Controversies in the Determination of Death
The President's Council on Bioethics
One of the implications of rejecting the neurological standard, as we saw in Part I of Chapter Five, is that non-heart-beating donors could become the exclusive source of organs for procurement and transplantation. In this chapter we examine non-heart-beating organ donation—specifically, the practice of controlled donation after cardiac death—in greater depth.
As was noted in Chapter One, the neurological standard for the determination of death was formulated in the 1960s, in the early years of successful cadaveric organ transplantation. Since then, the practice of transplanting organs from the dead has become commonplace, in large measure because the “heart-beating cadaver” has been accepted as a medical and legal reality. Heart-beating cadavers, however, are not the only source of organs from the deceased: Human beings whose deaths have been determined according to the more traditional cardiopulmonary standard may also provide organs. The earliest successful organ transplantations from cadaveric donors, in fact, made use of the organs of individuals whose deaths were determined in this way.i
A key challenge to procuring organs from non-heart-beating cadavers is posed by warm ischemic damage, caused by the lack of blood perfusion while the organs are still in the body of the newly deceased donor. To prevent or limit this damage, the time between the cessation of circulation (“effective asystole”) and the procurement of organs must be minimized. For this purpose, the final cardiac contractions must be controlled : In the moments following asystole, the body must be prepared for surgeons either to remove the organs promptly or to maintain their viability for transplantation in the ensuing hours. For the moment of cardiac arrest to be controlled in this way, the prospective organ donor must be a patient from whom a ventilator and perhaps other forms of life-sustaining treatment will be withdrawn, either in accordance with the patient's wishes or the wishes of a surrogate. In other words, a potential non-heart-beating donor, in the vast majority of cases, is an individual who is ventilator-dependent but not yet deceased according to today's neurological standard.ii The ventilator is then removed, the patient is watched and kept comfortable until the heart stops circulating blood through the body, a waiting period is observed (usually two to five minutes), and then the surgical procurement of organs begins.
Procuring organs in this way was very rare until the early 1990s—in most transplant centers throughout the United States, only heart-beating organ donors were used.iii This restriction meant that some individuals who had suffered accidents or injuries that left them with little hope of recovery were not eligible to become donors, even if a responsible decision to remove life-sustaining interventions was made. In some cases, families who knew that their relatives wanted to be donors in the event of an accident were disappointed to find out that making this gift was not possible. This fact—along with concerns about the general shortage of available organs compared to the growing need among potential recipients—led some pioneering institutions to develop and implement non-heart-beating donor protocols.iv These protocols codified the institutional practices for managing the withdrawal of treatment and for optimizing the conditions for organ procurement. The early controlled DCD protocols included both logistical procedures and ethical safeguards.
The work of these first institutions raised concerns among some ethicists, health care professionals, and members of the general public. For the purposes of this report, these concerns can be divided into two categories: concerns about the care of the dying patient at the end of life, and concerns about initiating organ procurement before the patient is dead. The second set of concerns is treated at length in Part II of this chapter. Regarding the first set of concerns, we shall address the principal points here while encouraging the interested reader to consult the companion volume to this white paper, Organ Transplantation: An Ethical Inquiry by the President's Council on Bioethics ; in Chapter Three of that report, the Council offers a more thorough investigation of these concerns about care of the dying patient in the context of organ transplantation.
Those who fear that the widespread practice of controlled DCD may adversely affect the quality of care at the end of life have raised several questions, including whether families offered the option of controlled DCD are pressured to make a decision in favor of withdrawal of life support; whether the steps necessary to optimize the circumstances of death for transplant purposes interfere with good palliative care for the patient in his or her last moments; whether the family's emotional needs will be respected, considering that loved ones must be kept “out of the surgeons' way” immediately after the patient's heart stops beating; and whether the patient's death will be hastened to ensure that the procured organs are maximally viable.v
In response to such concerns, institutions seeking to expand their transplant programs to include non-heart-beating donors have incorporated ethical safeguards into their transplantation protocols. To aid institutions in the development of ethically sound protocols, two reports by the Institute of Medicine on non-heart-beating donation were published.1 The Society for Critical Care Medicine also published an analysis of the issue, detailing the requirements for the practice to be carried out in an ethically sound manner. 2 More recently, a report was published by a National Conference on Donation After Cardiac Death.3 This report, while supportive of the practice, addressed medical and ethical concerns that have arisen as more institutions have started to procure organs in this way.
The practice of controlled DCD has expanded rapidly in the last decade. In 1997, for example, there were seventy-eight cases of organ procurement from DCD donors; in 2007, there were 793 cases. In 1997, only six organ procurement organizations (OPOs) reported five or more controlled DCD procurements within their respective areas; in 2007, forty-one OPOs reported five or more controlled DCD procurements.4 This expansion has been aided by requirements from such national bodies as the United Network for Organ Sharing (UNOS) and the Joint Commission (formerly the Joint Commission on Accreditation of Healthcare Organizations or JCAHO) that hospitals either institute controlled DCD protocols or, at the very least, address the practice in their statement of hospital policy.5
II. The Rush to Declare Death and the Problem of Irreversibility
Are those who donate organs under a controlled DCD protocol actually dead at the time of donation? It might seem somewhat surprising that this is a matter of controversy. After all, at the time of procurement, the donor's heart has stopped and he or she is no longer breathing—either spontaneously or with ventilator support. Thus, the individual would seem to meet the first (more traditional) standard for determining death, that is, in the wording of the UDDA, “irreversible cessation of circulatory and respiratory functions.”
The difficulty here stems from the crucial requirement that cessation of circulatory and respiratory functions be irreversible. In truth, there is reason to doubt that the cessation of circulatory and respiratory functions is irreversible, in the strict sense, in every case of controlled DCD. To call the loss of functions irreversible, it must be the case that the functions could not possibly return, either on their own or with external help. It is often possible, however, to cause circulation and respiration to return by administering cardiopulmonary resuscitation (CPR). If this were attempted after the “declaration of death” in controlled DCD, some patients would indeed regain—for a brief time, at least—a heartbeat and some capacity to breathe. If this were to occur, the patient would certainly not have been “resurrected,” but instead would have been (according to the cardiopulmonary standard of death) resuscitated, i.e., prevented from dying. Thus, the prior “declaration of death” would turn out to be questionable. The patient was, it could be argued, no more dead than a person who collapses in his or her home, loses heartbeat, and is resuscitated by paramedics who arrive moments later.
It is important to note that this hypothetical scenario of resuscitating a patient who has been prepared for a controlled DCD procurement is merely a “thought experiment.” In reality, attempting to revive such a patient would be ruled out ethically because the practice of controlled DCD is premised on the assumption that the individual's family has decided to allow withdrawal of life-sustaining interventions and would, therefore, want to abstain from any efforts to prevent the patient's death (perhaps by consenting to a “do not resuscitate” order). For this reason, many have argued that the word “irreversible” in this context should be understood in a weaker sense than that spelled out above: It should be understood to mean “cessation of circulatory and respiratory functions under conditions in which those functions cannot return on their own and will not be restored by medical interventions.”vi This looser sense of the term “irreversible” would seem to be a better fit in this context.
The problem of ascertaining irreversibility in controlled DCD procurement is illustrated by a series of procurements and transplants carried out in Denver, Colorado, between 2004 and 2007.6 In each of the three procurements conducted under a controversial controlled DCD protocol, the heart was taken from a severely brain-injured (but not “brain dead”) infant whose family had authorized removal of life-sustaining treatments. In one case, the procurement team waited three minutes after cardiac contractions had stopped before removing the heart from the donor. In the other two cases, they waited only seventy-five seconds—a shorter time than either the Society of Critical Care Medicine or the National Conference on Donation After Cardiac Death has recommended.7 In each of the three cases, the hearts were transplanted and the recipients were still alive six months post-transplant.
In analyzing these controversial cases, it is important to note that the fact that the recipients of these hearts were restored to a state of health does do not imply that the donors' conditions were anything less than terminal and dire. To state this another way, the heart, itself, in each of these cases, was healthy enough to become an effective organ again in the body of the recipient, even though the overall condition of the body of the dying donor made it impossible for the donor's health to be restored.
The ethically troubling point is brought into sharp relief when we think of the moment of procurement. In each case, the heart of the infant had stopped. But was the infant actually dead? Because the heart tissue itself was not beyond the point of resuscitation once transferred to the recipient, it is clear that (as in some other controlled DCD cases) the cessation of cardiac function in these infants might have been reversed, albeit only for a short time, had there been a desire to do so. What prevented this from occurring was not a certainty that resuscitation attempts on the would-be donor would fail, but rather the decision made by the parents that no effort to resuscitate should be made. In short, the donor had only suffered “irreversible cessation of circulatory and respiratory functions,” if the term “irreversible” is taken in the weaker sense that Youngner and others have proposed.
In any event, such debates about the proper meaning of “irreversible” do not address a more pressing ethical dilemma—the question of whether there is something wrong with rushing to make a determination of death because of external pressures to procure organs as expeditiously as possible. There is reason to worry that this practice—if carried out on a wide scale—could make the donor's death seem like a mere formality, with “patient dies” becoming simply another item to check off on a list of events required for a successful controlled DCD procurement.
There might be something to learn about this ethical concern from anecdotes of physicians who are called upon to declare death in more routine hospital cases. It is said that experienced physicians will take their time in proceeding to the patient's room to declare him or her dead. They may even deliberately linger to give the patient's body ample time to complete the dying process. This is not done, we can imagine, because of a lack of confidence in the physician's own judgment, but rather out of a sense that, in approaching the mysterious threshold between life and death, circumspection and caution are required. Rushing to make a declaration as quickly as possible is not viewed as respectful or appropriate. Unfortunately, such a deliberate demeanor is harder to maintain if the death is going to be “made use of” for the sake of other needy patients.
Health professionals should help families understand that the process of mourning their loved ones is likely to be disrupted by a controlled DCD protocol and that this disruption is part of the sacrifice they are making in giving their loved ones' organs to the sick. They should also help families understand the controversies about “irreversibility” that will always be a part of controlled DCD procurement. In addition to such steps that should be taken on a case by case basis, there ought to be a broader public discussion and debate about the propriety of controlled DCD as a standard practice in hospitals. With the spread of controlled DCD practice to more and more hospitals, it is clear that patients and their families will need to come to terms with this still novel aspect of end-of-life decision making.
The integrity of the procurement system has just as much at stake in the context of controlled DCD procurements as it has in the context of procurements made after diagnoses of total brain failure. The clinical preference for acquiring the most viable organs possible cannot be allowed to undermine the principle that death should be declared only when there are sound medical reasons to do so. In light of this, more research should be done to investigate the question of auto-resuscitation (mentioned in the footnote on page 84). The assurance that the heart will not restart on its own within the relevant time frame, combined with an informed decision by the patient and family in favor of controlled DCD, may or may not be sufficient as a moral warrant for declaring death, but it is certainly necessary. Again, the principle here is clear: Death should be declared on the sole basis of the medical facts of the patient's case. Since ruling out the possibility of auto-resuscitation is an essential step here—a point that has been emphasized by the IOM, the Society of Critical Care Medicine, and others—the prediction must be based on sound and publicly available evidence.
1. Institute of Medicine, Non-Heart-Beating Organ Transplantation: Medical and Ethical Issues in Procurement (Washington, D.C.: National Academy Press, 1997) ; and Institute of Medicine, Non-Heart-Beating Organ Transplantation: Practice and Protocols (Washington, D.C.: National Academy Press, 2000).
2. Society of Critical Care Medicine, “Recommendations for Nonheartbeating Organ Donation,” Crit Care Med 29, no. 9 (2001): 1826-31.
3. J. L. Bernat et al., “Report of a National Conference on Donation after Cardiac Death,” Am J Transplant 6, no. 2 (2006): 281-91.
4. Institute of Medicine, Organ Donation: Opportunities for Action (Washington, D.C.: National Academy Press, 2006), 142-43 ; Kevin O'Connor, Senior Vice-President, New England Organ Bank, Personal Communication, September 12, 2008.
5. For Joint Commission requirement, see Joint Commission, “2009 Accreditation Requirements: Accreditation Program: Critical Access Hospital,” (2008), Standard TS.01.01.01, Element 4. Available online at http://www.jointcommission.org/NR/rdonlyres/8093C734-2317-4CDC-A0BF-BFB2DF0FBAD1/0/CAH_AllChapters.pdf (accessed Sept.12, 2008). For UNOS requirement, see Appendix B to UNOS Bylaws, available online at http://www.unos.org/policiesandBylaws2/bylaws/UNOSByLaws/pdfs/bylaw_31.pdf (accessed Sept. 12, 2008).
6. M. M. Boucek et al., “Pediatric Heart Transplantation after Declaration of Cardiocirculatory Death,” N Engl J Med 359, no. 7 (2008): 709-14.
7. Society of Critical Care Medicine, “Recommendations for Nonheartbeating Organ Donation,” 1826-31; and J. L. Bernat et al., “Report of a National Conference on Donation after Cardiac Death,” 281-91.
i. Two useful sources on the early use of non-heart-beating donors are M. A. DeVita, J. V. Snyder, and A. Grenvik, “History of Organ Donation by Patients with Cardiac Death,” Kennedy Inst Ethics J 3, no. 2 (1993) : 113-129; and C. A. Zawistowski and M. A. DeVita, “Non-Heartbeating Organ Donation: A Review,” J Intensive Care Med 18, no. 4 (2003): 189-97. In addition to kidney transplantations from such cadavers in the 1950s and 1960s, the first liver transplantation in 1963 and the first heart transplant in 1967 were performed by taking the organ from a patient after life-support had been removed and the patient had become asystolic.
ii. Many candidates for controlled DCD are patients who are “very close to brain dead”—i.e., who are tested for total brain failure but show minimal signs of brain function that are nonetheless sufficient to preclude such a diagnosis. There are other candidates for controlled DCD, however, who are dependent on a life-sustaining technology for reasons other than a traumatic brain injury, e.g., patients with a high-spinal cord injury or patients at the end-stage of a neurodegenerative disease such as amyotrophic lateral sclerosis (ALS). For a case study of a fully conscious patient who chose to become a controlled DCD donor upon removal of the ventilator, see J. Spike, “Controlled NHBD Protocol for a Fully Conscious Person: When Death Is Intended as an End in Itself and It Has Its Own End,” J Clin Ethics 11, no. 1 (2000): 73-7.
iii. One exception is the University of Wisconsin Hospital, where controlled DCD has been performed contin u ously since 1974. See J. Lewis, et al., “Development of the University of Wisconsin Donation after Cardiac Death Evaluation Tool,” Prog Transplant 13, no. 4 (2003): 265-73.
iv. The first modern controlled DCD protocol was instituted at the University of Pittsburgh Medical Center in 1992. For details, see M. A. DeVita, J. V. Snyder, and A. Grenvik, “History of Organ Donation by Patients with Cardiac Death, ” Kennedy Inst Ethics J 3, no. 2 (1993): 113-29.
v. Some detail about this last concern: The patient who is designated a candidate for controlled DCD must not only be ventilator-dependent, but also likely to expire (become asystolic) within a short window of time after removal of the ventilator. The longer a patient removed from ventilation “lingers” before expiring, the more likely are the organs destined for transplantation to be damaged by warm ischemia. Institutional protocols specify the maximum time—generally between thirty minutes and two hours—that a patient removed from ventilation can linger before the organs are judged to be no longer viable, For a description of a technique used to estimate how long a particular patient will take to expire (and, thus, whether he or she can be a candidate for controlled DCD) see Lewis, et al., “ Development, ” 265-73. Concerns about the temptation to hasten death stem from the fact that hopes are raised and many expensive preparations are made on the expectation that death will occur quickly enough to allow procurement of transplantable organs.
vi. See, for instance, S. J. Youngner, R. M. Arnold, and M. A. DeVita, “When Is ‘Dead'?” Hastings Cent Rep 29, no. 6 (1999): 14-21. For the purposes of the argument made here, it is assumed that two to five minutes of waiting time is enough to guarantee that the heart will not auto-resuscitate, i.e., that it will not begin to beat again on its own. There is insufficient scientific evidence to decide whether this assumption is always justified, as has been noted by many commentators, including the National Conference on Donation After Cardiac Death. See J. L. Bernat, et al., “Report of a National Conference on Donation after Cardiac Death,” Am J Transplant 6, no. 2 (2006): 282.