Brain death, states of impaired consciousness, and physician-assisted death for end-of-life organ donation and transplantation

Neurological criteria of death

Historically, the absence of comprehensive criteria for death has led society to rely on a definition based on cardiorespiratory criteria (i.e., the prolonged absence of arterial pulse and respiration), which clearly delineate the line between being alive and being dead.

Technologic advances in life-support systems have made it possible to procure and transplant viable organs from patients who are in irreversible coma and apnea but have retained spontaneous functions of the heart and circulation. In recognition of this reality, a broader definition of death, i.e., brain death, was legislated in the UDDA (President’s Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research 1981). The UDDA explains that a person is determined to be dead upon sustaining either irreversible cessation of circulatory and respiratory functions or irreversible cessation of all brain function, including that of the brain stem (President’s Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research 1981). This determination must be made in accordance with accepted medical standards.

Since human death is a single phenomenon, a definition of death should incorporate contemporary knowledge and an understanding of human biological processes (Fig. 2). Prolonged arrest of blood flow causes irreversible ischemia (and necrosis) of the whole brain and cessation of all integrated neurological functions. From a neurophysiologic perspective, a minimum cerebral perfusion pressure of 15 mmHg is required to sustain blood flow and oxygen supply to the brain (Ivan and with contributions by Melrose 2007). Cerebral perfusion pressure is the gradient between mean arterial and central venous pressures subtracted from the intracranial pressure. The amount of brain necrosis that occurs when the cerebral perfusion pressure falls below this level determines the reversibility or finality of coma in human beings. Absent arterial blood inflow to and venous outflow from the brain on four-vessel cerebral angiography demonstrate complete cessation of intracranial circulation. Prolonged cessation of intracranial circulation eventually results in necrosis and irreversible cessation of integrated neurological functions in heart-beating brain-dead patients.

Fig. 2

Human death is a single phenomenon. Human death is a single phenomenon occurring gradually as a process over time. There is a gradual loss of capacity for somatic integration of the whole body because of an irreversible cessation of all vital and biological functions including circulation, respiration (controlled by the brain stem), and consciousness. The irreversibility of cessation of circulatory and respiratory functions is inter-linked to the onset of whole brain necrosis. Arbitrary neurological criteria and circulatory criterion redefining human death in states of impaired consciousness enable heart-beating and non-heart-beating procurement of transplantable organs, respectively

Although brain-dead patients have no intracranial circulation and (by definition only) irreversible loss of spontaneous respiratory drive and consciousness, the heart and whole body circulation continue to function spontaneously in these patients, i.e. without the use of vasoactive medications for hemodynamic support. Spontaneous circulation maintains many integrated biological functions in brain-dead patients that are indistinguishable from living human beings, and in some cases these patients can survive on mechanical ventilation for years (Shewmon 1998). Some of the biological functions include wound healing, body temperature regulation, growth to puberty, reproduction, successful completion of pregnancies and delivery of healthy infants (Truog 2007). Therefore, the concept of death based on only neurological criteria—i.e. irreversible cessation of all brain functions or brain death—does not fully encompass the notion of irreversible loss of integrative unity of the organism and its regulatory functions that are essential for life (Joffe 2007b; Karakatsanis and Tsanakas 2002; Shewmon 2001; Zamperetti et al. 2004). In its white paper entitled “Controversies in the determination of death,” The President’s Council on Bioethics sets out to rebut exactly that position and proposes a new approach to answering the question of whether a diagnosis of “whole brain death” means that the human being is dead (The President’s Council on Bioethics 2008, p. 10). The Council proposes to replace the term “brain death” with the term “total brain failure” for the clinical diagnosis that underlies the current neurological standard (p. 12). The Council recognizes that the central question of whether total brain failure equates with the death of the human being “cannot be settled by appealing exclusively to clinical or pathophysiological facts” (p. 49). In the absence of sufficient empirical evidence for the concept of brain death, to salvage the concept of brain death and to continue support of the current practice of procuring organs from heart-beating donors, the Council proposes to ground this concept in a completely new philosophical rationale. This rationale has not yet been the subject of public debate. It acknowledges that “A human being whose death has been determined according to a neurological standard is the ideal source of transplantable organs” (p. 8). It argues that living organisms preserve themselves and, for that, “must—and can and do—engage in commerce with the surrounding world” (p. 60). “If there are no signs of consciousness and if spontaneous breathing is absent and if the clinical judgment is that these neurophysiological facts cannot be reversed…a once-living patient has now died” (p. 64). The Council’s final conclusion is that based on this new rationale the current neurological standard for declaring death is still defensible. However, some Council members have expressed dissent (personal statements published in the white paper pp. 95–100 and pp. 107–119) on the proposed philosophical rationale equating total brain failure (brain death) with human death. In a separate commentary, Shewmon challenges the validity of the critical role of spontaneous breathing in defining living organisms (Shewmon 2009). The inner drive to breathe is absent not only in total-brain failure patients, but also in conscious patients with lower brain stem lesions and during sleep in patients with Ondine’s curse. Therefore, brain death as a state of impaired consciousness and apnea continues to be challenged as a valid concept of human death because of continued insufficient scientific evidence and a less than convincing philosophical rationale. Instead, as critics have postulated, the moment when a living organism is dead, and hence no longer alive is, conceptually, the moment when there is an irreversible cessation of integrative unity of the whole living organism (Byrne and Weaver 2004; Joffe 2007a; Maruya et al. 2008).

Defining death by neurological criteria has additional conceptual implications. The reduction of any definition of death to exclusively neurological terms ignores the anthropologic, cultural, and religious dimensions that many people value highly. The Committee on Increasing Rates of Organ Donation recognizes the role that cultural and religious beliefs play in consenting to organ donation (Committee on Increasing Rates of Organ Donation-Board on Health Sciences Policy-Institute of Medicine 2006). Cultural and religious traditions and beliefs about the treatment of the dead body, beliefs about life after death, and fears of mutilation can also influence decisions about organ donation (p. 35). Policies and practices for procurement of organs must be compatible with conditions deeply rooted in cultural, religious, and legal traditions (p. 4). These traditions, however, greatly vary among global communities and pluralistic societies. Reduction of the definition of death to brain-based criteria ignores that the concept of death is not simply bioethical or biomedical in nature but fundamentally shaped and driven by a series of important sociologic influences (Kellehear 2008). Even if there is medical consensus on brain-based criteria and determination of brain death, this consensus is insufficient to conclude that “brain death” is in fact death (Joffe 2007b). To the contrary, for many health care professionals and the general public, the concept of brain death is becoming increasingly abstract and socially disconnected from the nature of death (Kellehear 2008). This paradoxical death, a brain-based determination of death with the physical image of a normally functioning body, creates emotional and cognitive conflicts for many health care professionals and family members (Long et al. 2008).

At the November 9, 2007, meeting of the President’s Council on Bioethics, Shewmon (The President’s Council on Bioethics 2007b) pointed to the growing consensus that the neurological standard for determining death has become insufficient to appropriately and comprehensively explain brain death:

What has actually happened in the history of this topic is [that] in 1968 we start with the practice. Then there is a revision of statutory laws. Then there is an attempt to come up with diagnostic standards. Then there is a scramble to find rationales for the statutory laws, and there is incoherence and lack of consensus about why destruction or total brain failure, whatever you want to call it, should be death. And so the actual history of brain death has followed exactly the opposite sequence of events that ought to characterize an ideal major socio-legal medical change. So I think at this point in time it’s going to be very hard to change how transplantation is done because it’s already so ingrained. (The President’s Council on Bioethics 2007b)

Accuracy of determining the state of “brain death”

The concept of brain death has been defined as an irreversible state of impaired consciousness diagnosed by universally approved criteria. However, reaching consensus on the “moment of death” can be time consuming. Proponents also point out, what makes the determination of death accurate is the accuracy of following the rules: the practice guidelines that professional associations established for determining brain death (Ivan and with contributions by Melrose 2007). In 1995, The American Academy of Neurology published clinical guidelines for the clinical determination of brain death. These clinical guidelines include: (1) demonstration of coma; (2) evidence of the cause of coma; (3) absence of confounding factors, including hypothermia, the presence of drugs, or electrolyte or endocrine disturbances; (4) absence of brain stem reflexes; (5) absent motor responses; (6) apnea; (7) a repeat evaluation in 6–72 h, with the time based on the patient’s age; and (8) confirmatory tests only when specific components of the clinical testing cannot be reliably evaluated (The Quality Standards Subcommittee of the American Academy of Neurology 1995). However, over the past decade, critics have increasingly scrutinized the scientific validity of these clinical guidelines for demonstrating that all brain functions have ceased irreversibly (Joffe 2007b, 2009; Karakatsanis 2008; Karakatsanis and Tsanakas 2002; Shewmon 1997).

The clinical guidelines established for brain death determination and accepted as the medical standard (The Quality Standards Subcommittee of the American Academy of Neurology 1995; Wijdicks 2001) ought to confirm the irreversible cessation of all functions of the entire brain, including the brain stem, before organ donation (President’s Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research 1981); they do not do so. The ‘accepted medical standard’ must determine with clinical certainty that the brain injury has reached the endpoint of a process of self-perpetuating destruction of neural tissue (The President’s Council on Bioethics 2008, p. 38). Clinical and histopathologic observations support that there are serious flaws in the clinical criteria used to declare brain death for heart-beating organ donation. Several critical brain structures remain viable and continue integrated neurological functioning after clinically determined brain death. These include, among others, electroencephalogram activity, brain stem auditory and/or somatosensory evoked potentials and hypothalamic functions (Joffe 2009). First, clinical observations suggest that heart-beating organ donors have residual brain functions, including hormonal and neural responses to nociception and pain during the procurement process (Joffe 2007b). Surgical procurement, which is performed on donors without general anesthesia (Keep 2000), induces hemodynamic responses in donors that are similar to the responses of living organisms in distress (Young and Matta 2000). Second, histopathologic observations in one study suggest that, even when the clinical guidelines of brain death determination are applied appropriately, more than 60% of heart-beating donors have no or minimal structural disruption of the brain stem at autopsy (Wijdicks and Pfeifer 2008); the absence of neuropathologic features of ischemia or necrosis can also suggest reversibility of ceased brain stem functions. In this particular study, most of these donors were young patients who were declared brain dead by clinical examination and within 24 h of blunt-force traumatic brain injury (Wijdicks and Pfeifer 2008).

Reliance upon clinical examination to demonstrate irreversible loss of all brain functions is problematic for the following reasons: (1) the short interval between the initial acute brain injury and clinical declaration of death cannot exclude reversible neurological findings that are masquerading as brain death; (2) no confirmatory test is mandated to validate the complete absence of perfusion pressure and cessation of blood flow to the whole brain; and (3) brain stem ischemic changes, which are structural surrogates of irreversibility, are not always present on autopsy (President’s Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research 1981). The likelihood of incorrect declaration of brain death increases when the timeline necessary for determining irreversible cessation of critical brain functions is shortened because of early organ procurement. Transplant advocates have recommended against performing confirmatory tests when declaring brain death in order to avoid a delay or deferral of organ donation (Greer et al. 2008b). Not performing confirmatory tests may also lead to catastrophic errors in the clinical determination of brain death. The absence of neuropathologic findings of profound brain stem ischemia can suggest reversible causes of coma or perhaps retained neurological activity undetected by clinical examination (Walker 1978). In a Canadian survey, 46% of neurosurgeons considered the absence of moderate structural damage in both the brain stem and cerebral cortex on brain autopsy to be incompatible with a clinical determination of brain death (Joffe et al. 2007). More than one-third of the neurosurgeons surveyed also considered that some blood flow to the brain or a brain stem with minimal microscopic damage is incompatible with a clinical determination of brain death. Accepting questionable clinical guidelines as the medical standard for early declaration of brain death and organ donation can have fatal consequences when patients whose condition may be salvageable, i.e., amenable to treatment, are determined to be brain dead (Lifesitenews.com 2008; Morales 2008; The President’s Council on Bioethics 2008).

It is commonly claimed that, despite unresolved and debated issues about clinical determination of brain death, both the concept and the practice of declaring brain death for organ donation and transplantation have gained wide acceptance in most parts of the world. The development of and adherence to practice guidelines to determine brain death are believed to have elicited universal compliance (Ivan and with contributions by Melrose 2007). Recently, the inconsistency of the medical standard for the clinical determination of brain death for organ donation has been highlighted at leading US neurological hospitals (Laureys and Fins 2008). This inconsistency has opened the door not only for loosening the accepted medical standard for the determination of brain death, but also for potentially sacrificing neurologically salvageable individuals for the sole purpose of organ procurement. In a study by Mathur et al. (2008), organs were procured from 142 pediatric patients who were heart-beating donors, based on brain-death declaration, between January 2000 and December 2004 in southern California. The authors reported their study findings, which had been based on the medicolegal standard: “If it’s not documented, it wasn’t done.” One of 294 neurological examinations documented completion of all the elements required in clinical brain-death examination, 26% had the apnea test correctly performed, 15% had at least two examinations performed at the recommended time intervals, and 58% had cerebral angiography as a confirmatory test. Therefore, in this study, a significant proportion of donors may have been incorrectly declared as brain dead. Similar observations have been noted among adult donors as well; more than two-thirds of leading US neurological hospitals vary widely in their compliance with clinical guidelines for determining brain death before organ donation (Greer et al. 2008a).

Circulatory criterion of death

With the growing need for transplantable organs far exceeding the number of organs procured from brain-dead donors, an alternative criterion to declare death based on cessation of circulation was incorporated in the Pittsburgh NHBOD protocol (DeVita and Snyder 1993). The circulatory criterion to declare death can be used for procuring transplantable organs from patients with other clinical states of impaired consciousness but who cannot be declared clinically brain dead (Fig. 1). Immediately after the introduction of the NHBOD protocol, the criterion of 2 min of circulatory arrest for declaring cardiorespiratory death and commencing organ procurement became controversial (Lynn 1993). Circulatory arrest is determined by the loss of arterial pulse. The Institute of Medicine published a report on the practice and protocols of NHBOD in the United States (Committee on Non-Heart-Beating Transplantation II-The Scientific, Ethical Basis for Practice and Protocols-Division of Health Care Services-Institute of Medicine 2000). Over 90% of NHBOD protocols in the United States allow for organs to be procured at 5 min or shorter time periods following circulatory arrest (Howard M. Nathan-Medscape Transplantation 2005). Recently, Denver Children’s Hospital amended the NHBOD protocol to allow surgical procurement of hearts for transplantation after 75 s of circulatory arrest (Boucek et al. 2008).

Ongoing debates indicate that not only the scientific validity of the circulatory criterion is questionable (Joffe 2007a) but also that brain-based criteria of death contribute little, if anything, to defining the exact moment of death in NHBOD (Shemie 2007). In human beings, a coronary perfusion pressure of 15 mmHg is sufficient for continued viability of the heart muscle and return of spontaneous circulation after initial circulatory arrest (Paradis et al. 1990). Coronary perfusion pressure is the gradient between asystolic (diastolic) arterial and central venous pressure subtracted from the intrapericardial pressure. Autoresuscitation i.e. spontaneous return of circulation and recovery of heart and brain functions (also called the Lazarus phenomenon) has been reported after 10 min of circulatory arrest in human beings (Adhiyaman et al. 2007; Joffe 2007a). The presence of coronary and cerebral perfusion pressures after circulatory arrest can explain the Lazarus phenomenon and autoresuscitation in some of the reported cases. This observation is relevant not only because coronary and cerebral perfusion pressures are related to systemic blood flow, but it is also independently controlled by selective arterial and venous vasomotor tones (vascular smooth muscles contraction) and intrapericardial and intracranial pressures surrounding the heart and brain, respectively (Rady et al. 2007). In spite of this physiologic phenomenon, the Institute of Medicine has relied on the loss of systemic arterial pulse and circulation for 2–5 min as the exclusive circulatory criterion for determining death in NHBOD and with no requirement for the complete absence of coronary and cerebral perfusion pressures (Committee on Increasing Rates of Organ Donation-Board on Health Sciences Policy-Institute of Medicine 2006). Neuropathologic features of ischemia or necrosis of the whole brain (and brain stem) become integral observations to establish, with clinical certainty, irreversible apnea and unconsciousness when determining death by the circulatory criterion in NHBOD. The President’s Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research has recommended that circulatory arrest time should be longer than 15 min for the onset of brain ischemia and to the point that irreversible cessation of brain functions (including respiratory function of the brain stem) is certain (President’s Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research 1981). The criterion of a circulatory arrest time between 75 s and 5 min is unlikely to result in irreversible cessation of all integrated neurological functions and respiratory drive in NHBOD. Extracorporeal circulation and cardiopulmonary bypass initiated in NHBOD after 5 min of circulatory arrest invariably reanimate donors during surgical procurement of organs (Dejohn and Zwischenberger 2006). From a legal perspective on determining death with cardiorespiratory criteria, the irreversibility of cessation of circulatory function is interlinked with the irreversibility of cessation of respiratory function (of the brain stem) (National Conference of Commissioners on Uniform State Laws 1981). This point must be emphasized because the UDDA considers human death as a single phenomenon whether determined by neurological criteria or by cardiorespiratory criteria.

Considering death to be the total cessation of life processes characteristic of living organisms throws the practice of applying either circulatory or neurological criteria for declaring death for organ donation into question (Fig. 2). The exact moment when loss of circulation and loss of somatic integration occur is not known and, therefore, the concept of brain death adds little, if any, relevant information about how to determine the precise moment of death in human beings.

Accuracy of determining other states of impaired consciousness

Other states of impaired consciousness, besides brain death, include coma, akinetic mutism (locked-in syndrome), minimally conscious state, and vegetative state. Since 1997, pressure has been growing to expand the recovery of transplantable organs from patients with other states of impaired consciousness, such as those in a vegetative state (Hoffenberg et al. 1997). In September 2007, The President’s Council on Bioethics discussed the Draft White Paper on neurological determination of death and renewed the interest in organ procurement from vegetative state to meet an increasing demand for transplantable organs (The President’s Council on Bioethics 2007a). The renaming of the neurological standard for determining human death from “brain death” to “total brain failure” preempted the Council’s discussion expanding the recovery of transplantable organs from other states of impaired consciousness. In August 2008, a perspective roundtable discussion of organ procurement for transplantation echoed similar interest to abandon traditional neurological criteria for determining death (The New England Journal of Medicine (online) 2008; Truog and Miller 2008).

As highlighted in previous sections of this paper, several shortfalls exist regarding the scientific validity of the concept of brain death. The unassailability of the clinical guidelines for the determination of not only brain death but also that of vegetative state has also been called into question. The American Academy of Neurology has described vegetative state with the following criteria: (1) no evidence of awareness of self or environment and an inability to interact with others; (2) no evidence of sustained, reproducible, purposeful, or voluntary behavioral responses to visual, auditory, tactile, or noxious stimuli; (3) no evidence of language comprehension or expression; (4) intermittent wakefulness manifested by the presence of sleep-wake cycles; (5) sufficiently preserved hypothalamic and brain stem autonomic functions to permit survival with medical and nursing care; (6) bowel and bladder incontinence; and (7) variably preserved cranial nerves (pupillary, oculocephalic, corneal, vestibulo-ocular, gag) and spinal reflexes (Report of the Quality Standards Subcommittee of the American Academy of Neurology 1995). The term persistent vegetative state (PVS) is applied when the clinical criteria of vegetative state persist for at least 1 month after the patient has suffered impaired consciousness. The term permanent vegetative state is applied when the clinical criteria of vegetative state persist at least 3 months after nontraumatic brain injury and 12 months after traumatic brain injury in adults and children. Both brain death and PVS are possible outcomes of the comatose state, but, in PVS, the brain stem, which sustains functions such as respiration and circulation, has been spared injury. The American Medical Association has posited that medical treatment, such as artificial nutrition and hydration, may be withdrawn in patients who are in a PVS (Council on Scientific Affairs, Council on Ethical, Judicial Affairs of the American Medical Association 1990). Because these patients may be allowed to die, they may be considered for organ donation; however, unless physician-assisted death is legalized, patients in PVS are ineligible to become organ donors (Detry et al. 2008). Although some have advocated the acceptance of the concept of higher brain death so that organs from patients in permanent vegetative state can be used for transplantation (Hoffenberg et al. 1997), this concept of death has not been endorsed, accepted, or legalized in the United States.

It is often postulated that patients in PVS are devoid of conscious content and cognitive and affective functions and that they have no behavioral evidence of awareness of their external environment (Ivan and with contributions by Melrose 2007). A neurophysiologic explanation of impaired consciousness in PVS underpins this premise, although scientific observations and theory may not be capable of providing a complete account of consciousness (Zeman 2001). Critics have countered, however, that the evidence from human brain imaging studies as well as neurological damage in animals and humans suggests that some form of consciousness can survive the brain damage that commonly causes vegetative state (Panksepp et al. 2007). Neuroscientific evidence indicates that raw emotional or affective feelings (primary-process affects) can exist without cognitive awareness of those feelings. The fact that patients in a PVS can have the capacity to experience affective feelings in the absence of any reflective awareness represents a diametrically opposite position to the one claiming that these patients are devoid of conscious content and cognitive and affective functions. Many aspects of human cognition can go on in the absence of reflective awareness. Functional neuroimaging methods have demonstrated that aspects of speech perception, emotional processing, language comprehension, and even conscious awareness might be retained in some patients who behaviorally meet all of the criteria that define PVS (Owen and Coleman 2008). The diagnosis of PVS is made primarily based on clinical judgment and without performing static and functional neuroimaging and electrodiagnostic studies to confirm this diagnosis with certainty (Wijdicks and Cranford 2005). When diagnosing vegetative state, clinicians may not be as meticulous in their thought processes as they should be, particularly because the need for transplantable organs has risen to the level of what has been called a national health crisis. The risk of misdiagnosing another reversible condition as vegetative state is greater because the vegetative state is not as common as are other prolonged states of impaired consciousness. Shewmon has cautioned that the very use of the term vegetative state itself predisposes physicians to sloppy thinking and employing logical fallacies by implying that something is ‘by definition’ what is actually an intrinsically unverifiable hypothesis (Shewmon 2004). The clinical criteria for the diagnosis of vegetative state cannot be proven to be valid beyond doubt. The clinical criteria for the diagnosis of vegetative state are a challenge to the premise that good medical research involves, among other aspects, “hypotheses susceptible of proof and disproof, and a methodology that systematically rules out all other explanations” (Shewmon 2004). The accurate determination of PVS requires, among other criteria, clear and robust standards that can be followed in a consistent fashion by the medical community (Rifkinson-Mann 2003). Currently, there is neither consensus on the criteria that encompass the spectrum of PVS nor agreement on the criteria that distinguish this diagnosis from other states of impaired consciousness (Cusack et al. 2000). One retrospective study looked at the clinical records of 40 patients who had received a diagnosis of being in a vegetative state; 17 patients (43%) were found to be in other states of impaired consciousness, i.e., had a misdiagnosis of being in a vegetative state (Andrews et al. 1996). All of these open-ended questions and uncertainties surrounding the diagnosis of PVS (The House of Representatives of the 144th General Assembly of the State of Delaware 2008), as with brain-dead, indeed make it medically, ethically, and legally impossible to procure organs by following heart-beating procurement procedures in patients who are in a vegetative state. After a careful analysis of various viewpoints, the President’s Council on Bioethics has rejected the concept of higher brain or neocortical failure alone (with normal brain stem functions) for determining human death and procuring transplantable organs (The President’s Council on Bioethics 2008).

The scientific validity and accuracy of clinical guidelines declaring other states of impaired consciousness as unrecoverable within a few days after non-traumatic brain injury have also been called into question. The American Academy of Neurology formulated practice parameters for predicting unrecoverable coma within 3 days of acute brain injury after successful cardiopulmonary resuscitation (Wijdicks et al. 2006). The practice parameters were derived from retrospective analysis of published studies over a time period of 40 years. The predictive accuracy of the American Academy of Neurology practice parameters have not been prospectively validated in either multicenter or multinational studies. Nevertheless, the practice parameters and clinical guidelines declaring states of impaired consciousness or coma unrecoverable within days have become essential prerequisites to facilitate early recovery of transplantable organs (Fig. 1).