and adult brain … · 2020. 12. 8. · 16 age and sought to combine adult and pediatric guidance recommendations, updated with current

Draft for Public Comment

1

Consensus practice recommendations: Pediatric and adult brain death/death by neurologic 1

criteria 2

3

Report of the Guideline Subcommittee of the American Academy of Neurology 4

5

David Greer, MD, MA,1 Matthew Kirschen, MD, PhD,2 Ariane Lewis, MD,3 Gary Gronseth, MD,4 6

Alexander Rae-Grant, MD,5 Stephen Ashwal, MD,6 Maya Babu, MD, MBA,7 David Bauer, MD, MPH,8 7

Lori Billinghurst, MD, MSc,9 Amanda Corey, MD,10 Sonia Partap, MD, MS11 Michael Rubin, MD, 8

MA,12 Lori Shutter, MD,13 Courtney Takahashi, MD,1 Robert C. Tasker, MBBS, MD,14 Panayiotis 9

Varelas, MD,15 Eelco Wijdicks, MD, PhD,16 John J. Halperin, MD,17 10

11

1. Department of Neurology, Boston University School of Medicine, MA 12

2. Departments of Anesthesiology and Critical Care Medicine, Neurology, and Pediatrics, 13

Children’s Hospital of Philadelphia, Perelman School of Medicine at the University of 14

Pennsylvania, Philadelphia 15

3. Departments of Neurology and Neurosurgery, NYU Langone Medical Center, New York, NY 16

4. Department of Neurology, University of Kansas Medical Center, Kansas City 17

5. Prior Professor of Neurology, Cleveland Clinic Lerner College of Medicine, Cleveland, Ohio 18

(pending emeritus professor status) 19

6. Department of Pediatrics, Loma Linda University School of Medicine, CA 20

7. Neurosurgeon, Cleveland Clinic Martin Health, Port Saint Lucie, FL 21

8. Department of Neurosurgery, Baylor College of Medicine, Texas Children’s Hospital, Houston 22

9. Department of Neurology, University of Pennsylvania, Philadelphia 23


2

10. Atlanta VA Medical Center and Department of Radiology and Imaging Science, Emory 1

University, Atlanta, GA 2

11. Departments of Neurology and Pediatrics, Stanford University, Palo Alto, CA 3

12. Department of Neurology, University of Texas Southwestern Medical Center, Dallas 4

13. Departments of Critical Care Medicine, Neurology, and Neurosurgery, University of Pittsburgh, 5

PA 6

14. Department of Anaesthesia, Boston Children’s Hospital, MA 7

15. Department of Neurology, Albany Medical College, NY 8

16. Department of Neurology, Mayo Clinic, Rochester, MN 9

17. Department of Neurosciences, Overlook Medical Center, Summit, NJ 10

11

Address correspondence and reprint requests to American Academy of Neurology: [email protected] 12

13

Title character count: 96 14

Abstract word count: XX 15

Manuscript word count: XXX 16

17

Approved by the Guideline Subcommittee on XX, by the Quality Committee on XX, and the AAN 18

Institute Board of Directors on XX. 19

20

These consensus practice recommendations were endorsed by XX. 21

22

mailto:[email protected]


3

This practice guideline was developed with financial support from the American Academy of Neurology 1

(AAN). Authors who serve or served as AAN subcommittee members or as methodologists (G.G., 2

A.R.G., L.B., C.T, and J.J.H.) were reimbursed by the AAN for expenses related to travel to 3

subcommittee meetings where drafts of manuscripts were reviewed. 4

5

DISCLOSURE 6

D. Greer has received travel funding from Boston University; serves as Editor-in-Chief for Seminars in 7

Neurology; receives publishing royalties for 50 Studies Every Neurologist Should Know; has received 8

honoraria from American Academy of Neurology (AAN); has received research funding from Becton 9

Dickenson, Inc.; and has served as expert witness in legal proceedings. 10

M. Kirschen has received funding for travel from AAN/Child Neurology Society; has received research 11

support from Masimo, Infrascan, and the Neurocritical Care Society. 12

A. Lewis has received funding from AAN for travel to meetings; serves as Neurology Deputy Editor of 13

Disputes and Debates; and serves as Deputy Editor of Seminars in Neurology. 14

G. Gronseth has received personal compensation in the range of $10,000 for serving as an Editor, 15

Associate Editor, or Editorial Advisory Board Member for Neurology; has received personal 16

compensation in the range of $0-$499 for serving as an Editor, Associate Editor, or Editorial Advisory 17

Board Member for Brain & Life, and has received personal compensation in the range of $0-$499 for 18

serving as a member/evidence-based medicine consultant on the AAN Guideline Subcommittee. 19

A. Rae-Grant has received royalties for editing or coediting multiple textbooks with Springer Publishing 20

and Wolters-Kluwer publisher; works part time for Ebsco Industries editing point-of-care evidence-21

based information for DynaMed, a subscription-based point-of-care tool for clinicians. 22


4

S. Ashwal serves on the medical advisory board of the Tuberous Sclerosis Association and receives 1

publishing royalties as coeditor of Pediatric Neurology: Principles and Practice. 2

M. Babu has testified at a grand jury hearing relating to brain injury. 3

D. Bauer has received research funding from the National Institutes of Health (NIH) and Park-Reeves 4

collaborative. 5

L. Billinghurst has reported no conflicts of interest. 6

A. Corey has performed blinded case review of MRI and CT scans for RadMD, LLC; has received 7

travel funding from Georgia Radiological Society; and is employed by the Department of Veterans 8

Affairs. 9

S. Partap has received honoraria from AAN and the American Academy of Pediatrics; has served on 10

advisory boards for Bayer and GLG; and has received research support from the National Cancer 11

Institute. 12

M. Rubin has received travel funding from AAN and the Neurocritical Care Society; serves on the 13

editorial board for Neurology Today; and has received honoraria from the University of Texas Tyler and 14

Cooper Clinic. 15

L. Shutter serves on a scientific advisory board for SENSE NeuroDiagnostics; serves on the ACCM 16

Board of Regents; has received travel funding from SENSE NeuroDiagnostics; has received honoraria 17

from the AAN; has received research support from the Chuck Noll Foundation; receives publishing 18

royalties as coeditor of Pittsburgh Critical Care Medicine: Neurocritical Care; and has served as an 19

expert witness and counsel in legal proceedings. 20

C. Takahashi has reported no conflicts of interest. 21

R. Tasker has served as an NIH grant reviewer; serves as Senior Associate Editor for Archives of 22

Disease in Childhood; serves as ad hoc reviewer for Intensive Care Medicine; serves as editor of the 23


5

pediatric neurology section of Current Opinions in Pediatrics; receives publishing royalties from 1

Nelson’s Textbook of Pediatrics, edition 21, Oxford Handbook of Paediatrics and Child Health, edition 2

2, Roger’s Textbook of Pediatric Intensive Care, edition 5, UpToDate, 2 chapters: ICP treatment and 3

ICP recognition; has received research support from NIH and Boston Children’s Hospital. 4

P. Varelas serves on a scientific advisory board for Portola, has received honoraria from Portola; serves 5

on a speakers’ bureau for Portola; serves on an advisory board for UBC; serves on the editorial board of 6

Neurocritical Care; has received royalties for the book Seizures in Critical Care; has received research 7

funding from Marinus and Bard. 8

E. Wijdicks receives publishing royalties from Oxford Press for Brain Death. 9

J. Halperin has received personal compensation in the range of $500-$4,999 for serving as an Expert 10

Witness for Tri-Century Insurance Company; has received intellectual property interests from 3 11

publications relating to health care; has received personal compensation in the range of $0-$499 for 12

serving as a medical consultant with TelaDoc; and Dr. Halperin’s institution has received research 13

support from the NIH. 14

15

GLOSSARY 16

AAN: American Academy of Neurology 17

AAP: American Academy of Pediatrics 18

AEP: auditory evoked potentials 19

CNS: central nervous system 20

COI: conflict of interest 21

CPAP: continuous positive airway pressure 22

CV: curriculum vitae 23

ECMO: extracorporeal membrane oxygenation 24

ETT: endotracheal tube 25


6

HIBI: hypoxic-ischemic brain injury 1

MAP: mean arterial pressure 2

MRA: magnetic resonance angiography 3

BD/DNC: brain death/death by neurologic criteria 4

OCR: oculocephalic reflex 5

OVR: oculovestibular reflex 6

PEEP: peak end expiratory pressure 7

SBP: systolic blood pressure 8

SCCM: Society of Critical Care Medicine 9

SEP: somatosensory evoked potentials 10

UDDA: Uniform Determination of Death Act 11

VA: venoarterial 12

13

INTRODUCTION 14

Determination of brain death/death by neurologic criteria (BD/DNC) must be accurate and consistent. 15

BD/DNC, which occurs in individuals who have sustained catastrophic brain injury and demonstrate no 16

signs of neurologic recovery, comprises about 2% of adult in-hospital deaths and about 20% of deaths in 17

pediatric intensive care units.1, 2 The Uniform Determination of Death Act (UDDA), the legal foundation 18

for BD/DNC, recognized BD/DNC as irreversible cessation of all functions of the entire brain, including 19

the brainstem, and stipulated that determination of BD/DNC must be made in accordance with accepted 20

medical standards.3 In 2018, multiple medical societies agreed that the accepted medical standards for 21

determination of BD/DNC were the 2010 American Academy of Neurology (AAN) guideline for adults, 22

and the 2011 American Academy of Pediatrics (AAP), Society of Critical Care Medicine (SCCM), and 23

Child Neurology Society guideline for infants and children.4-6 These guidelines stipulate that BD/DNC 24

should be declared when a patient with a known cause of catastrophic brain injury has complete and 25


7

permanent loss of function of the entire brain, including the brainstem, resulting in: 1) coma, 2) 1

brainstem areflexia, and 3) apnea during a carbon dioxide challenge. 2

3

These guidelines are now a decade old and include portions that can be considered incomplete or 4

outdated, necessitating reexamination of the BD/DNC determination process to ensure diagnostic 5

accuracy. Additionally, they do not address challenging circumstances for determination of BD/DNC, 6

including patients on extracorporeal membrane oxygenation (ECMO), who are hypothermic or have 7

primary infratentorial pathology. 8

9

Thus, in concert with the Ethics, Law, and Humanities Committee, a joint committee of the AAN, the 10

American Neurological Society, and the Child Neurology Society, and with the support of AAN 11

leadership and the AAN Quality Committee, the AAN Guideline Subcommittee formed a 12

multidisciplinary panel with members from multiple medical societies to make updated formal 13

consensus recommendations about the process of BD/DNC determination. The panel felt that the 14

process for diagnosing death, either cardiopulmonary or neurologic, should be independent of patient 15

age and sought to combine adult and pediatric guidance recommendations, updated with current 16

literature and expert consensus opinion, into a single, unified guideline to ensure accuracy and 17

consistency in determining BD/DNC in persons of all ages, making note of where child-specific 18

principles apply. In addition, the panel sought to address the aforementioned clinical challenges that 19

were not included in prior guidelines. 20

21

22

AUTHOR PANEL FORMATION AND METHODOLOGY 23


8

Leadership from the AAN Guideline Subcommittee (A.R.G., J.J.H.) and the AAN Brain Death Working 1

Group (D.G., M.K., A.L.), a subgroup of the AAN Ethics, Law, and Humanities Committee, engaged a 2

broad panel of stakeholders from multiple medical societies to collaborate on an updated BD/DNC 3

guideline. Panel members were invited from the following organizations: AAP, Child Neurology 4

Society, Congress of Neurological Surgeons, Neurocritical Care Society, and SCCM. Clinicians with 5

expertise in BD/DNC were also invited. The author panel included experts (D.G., M.K., A.L., G.G., 6

S.A., M.B., D.B., L.B., A.C., S.P., M.R., L.S., C.T., R.C.T., P.V., and E.W.), process facilitators 7

(A.R.G., J.J.H.) and a methodologist (G.G.). Each potential author was required to submit an AAN 8

relationship disclosure form and a copy of his or her curriculum vitae (CV). The panel leadership 9

(A.R.G., J.J.H., G.G., D.G., M.K., A.L.) and AAN staff reviewed the relationship disclosure forms and 10

CVs for financial and intellectual conflicts of interest (COI). These documents were specifically 11

screened to exclude those individuals with a clear financial conflict and those whose profession and 12

intellectual bias would diminish the credibility of the guideline in the eyes of the intended users. Links 13

with organ procurement organizations were considered a direct conflict to participation. As required by 14

the AAN, a majority (51 percent) of the members (A.L., G.G., A.R.G., S.A., M.B., D.B., L.B., A.C., 15

S.P., M.R., C.T., R.C.T., E.W., and J.J.H.) of the author panel and the lead author (D.G.) are free of COI 16

relevant to the subject matter of this practice guideline. 17

18

The author panel met in person October 11th, 2019, to define the methodology and develop initial 19

rationales and recommendations. Because of the lack of high-quality evidence on the subject, a novel, 20

evidence-informed formal consensus process was used. This process relied on the panel experts’ detailed 21

knowledge of the literature surrounding BD/DNC to guide the development of preliminary 22

recommendations, followed by three iterative rounds of anonymous voting on each rationale and 23


9

recommendation (modified Delphi process). The intent was for recommendations to be conservative, yet 1

practical. After the first round of voting, the preliminary manuscript and initial rationales and 2

recommendations were reviewed by the AAN Guidelines Subcommittee for comment on January 11, 3

2020, and revised by panel members. After the second round of voting, the document was distributed for 4

public comment (December 14, 2020, through January 12, 2021) and all comments were analyzed and 5

integrated where appropriate. After the third round of voting, the rationales and recommendations were 6

finalized for the manuscript. 7

Recommendations were formulated according to the 2017 AAN Clinical Practice Guideline Process 8

Manual.7 For those recommendations that have not yet achieved consensus, space holder terms are 9

included in parentheses. 10

11

12

13

TERMINOLOGY 14

The term “brain death” has been used in common parlance, but the panel chose to use “brain death/death 15

by neurologic criteria” or BD/DNC to both embrace the colloquial term and emphasize that a 16

determination means more than death of the brain and that, rather, death of the person has occurred, 17

equivalent to death by cardiopulmonary criteria. This terminology was also used in the World Brain 18

Death Project.8 19

20

RECOMMENDATIONS 21

22

General principles for the BD/DNC evaluation 23


10

1. Recommendation 1 rationale. BD/DNC is declared when there is irreversible cessation of all 1

functions of the entire brain, including the brainstem (PRIN). Lack of meticulous attention to 2

stringent guidelines could lead to inappropriate or inaccurate diagnosis of BD/DNC (PRIN). 3

Consistently following a standardized process for BD/DNC determination will guard against 4

inappropriate or inaccurate diagnosis or variability in the diagnosis (INFER). The following 5

recommendations have been developed based on a critical review of the available evidence, 6

using a formal consensus-developing methodology (modified Delphi) among adult and pediatric 7

neurologists, neurosurgeons, and intensivists with expertise in the field, with input from the 8

AAP, SCCM, Child Neurology Society, Neurocritical Care Society, American College of 9

Radiology, and Congress of Neurologic Surgeons, have undergone extensive peer review, and 10

should provide the basis for determination of BD/DNC in United States institutions (PRIN). 11

• Recommendation statement 1. Unless otherwise legislated by local or regional authorities, 12

clinicians at institutions in the United States should follow the standardized process in 13

this document for the determination of BD/DNC in persons of all ages (Level A). 14

15

2. Recommendation 2 rationale. Clinical determination of BD/DNC is likely less reliable in infants 16

younger than 37 weeks corrected gestational age because some of the brainstem reflexes may not 17

be completely developed (EVID).9 18

• Recommendation statement 2. Clinicians (should) not evaluate infants younger than 37 19

weeks corrected gestational age for BD/DNC. 20

21

3. Recommendation 3 rationale. The purpose of BD/DNC evaluation is to determine whether a 22

patient meets criteria for BD/DNC (PRIN). The clinician’s primary responsibility is to the 23


11

interests of that patient (PRIN). A patient who meets criteria for BD/DNC may be a candidate for 1

organ/tissue donation (PRIN), which affects the interests of other patients. (PRIN). These 2

interests may conflict (INFER). 3

• Recommendation statement 3. To avoid conflicts of interest, clinicians involved in 4

BD/DNC determination (must) only consider the interests of their patient and avoid any 5

direct involvement in decision making regarding organ donation. 6

7

4. Recommendation 4 rationale. Before performing the BD/DNC evaluation, patients must be 8

observed for sufficient time to determine the severity and irreversibility of the brain injury and 9

exclude confounders (PRIN). During this time, patients require mechanical ventilation and often 10

hemodynamic support (PRIN). The BD/DNC evaluation requires apnea testing, which may lead 11

to hypoxemia and hemodynamic compromise (EVID).10 Apnea testing must be performed in a 12

setting with staff with appropriate expertise in managing the potential cardiopulmonary 13

complications of the test (INFER). 14

• Recommendation statement 4. Patients undergoing BD/DNC evaluation must be cared for 15

in an environment that allows for adequate observation to ensure the severity and 16

irreversibility of the brain injury and exclude confounders and has staff with appropriate 17

expertise in managing the potential cardiopulmonary complications of apnea testing 18

(Level A). 19

20

5. Recommendation 5 rationale. Spontaneous breathing, the absence of coma, intact brainstem 21

reflexes, or motor activity other than spinally mediated reflexes indicate brain function and are 22

inconsistent with BD/DNC (PRIN). 23


12

• Recommendation statement 5. Patients with any level of consciousness or preservation of 1

any brainstem reflex or who display motor movements that are centrally mediated or are 2

spontaneously breathing are not dead and must not undergo BD/DNC testing (Level A). 3

4

Qualifications of examiners 5

Previous AAN guidelines did not address qualifications for examiners; the 2011 pediatric guidelines 6

noted that 2 attending physicians, each of whom was a pediatric intensivist, neonatologist, pediatric 7

neurologist, pediatric neurosurgeon, pediatric trauma surgeon, pediatric anesthesiologist with critical 8

care training, or an adult specialist trained in neurology and/or critical care, were needed for an 9

evaluation of BD/DNC.4, 5, 11 10

11

6. Recommendation 6 rationale. Clinician competency in performing a BD/DNC evaluation is 12

important to optimize the accurate and consistent application of this process (PRIN) to avoid 13

erroneous BD/DNC determination. There may be different mechanisms by which clinicians can 14

acquire and demonstrate this competency (INFER). 15

• Recommendation statement 6a. Clinicians performing BD/DNC evaluations in adult 16

patients should demonstrate competency in the evaluation of adult patients for BD/DNC 17

by such means as completion of a supervised BD/DNC evaluation in a simulated or 18

clinical environment (Level B). Examples of appropriate training include, but are not 19

limited to, specific education in residency or fellowship or completion of a well-designed 20

on-line or in-person training course. 21

• Recommendation statement 6b. Clinicians performing BD/DNC evaluations in infants 22

and children should demonstrate competency in the BD/DNC evaluation of infants and 23


13

children by such means as completion of a supervised BD/DNC evaluation in a simulated 1

and/or clinical environment (Level B). Examples of appropriate training include, but are 2

not limited to, specific education in residency or fellowship or completion of a well-3

designed on-line or in-person training course. 4

• Recommendation statement 6c. All clinicians performing these evaluations (should) be 5

attending physicians. Advanced practice providers or trainees performing this evaluation 6

must be directly supervised by an attending physician who must be present at the 7

patient’s bedside or by telemedicine for the entirety of the evaluation. 8

9

Prerequisites for determination of BD/DNC 10

11

Identification of etiology of brain injury 12

BD/DNC can result from traumatic brain injury, vascular events (subarachnoid hemorrhage, 13

intracerebral hemorrhage, ischemic stroke), hypoxic-ischemic brain injury (HIBI), intracranial infection, 14

or metabolic disorders leading to malignant cerebral edema.1, 11, 12 However, there have been numerous 15

reports of reversible mimics of BD/DNC caused by other pathology, including, but not limited to, 16

Guillain-Barré syndrome, leptomeningeal carcinomatosis, snake bites, botulism, and high cervical cord 17

injuries.8 18

The 2010 AAN guidelines noted the necessity of neuroimaging to explain the etiology of coma, and the 19

2011 pediatric BD/DNC guidelines stated that neuroimaging “should demonstrate evidence of an acute 20

central nervous system (CNS) injury consistent with the profound loss of brain function,” 21

acknowledging that “early after acute brain injury, imaging findings may not demonstrate significant 22


14

injury, (so) in such situations, repeat studies are helpful in documenting that an acute severe brain injury 1

has occurred.”4, 5 2

3

7. Recommendation 7 rationale. BD/DNC evaluation can only be initiated when a patient has 4

sustained a catastrophic, irreversible brain injury, and the mechanism of the brain injury is 5

known to lead to BD/DNC (PRIN). When a patient is comatose with lack of brainstem reflexes 6

and apnea, and there is not an identified mechanism of brain injury that is known to lead to 7

BD/DNC, there is a risk that the clinical findings may be caused by a reversible process (PRIN). 8

Neuroimaging can assist with identifying the mechanism of brain injury and determining the 9

severity of the brain injury (PRIN). However, in patients with HIBI, there may not be visible 10

changes on initial neuroimaging studies to support the diagnosis of BD/DNC (EVID), and follow 11

up neuroimaging is advisable.13, 14 12

• Recommendation statement 7a. Clinicians should ensure that a patient has sustained a 13

catastrophic, irreversible brain injury caused by an identified mechanism that is known to 14

lead to BD/DNC before initiating a BD/DNC evaluation (Level B). 15

• Recommendation statement 7b. If a patient is comatose with lack of brainstem reflexes 16

and apnea, and there is not an identified mechanism of brain injury that is known to lead 17

to BD/DNC, the patient requires further diagnostic evaluation and should not undergo 18

evaluation for BD/DNC (Level B). 19

• Recommendation statement 7c. Neuroimaging that has been performed after sufficient 20

time has elapsed from the onset of the brain injury (should) show findings consistent with 21

a mechanism of catastrophic injury known to lead to BD/DNC. 22

23


15

Observation for irreversibility 1

The UDDA requires that BD/DNC determination only be made if an individual has sustained 2

“irreversible cessation of all functions of the entire brain, including the brainstem.”3 It was noted in the 3

2010 AAN guidelines that there is insufficient evidence to determine the minimally acceptable 4

observation period to ensure irreversible loss of function of the brain.4 The 2011 pediatric guidelines 5

recommended that BD/DNC evaluation be deferred for 24 to 48 hours or longer if there are concerns or 6

inconsistencies in the examination following cardiopulmonary resuscitation or other severe acute brain 7

injuries. The 2011 pediatric guidelines also recommended an age-based observation period between the 8

first and second clinical examinations for BD/DNC (24 hours for neonates 37 weeks up to 30 days and 9

12 hours for infants and children older than 30 days to 18 years).5 A conservative observation period 10

after brain injury and before evaluation for BD/DNC helps ensure the brain injury is irreversible. 11

Around the world, there is no standard observation period before the initiation of the evaluation for 12

BD/DNC. 13

14

8. Recommendation 8 rationale. The declaration of BD/DNC requires that the catastrophic brain 15

injury is irreversible (PRIN). A conservative observation period after the brain injury occurs and 16

before the initiation of the BD/DNC evaluation helps ensure that there is no potential for 17

recovery of brain function (PRIN). In patients with HIBI, an observation period of at least 24 18

hours is reasonable (INFER). There is no evidence to support a specific observation period 19

between brain injury of other etiologies and performance of the BD/DNC evaluation. 20

• Recommendation statement 8. Clinicians should wait a sufficient amount of time after the 21

brain injury occurs before initiating the BD/DNC evaluation to ensure there is no 22

potential for recovery of brain function (Level B). 23


16

1

9. Recommendation 9 rationale. Infants and young children with open fontanelles and unfused 2

sutures can experience different physiologic consequences of acute brain injury than older 3

children and adults (PRIN). The anterior fontanelle usually closes between 13 and 24 months 4

gestational age, but closure may be delayed beyond this period in patients with certain genetic or 5

metabolic conditions (EVID).9, 15 In patients with open fontanelles and unfused sutures, the 6

distensibility of the calvarium and dura may prevent the rise in intracranial pressure and 7

subsequent herniation syndromes caused by cerebral edema (INFER). Additionally, the 8

brainstem in infants is more resistant to HIBI than other brain regions (EVID).16-18 Thus, infants 9

and younger children may retain some brainstem function that only becomes apparent when 10

cerebral edema subsides, particularly after HIBI, which accounts for nearly two-thirds of brain 11

injury that leads to BD/DNC in pediatric populations (INFER).1 12

• Recommendation statement 9. For infants and young children, clinicians (should) wait 13

longer (e.g. at least 48 hours) after acute brain injury before clinical determination of 14

BD/DNC, particularly after HIBI. 15

16

Avoiding inaccurate declaration of BD/DNC caused by hypothermia 17

Hypothermia can affect the neurologic examination because of blunting of brainstem reflexes.12 This is 18

particularly notable when patients took or received medications that depress the CNS before being, or 19

while, hypothermic because hypothermia alters drug pharmacokinetics and pharmacodynamics, leading 20

to delayed elimination.19-22 The minimum temperature for BD/DNC evaluation for around the world 21

varies from 32˚C to 36˚C.23 The 2010 AAN BD/DNC guidelines required a minimum temperature of 22

36˚C, but the 2011 pediatric BD/DNC guidelines required 35˚C.4, 5 It is unknown how long hypothermia 23


17

can impact the neurologic examination, so there is no standard time period to delay evaluation for 1

BD/DNC after achieving normothermia.23, 24 There are 2 cases in the literature in which BD/DNC was 2

declared prematurely after rewarming from therapeutic hypothermia, and there was some subsequent 3

recovery of neurologic function.25, 26 4

5

10. Recommendation 10 rationale. Hypothermia may suppress brain function, resulting in a 6

comatose patient with brainstem areflexia and apnea, potentially leading to an inaccurate 7

determination of BD/DNC (EVID).25, 26 Hypothermia may result from either environmental 8

exposure or be induced as a neuroprotective therapy in some patients after cardiac arrest or other 9

types of acute brain injury (PRIN). Hypothermia delays the clearance of medications that depress 10

the CNS (EVID).19-22 There is no evidence to support a specific observation period between 11

establishment of a minimal core body temperature and performance of the BD/DNC evaluation 12

(PRIN). Providing a conservative observation period for patients who have been hypothermic 13

before initiating the BD/DNC evaluation should help ensure that there is no potential for 14

recovery of brain function after the hypothermia resolves and medications that depress the CNS 15

have had adequate time to be cleared (INFER). 16

• Recommendation statement 10a. Clinicians should ensure that patients’ core body 17

temperatures are ≥36˚C before performing a BD/DNC evaluation (Level B). 18

• Recommendation statement 10b. In patients whose core body temperature has been 19

35.5˚C, clinicians (should) wait a minimum of 24 hours after the patient has been 20

rewarmed to ≥36˚C before evaluating for BD/DNC. Clinicians (should) allow a longer 21

observation period in the presence of renal or hepatic dysfunction to ensure adequate 22

clearance of medications that can depress the CNS. 23


18

1

Avoiding inaccurate declaration of BD/DNC caused by hypotension 2

Hypotension may suppress brain function and lead to a false positive BD/DNC declaration. The 2010 3

AAN BD/DNC guidelines required that systolic blood pressure (SBP) be ≥100 mm Hg and the 2011 4

pediatric BD/DNC guidelines required that systolic or mean arterial blood pressure be not less than 2 5

standard deviations below age-appropriate norms.4, 5 6

7

There are no data to suggest a minimum blood pressure needed to ensure that a brain-injured patient’s 8

lack of responsiveness could not be attributable to lack of cerebral perfusion. Published reference ranges 9

for children are influenced by many factors, including age, gender, height, ethnicity, and measurement 10

method. Individual variability in blood pressure and underlying medical conditions (e.g., chronic kidney 11

disease, endocrinopathies, chronic cardiac or vascular conditions) can also affect baseline blood 12

pressure. 13

14

11. Recommendation 11 rationale. Hypotension may suppress brain function, which can result in a 15

comatose patient with brainstem areflexia and apnea and lead to an inaccurate determination of 16

brain death (PRIN). 17

• Recommendation statement 11a. Clinicians should ensure that the patient is not 18

hypotensive before performing a BD/DNC evaluation (Level B). Intravenous 19

administration of volume (crystalloid or colloid) vasopressors or inotropes as needed for 20

management of blood pressure before and during BD/DNC evaluation may facilitate this. 21

• Recommendation statement 11b. In patients older than 18 years, clinicians (should) 22

maintain SBP ≥100 mm Hg and mean arterial pressure (MAP) ≥75 mm Hg. In patients 23


19

aged 18 years and younger, clinicians (should) maintain SBP and MAP >10th percentile 1

for age (table 1). 2

• Recommendation 11c. For patients older than 18 years receiving venoarterial (VA) 3

ECMO, clinicians (should) target MAP ≥75 mm Hg. For patients aged 18 years or 4

younger receiving VA ECMO, clinicians (should) target MAP >10th percentile for age. 5

• Recommendation 11d. If an individual has a baseline blood pressure that varies 6

significantly from their age-based normal range, clinicians (should) target a blood 7

pressure that approximates the known chronic baseline for that individual patient. 8

9

Avoiding inaccurate declaration of BD/DNC caused by drugs/medications and metabolic 10

derangements 11

With regard to drugs/medications and metabolic derangements, the 2010 AAN BD/DNC guidelines: 1) 12

required exclusion of “CNS-depressant drug effect by history, drug screen, calculation of clearance 13

using 5 times the drug’s half-life (assuming normal hepatic and renal function), or if available, drug 14

plasma levels below the therapeutic range;” 2) noted that a determination of BD/DNC could not be 15

performed after recent administration of neuromuscular blocking agents, which could be ruled out by 16

presence of a train of 4 twitches with maximal ulnar stimulation; 3) noted the legal alcohol limit of 17

0.08% was a practical threshold below which determination of BD/DNC could be performed; and 4) 18

required exclusion of “severe electrolyte, acid-base, or endocrine disturbance (defined by severe acidosis 19

or laboratory abnormalities markedly deviated from the norm).”4 Similarly, the 2011 pediatric BD/DNC 20

guidelines: 1) required exclusion of drug or alcohol intoxication by checking levels, when available, to 21

ensure they were in the low to mid-therapeutic range before determination of BD/DNC or waiting 22

several half-lives; 2) noted that adequate clearance of neuromuscular blocking agents should be 23


20

confirmed via demonstration of a twitch response to a nerve stimulator; and 3) noted that severe 1

electrolyte, hyper- or hypoglycemia, severe pH disturbances, severe hepatic or renal dysfunction, or 2

inborn errors of metabolism may cause reversible coma and should be identified and treated before 3

evaluation for BD/DNC.5 The American College of Medical Toxicology provided a position statement 4

regarding BD/DNC determination after drug overdose in 2017.27 5

6

While we noted that there is no scientific rationale to define specific laboratory results that are/are not 7

acceptable in BD/DNC, and that a paucity of protocols around the world stipulate such values,23 we 8

recognized that the acceptability of a given value is commonly questioned. Acknowledging that a 9

multitude of derangements could result in coma and even impair brainstem reflexes, and that 10

determination of cutoff values is arbitrary, we proposed general guidance on specific laboratory result 11

abnormalities which, if identified via clinically indicated testing, warrant correction before the clinical 12

evaluation for BD/DNC and/or performance of ancillary testing (tables 2 and 3). It should be noted that 13

while most of the noted laboratory results are obtained routinely, ammonia or thyroid function are not 14

routinely measured. 15

16

12. Recommendation 12 rationale. Conditions such as metabolic derangements, intoxication, or 17

medications can depress the CNS and result in a comatose patient with brainstem areflexia and 18

apnea, which may lead to an inaccurate determination of BD/DNC (EVID).8 19

• Recommendation statement 12a. Clinicians must ensure that metabolic derangements, 20

intoxication, and medications that depress the CNS are excluded, adequately corrected, or 21

eliminated before evaluating patients for BD/DNC, as clinically appropriate (Level A): 22

Specifically, clinicians must: 23


21

a. Ensure a toxicology screen is negative if clinically indicated 1

b. Ensure the alcohol level is ≤80 mg/dL if clinically indicated 2

c. Ensure drug levels for medications that suppress CNS function, if available, are in 3

the therapeutic or subtherapeutic range and not felt to contribute to the neurologic 4

state: 5

1. Allowing at least five half-lives for all CNS depressing 6

medications or intoxicants to pass, and longer if there is renal or 7

hepatic dysfunction or if the patient was hypothermic 8

2. Accounting for age-dependent metabolism of potentially 9

depressing medications in infants and young children 10

d. Exclude severe metabolic, acid-base, and endocrine derangements 11

e. Exclude the effect of pharmacologic paralysis, if administered or suspected, through 12

use of a train-of-four stimulator or demonstration of deep tendon reflexes 13

• Recommendation statement 12b. If metabolic derangements are unable to be adequately 14

corrected, clinicians (may) proceed with the clinical BD/DNC evaluation, and if 15

consistent with BD/DNC, perform ancillary testing. 16

17

Performing the BD/DNC neurologic examination 18

19

Number of examinations required 20

In the first AAN practice parameter on BD/DNC (1995), following the initial examination for 21

determination of BD/DNC, a repeat clinical evaluation 6 hours later was advised (option), but no firm 22

recommendation could be provided, and it was acknowledged that the interval is arbitrary.11 The 2010 23


22

AAN guideline noted a single examination was the minimum standard for BD/DNC determination in 1

patients older than 18 years.4 In contrast, 2 examinations with a stipulated observation period between 2

them were required for BD/DNC determination in patients younger than 18 years in the 1987 and 2011 3

pediatric guidelines.5, 28 4

5

13. Recommendation 13 rationale. The neurologic examination is a required component of the 6

BD/DNC evaluation (PRIN). Patients who meet criteria for BD/DNC have not been found to 7

have recovery of neurologic function between a first and second evaluation for BD/DNC 8

(EVID).22 However, using 2 independent, qualified examiners for the BD/DNC evaluation helps 9

ensure the accuracy of BD/DNC determination (PRIN). However, there is no physiological 10

reason to require an observation period between these 2 evaluations (INFER). The observation 11

period after the brain injury and before the initiation of the BD/DNC evaluation ensures the 12

irreversible nature of the catastrophic brain injury (see rationale/recommendation statement 7). 13

There is no physiological reason for the number of examinations to differ between adult and 14

pediatric patients (INFER). 15

16

• Recommendation statement 13a. To perform a BD/DNC evaluation, clinicians (must) 17

review the medical record to determine that the patient has sustained a catastrophic, 18

irreversible brain injury with a mechanism of brain injury that is known to lead to 19

BD/DNC and that confounders to the evaluation have been excluded. 20

• Recommendation 13b. Two clinicians (should) perform 2 separate and independent 21

BD/DNC evaluations. There is no need to wait a specific period of time between these 2 22

evaluations. 23


23

1

14. Recommendation 14 rationale. Accurate reliable assessment of head, neck, and limb motor 2

responses can be confounded by severe neuromuscular disorders, sensory neuropathies, spinal 3

cord injuries, and/or facial trauma and swelling (PRIN). Accurate reliable assessment of the 4

pupillary light reflex can be impeded by corneal trauma, severe orbital/scleral edema/chemosis, 5

ophthalmic surgery, anophthalmia, and/or use of ocular or systemic anticholinergic medications 6

(PRIN). Accurate reliable assessment of the oculocephalic reflexes (OCR) and/or oculovestibular 7

reflexes (OVR) can be impeded by anophthalmia, severe orbital/scleral edema/chemosis, skull 8

fracture, cervical spine fracture or a ruptured tympanic membrane (PRIN). Accurate reliable 9

assessment of the corneal reflex can be impeded by anophthalmia and severe orbital/scleral 10

edema/chemosis (PRIN). Accurate reliable assessment of the gag and cough reflexes can be 11

confounded by high cervical injury (PRIN). Other injuries or confounders may also interfere 12

with the accurate and/or reliable evaluation of the neurologic examination for BD/DNC (EVID).8 13

• Recommendation statement 14a. When the accurate reliable evaluation of a component of 14

the BD/DNC neurologic examination cannot be assessed safely, clinicians must perform 15

ancillary testing to complete BD/DNC determination (Level B). 16

• Recommendation statement 14b. All elements of the BD/DNC neurologic examination 17

included here that can be assessed must be assessed, and findings must be consistent with 18

BD/DNC (Level A). 19

20

Components of the BD/DNC neurologic examination 21

The clinical examination findings supporting BD/DNC described in this guideline are consistent with 22

those included in prior guidelines, both in the United States and around the world (table 4).4, 5, 23 23


24

1

Assessment for unresponsiveness 2

15. Recommendation 15 rationale. The presence of coma with no response to noxious stimulation is 3

a key component of the BD/DNC neurologic examination (PRIN). 4

• Recommendation statement 15. Clinicians performing the BD/DNC neurologic 5

examination (must) ensure that the patient is comatose and unresponsive to visual, 6

auditory, and tactile stimulation. 7

8

Assessment for motor response 9

16. Recommendation 16 rationale. The absence of cerebrally mediated motor responses to noxious 10

stimuli is a key component of BD/DNC neurologic examination (PRIN). Retained spinally 11

mediated reflexes can be seen in the setting of BD/DNC and do not invalidate the diagnosis of 12

BD/DNC (EVID).29 It can sometimes be challenging to determine if a movement is cerebrally or 13

spinally mediated based solely on the clinical examination (PRIN). When such difficulties arise, 14

review with more experienced clinicians may be helpful (INFER). 15

• Recommendation statement 16a. Clinicians performing the BD/DNC neurologic 16

examination must ensure that the patient has no motor responses, other than spinally 17

mediated reflexes, of the head/face, neck, and extremities after application of noxious 18

stimuli to the head/face, trunk, and limbs (Level A). 19

• Recommendation statement 16b. If it is unclear whether observed limb movements are 20

spinally mediated, clinicians must not declare the patient BD/DNC without an ancillary 21

test (Level B). 22

23


25

Assessment of the pupillary light reflex 1

17. Recommendation 17 rationale. The pupillary light reflex is a brainstem reflex and part of the 2

BD/DNC neurologic examination (PRIN). 3

• Recommendation statement 17a. Clinicians performing BD/DNC neurologic 4

examinations must determine that there are no pupillary responses to light bilaterally 5

(Level A). 6

• Recommendation statement 17b. Constricted pupils (


26

1

Assessment of the corneal reflex 2

19. Recommendation 19 rationale. The corneal reflex is a brainstem reflex (PRIN) and part of the 3

BD/DNC neurologic examination. 4


examination must determine that there are no corneal reflexes bilaterally (Level A). 6

7

Assessment of the gag and cough reflexes 8

20. Recommendation 20 rationale. Medullary function in the BD/DNC neurologic examination is 9

clinically assessed by the cough and gag reflexes and the apnea test (PRIN). The gag reflex is a 10

brainstem reflex (PRIN). The cough reflex is a brainstem reflex (PRIN). 11


examination must determine that both the gag and cough reflex are absent (Level A). 13

14

Assessment of the sucking and rooting reflexes 15

21. Recommendation 21 rationale. The sucking reflex is a centrally mediated primitive reflex in 16

infants that becomes a voluntary reflex around 4 months of age (PRIN). The rooting reflex is a 17

centrally mediated primitive reflex in infants that disappears between 3 and 6 months of age 18

(PRIN). 19

• Recommendation statement 21. In infants younger than 6 months, clinicians performing 20

the BD/DNC neurologic examination must determine that there is no sucking or rooting 21

reflex (Level A). 22

23


27

Apnea testing as part of the BD/DNC evaluation 1

2

Number of apnea tests required 3

Both the 1995 and 2010 AAN guidelines for adults recommended performance of a single apnea test.4, 11 4

In contrast, the 1987 and 2011 pediatric guidelines recommended performance of 2 apnea tests.5, 28 In 5

seeking to determine the appropriate number of apnea tests to prescribe, the panel considered 1) the 6

desire to err on the side of being conservative and 2) the fact that there is potential for hemodynamic 7

instability during apnea testing, recognizing that individual practitioners/institutions could weigh the 8

risks and benefits of performing apnea testing more than once on a given patient. 9

10

22. Recommendation 22 rationale. Apnea testing is a required component of the BD/DNC 11

evaluation (PRIN). Apnea testing may lead to hypoxemia and hemodynamic compromise 12

(EVID).10 There are no physiological reasons or empiric evidence to support performing more 13

than 1 apnea test to declare BD/DNC (INFER/EVID).10 14

• Recommendation statement 22. Clinicians (must) perform at least 1 apnea test after the 15

final BD/DNC neurologic examination. 16

17

Procedure for performing the apnea test 18

Safe performance of apnea testing can be achieved by numerous means.8, 10, 23 Apneic oxygen diffusion 19

is the most commonly used technique in adults, with disconnection from the ventilator and providing 20

oxygen via a catheter placed just above the carina. Alternative methods for testing include delivering 21

100% oxygen via continuous positive airway pressure (CPAP) at 10 cm H20 while the patient is still on 22

the ventilator or with the use of a T-piece tube with a CPAP valve at the outflow end or delivering 100% 23


28

oxygen via a flow-inflating resuscitation bag with a functioning peak end expiratory pressure (PEEP) 1

valve. The level of PEEP using these methods is often set to the same PEEP as the ventilator before 2

disconnection. Atelectasis from the drop in mean airway pressure while disconnecting the ventilator can 3

cause shunt and hypoxemia. Lower rates of hypoxemia are described in adults using CPAP versus 4

tracheal insufflation, although CO2 rise and premature termination of the apnea test are similar between 5

the different methods of apneic oxygenation.31-34 The use of CPAP in children can prevent hypoxemia 6

and early termination of the apnea test.35 Other methods of apneic oxygenation include bulk diffusion36 7

and carbon dioxide augmentation,37 although the latter does not affect pH and thus is not preferred. 8

Performing apnea testing on the ventilator can be complicated by auto cycling caused by cardiac 9

pulsations or condensation in the tubing and difficulty distinguishing between patient- and autocycle-10

initiated breaths. 11

12

Although the 2010 AAN BD/DNC guideline indicated that the PaCO2 threshold for determination of 13

BD/DNC during apnea testing was ≥60 mm Hg (or 20 mm Hg over baseline normal), the 2011 pediatric 14

BD/DNC guidelines specified a threshold of ≥60 mm Hg and ≥20 mm Hg over baseline.4, 5 The 15

distinction between whether “baseline” referred to premorbid PaCO2 or PaCO2 before commencement of 16

apnea testing was not specified in either BD/DNC guideline. Selection of targets for this challenge is 17

arbitrary because no scientific data demonstrates a specific PaCO2 above which medullary 18

chemoreceptors would prompt respiration if they were functional. However, much of the world uses the 19

PaCO2 target of ≥60 mm Hg for determination of BD/DNC.23 20

21


29

Neither the 2010 AAN BD/DNC guideline nor the 2011 BD/DNC guideline dictate a pH target; 1

similarly, there is only a pH target for apnea testing in a few countries around the world.4, 5, 23 However, 2

the trigger for medullary chemoreceptors to initiate respiration is both hypercarbia and acidosis.38, 39 3

The procedure for safely and properly performing apnea testing for determination of BD/DNC described 4

in this guideline is consistent with that which is included in prior guidelines, both in the United States 5

and around the world.4, 5, 23 6

7

23. Recommendation rationale 23. Respiration is stimulated in the medullary chemoreceptors by 8

hypercarbia and secondary acidosis (PRIN). The BD/DNC evaluation requires an evaluation for 9

spontaneous respiratory effort in response to a hypercarbic and acidotic challenge by apnea 10

testing (PRIN). Spontaneous respirations are a sign of brainstem function and are not compatible 11

with BD/DNC (PRIN). Apnea testing can lead to complications, including hypoxemia and 12

hemodynamic compromise with cardiovascular collapse requiring cardiopulmonary resuscitation 13

(EVID).40-44 These complications can be prevented via preoxygenation, fluids, and medications 14

as appropriate (PRIN). Hypoxemia and hypotension before apnea testing are associated with a 15

higher risk for cardiopulmonary decompensation during apnea testing (EVID).10 Patients with 16

severe respiratory and/or cardiovascular failure may be at higher risk for cardiopulmonary 17

decompensation during apnea testing (INFER). Patients with chronic hypoxemia caused by 18

cyanotic heart disease are at higher risk of cardiopulmonary decompensation during apnea 19

testing (INFER). 20

• Recommendation statement 23a. Before attempting apnea testing, clinicians (must) 21

ensure that the patient’s risk of cardiopulmonary decompensation during apnea testing is 22


30

assessed and is acceptable. Specifically, clinicians (must) ensure that the patient is not 1

hypoxemic, hypotensive, or hypovolemic before starting the apnea test. 2

• Recommendation statement 23b. Clinicians (must) not perform apnea testing in patients 3

with chronic hypoxemia caused by cyanotic heart disease. An ancillary study must be 4

conducted instead to assist with determination of BD/DNC. 5

6

24. Recommendation 24 rationale. The apnea test evaluates for spontaneous respiratory effort in 7

response to a hypercarbic and acidotic challenge (PRIN). Before apnea testing, patients who are 8

not known to have chronic hypercarbia need to have normal PaCO2 and pH values (INFER). In 9

patients with chronic hypercarbia (e.g., chronic obstructive pulmonary disease), medullary 10

chemoreceptors can develop a diminished response to CO2 over time, especially at CO2 11

concentrations below the patient’s baseline level (PRIN). Patients with chronic hypercarbia need 12

to have their pre-apnea testing PaCO2 level be at their chronic baseline level before apnea testing 13

(INFER). 14

• Recommendation statement 24a. Clinicians must ensure that patients have normal PaCO2 15

(35-45 mm Hg) and pH (7.35-7.45) levels before apnea testing, provided the patient is 16

not known to be hypercarbic at baseline (Level A). 17

• Recommendation statement 24b. If a patient is known to have chronic hypercarbia, and 18

the patient’s chronic baseline level is known, the PaCO2 level before apnea testing should 19

be at the patient’s chronic baseline level (Level B). 20

• Recommendation 24c. If a patient is known to have chronic hypercarbia, but the patient’s 21

chronic baseline PaCO2 level is not known, the PaCO2 before apnea testing (should) be at 22


31

the patient’s estimated chronic baseline level. However, in this circumstance if apnea is 1

present, clinicians should perform ancillary testing in addition. 2

3

25. Recommendation 25 rationale. Apnea testing can lead to hypoxia, hypotension, arrhythmias, and 4

pneumothorax if appropriate technique is not used (EVID).10 Adherence to a protocol ensures the 5

proper performance of the apnea test and helps avoid potential complications (PRIN). A person 6

who demonstrates spontaneous respiratory effort during apnea testing does not meet criteria for 7

BD/DNC (PRIN). In the absence of respirations, the PaCO2 increases by approximately 3 mm Hg 8

per min (EVID) in normothermic individuals.10 Failure to abort the apnea test because of 9

hemodynamic instability or hypoxemia could lead to cardiovascular collapse and 10

cardiopulmonary arrest (EVID).10 11

• Recommendation statement 25a. Clinicians should pre-oxygenate the patient with 12

100% oxygen for at least 10 minutes before apnea testing to achieve a PaO2 >200 13

mm Hg (Level B). 14

• Recommendation statement 25b. To allow multiple arterial blood gas (ABG) 15

measurements and reliably monitor the patient’s hemodynamic status during apnea 16

testing, clinicians should ensure that the patient has an invasive arterial catheter 17

whenever possible (Level B). 18

• Recommendation statement 25c. Clinicians should perform ABG measurement 19

immediately before disconnecting the ventilator to determine the baseline PaO2, 20

PaCO2, and pH levels (Level B). 21

• Recommendation statement 25d. Clinicians (must) assure adequate oxygenation 22

during apnea testing by such means as (1) stopping intermittent mandatory 23


32

ventilation and disconnecting the ventilator from the patient’s endotracheal tube 1

(ETT)/tracheostomy and delivering 100% oxygen at a rate of 4˗6 L/min through a 2

catheter that is


33

• Recommendation statement 25h. Clinicians (should) abort apnea testing if the 1

patient experiences hemodynamic instability or hypoxemia, for example, if any of 2

the following occur at any point during the apnea test: 3

a. SBP ≤100 mm Hg or MAP ≤75 mm Hg in adults or SBP or MAP 200 mm Hg and re-establishing 17

normal PaCO2 and pH levels (return to Recommendation statement 21a). 18

• Recommendation statement 25k. If the patient experienced hypoxemia during apnea 19

testing and the pH and PaCO2 level criteria were not reached, clinicians (should) 20

either repeat the apnea testing using an alternative apneic oxygenation method that 21

maintains functional residual capacity (e.g., CPAP via a flow-inflating resuscitation 22


34

bag), repeat the entire BD/DNC evaluation when apnea testing can be safely 1

completed, or perform an ancillary test. 2

• Recommendation statement 25l. If the patient experienced hypotension during 3

apnea testing and the pH and PaCO2 level criteria were not reached, clinicians 4

(should) either repeat apnea testing after augmenting the blood pressure, repeat the 5

entire BD/DNC evaluation when apnea testing can be safely completed, or perform 6

an ancillary test. 7

• Recommendation statement 25m. If the patient developed a cardiac arrhythmia with 8

hemodynamic instability during apnea testing, and the pH and PaCO2 level criteria 9

were not reached, clinicians (should) repeat the entire BD/DNC evaluation when 10

apnea testing can be safely completed or perform an ancillary test. 11

12

Procedure for performing the apnea test in patients on ECMO 13

The methodology for apnea testing on ECMO was not included in the prior adult or pediatric BD/DNC 14

guidelines.4, 5 Despite this, apnea testing is performed regularly in patients on ECMO.34, 45-51 The 15

physiologic principles when apnea testing is performed while on cardiopulmonary mechanical support 16

with ECMO are identical to when apnea testing is performed off ECMO‒respiration is stimulated in the 17

medullary chemoreceptors by hypercarbia and secondary acidosis. 18

19

When not on VA ECMO, the PaCO2 and pH levels obtained from a distal arterial catheter approximates 20

the values in the cerebral circulation. Similarly, when a patient on VA ECMO has an absence of cardiac 21

contractility, blood sampled from a distal arterial line will approximate the cerebral circulation. 22

However, for a patient on VA ECMO who has some native myocardial function, blood flow from the 23


35

native heart and lungs mixes with blood from the ECMO circulation at a point in the aorta called the 1

“watershed” or “mixing point”.52, 53 Gas tensions measured from a distal arterial line in this situation do 2

not approximate the cerebral circulation, so ECMO arterial cannula position and the potential 3

contributions to cerebral blood flow from both the ECMO circuit and the patient’s native cardiac output 4

affect decision-making about the location of arterial blood gas sampling.53 Adjusting the sweep gas flow 5

rate or titrating exogenous CO2 into the ECMO circuit leads to increased PaCO2 levels.34, 49, 50, 54 6

7

26. Recommendation 26 rationale. Patients on ECMO may progress to BD/DNC, and modifications 8

to apnea testing are necessary because of differences in cardiopulmonary support (PRIN). 9

Preoxygenation through the membrane lung and ventilator minimizes hypoxia for patients on 10

ECMO (PRIN). The PaCO2 level for patients on ECMO can be increased via the addition of 11

exogenous CO2 in the ECMO circuit or a decrease in the sweep gas flow (PRIN). Locations of 12

arterial blood gas sampling from the patient must consider ECMO arterial cannula position and 13

the potential contributions to cerebral blood flow from both the ECMO circuit and the patient’s 14

native cardiac output (PRIN). 15

• Recommendation statement 26. Clinicians (should) adhere to the following protocol for 16

apnea testing on ECMO: 17

a) Preoxygenate by maximizing the oxygen on the ventilator and through the 18

membrane lung. 19

b) To achieve an adequate increase in PaCO2 level, either titrate exogenous 20

CO2 into the ECMO circuit or adjust the sweep gas flow rate to 0.2˗1 L/min. 21

c) Sample ABGs from both the patient’s distal arterial line and the ECMO 22

circuit postoxygenator for patients on VA ECMO (Level B). PaCO2 and pH levels 23


36

from both locations must meet BD/DNC criteria for the apnea test to be consistent 1

with BD/DNC. This ensures that independent of the mixing point, the PaCO2 and 2

pH levels in the cerebral circulation meet BD/DNC criteria. For patients on 3

venovenous ECMO, sample ABGs only from the patient’s distal arterial line. 4

d) Hypotension during apnea testing on ECMO can be managed by 5

increasing ECMO flows, intravenous fluid administration, or 6

vasopressor/ionotropic support. 7

8

Ancillary testing as part of the BD/DNC evaluation 9

10

Indications for ancillary testing 11

Under some circumstances, the entire BD/DNC clinical examination and apnea test cannot be 12

completed, such as in patients with injuries to the head and neck that preclude evaluation of cranial 13

nerve reflexes or when apnea testing cannot be completed safely because of the patient’s underlying 14

medical condition. Both the 2010 AAN and 2011 pediatrics guidelines state that ancillary tests can be 15

used when uncertainty exists about the reliability of parts of the neurologic examination or when the 16

apnea test cannot be performed because of the underlying medical condition of the patient.4, 5 The 17

pediatrics guidelines further state that ancillary testing can be used if a medication effect may be present 18

or to shorten the duration of the inter-examination observation period.5 19

20

27. Recommendation 27 rationale. In most circumstances, the diagnosis of BD/DNC is based on the 21

clinical demonstration of coma, absent brainstem reflexes, and apnea and does not require 22

ancillary testing (PRIN). There are some situations where the BD/DNC neurologic examinations 23


37

or apnea tests cannot be completed, or the findings cannot be interpreted adequately (PRIN). 1

Although ancillary tests have previously been used in the setting of confounding factors such as 2

hypothermia or high levels of sedating medications (PRIN), the effects of these factors on 3

ancillary tests and cerebral perfusion in particular are poorly defined (PRIN), confounding 4

interpretation of cerebral perfusion changes (INFER). The presence of an open fontanelle, skull 5

fracture, skull defect (e.g., craniectomy), or CSF diversion device does not preclude performance 6

of a complete clinical assessment and apnea test (PRIN). Hospital or state policies may dictate 7

indications for ancillary tests and ancillary test selection (PRIN). 8

• Recommendation statement 27a. Clinicians should only perform ancillary testing to assist 9

with the diagnosis of BD/DNC when the BD/DNC neurologic examinations or apnea test 10

cannot be completed, or the findings cannot be interpreted adequately (Level A). 11

• Recommendation statement 27b. Clinicians should use ancillary testing in the following 12

circumstances (Level B): 13

a. Injuries, such as fractures of the cervical spine, skull base, or orbits, or severe facial 14

injuries or abnormalities that preclude accurate pupillary light, corneal, 15

oculovestibular, oculocephalic, cough, or gag reflex assessment or injuries to the 16

cervical spinal cord that limit the adequate assessment of extremity movement or 17

spontaneous respirations 18

b. The inability to perform or complete the apnea test safely because of the patient’s 19

risk of cardiopulmonary decompensation or the inability to interpret the PaCO2 levels 20

in a patient with chronic hypoxemia for whom the chronic baseline PaCO2 level is 21

unknown 22


38

c. Neurologic examination findings that may be difficult to interpret, such as limb 1

movements that may or may not be spinally mediated 2

d. Metabolic derangements that are unable to be adequately corrected 3

• Recommendation statement 27c. Clinicians should not use ancillary tests to assist in the 4

diagnosis of BD/DNC in the setting of hypothermia or high levels of sedating medications or 5

to avoid performing otherwise testable elements of the BD/DNC assessment (Level B). 6

• Recommendation statement 27d. If ancillary testing is needed to diagnose BD/DNC, before 7

proceeding to an ancillary test, the BD/DNC neurologic examinations and apnea test still 8

need to be performed to the fullest extent possible, and findings (must) be consistent with 9

BD/DNC. 10

• Recommendation statement 27e. If any findings on the BD/DNC neurologic examinations or 11

apnea test are consistent with brain-mediated activity, the patient does not meet criteria for 12

BD/DNC, and ancillary testing must not be performed (Level A). 13

• Recommendation statement 27f. In patients who meet all clinical criteria for BD DNC, 14

clinicians (should) not perform ancillary testing solely because of the presence of an open 15

fontanelle, skull fracture, skull defect (e.g., craniectomy), or CSF diversion device. 16

17

Performance of ancillary tests 18

Several different modalities are available to evaluate for electrophysiological function or perfusion of 19

the brain. However, all ancillary tests have flaws, and, as discussed below, none are 100% sensitive and 20

specific (tables 5 and 6). In identifying the tests that could be used when ancillary testing is indicated, 21

specificity is paramount as a false positive result (i.e., patient is alive and ancillary test is consistent with 22

BD/DNC) could lead to an inaccurate diagnosis of BD/DNC. 23


39

1

Tests of electrophysiological function 2

Electroencephalography 3

The 2010 AAN and 2011 pediatric BD/DNC guidelines include EEG as an acceptable ancillary test, but 4

they do not include evoked potentials.4, 5 Although the EEG was the only ancillary test included in the 5

first U.S. delineation of the BD/DNC construct by a group at Harvard Medical School in 1968,55, 56 and 6

it has consistently been recommended ever since, it has at least 1 fundamental limitation–it assesses the 7

function of the cerebral hemispheres and not deeper structures, notably the brainstem. Even the absence 8

of any detectable EEG activity is not informative about the presence or absence of brainstem function, 9

which is problematic because one of the more common reasons for performing an ancillary test is the 10

inability to fully assess a patient’s brainstem function. 11

12

28. Recommendation 28 rationale. Neurophysiologic tests, including EEG, auditory evoked 13

potentials (AEP), and somatosensory evoked potentials (SEP) have historically been used as 14

ancillary tests to assist with BD/DNC determination. EEG only assesses the function of the 15

cerebral hemispheres and not the brainstem (PRIN). Thus, the absence of any detectable EEG 16

activity does not provide information about the presence or absence of brainstem function 17

(PRIN). Yet, one of the primary indications for ancillary testing is when the clinical evaluation 18

of brainstem function, including apnea testing, cannot be completed. Studies of the anatomically 19

limited somatosensory and brainstem auditory pathways provide incomplete information about 20

the entire brainstem (PRIN). Thus, the addition of AEPs or SEPs cannot completely compensate 21

for this inherent limitation of EEG for the purpose of assisting with BD/DNC determination. 22


40

• Recommendation statement 28. Clinicians should not use EEG, AEPs, or SEPs as ancillary 1

tests to assist with the diagnosis of BD/DNC (Level B). 2

3

Tests of cerebral perfusion 4

Four-vessel catheter angiography 5

29. Recommendation 29 rationale. Catheter angiography has long been considered the gold-standard 6

perfusion study, as it allows direct and dynamic assessment of intracranial flow through manual 7

injection under pressure with the catheter positioned in the great vessels leading to the brain 8

(PRIN).57 Flow arrests at the point of entry of the vessels to the dura in both the anterior and 9

posterior circulation, the result of tamponade from highly elevated intracranial pressure. The 10

absence of intracranial perfusion is a result of intracranial pressure that has increased beyond 11

MAP and is consistent with BD/DNC (PRIN). 12

• Recommendation statement 29. Clinicians (may) use conventional 4-vessel catheter 13

angiography as an ancillary test to aid in the diagnosis of BD/DNC. 14

15

Radionuclide perfusion scintigraphy 16

30. Recommendation 30 rationale. Radionuclide cerebral scintigraphy is practical (PRIN) and 17

widely used in adults and children to assist with BD/DNC determination (EVID). Standards 18

established by the Society of Nuclear Medicine and Molecular Imaging and the American 19

College of Radiology must be followed when applicable to ensure appropriate interpretation of 20

images (EVID).58, 59 99Tc-HMPAO (hexamethylpropylebeamineoxime) is lipid soluble and 21

brain-specific and is used with delayed planar or SPECT imaging to demonstrate absence of 22

intracranial blood flow and cerebral perfusion. SPECT imaging provides better visualization of 23


41

posterior fossa and brainstem structures but requires patient transport to the scanner (EVID). 1

Nonspecific, nonlipophilic agents such as 99Tc-DTPA (diethylene triamine penta-acetic acid) 2

use planar imaging and can be performed at the bedside but provide only vascular imaging and 3

do not assess tissue perfusion.58 4

Recommendation statement 30. Clinicians (may) use either SPECT radionuclide 5

perfusion scintigraphy with a brain-specific agent or planar radionucleotide angiography 6

with either a brain-specific or non—brain-specific agent as an ancillary test to aid in the 7

diagnosis of BD/DNC. 8

9

Transcranial doppler ultrasound 10

31. Recommendation 31 rationale. Transcranial doppler ultrasonography is widely available and 11

relatively easy to perform at the bedside. In approximately 10% of adults, skull thickness 12

precludes adequate insonation (EVID).60 Detection of oscillating flow or systolic spikes in 13

proximal large intracranial arteries (internal carotid, middle and anterior cerebral, posterior 14

cerebral, basilar and vertebral arteries) is consistent with BD/DNC (EVID).60, 61 It is not difficult 15

to differentiate between the absence of any signal for technical reasons and the flow pattern seen 16

in patients who are BD/DNC (PRIN). It has not been validated in patients younger than 18 years 17

(EVID). 18

• Recommendation statement 31. Clinicians (may) use transcranial doppler 19

ultrasonography in patients aged 18 years or older as an ancillary test to aid in the 20

diagnosis of BD/DNC. 21

22

CT and magnetic resonance angiography 23


42

32. Recommendation 32 rationale. Because of the availability of CT angiography, this technique has 1

been used with increasing frequency to assist with BD/DNC diagnosis (EVID).62, 63 However, 2

the available data demonstrate a lack of validation and the potential for false positives (EVID).62, 3

64 Variations in protocols with different sensitivity and specificity, contrast injection parameter 4

variation, and limited data that compare this technique with other tests of cerebral blood flow 5

currently relegate CT angiography as only investigational as an ancillary test to assist with 6

BD/DNC determination (EVID).63, 65 7

• Recommendation statement 32. Clinicians should not use CT angiography as an ancillary 8

test to aid in the diagnosis of BD/DNC (Level B). 9

10

33. Recommendation 33 rationale: Because of the availability of magnetic resonance angiography 11

(MRA), this technique has been used to assist with BD/DNC diagnosis (EVID).66, 67 It can be 12

challenging to perform in critically ill patients, and the available data do not demonstrate 13

adequate validation (EVID).66-68 14

• Recommendation statement 33. Clinicians should not use MRI or MRA as an ancillary test 15

to aid in the diagnosis of BD/DNC (Level A). 16

17

Special considerations 18

There are a number of special considerations not included in the prior adult and pediatric guidelines for 19

BD/DNC, which we address here.4, 5, 11, 28 These include 1) the need for consent before evaluation for 20

BD/DNC, 2) time of death and discontinuation of organ support, 3) evaluation of BD/DNC in a patient 21

who is pregnant, 4) neuroendocrine function, and 5) evaluation of BD/DNC in a patient with primary 22

posterior fossa injury. 23


43

1

Obtaining consent for BD/DNC evaluation 2

In recent years, the question of whether consent is needed before evaluation for BD/DNC has been 3

debated in medical, legal, ethical, and philosophical forums.69-77 According to the UDDA, which was 4

written in 1981 by the President’s Commission for the Study of Ethical Problems in Medicine and 5

Biomedical and Behavioral Research, the American Bar Association, and the American Medical 6

Association, death can be declared if there is irreversible cessation of cardiopulmonary function or 7

irreversible cessation of whole brain function.3 The need for consent before evaluation for either 8

BD/DNC or death by cardiopulmonary criteria is not addressed in the UDDA. However, some states 9

(New York and Nevada) legally stipulate that consent is not required before evaluation for BD/DNC.78, 10

79 When this matter has been adjudicated in court, some courts determined that consent is not required 11

before evaluation for BD/DNC or discontinuation of organ support,77, 80 while others ruled that a 12

person’s surrogate has the right to choose or refuse whether or not any medical procedure, including 13

evaluation for BD/DNC, is performed.76, 81 14

15

A 2015 survey of adult neurologists in the AAN found that 78% of respondents believed physicians 16

should not need to obtain consent before an evaluation for BD/DNC.82 Similarly, a 2016 survey of 17

pediatric neurologists and intensivists found that 72% believed physicians should not need to obtain 18

consent before an evaluation for BD/DNC.83 Accordingly, in a 2019 position statement, the AAN noted 19

that clinicians should inform a patient’s surrogates about the intent to perform an evaluation for 20

BD/DNC, but that there was no obligation to obtain informed consent before performing the 21

evaluation.84 22

23


44

34. Recommendation 34 rationale. Death by BD/DNC criteria is equivalent medicolegally to death 1

by cardiopulmonary criteria (PRIN). Evaluating a patient for death is part of the patient’s general 2

medical care. Consent is not required to evaluate a patient for death (INFER). The decision to 3

initiate an evaluation for BD/DNC after catastrophic, irreversible brain injury is a clinical 4

decision made by the attending physician. 5

• Recommendation statement 34. Clinicians do not need to obtain consent before an 6

evaluation for BD/DNC (Level B). 7

8

35. Recommendation 35 rationale. Loss of a family member is a traumatic event for families 9

(PRIN). Being informed of the BD/DNC process before its initiation and having the opportunity 10

to witness the BD/DNC evaluation may aid in acceptance of BD/DNC (INFER). 11

• Recommendation statement 35a. Clinicians planning to evaluate a patient for BD/DNC 12

(should) make a reasonable attempt to inform the patient’s family of the plan to perform a 13

BD/DNC examination. 14

• Recommendation statement 35b. Clinicians evaluating a patient for BD/DNC (may) 15

provide the option for the family to observe the clinical evaluation, including apnea 16

testing. Clinicians (should) inform families that patients may have reflexive movements 17

caused by activity from the spinal cord, muscles, or nerves during the BD/DNC 18

evaluation and that these movement do not preclude declaration of BD/DNC. 19

20

Time of death 21


45

36. Recommendation 36 rationale: Death certificates require that the time of death be recorded 1

(PRIN). The process of assigning the time of death for patients who meet criteria for BD/DNC 2

should be consistent (PRIN). 3

• Recommendation statement 36a. For patients who meet clinical criteria for BD/DNC, 4

clinicians (must) assign the time of death as the time during the apnea test that the ABG 5

results are reported and demonstrate that the PaCO2 and pH levels are consistent with 6

BD/DNC criteria. 7

• Recommendation statement 36b. For patients in whom an ancillary test is required and 8

performed, clinicians (must) assign the time of death as the time an attending physician 9

documents in the medical record that the test results are consistent with BD/DNC. 10

• Recommendation statement 36c. The time of death for pregnant women who meet criteria 11

for BD/DNC should be assigned as the time BD/DNC is declared as specified in 12

recommendation statement 38 (Level B). 13

14

37. Recommendation 37 rationale. Once BD/DNC has been declared, the mechanical or 15

pharmacologic support that is maintaining respiratory or cardiac function should not be 16

continued unless organ donation is being considered (PRIN). It is reasonable for this organ 17

support to be continued for a period of time, providing the family with a reasonable but limited 18

amount of time with the deceased patient (PRIN). Some U.S. states legally require 19

accommodation of requests to continue organ support after BD/DNC determination under 20

specifically defined circumstances (PRIN). 21

• Recommendation statement 37a. Organ support (may) be continued for a period of time 22

after BD/DNC, the length of which is based on the judgement of the attending physician 23


46

of record, to provide the family with a reasonable but limited amount of time with the 1

deceased patient before the discontinuation of this support. 2

• Recommendation statement 37b. Following the declaration of BD/DNC in a pregnant 3

woman, the clinicians providing her care, assisted by clinicians knowledgeable in 4

maternal-fetal medicine, child neurology, and neonatology, as needed, should educate 5

and discuss with lawful surrogates the risks and benefits to the fetus of continuing 6

maternal organ support (Level B). 7

8

Evaluation of BD/DNC in a patient who is pregnant 9

Women who are pregnant can develop catastrophic, irreversible brain injuries requiring declaration of 10

BD/DNC.85-87 It is feasible to continue organ support in a pregnant woman after BD/DNC with the goal 11

of delivery of a viable fetus. This has been attempted in women with a fetus of gestational age ranging 12

from 1 to 40 weeks, but aggressive medical management was needed to address the multiorgan 13

dysfunction which developed before and after BD/DNC, and there is a risk of cardiopulmonary arrest at 14

any time after BD/DNC; thus, while some cases of continuation of organ support after BD/DNC in a 15

pregnant woman resulted in delivery of a developmentally normal infant, others resulted in spontaneous 16

abortion, intrauterine death, or congenital defects.85, 87 In a 2019 position statement, the AAN noted that 17

if a pregnant woman is found to be BD/DNC, the family should be “educated by knowledgeable 18

professionals about relevant law as well as fetal outcome, which is often uncertain,” and that decision-19

making regarding continuation of organ support versus discontinuation of support should be based on 20

consideration of the viability of the fetus, fetal brain injury, the law, the patient’s surrogate, and prior 21

wishes of the patient regarding care of her unborn child.84 22


47

38. Recommendation 38 rationale. Women who are pregnant can develop catastrophic, irreversible 1

brain injuries and may be declared BD/DNC (PRIN). In these situations, the fetus may still be 2

viable (PRIN). Continued organ support in a pregnant woman after BD/DNC declaration may 3

lead to the delivery of a viable fetus (EVID).85, 87 The ethical analysis of whether to continue 4

organ support in a pregnant woman declared BD/DNC should largely focus on the welfare of the 5

fetus.84 6

• Recommendation statement 38. Clinicians should assess and diagnose pregnant women 7

with catastrophic, irreversible brain injuries for BD/DNC when clinically appropriate 8

(Level B). 9

10

Neuroendocrine function 11

The hypothalamic-hypophyseal axis may remain intact after BD/DNC, in part because of variable 12

extracranial vascular supply, reducing the effect of ischemia.88, 89 Thus, diabetes insipidus only develops 13

in BD/DNC in 9˗90% in adults and 38˗41% in children.90-95 In a 2019 position statement, the AAN 14

noted that neuroendocrine function can persist in patients with irreversible injury to the brain and 15

brainstem.84 16

39. Recommendation 39 rationale. The concept of BD/DNC has its legal origin in the United States 17

in the UDDA, which in turn is rooted in the Harvard Criteria (EVID).3 The UDDA states, “an 18

individual who has sustained… irreversible cessation of all functions of the entire brain, 19

including the brainstem, is dead. The determination of brain death must be made in accordance 20

with accepted medical standards.” Since the publication of the Harvard criteria, these accepted 21

medical standards have been anchored in “unreceptivity and unresponsiveness, absent breathing 22

or movements,” and absent brainstem function—standards reaffirmed in the 1995 and 2010 23


48

AAN brain death Guidelines as well as the 1987 and 2011 Pediatric Brain Death Guidelines from 1

the AAP, SCCM, and Child Neurology Society (EVID). Neuroendocrine function may persist in 2

patients with catastrophic, irreversible brain injury (EVID),90-95 in part because of variable 3

extracranial contribution to the vascular supply of the hypothalamus and pituitary, reducing the 4

effect of ischemia (EVID).88, 89 5

• Recommendation statement 39. Clinicians may initiate a BD/DNC evaluation and declare 6

a patient BD/DNC despite evidence of neuroendocrine function (Level B). 7

8

Primary posterior fossa pathology 9

Irreversible loss of function of the entire brain, including the brainstem, usually begins with 10

supratentorial injury followed by herniation leading to infratentorial injury. In rare cases, the primary 11

injury takes place in the posterior fossa, and edema, obstructive hydrocephalus, and upwards herniation 12

subsequently lead to injury of the supratentorial structures.96 13

14

40. Recommendation 40 rationale. Patients with primary posterior fossa pathology may be clinically 15

comatose with brainstem areflexia and apnea; however, they may retain some cortical function 16

(EVID).96 17

• Recommendation statement 40. To avoid declaring BD/DNC in patients with primary 18

posterior fossa pathology and retained supratentorial function, clinicians (should) ensure 19

that the posterior fossa process has also led to catastrophic supratentorial injury as 20

demonstrated on a conventional neuroimaging study before initiating the BD/DNC 21

evaluation. 22

23


49

SUGGESTIONS FOR FUTURE RESEARCH 1

There are several topics related to BD/DNC that warrant further research, including but not limited to: 2

1. Importance of number of examinations and qualifications of examiners 3

2. Appropriate observation periods to ensure irreversibility 4

3. Further development and validation of ancillary testing 5

6

DISCLAIMER 7

Documents

Consensus practice recommendations: Pediatric and adult brain … · 2020. 12. 8. · 16 age and sought to combine adult and pediatric guidance recommendations, updated with current