Fifty years after the 1968 publication of the report by the ad hoc committee of the Harvard medical school on a definition of irreversible coma,1 ∼70% of countries have a legal provision for brain death, predominantly those with transplantation programs. Publication of the Harvard report initiated a prospective study, a number of committee reports, and eventually settled with the American Academy of Neurology guideline for the determination of brain death in 1995, which was updated with minor modifications in 2010.2 The neurological determination of brain death involves a complex set of careful considerations and clinical tests before the diagnosis of death can be irrefutably made. While the concepts of death that underpin the diagnosis of brain death are settled, the main challenges for practicing clinicians are to reduce local, regional, and international variability in practice,3 and resolve the implications of brain death determination in patients with the so called “isolated” brainstem lesions, which can theoretically result in a patient being legitimately declared dead in one country but not in another (where protocols demand additional tests to demonstrate supratentorial hemispheric destruction). This situation has generated a controversy, which we would like to discount.
The biggest need in brain death determination remains education and maintenance of the skills of examination of a patient suspected of being brain dead. Details aside, the need for consistency in brain death determination is well recognized, and international efforts to achieve this continue to be made.3 However, variability remains, and much of this relates to the length of the observation period before testing (particularly after hypoxic brain injury), conduct of the apnea test, the number of times the tests are undertaken, exclusion confounders, a requirement for ancillary tests, and the degree of expertise of the physician undertaking the testing. Despite this variability, virtually all guidelines and protocols consistently recommend a 3-staged approach to brain death determination: establishing a known cause for the diagnosis; excluding reversible causes and confounders; and, finally, when these preconditions have been met, a clinical examination to confirm the absence of all brainstem function, including the capacity to breathe. This simple and trustworthy approach allows a foundation on which to achieve international consensus for the determination of brain death.
Brain death in patients with an isolated brainstem lesion remains controversial and the only outstanding issue in the “transatlantic divide” in brain death determination.4 The term “isolated brainstem lesion” here refers to an intrinsic destructive brainstem lesion, such as a massive pontine hemorrhage or severing of the pons with hanging, or an extrinsic lesion compressing and destroying the brainstem, such as a massive cerebellar hemorrhage. The term arose because of the historically and conceptually different formulations for brain death developed on either side of the Atlantic. The “whole brain” formulation in the United States was based on demonstrating irreversible loss of all functions of the brain including the brainstem, whereas the “brainstem” formulation in the United Kingdom required only demonstration of the irreversible loss of brainstem functions. While previous debate has focussed on which approach was the correct one, it is increasingly recognized that, while the 2 formulations are semantically different, they are clinically wholly synonymous.4 A strong case can, therefore, be made for abandoning whole brain or brainstem death terminology as redundant, confusing, and unnecessary. Indeed, the latest UK guidance on the confirmation of death no longer uses the term brainstem death but, rather, the diagnosis of death following irreversible loss of brainstem function and reflexes.5 Both whole brain and brainstem formulations adopt the same 3-stage approach for the determination of brain death described above, and, in many jurisdictions, there is no requirement for additional or ancillary tests that may allow a differentiation between the 2. This perhaps reflects that about 98% of all clinical examinations consistent with the confirmation of brain death involve demonstrating the infratentorial (brainstem) manifestations of a primarily supratentorial (whole brain) insult such as subarachnoid or intracerebral hemorrhage, or traumatic or hypoxic brain injury.
Brain death from an isolated brainstem lesion is exceedingly rare. In the Henry Ford Hospital’s case series, only 2% of patients meeting the clinical criteria for diagnosing brain death had an isolated brainstem lesion as the underlying cause.6 All these patients had persistent supratentorial blood flow that was lost over time,6 meaning that all initially met the “brainstem” requirements for confirming brain death and eventually all also met the “whole brain” criteria. The rarity of isolated brainstem lesions—if seen in posterior circulation ischemia, both thalami are involved—as the underlying cause of brain death means that many physicians will never confirm death in these circumstances. Also, the current investigations undertaken primarily to establish the underlying diagnosis, rather than as an ancillary test to support the determination of brain death, will increasingly demonstrate the occasional persistence of supratentorial blood flow. It is likely that some colleagues will feel uncomfortable and less confident in confirming brain death in these circumstances. Some authors recommend that demonstrating an isoelectric electroencephalogram (EEG) or absent supratentorial blood flow is mandatory before confirming brain death in cases of an isolated brainstem lesion.7 They attempt to justify this recommendation on the basis of an unproven theoretical possibility of sparing of the meso-pontine tegmental reticular formation by an isolated brainstem injury, with potential for a total apneic locked-in syndrome mimicking brainstem death.7 This justification requires knowledge of the boundaries and exact position of the reticular formation and implausible ellipsoid lesions sparing all brainstem nuclei and tracts, none of which are known or seen. There are, however, circumstances when we can be confident of the permanent loss of any capacity for consciousness even with preservation of some supratentorial blood flow (Fig. 1A). This raises 2 crucial questions.
First, whether persisting supratentorial blood flow is associated with any brain function, particularly consciousness. There are no reports of patients in whom the determination of brain death was correctly undertaken using either the whole brain or brainstem formulations ever regaining consciousness or starting to breathe again. Furthermore, persisting supratentorial flow or EEG activity does not equate to the presence of brain function. This question also needs to take into consideration, the fact that 15% of all patients with supratentorial lesions who meet the criteria for the determination of brain death have persistent supratentorial flow,8 and 20% have persistent EEG activity,9 both of which are of questionable significance.9,10 Such persistent EEG activity may be considered as the brain equivalent of ECG activity without cardiac function in pulseless electrical activity. In these circumstances, many clinicians would still be confident in confirming death using neurological criteria, and, in practice, the determination of brain death is frequently undertaken without clinicians being aware of such persisting blood flow or electrical activity, because ancillary tests are not considered mandatory on most occasions.2 It remains essential to differentiate between brain function that can be detected at the bedside examination and brain activity that is not associated with any detectable function. Uniformly, on both sides of the Atlantic, confirmation of brain death is based on the irreversible loss of function rather than on demonstration of loss of brain cellular activity using any advanced technology. Functional magnetic resonance imaging is increasingly used to investigate minimally conscious states, and it is likely to be similarly investigated following brain death, but the issue of differentiating responses that demonstrate brain function rather than residual activity will remain.
The second crucial question is whether it is necessary to await the loss of any persisting supratentorial cerebral blood flow (Fig. 1B) before diagnosing brain death in patients with an isolated brainstem lesion. Demonstrating the eventual loss of supratentorial blood flow in these rare cases may increase the confidence of some physicians in making the diagnosis. However, in the United Kingdom, there is no difficulty diagnosing brain death in such circumstances, and only a few physicians in the United States may wait for the complete loss of flow. To mount a challenge to the concept of death in a patient with a dead brainstem would require evidence of reversibility of important functions or other evidence of a functioning circuit producing awareness. None exist. Moreover, if all patients with an isolated brainstem lesion are shown to lose supratentorial blood flow over time, performing such a test may no longer be necessary. In the meantime, continuing with the current approach for the confirmation of brain death is justified by 50 years of worldwide experience in making the diagnosis with 100% specificity and sensitivity.
The perceived divide between whole brain and brainstem death is now kept “alive” only by a minority. It has more to do with emotive concepts rather than hard neurobiological facts, and represents a failure to accept the centrality of the brainstem in defining life or death. Once the functionality of the mesencephalic and pontine regions is irreversibly lost, and the lesion reaches the caudal medulla, all vital functions are also irreversibly lost.
Alex Manara, MB, BCh, FRCA, FRCP, FFICM*
Panayiotis Varelas, MD, PhD, FNCS, FAAN†
Eelco F. Wijdicks, MD, PhD‡
*Southmead Hospital, North Bristol NHS Trust, Bristol, United Kingdom
†Neuro-Intensive Care Unit, Department of Neurology, Henry Ford Hospital, Detroit, MI
‡Department of Neurology Mayo Clinic, Division of Critical Care Neurology, Rochester, MN
1. A definition of irreversible coma. Report of the ad hoc committee of the Harvard Medical School to examine the definition of brain death. JAMA. 1968;205:337–340.
2. Wijdicks EFM, Varelas PN, Gronseth GS, et al. Evidence-based guideline update: determining brain death in adults Report of the Quality Standards Subcommittee of the American Academy of Neurology. Neurology. 2010;74:1911–1918.
3. Shemie SD, Hornby L, Baker A, et al. The International Guidelines for Determination of Death phase 1 participants, in collaboration with the World Health Organization. International guideline development for the determination of death. Intensive Care Med. 2014;40:788–797.
4. Wijdicks EF. The transatlantic divide over brain death determination and the debate. Brain. 2012;135:1321–1331.
6. Varelas PN, Brady P, Rehman M, et al. Primary posterior fossa lesions and preserved supratentorial cerebral blood flow: implications for brain death determination. Neurocrit Care. 2017;27:407–414.
7. Walter U, Fernández-Torre JL, Kirschstein T, et al. When is “brainstem death” brain death? The case for ancillary testing in primary infratentorial brain lesion. Clin Neurophysiol. 2018;129:2451–2465.
8. Kramer AH, Roberts DJ. Computed tomography angiography in the diagnosis of brain death: a systematic review and meta-analysis. Neurocrit Care. 2014;3:539–550.
9. Grigg MM, Kelly MA, Celesia GG, et al. Electroencephalographic activity after brain death. Arch Neurol. 1987;44:948–954.
10. Sawicki M, Sołek-Pastuszka J, Chamier-Ciemińska K, et al. Computed tomographic perfusion is a useful adjunct to computed tomographic angiography in the diagnosis of brain death. Clin Neurorad. 2017. Doi: 10.1007/s00062-017-0631-7. [Epub ahead of print].