Abstract
Objective: The purpose of this study is to investigate the potential availability of hearts from adult donation after cardiac death (DCD) donors within an acceptable hypoxic period. Methods: We retrospectively reviewed a donor database from the University of Wisconsin Organ Procurement Organization Donor Tracking System between 2004 and 2006. The DCD population (n = 78) was screened using our inclusion criteria for DCD cardiac donor suitability, including warm ischaemic time (WIT) limit of 30 min. In the same period, 70 hearts were donated from brain-dead donors. Results: Of 78 DCD donors, 12 (15%) met our proposed DCD cardiac donor criteria. The mean WIT of these 12 DCD donors was 21 min (range 14–29 min). When inclusion criteria are further narrowed to (1) age ≪30 years, (2) WIT ≪20 min and (3) male gender, only two out of 12 met the criteria. Conclusions: Based on our proposed DCD cardiac donor criteria, the potential application of DCD cardiac donors would represent an increase in cardiac donation of 17% (12/70) during the 3-year period. When the criteria were narrowed to the initial ‘ideal’ case, only two donors met such criteria, suggesting that such ‘ideal’ DCD donors are rare but they do exist.
1 Introduction
The shortage of donor organs is the most critical problem in transplantation. The disparity between the number of grafts and the number of possible recipients is still growing [1]. In fact, the Organ Procurement and Transplantation Network (OPTN) reports indicate that the number of registrations on the heart waiting list is approximately 3000 in the recent 3-year period and 15% of patients die in 1 year while awaiting a donor heart (http://www.optn.org/). To maintain or increase the present transplantation rate, donation after cardiac death (DCD) donors have been proposed as another donor source. The concept of heart transplantation using DCD donors existed in the 1960s before brain death became legally accepted and the first successful clinical human heart transplantation in an adult was performed using a DCD heart allograft in 1967 [2]. Recently, Boucek et al. reported the short-term results of three infants undergoing successful orthotopic heart transplantation from DCD donors [3]. For non-cardiac organ transplantation, clinical studies from several institutions show that DCD donors are reliable donor sources for organs such as kidney [4], liver [5] and lung [6]. Steinbrook reported a progressive increase in the rate of organ recovery from DCD donors (645 DCDs in 2006, as compared with 189 in 2002) and that these donors accounted for 8% of all deceased donors in 2006 [7]. In particular, at several Organ Procurement Organization (OPO) programmes, DCD donors accounted for more than 20% of all deceased donors (http://www.optn.org/). Although the use of DCD donors for non-cardiac organ transplant has been increasing, the potential for heart transplantation from DCD donors remains unrealised, because of potentially severe myocardial damage due to unavoidable warm ischaemia (from the withdrawal of life support to organ recovery).
According to the 1995 Maastricht categories [8], DCD donors are classified as ‘uncontrolled’ or ‘controlled’ donors. Uncontrolled donors are dead on arrival (category I) or had an unsuccessful resuscitation (category II). Controlled donors have an awaited cardiac arrest (category III) or develop cardiac arrest while brain dead (category IV). The so-called Maastricht category III, withdrawal of donor treatment (usually in the intensive care unit or operating room), is a controlled DCD technique and the only one presently in use in the United States, including our institution.
Over the past decades, many animal experiments have been undertaken to evaluate the feasibility of DCD heart transplantation, using various reperfusion strategies and animal DCD models. Some researchers indicated excellent results in animal DCD heart transplant experiments. Shirakura et al. [9] reported satisfactory functional recovery of canine hearts after a 24-h period of preservation in a 30-min warm ischaemic DCD transplant model, and Gundry et al. [10] achieved long-term survival of baboons receiving transplantation of hearts harvested from 30-min warm ischaemic DCD donors. However, the successful outcomes of these experiments are largely attributable to the application of multiple pretreatments, which, in humans, would be ethically unacceptable. In our recent animal studies [11,12], we evaluated acute post-transplant graft function using a 30-min warm ischaemic pig DCD model without any cardioprotective pretreatments except for heparin, and this graft functional recovery rate was approximately 80% of the pretransplant (normal) value. These animal studies support the potential acceptability of 30-min warm ischaemic DCD cardiac grafts. On the other hand, irreversible change in the myocardial tissue can occur after only 20 min of warm ischaemia [13]. Thus, to simulate the ‘ideal’ case for the initial clinical application of adult DCD heart transplant, the limit of warm ischaemic time (WIT) should be set within 20 min.
Singhal et al. [14] retrospectively reviewed data from 119 DCD donors from the Gift of Life Donor Programme (GLDP) between 2001 and 2003, and screened the DCD population using their proposed DCD cardiac donor criteria, including WIT ≪30 min, age ≪45 years, negative donor history and no cardiac functional abnormalities. Fourteen DCD donors met all their proposed cardiac donor criteria, and they estimated that 12% (14/119) of DCD donors for other organ transplants were potential heart donors, representing a 4% increase over of the number of heart transplants performed during the same time interval. Although all DCD cardiac allografts suffer from warm ischaemic myocardial damage, not observed in brain-dead hearts, this number was estimated by applying the criteria based on current practice of brain-dead cardiac donors and the actual suitable case for the initial adult DCD heart transplant has not been simulated. In the University of Wisconsin Hospital and Clinics Organ Procurement Organization (UWHC-OPO), our local OPO, the number of DCD donors has been significantly increasing since 2003 with the increase of health-care professionals’ support for DCD [15]. Therefore, the purpose of this study is to investigate the potential availability of acceptable hearts from adult DCD donors in the UWHC-OPO network in the recent era and to identify the frequency of ‘ideal’ cases for the initial clinical application of adult DCD heart transplant.
2 Patients and methods
We retrospectively reviewed a donor database from the UWHC-OPO Donor Tracking System between January 2004 and December 2006. Deceased donor demographic data included age, gender, ethnicity and cause of death. In addition, the DCD donor information included data concerning the presence of cardiac abnormalities, such as (1) elevated creatine kinase (CK-MB) or troponin, (2) inotropic support, (3) cardiac contusion or sternal fracture, (4) reduced ejection fraction by echocardiography, (5) significant coronary artery disease, (6) previous history of heart disease or (7) history of illicit intravenous drug abuse. WIT, defined as the time from withdrawal of life support to reperfusion of organs, was also included in the database. The Institutional Review Board committee approved this study.
The inclusion criteria to determine potential DCD cardiac donor suitability were created prospectively before reviewing donor records. Table 1 shows the inclusion criteria for DCD cardiac donor suitability. These criteria included donor age ≪50 years, no cardiac abnormalities and a WIT limit of 30 min. The DCD population was screened using our inclusion criteria to assess the potential suitability for cardiac donation. In addition, we simulated the ‘ideal’ case for the initial clinical application of adult DCD heart transplant using further narrowed inclusion criteria (Table 2 ), which included age ≪30 years, no cardiac abnormalities, WIT ≪20 min and male gender.
Inclusion criteria for DCD cardiac donor suitability.
Inclusion criteria for DCD cardiac donor suitability.
Inclusion criteria for the ‘ideal’ adult DCD cardiac donor for initial clinical application.
Inclusion criteria for the ‘ideal’ adult DCD cardiac donor for initial clinical application.
2.1 DCD programme and evaluation process in the UWHC-OPO
Our institution has a long history of using kidneys from controlled DCD donors [4]. In 1993, we expanded our programme to include transplantation of extrarenal organs such as liver, simultaneous pancreas–kidney and lung [16]. Since 2003, the UWHC-OPO has evaluated all referrals of patients with severe brain damage, but who do not meet criteria for brain death, and whose families have chosen to withdraw life-sustaining therapies. Because DCD donors frequently have preservation of brainstem reflexes, the possibility of continued respiration after extubation must be discussed when obtaining consent from the family. When consent is obtained for DCD, the family is informed that multi-organ donation will not occur beyond 2 h following withdrawal, after which the patient will be returned to the ward, where end-of-life care is given, until they die. During the time that consent is obtained, the family is fully informed about the procedure, including any medications that may be given such as phentolamine and heparin. Once death is declared by cardiopulmonary criteria, an additional 5 min elapse, as described in the 1997 Institute of Medicine Guidelines [17], before solutions are infused and the incision is made.
2.2 Organ procurement and preservation
Our techniques of organ procurement and preservation during DCD are previously described [5,16]. In some instances, femoral arterial and venous cannulas were placed under local anaesthesia. While the patient was fully supported, 30 000 units of heparin and 10 mg of phentolamine were given intravenously to prevent vasospasm and to facilitate subsequent organ flushing. The patient’s physician of record withdrew life support by stopping intravenous medications and extubation. Variable periods of hypotension and hypoxia occurred after the withdrawal of life support and before cessation of all cardiopulmonary function. During this time, the patient was monitored with an arterial line, continuous electrocardiogram and physical examination. Five minutes after the declaration of death, cold UW (UW – University of Wisconsin) solution was infused into the femoral arterial cannula and the femoral venous cannula was opened to gravity. Median sternotomy and a midline abdominal incision were made and the intra-abdominal organs were removed en bloc. In those instances where femoral cannulas were not placed, the distal aorta was cannulated immediately upon entry into the abdomen. Approximately 2–3 l of UW solution was infused in situ, and the entire en bloc preparation was stored in UW solution at 4 °C and separated either immediately or upon return to our centre.
2.3 Statistical analysis
Categorical data were summarised with frequency distributions and percentages. Values of continuous variables were expressed as means ± standard deviations. Continuous variables were compared by the Mann–Whitney U-test, whereas nominal variables were compared using the Fisher’s exact test. A p-value less than 0.05 (two-sided) was considered to be statistically significant. All analyses were performed using the SPSS statistical software program (SPSS for Windows version 14.0, SPSS Inc., Chicago, IL, USA).
3 Results
3.1 DCD donor demographics
There were 78 DCD donors in the 3-year period for whom a complete data set was available. The demographics and organ utilisation of 78 DCD donors are summarised in Table 3 . Mean age was 46 ± 15 years, with 28 (36%) female and 50 (64%) male. Anoxia was the most frequent cause of severe brain damage leading to withdrawal of life support (38%). Mean WIT was 28 ± 16 min (range: 10–103 min). Fifteen (19%) of the DCD donors had femoral cannulas inserted before withdrawal of life support.
DCD donor demographics and organ utilization in the UWHC-OPO from 2004 to 2006.
DCD donor demographics and organ utilization in the UWHC-OPO from 2004 to 2006.
3.2 Potential DCD cardiac allograft application
All 78 DCD donors were screened by using the inclusion criteria for DCD cardiac donor suitability (Table 1). The diagram for screening 78 DCD donors is shown in Fig. 1 . Of the 78 DCD donors, 41 (53%) were under 50 years. Among these DCD donors, 18 (24%) had no cardiac abnormalities, defined as in Table 1, and the most common cause for exclusion was the presence of inotropic support before donation (n = 15, Fig. 1). On applying the criteria of WIT within 30 min to these 18 DCD donors, 12 (15%) ultimately met our proposed DCD cardiac donor criteria. When inclusion criteria were further narrowed, as shown in Table 2, to simulate the ‘ideal’ case for the initial clinical application of adult DCD heart transplant, only two out of 12 met these criteria (Fig. 1).
The diagram for screening 78 DCD donors in the UWHC-OPO between 2004 and 2006. DCD, donation after cardiac death; UWHC-OPO, University of Wisconsin Hospital and Clinics Organ Procurement Organization; WIT, warm ischemic time; CK-MB, creatine kinase-MB; EF, ejection fraction.
The diagram for screening 78 DCD donors in the UWHC-OPO between 2004 and 2006. DCD, donation after cardiac death; UWHC-OPO, University of Wisconsin Hospital and Clinics Organ Procurement Organization; WIT, warm ischemic time; CK-MB, creatine kinase-MB; EF, ejection fraction.
Demographics in 12 DCD ‘potential’ cardiac donors were compared with those of brain-dead cardiac donors between 2004 and 2006 in our UWHC-OPO programme (Table 4 ). Mean age in DCD ‘potential’ cardiac donors was 37 ± 9 years and was significantly older than that of brain-dead cardiac donors (p = 0.02). Anoxia was the most frequent cause of severe brain damage leading to withdrawal of life support among DCD ‘potential’ cardiac donors (42%); meanwhile, for the brain-dead cardiac donors, trauma was the most common cause of death (67%). The mean WIT of these 12 DCD donors was 21 ± 5 min (range 14–29 min).
Demographics in DCD ‘potential’ cardiac donors and brain-dead cardiac donors in the UWHC-OPO program from 2004 to 2006.
Demographics in DCD ‘potential’ cardiac donors and brain-dead cardiac donors in the UWHC-OPO program from 2004 to 2006.
4 Discussion
We identified the potential DCD cardiac donors by reviewing the recent 3-year DCD donor data from the UWHC-OPO Donor Tracking System. These 78 DCD donors for non-cardiac organ transplant were screened using our proposed inclusion criteria to estimate the potential impact of the application of DCD cardiac allografts on the number of heart transplants that could have been performed during the 3-year period. If the DCD cardiac allografts identified by our criteria could have been used for the heart transplant candidates during this study period, 12 additional hearts would have been provided over a time interval in which 70 heart transplants were performed. Thus, between 2004 and 2006, the potential applicability of DCD cardiac donors would have represented an increase in cardiac donation of 17% (12/70) for the UWHC-OPO programme.
Singhal et al. [14] have reported their estimated number of potential DCD cardiac donors by analysing the data from 119 DCD donors from the GLDP between 2001 and 2003 using their criteria for DCD cardiac donor suitability. They indicated that 14 DCD donors met their proposed criteria, and their estimation was a 4% (14/334) increase in the number of heart transplants during the same time interval. Although the actual number of potential DCD cardiac donors was similar to that in our programme, the discrepancy of expected increase in heart transplants could be due to a difference in the size of the donation service area covered by each programme, that is, the total cases of brain-dead heart transplants performed during the 3-year period (70 in the UWHC-OPO programme and 334 in the GLDP). In addition, a disparity in the number of the DCD organ recoveries between these two programmes could have affected this result. In fact, according to the Association of Organ Procurement Organizations annual survey in 2004, DCD donors accounted for 20% of all deceased donors in the UWHC-OPO programme, but only 12% in the GLDP [18]. However, we believe that the recent rapid increase in the number of organs recovered from DCD donors accounts for our observation: national data reported by the OPTN revealed approximately a threefold increase in the number of DCD donations during a recent 4-year period (from 189 in 2002 to 645 in 2006) [7] and our study period was more recent (2004–2006), compared with that of Singhal’s analysis (2001–2003). Currently, DCD donors have entered the mainstream of expanding the donor pool for organ transplant therapy.
One could argue that our criteria for DCD cardiac donor suitability (Table 1) are too optimistic, because these criteria, except for the WIT limitation, are based on current practice for brain-dead cardiac donors and previous published studies [19]. All DCD cardiac allografts suffer from some myocardial damage due to warm ischaemia that is not observed in brain-dead hearts. In addition, hypoxic cardiac arrest can produce catecholamines during the agonal phase, which may also contribute to myocardial damage prior to the start of warm ischaemia. In order to simulate the actual suitable case for the initial adult DCD heart transplant, we have modified our proposed criteria, in particular, donor age, gender and WIT. Older donor age is one of the significant risk factors for post-transplant mortality. The 24th official report by the registry of the International Society for Heart and Lung Transplantation (ISHLT) illustrated a linear relationship between increasing donor age and increasing mortality [1]. Some large single-centre reports [20] have shown that the use of a heart from a donor aged 40 years or more was associated with increased mortality, independent of other risk factors. In addition, Lietz et al. [20] demonstrated a relationship between donor age >30 years and time to the onset of transplant-related coronary artery disease. On the basis of these analyses, the donor age criteria for the initial case of DCD cardiac donors in our study was narrowed from 50 to 30 years. In terms of donor gender, several studies have pointed to the detrimental effect of female donor gender on post-transplant survival [21]. According to the analysis of ISHLT registry data, female donor gender was a risk factor for mortality within 1 year after adult heart transplant only in the period 1995–1998, whereas analysis done for the period 2002–2005 did not identify female donor gender as a significant risk factor [1]. However, this registry data showed that female gender was still a significant risk factor of developing cardiac allograft vasculopathy. Thus, male gender was added to the criteria for initial DCD donor application.
Warm ischaemia, commonly defined as the interval of time between extubation (as the definitive withdrawal of treatment) and the initiation of cold perfusion for organ preservation, is the most critical concern for DCD allografts, because it is not observed in the hearts of brain-dead donors and causes more severe graft injury than cold ischaemia. In general, the cardiac allograft is more vulnerable to ischaemia that other solid organs. Moreover, DCD hearts would be forced to attempt to support an entire circulatory load while functioning in an increasingly hypoxic environment during the period from extubation to cardiac arrest, whereas the brain-dead cardiac allografts do not suffer such load impairments induced by hypoxia because cardiac arrest is initiated by using cardioplegic solution and venous drainage. In our previous animal study, we clearly documented the changes in pressure and volume loads during the increasing hypoxia in the DCD animal model: the peak distending pressure was about 150% of baseline and the peak ventricular distended volume was 132% of baseline [22]. A number of animal DCD heart transplant studies have been conducted to elucidate the feasibility of using DCD cardiac allografts under these unfavourable conditions, to evaluate their functional recovery after transplant and the acceptable limits of WIT [9–12]. Although several species such as pig [11,12], dog [9] and baboon [10] were used and various strategies of reperfusion and organ preservation were applied to these studies, the results support the potential acceptability of up to 30 min of WIT. Gundry et al. [10] achieved long-term survival of baboons receiving transplantation of hearts harvested from DCD donors within 30-min WIT. Therefore, we have proposed a 30-min of WIT limitation as one of the criteria to identify potential DCD cardiac donors in the present study (Table 1). However, at the structural level, global warm ischaemia of 20 min can result in irreversible ischaemic injury, that is, isolated cell death or tiny islands of cell death [13]. If the DCD cardiac allografts could be recovered or preserved within 20 min of WIT, it might be possible to prevent this irreversible injury. Thus, to simulate the ‘ideal’ case for the initial clinical application of adult DCD heart transplant, we considered that the limitation of WIT should be set within 20 min. Applying this narrowed criterion (Table 2) to 12 potential DCD cardiac donors screened by our first proposal (Table 1), only two out of 12 met this criterion between 2004 and 2006. Despite the small estimated number, these ‘ideal’ cases for initial clinical application of adult DCD heart transplant do exist.
Levvey et al. suggested that the appropriate definition of WIT should be considered as the interval between a systolic blood pressure less than 50 mmHg and the initiation of cold flush preservation of the organs for controlled DCD donors [23]. Currently, in the UWHC-OPO, the changes in haemodynamics from extubation to cardiac arrest are recorded. However, these data were not documented in this study period. In terms of the exact time between cardiac arrest and cold perfusion of the organs, the data were documented in 70 out of 78 DCD donors. This mean time was 7 min and the range was from 5 to 19 min. When the WIT was defined as the period between cardiac arrest and cold perfusion of the organs, 14 out of 70 DCDs met our proposed DCD cardiac donor criteria (Table 1). Three of these 14 DCDs met the criteria for the ‘ideal’ case for the initial clinical application (Table 2).
Recently, Boucek et al. reported the excellent short-term result of three infants undergoing successful orthotopic heart transplantation from DCD donors [3]. According to this report, the mean WIT was 18.3 min (11.5, 16.0 and 27.5 min, respectively) and the 6-month survival rate was 100% for the three DCD heart transplant recipients, compared to 84% survival for 17 control infants who received transplants procured through standard organ donation. These DCD heart transplant recipients have had functional and immunologic outcomes similar to those of controls. A few ethical and legal issues need to be discussed and solved for cardiac donation after cardiac death [24]. In terms of physiological features in immature hearts, a possible greater tolerance for global warm ischaemia of the immature myocardium is suggested as compared with the adult heart [25]. However, the successful experience in infant cases could encourage the initiation of heart transplantation from adult DCD donors.
In conclusion, we identified potential DCD donors for heart transplantation among the recent 3-year DCD donors for non-cardiac organ transplantation in our local OPO by using current brain-dead cardiac donor criteria and acceptable limits of the hypoxic period. The transplantation of such organs could significantly increase the number of cardiac donations. When the criteria were narrowed for the ideal case, only two donors met such criteria in the 3-year period. The clinical application of adult DCD heart transplantation could be challenging. Nevertheless, such ideal cases exist and could be acceptable as the first adult DCD cardiac donor.
