Dr. Edmund D. Pellegrino, M.D.
Controversies in the Determination of Death
Personal Statement of Edmund D. Pellegrino, M.D.
The Chairman’s first obligation concerning any Council report is to ensure that it fairly and accurately reflects the opinions of the Council members and that the evidence and research supporting those opinions is complete and reliably presented. Having participated personally in the preparation of this white paper, I believe it satisfies those conditions.
Like any Council member, the Chairman is free to express his personal views on the debated issues. To that end, I offer my own interpretations of some of the evidence and arguments employed in the white paper. I do so in the spirit of “good faith” urged in the white paper’s closing exhortation “…to re-examine the human experience of death in light of new evidence and novel argument.”
After extended deliberation, the Council made these recommendations: (1) to reaffirm the ethical propriety of the “dead donor rule” (DDR); (2) to reaffirm the ethical acceptability of the neurological standard (total brain failure, including the brain stem) as well as the cardiopulmonary standard (irreversible cessation of both cardiac and respiratory functions); and (3) to reject the use of patients in permanent vegetative states as organ donors.
I am in general agreement with these recommendations, but I differ with some of the arguments advanced for them. This contribution focuses on four issues: (1) the matter of definitions, (2) the significance of the DDR, (3) the relative merits of the neurological and cardiopulmonary standards, and (4) the places of prudential reasoning and futility in remedying some of the problems with both standards.
The Matter of “Definition”
The so-called “definitions” of death fall into two categories: the philosophical and the empirical. The first seek a conceptual understanding of the essential differences between life and death. The second seek to determine the clinical signs, tests, or criteria which separate life and death most accurately. Ideally, a full definition would link the concept of life (or death) with its clinical manifestations as closely as possible. So far, this linkage has been the subject of controversy because of its pivotal role in ethically justifying the removal of vital organs from donors in transplantation protocols.
My colleague, Professor Gómez -Lobo, in his personal commentary, outlines the requirements for a philosophically valid definition of death and the failure of present attempts at definition to satisfy those requirements. I agree with Professor Gómez -Lobo’s analysis. Here, I need only remind us about the difficulties inherent in all definitions. As Aristotle would have it: “Clearly, then, a definition is the easiest of all things to demolish, while to establish one is the hardest.”1 As the debates in this white paper attest, this is especially true for so-called definitions of death and life. Each is defined in terms of the absence of the other. Rather than being defined conceptually, each is identified with a set of empirically observable criteria.
Shortly after the proposal of the Harvard criteria for total brain death was advanced, the philosopher Hans Jonas described the central philosophical difficulty in this way:
Reality of certain kinds—of which the life-death spectrum is perhaps only one—may be imprecise in itself, or the knowledge obtainable of it may be. To acknowledge such a state of affairs is more adequate to it than a precise definition which does violence to it. I am challenging the undue precision of a definition and its application to an imprecise field.2
On Jonas’s view, the intrinsic connection between the empirical realities of biological death and its conceptual formulation is too weak to support the moral weight required to justify removal of vital organs. Jonas recognizes the logical fact that states of imprecision in our perception of reality cannot give rise to precision in our concept of reality. Similarly, states of doubt cannot give rise to certitude.
Aristotle issues the same kind of warning when making ethical judgments:
Our discussion will be adequate if it has as much clearness as the subject matter admits of, for precision is not to be sought for alike in all discussions, any more than in all the products of the crafts.3
This warning is especially relevant for any attempt to arrive at moral judgments in the presence of reasonable doubt about the clinical criteria for death.
These same criticisms apply to proposed philosophical definitions of death in the absence of indisputable objective signs of death. Four such attempts to define death in philosophical terms are considered, i.e., loss of integrative functioning of the whole organism, failure to engage the environment spontaneously by respiration, loss of consciousness and sentiency, and the separation of some vital principle from the body.
Integrative Functions: Loss of somatic integration of the organism as a whole as a result of brain death was proposed nearly thirty years ago by the then-President’s Commission.4 In recent years, this criterion has been cast into doubt by a long series of clinical observations. A list, from the work of neurologist Alan Shewmon, is presented in Chapter Four of the white paper.5 Strenuously debated in the past, the criterion of somatic integration enjoys waning support today.6
Engagement with the Environment: The Council’s white paper proffers a more attractive, philosophical argument, i.e., loss of the capacity by the apneic patient for active spontaneous engagement with the environment through the function of breathing. The patient lacking this capacity is said to be “dead,” even if respiratory function and cell metabolism are sustained by mechanical ventilation because they are not, then, the result of “spontaneous” respiration. However, other patients kept alive “artificially”—by pacemakers, defibrillators, vasopressors, ventricular assist devices, artificial nutrition and hydration, etc.—are not, by that fact alone, considered to be “dead.” Patients with respiratory paralysis due to poliomyelitis or cervical spine trans-section have lived with the assistance of respirators for many years. Few would embalm or bury these supposed “living cadavers” before their hearts had stopped irreversibly.
Loss of Sentience and Mental Capacity: Some have proposed a philosophical definition of death based in total loss of conscious mental capacity. Mental capacity, it is argued, is a fundamental capability specific to human life. If it is completely lost, it is argued, the subject is no longer entitled to the moral status of a member of the natural kind we call “human.” “A human body that can only function biologically without inward mental life does not sustain a moral agent.”7 This argument, again, identifies death of the organism with death of one organ —the brain. The sufficiency of the proposed criteria for “death” of the brain is, however, precisely what is currently being debated.
Separation of Soul and Body: Finally, the metaphysical definition of death as separation of the body from its vital principle is still held as the authoritative definition by many worldwide. Plato put it most bluntly: “Death in my opinion is nothing else but the separation from each other of two things, soul and body.”8 No precise congruence of this concept with any observable set of clinical facts has ever been agreed upon.
In the end, attempts at philosophical definitions lack the empirical precision required for a definition, as Jonas has pointed out. Until an empirically sound criterion for death is found, the lack of a conjunction between concept and reality remains a problem. Most such attempts now end in some form of circular reasoning—defining death in terms of life and life in terms of death without a true “definition” of one or the other. Plato recognized the circular reasoning in this way: “….about life and death, do you not admit that death is the opposite of life?”9
Each philosophical definition builds on clinical criteria that are still debated. While necessary , these criteria are not sufficient per se to define death. The search must continue for better physiological criteria if there is to be satisfactory closure of the gap between philosophical concepts and clinical reality.
The only indisputable signs of death are those we have known since antiquity, i.e., loss of sentience, heartbeat, and breathing; mottling and coldness of the skin; muscular rigidity; and eventual putrefaction as the result of generalized autolysis i of body cells.10 There is no biomarker to tell us when this trajectory begins. Instead, we make judgments that the process of autolysis is underway with sufficient certainty to embalm, dissect, cremate, mourn, and bury the body long before signs of putrefaction are evident.
The possibilities of organ transplantation have forced us to shorten the time for observation and deliberation in the interests of preserving the vitality of organs to be transplanted. In place of a prudent waiting period, we must declare a donor to be dead as soon as possible, by one or the other of two standards, both of which are subject to increasing uncertainty about their validity. Most of this debate is covered extensively in the Council’s forthcoming report on organ transplantation. Allusion here will be made only to the comparative reliability of the neurological and the cardiopulmonary standards.
Preserving the “Dead Donor Rule”
The DDR has been the anchor for the moral and social acceptability of organ transplantation protocols from their earliest days. This rule requires assurance of the death of the donor as the first step in any ethically legitimate transplantation protocol (other than those involving healthy, living donors). In addition, the death of the patient must not be hastened, nor end of life care compromised in any way, to accommodate transplantation protocols. No protocol can claim moral sanction without fidelity to this rule.
Today, as serious doubts about the reliability of brain death criteria and the neurological standard persist, some bioethicists are proposing modification or even abolition of the DDR. Robert Truog and Franklin Miller hold that since both the brain death and cardiac death criteria have “…never been fully convincing,” the only ethically valid precondition for transplantation is the consent of the donor or his or her surrogates prior to the withdrawal of life support.11 Robert Veatch also doubts the reliability of both the neurological and the cardiopulmonary standards. He proposes that donors or families should be allowed to choose the definition of death that fits best with their personal values.12
Recently, the DDR has been seriously compromised in three protocols involving infants in which the recommended time after the cessation of the heartbeat was reduced from the minimum recommended time of two- to five-minutes to seventy-five seconds.13 This is an unacceptably dangerous assault on the DDR. The uncertainties of death determination in infants are notoriously formidable. Extrapolations based on these cases are exceedingly perilous and border on the irresponsible.
Also recently, Miller and Truog14 have expanded their attack on the DDR, calling for its abolition and replacement by the autonomous consent of the donor or his or her surrogate. Their line of reasoning is a utilitarian device: They abolish the DDR, replace it with autonomous choice, and declare that such a move makes removal of vital organs from the donor ethically defensible. Indeed, on their view, this revision is ethically laudable because it removes the “veneer” hiding the fact that both the neurological and cardiopulmonary standards result in killing donors. They reject using the word “killing” as counterproductive, although they admit it could be called killing. Their attempt to justify the taking of vital organs from living donors will undoubtedly be seriously debated.
Relaxation of the DDR is a morally unacceptable and logically specious way to deal with the uncertainties of the criteria for death of the donor. It leaves the choice of the criteria for death to individual preference, amounting to eventual abolition of any stable criteria for death. Some additional dangers are: the use of assisted suicide to facilitate organ donation; legitimizing the use of patients in permanent vegetative states or of “less-than-perfect” infants as donors. It exposes “undeclared” patients to “presumed” consent to donation. Given the expanding cultural and ethical pluralism of the U.S. about all aspects of life and death, eliminating the DDR promises a future of moral and legal chaos. Above all, it exposes the vulnerable or gullible patient to an increased danger of exploitation for the benefit of others.
The need for organs, the desire to prolong life, and the potential “good” to be done are forces difficult to control when death can be defined on one’s own—or one’s guardian’s—terms. As experience has repeatedly shown, personal autonomy without moral constraint ends in divisive moral atomism. As difficult as the search for a common definition of death may be, that difficulty cannot justify abandoning the effort to establish a common definition or manipulating the DDR to meet the need for more organs.
Which Standard Should Be Preferred—the Neurological or the Cardiopulmonary?
Since the 1970s, the dominant criteria for death of the donor have been those of the neurological standard (a heart-beating donor). In recent years, there has been a resurgent interest in the cardiopulmonary standard (a non-heart-beating donor) under the auspices of protocols known as “controlled donation after cardiac death,” protocols that have recently received approval of prestigious medical bodies in the United States and Canada.15 The Council’s opinion is that both the neurological and the cardiopulmonary standards are ethically acceptable, but some Council members have expressed reservations about the use of the cardiopulmonary standard in “controlled donation after cardiac death protocols.”
In my view, the reasons that favor the neurological standard are not compelling. The clinical tests and signs that support it are as subject to doubt as those of the cardiopulmonary standard. The philosophical arguments for both suffer from the same conceptual and empirical difficulties already identified and examined.
Some favor the neurological standard over the cardiopulmonary standard because of the possibility of auto-resuscitation with the latter. Auto-resuscitation has occurred in rare instances, but not in controlled non-heart-beating protocols if the time after withdrawal of respirator support was five minutes or more.16
Others oppose the cardiopulmonary standard because of doubts about possible lack of irreversibility of cardiac function. How, they ask, can a supposedly “dead” heart be transplanted from a “dead” donor and “revived” to function in a recipient?17 But the fact is that the donated heart is transplanted from a physiological environment that could not support the metabolism of myocardial cells to a physiological environment suitable to cell metabolism. The heart from a patient declared dead by the cardiopulmonary standard can beat when transplanted because autolysis occurs at different rates in different organs. An organ removed when cellular autolysis has not yet advanced to a point that would cause organ failure can function when placed in a physiologically normal, natural or artificial environment that meets its metabolic needs.
Furthermore, if valid, the same argument would apply to other organs as well. Kidneys, intestines, lungs, etc., are taken from donors pronounced “dead” by both the neurological and cardiopulmonary standards. These organs are taken from a supposedly “dead” donor in full expectation that they will function in a recipient.
Taking everything into account, I believe there is as much—or more—moral assurance of death of the donor with the cardiopulmonary as with the neurological standard. In fact, with the cardiopulmonary standard, there is a higher degree of certainty of death than there is with a heart-beating donor, because heart, lung, and brain have all ceased functioning.
Can Transplantation Be Ethically Legitimated?
Do the considerable doubts about both the neurological and cardiopulmonary standards make organ transplantation ethically indefensible? One way to examine this question is as a problem of prudential or practical reason—that is, all things considered, how can we act ethically in the face of relative clinical doubt? This is the context within which many important clinical decisions must be made in modern medicine and have always been made in medicine’s past.
Many, if not most, serious clinical decisions must be made in a matrix of concrete, contingent, and uncertain realities in the present and the future. This is why medicine has rightly been called by one of its wisest practitioners “…a science of uncertainty and an art of probability.”18 The moral danger of removing vital organs from a donor whose death is not reasonably certain is significant. However, if absolute certainty is not possible, a sufficient degree of moral certainty to warrant such an action may be attainable if the requirements for prudent decision-making are satisfied.
Prudence in both the ethical and clinical aspects of bedside decisions is an indispensable practical virtue for the good clinician. It is the capability to choose, in any difficult decision, the means that will most closely fit the good intent of the clinical act. Prudence serves the first rule of all ethics, to do good and to avoid evil, even in the most difficult circumstances. Prudence activates two principles of moral action in the face of doubt: the precautionary principle (i.e., when in doubt, act to avoid the gravest danger) and the principle of proportionality (i.e., properly balance the benefits and the burdens of treatment). Prudential reasoning deserves closer attention than it has received in the current debate about the determination of death.
Relative moral certitude does not substitute for scientific certitude. But, properly weighed, it can give a legitimate warrant for necessary action in the face of unavoidable uncertainty. This is the situation within which ordinary decisions in daily life are made. Clinical prudence seeks to avoid both the error of inaction, which would deprive the recipient of a needed transplant, and the error of premature action, which would deprive the donor of life. Fidelity to beneficence and the prudential approach to decisions aim to avoid both the paralysis of inaction and the harmful use of ineffective medical treatments. Prudence must not be confused with self-protective cowardice. It is the decision to act for a good end in the morally optimal way despite persistent uncertainty about the outcome.
Futility in the Decision Process
The moral quality of a non-beating heart donation can be improved as can the degree of compliance with the DDR by use of the principle of futility. When considering the moral quality of a decision to initiate a donation protocol, the relevance of clinical futility deserves attention. Futility is that state in the history of a patient’s disease when he or she is beyond medical rescue, i.e., beyond the powers of medical technology to help. Clinical futility is present when any medical intervention is: (1) ineffective , i.e., unable to change the natural history of a disease or its trajectory towards death; (2) non-beneficial , i.e., unable to satisfy any good or value perceived by the patient or his or her surrogate; and (3) disproportionately burdensome to the patient, physically, psychologically, or financially.19 Balancing the relationship among those three criteria is at the heart of prudent, precautionary, and proportionate action. This formulation accommodates the physician’s expertise with respect to effectiveness and the patient’s values with respect to benefits, and it results in shared decision-making regarding the proportionality of benefits and burdens.
Care of the patient is never futile. Provision of comfort, pain relief, easing of suffering, and palliative care are always morally mandatory. They are the indispensable and absolute requirements for morally valid care at the end of life for every patient. However, under the conditions of ethically legitimate futility, medical interventions may be discontinued. Indeed, if continued, clinically futile medical treatment can convert beneficence into maleficence. Valid use of the principle of futility is not medical abandonment. It does include patient participation as well.
When the criteria for clinical futility are properly met and attested to by the attending physician together with the family, the patient’s life-support can ethically be removed. Following removal of life-support and cessation of both respiration and cardiac function, initiation of a “controlled donation after cardiac death” (controlled DCD) protocol may begin.
Ethically sensitive controlled DCD recovery protocols are being developed and used by institutions providing transplant services. My purpose is not to compare and contrast those protocols or their comparative ethical acceptability. Some of the ethical constraints by which any transplant protocol may be judged, however, include the following: both the prospective donor and the prospective recipient must be treated as patients in their own right—each with his or her own attending physician whose major concern is the patient’s needs and not the needs of the protocol; the donor’s physician must, together with the family, determine whether further treatment would be futile; quality of life can be a valid factor only if expressed by the patient or the patient’s morally valid surrogate; care and relief of pain and suffering are never futile, nor is palliative care. The usual ethical requirements for morally and legally valid consent must be observed; the declaration of death must be neither premature nor overdue.
The prospective donor and the prospective recipient must therefore be treated in accord with the principle of beneficence. If we grant, as I do, that Jonas’s judgment about the imprecision of decisions at the life-death interface is cogent, then his advice to “lean” in the direction of protecting life is ethically sound and grounded in a valid precautionary principle. Within these ethical constraints, the dignity of the donor and recipient, and the benefits of organ transplantation, can be protected. The act of organ donation could retain its moral credibility without compromising care of the donor.
Remaining Ethical Uncertainties
Clinical and ethical uncertainties remain with both brain death and controlled DCD protocols. When the respirator is removed, there is the possibility that electrical activity of the myocardium may return after the prescribed five minute waiting period. Return of electrical activity is not synonymous with effective mechanical pumping. Nevertheless, some may feel constrained to consider cardiopulmonary resuscitation. This possibility can be anticipated by a prior, valid “do not resuscitate” (DNR) order. This order is invoked only when medical treatment is truly futile, death is foreseeable within a short period, and further cardiopulmonary resuscitation can itself be expected to meet the criteria of futility. A valid DNR order implies that death is to be allowed to occur naturally as the final event of an illness whose inevitable natural clinical trajectory is death within a foreseeable interval. Under these circumstances, cardiopulmonary resuscitation would not be in the patient’s interest.
The danger of hastening the decision to discontinue treatment is present with both the neurological and the cardiopulmonary standards for determination of death.ii In both, the only safeguard against such a decision is the ethical fidelity of physicians and families to the patient’s welfare. The decision not to resuscitate must be based on the futility of all technical medical procedures, not the need, however great, for a transplantable organ by a particular patient.
Not to be ignored are the social costs of shifting the criteria for organ procurement from brain death to the cardiopulmonary standard. Such a move places the good of the dying potential donor over that of a dying recipient in need of a life-saving organ. This change may result in delays, greater procedural complexity, loss of vitality of the more sensitive organs, and, perhaps, fewer organs available for transplantation. Here, the vexed issue of social versus individual good arises and sharp differences of opinion between and among interested parties seem unavoidable.
I have chosen to give priority to the welfare of the patient before he or she becomes a donor on grounds that harm must not be done even if good comes from it. No person should be sacrificed as a means for the good of another. This is a moral precept that recognizes the intrinsic worth of every human being.
The significant controversies and doubts about when the donor is dead cannot justify abandonment of the DDR. Rather, it should stimulate research into how to extend the viability of potentially transplantable organs. It might also accelerate the process well under way to grow organs extra-corporeally for transplant purposes. This has already been done successfully with at least a diseased urinary bladder and a trachea grown from cells from the patient’s body. In the case of the trachea, that organ was removed from a dead donor, stripped of the donor’s cells, and a new trachea was formed using the remaining cartilaginous skeleton and the recipient’s own stem cells.20
Ultimately, the central ethical challenge for any transplantation protocol is to give the gift of life to one human being without taking life away from another. Until the uncertainties and imprecision of the life-death spectrum so clearly recognized by Hans Jonas are dispelled, his moral advice must be our guide for all transplant protocols:
We do not know with certainty the borderline between life and death, and a definition cannot substitute for knowledge. Moreover, we have sufficient grounds for suspecting that the artificially supported condition of the comatose patient may still be one of life, however reduced—i.e., for doubting that, even with the brain function gone, he is completely dead. In this state of marginal ignorance and doubt the only course to take is to lean over backward toward the side of possible life.21
ii. One glaring example is a $1.5 million grant over three years to New York City from the federal Health Resources and Services Administration to study the use of a special ambulance fully equipped for removal of organs for transplantation. This vehicle would be rushed to the site of traumatic incidents, ready to remove organs as soon as the donor’s death was declared. (Cara Buckley, “City to Explore a Way to Add Organs” ( New York Times June 1, 2008. Available at http://www.nytimes.com/2008/06/01/nyregion/01organ.html. Accessed December 10, 2008).
2. Hans Jonas, “Against the Stream,” in Philosophical Essays, From Ancient Creed to Technological Man , (Englewood Cliffs, New Jersey: Prentice-Hall, 1974), pp. 132-140. See also Controversies in the Determination of Death, A White Paper of the President’s Council on Bioethics , 33-37.
3. Aristotle, Nicomachean Ethics (1094b 13-15) in The Complete Works of Aristotle , The Revised Oxford Translation, Vol. 2, ed. Jonathan Barnes (Princeton, New Jersey: Princeton University Press, 1971), p.1730
4.President’s Commission for the Study of Ethical Problems in Medicine and Biomedical and Behavioral Research, Defining Death: Medical, Legal and Ethical Issues in the Determination of Death (Washington, D.C.: U.S. Government Printing Office, 1981).
5.Cf . D. Alan Shewmon, “The Brain and Somatic Integration: Insights into the Standard Biological Rationale for Equating ‘Brain Death’ with Death.” Journal of Medicine and Philosophy 26, no. 5 (2001): 457-78. See also D. Alan Shewmon, “Mental Disconnect: ‘Physiological Decapitation” as a Heuristic for Understanding ‘Brain Death’,” Working Group on the Signs of Death 11-12 September 2006, ed. H.E. Msgr. Marcelo Sanchez Sorondo ( Vatican City : Pontifica Academia Scientiarum, 2007), pp. 292-333.
6.Controversies in the Determination of Death, A White Paper of the President’s Council on Bioethics. See also K. G. Karakatsanis. “‘Brain Death’: Should It Be Reconsidered?” Spinal Cord 46, no. 6 (2008): 396-401.
10. See Liang Cheng and David Z. Bostwick, eds., “Decomposition,” in Essentials of Anatomic Pathology 2d ed. ( Totowa , New Jersey : Humana Press, 2005), 87-92; Arpad Vass, “How Long Does Cellular Metabolism Persist After Death?” Scientific American September 2008, p. 116; Cell Death and Differentiation 15, no. 7 (July 2008).
13.Mark M. Boucek, Christine Mashburn, et al., “Pediatric Heart Transplantation after Declaration of Cardiopulmonary Death,” New England Journal of Medicine 359, no. 7 (August 14, 2008):709-714. Special Task Force, American Academy of Pediatrics, “Guidelines for the Determination of Brain Death in Children,” Pediatrics 80 (1987): 298-300.
15. Institute of Medicine , Organ Donation: Opportunities for Action ( Washington , D.C. : National Academy Press, 2006; Society of Critical Care Medicine, “Recommendations for Nonheartbeating Organ Donation.” Critical Care Medicine 29, no. 9 (2001) : 1826-31; Joint Commission. “2009 Accreditation Requirements: Accreditation Program: Critical Access Hospital ,” (2008).
16.Bruce Ben-David, et al., “Survival after Failed Intra-operative Resuscitation: A Case of the Lazarus Syndrome,” Journal of Anesthesiology and Analgesia 92 (2001):690-692. See also M.A. Da Vita, “The Death Watch: Certifying Death Using Cardiac Criteria,” Progress in Transplantation 11, no. 1 (2001):58-66.
18.William Osler, Sir William Osler Aphorisms; From His Bedside Teachings and Writings, collected by Robert Bennett Bean, edited by William Bennett Bean (Springfield, Ill.: Charles C. Thomas, Publisher, 1968), p.129.
19.Edmund D. Pellegrino, “Decisions at the End of Life: The Use and Abuse of the Concept of Futility,” in The Dignity of the Dying Person (Proceedings of the Fifth Assembly of the Pontifical Academy for Life (February 24-27, 1999 ) , ed. Juan De Dios Vial Correa and Elio Sgreccia (Citta del Vaticano: Libreria Editrice Vaticano, 2000), 219-241. See also Edmund D. Pellegrino, “Futility in Medical Decisions: The Word and the Concept,” Healthcare Ethics Committee Forum 17, no. 4 (December 2005):308-318.