“[Magoun’s] The Waking Brain … with its appearance the brain of man and animals ceased to be regarded as a purely passive responding apparatus [but] … it did not attempt to analyse the fundamental forms of human concrete psychological activity.”
—A. R. LURIA, The Working Brain, 1973
“I like my head. How about you? It lets me know I’m alive.”
—BARNEY
Robert Schwab first advised colleagues to turn off a respirator in 1954, based upon EEG and clinical exam findings. The criteria he proposed for the Harvard Report, however, evolved through a series of studies, published through the 1960s, that looked at the outcomes of comatose patients. These studies, and the uses of EEG and examination to guide care for those with severe coma, drew in part upon EEG research on the “waking brain.” Many commentators critical of the Harvard criteria faulted its lack of an empirical basis and failure to explicitly describe a concept of death. However, any assessment of the Report should include attention to the actual care of these patients at MGH, and to Schwab’ s efforts to develop and use the criteria he presented to the Committee.
Schwab was clearly conversant with research and controversies surrounding the use of EEG to explain brainstem and subcortical mechanisms in brain function. He played a central role in the institutional growth and research base of electroencephalography. In 1937, at MGH, he established one of the first, if not the first EEG laboratory ever to be sited in a hospital.1 He was a founding officer (secretary) of The American Society of Electroencephalography, which was established in 1947 with Jasper as its first president, a post Schwab would later assume. Schwab was also the founding managing editor of the journal EEG and Clinical Neurophysiology, through which much of the key work on establishing reticular and subcortical roles in central nervous function appeared.
Schwab did not use EEG in order to advise colleagues to end treatment simply because some mechanism of consciousness was no longer functioning. He wanted to solve a concrete problem for his colleagues: Could interobserver replicable findings indicate when these patients were no longer subjects for medical attention? The interaction of EEG and neurological research, expectations about extraordinariness of care, the specific capabilities and limitations of medical practice in the 1960s, and the behavior of the brain and the body under these circumstances contributed to the possibility that brain death was death. So was brain death a straightforward “mere” fact, or a contexted practice? Brain death, as a President’s Council on Bioethics senior consultant put it, “is far from a mere social construct. It is a biologically well-grounded response to the lived phenomenon of the ‘brain dead’ patient.”2 Medical practices and the empirical strategies from which they emerge as “biologically well-grounded responses,” have to be central to any serious look at social uses, benefits, and implications of these facts. But at the same time, and as I will argue in the following chapter, the social contingency and purposes of medical facts are also crucial to understanding them and bothering to use them. Later bioethics debates often simplified this complexity and took brain death out of its sources.
Very little of the initial April 11 Report draft detailed the criteria of “irreversible coma” but where it did, it featured what Schwab referred to as his “triad” of symptoms describing brain death.3 Two sections took on this task: “Definition of Irreversible Coma” by Beecher, and a subsequent section entitled “Characteristics of Irreversible Coma,” which included material Beecher attributed to Schwab’s work. These few paragraphs contain varying descriptions of the symptoms and signals the criteria were meant to cover. “Characteristics” described Schwab as a pioneer in using EEG to diagnose “brain death.” In a subsequent “discussion,” Beecher lamented using extraordinary means to sustain persons who had “no hope of recovery of consciousness” and stated that the definition of irreversible coma permitted removing extraordinary means with “death to follow.” He then observed that the “moment of death” is seen to “coincide with irreversible coma while the heart continues to beat.” The April 11 draft was addressed to the “Committee to Define Irreversible Coma,”4 while later notes and memoranda describe it as the as “Ad Hoc Committee to Examine the Definition of Brain Death” or “Ad Hoc Committee on Brain Death.”5 When presenting his autopsy study of ninety cases meeting his triad (a study which was cited in this early draft) at a scientific meeting during the Committee’s tenure, Schwab remarked, “we say the time of death is the time the heart stops beating”6 [emphases mine]. Although, as will be seen in MGH medical records, Schwab actually distinguished what he meant by “death” in several ways.
From the outset, then, the Report contained various ways to understand its opening claim that “our primary purpose is to define irreversible coma as a new criterion for death.” The Report, especially in its earlier drafts, alternated between descriptions of a coma that predicted heart-based death and a coma that was death itself. While this became a focus of criticism in the decades following the appearance of the Report, this potential ambiguity, and its evolution in Report drafts, reflects an important moment in medical practice that needs to be understood—as does the discomfort it generated.
Brain death was, and is, bound up with both ontological and prognostic purposes: it describes death and its imminence. The conditions of death before dying brought the tasks of defining death and defining futility closer together, such that they were indistinguishable within the context of care as understood by those who first framed it, and within a set of attitudes described earlier around experimentation and medical knowledge. Commentators later sought to resolve this overlap of descriptive and prognostic purposes by crafting more careful conceptual distinctions between death and dying. As the Report most unambiguously stated, though, its starting point was technically narrower—to describe the “characteristics of a permanently non-functioning brain.” That such a state meant death to these authors was by virtue of Beecher’s pragmatism, Schwab’s empiricism, and experience with comatose bodies in the hospital, which is the focus of this chapter.
Schwab responded to the April 11 draft by emphasizing that what the Committee was trying to do was to identify a particular form of irreversible coma that predicted death with certainty in the absence of life support. On his copy of the early Report draft that he sent to Beecher, he scribbled in his characteristically large hand that “it is not necessary to change definition of death to define Brain Death. If we establish the concept of irreversible coma with cessation of function at all levels of the CNS it will not be difficult … to withhold or discontinue [treatment.]”7 Schwab initially argued that the Committee’s focus on redefining death was not the right posture, strategically. The key function of the Report, he reasoned, was to define irreversible coma as a distinct and unique neurological entity, and assumed that the consequences of that definition would follow. Schwab shared (apparently at the prior Committee meeting) a confidential memorandum summarizing initial findings from what would be a widely quoted study conducted by the American EEG Society on EEG, survival, and coma that appeared the following year. He wrote to Beecher in May 1968 about the first set of what would be a review of over two thousand reports of EEG and coma outcomes. While over seven hundred reports had been gathered by that point, no patients with a “flat” EEG had survived. However, there were many “flat EEGs” that did not meet his full triad.8 This seemed to underscore his own experience with varying presentations of isoelectric coma and to the potential confusion over what signs constituted irreversible and compete loss of brain function—an issue Schwab explored with colleagues for at least decade, as will be reviewed here.
The first order of business, then, was to get consistency and rigor around the description of this coma. In the memorandum to Beecher Schwab therefore advised, “Do not attempt to redefine death. Concentrate on agreement as to what constitutes irreversible coma.”9 The medical records of the patients Schwab saw, and for whom he advised withdrawal of treatment, included many patients for whom care was discontinued based on reasons other than meeting this kind of criteria. So, discontinuation of care did not need this definition. But this kind of coma needed care to end. This coma was different and needed to be reliably distinguished in the ways it was different. His experience with patients, and the potential for confusion without informed and shared criteria as evidenced by this growing data set, led him to advise Beecher in his memorandum to stay focused on communicating a standard or guideline for definition of irreversible coma itself. His advice to Beecher did not mean he did not think it reflected death, but underscored how death was described in those details and so emphasized getting those details understood and accepted. Those details essentially boiled down to finding the full triad, a twenty-four hour duration of those findings, and the absence of confounding conditions (e.g., barbiturate overdose). It “may be too conservative but it will serve as a guideline.”
Schwab offered detailed language in the next draft that built on the more sketchy April 11 version of the criteria under the “characteristics” section, which was essentially what Beecher copied as his gloss on Schwab’s work from Eliasburg. The new proposed section, entitled “Details to Establish Irreversible Coma with no CNS Activity,” brought together descriptions of apnea and nonresponsiveness (from an earlier section of the draft), and of flat EEG (from a later section) in order to more compactly describe Schwab’s “triad” of “no movements or breathing,” “no reflexes,” and a “flat encephalogram.” Beecher edited this section for style but not content; in the next draft that appears in the archives, this remained the only section detailing the actual criteria, until further edits were made by Adams in June.
In the interim, other changes by Beecher show that some of Schwab’s overall comments rubbed off in edits that remained in the published version. The introductory sentence from the April “background” section stayed in, but was now preceded by a new sentence, a la Schwab, that read: “Irreversible coma can have many causes, but we are concerned only with those comatose individuals who have no discernible central nervous system activity.”10 This was the only italicized sentence in the published Report, as it was underlined by Beecher in his surviving manuscript. Schwab’s triad itself then followed under the unchanged heading of “Characteristics of Irreversible Coma,” and a new section called “Other Procedures” described how to inform family, nurses, and involved colleagues when the criteria were met. Only at this point, with some of the ambiguity of the prior drafts removed, did the subsequent drafts and final version of the Report state that “death is to be declared and then the respirator turned off.”
Curran’s “Death and the Law” section came next, followed by a new “discussion” section, and closed finally by a summary that contained the most explicit description in the Report of the underlying understanding of the mechanisms of consciousness. The “Summary” focused on detailing how “Cerebral, cortical and thalamic involvement are indicated by a complete absence of receptivity of all forms of sensory stimulation and a lack of response to stimuli and to inner need.” This unresponsive and unreceptive state always has “coincident paralysis of brain stem and basal ganglionic mechanisms” with absent cranial and postural reflexes:
Involvement of the spinal cord, which is less constant, is reflected usually in loss of tendon reflexes … Of the brain stem–spinal mechanisms which are conserved for a time the vasomotor reflexes are the most persistent … [and are] responsible for the paradoxical state of retained cardiovascular function … in the face of widespread disorder of the cerebrum, brain stem and spinal cord.
All of these changes persisted into publication except for consequential additions by Adams, who edited Schwab’s triad to better emphasize exceptions in case of the presence of depressant substances or medications, and to make more prominent and detailed (and perhaps redundant) the notion of unresponsivity. These edits reshuffled the substance of the first two legs of the triad into three. The first criteria was “Unreceptivity and Unresponsivity,” which overlapped in terms of features of physical examination with the following two but provided a clarifying statement about the irreversible functional losses of the body at stake here—a formulation (unreceptivity, unresponsivity) not too far from that which, many decades later, continued to animate commentary over the underlying theory of death implicit in brain death. The criteria that followed were then “No Movement or Breathing,” which described the need for apnea, properly tested, and no movements or responsiveness to stimuli. After that came “No Reflexes,” which Adams also edited in detail to describe specific cranial and other reflexes. “Flat Encephalogram” thus became a fourth criteria, “of great confirmatory value.”
Of note, Joseph Murray’s suggested edits for the April 11 draft consisted primarily of pointing out references to “irreversible coma” and recommending either rewriting them so that only the word “death” be used, or else clearly indicating that the two terms were synonymous. This included a suggestion to change the first sentence to read simply that the report’s purpose was to “define death.” None of these suggestions were taken. However, between the June 3 and 7 drafts, Beecher did change the first sentence, from “our primary purpose here is to define irreversible coma,” to “our primary purpose is to define irreversible coma as a new criteria for pronouncing death” [emphasis mine].
This formulation—that this coma pronounces but is not itself explicitly death—smoothed over much of the ambiguity, if not outright contradiction, in the original working drafts, but still presented to the world an interesting medical fact. Irreversible coma was indeed unique. It was a coma with “no discernible … activity,” an as-if-dead state, allowing practices usually linked to a body absent of motion to now be associated with a body unable to experience and respond: a neurologically empty body. The writing of the Report resisted calling irreversible coma simply death, but at the same time made it an event which justified “pronouncing” death. And yet, while its title describes irreversible coma, the Report describes the Committee as determining “brain death.”
The Report did not provide a worked out, consistent notion of death—or even of life. Why not Murray’s edits? Why not black and white clarity, calling irreversible coma death? Why did Beecher instead add language about “pronouncing” death, implying (but not outright stating) the equation of irreversible coma with death itself? Press reports and subsequent discussions characterized the Report as defining death, and Beecher himself later regularly referred to it as such.
In these edits we witness a change wherein the boundaries of death, themselves coherent in the context of the limits of the medically possible, reflected how what was medically possible was itself increasingly shaped and experienced at the bedside with these patients by the limits of the nervous system. Conditions of death before dying required facing the implications of caring for individuals essentially absent a nervous system. The resulting ascendancy of nervous system function as a factor in “the calculus of suffering” wasn’t advocated, it was described and managed.
In 1953, Raymond Adams was asked to examine and advise in the treatment of a housewife with a brain tumor. The neurological exam found “pupils fixed, equal at 3 mm … areflexic throughout including corneals. Extremities flaccid. No response to pinprick.” She died early that night when she stopped breathing. A few years later, the same examination findings had different significances within the context of other findings that had become possible because of the respirator. What those findings meant and caused to happen, or caused people to think about, would change. Fixed pupils and areflexia were not simply “discovered” as “real” death. The fixed pupil and the nonresponsive reflex arc were physical events whose nature and significance were themselves changed by the respirator and by additional new methods for sustaining blood pressure and other vital functions.
Attention to these comatose patients paralleled their increasing presence in hospital care. Consider that in 1958 at MGH, sixty-six patients required ventilator support for more than twenty-four hours. By 1964, that number was 398.11 Beecher realized the need to put individuals requiring “oxygen treatment” into a concentrated area in the late 1950s, as he recognized that the increasing expertise and labor required for the management of such patients was inefficiently dispersed through the hospital. Growing awareness of such inefficiency, and in particular the death of an MGH service chief’s relative from myasthenia gravis (a paralyzing neurologic disorder), combined to secure resources for a dedicated respiratory unit. It opened at MGH in October of 1961 to specialize in the care of respirator-dependent patients.12 This effort paralleled similar recognition by many hospitals of the challenges raised by growing numbers of these patients, accelerating concern by some within medicine about the aggressiveness that increasingly characterized this care. Those cared for in the MGH unit during the early to mid-1960s—when Schwab was particularly active pruning his criteria—did not, for the most part, fit the description of the “irreversibly comatose.” It is hard to estimate how many patients did. Schwab reported that he tallied three hundred cases in which he was involved from 1954 to 1970, mostly during the 1960s.13
Schwab was not alone in wanting to draw lines and limits around the new technologies of intensive care at MGH. Neurosurgeon Hannibal Hamlin made the following argument:
Respirators and heart stimulators can maintain the look of life in the face of death while agonizing and expensive prolongation of false hope continues for all concerned. When the brain is so compromised, the EEG can signal the point of no return … interpretation of the EEG should gain medical approval for legal pronouncement of human death.14
Hamlin reported that Schwab’s criteria, as set forth in a 1962 paper, were used to end life support fifteen times at MGH. Presented at the American Electroencephalographic Society, that paper outlined the following criteria: absence of spontaneous respirations for one hour; no tendon reflexes or pupil reflexes; dilated pupils; no arousal to stimuli on EEG; an isoelectric EEG for one hour; and no heart range change to eyeball pressure. Hamlin reported that ten patients had been diagnosed this way, seven “in the past two years.”15 The following year, additional experience with thirty-four more cases was presented.16 In 1964, Hamlin announced to the national medical audience that Schwab, using these basic criteria, had “accepted responsibility with support of his colleagues for death pronouncement by EEG some 15 times during the past 21/2 years. The experience,” Hamlin went on, had “proved eminently satisfactory to all concerned, diminishing grief and anxiety for the family of the victim in a state of limbo, also relieving the travail and expense of skilled personnel, special equipment, and service.”17 William Sweet recalled that Hamlin may have made the connection for Schwab between his work on EEG correlates to nonsurvival and a legal process of defining death;18 at least one newspaper account described Hamlin as having “set up a set of criteria for certifying brain death” with Schwab.19 In his 1962 presentation, Schwab observed:
The new cardiac stimulation, other techniques, and compact respirators, have made it increasingly possible to revive the apparently dead. Fortunate cases recover both respiratory and higher central nervous function as well as normal cardiac activity. This communication is not involved with these patients … In these cases a human heart-lung preparation results that may be viable for many days … In such cases the prolongation of cardiac circulation serves no purpose, is a tremendous financial and emotional stress to the relatives, and a severe demand on hospital personnel and equipment.20
Schwab also struck up a friendship with Sidney Rosoff, who was at that time an attorney for the Euthanasia Society of America and who would later become its president, as well as the president of the Hemlock Society, which advocated assisted suicide. When Schwab needed funds to support travel to (and participate in) the USSR Academy of Medical Sciences Symposium on Reanimation, held in Moscow from November 25–29, 1968, Rosoff facilitated access to funds from the Euthanasia Society.21 Schwab, said Rosoff, “was ahead of his time” and saw brain death as part of broader questions that needed more attention regarding patient preference about life support termination.22 Rossof was co-author with Schwab of a 1967 paper, the one Curran commented on to Beecher, read before a later American Electroencephalographic Society gathering that reported on the use of the criteria at MGH for over 150 cases.23 Schwab’s interest in refining the criteria was to validate the ability to identify those beyond coma, not to advance transplant. Fred Plum, by then a world leader in the study of coma, recalled that his and his colleague’s understanding of the Harvard Report, at the time of publication, was that it reflected an interest not in transplant but in the work he and others had grappled with for decades, on the implications of and mechanisms behind severe coma.24 My own review of care at MGH, described further in the following pages, draws on medical records of 421 patients in coma who had EEGs performed by Schwab’s EEG service during the decade before the Report’s criteria for brain death were published (1958–1967). Seventy-one of these patients met what would later be established as the Harvard criteria for brain death. Interest in transplantation is not evident from these records. The issue of organ donation was documented for consideration for transplantation in the medical records of only two. In neither case were organs harvested. But the emerging utility, content, and purpose of the criteria, as well as the applicability and use of EEG and neurologic examination in the circumstances of this new irreversible coma, can be seen in these records.
Growing and concentrated numbers of respirator-bound patients spurred several efforts to describe the examination, diagnosis, and prognosis of coma—most notably by Plum, as well as by C. Miller Fisher, then the director of MGH’s Stroke Service.25 Their work set the standard for neurological examination of comatose patients through the latter twentieth century.
Increased interest in the use of neurologic signs for clinical prognosis meant rethinking which neurologic signs mattered. Admitted in 1966 for a skull fracture, a twenty-nine-year-old mother of three is described as being revived from a cardiac arrest only to be “areflexic—anticipate EEG today will be without cortical activity and thus resuscitative measures should be abandoned.” Schwab subsequently reported that her “tracing is flat,” as were two repeat EEGs twenty-four hours and three days later, at which point such measures were abandoned. However, another patient in 1959 was similarly apneic, with absolutely no activity in any of her cranial nerves, no spinal reflexes, and no movement or responsiveness to any stimuli presented. In this case, resuscitation was vigorously attempted. One difference between these two patients was that the second woman overdosed on barbiturates, which might have only temporarily affected brain functions. Consensus over the importance of distinctions like these and their impact on the reliability and use of different neurological signs took work and cumulative experience, and even then were varyingly adopted in practice. Schwab—who was called upon by colleagues to assess coma with EEG, and was well versed in EEG-based research on brain function and consciousness—explored these questions. Which findings were reliably irreversible? What neurologic signs indicated that the brain no longer functioned? That treatment was no longer relevant to outcome?
The circumstances in which Schwab and his colleagues wrestled with these questions, and the patient histories that informed their studies, can be found in surviving medical records at MGH. Records for this review were identified through index card summaries of EEGs (performed by Schwab’s MGH EEG service), which were available in the Department of Neurology archives and dated back to near the outset of Schwab’s tenure at MGH. Preserved on microfilm, these brief summaries contained the results of approximately10,000 EEGs performed from 1958 through 1967, the decade preceding the Committee’s work as well as the period during which Schwab used some of these cases to evaluate and revise his criteria. Of these summaries, most of which were routine studies done on outpatients, I selected all EEG reports that mentioned the word “coma.” This yielded over six hundred medical record identifier numbers and of these, 421 medical records could be retrieved and reviewed.
These hospital records described patients with markedly poor prognoses but widely varying approaches to the determination of when care ended. Why were these EEG consults requested and how were they used? To start, in only 163 (39 percent) of these cases was the EEG even noted to have been read or reviewed by the patient’s physician. The patterns of how EEGs were used, however, are consistent with discriminating and focused purposes that suggest that Schwab’s irreversible coma, as his edits to Beecher also suggested, was perceived as a unique entity—one which presented challenges distinct from (though overlapping with) other decisions to remove care among those severely ill. I return to this detail as it becomes relevant to post-Report commentary, as it did to Beecher’s early critics in Chapter 2. To those contemporaries of Beecher, the Committee was often described to simply be looking for a reasonable point to justify withdrawal of treatment and overdid it by feeling it had to call this point death. This would mean that the criteria were unnecessary at best and disingenuous at worst. But the medical records suggest that thresholds for discontinuation of care, and irreversible coma, were quite different.
For example, this new type of coma differed from other discontinuation cases and from other comas in the use of EEG. Of the total records reviewed, sixty-six (16 percent) had their treatment discontinued and seventy-one (17 percent) retrospectively met the Harvard criteria based on symptoms described in the record. These were not all the same cases, although those meeting the criteria were much more likely to have treatment ended. Of those who retrospectively met the criteria, twenty-seven (38 percent) had their treatment discontinued. Among the 352 who did not retrospectively meet the criteria, thirty-nine, or 11 percent, had notations indicating that their treatment was discontinued. I categorized the stated rationale in the record for deciding to discontinue care as either based on the EEG (even if other relevant findings were present), on the EEG and other physical or laboratory findings, or only on other findings. Among those who met the criteria and had treatment discontinued, in nineteen cases (70 percent) the EEG was the reason specifically given for discontinuation, whereas it was the singular attributed reason in ten of the thirty-nine patients (26 percent) who had treatment discontinued but did not meet the Harvard criteria. So from this sampling, EEG increasingly became a benchmark for prognosis and ending of treatment in a general sense, but was also used much more frequently and specifically for those who would meet the criteria. While drawn from a convenience sample of records, this pattern suggests that features of these patients compelled a distinctly closer look at the “working brain” than did other patients for whom care was withheld. The criteria appear then to have captured and crossed more than a threshold for legitimate discontinuation or withdrawal of care in general, but captured a distinct condition and different kind of futility.
Of note, few cases of discontinuation of care involved discontinuation of respirators—only eleven patients across this decade from within this specific sampling of those seen by Schwab’s team and whose records were still accessible had a respirator turned off. While this is a very small subsample, it appears that hesitancy cut across cases meeting and not meeting the criteria, as six among the eleven retrospectively met criteria and five did not. It was resisted in patients for whom all other critical supportive care (e.g., vasopressors maintaining blood pressure) was discontinued. In these records, and in much of the literature reviewed herein, ending respirator care during this period seemed to be regarded as more of a commission than an omission, irrespective of condition. Consider, for example, one sixty-seven-year-old woman admitted to MGH in early 1963: She followed Fisher’s step-by-step path toward a nonfunctioning brain, but that was not sufficient to stop treatment. This patient was “flaccid and quadroparetic and unresponsive to all but pain stimuli. Her eyes are fixed … Neither move to vigorous head turning. Both pupils … react very weakly to light. Seems quite certain she has coned. I doubt process is reversible at this time.” Things soon further deteriorated: “The present situation seems almost hopeless: although her pupils … still respond to light, she is quite unresponsive with flaccid paralysis, supported with respirator and with aramine in a precarious state.” Hours later it was reported: “Pupils fixed, no calorics.” Finally, “There was no BP or pulse palpable and ECG showed no activity and respirator was taken off the patient, who expired quietly @ 2:10 AM.” After her heart stopped, the respirator was discontinued. But the physicians writing entries here noted that the patient “died” only after the respirator was discontinued. The visibility of the dead nervous system and even, in this case, a nonbeating heart, did not outshine the visible appearance of “breathing.” This illustrates the distance to be filled for neurological signs to be adequate for considering a patient already dead, and underscores the power of the respirator and the respiring body in shaping the experience of omission versus commission. Even decades later, routine use of brain death still involves declaring death first, then removing the respirator, as it did here and in other medical records where death was declared after irreversible cardiac arrest. At this time at MGH, in these records, respirators (as opposed to ending other life supporting treatment) usually were turned off after death. So in fashioning criteria, Schwab and the Committee were describing an equivalent to the end of heartbeat, a set of circumstances on par with respirator discontinuation as a marker of death itself. Brain death, to those who developed the criteria for it, emerges less as a proxy term for permission to withhold care than as an imperative for doing so. Reading through these cases reinforces this impression.
In 1959, near the outset of this decade of activity, one man entered the hospital with multiple worsening problems: likely renal failure, heart failure, and “dilated fixed pupils, centrally fixed eyes.” His continued capability for “breathing in gasps” reflected some preserved brain function, but the signs pointed to his being “terminal from combined organ failure … Cannot forseeably [sic] reverse these multiple disorders. Supportive care.” Although vague and often fragmented, descriptions like this one of the neurologic features of these “multiple disorders” determined the gap between aggressive treatment and “supportive care.” One patient had mixed intact cranial reflexes (some absent, some present) and was responsive to pain, but had no vertical eye movements. He was described as “alert, able to blink” but showed enough impairment to qualify as “unresponsive” long enough to discourage significant treatment: “plan has been not to treat other than for his comfort, in view of age plus 8 weeks of unresponsiveness.” The vague and varying barometer of neurologic status in these accounts was enough to shape therapeutic imperatives. The chart of one patient, in September 1962, contains this dire assessment: “Waiting to see if neurological function returns will continue and make it wise to proceed with 4+ push to solve his other difficulties if possible.” Later, “Pt somewhat more obtunded this evening—Prognosis still poor—will not add antibiotics.” Over the next two days the patient however was “a bit lighter” and “continues to brighten up. Now talking a bit and showing much more movement in bed … in view of improvement begun on Pen. rx.” However imprecise, neurologic status was increasingly relied upon for making decisions to limit care, whether or not it presented as irreversible coma. Commenting on the plight of a twenty-two-month-old child brought to MGH in October, 1962, a practitioner noted: “The general medical problems here seem to overshadow the basic neurological disturbance, however, the bulging extruding brain will probably supply the prognosis.”
The conditions of death before dying elevated the visibility of neurologic signs, especially as prognostic signs. Neurologic criteria assumed, and required, greater precision because of this. In reading through the cases of patients at MGH whom Schwab and his colleagues together encountered, this process is visible—a gradual, fluctuating use of and comfort with a credible and reliable set of core signs that identified when a subset of these patients with irreversible consciousness were also in a state proximate to the full loss of bodily functioning.
Any effort to list the characteristics of irreversible coma that distinctly placed it beyond coma, beyond the moving neurological parts of the EEG research literature that supported consciousness and physiological survival, not surprisingly referred back to the EEG itself. The presumed objectivity, replicability and visibility of EEG seemed perfectly suited to meet this apparent need for greater precision of determining brain function and the consequences of neurological failure. EEG was referred to in the Harvard criteria as a “confirmatory” test. However, its role in actual decision making, at least in these patients at MGH—and especially in the composition and validation of the criteria—proved more important than that. The varying and evolving role of EEG helps describe how brain death seemed to make sense to Schwab, and to the physicians at MGH who requested EEGs from him.
This use of EEG to arbitrate standard work for severe coma evolved over time. For one thing, many cases cautioned drawing conclusions from EEG at the bedside. That caution is also ultimately reflected in the “whole brain” set of conditions and timelines in Schwab’s “conservative guideline.” For one thing, EEG could itself be misleading. In November of 1965 at MGH, a fifty-year-old woman with meningitis was found, in an EEG report penned by Schwab, to have had “no clear evidence of EEG activity … An EEG of 12/28 was flat, and one on 12/29 showed only very slight activity … Very little we can do … Almost certain she is terminal.” But several weeks later, she “ma[de] copious spontaneous movements of all extremities,” and left the hospital conversing, interacting, and moving all extremities. While not very common among the hundreds of cases reviewed, such turnarounds cautioned that other evidence was needed to declare death before dying.
A more common challenge was the ambiguity of the tracings themselves. A series of EEG readings tracking the condition of a pregnant patient with a barbiturate overdose in September of 1963 found first that “there is much muscle and movement artifact, but very little real brain activity here.” Two days later, in the same patient, the “EEG shows a little more activity than the previous one … diffuse very low voltage fast and intermediate fast activity in all leads.” Three days after that, “at regular gain, there is no EEG. No responses in EEG … NO clear EEG present.” Another chart of a patient admitted to MGH that same year read: “EEG shows almost a complete absence of cortical activity in dominant hemisphere.” In June, 1964, a man who became comatose after suffering an MI had an EEG showing “low voltage again of a very questionable sort. I am unwilling to say whether there is definite cerebral activity in this tracing … Markedly abnormal, probably no brain wave activity.” Another 1964 report read: “EEG again is flat but shows bursts of 3–7/sec waves that I cannot explain.” Later that day, a note in the chart interpreted this report to mean that the “EEG today [was] still essentially flat.” A repeat EEG was needed to clarify things: “This time, the curious theta activity is not seen … no evidence of cerebral activity.” For a fifty-eight-year-old patient admitted unconscious after a fall in August of 1965, withdrawal from pain was the only sign of a working nervous system: “Will get EEG to check cortical activity.” That report found literally mixed signals: “With normal gains there is almost no activity. With the gains up at maximum there is a little amount of fluctuant activity which may be cerebral activity … Doubtful evidence of cerebral activity” [all emphases mine]. Caveats with respect to “artifact” were common, as were characterizations of the presence of waveforms as “very little,” then “a little more,” “probably,” “maybe,” “questionable,” “doubtful,” or “almost” “real” brain activity.
In Schwab’s EEG reports, ambiguity of interpretations as to when the brain no longer worked diminished over time. Precise prognoses from EEG output often came not from the EEG alone but from how it interacted with and reinforced key physical findings. For a sixty-seven-year-old man, a surgery in March 1964 turned into an unexpected emergency with severe brain anoxia. The EEG found “no normal EEG” but some activity. A suggested repeat EEG found that “it is slightly encouraging that the record wasn’t flat but the evidence for severe brain damage is present nevertheless. Pt responds a little more today to various stimuli but the overall prognosis is very poor.” An additional EEG found “further improvement but clinically patient remains unchanged.” EEG and physical exam were tethered—shared direction and intensity reinforced each other. Schwab’s criteria ended up being those that described the point at which the outer limits of nervous function, discernible by physical examination, coincided with the limits of EEG to capture deterioration—as they did with a three-year-old who came to MGH in February of 1966 showing clinical evidence of absent neurologic functioning:
EEG … shows no evidence of electrical activity originating in the brain—pupils stopped responding to light at 4:40 PM. Blood pressure barely obtainable, No urine formed since this morning. In view of these findings the mechanical respiration was discontinued at this time. Spontaneous respirations did not resume—heartbeat ceased at 5:35 PM and patient declared dead at this time.
Schwab’s criteria for absence of function reflected his accumulating experience with patients and his response to colleagues’ requests. This meant following the course and reliability of physical and electroencephalographic signals—their fluctuation one with the other—in order to identify the conditions under which absent EEG activity also reflected physical findings that made sense. In April 1967, neurological examination of a thirty-six-year-old with encephalitis found that “his neurological status is unchanged and it does not appear likely to be changed it at all—He will be on Bird [respirator] for forseeable future and so I agree we need tracheostomy—and will so arrange.” An EEG found that “in between these jerks the record is usually flat although not always so but these artifacts appear to obscure anything else. Thus no true cortical activity can be recognized although the possibility of its existence is not excluded entirely.” Days later, as the clinical condition pronounced itself more unambiguously, so did the EEG: “EEG shows no definite cortical activity. Situation is hopeless.” This young patient died soon thereafter. On the other hand, early the following month, another patient with a dismal EEG record had this follow-up:
Repeat EEG today shows “remarkable recovery” from yesterday with evidence of some activity whereas there was complete absence yesterday. … She is totally a-reflexic, bilateral upgoing toes, no response to noise or painful stimulus. Absent corneal reflexes, minimal pupillary reaction to light. No spontaneous respiration off the respirator. The situation continues to be extremely grim. In spite of improved EEG, it is still markedly diffusely abnormal.
Schwab’s criteria for irreversible coma sought then to describe the coincident irreversible loss of sustainable physiology and the machinery of consciousness. The apparently objective, visible qualities of EEG, along with efforts to standardize the physical, neurological examination of coma, mutually reinforced each other for that purpose.
Disagreement about whether to emphasize brainstem versus “whole” brain (brainstem and cortex) signs in the criteria continued among neurologists long after the Report’s publication. This was taken by commentators on brain death to be a sign that brain death was really a dispute over conceptual biases regarding whether death resulted from loss of “higher” consciousness and personhood on the one hand, versus loss of “lower” functions and biological integration on the other. Within the medical thought-style, for these physicians the debate between brainstem and whole brain positions was not a conceptual one but a technical one. Either position sought to capture a “whole” loss of higher and lower brain function. They differed over what parsimony of symptoms offered an acceptably accurate reading of those losses. Rather than a dispute over meanings or priority of mental versus biological aspects of personhood, these disagreements were about medical knowledge, about the sensitivity and specificity of certain outward physical signs to describe inward details, and the conditions and reasons for knowing them. In this way it was also a difference over how to understand the findings of Moruzzi and Magoun, and how that generation of research described in Chapter Four translated to the bedside—whether among its implications was that “an isolated cortex is useless to an animal.”
Since at least 1955, MGH neurologist C. Miller Fisher had focused on these physical signs. Independent of Schwab, he developed a physical examination to determine what was happening in the brains of comatose, respirator-bound patients. Fisher recalled that as director of the Stroke Service at MGH, he began that year to try to identify physical characteristics that correlated with level of coma and that could be used to predict outcome: “We had to get into it … we had no language to describe them or rate them or anything … the examination of the sick [comatose] patient was chaotic.”26 Serial editions of Roy Grinker’s Neurology, a prominent neurology textbook published from the 1930s through the 1960s, for example, included minimal descriptions of the examination of the comatose patient.27 Fisher drew upon observations of brainstem and higher level function that appeared in the clinical literature, such as those involving maintenance of gaze upon rapid head movement (the oculocephalic reflex), or eye movements resulting from pouring iced water in a person’s ears (caloric stimulation) and lightly touching the corneas (the corneal reflex), as well as responses to general motor reflex stimulation; he further considered how all these correlated with lesions at autopsy. Fisher recalled that by 1957, he was advising decisions to turn off respirators based on confidence in his exam to predict the irreversible loss of the central nervous structures’ ability to maintain any cortical activity or independent vegetative functions. With few exceptions, families participated in these decisions with little fanfare or controversy.28
Fisher coined what later became commonplace names for neurological syndromes and terms; in particular, he established a nomenclature to describe brain hemorrhage and infarcts by combining careful patient observation and pathological correlation.29 His examination of the comatose patient, not published in detail until 1969, reflects this kind of approach. It was an inventory of symptoms and a description of their functional and prognostic significance. Fine detail mattered in order to accurately describe the nature and location of the coma’s cause. “Athetoid restlessness,” “rhythmic myoclonus,” “fishmouthing (dropping of the lower jaw at inspiration),” and “ocular bobbing” all were often missed or poorly understood signs that could provide diagnostic information above and beyond the more traditional examination of pupillary light response, reflexes, respirations, and response to pain and other stimuli. Fisher detailed the value of certain cranial nerve and brainstem reflexes, easily missed combinations of symptoms, or order of appearance of symptoms that revealed the neurological condition within.
Fisher described two general causes of coma: severe injury to the cerebral hemisphere, or “indirect functional paralysis of cortical function consequent upon inactivity of the ascending reticular activating system.”30 The latter might be captured through unique physical findings on examination. While Bremer, Magoun, Moruzzi, Penfield, and others “accumulated an extensive knowledge of the relation of alertness and consciousness to the functions of the brain stem … when human coma cases are discussed the details of the neurologic examination are almost totally lacking.”31 That examination had to focus on objective correlates of brainstem functioning: “Coma can be accurately defined only in terms of a complete neurological examination.”32 This relied on replicable physical findings, ones far more reliable than vague descriptions of awareness or consciousness: “Introduction of the terms conscious and unconscious has been avoided here, for these have to do with the content of ‘psychic’ experience, and although dependent upon ‘alertness’, alertness by no means insures full consciousness.”33 Fisher went on to explain these findings in greater detail:
Recently it has been suggested that the death of a patient be defined in terms of an electroencephalogram, namely electrical silence or the absence of brain waves. Preferable to this method is a meticulous assessment by a neurologist skilled in the interpretation of comatose states and the recognition of extensive irreversible damage to the brain stem—cessation of all respiratory movements, fixed dilated pupils, fixed eyeballs in the central position, absent corneal reflexes and flaccid limbs. Since the introduction of modern respirators in 1955, we have followed the course of over 200 comatose patients with these signs and none with a structural non-metabolic insult to the brain has made any recovery despite vigorous supportive measures carried out for periods of up to three weeks. Lesser deficits may also be incompatible with useful recovery but such states are difficult to define.34
Fred Plum also felt that physical clues alone about the integrity of the brainstem were sufficient. In 1966 Plum published, with Jerome Posner, a still authoritative book on the examination of coma called The Diagnosis of Stupor and Coma; the second edition was published in 1972. The change in content across these two volumes reflects a change in the ways that observations of coma, along with work on EEG and arousal mechanisms, guided conclusions about the ends of treatment and the end of life.
The book, essentially an elaboration of a paper Plum published with Donald McNealy in 1962, laid out how Plum considered the steps leading to damage to the upper brainstem as the usual cause of coma.35 In this book, Plum revisited the haunting decerebrate patients of Wilson, Cairn’s cases of akinetic mutism, von Economo’s experience with encephalitis, and a range of observations in the clinical literature of “pathological sleep” and sensory-motor signs in brainstem and third ventricle injuries.36 He also reviewed well-described syndromes of supratentorial (above the brainstem structures) lesions, and patients for whom edema from cerebral anoxia and trauma also caused swelling and impingement on lower brain structures coincident with worsening responsiveness and nervous system function. Taken together, these cases led Plum to conclude that most roads to coma traveled through the brainstem. The work of Moruzzi, Magoun, and Lindsley on the RAS bolstered the view that these clinical signs should be interpreted to mean that the brainstem was the final common pathway to coma. Plum considered diagnosis of coma to be the task of distinguishing between (1) supratentorial lesions causing downward damage that eventually reached the brainstem; (2) direct brainstem lesions; (3) more unusual severe bilateral hemispheric injury that usually derived from metabolic and, in rare cases, from catastrophic anoxic injury; and (4) psychogenic causes. The book was organized around detailed descriptions of these four possibilities, and the clinical correlates that should alert the clinician to each.
The first edition of the book included research by Jouvet on the complex regulation of sleep, along with Lindsleys’s musings on the RAS and thalamic reticular systems as the mechanism of true consciousness defined as selective attention and sensory ordering. Plum focused on the argument that coma usually meant brainstem damage and was generally different from the clouding of consciousness found in cortical damage. Exceptions to this rule were easy to identify—barbiturate or anesthesia use, or massive bilateral cortical damage. Plum was more circumspect in following through on the implications of this shift in focus from cortical to brainstem damage as the core cause of coma in terms of what it meant for drawing conclusions about irreversible coma. While helpful, EEG work on the brainstem was not definitive in this regard. Magoun, Plum would later point out, was talking about the RAS and EEG waveforms, not necessarily consciousness: “All consciousness requires arousal but all unconsciousness is not the absence of arousal. This was still a hard concept, even in the 1960s.”37 The irreversibly and completely unconscious body did not directly translate into the cerveau isole, at least not generally in the US experience at this time. Again, Bard’s response to Magoun—that it may be impossible to know what a decerebrate animal experiences—remained quite relevant. The debate over brainstem approaches versus whole brain approaches (usually meaning loss of cortical EEG as well as loss of brainstem signs and greater rigor about absence of certain signs such as peripheral reflexes) would distinguish efforts in the United States from those in Britain to establish criteria for brain death, and capture the continued relevance of Magoun’s question.38
But Plum and Posner’s second edition showed more confidence in asserting that brainstem damage was the basis for coma and, then further, that if the brainstem were no longer functional, this provided enough certainty as to irreversible loss of the capacity for consciousness and related cortical activity. Significant research supporting that conclusion, even though published before the first edition appeared, was only included in this second edition. This work was cited to bolster the primary role of the brainstem in coma and to address more directly how loss of brainstem structures reliably meant the loss of consciousness. The second edition also elaborated more on describing RAS, this time including transcriptions of the 1958 symposium that had been edited by Jasper to summarize work on the reticular formation.39 Plum described consciousness as “awareness of self and environment” and coma as “unarousable unresponsiveness.” He endorsed the British Medical Research Council definition of coma as the “absence … [of] psychologically understandable response to external stimulus or inner need,” which was similar to the phraseology that Adams added to the Report.40 While Plum assigned to the brainstem and cortex the roles of maintaining conscious arousal and content, respectively, he agreed that there was enough evidence to say that both ultimately relied on brainstem functions. Therefore this edition more firmly indicated that loss of those functions meant irreversible loss of consciousness, irrespective of how the cortex was doing. He also included Mollaret and Goulon’s paper coining “coma dépassé,” and work on the pathology of brains of respirator-bound, comatose patients—all of which were not included in the 1966 edition. The very same citations of work by Moruzzi and Jouvet on the mechanisms of sleep in the first edition were used in the second to draw a different conclusion, reflecting the difference these interim years made. In the first edition, these citations supported vague acknowledgements that coma and sleep mechanisms might overlap. In the second edition, Plum is comfortable enough in his conclusions about the brainstem to use these same references to justify the conclusion that their mechanisms were in fact quite different.
In the space of five or six years, Plum became more secure in describing the prominence of the brainstem for coma, a transition he explicitly recognized.41 This confidence was critical for a new section appearing in the second edition of Diagnosis to address a subject unremarked upon in the first: the subject, that is, of which signs constituted an irreversible coma indicating nonsurvival and perpetual unconsciousness.
In Plum’s reading of the literature and his experience with patients, if examination found a nonfunctioning brainstem by reviewing the reflexes and responses served by that structure, then EEGs were unnecessary to determine irreversible coma. He felt Beecher and his Committee were too conservative. Similarly, preservation of some primitive stretch and postural reflexes unrelated to brainstem integrity did not exclude the diagnosis. His own criteria thus included total unawareness, lack of purposeful response to stimuli, and absence of brainstem function—that is, no spontaneous breathing or apnea, unreactive pupils, and absent oculocephalic and oculovestibular reflexes. These basically “British” brainstem criteria were still markedly similar to the whole-brain Harvard criteria. The latter was differentiated by a broader exclusion of peripheral reflexes (though also contingent upon them), as well as by reference to EEG and confirmatory testing to more completely avoid having to answer Magoun’s question. Over the next decades, these differences in criteria would narrow due to the accumulating—though still disputed—evidence of the redundancy of these confirmatory tests.
The whole-brain direction of the Committee not only reflected more caution as to whether some surviving cortex in the context of irreversibly nonfunctioning brainstem was relevant for consciousness. It also turned on the degree to which these signs described physiological survivability as well. For Schwab, and the Report, Plum’s conclusions about permanent unconsciousness were necessary but not sufficient for a neurologically based finding of death—and also not fully conclusive.
Reliable signs of the absence of capacity for consciousness were necessary but not sufficient to consider irreversible coma as a pronouncement. Regardless of whether one would argue that consciousness was the sine qua non of life and personhood, it could not be argued that a dead person could still have the capability for consciousness—one could not be considered dead if one were conscious. The neurologically empty body described by the criteria was one that, in this context of care and other physical signs, appeared to show itself to be physiologically empty as well, such that Schwab in his medical record notes began distinguishing the “cerebral” or “cortical” death from the “clinical” and “physiological” death of the irreversible coma patient.
For a patient seen in November 1958 with significant unresponsiveness to pain, but signs of continued brain and cranial nerve functioning, it was suggested that a “special EEG [be done] today to see what state of consciousness” the patient maintained. However, the “EEG [was] not contributory as hoped it might be.” In the cases Schwab’s EEG team was asked to see at MGH, “special” EEGs would become less special and more common guides for what they described, not only about the survivability of consciousness but also about overall physiological functioning. In May of 1959, a patient with pneumonia had a neurological examination generally consistent with what would become the Harvard criteria, save for the persistence, albeit poor, of spontaneous breathing. The record indicated that “the presumable amount of brainstem involvement is probably not reversible” and several days later noted: “on optimum antibiotics—appears jig is up. Cannot survive long.” The following entry reported that the “Pt. died a respiratory death at 11:41 PM.” No EEG was obtained, and in this case the interest in brainstem functioning arose, but in order to understand the likelihood of continued breathing. There is clear evidence in these records of a working understanding of the role of the brainstem in maintaining consciousness. Indeed, In March of 1962, Schwab’s colleague Hannibal Hamlin suggested pushing the degree to which RAS integrity mattered, and could even manipulated, when he wrote the following about a patient:
Coma seems fairly deep—probably due to RAS depression … Can brainstem be stimulated to improve state of awareness by external EST [electric stimulation] … Would doubt it because of other factors assoc with or contributing to RAS suppression. But if EEG is not too disorganized … EST could be tried … If no response—brainstem damage might be considered irreversible.
EST as a means toward stimulating more direct repair of injured brainstem was not a successful strategy (though there have been intermittent efforts to revive interest in that possibility since). But as with the case described above where mention of the brainstem was in the context of assessing the status of respiratory function, not consciousness, these records tended less commonly or directly to focus specifically on the condition of the brainstem.
The pattern, at least in these records with disproportionate use of EEG among those who would retrospectively meet criteria seems to reflect a belief that redundant and more direct and visible validation of cortical loss was necessary. But, again, only part of the work brain death criteria had to do was to validate irreversible loss of consciousness. The whole-brain, as opposed to brainstem, direction of the Harvard criteria also reflected a response to inconsistency in the degree that this loss of cortical activity alone also captured central nervous system and overall functional loss more generally.
One forty seventy-one-year-old patient on Fisher’s Stroke Service in February of 1961 showed significant unresponsiveness, but still had some rotation on one limb in response to pain and symmetrical reflexes. However, one physician noted “no cortical activity by EEG,” and another that “the situation seems entirely hopeless.” Orders were made to “D/C aramine streptomycin penicillin” to maintain the patient’s blood pressure. In another situation in April 1962, a patient after a cardiac arrest was felt to need “no additional therapeutic suggestions. No heroics are indicated for indefinite cardiac maintenance.” The medical record a few days later reinforced the physician’s initial stance: “I really do not know if we have to continue the situation. The anoxic brain damage seems quite [unreadable].” But others noted: “He is making swallowing movements but does not direct gaze toward spoken voice.” Nonetheless, “The EEG report is very close to being a pronouncement of no activity. Under the circumstances the prognosis is probably hopeless and should discourage anything except palliative [treatment].” At this point, medications supporting the patient’s blood pressure were removed. These cases showed ambiguous signs of continued physical responsivity in light of evidence of irreversible loss of consciousness, and would not have met the Report’s criteria. They illustrate both the degree to which brain function factored into decisions to determine and discontinue futile care, as well as the degree to which the eventual brain death criteria, in contrast and on purpose, went beyond that threshold to describe those who showed a further state of physiological failure.
Schwab established a routine that included the use of EEG to validate this state—first, a finding of a total lack of neurological responsiveness by exam; then a flat EEG, confirmed by a second; followed by withdrawal of treatment. Completely unresponsive except for spinal reflexes, a five-year-old girl had an EEG in February of 1965. The record showed that “after discussing findings with the parents, the mother requested that the Byrd machine be turned off. This was done as requested since there was no activity of EEG record.” A twenty-five-year-old head injury victim was admitted that same month in “flaccid, areflexic” condition, with “pupils fixed and dilated.” The record further noted that “pt may be candidate for EEG determination if there is any CNS function.” That “determination” found “no EEG seen any leads any time in spite of increased gains. Probably cerebral death. Repeat in 24 hours if desired.” The notes went on to say that the “EEG yesterday thought to show cortical death by Dr Schwab. Repeat today. I should think we should refrain from further transfusions, etc. Family aware of the situation.” Another patient who fit this pattern was a fifty-nine-year-old man, admitted in May of 1966, who had stopped breathing. He met what would become the Harvard criteria: fixed pupils, no responsiveness or reflexes, and no breathing. The EEG was reported by Schwab as “essentially flat with no recognizable EEG. Should be repeated in 24 hours.” That same day the chart read: “The EEG as reported by Dr. R. Schwab was ‘no evidence of cortical activity.’ After discussion between Mr. B and Dr. S it was Mr. B’s decision to omit aramine fom IV’s. The full implications of this were explained to Mr. B.”
Consistent correlation between the tangible, flat EEG and the physical signs, nonsurvival, and irreversible loss of consciousness was reported in a series of three studies published from 1963 to 1968 in which Schwab worked to elaborate and, most significantly, confirm the prognostic value of his criteria. First with a series of ten patients, and finally with a large series of ninety cases with autopsy correlations, he felt he showed that his triad of clinical findings reliably indicated a dead brain and nonsurvival.42 The first study, presented in 1962 and published in 1963, summarized a series of ten cases over the prior two years. These cases showed no spontaneous respiration or pupillary reflexes, absence of EEG, and no response to noxious stimuli (i.e., no heart rate change with eyeball pressure). This triad predicted brief survival and led Schwab to conclude that “in spite of cardiac actions, the patient is dead.” Based on experience presented the following year with thirty-five more patients, Schwab likely was spurred to lengthen the time of isoelectric EEG to a twenty-four hour minimum between readings, a length that appeared in subsequent versions of his triad.43
In 1965, Schwab published a study of what he described as all of the cardiac or respiratory arrest patients on whom EEGs were performed at MGH. In those with flat EEG records, or records with markedly depressed voltage, death occurred in all cases with no interval of a return of consciousness.44 Finally, in 1968, Schwab reported a series of ninety patients who met his now condensed “triad” of criteria for “irreversible coma”: no reflexes, no spontaneous respirations, and an isoelectric EEG. This triad correlated with marked necrosis of brain tissue at autopsy. The criteria predicted nonsurvival as well as destruction of brain tissue. The Harvard Report emerged from this mixture of reliable, brainstem-based but cortical EEG-confirmed signs of a lost capability for consciousness, which also indicated nonsurvival. The Report summarized irreversible coma as the “Complete absence of receptivity of all forms of sensory stimulation and a lack of response to stimuli and inner need.” This unresponsiveness “always [had] coincident paralysis of the brainstem and basal ganglionic mechanisms.” Brainstem loss, up through the thalamus, was the physical finding correlating with loss of “response to … inner need.” By specifying in this “summary” section that brainstem pathology was at the root of irreversible coma, the Report aimed to distinguish spurious spinal reflexes from relevant responses—a point the “British” view underscored. Why, then, didn’t the criteria simply and directly adopt Plum’s eventual confidence in the brainstem?
To start, for Adams, work on RAS—the “diencephalic regulator”—still fell short of a completely satisfying account of consciousness.45 Conditions like akinetic mutism left unsettled the nature of the brainstem and consciousness. Adams recalled Committee discussions of syndromes such as akinetic mutism and what would come to be called persistent vegetative state. These were syndromes against which the Committee needed to set boundaries. They were not ready to judge these as markers of “higher” death of the brain or person—not because of a starting point of philosophical commitments, but because of limited confidence in what these clinical syndromes reflected physiologically and functionally, and lack of reliable terminology with which they could be described.
Knowledge about the “diencephalic regulator” lent confidence, but also the expectation that the scope of the criteria described the lack of a “waking” brain as well as the absence of a “working” brain and body. Based on comments by William Sweet, the “summary” section likely emerged out of debates between Sweet, Adams, and Schwab around how completely physical signs of brainstem dysfunction alone described irreversible coma and death. Why suggest a confirmatory EEG? Why require (and then later seem to backtrack on) absent spinal reflexes and all movement? Sweet recalls taking the British view that all that was needed was clinical evidence of no brainstem, which primarily included apnea, absent cranial reflexes, and unresponsivity. But the other three “beat on me,” and remained concerned about whether evidence was indeed adequate that there would be no recovery of consciousness if there were remnants of EEG activity (e.g., nests of cells in the cortex) as the British view could allow. Sweet felt the more restrictive view of the Report would “hold up” transplantation.46 Adams, though, proved decisive. He cited the line of research, beginning with Moruzzi, Magoun, Dempsy, and Morison, which established that the regulator of consciousness was in a small part of the brainstem. But, as with Sweet’s recollection, Adams also was mindful that for the purposes of describing a coma that pronounced death, the data (such as Schwab’s) that more reliably demonstrated not just irreversible unconsciousness but nonsurvivability pointed to broader criteria.
Practice at MGH relied upon this mutual reinforcement of EEG and physical findings in ultimately describing the features of the brain dead condition. Beecher’s confidence in Schwab’s triad, both at Eliasburg and in drafting the Report, rested to a large degree on his experience at MGH (and the growing practice in other centers) of including isolelectric EEG as a criterion. Almost eighteen months after the Report appeared, Beecher gave advice to a Johns Hopkins surgeon—who was also leading an effort in Maryland to establish guidelines for brain death—reiterating the importance of EEGs within Schwab’s timeframe and emphasizing that data did not yet support a change in practice.47 Adams’ edits to the criteria positioning EEG as a fourth and conditional criteria don’t seem, then, to have been intended to diminish EEG use, but more likely to underscore, as he later put it, that “the point of the article was that you could circulate with no brain and that society should recognize that to be without a brain was more equivalent to death then stopping of the heart.”48 Eventually, though, additional data and clinical consensus would move more toward agreement on a limited need for EEG and on Adam’s highlighting the loss of receptivity and response to inner need as key to what the criteria described.
The still tentative connection between what was known from research, and what was found with patients, called for more conservative criteria, in the view of the Committee—especially as a basis for the standardization of a condition that pronounced death. Schwab was explicitly asked if his criteria established “death” of the “higher” or “lower” centers, or both. Schwab replied that the criteria established both, but attributed this to the loss of the “lower centers,” which Sugar and Gerard,49 his most cited source, showed had the longest survival times among brain regions in the face of anoxia. If the lower centers were dead, “one can certainly assume that the higher centers are also dead.”50 But to know that from the outside required findings that might exceed simply testing brainstem integrity.
This was further underscored by cases where it seemed that brainstem function remained but cortical function apparently did not. In 1963, for example, a fifty-year-old man was unresponsive with absence of most midbrain and brainstem findings, such as response to pain, pupillary reflexes, and other cranial, ocular, and auditory nerve reflexes. But on exam, he had some residual lower brainstem reflexes, including slight gag and tongue reflexes, and the record noted this activity:
[The patient was] showing functions of the lower pons and medulla but little evidence of function of any higher level of nervous activity. The prognosis is extremely poor and there is little to suggest that the most major therapeutic efforts would result in significant patient salvage. An EEG would be of aid in evaluating level of cerebral cortical damage. [That] EEG shows no electrical activity (completely flat record) over 20 minutes of recording. In our experience this would indicate that the prognosis for recovery is nil.
So some cases suggested that when it came to survival, irreversible cortical loss, may have been a more sensitive sign. Reinforcing this learned need to trust but verify the consequences of lost brainstem function, and to distinguish cortical death from just grave prognosis, one physician caring for a patient with coma in January of 1966 noted that an “EEG repeated today again fails to show any evidence of cortical activity. Imp—There is no clinical or electrical evidence of function of the cerebral cortex despite continuing brain stem activity. The likelihood of a return of cerebral function to anywhere near a normal state in this patient is now very small” [emphasis mine]. This conclusion was explained by Young: “[EEG] is as flat a record as I have seen, implying widespread cerebral death. This is not the picture following brainstem infarction (cortex remains electrically active with diffuse slow waves) … Therefore I think we have to ascribe deep coma [without] any reflexes except [decreased] plantars to widespread brain damage following cardiac and/or respiratory arrest.” Similarly, there was risk of EEG overdetermining complete loss of cortical activity as the gold standard endpoint. In another patient with a flat EEG, seen in December 1967, the EEG report conveyed that “it is obvious that there is some cerebral activity still present in view of the brisk pupillary response. It would be reasonable to carry on with present regime of Rx.” There often appeared a preference for cortical functioning or reliable evidence of its loss in making prognoses more generally, a kind of anti-British view: “Prognosis is dismal in view of EEG and physical status. … No corneals, no EOM on Doll’s head, no pupillary reaction with pupils … Flaccid, no DTR’s, no reaction to painful stimuli.” That said, another note read: “She still manifests a few brainstem reflexes, but the important fact here is that the cerebral hemispheres are essentially dead. Please do not order any chemistries or other diagnostic tests” [emphasis mine].
Whole, cortical, and brainstem-only signs were complexly distinguished and sorted. The “beating up” Sweet described by Schwab and Adams makes sense in the context of these cases. If there was to be a standard practice, it would be better to err on more sensitive, rather than specific, criteria. In these cases at MGH where Schwab’s team was asked for EEG consultation, physicians making decisions showed this preference for the collection of signs that the whole brain was absent activity by reading the cortex. As quoted above, Schwab increasingly used the term “cortical death” as the ultimately important factor in either EEG tracings (which measured activity of the cortex) or physical neurological examination. Remarking on the case of one unfortunate fifteen-year-old, who was injured in a motor vehicle accident in April of 1967 and who met what would be the Harvard criteria as well, Schwab wrote that despite the lack of the physical examination findings at the core of those criteria, ultimately there was “no evidence of electrical activity … This is compatible with cortical death … There is no evidence of cortical electrical activity in this study.” A few months before the Report came out, Schwab’s lab reported on the EEG of a forty-four-year-old woman: “There is no EEG in this record … this second flat EEG, 48 hrs after the first, means that there is no chance of cortical recovery.” A fifty-one-year-old man treated in 1964 for a cardiac arrest after withdrawal from alcohol was described as “a desperate situation where there seem to be too many strikes against the patient to consider survival possible. At this point I would give no further pressor agents but treat symptomatically, continuing respirator and IV fluids, until heart action stops.” A subsequent EEG was read by Schwab to show “no EKG or EEG seen; patient clinically, as well as physiologically, dead.” The relevance of cortical loss as the measure of coma dépassé depended on the other legs of the triad that reinforced the connection between irreversible loss of consciousness (through redundant signs of brainstem failure) and overall physiological failure. Read another record in 1966:
The pt’s upper brainstem (midbrain) was functionally transected in the EW as shown by fixed dilated pupils s globe movement to doll’s head or ice water calorics—now the medulla is failing as shown by absent spontaneous respirations and now falling BP. Would favor EEG today and in 24 hrs if pt ‘lives’ that long—anticipate both will be flat.
Schwab’s evolving phraseology reflected this experienced redundancy of cardiac standstill as a marker for physiological death when what he called clinical death, in terms of irreversible cortical and brainstem failure, were found. Summarizing the condition of a two-year-old patient, in coma after a fall and subsequent cardiac arrest and admitted in December of 1966, one physician noted the situation this way: “EEG x 2 with no cort. activity demonstrable makes any statement about heart, etc. academic.” The EEG report captured the concurrent “deaths” that were visible and managed in this world of death before dying: “Recording on a deceased pt is abnormal because of absence of all electrical activity … This is clinical death.” (emphasis mine). Of another patient cared for after a stroke in October 1966, a physician wrote: “I suspect we could keep cardiorespiratory system functioning for an indeterminate period, but with no cerebral activity there is little rational reason for doing so.” Reliance on cortical death, understood as a reinforcement of brainstem death, became more routine and explicit in the records of cases Schwab’s team was called to see for EEG. In July of 1967, one of Fisher’s former Fellows wrote the following report for an eighteen-year-old who sustained head trauma after a car accident and met the full set of the Harvard criteria:
I have discussed in detail with Mr. and Mrs. C—about the significance of the EEG with absent brain waves; they understand fully that his respirations are being controlled completely and that his heart continues to function because of respiratory assistance. They have agreed that if additional EEG shows no electrical activity that the respirator be discontinued.
These MGH records also show the contradictions and ambivalence that came with more technologically intensive medicine and that appeared long before, and continued through and presumably long after, the appearance of brain death. In 1958, about when Schwab began cataloguing cases, the record for a patient with massive, terminal, internal bleeding documents that she was regularly “begging to be allowed to die pain free. Further attempts at transfusion not carried out.” Yet when found with no heartbeat, the patient underwent tracheal suction, chest compressions, and intracardiac injections of epinephrine. A year later, a physician’s note about whether to change a breathing tube of his six-year-old comatose patient as a precaution against future infection reflects much of the back and forth around the calculus of omissions and therapeutics in end of life decision making that was just emerging medicine at this time: “Of course she will probably not survive long enough for … tracheal inflammation to become apparent—but as long as we are making an attempt it seems worthwhile to make it in such a way that it won’t be self-defeating if by remote chance it should be successful.” In 1962, as Schwab’s EEG lab linked his triad of symptoms (the lack of any reflexes, spontaneous breathing, or EEG) to nonsurvival, his team consulted on a patient without evidence of any brainstem function as tested via examination of the cranial nerves: “No EOM, no corneal, no facial grimace, no gag. No DTR’s, no response plantars flaccid. No response to pain … EEG still flat this AM … Beyond maintaining her cardiorespiratory function to some extent, there is no evidence of viable CNS at this time.” However, when her heart stopped, vigorous efforts were still made to revive her.
These medical records show repeated cases in which caregivers identified and agreed with the criteria and what they meant, but aggressively treated patients even when the heart stopped. Just a year before the Report appeared, a woman with cardiac arrest had an “official EEG report [which] has been read as no activity and will be repeated tomorrow.” The next day, the patient “remain[ed] hypotensive on levophed without detectable CNS activity on 2nd EEG.” But when she had another cardiac arrest, she was aggressively treated with multiple cardiac shocks and epinephrine injections. In 1966, another unfortunate patient, who was comatose, had a recording that was “grossly abnormal because of absence of electrical activity … This is compatible with clinical death.” A same-day confirmatory EEG found “no change … Findings still are those of clinical death.” And yet, aggressive treatment continued.
Beecher’s goal, in part, was to make a new effective standard, or set of standards, in this context. Robert Young, Schwab’s Fellow, observed that a prominent tension in the Committee meetings he attended arose over whether individual clinicians—considering their knowledge of pathology and familiarity with Fisher’s experience—could diagnose irreversible coma based on their “judgment” alone. Likely as a result of seeing mixed and unpredictable practices and applications of available examination methods, Beecher and Schwab argued for standardization of criteria.51 About the time Beecher moved to form the Committee, some signs of standardization appeared. One note confidently and crisply surmised that a forty-year-old comatose man admitted in October of 1967 had an “EEG x 2 absent reflexes for legal death.” And earlier that year, a new term appeared in these medical records: “I see absolutely no chance of any recovery whatsoever and pronounce the brain dead. With husband’s consent I recommend turning off mechanical aids and declaring pt dead” [emphasis mine]. The patient met the eventual Harvard criteria of no cranial reflexes, responsiveness, or ability to breathe. After describing that examination, a note appears in the record written by the patient’s husband: “I authorize the termination of automatic artificial respiration on my wife.”
The relationship between the brain’s death and physiological dissolution was more an evolving finding than a principled position. Bioethics critics would argue that in reality, however, the brain death criteria outlined by the Report did, at least implicitly, adopt a conceptual position of life as biological integration, and of the brain as the integrator-in-chief. That position, it was argued, became more untenable as life support methods improved. For Beecher, and it appears similarly so for many working at the bedsides at MGH, any such position was an endpoint, not a starting point. Dis-integration did not have to be proven to be a legitimate guiding concept. It was a lived reality and target on the wards at MGH and elsewhere. If ICU technology has changed the way disintegration is understood since the criteria appeared, it has done so through changes in how to prove, describe, make visible and test things in the medical context, and the purposes that driven them. Those changes deserve our deep understanding, rather than the conceptual labels that may describe those changes. Preoccupation with the latter labels rather than the former knowledge practices would prove to have substantial consequences, as discussed in the next chapter, for example in changes to how death was seen as necessary before removing organs for transplant.
The Harvard criteria described (as any death criteria should) how the longstanding distinction between omission and commission at the heart of guiding medical action become so thinly separated as to have no meaning in practice. If commission (intensive care on a neurologically empty body) was unethical experiment, then its omission was not an act of removal but recognition of what was so: that death was present before dying. These touchstones of medical action and experimentation were not defined through discursive, conceptual, analysis and debate, but through pragmatic use, accumulated evidence, and comparative experience with death and dying in the medical context.
Taking on these meanings not from some external conceptual perspective, but from the perspective of how meanings pragmatically worked in practice through use of inter-observable facts between participants, leaves us facing larger questions about the consequences of choosing to heal bodies through manipulation of biological mechanisms. What are the tools for generating, verifying, and living with the sorts of knowledge needed to do that? How much are these constrained by the body, or instead open to our own ends, such that this knowledge serves goals we value? These questions invite taking seriously the ontology of medicine; the ways that scientific methods of verification, however imperfect, unavoidably shape reality and broker values. They also acknowledge the constraints that raises for hard choices that ethical analysis, explored in the following chapter as applied to brain death, could neither solve nor remove.
Medical knowledge and action indeed shape the messy tangle we carry on with our biological selves and what we consider human nature to be. That knowledge and action has to be deeply engaged. Beecher’s pragmatic response and description of how that knowledge works, and has to work, as a starting point for addressing these dilemmas, was a substantive one. Histories have instead generally consigned Beecher and the Report, to a role in justifying the bioethics narrative, which he, and the events described in this book, challenged.
1. Robert Young, who succeeded Schwab as director of the Brain Wave Laboratory at MGH, confidently claimed first status in an obituary in the Archives of Neurology. See “Robert S. Schwab, M.D.,” Archives of Neurology 27 (September, 1972): 271–72. Schwab’s primary research assistant at the lab, John Barlow, eulogized Schwab as well, with the more qualified “one of the first” claims in “Obituary,” Journal of the Neurological Sciences 19 (1973): 257–58. Most of the literature that I have come across describing Schwab’s life and/or development of EEG echoes Young’s version.
2. Alan Rubinstein, “And she’s not only merely dead, she’s really most sincerely dead,” Letters, The Hastings Center Report 39, no. 5 (September–October 2009): 4–5.
3. I rely on surviving copies of the draft that belonged to Schwab and Murray throughout this study and thus may not specify which version when quoted, unless specific marginalia or source of edit evidence is unclear. They can be found respectively in Box 6, Folders 23 and 21, Beecher Papers. The same is true for the other versions. Changes resulting in and to the June 3 and June 7 draft are primarily from Beecher’s copies, which can be found in Box 6, Folders 32 and 33, Beecher Papers. Adams’ copies also contain other confirmation attributing edits to him (as do notes on Beecher’s copies) in Box 6, Folder 18. June 13 version edits rely upon Beecher’s versions in Folder 33 and Adams’ versions in Folder 18. Unless specifying a particular issue, I will not cite these sources again below.
4. Box 6, Folder 23, Beecher Papers.
5. Box 6, Folder 36, Beecher Papers.
6. J. F. Alderete, F. R. Jeri, E. P. Richardson Jr., S. Sament, R. S. Schwab, and R. R. Young, “Irreversible coma: a clinical electroencephalographic and neuropathological study,” Transactions of the American Neurological Association 93 (1968): 16–20. He similarly wrote to Beecher in comments on the April 11 draft that ending care for those with irreversible comas will “allow the heart to stop which is the moment of death.” Schwab to Beecher, Box 6, Folder 23, Beecher Papers.
7. Schwab to Beecher, Folder 23, Beecher Papers.
8. Schwab to Beecher, Folder 23, Beecher Papers.
9. Robert Schwab, Memorandum, “American EEG Society Committee on Problem of Cerebral Death,” Box 6, Folder 36, Beecher Papers.
10. Beecher’s copy of June 3, 1968 draft, Box 6, Folder 32, Beecher Papers.
11. H. H. Bendixen et al., Respiratory Care (St. Louis: CV Mosby Co., 1965), 4.
12. See Henning Pontoppidan, “The development of respiratory care and the Respiratory Intensive Care Unit (RICU): a written oral history,” in “This is no humbug!” Reminiscences of the Department of Anesthesiology at the Massachusetts General Hospital, ed. Richard J. Kitzler (Boston: MGH), 151–77; and Pontoppiden, interview with the author.
13. Schwab, “Manuscript outline for book proposal, ‘Medico-Legal Aspects of Cerebral Death.’ ”
14. Hannibal Hamlin, “Life or death by EEG,” JAMA 190, no.2 (October 12, 1964): 120–22, 120.
15. R. S. Schwab, F. Potts, and A. Bonazzi, “EEG as an aid in determining death in the presence of cardiac activity (ethical, legal and medical aspects),” EEG and Clinical Neurophysiology 15 (1963): 147–48. See also “Criteria to determine death are based on data from EEG,” Medical Tribune, July 9, 1962, 1, 22. An accompanying editorial remarked how modern medicine is challenging distinctions between life and death, as well as the reliability of signs of cardiac function and respiration. Of note, no mention is made of transplantation.
16. “EEG termed guide to resuscitation,” Medical Tribune, November 8, 1963, 1, 27.
17. Hannibal Hamlin, “Life or death by EEG,” JAMA (October 12, 1964): 120–22, 121–22.
18. William Sweet, interview with the author, September 9, 1991.
19. Earl Ubell, “Death Defined,” New York Herald Tribune, Sunday, February 14, 1965. See also a Boston newspaper account of Hamlin’s address before the American Medical Association that informed the JAMA paper: Loretta McLaughlin, “Brain Waves Sign of Life,” Boston Herald-Traveler, 1964. In both of these newspaper accounts and in other press coverage of this work at MGH, no connection to transplant is made by Hamlin, Schwab, or the authors.
20. R. S. Schwab, F. Potts, and A. Bonazzi, “EEG as an aid in determining death in the presence of cardiac activity (ethical, legal and medical aspects),” Electroencephalography and Clinical Neurophysiology 15 (1963): 147–48, 147.
21. See letter from Rosoff’s secreatry to Schwab, Ocotober 31, 1968, copied from Rosoff’s personal papers and used with his permission. The Soviet conference is in itself of interest. Negovsky, some of whose work was translated in the West, invited Schwab and directed a large institute devoted to “reanimation” utilizing a range of techniques, such as transfusing super-oxygenated blood. A wide range of work was resented by the Soviets focusing on how to prolong and revive the presumably comatose and brain-injured. However, they were very interested in how to prognosticate recovery of brain functioning. See G. Pampiglione, “A Report on the Symposium on Recovery after Resuscitation Moscow, November 1968,” manuscript copy, Joan Schwab Papers; and Robert Schwab, “Report on the 5-Day Symposium on Recovery Period After Resuscitation Pathophysiology and Treatment in Experiment and Clinics,” manuscript copy, Joan Schwab Papers. Schwab presented the triad as submitted to the Ad Hoc Committee. The papers were published as Vostanoveetyelnee Peryod Posle Ozhyblyenya, (or “The Recovery Period after Resuscitation”) in 1970. See also V. A. Nogovsky, “Some physiopathologic regularities in the process of dying and resuscitation,” Circulation 23 (March 1961): 452–57. Here, Negovsky made the point that life support interventions now were being applied to manage the process of dying. That process required a shift to assessing the state of the brain in order to assess the appropriateness of such interventions.
22. Sidney Rosoff, interview with the author, April 20, 1998.
23. “The EEG in Establishing Brain Death: A 10 Year Report with Criteria and Legal Safeguards in the 50 States,” manuscript copy, Joan Schwab Papers. The copy states it was presented June 8, 1967, and that the criteria—now simplified to no hypothermia or anesthesia, “no reflexes or spontaneous breathing, or muscle activity, no clinical or EEG response to noise or pitch” and repeatable “24 or 48 hours later”—are notes to have “been fully approved by the authorities at the hospital, various medial examiners in the area, two district attorneys, and the Medico-legal department at Harvard Medical School.”
24. Fred Plum, interview with the author, April 20, 1998.
25. As described further below, Fisher was motivated by facing comatose patients in large numbers as the director of the new Stroke Service, which consolidated the examination and care of these patients. Henning Pontoppidan, who was the first director of Beecher’s respiratory unit, credits the collection of such patients with the enhanced visibility of the problem of futility, as rare decisions by many practitioners segued into many such decisions by fewer specializing practitioners. Pontoppidan, interview with author, November 19, 1991.
26. C. Miller Fisher, interview with the author, August 30, 1991.
27. For example, see successive volumes through much of the twentieth century of Roy R. Grinker’s text Neurology, published in serial editions in 1931, 1937, 1943, 1949, 1960, and 1966.
28. C. Miller Fisher, interview with the author, December 6, 1996. Fisher’s chronology of developing and using such an examination is corroborated by Fellows, who worked with him at this time. Vincent Perlo, interview with the author, July 30, 1991.
29. For a brief summary, see Raymond D. Adams and Edward P. Richardson Jr., “Salute to C. Miller Fisher,” Archives of Neurology 38 (March 1981): 137–39.
30. C. Miller Fisher, “The neurological examination of the comatose patient,” Acta Neurologica Scandinavica: Supplementum 36, no. 45 (1969): 6.
31. Ibid., 5.
32. Ibid., 6.
33. Ibid., 10.
34. Ibid., 53–54.
35. Donald McNealy and Fred Plum, “Brainstem dysfunction with supratentorial mass lesions,” Archives of Neurology 7 (July 1962): 26–48.
36. Prominent among these is a paper co-authored with Ray Adams on syndromes found with patients who had occlusions of the basilar artery, a main arterial supply to the brainstem. See Charles S. Kubik and Raymond D. Adams, “Occlusion of the basilar artery: a clinical and pathological study,” Brain 69, no. 2 (June 1946): 6.
37. Fred Plum, interview with the author, April 20, 1998.
38. Eelco F. M. Wijdicks, “The transatlantic divide over brain death determination and the debate,” Brain 135, no. 4 (2011): 1321–31.
39. Herbert Jasper et al., Reticular Formation of the Brain (Boston: Little, Brown and Company, 1958).
40. Plum and Posner, Diagnosis of Stupor and Coma, 2nd ed. (Philadelphia: FA Davis Company, 1972), 2, 5.
41. Plum, interview with the author, April 20, 1998.
42. R. S. Schwab, F. Potts, and A. Bonazzi, “EEG as an aid in determining death in the presence of cardiac activity (ethical, legal and medical aspects),” Electroencephalography and Clinical Neurophysiology 15 (1963): 147–48; Judith Hockaday, Frances Potts, Eve Epstein, Anthony Bonazzi, and Robert S. Schwab, “Electroencephalographic changes in acute cerebral anoxia from cardiac or respiratory arrest,” Electroencephalography and Clinical Neurophysiology 18 (1965): 575–86; J. F. Alderete, F. R. Jeri, E. P. Richardson Jr., S. Sament, R. S. Schwab, and R. R. Young, “Irreversible coma: a clinical electroencephalographic and neuropathological study,” Transactions of the American Neurological Association 93 (1968): 16–20.
43. Alderete et al., “Irreversible coma.”
44. Hockaday et al., “Electroencephalographic changes in acute cerebral anoxia.”
45. Adams, interview with the author, January 25, 1997.
46. William Sweet, interview with the author, April 22, 1998. Indeed, Sweet talked about approaching the criteria to make kidneys available. Given this advocacy, the rejection of his views underscores the peripheral importance of this issue.
47. Beecher to G. M. Williams, Professor of Surgery at Johns Hopkins, December 19, 1969, Box 13, Folder 14, Beecher Papers.
48. Adams, interview with the author, January 25, 1997.
49. Sugar and Gerard, “Anoxia and brain potentials,” Journal of Neurophysiology 1 (1938): 558–72.
50. J. F. Alderete, et al., “Irreversible coma,” 20.
51. Robert Young, interview with the author, April 30, 1998.