Clinical Ethics in Anesthesiology. A Case-Based Textbook
2. End-of-life issues
19. Brain death
Robert B. Schonberger and Stanley H. Rosenbaum
A 35-year-old patient is comatose following a drowning. The treating neurologist has found the patient to be irreversibly unconscious with absent brainstem reflexes and no respiratory effort during an apnea test. Because of persistent limb movements thought to be most likely spinal in origin, the primary team elects to get a confirmatory EEG that indicates no brain activity. At the conclusion of EEG testing, the attending neurologist writes in the chart, “3/15/09 20:00. The patient is now deceased.” The next of kin decide the following morning not to donate the organs, and respiratory support is promptly removed with an accompanying chart note saying, “3/16/09: The patient has died. Time of death 0700.” Two weeks following this incident, the state medical examiner contacts the care team and inquires as to the proper time of death.
The historical development of a neurological standard of death
While the determination of death among many traditions has for thousands of years relied on the cessation of the pulse or respirations, twentieth-century medical care brought about both new possibilities and incentives for redefining death in terms of a neurological standard. The possibility of brain death in the presence of continued cardiovascular function emerged during the 1950s, largely as a consequence of new developments in the medical care of the critically ill. Shortly thereafter, the incentives for a new definition of brain death expanded and took on new practical urgency with the advent of deceased-donor organ transplantation.
Among the medical developments that led to the possibility of a new concept of brain death was the improvement in techniques of pulmonary support via mechanical ventilation during the polio epidemics of the 1940s and 1950s.1 Other techniques of cardiovascular support soon followed, including the development of cardiopulmonary bypass in the 1950s and the formalization of cardiopulmonary resuscitation in the late 1950s and early 1960s. As society was confronted with the possibility of an apparently alive body in an irreversibly brain dead patient, medicine was ripe for a neurologically based definition of death that could guide the ethical discontinuation of cardiopulmonary support in such patients. In 1959, the concept of irreversible coma was introduced by Mollaret and Goulon in France. However, it may have been the subsequent development of organ transplantation, and the accompanying increase in the practical import for determining brain death, that led 10 years later to the creation of the first widely accepted standard of brain death.
In 1968, 1 year after the first successful heart transplant, a well-publicized effort in the United States to reexamine the definition of death in terms of a neurological standard was conducted by a committee at Harvard Medical School.2 This committee, lead by the chairman of Anesthesiology at Harvard Medical School, Henry Beecher, MD, published what subsequently became known as the Harvard Criteria for determination of brain death which included: (1) unreceptivity and unresponsivity; (2) no movements or breathing; and (3) no reflexes (including deep tendon reflexes). The criteria also suggested checking an electroencephalogram (EEG) when available for its “great confirmatory value.”2 An isoelectric EEG should demonstrate no brain activity, and there should be no muscle movement. The Harvard Criteria mandated these tests be done twice separated by 24-hours. In addition, it was necessary to confirm that the patient was neither sedated nor hypothermic. While the Harvard criteria would be modified by various authorities over the subsequent four decades as described below, these guidelines represented the inaugural moment for a neurological standard of death into medical practice.
Current international neurological standards for determination of death
The whole brain death standard
While the concept of brain death is now well established, the specific criteria for its determination vary among different countries. The legal brain death standard that has been adopted in the majority of the United States is based on the advice of the National Conference of Commissioners on Uniform State Laws3 which formed a model of brain death legislation in 1980 that was subsequently adopted by both the American Medical Association and the American Bar Association as well as a large majority of state legislatures. This standard requires the determination of “irreversible cessation of all functions of the entire brain, including the brain stem.” Similar so-called “whole brain death” standards have also been adopted in Canada, Australia, and South America, as well as most European countries with the notable exception of the United Kingdom (see below).
East Asian countries have seen more recent acceptance of a neurological standard, with brain death criteria analogous to the US standards having been adopted in Japan in 19971 and in South Korea in 2000. A major meeting of Chinese officials to discuss the establishment of neurological standards of brain death occurred in 2008, but no English language summary of their findings was located by the present authors.
The determination of whole brain death, when made on clinical grounds, requires the demonstration of three things: (1) an irreversible comatose state; (2) the loss of brainstem reflexes; and (3) brainstem inactivity leading to apnea. After reversible causes of apparent coma, such as hypothermia, intoxication, severe metabolic derangement, or residual neuromuscular blockade have been ruled out or otherwise corrected, a patient meets the standard by demonstrating the absence of responses to painful stimuli including to cranial nerve territories (coma), the lack of all cranial nerve reflexes (brainstem areflexia), and the absence of respiratory efforts in the face of a hypercarbic challenge (brainstem inactivity). Of note, the whole brain death standard differs from the Harvard criteria in making no mention of spinal cord function or the total absence of muscle movements. This standard also requires no documentation of hypothalamic failure. Although hypothalamic dysfunction including diabetes insipidus is often seen in patients diagnosed as brain dead, both hypothalamic and pituitary function as well as spinal reflexes are commonly found to continue in patients who meet the current whole brain death criteria.
Protocols for the clinical determination of brain death vary among institutions but must generally be made by more than one doctor in one of several relevant specialties (usually some combination of neurology, neurosurgery, trauma, or anesthesiology). In some institutions, most commonly in reference to pediatric patients, clinical assessment must be repeated over various intervals of time before the declaration of death can be properly made.
Sufficiency of the clinical diagnosis of brain death versus neurophysiologic testing
The majority of the US as well as most European Union countries have followed the lead of the Harvard committee in specifying that the clinical diagnosis of brain death in adults is sufficient in itself for the determination of death, without the need for confirmatory neurophysiologic testing. In these jurisdictions, neurophysiologic assessment is reserved for young pediatric patients as well as for circumstances in which there is an inability to perform any of the required clinical tests of brain death or where, for whatever reason, some doubt exists about the clinical diagnosis of brain death. For example, a patient too unstable to tolerate an apnea test, may instead undergo alternative neurophysiologic testing. Exceptions to the sufficiency of the clinical standard in adults include France, Italy, Luxembourg, and the Netherlands, each of which requires some type of confirmatory neurophysiologic testing even in the presence of clinical brain death.4
Neurophysiologic testing generally falls into two categories that serve either “to confirm the loss of bioelectrical activity of the brain” or to “demonstrate cerebral circulatory arrest.”4 Confirmation of brain death via examination of bioelectrical activity can itself be conclusive via the finding of an isoelectric EEG in the absence of other causes for central neurological depression. The absence of brainstem auditory evoked potentials or somatosensory-evoked potentials can also be useful in certain circumstances. These tests have the advantage of being less susceptible than EEG to distortion by sedative drugs, but evoked potential monitoring, while suggestive, cannot in isolation conclusively determine whole brain death.
Neurophysiologic determinations of cerebral circulatory arrest include four-vessel cerebral angiography as well as various scintigraphic perfusion studies. Angiography demonstrating absence of blood flow to the brain is widely accepted throughout Europe and the US as a valid standard, while scintigraphic perfusion studies, which may be less sensitive markers for the absence of brain perfusion, are accepted in some United States institutions as well as in Germany and Switzerland.4 Finally, transcranial Doppler may be used as a valid confirmatory test of brain death only in Germany but is not in itself sufficient for the diagnosis of brain death due to the possibility of an insufficient bone window for valid blood flow determinations.4
The brainstem death standard
The UK and India are unique in having a neurological standard of death that specifically focuses on brain stem dysfunction. In 2008, the UK’s Academy of Medical Royal Colleges published its “Code of Practice for the Diagnosis and Confirmation of Death.”5 (UK) The authors reaffirm the UK’s original 1976 neurological standard of death which was based on “the irreversible cessation of brainstem function.” In India, a similar declaration was part of the Transplantation of Organs Act of 1994. The Academy of Medical Royal Colleges argues that loss of brainstem function entails both apnea as well as the loss of consciousness and therefore suffices for the determination that the “death of the individual” has occurred. For the exclusively clinical diagnosis of brain death, there is little practical difference between the UK definition and a whole brain definition of death. If neurophysiologic testing is used, however, patients with continued cortical EEG activity or cortical blood flow could still be considered dead according to the UK and Indian standards but not by the other countries considered in this review.
Practitioners involved in brain death determinations or engaged in the care of deceased organ donors would be well advised to familiarize themselves with their own institutional guidelines as well as the laws of their particular jurisdiction, as the nuances and procedural requirements for determination of brain death may be critical in determining the status of suspected brain dead patients. Moreover, some jurisdictions give the next-of-kin the right to deny that a neurological standard of death be used at all.
The ethical foundations for a neurological standard of death
Philosophical underpinnings of differing neurological standards of death
The philosophical question of why brain death should be equated with the death of a patient has been answered largely either in terms of sociological traditions or bio-philosophical arguments. Several arguments have been made in support of a neurological standard, and they have great import for the resultant neurological standard, if any, that is advocated. The current review will focus briefly on three common ethical foundations used in support of a neurological standard of death.
Formulations of a neurological standard based on a reliance on cultural traditions
Defenders of various formulations of the neurological standard of death sometimes reference or explicitly rely on sociological traditions for their justification of a particular standard. For example, neurologist Christopher Pallis,6 one of the early developers of the UK standard, justified his country’s standard in reference to a tradition that equates both breathing and consciousness to the existence of the human soul. The tradition focusing on apnea and unconsciousness does indeed have quite a strong history within the Judeo-Christian framework of which Pallis was a part.
Genesis 2:7 recounts “and the Lord God formed man of the dust of the ground and breathed into his nostrils the breath of life, and man became a living soul.”7 This tying together the idea of breath – albeit a positive pressure breath! – with the essence of life has resonated deeply through Western culture. The Christian doctrine that the soul departs the body at the time of death has been interpreted by many theologians, Pope Pius XII among them, as meaning that the soul leaves the body at the time of irreversible loss of consciousness. It is of interest to note that the sole reference listed in the Harvard committee’s original report was a speech by Pius XII.
It is, of course, possible to advocate an apnea and unconsciousness standard on non-religious grounds. In addition, it is possible to advocate for other neurological standards – or none at all – based on tradition. Any standard thus formulated is subject to all the strengths and weaknesses of all arguments from tradition. Some thinkers might argue that all neurological standards share at some level such a philosophical foundation.
Formulations of the neurological standard based on “loss of personhood:”
Some justifications for a neurological standard resort to a “loss of personhood” argument, asserting that the absence of a particular set of functions – often permanent lack of consciousness and reasoning ability – even when they occur independently from the death of the human organism, may nevertheless suffice for the ethical and legal determination of death of the person.
Such a theory finds resonance in the Western canon at least as far back as Aristotle who argued that the rational soul was the unique and essential characteristic of man. Some argue that an individual who lacks such capability has lost his or her essential claim to personhood. This stance may have provided the ethical foundation for a decision by the American Medical Association Council on Ethical and Judicial Affairs in 1995, that advocated for the use of anencephalic neonates as organ donors.1 Such neonates are born without a brain cortex and could reasonably be assumed to lack the possibility of a meaningfully conscious life.
In accord with the AMA council’s philosophical view of personhood, it might have been reasonable to assume (though the council disputed this suggestion) that not only anencephalic newborns, but any persistently vegetative patients who lacked function beyond primitive brainstem activities might reasonably qualify to be considered dead-as-people. The council subsequently reversed their stance regarding anencephalic babies,1 presumably concluding that denying personhood based on an ill-defined threshold of higher order brain functions was a hopelessly messy enterprise, with possibly untoward slippery slope implications. As a matter of law, there is no jurisdiction that currently equates persistently vegetative states with death.
Loss of personhood arguments, however, can also be used to advocate for other standards, including the whole brain death standard. In separating the death of the person from the ultimate death of the patient, such a formulation resonates with the common desire to call a person with a beating heart still living. In doing so, however, such arguments shy away from advocating for a truly neurological standard of death equivalent to the cardiovascular standard. It may be this argument that has prevailed in some jurisdictions where next-of-kin may refuse a neurological standard altogether. If the neurological standard of death were truly equivalent to a cardiovascular death in the eyes of the legislators who passed such laws, it would be difficult to imagine that they could grant a family the right to consider a dead patient alive, no matter what the family’s beliefs.
Foundations specific to the whole brain death neurological standard
So far in this discussion, we have seen that even among people who agree that a neurological standard of death should exist, the specifics of what such death should entail as well as the arguments used in their support can be vastly different. Of the few positions outlined above, one relies on an exclusively brainstem-based standard while another relies on an exclusively cortical standard. We will now turn to an ethical foundation that may be specific to the whole brain death standard.
The bio-philosophical position that was expounded by some members of the United States President’s Council on Bioethics in their 2009 report advocates for the whole brain death standard. In contrast to the “loss of personhood” standard, they dispute that there can be the death of the person independent from some more fundamental phenomenon of organismal death.
Specifically, they assert that there is only one death, and that permanent and total brain failure, including both cortex and brainstem, is its necessary harbinger. Underlying their stance is the argument that a living human organism must of essence exhibit “self-preserving commerce with the world” including some inchoate expression of neediness and actions to satisfy those needs. The necessary qualities for human organismal life have been expressed by one author as “openness to the surrounding environment, ability to act upon that environment, and inner experience of need.”6
In accord with this philosophical position, the neurological standard that would warrant a declaration of death is more stringent than both a “loss of breath and consciousness” standard as well as most incarnations of a “loss of personhood” standard. This philosophical stance requires that function of the entire brain – including cortical and brainstem activity – cease prior to the determination of death. The majority of the Presidential council that delineated this standard supported the current United States standard of whole brain death even as they provided a new philosophical grounding for it.
They rejected an exclusive reliance on brainstem death – the apnea and unconsciousness standard of England and India – largely based on epistemological grounds. The principal objection was that the inner state of a person with cortical electrical activity after brain stem death is fundamentally unknowable, and it would therefore be problematic to conclude that there was a definitive lack of consciousness in cases of continued cortical activity. There may be good reason for such doubt, given recent strong evidence for the preservation of higher-order consciousness in some patients previously considered to be in persistently vegetative states.8
By their standard, they also would reject a purely cortical definition of brain death, since in their view a spontaneously breathing but unconscious person would still possess a primitive neediness, openness to the world, and ability to act and would thus fulfill the criteria for continued life. It is notable that hypothalamic activity was not addressed in detail by the President’s council in their report, which leads to the question of why temperature autoregulation (a feature of hypothalamic activity) would not also qualify as an inchoate self-preserving commerce with the world that was worthy of being called life.
The whole brain death standard – with slight variations – has come to be the most widely held international neurological standard of death. While the philosophical foundations underlying the whole brain death standard may include arguments from tradition, loss-of-personhood arguments, or other arguments not described here, it seems that the “self-preserving commerce with the world” stance may be most specific to a whole brain death standard.
Determining the time of death – the essence of the neurological standard
From a legal standpoint, most jurisdictions hold that when the neurological standard of death has been satisfied, it is legally equivalent to a declaration of death based on cessation of cardiovascular function. Such jurisdictions would hold that the neurologist’s first declaration of death was the correct one.
From an ethical standpoint, according to the “self-preserving commerce with the world” stance, as well as many arguments based on sociological tradition, a brain dead patient is fundamentally equivalent to a corpse. As such, a person fulfilling the whole brain death criteria should be considered not just dead as a person, but dead as an organism. Such an ethical standpoint therefore would agree that the neurologist’s original declaration of death was correct and that the subsequent declaration the following day misunderstood the essence and practical import of a neurological standard of death.
In this regard, the “loss of personhood” stance fundamentally differs from the other philosophical stances since it maintains a distinction between death of the person and death of the organism. Although few jurisdictions follow this distinction, an argument could be made under the loss of personhood standard that both declarations of death in the case above were correct – the neurologist’s declaration of death was in reference to the death of the person and second declaration was in reference to the death of the organism. Such a distinction is still widely held in the public and the press as evidenced by frequent reports of brain dead patients “being kept alive” on life support pending final care decisions. For most advocates of a neurological standard, such a statement is accurate only in reference to persistently vegetative patients.
Whether or not a ventilator remains on or off and whether or not there is a blood pressure, most jurisdictions hold that a brain dead patient is a corpse. This is the essence of the neurological standard of death, and both doctors and society at large continue to struggle with its justification.
• Advances in medical technology and transplantation lead to the re-defining of death to include not only cardiopulmonary death, but death by virtue of “brain death.”
• The medical and legal definitions of brain death vary slightly among countries – with some using a “whole brain” definition of death in which no cortical activity may be present, and some relying on a “brainstem” standard, in which some cortical activity may be detectable.
• In all countries where brain death is recognized legally, the diagnosis rests with physical examination, at times supported by further medical testing.
• Philosophical arguments for the integrity of brain death as a definition of death rest in historic religious and social concepts of what constitutes life, or with ideas that loss of “personhood” may be equivalent to, or another form of, death.
• Because of differences in social, religious and cultural beliefs, “brain death” is by no means universally accepted, and continued debate can be expected regarding what, if any, neurological standard of death should be recognized.
1* Van Norman, G.A. (1999). A matter of life and death: what every anesthesiologist should know about the medical, legal, and ethical aspects of declaring brain death. Anesthesiology, 91(1), 275–87.
2* A definition of irreversible coma: a report of the ad hoc committee of the Harvard Medical School to examine the definition of brain death. (1968). JAMA, 205, 337–40.
3* Uniform Determination of Death Act. (1980). National Conference of Commissioners on Uniform State Laws. Chicago, IL. http://www.law.upenn.edu/bll/archives/ulc/fnact99/1980s/udda80.htm
4* Haupt, W.F. and Rudolf J. (1999). European brain death codes: a comparison of national guidelines. J Neurol, 246(6), 432–7.
5* A Code of Practice for the Diagnosis and Confirmation of Death by the Academy of Medical Royal Colleges. (2008). http://www.aomrc.org.uk/aomrc/admin/reports/docs/DofD-finalpdf 2008.
6* Controversies in the Determination of Death: A White Paper by the President’s Council on Bioethics. (2009) http://www.bioethics.gov/reports/death/ 2009.
7 Holy Bible, King James version.
8* Owen, A.M., Coleman, M.R., Boly, M., et al. (2006). Detecting awareness in the vegetative state. Science, 313, 1402–3.
Aita, K. (2009). Japan approves brain death to increase donors: will it work? Lancet, 374(9699), 1403–4.