Although brain death is a phenomenon of only the last half-century, philosophical and practical issues concerning the definition and determination of human death have been discussed since antiquity. The ancients addressed an important question: what organ of the body represents the essence of life, because, without its function the body is dead? Greek physicians held that the heart was “the seat of life; the first organ to live and the last to die.”¹ Because the heart was the essence of life and created the vital spirits, the absence of heartbeat was regarded as the principal sign of death. Greek and Roman medical experts believed that neither respiration nor brain function was essential to life, although both Hippocrates and Galen held that the brain was the source of reason, sensation, and motion.

The 12th century rabbi and physician-scholar Moses Maimonides can be regarded as the father of brain death because he first argued that a decapitated person was immediately dead, despite the presence of movements of some muscles. For the preceding millennium, ancient Hebrew law (Halachah) had provided that breathing, not heartbeat, was the essence of life and accordingly, cessation of respiration was the defining aspect of death. Maimonides took a step further toward the concept of brain death when he asserted that the “spasmodic jerking” of the decapitated body did not represent evidence of life because the muscle movements were not indicative of a central control.² Thus, Maimonides believed that the central controlling mechanism of locomotion was as essential to life as was breathing. Today, these profound insights remain essential to the concept of brain death.

Throughout the 18th and 19th centuries, the general public expressed fear of incorrect death determination and resultant inappropriate burial of the living. These fears were the subject of several short stories by Edgar Allen Poe³ and resulted in the marketing and sale, by the Russian Count Karnice-Karnicki and others, of above-ground bell-ringing and flag-waving devices attached to caskets to detect breathing that could alert people to exhume a recently buried person who was erroneously declared dead and buried. The omnipresence of the fear of premature burial has continued to impel physicians to design tests of death that have zero false-positive determinations.

The implementation of the positive-pressure mechanical ventilator in the 1950s ushered in the contemporary era of brain death. For the first time, patients who had suffered irreparable damage to the brain that had rendered them permanently apneic could have their heartbeat and circulation sustained temporarily by positive-pressure support of their respiration. The depth of unresponsiveness associated with this profound degree of brain damage had not been observed before because, previously, all such patients had died immediately from apnea.

The first careful descriptions of patients on ventilators who had undergone functional destruction of the brain were reported by French neurologists, neurosurgeons, and neurophysiologists. Mollaret and Goulon coined the term le coma dépassé (a state beyond coma, irretrievable coma) to acknowledge that the constellation of utter unresponsiveness, deep tendon areflexia, cranial nerve areflexia, apnea, and absent electroencephalographic activity was unprecedented in the annals of coma observation. Although initially they did not conclusively assert that such patients were dead, they did believe that le coma dépassé was qualitatively unlike any form of coma previously reported.⁴

A number of isolated case reports and commentaries were published over the next decade that further consolidated the clinical features of the state that came to be called brain death. Most patients reached this tragic state because they were resuscitated and ventilated following severe head trauma, hypoxic-ischemic brain damage from cardiac arrest or asphyxia, or massive stroke. With aggressive ventilatory and other intensive support, physicians could sustain cardiac activity for several days or, more rarely, weeks. But were such patients alive or dead?

In a landmark paper published in 1968, an ad hoc committee of Harvard Medical School faculty, including ten physicians, a theologian, a lawyer, and a historian of science, first formulated criteria to assert that patients with irreversible apnea, areflexia, and complete unresponsiveness from devastating brain damage were legally dead.⁵ Analyses of the writings of the committee chairman, Henry K. Beecher, suggest that the committee was motivated by a desire to permit withdrawal of treatment in hopeless cases and to facilitate organ transplantation. Despite the misleading title of their report, the committee stated that their “primary purpose [was] to define irreversible coma as a new criterion of death.”⁶

Over the succeeding two decades, numerous investigators and commentators added refinements to the concepts and tests proposed by the Harvard Ad Hoc Committee, which further consolidated the clinical findings in brain death and the philosophical basis for regarding brain dead patients as dead.⁷ The attraction of the brain death concept was so strong that only two years after the Harvard report, the state of Kansas enacted the first statute of death incorporating the new concept of brain death, even before it received a rigorous biophilosophical justification.⁸ In 1981, the President’s Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research chose Defining Death as the topic of their first book. The President’s Commission strongly endorsed both the concept of brain death and the desirability of adopting uniform statutes of death among jurisdictions.⁹ They cited the analysis of death authored by my Dartmouth colleagues, Bernard Gert and Charles Culver, and me¹⁰ as the rigorous philosophical justification that brain death was the equivalent of human death.¹¹ A contemporaneous report by the Law Reform Commission of Canada also reached the conclusions that brain death was a valid concept and should be uniformly incorporated into law.¹²

By the turn of the 21st century, the practice of declaring brain death had become so well-accepted throughout the developed world and much of the developing world that only a minority of scholars continued to regard it as a philosophical or ethical controversy. The worldwide acceptance of brain death continues to grow: it is practiced in over 80 countries at present, and even countries such as Denmark, Japan, and Saudi Arabia that previously had questioned or rejected brain death now accept it.¹³

Until the past generation, philosophical discussions of death concerned metaphysical rather than biological considerations. Prior to the advent of positive-pressure ventilators, death was a biologically unitary phenomenon: when one organ system critical to life failed irreversibly, other organ system failures followed rapidly and unequivocal death occurred within minutes. Patients who stopped breathing quickly lost heartbeat and blood pressure. Hypoxic-ischemic destruction of the brain and other organs followed with subsequent rigor mortis, putrefaction, and dissolution of the organism. A similar chain of events could be triggered by loss of heartbeat. One did not have to consider whether an organism that had lost certain vital functions while retaining others was alive or dead because such an occurrence was technically impossible.

With the development of the positive-pressure mechanical ventilator and characterization of le coma dépassé, the life status of the “brain-dead” patient became ambiguous. Brain-dead patients possess certain characteristics previously associated with life: they are warm, have spontaneous heartbeat and systemic circulation, produce urine, and absorb and metabolize administered food. Yet they also possess certain characteristics previously associated with death: they are utterly unresponsive to even the most noxious stimuli, are apneic, make no spontaneous movements, and have few internal mechanisms of physiologic homeostasis. Are such patients alive or dead?

The new ambiguity in death determination was a technological artifact. Technology had produced an example showing us that our previous understanding of death was ambiguous and incomplete. Now that it was possible for a patient to have lost one set of vital functions (brain functions) while retaining others (circulation, albeit mechanically supported), physicians alone could no longer determine death. Physicians could not develop and apply clinical tests and examination procedures to diagnose death at the bedside because it was unclear what the tests were intended to measure. A concept or definition of death had to be agreed upon before the tests to measure it could be formulated and used.

A rigorous philosophical analysis of death should be conducted in four sequential phases that proceed from the conceptual and intangible to the measurable and tangible.¹⁴ First, scholars must formulate the paradigm of death by agreeing on the nature and boundaries of the topic to be discussed. Second, philosophers must propose a concept or definition of death. The task is to analyze our intuitions, attitudes, and practices of death determination, and from these, distill the essence of the ordinary meaning of the word “death” that has been rendered ambiguous by technology. In short, this is the philosophical task of making explicit the traditional implicit concept of death that we all hold. The third task is the philosophical and medical task of identifying the criterion of death. The criterion of death is the general, measurable standard of death that shows the definition has been fulfilled by being both a necessary and sufficient condition for death. The criterion can be used in a statutory or operational definition of death. Finally, physicians have the medical task of developing and validating a battery of bedside tests, operations, and procedures to show conclusively that the criterion has been satisfied.¹⁵ With a few exceptions, most scholars in this area have accepted our method of analysis, if not our conclusions.¹⁶

The next task is for physicians to devise sets of tests and procedures to show that the criterion of death has been satisfied. Most patients who die are not mechanically ventilated prior to death. In such instances, “cardiopulmonary” tests that show the permanent absence of breathing and circulation are sufficient. Brain death tests are necessary only for the patient receiving mechanical ventilation.

Numerous sets of brain death tests have been studied since the 1968 Harvard Ad Hoc Committee Report, but those proposed by the 1981 President’s Commission medical consultants have achieved the widest acceptance.⁴⁸ In 1995, these tests were subjected to an evidence-based analysis by Eelco Wijdicks, which culminated in a practice guideline published that year by the American Academy of Neurology.⁴⁹ The tests were delineated in such a way as to eliminate the possibility of false-positive determinations, even though doing so generated the probability of some false-negative determinations. The tests have been validated thoroughly. There has never been an established case in which a patient fulfilled properly performed and interpreted tests for brain death and recovered at all. This validation is in ironic contrast to the occasional instance in which a physician incorrectly diagnoses death using cardiopulmonary tests.

The test batteries in use today are essentially identical, requiring a demonstration that all clinical brain functions are absent and that
the loss of functions results from an irreversible structural lesion, and not from a potentially reversible metabolic or toxic condition.⁵⁰ In practice, these goals are accomplished by performing a bedside examination showing utter unresponsiveness to noxious and all other stimuli, true apnea (tested correctly), and the absence of all reflexes served by the brain stem and cranial nerves, such as pupillary, corneal, gag, cough, and vestibulo-ocular reflexes. Deep tendon reflexes and other reflexes mediated by the spinal cord may persist. Any possible contribution to absent brain functions resulting from metabolic and toxic disorders must be excluded, such as depressant drug intoxication, the use of a neuromuscular junction blocking agent, hypothermia, and shock.

The absence of clinical brain functions must persist for a predetermined time, after which a second examination is performed to demonstrate the continued absence of clinical brain functions. It is presumed that clinical brain functions were absent throughout the interval between examinations. The necessary duration of the interval between serial examinations varies as a function of the age of the patient, the disorder producing brain death, and the use of confirmatory laboratory tests. In general, the interval is longer in younger patients and in those whose disorders were produced by hypoxic-ischemic damage. The specific methods and procedures of the bedside tests and of the laboratory confirmatory tests for brain death are beyond the scope of this chapter, but are available in standard texts.⁵¹

The moment of death is the time at which the patient fulfills the second set of brain death tests. Until that time, the patient is considered incipiently brain dead. Although such a practice at first appears arbitrary, declaring death at that time is consistent with the practice of determining death using cardiopulmonary tests. A physician called to the bedside of a hospitalized patient who has been found dead in bed will declare the time of death to be the time he examined the patient and ascertained a prolonged absence of breathing and heartbeat. The patient may have died several hours earlier. The somewhat arbitrarily delineated moment of death shows that brain death can be determined only in retrospect.⁵²

Electrophysiologic and neuroimaging tests for brain death can be performed to expedite and confirm the clinical determination. These tests are useful in several circumstances: (1) to facilitate timely organ procurement; (2) to make the diagnosis when a complete clinical examination cannot be performed; and (3) to assist less experienced examiners in making the diagnosis. An adequate electrophysiologic determination includes EEG, brain stem auditory evoked potentials, and somatosensory evoked potentials, all of which must be absent. It is preferable to perform a study showing the absence of intracranial blood flow because this finding proves the irreversibility of the process.⁵³ Imaging studies to prove absent intracranial blood flow include radionuclide angiography, transcranial Doppler ultrasonography (TCD), magnetic resonance angiography or perfusion scanning, computed tomographic angiography, and SPECT.⁵⁴ Most physicians declaring brain death rely on a confirmatory study. Radionuclide angiography is ordered most commonly but younger physicians prefer TCD.⁵⁵ There is evidence that neurosurgeons and critical care physicians conducting brain death determinations rely more on intracranial blood flow tests and neurologists rely more on clinical tests.⁵⁶

The determination of brain death in young children and neonates differs from that in adults in one critical respect: how thoroughly the bedside tests for brain death have been validated in this age group.⁵⁷ The Medical Consultants for the Diagnosis of Death of the President’s Commission cautioned that the tests for brain death had not been validated sufficiently in children under five years of age.⁵⁸ Subsequently, the Task Force for the Determination of Brain Death in Children was impaneled from representatives of the American
Academy of Neurology, the American Neurological Association, and the American Academy of Pediatrics to study the available data on brain death in children and neonates and make recommendations for physicians.

Their 1987 report Guidelines for the Determination of Brain Death in Children permitted brain death declaration in any full-term infant over the age of one week. Children over the age of one year were treated exactly the same as adults. Children from two months to one year must have a confirmatory test such as electroencephalogram or radionuclide intracranial blood flow study in addition to fulfilling the clinical tests showing absence of all clinical brain functions for an interval of 24 hours. Infants from one week to two months must fulfill the same clinical and confirmatory test criteria over an interval of 48 hours.⁵⁹

The Guidelines generally were accepted by neurologists and pediatricians, except for those that pertained to determining brain death in infants under the age of two months. Joseph Volpe pointed out the practical limitations of attempting to determine brain death in neonates. Most neonates with this degree of brain damage have suffered perinatal injuries, especially diffuse hypoxic-ischemic insults. Often, there is not sufficiently accurate historical information about the duration and severity of this insult to confidently predict irreversibility. Second, the role of hypotension cannot be fully appreciated. Third, brain imaging techniques cannot accurately quantify the extent of brain damage. Thus, the essential criteria of irreversibility and totality of functional loss often cannot be met.⁶⁰

Alan Shewmon pointed out that the false-positive error rate in neonatal brain death determination is no less than 0.02. He showed that even with more recorded experience, this error rate will not fall significantly. This problem, he argued, results from reliance for the declaration of brain death on tests of brain function rather than on tests of brain structure. Shewmon believes that neonatologists should rely only on unequivocal evidence of structural, irreversible brain damage, such as completed rostrocaudal, transtentorial brain herniation.⁶¹

From a practical perspective, the principal indication for formal brain death determination in a neonate or young infant is to permit the brain-dead infant to serve as an organ donor. The determination of brain death never has been a prerequisite for termination of medical treatment in the setting of hopeless brain damage. (See discussions on withholding and withdrawing life-sustaining treatment in chapters 8, 12, and 13.) Reports of the experience of conducting brain death determination in pediatric and neonatal intensive care units show results similar to those in adult ICUs,⁶² unfortunately, including the same disappointingly high degree of noncompliance by physicians in using established test batteries.⁶³ Recent experiences have been published describing the challenging medical and ethical issues involved in pronouncing brain death in children.⁶⁴

The final task is to draft a statute of death recognizing the irreversible cessation of brain functioning as the legal criterion of death. Statutory recognition of brain death is desirable because it both acknowledges biological reality and provides physicians the authority to determine death when patients are receiving ventilatory support. Since 1970, when Kansas became the first state to incorporate brain death in a statutory definition of death, all other states in the United States have enacted statutes of death incorporating brain death, except New York, which issued an administrative regulation legalizing brain death.⁶⁵ By the turn of the 21st century, nearly all Western and developed countries had enacted brain death laws and many other countries also had done so.⁶⁶

An ideal statute of death should accurately recapitulate the whole brain concept of death and offer practical guidance to physicians regarding when the brain death tests should be employed and when the cardiopulmonary tests are sufficient. An ideal statute first should state the criterion of death as the permanent cessation of the clinical functions of the brain. Then it should make clear that death can be determined in one of two ways: in the patient without mechanical ventilation, by showing the permanent cessation of respiration and
circulation; and in the patient with mechanical ventilation, by the permanent absence of clinical functions of the brain.

In 1972, Alexander Capron and Leon Kass designed a model death statute incorporating brain death, which they revised in 1978.⁶⁷ This statute created two equal criteria of death instead of explaining that death had a single criterion but could be determined in two ways. In 1975, the American Bar Association created another model death statute with a correct single brain criterion that failed to clarify that most determinations of death rely on permanent cessation of respiration and circulation.⁶⁸

By far, the most influential American model death statute was published in 1981 by the President’s Commission in their work Defining Death. Their model statute, the Uniform Determination of Death Act (UDDA), was proposed for adoption by all American jurisdictions. With the assistance of the National Conference of Commissioners of Uniform State Laws, the UDDA eventually was adopted by about half of states in the United States as their statute of death and most of the others adopted a variation of it. The UDDA provides:

In 1981, Charles Culver, Bernard Gert, and I drafted a model statute clarifying the criterion and tests of death, which we modified in 1982 to incorporate language compatible with the UDDA.⁷⁰ We criticized the UDDA as incorrectly elevating the tests of death into criteria of death rather than maintaining a single whole brain criterion with two tests. Our model statute provides:

Our statute has the advantage of conceptual clarity and practical utility. It clarifies that irreversible cessation of brain functioning is the sole criterion of death, but that the criterion can be determined by two separate tests, depending upon the clinical circumstances. Although this statute represents an improvement over the UDDA, states choosing a brain death statute in the future probably should select the UDDA. At this point, the UDDA’s practical advantage of obtaining uniformity in statutory language among states probably outweighs our statute’s abstract advantage of greater conceptual clarity.

A group of scholars has opposed the concept of brain death since it was first introduced in the 1960s.⁷¹ I discussed the opposition to the whole brain criterion by the advocates of the higher brain formulation. Despite Robert Veatch’s exaggerated claims that the concept of whole brain death is in “impending collapse” or is “dead” itself,⁷² after over three decades of writings, the higher brain theorists have failed to convince any medical societies in the world to change their criterion of death or any jurisdictions to convert their brain death laws from the whole brain to the higher brain formulation.

During the 1990s, a new group of opponents emerged to refute not simply the whole brain criterion but the entire concept of brain death. The current opponents argue three positions: (1) brain death is philosophically incoherent as a concept of human death; (2) brain death is a legal fiction that once may have been justified but now no longer serves a useful social purpose; and (3) brain death is incompatible with the belief systems of Christianity and Judaism.