Potential Yield of Imminent Death Kidney Donation

Abstract

About 99,000 people are waiting for a kidney in the US, and many will die waiting. The concept of “imminent death” donation, a type of living donation, has been gaining attention among physicians, patients, and ethicists. We estimated the number of potential imminent death kidney donors at the University of Wisconsin Hospital and Clinics by estimating the number of annual deaths in individuals with normal kidney function. Based on a previous survey suggesting that 1/3 of patients might be willing to donate at imminent death, we estimate that between 76 and 396 people in the state of Wisconsin would be medically eligible and willing to donate each year at the time of imminent death. We extrapolated these numbers to all transplant centers in the US, estimating that between 5925 and 31,097 people might be eligible and willing to donate. Our results suggest that allowing donation at imminent death and including discussions about organ donation in end-of-life planning could substantially reduce the nation’s kidney waiting list while providing many more donors the opportunity to give this gift.

INTRODUCTION

Over 120,000 adults and children are waiting for an organ in the US, and another patient needing an organ is added to the transplant list about every 10 minutes. The organ in highest demand is the kidney, with approximately 99,000 people waiting, 30,000 new patients added to this list every year, and 4500 who die on this list every year. The incidence of end stage renal disease (ESRD) has been increasing and represents a significant public health and economic burden, as nearly 900,000 Americans are being treated for ESRD and over 20 million have some form of chronic kidney disease (CKD).(2) Treatment of ESRD requires either dialysis or kidney transplantation. Kidney transplantation is the most effective therapy for kidney failure,(3) and benefits over dialysis include significantly better quality of life, fewer medical complications, and longer survival.(3–5), Transplantation is more cost-effective than dialysis.(6, 7) Successful kidney transplantation to candidates awaiting transplants could save $10 billion per year (approximately $55,000 per year for the life of every functioning transplant).(8) Despite the benefits of transplantation over dialysis to treat ESRD, rates of living kidney donation have decreased in recent years.(9) To save lives while potentially decreasing healthcare expenditures, there is a need to increase the number of kidneys available for transplant.

Currently, organ transplant policy and laws in the US allow for organ procurement from three sources: neurologic deaths, controlled circulatory deaths, and live organ donors. Past efforts to increase donors have focused on education campaigns, typically encouraging people to sign up to be a donor when getting their driver license. However, these campaigns quickly lose impact over time(10) and have not focused on increasing living donation.

Some individuals with terminal illness wish to donate their kidneys shortly before they die – at the time of imminent death(11) – but there has not been a single reported case of kidney donation at the time of imminent death (i.e. before the imminently dying patient has been declared “dead”). The ethical and legal dimensions of imminent death donation (IDD) have been addressed elsewhere.(11–18) The purpose of this paper is to provide an estimate of the number of potential annual kidney donors in the US if IDD were to become available. Herein, we estimated the number of potential imminent death kidney donors at the University of Wisconsin Hospital and Clinics by assessing the number of annual deaths in individuals receiving end of life care with apparently healthy kidneys. We extrapolate these numbers to other hospitals with transplant centers in the US. Our results suggest that allowing donation at imminent death and including discussions about organ donation in end-of-life planning could substantially reduce the nation’s kidney waiting list.

METHODS

We carried out a retrospective study of medical records at the University of Wisconsin Hospital and Clinics. Potential subjects were identified using the institution’s electronic health information system, including all patients who died between January 2008 and December 2015 and were 18 years or older at the time of death, though this project was stimulated by a terminally ill 11-year-old boy whose parents asked if he could be a donor.

We performed two queries of our patient health records. The first query, using strict criteria, excluded patients if they met any of the following criteria: donated or received an organ, any previous or current diagnosis of cancer, most recent systolic blood pressure > 140 or < 60 mm Hg, most recent diastolic blood pressure >100 or < 60 mm Hg, use of pressors (epinephrine, norepinephrine, dopamine, vasopressin), HIV/AIDS, diabetes, hepatitis at time of death, pregnancy at time of death, on dialysis at time of death, age under 18 years or over 60 years. The rationale for each of these criteria is described in Table S1. The second query excluded patients < 18 years old and patients with metastatic cancer, as successful transplants have been reported from individuals with some of the exclusionary criteria used for the first query, including donors over age 60(19, 20), donors with bacteremia and other bloodstream infections (21–23), and HIV-positive to HIV-positive transplants. (24) We estimate that the actual number of potential imminent death kidney donors falls between these two estimates. For both queries, we counted the number of individuals in our health care system who died with normal kidney function, defined as eGFR > 90 at any one of 3 times: (1) most recent value; (2) at any point in their last year, and (3) within the last 7 days of life. We stratified patients by decade of life and whether or not they died as an inpatient at our hospital, at another hospital, or outside of a hospital. The number of patients was counted by age group (18-29, 30-39, 40-49, 50-59, 60-69, 70-79, ≥80), by year, and by death as an inpatient at our institution versus outside our institution (which could have been at an outside hospital or not at a hospital). Our institution uses the CKD-EPI equation to assess eGFR. The raw counts are shown in Tables S2–S4.

Each of these measures had a slightly different rationale. For criterion 1 (i.e. most recent eGFR was normal), we wanted to know if the patient’s last eGFR was normal because this would at least suggest that the patient died with kidneys that could have potentially been donated before the patient died. For criterion 2 (i.e. normal eGFR in the last 7 days of life), we wanted to know if there was any normal eGFR value in the last 7 days of life because the events precipitating the patient’s death may have affected the patient’s kidneys and the eGFR. However, a patient in this group might have been able to donate a kidney just before the dying process. For example, some patients may have had normal eGFR in the last week of life, but withdrawing supportive care may have compromised renal function; this patient’s kidney could potentially have been recoverable prior to withdrawing supportive care. This same rationale was used for criterion 3 (i.e. any normal eGFR at any point in the last year). Our primary analysis was conducted using the first eGFR criterion (i.e. most recent eGFR before death was normal), and additional analyses were conducted using the second and third criteria.

All calculations are based on the assumption that one kidney could be recovered from each imminently dying donor.

RESULTS

For our primary analysis, we counted the number of individuals who died within the UW health system (both at UW Hospital and outside our hospital) by year from 2008-2015. As described in the Methods section, we performed 2 queries: query 1 using strict donor criteria provides a lower estimate, and query 2 using less stringent criteria provides an upper estimate. Our lower estimate was 69 potential donors per year, while our upper estimate was 362 per year (Table 1). Based on previous survey data that 1/3 of patients might be willing to donate at imminent death, (12, 25) our lower estimate is 23 potential donors and upper estimate is 121. The estimate that 1/3 of patients would be willing to donate was used for all estimates in our analysis. If we extrapolate these numbers to all three transplant centers in Wisconsin, we estimate that there are between 76 and 396 eligible and willing donors every year. In Wisconsin, there are about 2000 candidates waiting for a kidney, 743 candidates get added every year and 608 candidates get removed every year (OPTN data, averaged from data from 2008-2015 in Wisconsin);(1) people get removed from the list if they receive an organ, become too sick to get an organ, or die. If we use the mean of our upper and lower estimates (236 per year) and assume that one kidney could be recovered from each imminently dying donor, it would take approximately 19.8 years to eliminate the kidney waiting list in Wisconsin (Figure 1A). However, query 1 (strict criteria) likely reflects the potential number of donors more accurately (76 potential eligible and willing donors). Nevertheless, this is potentially 76 more kidney donations that could be performed every year in the state of Wisconsin.

Table 1.

Summary of estimates of potential imminent death kidney donors per year.

	Wisconsin		Transplant Centers
	Query 1 (lower limit)	Query 2 (upper limit)	Query 1 (lower limit)	Query 2 (upper limit)
Criterion 1 (most recent eGFR >90)	76	396	5925	31097
Criterion 2 (eGFR >90 in last 7 days of life)	5	38	364	2972
Criterion 3 (eGFR >90 in last year of life)	49	365	3866	28618

^*Lower and upper estimates were determined from the two queries of the data described in Methods. Three different criteria were used for counting patients based on their eGFR. Data from the University of Wisconsin Hospital (total beds = 544) were extrapolated to Wisconsin’s three transplant centers (total beds = 1786) and all transplant centers (total beds = 140,076). Counts were divided by 3 to take into account our estimate that 1/3 of medically eligible donors would be willing to donate at imminent death.

Figure 1. Kidney supply and demand estimates.

Lines demonstrate the estimated number of kidney transplant candidates (Demand, solid line) based on OPTN data from 2008-2015 (using the average of the annual difference between wait list additions and removals and assuming that these trends continue) compared to estimated organs procured from imminent death donation (Supply Estimates, dotted lines). Estimates are made for the state of Wisconsin (A) and the US (B). Two supply curves are shown, demonstrating estimates from the 2 queries made of our data and based on criterion 1 (i.e. last eGFR of person’s life was >90) and the estimate that 1/3 of potential donors would be willing to donate. We assume that one kidney could be recovered from each imminently dying donor. To estimate demand, we subtracted the annual additions to the wait list from the annual removals from the wait list and averaged this difference for the time span from 2008-2015; this value became the slope (542 for Wisconsin, 5319 for US). The y-intercept was set to the current number of candidates on the waiting list (2000 for Wisconsin, 99,393 for US).

Next we extrapolated our data to estimate how many potential kidney donors there might be nationwide. According to UNOS and the National Kidney Center, there are 244 kidney transplant centers in the US. (26) We excluded children’s hospitals, leaving 218 centers with 140,076 beds (Table S5). We extrapolated the counts from our institution to the nation’s other transplant centers, as living organ donation is restricted to transplant centers by UNOS regulations. As above, we projected the number of kidneys that would become available if 1/3 of these patients became donors. We estimate that the number of potential interested donors at transplant centers in the US falls between 5925 and 31,097 each year (Table S5).

Next we assessed how this number of potential donors could affect the size of the nation’s waiting list. There are currently 99,393 candidates waiting for a kidney; 34,576 candidates are added to the waiting list every year, and 29,257 candidates get removed from the list every year (OPTN national data, averaged from 2008-2015) (1), Using the mean of our upper and lower estimates from our extrapolation to all transplant centers in the US (18,511 donors per year), we find that the national kidney transplant waiting list could be substantially reduced. (Figure 1B). As stated previously, query 1 (strict criteria) likely reflects the potential number of donors more accurately, meaning there are at least 5925 potential eligible and willing imminent death donors every year at US transplant centers.

Because a sizable portion of potential imminent death donors may be unable to give first person consent, the actual consent rate may be lower than our estimated 33%. Therefore, we assessed the impact of different rates of consent on the donor pool, and these data are summarized in Table S6. For example, if just 10% of medically eligible donors consented to donate at imminent death, this would yield between 25,748 and 257,704 donors annually in the US.

DISCUSSION

In this study, we sought to determine how many people are potentially being excluded from being living kidney donors at the time of imminent death. Our results suggest that offering patients, or their surrogates, the opportunity to donate kidneys at the end of their lives but prior to cardiac death has the potential to substantially reduce the US kidney waiting list. The limiting factors would then become issues other than the willingness of potential donors, including possible objections from, ethics committees, UNOS, legislatures, or courts. Past efforts to expand live donor kidney transplantation have included incompatible kidney transplantation, kidney-paired donation, non-directed donation, and the use of organs from live donors who may have previously been excluded because of hypertension or older age. Despite these efforts, 99,000 people remain on the waiting list for a kidney in the US, approximately 4500 of whom die every year on this list. There is a need to find better ways to address this critical and worsening shortage. Unlike other organs, such as lungs and hearts, we do not have a shortage of healthy kidneys that could potentially be transplanted.

Today, many terminally ill patients’ best, and often only, chance of donating organs is by donation after circulatory death (DCD), where life supporting treatment is withdrawn while an organ-recovery team stands by. In order for organs to be successfully transplanted in this manner, the donor needs to die within 1-2 hours of being taken off life support, or before impaired perfusion makes the organs unsuitable for transplantation; this unsuccessful donation occurs about 1/3 of the time.(28, 29) Even when DCD organ donors die in the allotted time, fewer organs may be recovered than from brain-dead donors, whose bodies are not subjected to prolonged ischemia.(11) Therefore, IDD has the medical advantage over DCD of obtaining better perfused organs and therefore has greater potential for improved graft success and recipient outcomes and survival. Furthermore, this also benefits organ recipients (more organs available), donor families (emotional fulfilment), and donor hospitals (less wasted time and resources compared to DCD).(14) A recent OPTN/UNOS Ethics Committee white paper on IDD acknowledged these potential benefits of IDD, but UNOS will not develop formal policy on IDD until the ethical, clinical, and practical concerns can be addressed.(30) One potential problem with imminent death donation is that if only one kidney were donated, as our estimates assumed, it could yield fewer organs than DCD (18). However, removal of both kidneys would not be likely to cause the donor’s death, as anephric patients typically do not die of uremic complications for days or sometimes weeks (17), and imminently dying patients are, by definition, expected to die of other causes soon after the donation. Additionally, it would be possible for donation of organs not required to sustain the imminently dying donor (e.g. one or even both kidneys) to be followed by DCD for the remaining organs. Further, the potential increase in overall IDD eligible donor pool should reduce the potential loss of organs from DCD.

If the patient provides first-person consent prior to donation, IDD respects the autonomous preferences of dying patients and provides psychosocial benefits for the donor’s family. However, our current system and policy do not always require first person consent for organ donation, or end of life decisions in general, as it allows authorized surrogates to make judgments about the patient’s interests and his/her likely preferences. Some have argued that it is ethically justified to perform non-harmful premortem interventions on dying people as long as they are consistent with the interests and likely preferences of the dying patient.(31) In a national survey of the American public, 71% of the sample agreed that it should be legal for an individual in a coma to donate organs even if procurement caused the donor’s death.(32) In addition, procuring organs from a terminally ill patient prior to removing life-sustaining therapy would provide the well known benefits for the donor’s family.

As outlined by Morrissey, (14) in the patient with terminal illness (e.g. stroke, hypoxic encephalopathy) who has become unconscious, the patient’s surrogate could be given the option of premortem organ donation. the patient would be taken to the operating room, donate one or both kidneys (neither of which would end the patient’s life), taken back to the family, end-of-life care instituted according to the family’s request (as they would have done regardless of transplant) and in a more relaxed time period. This allows the family to grieve at the decedent’s bedside in a more private setting.

The major principle of the dead donor rule is that a patient should not be killed to obtain his/her organs. (33) IDD would not be expected to kill the patient and clearly does not intend to hasten the patient’s death. The cause of death would be the same as if there were no organ retrieval: withdrawal of life sustaining medical treatment, as requested by the patient or his surrogate. It is possible that IDD might lead to a death sooner than would otherwise have occurred, but these are patients whose death, by definition, is anticipated and thought to be in the patient’s interest.

There are concerns that a transplant center performing a donation at the time of imminent death could be in jeopardy of being penalized because of a donor death soon after donation. Current UNOS policy requires reporting of all donor deaths, regardless of cause. The purpose is to identify deaths that were the result of donation. IDD, as proposed herein, would not be the cause of death. It would be important for UNOS policy to acknowledge this so transplant centers can be confident that they will not be punished.

If UNOS policy allowed patients to donate as living donors just before cardiovascular death, with appropriate consent and without negative consequences to the transplant centers, one could envision the following situation: a competent patient with a chronic disease that is expected to end his/her their life but have no effect on the health of the kidneys could donate a kidney when his/her physician(s) determine that death is imminent. Patients with cystic fibrosis and amyotrophic lateral sclerosis are able to contemplate and prepare for their premature death for many years, and consider the possibility of donation before death. They are able to provide primary consent – i.e., consent from the actual donor – rather the secondary consent from surrogates who may have imperfect information about the preferences of the dying patient. Another benefit to this proposal is that it allows these patients and their families to look forward to a meaningful, rewarding experience and legacy at the end of their lives, which many patients desire. (11) We have had four such requests at our hospital, 3 from adults, initiated by the patients, and one from parents of a child with non-metastatic brain cancer.

There are several limitations to this analysis. We may be underestimating the potential number of donors, as we excluded patients if there was no creatinine measurement available. Further, our extrapolation data may also underestimate the true number of potential donors, as we only extrapolated to hospitals with transplant centers for our primary analysis. Another cause of underestimation is our exclusion of patients receiving vasopressors from our lower estimate, as some of these patients on vasopressors would still be medically suitable donors. Many brain dead donors are hemodynamically unstable and on vasopressors, and many DCD donors are on vasopressors prior to withdrawal of life-sustaining treatment in the operating room. Another potential limitation is that we based our analysis on preliminary data that about 1/3 of potential donors would be willing to donate a kidney at imminent death; this data came from a small survey of patients with cystic fibrosis,(25) and it is unclear if this estimate is representative of the population as a whole. To address this limitation, we included data based on different estimates of the consent rate (Table 2). Additionally, there may be other barriers that would reduce the number of eligible IDD donors, including: (1) the patient dies too quickly for the team to coordinate donation; (2) the patient becomes unstable during organ procurement and the donation is aborted. Further, we acknowledge that these estimates are crude, and further analysis is needed; however, the purpose of this article is to demonstrate that the number of potentially eligible and willing imminent death donors is not trivial. Lastly, there are several reasons why our estimates may also overestimate the number of potential imminent death donors. First, for some cases of death in the hospital, the family and/or the patient are fighting to survive, and end of life planning happens too late or not at all. A discussion about living organ donation would often be unwarranted until an exclusion criterion like major organ dysfunction had already developed, thereby making them ineligible to donate. However, a majority of decisions to withdraw life-sustaining therapies take place hours or days before death. Second, when extrapolating, we assumed that each hospital bed is the same; however, there are significant differences in the acuity of different hospital beds, and our data do not take this into consideration.

Table 2.

Estimated imminent death donors based on different potential consent rates.

	Wisconsin		Transplant Centers
Percent consenting	Query 1 (lower limit)	Query 2 (upper limit)	Query 1 (lower limit)	Query 2 (upper limit)
10%	8	40	593	3110
20%	15	79	1185	6219
30%	23	119	1778	9329
40%	30	158	2370	12439
50%	38	198	2963	15549
60%	46	238	3555	18658
70%	53	277	4148	21768
80%	61	317	4740	24878
90%	68	356	5333	27987
100%	76	396	5925	31097

^*Our primary analyses were based on an empiric estimate that 1/3 would be willing to donate. This table demonstrates lower and upper estimates (based on the two queries) using criterion 1 (last eGFR measurement of patient’s life >90) based on different percentages of patients consenting (10-100% in 10% increments).

There remain about 99,000 people waiting for a kidney, many of whom will die without receiving a kidney. Previous efforts to increase living or post-mortem organ donation have not succeeded in closing the gap between supply and demand. Our data suggest a potential untapped source of kidneys that could save lives and provide a valuable experience and legacy for donors at the end of their lives. Heretofore, it has been unclear how many individuals would be medically eligible and willing to donate kidneys at the time of imminent death. Whatever the actual number may be, it is not zero.

Supplementary Material

Supp TableS1-5

Table S1: Rationale for criteria used to estimate potential imminent death donors.

Table S2: Estimates of potential imminent death kidney donors. Lower and upper estimates were made based on the 2 queries described in Methods. This table utilizes criterion 1, showing the number of deceased people whose last eGFR measurement was >90. These were patients of the University of Wisconsin Hospital and Clinics that died during the indicated year. Our primary analyses were performed using this criterion. We utilized the average number of deaths from 2008-2015.

Table S3: Lower and upper estimates were made based on the 2 queries described in Methods. This table utilizes criterion 2, showing the number of deceased people per year who had an eGFR measurement >90 in the last 7 days of life. These were patients of the University of Wisconsin Hospital and Clinics that died during the indicated year.

Table S4: Lower and upper estimates were made based on the 2 queries described in Methods. This table utilizes criterion 3, showing the number of deceased people per year who had an eGFR measurement >90 at any point during the last year of life. These were patients of the University of Wisconsin Hospital and Clinics that died during the indicated year.

Table S5: Extrapolation to the nation’s transplant centers. We divided estimates from our upper and lower estimates from the University of Wisconsin Hospital and Clinics by the number of hospital beds, then multiplied by the number of beds at all the nation’s transplant centers. As detailed in Methods, we utilized upper and lower estimates for 3 different criteria, as described in Methods: (1) most recent eGFR within >90; (2) most recent eGFR >90 and must have been within the last 7 days of life, and (3) eGFR >90 at any point within the last year of life.

Abbreviations

CKD

chronic kidney disease

DCD

donation after cardiac death

eGFR

estimated glomerular filtration rate

ESRD

end stage renal disease

IDD

imminent death donation

OPTN

Organ Procurement and Transplantation Network

UNOS

United Network for Organ Sharing