Clinical manifestations

Presentation and course

BD/DNC should be suspected in any patient rendered deeply comatose and apneic from a profound diffuse brain insult. The cardinal diagnostic criteria for BD/DNC are: (1) the presence of a structural brain lesion sufficient to produce the clinical findings, (2) permanence by excluding potentially reversible metabolic and toxic factors, (3) profound coma with total unresponsiveness, (4) absence of all reflexes served by the brainstem and cranial nerves, and (5) apnea despite induced hypercapnia and acidosis (78; 131; 132; 16; 137; 85; 86; 41; 02). The authors recommend that BD/DNC determinations be performed in an intensive care unit setting where resources and subsequent discussions and care are most appropriate.

Presence of a structural brain lesion. Neuroimaging evidence of a catastrophic brain lesion (meaning one that will invariably cause loss of all cerebral blood flow and cessation of all brain function) is strongly recommended before initiating BD/DNC testing. If no adequate structural diffuse brain lesion can be identified, the clinician is obligated to consider alternate diagnoses and not declare BD/DNC (02).

Permanence. The demonstration of coma, apnea, and cranial nerve areflexia shows the absence of clinical brain functions. To show that their absence is permanent, the clinician must demonstrate that the lesion responsible for the patient's neurologic condition is structural and severe enough to cause cerebral circulatory arrest. Demonstration by CT or MR that the lesion is structural is easily accomplished in cases of massive head trauma or massive stroke. However, in cases of diffuse hypoxic-ischemic neuronal damage suffered during cardiopulmonary arrest, early CT or MR imaging may be normal or nondiagnostic, and delayed imaging may be required. A diffuse hypoxic-ischemic insult can be shown to be structural and permanent only if the patient satisfies BD/DNC criteria after an appropriate time interval and if all contributions from potentially reversible metabolic and toxic disorders can be excluded. Although BD/DNC declaration should be done expeditiously once a clinician suspects the diagnosis, there is potential for recovery of pupillary light reflexes in up to 30% of cardiac arrest patients, especially those treated with therapeutic hypothermia (22). As such, the guidelines recommend that in cardiac arrest patients who undergo therapeutic hypothermia, BD/DNC testing should be delayed for at least 24 hours post-rewarming, longer if there is concern for drug metabolism or other confounders. In overdose patients, consultation with a medical toxicologist or clinical pharmacist may be very helpful. If imaging findings already show diffuse cerebral edema with tonsillar herniation, the diagnosis of BD/DNC is probable, and delays in declaration may not be necessary.

A study questioned whether a second BD/DNC examination is necessary. In a review of 1229 adult and 82 pediatric BD/DNC determinations, Lustbader and colleagues showed that no patients with clinical findings of BD/DNC on the first examination regained any brain functions by the second examination (73). They also found, as expected, a negative impact on ICU bed utilization and a reduction in organ donation as a consequence of waiting for the second examination, which occurred after a mean interval of 19.9 hours. They concluded that in patients older than 1 year of age, a single examination was sufficient to determine BD/DNC. A second retrospective study conducted over 39 months by one of the investigators confirmed these findings in a separate patient population (125). A study of BD/DNC determinations in Canadian pediatric ICUs showed less than 1% discordance between the two examinations (1 in 110) (55).

The impact of these studies on clinical practice guidelines has been variable. The 2010 American Academy of Neurology BD/DNC guideline for adults no longer requires the second examination (137), but the 2011 multi-society guideline for diagnosing BD/DNC in infants and children continues to require two sequential examinations (85; 86). The 2023 AAN guidelines require one examination in adults but strongly encourage a second examination; only one apnea test is required in adults. In contrast, children (defined as 37 weeks gestational age to 18 years) require two examinations and two apnea tests, separated by 12 hours. Many state laws and hospital policies continue to mandate two serial examinations, sometimes requiring that they be performed by different physicians. Physicians declaring BD/DNC should know the requirements of their hospital policies and state laws.

A prerequisite for BD/DNC determination to exclude reversible causes is that significant levels of CNS-depressing drugs and neuromuscular blocking agents must be absent and that the patient is not severely hypothermic (temperature is less than 36°C). The patient must also be normotensive; per the new 2023 AAN guidelines, this means both a systolic blood pressure of at least 100 mm Hg and a mean arterial pressure of 75 mm Hg. These pharmaceutical agents and other confounders, as well as hypothermia, can induce a set of clinical findings similar to those seen in BD/DNC yet be potentially reversible with appropriate treatment. Gray and colleagues showed that usual therapeutic doses of neuromuscular blocking agents do not affect the integrity of pupillary light reflexes (40). No severe electrolyte, acid-base balance, or neuroendocrine disorder should be present that may be a reversible yet contributory factor to depressed CNS function. The American College of Medical Toxicology published a position statement on the determination of BD/DNC in the setting of drug overdose (88). They recommended that (1) drugs and toxins be identified by history and targeted testing; (2) at least five drug half-lives be considered an absolute minimum to assure drug clearance; and (3) a clinical or medical toxicologist be consulted in unclear situations. In patients with organ failure or those who have been subjected to hypothermia, five half-lives may not be sufficient for drug clearance, and the authors recommend either direct toxicology or ancillary testing.

The interfering role of therapeutic hypothermia in BD/DNC determination was dramatically reported in the case of a 55-year-old man who underwent therapeutic hypothermia after suffering cardiac arrest and was subsequently diagnosed as BD/DNC using accepted guidelines (130). However, when he arrived in the operating room for organ donation 24 hours later, he was observed to regain corneal reflexes, cough reflexes, and spontaneous respirations. This well-documented case of “reversible BD/DNC” shows that therapeutic hypothermia can induce a prolonged but potentially reversible metabolic encephalopathy that may violate the irreversibility condition for making a diagnosis of BD/DNC. One possible explanation is that the metabolism of depressant drugs administered before and during therapeutic hypothermia may be slowed by the hypothermia, so these drug effects are more prolonged than expected. Physicians caring for cardiac arrest patients who have undergone therapeutic hypothermia should delay BD/DNC determination for at least 24 hours after rewarming to 36°C until the reversible metabolic encephalopathy caused by hypothermia has abated. In the United States, primary brainstem injuries provide another diagnostic dilemma in the declaration of BD/DNC. Some patients who have suffered a direct injury to their brainstem will satisfy the examination criteria for BD/DNC, but they would not satisfy the “whole-brain” criteria as cortical perfusion and electrical activity may be spared initially. In these patients, it is now required in the 2023 AAN guidelines that there also be neuroimaging evidence of catastrophic supratentorial injury prior to proceeding with BD/DNC determination. Furthermore, in patients who have been hypothermic <35.5°C, either iatrogenically or due to environmental exposure, their temperature must be maintained at 36°C or greater for 24 hours before clinical testing can ensue.

Coma and unresponsiveness. Lack of a cerebrally-mediated response to external stimuli is required to declare BD/DNC. The patient should have no movements--no withdrawal or posturing (flexor or extensor), spontaneously or to noxious stimulation (including sternal rub, nailbed pressure, and noxious cranial stimuli, such as supra-orbital pressure). Because deep-tendon reflexes are integrated at a purely spinal cord level, their presence or absence is not directly relevant to brain functioning.

Spinally-mediated reflexes and automatisms are allowed in the diagnosis of BD/DNC. If reflexes are stimulus-induced, automatisms may occur spontaneously. These may include undulating toes, facial myokymia, multifocal spinal myoclonus, hugging-like movements, eyelid opening, or even head turning (51). A dramatic reflex known as the Lazarus sign occurs rarely in BD/DNC patients, usually during apnea testing. The patient may slowly elevate both arms and progressively adduct them across his or her chest. This bizarre sign is believed to result from involuntary motor discharges arising from previously intact cervical spinal cord anterior horn cells rendered progressively hypoxic during apnea testing (123). To prevent the misleading impression of intact brain function, family members should be informed of the presence of these movements and of their spinal origin to clarify that they are neither voluntary nor represent evidence of residual brain function.

Brainstem areflexia. Areflexia of those reflexes subserved by the cranial nerves must be present for BD/DNC to be determined. There can be no pupillary light or dark reflexes or oculovestibular, oculocephalic, corneal, gag, or cough reflexes. The absence of these reflexes must be ascertained accurately.

Pupillary light reflexes should be tested with a strong point-of-light source in a darkened room. A magnifying glass should be used in order to detect small pupillary changes, and a pupillometer may also be used as an adjunct measure, although this has not yet been validated for use in BD/DNC. There must be neither constriction to light nor dilatation to dark. Pupils are usually midposition (4 to 9 mm) because of sympathetic and parasympathetic denervation. Smaller pupil size (less than 2 mm) suggests a possible underlying drug effect, and potential confounding must be determined. Pupillary reflexes are unaffected by atropine given intravenously during resuscitation (36) and are also unaffected by therapeutic neuromuscular blockade during anesthesia (40). Midposition pupils of BD/DNC patients usually dilate further following circulatory arrest (64).

Corneal reflexes should be tested with a cotton-tipped applicator applied firmly to the cornea; reflexes must be absent. A sterile saline or water “squirt” may be used as a screening tool, but use of a cotton-tipped applicator with light pressure at the border of the iris is recommended to be definitive (75). Gag and cough reflexes can be assessed by suctioning the patient's endotracheal tube and stimulating the posterior pharynx bilaterally using a tongue depressor. Coughing or gagging resulting from pharyngeal or tracheal stimulation must be absent in BD/DNC.

Oculovestibular reflexes should be tested using the technique of "ice water caloric stimulation." An otoscopic examination is performed to ensure that the tympanic membranes are unruptured and that the external auditory canals permit passage of the ice water to the tympanic membranes. The patient's head is elevated to 30 degrees above horizontal. Through a flexible, dull-ended tube such as a butterfly plastic tube with the needle removed, 50 mL of ice water is infused over one minute into each external auditory canal as a second person holds open the patient's eyes. The BD/DNC patient should demonstrate no response to this test, making neither eye movements, limb movements, nor grimace (46). There should be an intervening time of at least 5 minutes before testing the opposite ear to allow the temperature of the endolymph to reequilibrate in the tested ear, and both sides must be tested.

Oculocephalic reflexes should be assessed only if spine and skull base integrity is ensured. Proper testing involves rapid movement of the head from side-to-side, and there should be no movement of the eyes in BD/DNC. If the oculocephalic reflex cannot be performed, this is the one part of BD/DNC clinical testing that can be omitted, provided that the oculovestibular test can still be fully performed.

Apnea. Apnea is a cardinal sign of BD/DNC, but many clinicians test for it incorrectly (26; 79). Lack of spontaneous respirations should be performed by disconnecting the ventilator. Merely turning off the ventilator for a few minutes and observing the absence of respiratory effort is an inadequate test for apnea because in all patients acidemia, not hypoxemia, is the primary stimulus to breathe. The PACO2 increases only approximately 3 mm Hg per minute of apnea; thus, observing a few minutes of apnea is an inadequate test (109). The technique, problems, and safety of apnea testing in BD/DNC have been reviewed (63; 111). It is worth noting that the 2023 AAN guidelines require both a PACO2 goal and a pH goal (< 7.30).

A proper apnea test must permit physicians to observe for respiratory effort in the setting of an elevated PACO2 while protecting the PAO2 from falling to dangerously low levels. Of the several strategies devised to achieve these goals, the most commonly used is the technique of apneic oxygenation. First, the ventilator settings are adjusted to bring the PACO2 to 35 to 45 mm Hg (assuming a baseline premorbid normocarbic patient) and a normal pH (7.35 to 7.45), and the patient is preoxygenated with an FiO2 = 1.0 (100% O2) to a PaO2 of greater than 200 mm Hg. At this point, passive oxygenation can be accomplished by placing the patient on blow-by or by inserting a cannula down the endotracheal tube to the level of the carina and infusing 100% O2 at 4 to 6 L per minute; higher flow rates may wash out CO2, making the determination uncertain. By starting with a markedly elevated PAO2 and passively oxygenating, despite 8 to 10 minutes of apnea, the PAO2 usually does not fall to dangerously low levels (03). The goal PACO2 is greater than 60 mm Hg and 20 above the baseline value, as well as pH less than 7.30. A study of BD/DNC determination from the Mayo Clinic confirmed that apnea testing is safe when performed using apneic oxygenation (136). In approximately 10% of BD/DNC determinations, apnea testing could not be performed, and ancillary testing was required to establish the diagnosis. The most common reason that apnea testing cannot be performed is when acute lung injury impairs the A-a gradient, blocking the rise of PAO2 to levels safe enough to conduct the examination (140).

The duration of apnea necessary to raise the PACO2 to greater than 60 mm Hg can be estimated by knowing the PACO2 when the apnea test was begun and using the rate of increase of PACO2 of approximately 3 mm Hg per minute of apnea. Generally, observing the patient during 7 minutes of apnea is sufficient to raise the PACO2 to greater than 60 mm Hg if the PACO2 was 40 mm Hg at the beginning of the test. Any respiratory effort, including sighing, counts as respiration and negates the apnea test. Major complications of the apnea test are hypotension, acidosis, cardiac arrhythmias, and hypoxemia. These complications are much more likely to occur in the setting of inadequate preoxygenation (53; 39). False negative apnea testing in BD/DNC may result from ventilator auto-triggering, also known as auto-cycling (23; 77). Respiratory therapy personnel assisting with apnea testing should be trained to recognize this phenomenon and not interpret it as indicative of brainstem-generated breathing.

In a patient with chronic hypercarbia in which their baseline PACO2 is known, that is the baseline value for that patient; they must achieve 20 mm Hg above that value during the test in addition to the pH < 7.30. In a patient with suspected chronic hypercarbia in which the baseline PACO2 is NOT known, it should be estimated, at the PACO2 values greater than 60 mm Hg and greater than20 mm Hg above the baseline, the pH <7.0, AND an ancillary test is required.

In the setting of ECMO, the 2023 AAN guidelines provide the following guidance:

Prognosis and complications

No patient declared BD/DNC by properly performed and interpreted tests has ever recovered. Generally, BD/DNC patients develop generalized vasodilatation because of destroyed medullary blood pressure centers, resulting in the development of severe hypotension requiring intravenous vasopressors and, often, high-output congestive heart failure (35). Usually, they also develop diabetes insipidus because of destruction of the hypothalamic neurons secreting antidiuretic hormone (92). Some BD/DNC patients do not develop diabetes insipidus, presumably because the independent circulation to the posterior hypothalamus may be preserved (84). Recently, the presence of renal dysfunction has been noted to impact the development of diabetes insipidus as well (124). Irrespective of treatment, most patients become asystolic within several days or weeks, but the heartbeat and circulation of a few young patients have remained functioning for months to even years. The major risks to transplantable organs in the BD/DNC donor that may require immediate treatment are pulmonary edema, hypotension, and polyuria (131). The effect of BD/DNC on systemic physiology has been reviewed (100).

Clinical vignette

A previously healthy 28-year-old man suffered a sudden, severe headache, vomited, and collapsed unconscious in a restaurant. He was taken to a hospital emergency room within 15 minutes, where he was intubated and ventilated. Examination revealed deep coma with no movements, no breathing above the set rate on the ventilator, absent brainstem reflexes, and weak bilateral decerebrate posturing to noxious stimuli. CT scan disclosed a massive intracranial hemorrhage originating in the left thalamus or basal ganglia with intraventricular rupture and probable bilateral uncal herniation.

He was treated with hyperventilation, intravenous mannitol, and vasopressin. There was no improvement on serial examinations at 4 hours and 12 hours, and the decerebrate posturing ceased. The neurologic consultant suggested incipient BD/DNC, and the organ procurement organization was notified for the possibility of organ donation. A formal BD/DNC evaluation was performed at 16 hours.

This examination disclosed unresponsiveness to all noxious stimuli. Ice water caloric irrigation of 50 cc in each external auditory canal was without oculovestibular reflex or other response. Pupillary, corneal, oculocephalic, gag, and cough reflexes were absent. Apnea testing disclosed no breathing effort after 10 minutes off the ventilator, with a PACO2 of 64 mm Hg and a pH of 7.27. Death was declared.

Autopsy restricted to the brain disclosed massive hemorrhage, bilateral uncal herniation, early autolysis, and remnants of an arteriovenous malformation.

Biological basis

Etiology and pathogenesis

In adults, BD/DNC is caused most commonly by massive head trauma, hypoxic-ischemic neuronal damage from prolonged cardiopulmonary arrest, and massive intracranial hemorrhage. In children, BD/DNC is produced most commonly by traumatic brain injury, bacterial meningitis, asphyxia, and drowning (121). In the largest reported study of 1844 BD/DNC patients in Spanish intensive care units, 42% had intracerebral hemorrhage, 19% had traumatic brain injury, and 14% had subarachnoid hemorrhage (27).

The pathogenesis of BD/DNC usually features three sequential phases. The first phase encompasses the primary brain insult, such as traumatic brain injury or diffuse neuronal injury from cardiopulmonary arrest. The edema produced by the primary insult raises intracranial pressure to a level greater than that of mean arterial blood pressure. At this point, intracranial blood flow ceases, and cerebral herniation occurs. During this second phase, the remainder of cerebral hemispheric, cerebellar, and particularly brainstem neurons not destroyed during the primary phase are killed as the result of complete cessation of intracranial blood flow. Thereafter, intracranial pressure falls, and during the third phase, there may be reperfusion of the infarcted brain (110).

In studies of the neuropathology of BD/DNC in the 1970s, the brain was autolyzed at autopsy. The severity of gross autolysis, however, is proportional to the duration of continued systemic normothermic circulation after BD/DNC has occurred (127). Therefore, relatively little gross evidence of autolysis may be present if the BD/DNC determination has been rapid. A study of the pathology of BD/DNC confirmed that, in the contemporary era, the autolysis seen in studies from the 1970s usually no longer is present because the patients are declared dead within a day or two of the global brain injury and circulation ceases before the gross pathological changes have had time to occur (135).

The widespread physiological and neuroendocrine abnormalities produced by BD/DNC have been studied in patients undergoing multiorgan procurement (98; 19).

Epidemiology

A large study in the United Kingdom found that 10% of all patients dying in hospital intensive care units were declared BD/DNC (38). This percentage is probably higher than in most American intensive care units. One Canadian study found the percentage of traumatic brain injury patients who progressed to BD/DNC decreased over several years, likely owing to improved medical care (61). Staworn and colleagues found that BD/DNC was the discharge diagnosis for 0.9% of infants admitted to a busy pediatric intensive care unit (121).

Prevention

BD/DNC is the end result of any injury that will inevitably result in elevated intracranial pressure so severe as to lead to the total loss of cerebral perfusion. Although once a brain injury becomes catastrophic, the secondary mechanisms leading to herniation can sometimes be avoided with aggressive medical or surgical care; however, this is not commonly undertaken when it is clear that the prognosis is poor and aggressive care would be futile.

Differential diagnosis

BD/DNC should be suspected when a patient has been rendered comatose and apneic and the brainstem rendered areflexic by a permanent, diffuse, structural brain insult. If the patient satisfies the BD/DNC criteria previously listed, it is unlikely that any other diagnosis is responsible for the patient’s examination findings; the patient is BD/DNC and is declared dead. Other differential diagnoses of coma and brainstem deficits should be considered if patients do not satisfy all of the previously described criteria. (97).

There are several pitfalls in the diagnosis that the experienced examiner should carefully avoid (14). The examination findings of BD/DNC can be produced by drug intoxication or profound neuromuscular blockade, such as from delayed vecuronium clearance. Therefore, the absence of these drugs in concentrations sufficient to produce physical signs is a prerequisite for BD/DNC determination (58; 91). Occasional cases of severe Guillain-Barré syndrome or botulism causing profound paralysis may mimic many of the findings in BD/DNC (49; 34; 33).

Pupillary light reflexes, oculovestibular reflexes, and gag reflexes may be absent because of preexisting disease. Pupillary light reflexes rarely can be affected by ingested or prescribed medications, such as atropine. Oculovestibular reflexes can be abolished by aminoglycoside antibiotics. Patients with CO2-retaining chronic obstructive pulmonary disease may have false-positive apnea tests because they may breathe based more on their hypoxemic drive and not on their hypercapnic drive. Ventilator auto-cycling despite true apnea can falsely suggest an intact respiratory drive. This phenomenon, if present, should be evaluated by experienced respiratory therapists and anesthesiologists (134; 23; 77).

The vegetative state should be distinguished from BD/DNC. Patients in the vegetative state have eyes-open unawareness resulting from profound dysfunction of the cerebral hemispheres, yet these patients have retained sleep-wake cycles and variably intact brainstem functions, including breathing and cranial nerve reflexes (82; 52; 06).

Diagnostic workup

Algorithms to guide the less-experienced clinician in the determination of BD/DNC are available (41; 02; 68). Only experienced clinicians should perform BD/DNC determinations; trainees and advanced practice practitioners may perform BD/DNC testing only under the direct supervision of an attending physician unless otherwise stipulated by hospital policy. Non-neurologists, including intensivists and neurosurgeons, should also undergo proper training, which may include simulation courses (74) or online courses, such as those provided through the Neurocritical Care Society.

Several centers have produced simulation training programs to teach accurate BD/DNC determinations (74; 47). One advantage of simulation training is learning to overcome confounders and pitfalls in BD/DNC determination and to learn the best ways to explain BD/DNC to family members.

It is essential for the clinician not only to fully and carefully examine the patient but also to carefully record the clinical findings that support the diagnosis of BD/DNC. One study of 121 BD/DNC chart notes disclosed the disturbing findings that physicians recorded the results of tests of pupillary reflexes in only 86%, gag reflexes in 78%, and corneal reflexes in 57% of cases (129). Another study showed a surprisingly high variability rate of hospital policies for BD/DNC determination (101). A study of pediatric BD/DNC determination among hospitals in southern California showed a disturbing pattern of variability and lack of compliance with accepted test batteries (76). The authors advocated requiring physicians to complete a checklist when determining BD/DNC. Completing a checklist ensures that no examination element is omitted and that the clinician has made adequate documentation of the absent signs to justify the declaration of death. Both the 2020 World Brain Death Project and the 2023 AAN guidelines provided useful checklists (41; 02; 68). The role of laboratory and radiological evaluations is to confirm the bedside diagnosis of BD/DNC and to investigate the state of brain structure and function. Ancillary tests may be useful to supplement the clinical examination in three situations: (1) when an adequate clinical examination or apnea test cannot be performed, such as in states of trauma when the eyes, tympanic membranes, or lungs have been damaged; (2) in medicolegal circumstances in which the clinician wishes to have "objective" evidence to submit in addition to his or her own bedside examination findings; and (3) when concurrent metabolic or toxic factors may be affecting the clinical findings and cannot be corrected. In the last case, it is important to note that the entirety of the clinical examination, including the apnea test, still needs to be performed, and if signs of brain function are present, the patient is clearly not BD/DNC, and ancillary testing is not indicated. The ancillary test chosen will depend primarily on which test is available and the stability of the patient. Both the World Brain Death Project and the 2023 AAN guidelines downgraded electroencephalography (EEG), as it does not measure brainstem function or integrity. The authors recommend that a detailed clinical evaluation of BD/DNC criteria be performed and documented before starting ancillary tests, as these should only be undertaken if there is sufficient clinical evidence. Ancillary tests of BD/DNC now include only tests of cerebral blood flow. A survey of neurointensivists showed that radionuclide blood flow studies are used most frequently, and younger physicians are more likely to use transcranial Doppler ultrasound (11). Note that there remains variability among countries as to which ancillary tests are acceptable, and the reader is encouraged to investigate these accordingly.

Cerebral blood flow ancillary tests showing the absence of intracranial blood flow include catheter-based cerebral angiography (37; 32; 24; 143), single-photon emission computed tomography (SPECT) perfusion imaging with 99m technetium-hexamethylpropylene amineoxime (HMPAO) (105; 138; 141; 89; 83), and transcranial Doppler ultrasonography (99). Neither CT angiography nor MRI angiography has been validated for use as an ancillary test in BD/DNC, and both the World Brain Death Project and the 2023 AAN guidelines recommend against their use. Each of these tests exploits the fact that all intracranial blood flow ceases at some time during the process of BD/DNC.

Transcranial Doppler ultrasonography is the simplest of the intracranial blood flow tests and demonstrates reliable findings in BD/DNC. It is especially useful for patients deemed too unstable to transport out of the intensive care unit. A transcranial Doppler ultrasound test confirms BD/DNC if one of the following patterns is present: sharp systolic peaks when intracranial pressure exceeds systolic blood pressure and "reverberating" blood flow with reversal of flow or zero values during diastole when intracranial pressure exceeds diastolic blood pressure but is lower than systolic blood pressure (94; 25). Proper testing requires insonation both anteriorly and posteriorly and bilaterally, with two examinations separated by at least 30 minutes. Cranial insonation may be performed transorbitally with a high success rate (62). A meta-analysis of 22 studies found that transcranial Doppler ultrasound is a highly accurate ancillary test for BD/DNC with a pooled sensitivity of 90% and a pooled specificity of 98% (18).

A few studies exist comparing the relative reliability of the tests confirming BD/DNC. In a 2008 review, Heran and colleagues advocated for HMPAO-SPECT as the ancillary test with the highest positive and negative predictive values for BD/DNC (45). Two studies showed detectable blood flow imaged by CTA when the patient was clinically BD/DNC, and the EEG was flat (104; 60). Another case report showed absent blood flow by CTA when intracranial flow was detected by transcranial Doppler (42).

Some data on physicians’ practice preferences regarding ancillary testing are available. Radionuclide angiography is ordered most commonly in practice, but younger physicians prefer transcranial Doppler ultrasound (11). There is evidence that neurosurgeons and critical care physicians performing BD/DNC determinations rely more on intracranial blood flow tests, and neurologists rely more on clinical tests (87).

Management

BD/DNC patients should be declared dead and not treated with the goal of trying to alter the patient’s outcome. Exceptions include temporary treatment for prospective organ donors prior to procurement (21), patients whose religions preclude BD/DNC determinations (15; 107), and some pregnant women in whom the decision has been made to maintain oxygenation and circulation temporarily to permit the subsequent birth of the infant (28). Shaner and colleagues have published practical guidelines from the California Medical Association Ethics Committee for optimally implementing BD/DNC examinations and requests for organ procurement in clinical practice (112). The Canadian Medical Association published similar guidelines (114). However, several complications associated with herniation may require medical management either before the declaration of BD/DNC or in preparation for organ procurement in the BD/DNC donor.

Hemodynamic instability. Blood pressure and heart rate variability are commonly encountered in BD/DNC due to autonomic dysregulation from injury to sympathetic pathways. Patients frequently require vasopressors to maintain a systolic blood pressure above 100 mmHg and a mean arterial pressure above 75 mm Hg, which is a prerequisite for the determination of BD/DNC. Vasopressors frequently used include norepinephrine, dopamine, phenylephrine, and epinephrine (59). Tachyarrhythmias may also be seen, and treatment should be aimed at correction of the precipitating factor (139). The authors recommend short-acting therapies to avoid reflex bradycardia. Bradyarrhythmias occurring in the setting of herniation and BD/DNC tend to be refractory to atropine and may require epinephrine (139).

Endocrine dysfunction. Pituitary dysfunction may lead to endocrine derangements such as diabetes insipidus. In BD/DNC patients who develop significant polyuria, the subsequent hypovolemia may worsen hypotension. In patients who meet the criteria for diabetes insipidus (elevated urine output, hypernatremia with sodium greater than 145mEq, decreased urine osmolality with specific gravity less than 1.005), appropriate fluid repletion and hormonal replacement are required. We recommend desmopressin (ddAVP) intravenously or vasopressin continuous infusion if urine output exceeds 300 mL/hr. The patient’s urine output should also be replaced with equal volumes of a hypotonic solution, such as half normal saline. Serum sodium and urine-specific gravity should be checked frequently to ensure titration of fluids and vasopressin are adequate.

Legal considerations. Laws and regulations in the United States require physicians to offer organ and tissue donation to families of patients declared dead. The benefits to society of increasing the organ pool are obvious, especially given the growing disproportion between the number of patients who need organs and the inadequate donor organ pool. It is clear that families of organ donors benefit as a result of the meaning conferred by the gift of the organ from the dead patient that maintains the life or health of the organ recipient; however, how the organ donation consent discussion with families should be optimally conducted remains controversial, particularly the role of the organ procurement organization staff (106). It is imperative that the family not be given the impression that the opportunity for organ donation is driving the BD/DNC determination and that valid informed consent is obtained for organ donation (50).

All states in the United States have either enacted statutes of death or drafted administrative regulations supporting the use of BD/DNC as a legal determination of death (96). Similar laws exist in most other Western and industrialized nations (52). In cases where the validity of BD/DNC determination has become the subject of judicial review, all courts in the United States have upheld the practice of BD/DNC (12). California, New Jersey, and New York have “accommodation clauses” that families can incur in order to continue life support in BD/DNC patients. Physicians who practice in states where such religious exemptions have been enacted should familiarize themselves with these laws (95).

Physicians should be sensitive to the emotional needs of family members, particularly in those cases in which family members request or demand continued treatment. These demands have been reported by nearly half of American neurologists in a survey (65). One reason commonly noted is the belief of family members that the BD/DNC loved one will recover if treatment is continued. This false belief may be furthered by misleading or incorrect depictions of BD/DNC patients in film and television and in the print media (20; 71). Physicians faced with such cases should consult the thoughtful summary of practical guidelines for their management by Flamm and colleagues (31). Techniques to enhance communication between physicians and family members can increase family member understanding of BD/DNC. When implemented, they improve the quality of end-of-life care. One technique that enhances family acceptance is to allow the family members to observe the BD/DNC determination (90). However, one trial of this technique in practice failed to improve family acceptance (57). A review of American hospital BD/DNC protocols showed that 80% provided no guidance on how to resolve disputes with family members over BD/DNC (70).

Special considerations

Pregnancy

Several cases have been reported of BD/DNC pregnant women whose heartbeat and circulation were maintained temporarily to permit the birth of a healthy fetus. These cases raise interesting ethical and legal problems of whether the continued therapy is justified, as well as challenging technical problems of maintaining physiological functions in the absence of clinical brain functions (30; 10; 56; 17; 120; 29; 28). One survey showed that BD/DNC in pregnancy was not provided for in hospital BD/DNC policies (69). The investigators found that 2.5% of hospital policies eliminated pregnant patients from BD/DNC consideration and that, in the remainder, which did permit BD/DNC determination, only 6.2% provided guidance in this situation and only 1% indicated who was responsible for making decisions for the fetus. The AAN provided a position statement regarding both pregnancy and accommodation in 2019 (108), and the 2023 AAN guidelines state, “Pregnancy in and of itself is not a contraindication to BD/DNC evaluation. Clinicians should assess and diagnose pregnant persons with catastrophic, permanent brain injuries for BD/DNC.” They further state that “After the determination of BD/DNC in a pregnant person, the clinicians providing care, assisted by clinicians knowledgeable in maternal-fetal medicine, child neurology, and neonatology, as needed, should educate and discuss with surrogate decision-makers the risks and benefits to the fetus of continuing maternal organ support.”

Sperm retrieval

One report described a BD/DNC man from whom viable sperm were retrieved by the technique of electro-ejaculation (122).

Anesthesia

General anesthesia is not necessary when BD/DNC patients undergo organ donation because there is no possibility of pain perception or suffering.