Introduction

The term “brain death” (BD) encapsulates two correlated concepts: the death of the brain and the patient’s death certified by neurological criteria. In this sense, it is the basis of the current policies of organ retrieval for transplantation from brain-dead heart-beating donors. Several authors have challenged the fact that medicine can demonstrate the righteousness of such concepts. This situation is difficult to justify on medical grounds and could contribute to public confusion or disquiet. This review will summarise the most important points of disagreement, with the intention of stimulating a debate within the scientific community.

A concise historical review

In the early 1950s, mechanical ventilators made it possible to support respiration in patients with irreversible total brain damage. This gave rise to questions about the timing of death and the duty of care of clinicians towards such patients. In 1957, Pope Pius XII stated two principles: that resuscitation techniques, being ‘extra-ordinary’ means, could be withdrawn before circulatory arrest occurred and that the precise moment of death cannot be deduced from any religious or moral principle, and is therefore a matter for the competence of clinicians [1]. Two years later, Jouvet suggested the possibility of diagnosing the death of the central nervous system using the electroencephalogram (EEG) [2]. In the same year, Mollaret and Goulon described the clinical picture associated with this condition [3].

Subsequent discussion started to consider the ethical and legal aspects of irreversible coma [4]. Progress in surgical techniques and immunosuppression and the diffusion of renal dialysis (which decreased the urgency for kidney transplantation) all contributed to a growing demand for organs which living, related donors could not supply. This led to consideration of brain-dead patients as a potential source [5, 6, 7].

In autumn 1967 the first heart transplantation took place. One month later an “Ad hoc Committee to Study the Problems of the Hopelessly Unconscious Patient” was convened at Harvard Medical School. By the end of this work, it changed its title to “Ad hoc Committee to examine the Definition of Brain Death” [7]. Its report defined “irreversible coma as a new criterion for death” (Appendix, quotation 1): it described the signs of BD, the appropriate procedures to declare its presence, the most relevant legal issues and the justifications for this new criterion [8].

A rational approach to BD was developed by Capron (1978) [9] and Bernat (1981) [10], in order to demonstrate the equivalence between cardiorespiratory and neurological standards for determining death.

In 1981, the President’s Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioural Research proposed two alternative ways to justify BD: the “primary organ” concept, according to which cardiorespiratory functions are mere prerequisites for brain function (Appendix, quotation 2) [11] and the “loss of integrated functioning of the organism”. The Commission also produced a model statute, the Uniform Determination of Death Act (UDDA) (Appendix, quotation 3).

In less than 20 years, nearly all developed countries enacted laws or statutes that encapsulated the principles of the report.

A continuing controversy

The most important controversies regarding BD are reported below, grouped according to issues for convenience of presentation.

The utilitarian aspect of the brain death concept

Some authors criticised the strict utilitarian connection between transplant policies and BD [7, 12, 13], whose definition appears to be “the product of conceptual gerrymandering to solve a social problem” [12], or “a political decision (...) prompted by a growing need for organs for transplantation” [13]. Such an “inherently unstable” [12] utilitarian position could lead to attempts at stretching the definition, as has already happened [14, 15, 16].

All these statements might be somewhat morally disturbing. Nevertheless, they do not mean that the BD concept is per se scientifically untenable.

Definitions and criteria

A primary problem is that BD is defined in two different ways: in the USA and in most European countries a “whole brain” definition is adopted, while in the UK a “brainstem” definition is used (Table 1) [17]. Besides, major diversities exist among different countries in the procedures for diagnosing BD (Table 2) [18].

Table 1 Different definition of brain death (from Young [17])
Full size table
Table 2 Some of the most important differences in criteria for determining brain death in Europe and in Canada, USA, Australia, New Zealand and Japan, as derived from Wijdicks [18]
Full size table

Perhaps more pressing is the discrepancy between the “whole brain” definition (Table 1) and the correspondent criteria, which demonstrate the irreversible cessation of only a part of all known intracranial functions [19, 20, 21, 22, 23, 24, 25]. Actually, the activity of the pituitary gland, the control of cardiovascular tone and thermoregulation are virtually ignored.

It is not clear why neurological signs of BD were privileged, compared to the neuro-endocrine and autonomic ones. As a matter of fact, neither indispensability for biological survival [25] nor possibility of being replaced can be assumed as definite reasons for such a choice. Even control of ventilation, which can be considered the most vital intracranial function, is replaceable by either artificial ventilation or diaphragmatic pacing. Not surprisingly, the variable persistence of neuro-endocrine and autonomic intracranial functions has been described in many brain-dead patients [19, 21, 23, 24, 25, 26, 27, 28, 29, 30]. Furthermore, some residual electrical activity has also been demonstrated in conditions consistent with BD [19, 20, 21, 23, 24, 29, 30, 31].

Acknowledging that “the designation ‘whole brain’ death is an approximation”, Bernat suggested the identification of BD as the destruction of a theoretically determinable critical neuronal population [29] and quoted Pallis’ proposal of defining BD as the permanent cessation of functioning of the brain as a whole (not of the whole brain) [32]. Yet, how much of the brain must die has neither been clearly stated nor incorporated in the definition.

All these problems are attenuated by the necessity to demonstrate a sufficient cause of BD and to exclude every confounding factor, as required nearly everywhere.

The purported equivalence between brain death and death

In the last decade, what—with very few exceptions [30, 33, 34]—was an extremely united front of BD supporters started to break up. Some authors challenged the equivalence of BD to the death of the patient [21, 22, 23, 24, 25, 35, 36, 37, 39, 40]. The most authoritative view justifying this equivalence is the theory of “the brain as the central integrator of the body”. BD must be legitimately regarded as death of the individual because it induces a loss of the somatic integrative unit: in BD, the body is no more an integrated organism but a mere and rapidly disintegrating collection of organs which have lost forever the capacity of working as a co-ordinated whole (Appendix, quotation 4) [10]. This rationale is the only one officially accepted [11, 12, 35].

Such a theory has been strongly criticised, as biological death cannot be proven with certainty in BD. Actually, not all brain-dead patients inexorably deteriorate to cardiovascular collapse in a short time and some of them show an adequate level of biological integration for weeks or months (up to more than 14 years in one case) [12, 22, 24, 35, 36, 37, 39, 41, 42]. A brain-dead patient can assimilate nutrients, eliminate wastes, fight infections, heal wounds, carry out a pregnancy and so on.

Somehow, it seems that the “central integrator” theory rests more on what we think should happen to brain-dead patients, according to the definition, than on what can happen in reality. In particular, Shewmon outlined that this theory appears to be an a posteriori philosophical and moral elaboration of what had already been codified into laws and implemented in practice [35]. He confirmed his position by quoting the statement of Cranford and Smith about patients in permanent vegetative state (PVS): “It would be tempting to call [permanently unconscious patients] dead and then retrospectively apply the principles of death, as society has done with brain death” [43].

An alternative rationale could be that BD causes the loss of essential human properties (conscience, reasoning, feelings, memories, ...) and personhood. The permanent loss of these functions would identify the patient’s death (the “cortical death”), even in the presence of some lower neurological activities (breathing and swallowing), of spontaneous but unintentional movements and of thermoregulation [19, 20]. The major problems of this approach are clinical (the certain and permanent absence of all upper neurological functions is very hard to predict [44, 45]), moral (inclusion in BD of patients in PVS, dementia and anencephaly) and procedural (separation of the death of the person from the death of the organism, or the possibility of burying still breathing “cadavers”). This approach is far from finding unanimous favour and has not found any legal application yet.

Organ retrieval without brain death: forward to the past (the non-heart-beating organ donor protocols)

The definition of the vital status of brain-dead donors has been circumvented by the introduction of non-heart-beating organ donor (NHBD) protocols. The Ethics Committee of the Society of Critical Care Medicine (SCCM) recognised two patient subsets [46]: uncontrolled NHBD (organ retrieval following unsuccessful cardiopulmonary resuscitation after unexpected cardiorespiratory arrest) and controlled NHBD (organ procurement following a planned withdrawal of life-sustaining therapies which are considered futile or excessively burdensome by the patient and/or his family). Obviously, as for controlled NHBD patient, the cessation of cardiorespiratory functions is intended as “spontaneously irreversible” [47].

The period of time between asystole and organ harvesting is variable. The Institute of Medicine, after surveying the existing NHBD protocols (waiting time ranging from 0–5 min), recommended a 5-min interval [48]. The SCCM Ethics Committee stated that “no less than two minutes is acceptable, no more than five minutes is necessary” [46].

Actually, at the time of organ retrieval, the donor will probably not be dead according to either of the UDDA criteria (Appendix, quotation 3). After less than 5 min of asystole, irreversible cessation of circulatory and respiratory functions can not be demonstrated: harvested hearts have been successfully implanted [49, 50] and spontaneous restoration of cardiac function after more than 5 min of asystole has been reported [51, 52, 53, 54, 55]. Also, given the short time interval, the irreversible cessation of all intracranial functions can not be claimed as certain [46].

This whole issue is still actively debated [46, 47, 48, 49, 50, 56, 57, 58, 59, 60, 61, 62, 63], especially because of the potential yield of NHBD protocols [50, 60, 63].

Organ retrieval without death: revising the dead donor rule

Some authors deem that current BD criteria, though not satisfactory for the diagnosis of death, are sufficient for organ harvesting from previously consenting patients; consequently, they have suggested abandoning the dead donor rule, which states that patients must be dead before organ retrieval and death must be neither caused nor hastened by the retrieval [12, 21, 22, 24, 40]. Arnold and Youngner, for instance, proposed relying entirely on informed consent and on a “violate-no-interest” approach [12], which is nonetheless highly problematic as it could be considered active killing of a previously consenting patient [21].

Adapting rules and definitions

Different solutions have been proposed in order to reconcile the reality of BD and transplant medicine. Halevy and Brody proposed abandoning the hypothesis of the possible definition of BD as an event that sharply separates life and death at an arbitrary point. Yet, the satisfaction of current tests for BD can be considered an appropriate criterion for organ retrieval [23].

In Denmark, after very large community involvement, the Danish Council of Ethics recommended to maintain the traditional cardiorespiratory criterion for death, but recognised the particular significance of the situation identified by BD (“a condition ... which absolutely excludes the possibility of stopping the death process “ [64]); in this condition, every support should be forgone or maintained to permit organs retrieval: “such an intervention is cause for the conclusion of the process of death, but is not the cause of death” [64]. Yet, the Danish Parliament preferred to base the law on the more customary concept of BD.

On legal grounds, a new German law on transplants allows organ retrieval after irreversible loss of all intracranial functions without expressly recognising that donors are dead [37]. In the USA, laws in New Jersey and New York enable the application of BD definition only when respecting the stated view of the patient [65]. A recent Japanese law provides that only the patients who have agreed to donate their organs can be declared brain-dead [66].

Revising the brain death definition

The above-described state of reflection on BD can appear more complex than the everyday clinical activity in ICUs would suggest. Nevertheless, some authors maintain that there is only a superficial and fragile consensus among health care workers on this issue, beneath which little agreement (and sometimes great confusion) can be found [22, 35, 67].

Our view is that, as was originally proposed by the Harvard Committee, BD is much more a moral than a scientific concept. According to the Committee’s Report (whose only reference was the Pope’s address [1]), there were two reasons to introduce this concept (both clearly moral): to allow withdrawal of life support in irreversibly comatose patients and to provide justification for the retrieval of vital organs (Appendix, quotation 1). And Henry Beecher, chairman of the Committee, stated:

“At whatever level we choose to call death, it is an arbitrary decision. Death of the heart? The hair still grows. Death of the brain? The heart may still beat. The need is to choose an irreversible state where the brain no longer functions. It is best to choose a level where, although the brain is dead, usefulness of other organs is still present. This, we have tried to make clear in what we have called the new definition of death. (...). Here we arbitrarily accept as death, destruction of one part of the body; but it is the supreme part, the brain. (...). Dying is a continuous process; while death may occur at a discrete time, we are not able to pinpoint it.” [68].

The Committee’s reasoning seems to have been the following: in the process of dying it is possible to identify clinically a condition in which patients can be defined irreversibly sufficiently dead in order to forgo life support and to retrieve vital organs. This is clearly a moral proposal, and such it remains even if the definition of the criteria at which time it applies is a scientific matter.

In 1981, Bernat tried to demonstrate that BD scientifically identifies the biological death of the organism, putting forward the theory of the “brain as the central integrator” (Appendix, quotation 4) [10]. Unfortunately, as already seen, such a position is empirically untenable [12, 22, 24, 35, 36, 37, 38, 39, 40].

So, it seems that we are back where we started. Yet, though we are still unable to pinpoint death, in 35 years we have learnt at least three things:

  1. 1.

    In some cases of “brain dead”, some residual intracranial functions can be retained;

  2. 2.

    A level of biological integration can also remain, sufficient for prolonged biological maintenance;

  3. 3.

    Even in such cases, however, recovery of those intracranial functions upon which the BD declaration is based never occurred, confirming that their loss is irreversible.

In our opinion, the first two points are irrelevant to the social and moral acceptability of the original Committee’s position, which is conversely strengthened by the third one. Yet, they strongly question the use of the term “brain death”, as—to be rigorous—the clinical condition they refer to corresponds neither to the loss of all intracranial functions nor to the patients’ biological death (at least not always).

A simple alternative could be to revert to the original term of “irreversible coma” (Appendix, quotation 1) or, more precisely, “irreversible apnoeic coma” (IAC), understood not as equivalent to death, but as describing a particular condition in which life support should be legitimately forgone and organs can be retrieved from consenting patients. This term clearly differentiates IAC from deep coma (in which brainstem functions are maintained) (Table 3), a condition which—if recovery does not occur—typically turns into either a vegetative state, in which wakefulness (eye opening) is regained, or IAC [17]. Furthermore, this definition should have no consequences for transplantation: the patient’s condition, the previously obtained informed consent and the nobility of the action could preserve the process of organ retrieval from such donors as socially acceptable, even admitting the problematic definition of their vital status.

Table 3 Major differences among vegetative state (VS), coma and irreversible apnoeic coma (IAC)
Full size table

In other words, we think that this semantic change, while better describing the situation, neither substantially challenges the status quo nor undermines the righteousness of the traditional criteria, whose usefulness in identifying this condition has been confirmed. Obviously, it has to be adequately explained to lay public.

Maybe the problem could be re-assessed by reconsidering how intensive care alters dying and death. In this sense, the task of the new-born ESICM Working Group on the diagnosis of death and on post mortem medical interventions performed in intensive care units is an essential step in order to (re)define the vital status of these patients, taking account of both residual intracranial functions and biological integration, which can be present in IAC.

Conclusions

Medicine can demonstrate the irreversible loss of cortical and brainstem functions, thus identifying an extremely advanced point of no return in the dying process, the irreversible apnoeic coma (IAC). The definition of the vital status of IAC patients is more a task for society at large than a medical one. Yet, we argue that, in such patients, the retrieval of vital organs can be morally and socially permitted subject to legally acceptable consent.

We believe that, in line with the views expressed by the Danish Council of Ethics [64], this approach creates consistency between current practice and underlying reality, it is clear and understandable to the public and leaves much scope for the involvement of society at large.