Historical Perspective

The increasing gap between the number of organs available for transplantation and the number of patients listed for transplantation has become the rate-limiting step in reducing both wait times and wait-list deaths in patients with end-organ disease awaiting transplantation. Prior to the passage of the first U.S. brain death law in the state of Kansas in 1970,¹ donation after cardiac death (DCD, or non-heartbeating donation) was the primary mode of organ donation in this country. Death in DCD donors was determined according to traditional cardiopulmonary criteria. Early organ procurement strategies were somewhat crude and variable, and consequently, warm ischemia time (time from donor circulatory arrest to cold perfusion) in the DCD donor was often prolonged and outcomes were poor.² The impact of the type of graft on DCD outcomes was not apparent until experience with organs from donors declared brain dead (DBD) grew. Transplant centers in several states such as Nebraska, Ohio, North Carolina, and Illinois flourished after adoption of DBD in their respective states, and transplant volumes grew.³

The DBD phenomenon was a culmination of critical care physicians’ growing ability to maintain physiologic organ function in patients with little or no hope of neurologic recovery from severe insults to the central nervous system. A new debate was sparked over the precise definition and timing of death and the concept of futile care. This concept was introduced at a CIBA Foundation meeting in England in 1965 and subsequently endorsed with formal diagnostic criteria by Harvard Medical School in 1968.^1,4 Acceptance of this medically, philosophically, and legally novel concept radically changed the face of transplantation. The revolutionary ability to certify death while perfusing the donor with oxygenated blood guaranteed procurement with minimal warm ischemia and graft damage and better recipient outcomes. As early experience with the DBD organs demonstrated superior outcomes, the use of DCD organs declined and was subsequently abandoned.⁵

As a result of the success seen with DBD organ donation, the number of U.S. transplants performed annually increased exponentially. Based on Organ Procurement and Transplantation Network (OPTN) data, in 1988, the first year for which reliable national data were available, 10,794 deceased-donor transplants were performed.⁶ Just 6 years later, annual volumes increased by nearly 50% to 15,210 total transplants. Most dramatically, the number of lung grafts from deceased donors increased from 33 to 708.⁶ Moreover, intestinal transplantation gained clinical success with the introduction of DBD donors (in addition to refined medical and surgical techniques). The first case was performed in 1990; by 1994, 96 patients with intestinal failure had received intestinal transplants.⁶ Concomitantly, advances in critical care resulted in reduced mortality in patients with end-stage organ disease, thereby resulting in increasing additions to and decreased attrition from the wait list, often referred to as the growing “gap” between supply and demand in transplantation. For example, despite a burgeoning number of transplant centers, rapid increase in transplants performed, and increased utilization of living donors, in 1995, only 33% of listed registrants waiting for kidney transplant (33,167) were transplanted (11,081).⁶ Unfortunately, the rate of transplantation fell to 10% of the list in the subsequent era of 1998 to 2002.⁷

Exacerbating the impact of this trend, numbers of young, previously healthy DBD donors stagnated due to several statutory changes in the areas of gun control, automobile safety (air bags, seat belts, lowering of legal blood alcohol limits), and cyclist helmet use, thereby reducing traumatic fatalities and consequently changing the face of DBD organ donors in the process.⁸ The demographics and mode of death of the typical DBD donor transitioned from a young, healthy person rendered brain dead as a result of a devastating head trauma toward an older person rendered brain dead from a neurovascular insult. The change in median donor age and mode of death ultimately eroded some of the benefit of utilization of the DBD donor and prompted a search for additional options.

Scientific and legal strategies such as xenotransplantation (use of grafts derived from donors), presumed consent (requiring individuals to formally opt out of organ donation in order to be excluded as a donor at the time of his or her death), and living donation have been explored as tools to meet the growing demand for organ donors. These strategies have been rejected because of strong philosophical objections or have been met with little or no success. The transplant community has therefore revisited the use of DCD donors with much enthusiasm.

As noted by DeVita, in 1993, the University of Pittsburgh Medical Center (UPMC) introduced the nation’s first institutional policy to permit and regulate DCD donation.⁹ The need for such a policy arose when several patients and their families asked to participate in organ donation after previously electing withdrawal of life-sustaining treatment. This was a request that fell outside the parameters of donation policies and guidelines then in effect. The UPMC policy became the first concrete model for the use of cardiopulmonary criteria to determine death for the purposes of organ procurement,¹⁰ and it highlighted a milestone in the evolution of the practice of transplantation in this country. Since, DCD utilization has been adopted by many organ procurement organizations (OPOs) and hospitals nationwide. By December 2006, OPTN bylaws required that all OPTN members have a DCD donor protocol in place.¹⁰ Moreover, the Joint Commission now requires that all accredited institutions develop and implement standardized DCD policies.¹¹

After more than a decade and a half of ongoing scrutiny surrounding ethical issues and assessment of outcomes, several key issues regarding DCD organ donation remain controversial in both the lay and medical communities—namely, (1) how best to identify potential DCD donors, thus avoiding the financial and emotional burden of “failed” DCD donation, (2) how best to optimize DCD donor management, which by its very nature leaves little room for error, and (3) how best to standardize DCD procurement protocols to ensure a multidisciplinary effort and reproducible results. These issues will be explored in the remainder of this chapter after a brief discussion on the current status of DCD donation.

Current Status of DCD Donation

Outcomes

Though fraught with ethical controversy over the years, the real barrier to widespread acceptance of DCD graft utilization is based primarily upon the poor outcomes seen in the early DCD experience. Suboptimal organ function characterized by primary nonfunction (PNF), delayed graft function (DGF),¹² and/or abbreviated graft survival have traditionally been a threat to success with DCD donors organs because of the warm ischemic insult associated with cardiopulmonary arrest. Although these observations were valid at the time, they were accumulated during the early experiences with transplantation and are thus inherently confounded by era bias.

The primary lesson from the early DCD era was that the metabolically active renal cortex, biliary epithelium, pulmonary alveoli/central airways, and islets are sensitive to ischemia, with warm ischemic injury manifesting as acute tubular necrosis (ATN), ischemic-type biliary strictures (ITBS), bronchial dehiscence, and impaired beta cell function. These complications have been postulated to translate into and account for both poor initial graft function and long-term complications, seen particularly in the early era.^12,13,14 Droupy and Abt, however, in separate studies, report that outcomes have improved; that intermediate and long-term patient/graft survival in recipients of controlled DCD kidney and liver grafts, respectively, are equivalent to or approach that of DBD.^15,16 Per Droupy, DCD and DBD renal grafts followed for 10 years demonstrate equivalent survival despite a higher initial incidence of DGF for the DCD cohort. Salvaggio and colleagues, in an analysis of UNOS/OPTN data, concur¹⁷ (Figure 212-1, A).

Figure 212-1 A, Current outcomes after kidney transplant, DBD versus DCD. B, Current outcomes after DCD SPK transplant, DBD versus DCD. C, Current outcomes after DCD pancreas alone transplant, DBD versus DCD. DBD, donors declared brain dead; DCD, donation after cardiac death; SPK, simultaneous pancreas-kidney.

(From Salvaggio P, Davies D, Fernandez L et al. Outcomes of pancreas transplantation in the United States using cardiac-death donors. Am J Transpl 2006;6:1059-65.)

DCD liver outcomes have improved, though less dramatically. Some results of liver transplants using DCD donors were discouraging; both graft and patient survival rates were thought to be significantly lower when compared with DBD donors.¹⁸ Morbidity rates were higher as well.^19,20 However, reviews isolating recent data—as, for example, data outlined by Abt—demonstrate that 1- and 3-year patient survival rates for liver DCD are now similar to those from DBD, although lower graft survival rates for DCD liver grafts persist¹⁶ (Table 212-2).

TABLE 212-2 One- and Three-Year Outcomes after DCD and DBD Liver Transplantation

Much of the available outcome data for pancreatic DCD organs are derived from cases of simultaneous pancreas-kidney (SPK) transplants, demonstrating pancreatic graft and patient survival rates similar to those for DBD¹⁷ (see Figure 212-1, B). Utilization of pancreas-alone DCD grafts under the same protocols as SPK DCD grafts, while less frequent over the last decade, have been favorable¹⁷ (see Figure 212-1, C). Ongoing clinical experience will determine whether outcomes will reach those of SPK DCD grafts.

Oliveira et al. have, in the largest single-center series to date, demonstrated that lung grafts from DCD donors can also confer graft and patient survival rates equivalent to those from DBD donors.¹³ Of interest, these outcomes have been achieved in many settings in recipients who have been disproportionately more ill prior to transplant but deemed reasonable potential DCD recipients because of the long potential wait for DBD grafts. Consequently, clinicians now consider use of grafts that were previously routinely declined.²¹

The net effect of current practice and improved outcomes has been to shift the paradigm of the binary cadaveric donor (DBD versus DCD) to a spectrum of standard criteria to extended criteria, with the DCD donor potentially falling along several points on that spectrum based on specific factors. Careful evaluation of those factors, as discussed later, may allow the transplant community to unravel the issues preventing expansion of the successful use of DCD organ donors in contemporary transplant practice.

Identification and Categorization of the Potential DCD Donor

An important initial step in the process of DCD organ transplantation is recognizing the potential suitable donor. A significant consideration is the need to minimize organ ischemia in the presence of an unanticipated uncontrolled cardiac arrest; thus while organ procurement from DCD donors under uncontrolled conditions is technically feasible, it remains rare in contemporary practice.^22,23 Similarly, graft quality is compromised in situations in which a patient’s wishes regarding organ donation are unknown. Organ suitability declines while attempts are made to locate family members to obtain consent. The once popular practice by some institutions to manage potential DCD donors brought to the emergency department by placement of vascular and/or intraperitoneal catheters in order to infuse cold organ preservation solution before consent for procurement became available²⁴ has been largely abandoned. The practice stimulated contentious debate from opponents in both the medical and lay communities; unlike several European countries, no U.S. state at the time of the writing of this chapter has adopted presumed consent into law.

Remaining potential DCD donors are patients consented for donation with impending cardiopulmonary death, the timing of which is either unpredictable or predictable based upon patient/family-requested withdrawal of care, or unpredictable with premature arrest before withdrawal. Understandably, each type of DCD confers a varying risk of ischemic injury. A discussion of the management of DCD donors is facilitated by use of a classification scheme developed at a donor conference convened in 1994 by Maastricht, Netherlands, investigators.²⁵ The Maastricht Categories define potential donors by the circumstances under which their cardiovascular death occurs. A distinction is made between those donors whose cardiopulmonary failure is uncontrolled or emergent (categories 1, 2, and 4) and those donors whose death by cardiopulmonary criteria occurs in a controlled, planned fashion by withdrawal of futile life-sustaining support (category 3). Maastricht Categories are outlined in Table 212-3.

TABLE 212-3 The Maastricht Classification for DCD Donors

From Koostra G, Daemen JHC, Oomen APA. Categories of non-heartbeating donors. Transplant Proc 1995;27:2893-4.

Because only some single-center reviews report DCD results by Maastricht category, it is difficult to stratify nationwide DCD outcomes by category. Category 3 donors constitute the majority of the DCD procurements reflected in data from U.S. centers; a small but unknown fraction are category 4 donors, wherein the patient meets brain death criteria but subsequently loses circulation.

Of note, recent initiatives in the northeast United States involve training prehospital personnel in the rapid conversion of preconsented victims of unsuccessful resuscitation after cardiopulmonary arrest (category 2) to potential DCD donors.²⁶ For the sake of uniformity, the remainder of this chapter will be devoted to discussion of category 3 donors.

In addition to understanding the classification scheme and expected outcomes based upon absence or presence of controlled ischemia, the intensivist and OPO personnel must be familiar with the diagnoses and clinical circumstances qualifying a patient as a potential DCD donor. Again, candidates are patients in whom withdrawal of futile life-sustaining treatment is being planned. As shown in Table 212-4, the UNOS Critical Pathway for DCD,⁶ typical patients may have the following characteristics: absent or hyperactive respiratory drive, lack of adequate respiratory muscle strength, severe hypoxemia, or inadequate circulation in the absence of inotropic or vasopressor drugs. Such patients are usually supported by ventilators or mechanical circulatory assistance such as ventricular-assist devices (VAD) or intraaortic balloon pumps. They are often patients who have also suffered a severe neurologic insult. Conscious patients are usually suffering from degenerative neuromuscular diseases or end-stage cardiopulmonary disease and are often ventilator or VAD dependent. These patients or their families may decide to discontinue their support devices and request that their organs subsequently be donated.

The next important step in identification of the potential DCD donor is predicting when rapid physiologic deterioration and death are expected occur in a period of less than 30 to 60 minutes (depending on the organ to be procured) after withdrawal of life-sustaining treatment.²⁷ Failure of a potential donor to progress to cardiac death within the prescribed time disqualifies the potential donor owing to the extent of warm ischemic injury sustained by the organs. Factors such as age, comorbidities, and preterminal pressor requirement have been shown to have predictive value, but no strict criteria have been universally adopted.²⁸ Kaufman et al. have proposed four readily obtainable clinical criteria: (1) requirement for vasopressors to support arterial blood pressure, (2) absence of primary brain injury, (3) history of 6 or more days on mechanical ventilation, and (4) respiratory rate less than 20 breaths/min (in the absence of mechanical ventilatory support).²⁹ They noted that the presence of two or more of the indicators accurately predicted death within 60 minutes after life-supporting treatments were withdrawn, with a sensitivity and specificity of 81% and 78%, respectively.

Adoption of accurate predictive indices would enable intensivists and OPO staff to more precisely identify potential organ donors, help minimize the financial impact and resource drain for hospitals and procurement teams of the donor who “fails to progress,” and prevent unnecessary stress and disappointment for families during a psychologically vulnerable time.

Lastly, familiarity with relative and absolute contraindications for DCD donation, some of which overlap with those associated with DBD, some unique to DCD, is important. These include the multiply operated abdomen, active sepsis, active or recent extracranial primary malignancy, and human immunodeficiency virus (HIV) and hepatitis B infection. With regard to virologic status, OPOs are versed in rapid serologic testing necessary to rule out latent viral infections and should be involved as early as feasible to initiate testing.

Principles of DCD Donor Management

Appropriate management of the DCD organ donor requires integration of several fundamental principles exercised to protect the rights and interests of the donor and simultaneously prevent the care of the organ from superseding the care of the dying patient. The debate arises in the paradox that can emerge from attempts to protect those interests while preserving suitability of the potential grafts. Hence, the role of the intensivist and/or palliative care physician (terminal care of the patient, pronouncement of death) and the managing OPO (facilitating organ procurement) must be rigidly defined; the two factions must travel distinct paths to achieve their goals. As Ozark states, “as a general rule two discussions—whether to forego life-sustaining therapy and whether to donate organs—must be made separately and on their own individual merit.”³⁰ Ideally, discussion regarding withdrawal of life-sustaining treatment should come first so as not to be biased by the issue of transplantation.

The push for optimal palliative care has been a hallmark of recent critical care management initiatives.³¹ The dying patient who also wishes to be a DCD organ donor presents a special challenge, requiring care that is not only comparable to that afforded to all dying patients but also sensitive to the concerns already described. The Society of Critical Care Medicine (SCCM) has offered recommendations specific to DCD donation.^32,33 These guidelines, supplemented by individual transplant center reports and the UNOS pathway (see Table 212-4), provide direction for intensivists caring for patients who wish to become DCD donors. It is vital that all healthcare providers involved in this process be comfortable with, and knowledgeable about, their specific role such that the patient’s wishes can be respected.

Pain relief is the single most important goal in palliative care in the final hours of life, and there is firm ethical, legal, and medical justification for use of analgesics and anxiolytics in this scenario. Some patients require higher doses than others, so doses are given with the knowledge that unintended effects such as hypotension or respiratory depression may compromise organ viability. It is critical, then, that the interest of the dying patient be represented by a completely different entity than that responsible for representing the interests of the donor. If any question arises as to the practitioner’s ability to maintain an objective position, consultation from the hospital’s palliative care and ethics teams should be sought.

In 2008, in the first criminal case brought against a transplant surgeon for the death of a donor, the defendant allegedly administered high doses of analgesic and anxiolytic to a potential donor to hasten his death prior to procurement.¹¹ Although the surgeon was acquitted, the case highlighted the potential legal ramifications of the recovery team’s involvement in the care of a dying patient. There is a consensus that “medications given to provide comfort are reasonable, even if they might hasten death” but “no medication whose purpose is to hasten death should be given to the patient.”³⁴ Failure to attend to potential DCD donors’ comfort in contemporary practice is considered suboptimal end-of-life care but is not in any circumstance managed by anyone other than the physician(s) caring for the patient.

Active debate exists as to the optimal location of withdrawal of support. Arguments for ICU withdrawal stem from proponents who prefer to provide grieving families as “normal” a setting as possible to grieve, albeit briefly, at the bedside of their loved one. Others argue that effective and expedient progression to donor mode allows for the most successful procurement and can only transpire in the operating room. Currently there is no standard, and each facility is responsible for dictating protocol for their institution.

DeVita notes that “The initial University of Pittsburgh policy called for the withdrawal of care to occur in the operating room, which offered the advantage of minimizing the need to transport the patient after death and permitting the prepping and draping of the patient before death. This protocol was denounced for subjecting the patient to ‘a desolate, profanely “high tech” death’ surrounded by ‘masked, gowned, and gloved strangers.’³⁵ The initial experience in Pittsburgh found some truth to the proposition that presence of family at the patient’s bedside at the time of death may be more important to patients and families than organ donation or location of death. When three of the first four families approached about non-heartbeating donation agreed to consent only if they could be physically present at the time of death, the Pittsburgh policy was changed to allow families into the operating room or to move the withdrawal of care to an operating room ‘holding’ area. The area selected for withdrawal of support should allow family members to be present, accommodate the necessary monitors and equipment, and be close enough to the operating room to allow rapid transport immediately after death.”⁷ Other programs followed in kind; according to the 2000 IOM report on DCD, the family’s need to be present and involved in the dying process is generally widely cited and respected in the development of hospital policies on the setting for withdrawal of care.³⁴

DCD Procurement: an Opportunity for Standardization

Because every organ from a DCD donor sustains some degree of unavoidable warm ischemic damage, several aggressive methodologies to protect graft viability have been proposed. Strategies can be considered in the premortem or intraoperative phases.

Operative

The conduct of the operative procedure is dictated by the tenets already mentioned: the procurement team is not physically present at the time of death, and recovery of organs must be accomplished expeditiously with careful coordination of numerous personnel, equipment, and resources. To do so, the operative team prepares and drapes the patient upon arrival to the operating room. The team outlines the necessary instruments and maneuvers requested of the OPO staff to ensure a seamless procedure. The team, gowned and gloved, is escorted from the operating room and is notified by OPO staff if the patient progresses within the prescribed time frame. Optimally, after withdrawal and prior to incision, the OPO staff will complete a data form (Figure 212-2) allowing for recording of minute-by-minute hemodynamic and oxygenation data. The benefit of compiling data in such a fashion is that as DCD procurements become standardized, data can be collated without the encumbrance of shifting definitions of the initiation of warm ischemic time (and hence acirculatory status) that currently confounds comparison of outcomes between hospitals and OPOs. For instance, the definition of the warm ischemia start time remains variable, with some advocating that a threshold of systolic blood pressure of less than 80 mm Hg be used, others, a mean arterial pressure of 50 mm Hg, others a systolic blood pressure less than 50 mm Hg, and yet others when the arterial oxygen saturation decreases to less than 80%.¹⁶ The lack of a universal definition renders comparison of outcomes between centers and organs difficult.

Figure 212-2 Donation after cardiac death (DCD) monitoring form.

(Reprinted with permission from Lifegift Organ Donation Center, 2009.)

Whereas the asystolic interval to be observed remains a matter of institutional policy, it seems prudent to recommend that standardized criteria be developed. Highly sensitive maneuvers used to document the absence of circulation, such as intraarterial pressure monitoring or echocardiography, may be helpful if a short asystolic interval will be used.

Once the asystolic interval has passed, the operative team returns and infuses cold perfusate via either a premortem-placed cannula or a standard terminal aorta cannula placed after rapidly accessing the abdomen. Rapid but careful in situ cold dissection ensues, as the potential for vascular injury is increased without the benefit of pulsatile flow to assist with the identification of aberrant anatomy. The pancreas is sacrificed if a replaced right hepatic artery originating from the superior mesenteric artery appears to be present. Organs are packaged and implanted on the recipient end as soon as possible in order to mitigate the impact of cold ischemic injury.

The two most common contingencies the team must be prepared for are unexpected cardiac arrest while awaiting withdrawal and failure to progress after withdrawal. Appropriate intravenous access, a ventilator, and a special supply cart must be available and stocked with an oxygen tank, cardiac monitor, and an adequate supply of sedatives and narcotics. Of course, the patient’s wishes regarding resuscitation in order to ultimately donate must be determined from the patient or family as early in the process as possible.

Future Directions

Despite the success with kidneys, livers, and pancreatic grafts realized with use of DCD grafts, there remain a number of unmet challenges and questions regarding the transplantation of organs from DCD donors. They fall within the realms of medical as well as ethical concerns.

Owing to unavoidable warm ischemia, heart and intestinal grafts are difficult to utilize when procured from DCD donors. Nevertheless, cardiac DCD transplantation is technically possible. The first human cardiac transplant 40 years ago was made possible using a DCD donor heart, which started and functioned well after a single electric shock.⁵¹ It is conceivable, then, with the advent of preconditioning therapy, a new era of DCD cardiac transplantation may begin.

A second area of likely future focus concerns the real numeric advantage of using DCD grafts. What, if any, increase in available donor organs can be expected by promoting the utilization of DCD donors? Will the push for identification of DCD donors simply convert would-be DBD donors to donate as DCDs? Additionally, if the number of wait-list deaths decreases, and the gap is reduced, is the cost and burden shifted from wait-list complications to posttransplant complications with longer length of stays, increased readmissions, increasing complexity of diagnostic evaluations and immunosuppressive drugs regimens, and an increased rate of recipient morbidity and mortality? The transplant community will be in a better position to answer these questions once clean, prospective data are collated nationwide as suggested earlier.

Ethically, how, if at all, will DCD donation affect public trust in the healthcare system and the organ procurement and transplantation process? Furthermore, on the recipient end, will the potential for differential results of DBD versus DCD transplantation mandate that a recipient be afforded the prerogative to decline an organ based on knowledge of physicians’ concerns regarding the DCD process? As noted by the UPMC group and the IOM, an emphasis on patients’ and families’ wishes is paramount in the success of any DCD program. The public will be guided by medical information, but the practice of transplantation, which is donor driven, will be guided by the way in which that information is buttressed by open communication with the layperson.

Furthermore, the concern by the lay public is that physicians caring for patients who are potential donors have shifted the focus of care from the dying patient, and that there exists now more than ever, the latitude to violate the “dead donor rule” (comprising two complementary ideas: that the patient must be dead before the initiation of organ procurement; and that organ procurement itself must not be the cause of the donor’s death).⁴⁰ This concern may translate in the public mind into a fear that their likelihood of receiving aggressive life support will be compromised by consenting to organ donor status. These misperceptions can be particularly damaging when the overriding goal of the transplant community is to maintain and build public support of maximizing organ donation.

Conclusion

The widening gap between suitable donors and patients in need of transplant continues to be the single issue that keeps solid-organ transplantation from achieving its full potential for relieving suffering and improving survival in patients with end-stage organ disease. As current practice and outcomes have shifted the paradigm of the binary cadaveric donor (DBD versus DCD) to a spectrum of standard criteria to extended criteria, with factors mentioned earlier affecting where on this spectrum a DCD organ may fall, the ultimate impact of the estimated unrealized annual 22,000 DCD donors⁸ on the actual number of organs available for transplantation will remain unclear until sufficient data obtained under similar protocols are obtained. Whereas ethical questions regarding DCD donation persist, the process, as it is increasingly practiced in a standardized fashion, has proven to accommodate the needs of dying patients, as well as those awaiting transplantation, with improving success.³⁸ A number of organizations including the SCCM,³² UNOS,⁶ and the IOM³⁵ have endorsed the concept and issued relevant guidelines.

As experience grows, attitudes change, and outcomes continue to improve, DCD donation may yet have a significant impact on the number of organs available for transplantation and thus the quality of life for those waiting for and ultimately receiving cadaveric organs.

Key Points