Brain death: justifications and critiques

Introduction

An authoritative review of the concept of ‘brain death’ published in the New England Journal of Medicine in 2001 began ‘Physicians, health care workers, members of the clergy, and laypeople throughout the world have accepted fully that a person is dead when his or her brain is dead.’ ¹ If this were true, then an examination of the justifications and critiques of the concept of ‘brain death’ would not be worth the effort. But an editorial that accompanied this review, written by Alex Capron, one of the authors of the American brain-death legislation, offered what we regard as a more honest assessment: ‘If one subject in health law and bioethics can be said to be at once well settled and persistently unresolved, it is how to determine that death has occurred.’ ²

Capron captured an important truth about the diagnosis of ‘brain death’. On the one hand, it occupies a place deep within the fabric of contemporary medicine, serving as a foundational concept for the life-saving industry that has grown up around organ transplantation. It is likely that the great majority of practising physicians have no idea that any controversy surrounds the concept at all.

On the other hand, one does not need to dip very far below this calm surface to find strong currents of controversy. Since 1974, when Hans Jonas published his aptly named critique ‘Against the Stream’, ³ there has been a continuous flow of dissent. Doubts about the diagnosis frequently bubble up from below, as when family members struggle to understand how their loved ones can be regarded as ‘dead’ when they are breathing (with the aid of the ventilator), their hearts are beating and they are warm to the touch – far from a cold corpse.

The issue has been especially contentious in Japan, where under pressure from the transplantation community to legalize the concept of brain death, Japan adopted a half-way measure that permits patients who meet brain death criteria to be organ donors, but does not stipulate that brain death constitutes death. ⁴ In the USA, new controversy emerged in 2008 when the President's Council on Bioethics published a white paper on the diagnosis of death. ⁵ While disagreeing with all previous formulations explaining why the criteria for brain death constitute death, the Council created yet another rationale, which we will discuss below.

In this essay, we will review the origins of this controversy, attempt to show that proposed solutions to the problems are inadequate, explore the implications of this critique and close with some thoughts about where we may be headed in the future.

Historical foundations

Until the last several decades, there was little controversy about the definition of death. People or animals were dead when they were cold and stiff. Some ambiguity was recognized, however, around the phenomenon of decapitation. For example, after witnessing the execution of a convict by the guillotine, a Parisian pathologist wrote, ‘One hour after the execution the heart still beat; yet this man's existence was over, he had lost his personality, and yet his heart was beating!’ ⁶ Yet, these ambiguities were of little practical relevance, as whether death was defined in terms of the brain or the heart made little difference – the failure of either to function quickly led to the failure of the other.

All of this changed in the mid-20th century, however, with the development of mechanical ventilation. Now it was possible to keep patients with profound brain injury alive for an indefinite period of time by keeping them on a ventilator. This phenomenon was first described by French physicians as ‘coma depasse’ or ‘beyond coma’. ⁷ Analogous to anatomic decapitation, in which the brain is physically separated from the body, this condition was considered to be a form of physiological decapitation.

These ideas became of more than just academic interest in 1968, when Christian Barnard performed the first heart transplant in the ‘miracle of Cape Town’. ⁸ The transplant raised an urgent question: was the donor dead at the time that his heart was removed, and if not, then were the physicians responsible for killing him? Henry Beecher at Harvard Medical School immediately recognized the significance of this question, and chaired an ad hoc committee that published one of the seminal articles in the field: ‘A definition of irreversible coma, Report of the Ad Hoc Committee of the Harvard Medical School to Examine the Definition of Brain Death’. ⁹

The title reveals a gap in logic that has never been bridged. While the authors were confident in claiming a criterion for defining irreversible coma, they were only implying (without argument) that this might also be a criterion for determining death. A parallel process occurred in the UK, where in 1976 a medical committee stated that ‘permanent functional death of the brainstem constitutes brain death and that once this has occurred further artificial support is fruitless and should be withdrawn’, clearly asserting that ‘brain death’ was a criterion for justifying the withdrawal of life support, while being ambivalent about whether it was a criterion for declaring death. ^10,11 On both sides of the Atlantic, the concept of ‘brain death’ was developed in terms of other concepts (irreversible coma or the justified withdrawal of life support) without ever providing any clear argument or justification for why ‘brain death’ constituted death. In short, the point of this paper is to show that this challenge has never been convincingly addressed.

Justifications and critiques

In the USA, ‘brain death’ is defined by the Uniform Determination of Death Act (UDDA), using the so-called ‘whole-brain criterion’: ‘An individual who has sustained either (1) irreversible cessation of circulatory or respiratory functions, or (2) irreversible cessation of all functions of the entire brain, including the brainstem, is dead. A determination of death must be made in accordance with accepted medical standards.’ ¹² The first set of criteria for making this diagnosis were published in tandem with the UDDA, and with relatively minor modifications these are the standards still in use today. ¹³ They require showing that the patient is comatose, has lost a battery of brainstem reflexes as well as the drive to breathe, and that these changes are irreversible.

The critique of the American definition begins with the observation that the current standards require the testing of only a small number of brain functions, mostly limited to the brainstem. When more thorough testing is done, one finds that patients who meet the standard do, in fact, retain many brain functions, including the secretion of hormones, temperature regulation, etc. ^14–17 Hence, many patients who fulfil the diagnostic standards for brain death do not fulfil the legal definition, which requires the cessation of all brain functions.

This flaw was recognized almost immediately after publication of the UDDA, and was addressed in an influential paper by James Bernat et al. in 1981. ¹⁸ They suggested that death should be defined as ‘the permanent cessation of functioning of the organism as a whole’. The idea they advanced was that the brain is the central regulator of the body's homeostasis, and they pointed to evidence that when the brain loses the functions included in the standard diagnostic criteria for brain death, then functioning of the whole is lost (as manifested by the development of cardiac arrest within a short period of time, regardless of the intensity of life support provided). At the time, this was an elegant solution to the problem of defining ‘brain death’, since it provided a rationale for why it is not necessary to document the loss of all brain function (as required in the UDDA), but only those that are necessary to maintain the functioning of the organism as a whole.

We endorse Bernat's definition of death, and agree that death is defined as the loss of integrated functioning of the organism as a whole. In a work published since 1981, however, it has become absolutely clear that the diagnostic criteria for brain death no longer define the permanent loss of integration. Most of the careful work on this question has been done by neurologist Alan Shewmon. In a report of 175 patients who had been diagnosed as brain dead and had survived for longer than a week, he found many examples of prolonged survival, with one dramatic case of a boy who survived for more than 20 years. ¹⁹

Why did the evidence show that the diagnostic criteria for ‘brain death’ predicted the loss of functioning of the organism as a whole in 1981, but failed to show this in more recent years? Shewmon has suggested that the answer may lie in an analogy between ‘brain death’ and high cervical cord transection (which leads to cervical quadriplegia). ²⁰ These conditions are similar: in both there is a rapid loss of cerebral influence over the rest of the body, in the former case because of brain damage and in the latter case because of disruption of the neural pathways connecting the brain with the body. In both conditions, the sudden loss of modulatory input from the brain to the body is initially very disruptive, often resulting in a life-threatening form of haemodynamic instability known as spinal shock. In the absence of the medications and monitoring capabilities of the modern intensive care unit (ICU), patients with spinal shock rarely survive, explaining the findings observed with brain-dead patients in the 1960s and 1970s.

With the development of the modern ICU, however, it became possible to better support patients through the acute period following either ‘brain death’ or high cervical cord transection. Indeed, a significant part of the survival benefit of modern ICU care comes from the use of technologies that function as a ‘surrogate brainstem’, regulating those haemodynamic and respiratory functions that are critical to life. Once this acute phase has passed (generally a month or two after the injury), the body develops the capacity to self-regulate, and if provided with minimal ‘surrogate brainstem’ functions (such as mechanical ventilation and supplemental hormones), can survive for many years.

If this is the case, then why is it that our hospital ICUs are not full of patients who have been diagnosed as ‘brain dead’? The reason, of course, is that the rapid onset of death after the diagnosis is made has become a self-fulfilling prophesy. Once families understand that their loved-one has no hope of regaining consciousness or any meaningful existence, they agree to either organ donation or withdrawal of the ventilator. But even for the small number of families who refuse either of these, physicians are legally empowered to withdraw mechanical ventilation against the wishes of the family, since ‘brain death’ is nationally recognized as legal death, and physicians have no obligation to ventilate a corpse.

It is important to appreciate, however, that even though patients who previously met ‘brain death’ criteria rapidly progressed to cardiac arrest while on life support, it does not follow that they were dead at the time that ‘brain death’ was diagnosed. To claim that they were dead confuses a diagnosis of death with a prognosis of imminent death. In other words, being in a state likely to lead to death in a short period of time is not the same as being dead.

In current practice there is one situation, however, in which physicians sometimes agree to continue life support for a patient declared ‘brain-dead.’ This is when ‘brain death’ occurs in a woman early in her pregnancy, and her family asks the physicians to keep her alive until the baby can be delivered at a viable gestational age. Many such cases have been reported, ²¹ and the fact that they are successful vividly illustrates that the diagnosis of ‘brain death’ does not imply the loss of functioning of the organism as a whole. Otherwise, we are faced with the paradox that a corpse can gestate a viable fetus.

This evolution in our understanding of brain death was encapsulated in the 2008 report from the President's Council on Bioethics on ‘Controversies in the Determination of Death’. In what has been called a courageous document, the Council fully endorsed the central point argued above – that ‘brain death’ does not represent the loss of integrated functioning. After reviewing all of the integrated functions retained by patients diagnosed as ‘brain dead’, it concluded that ‘If being alive as a biological organism requires being a whole that is more than the mere sum of its parts, then it would be difficult to deny that the body of a patient with total brain failure (the Council's term for ‘brain death’) can still be alive, at least in some cases’. ⁵

The Council fully understood the profound effect that this conclusion could have on the life-saving practices around organ donation and transplantation. And so, perhaps to avoid these consequences, they proposed yet another rationale for why ‘brain-dead’ patients should be considered dead. While abandoning ‘the false assumption that the brain is the ‘integrator’ of vital functions,’ they instead appealed to the novel notion of ‘the vital work of a living organism – the work of self-preservation, achieved through the organism's need-driven commerce with the surrounding world’. The living organism does its vital work when it satisfies the following three criteria: it exhibits (1) ‘receptivity to stimuli and signals from the surrounding environment’, (2) ‘the ability to act upon the world to obtain selectively what it needs’ and (3) ‘the basic felt need that drives the organism to act as it must, to obtain what it needs’.

The white paper claims that patients with a diagnosis of total brain failure completely fail to meet each of these criteria. In contrast, it asserts that patients in a persistent vegetative state (PVS), who breathe spontaneously but show no signs of consciousness, completely fulfil these criteria. As we have described in more detail elsewhere, ²² it is difficult to understand how this argument can be sustained. Consider a pregnant woman diagnosed as brain dead and gestating a fetus for several months in an ICU. How can the Council's criteria be applied to conclude that this patient is ‘dead’ and a similarly unconscious patient in PVS is ‘alive.’ Indeed, the Council's report does not provide the detail necessary to understand how their criteria could sustain these distinctions.

In contrast to the USA, the UK has never endorsed the concept of whole-brain death that is the basis for the UDDA. ^10,11 Instead, the definition used in the UK emerged primarily from the work of the philosopher David Lamb and neurologist Christopher Pallis, and requires only the death of the brainstem. ^23,24 Briefly, they defended a definition of death based solely upon the irreversible loss of the capacity for both spontaneous breathing and consciousness. Although consciousness is generally assumed to be an emergent property of the brain as a whole, they argue that a structure in the upper brainstem, known as the reticular activating system (RAS), essentially functions as an on/off switch regulating arousal and consciousness. Similarly, respiration is neurologically controlled by several discrete centres within the brainstem. As a result, death of the brainstem leads to loss of both consciousness and spontaneous respiration.

An advantage of the British definition is that it focuses on the loss of two discrete functions, rather than the ‘complete loss of all functions of the entire brain’. While the loss of the brainstem centres driving respiration is directly tested in the apnoea test, there is no way to directly test the functions of the RAS. Instead, by testing for the destruction of a variety of nearby brainstem centres (essentially the same battery of tests used in the American criteria), the definition relies on the assumption that if all the nearby centres are destroyed, then the RAS must be destroyed as well. While this may generally be the case, patients with the ‘locked-in’ syndrome are a striking example of where consciousness may be retained despite the loss of virtually all other brainstem functions.

Since the British definition does not require the loss of functioning of the organism as a whole, it may be thought to be immune from the criticism levelled at the American approach. However, it must be defended against the question of why loss of these two functions should constitute the death of the person when a host of other basic biological functions indicative of life persist in patients with loss of brainstem functioning maintained on mechanical ventilation, including respiration, circulation, excretion of wastes, wound healing, temperature regulation, etc. Clearly, the loss of either of the two key functions of consciousness or spontaneous respiration alone is not sufficient. Patients who are in a PVS but who continue to breathe spontaneously are not dead by the brainstem definition, and patients who are conscious but who do not breathe spontaneously (like the late Christopher Reeves) are not dead either. What is it about the combination of the two that makes the difference?

Exploring this point further, consider a thought experiment involving two men who are in good health and identical except that one of them has cervical quadriplegia and requires a mechanical ventilator to be alive. Imagine that both suffer an identical exposure to carbon monoxide, rendering both of them as permanently unconscious. The man who was normal is now in a PVS (permanent loss of consciousness). The man with cervical quadriplegia is now dead by brainstem criteria (permanent loss of consciousness and spontaneous respiration). How could two equally alive individuals suffer an identical injury, with the result being that one of them is still alive and the other dead?

When questioned on why he believes that only these two functions should matter, Pallis departed from a biological definition of death and defended this choice based upon the importance of the sociological context to defining basic concepts such as life and death: ‘The single matrix in which my definition is embedded is a sociological one, namely Judeo–Christian culture… The ‘loss of the capacity for consciousness’ is much the same as the ‘departure of the conscious soul from the body,’ just as ‘the loss of the capacity to breathe’ is much the same as the ‘loss of the breath of life.’ ²⁴

In evaluating this position, the President's Council raised two interesting questions: ‘First, are consciousness and breathing the only or the most important culturally significant features of the soul? Second, does this argument about traditional beliefs, bound to a particular culture, provide a sufficient rationale for a standard applicable to the transcultural, universal phenomenon of human death?’ ⁵ By basing death on cultural concepts, the brain stem criterion arbitrarily makes human death essentially different from the death of other living organisms that do not have souls or consciousness.

Implications of the critique

The purpose of this paper is to outline some of the key formulations of the concept of ‘brain death’, along with their justifications and critiques. If we are correct, however, that no versions of the concept can withstand critical scrutiny, then the implications of this conclusion could be profound. Most importantly, it would mean that our routine practice of procuring organs from patients who have been diagnosed as ‘brain-dead’ violates the dead-donor rule, which stipulates that vital organs can only be procured ethically from dead human beings.

This point was not lost on the President's Council, which to their credit were blunt about the consequences of this conclusion: ‘If indeed it is the case that there is no solid scientific or philosophical rationale for the current “whole brain standard”, then the only ethical course is to stop procuring organs from heart beating individuals’ (emphasis in original). ⁵ As we have discussed, they attempted to side-step this conclusion by offering a defense of determining death by neurological criteria with the new concept of ‘total brain failure’, but unless a more robust defense of this approach is forthcoming it would appear to be seriously flawed, bringing us back to the conundrum they pose.

We believe that the outcome most feared by the Council is indeed unavoidable – the concept of ‘brain death’ cannot underwrite the practice of organ donation if it is to comply with the dead-donor rule. We do not believe, however, that this makes the current practice unethical. More generally, in other work, we provide a detailed account of how the principles of consent and non-maleficience can serve as a foundation for reconstructing the ethics of organ transplantation and end-of-life care. ^21,25–28 Patients who meet the diagnostic criteria for ‘brain death’ have reached a point of no return, from which it is impossible to regain consciousness. As such, they are neither harmed nor wronged when vital organs are procured prior to stopping life support. Moreover, stopping life support will result in their death, making it ethically immaterial that organs are retrieved prior to withdrawing treatment, provided that valid consent is obtained.

To the extent that Thomas Kuhn's views about the structure of scientific revolutions can be applied to ethical controversy, one might draw relevant parallels between the two. ²⁹ Kuhn hypothesized that dominant scientific paradigms can withstand and absorb anomalies and conflicting data up to a certain critical threshold, but beyond that point they are vulnerable to a wholesale revision that shifts the entire foundation of thought. As an ethical construct, we suggest that the concept of ‘brain death’ may be nearing that critical threshold. Unlike the President's Council, however, we believe that this outcome is not to be feared but welcomed, and that this will lead to a more open and transparent understanding of the ethics of organ transplantation and end-of-life care, one that will better serve us in the decades to come.

Conflicts of interest: The authors report that they have no conflicts of interest to report related to the content of this work.

Declarations: This paper has not been nor will be published elsewhere. The opinions expressed are the views of the authors and do not necessarily reflect the policy of the National Institutes of Health, the Public Health Service or the US Department of Health and Human Services.