American Journal of Transplantation 2014; 14: 21–26 Wiley Periodicals Inc.
C
Copyright 2013 The American Society of Transplantation and the American Society of Transplant Surgeons doi: 10.1111/ajt.12557
Personal Viewpoint
Revisiting Multi-Organ Transplantation in the Setting of Scarcity P. P. Reese1,2,3,*, R. M. Veatch4, P. L. Abt5 and S. Amaral2,6 1
Renal-Electrolyte and Hypertension Division, Perelman School of Medicine at the University of Pennsylvania, Philadelphia, PA 2 Department of Biostatistics and Epidemiology, University of Pennsylvania, Philadelphia, PA 3 Leonard Davis Institute of Health Economics, University of Pennsylvania, Philadelphia, PA 4 Kennedy Institute of Ethics, Georgetown University, Washington, DC 5 Department of Surgery, Penn Transplant Institute, Hospital of the University of Pennsylvania, Philadelphia, PA 6 Division of Nephrology, The Children’s Hospital of Philadelphia, Philadelphia, PA Corresponding author: Peter P. Reese, [email protected]
In the setting of organ scarcity, the ethics of multiorgan transplantation (MOT) deserve new examination. MOT offers substantial benefits to certain recipients, including avoiding serial surgeries. However, MOT candidates in the United States commonly receive priority for their nonprimary organ over many individuals who need that organ, which may undermine equity. The absence of standard criteria for MOT eligibility also enables large and unfair regional variation in MOT, such as simultaneous liver–kidney transplantation. Unfortunately, MOT may also undermine utility (optimal patient and graft survival) in circumstances where providing multiple organs to one person fails to achieve the greater collective benefit attained by providing transplants to multiple people. Policy reforms should include the adoption of minimal clinical criteria for MOT candidacy with the attendant goal of decreasing regional variation in MOT. In the future, these minimal criteria can be revised to accommodate new research about which patients derive the most benefit from MOT. Incentives to perform MOT should also be reduced, such as by including MOT outcomes in center-specific reports. These reforms run the risk that the transplant community could be perceived as abandoning MOT candidates, but offer an opportunity to align transplant practice and ethical principles. Keywords: Allocation, ethics, multi-organ transplantation
Abbreviations: eGFR, estimated glomerular filtration rate; ESRD, end-stage renal disease; HRSA, Health Resources and Services Administration; MELD, Model for End-Stage Liver Disease score; MOT, multi-organ transplantation; OPTN, Organ Procurement and Transplantation Network; SLK, simultaneous liver–kidney transplantation; SPK, simultaneous pancreas–kidney transplantation; STA, simultaneous thoracic and abdominal transplantation; US, United States Received 07 June 2013, revised and accepted for publication 23 October 2013
Background Over the last 50 years, clinical innovations have brought transplantation to diverse patients who were historically considered too high risk—including multi-organ transplantation (MOT) recipients (who receive two or more organs during the same surgery). Unfortunately, innovations in expanding the organ pool have not been equally successful, leading to ever-lengthening waiting lists and waiting times. In the setting of organ scarcity, MOT is an arena that presents opportunities for reframing key ethical concepts around allocation reform. The practice of MOT has become common, but striking heterogeneity in patient selection and listing practices across the United States reveals problems of equity and utility. In simultaneous liver–kidney (SLK) or simultaneous pancreas–kidney (SPK) transplantation, when the MOT candidate receives priority for the kidney over kidney-alone candidates, MOT may undermine equity. While MOT provides substantial clinical advantages to many recipients, the advantages provided by the additional organs must be weighed against the claims and likely outcomes of other candidates for those organs. Reforms should include a reduction in incentives to perform MOT, innovative research to better establish minimal clinical criteria for MOT eligibility and a transparent public process for determining allocation priority for MOT. Reforms to organ allocation policy and practice should be responsive both to general ethical principles as well as the specific clinical circumstances for different types of MOT. 21
Reese et al
Variation in MOT Allocation and Ethical Implications The most common MOT is SPK (>19 000 performed in the United States from 1988 to 2013), followed by SLK (>5000 performed). Most pancreas allografts are allocated through the kidney–pancreas waiting list, determined primarily by waiting time, with extra priority for sensitization (1). When one of the required organs is a liver, heart or lung, the MOT candidate is registered on the individual waiting list for the primary organ. If the candidate is allocated the primary organ, the additional required organ from the same donor is sequestered by that candidate. In these diverse clinical scenarios, MOT candidates commonly receive higher priority for the nonprimary organ (usually a kidney) than all other patients waiting for a single transplant with the same organ. These allocation systems raise serious equity concerns, particularly when those patients ‘‘passed over’’ for the nonprimary organ have strong ethical claims such as pediatric status or long waiting time. Pancreas transplantation improves quality of life and reduces the risk of complications for candidates with Type 1 diabetes and severe hypoglycemic episodes (2,3). In SPK transplantation, the pancreas improves kidney allograft survival (vs. kidney transplant alone), and possibly improves patient survival (4,5). SPK candidates receive higher priority for the transplant than kidney-alone candidates with similar or greater waiting time. This allocation priority is available even to some SPK candidates with Type 2 diabetes, for whom the benefits of pancreas transplant are uncertain (6). Liver and heart allocation are based on the ‘‘sickest first’’ principle. Liver transplant candidates are classified as status 1A or 1B based on medical urgency, then ranked according to their Model for End-Stage Liver Disease (MELD) score, which reflects mortality risk on the waiting list (7). For heart transplantation, candidates are classified as status 1A or 1B based on ventilator needs, circulatory status, inotropic support and life expectancy without transplant. For both liver and heart allocation, waiting time is considered within categories of similar medical urgency. Lung transplant allocation strives to maximize overall utility by integrating estimates of pre- and posttransplant survival. Lung candidates receive a Lung Allocation Score; higher scores indicate greater expected survival benefit from transplantation and lead to higher priority (7). In contrast, kidney allocation has historically been based primarily on waiting times, while also giving priority to zeroantigen mismatches and subgroups such as sensitized patients. Notably, unlike patients with liver or lung failure, patients with end-stage renal disease (ESRD) can often survive for prolonged periods without transplantation because of chronic dialysis. Consequently, kidney-alone candidates may be perceived as having less medical urgency than subjects with primary heart, liver or lung failure and some degree of renal insufficiency. In MOT allocation, the kidney is 22
Figure 1: Distribution of kidney donor profile index (KDPI) scores among recipients of kidney-alone and multi-organ transplant recipients. Figure generated using Organ Procurement and Transplantation Network data and represents a cohort of adult (18 years) solid organ transplant recipients who received a deceased donor kidney transplant from February 2002 to April 2013. p < 0.001 for each comparison of kidney-alone recipients to each other multi-organ transplant group.
always relegated to nonprimary organ status. In the year 2011, 548 kidneys (5% of all transplanted deceased donor kidneys) were allocated to MOT recipients. As shown in Figure 1, these kidneys are among the highest quality organs procured. Meanwhile, the percentage of adult patients receiving a deceased donor kidney transplant within 3 years has been declining since 1991. Among kidney-alone candidates, the percentage of individuals who will receive a kidney transplant by 3 years varies from 17% to 18% for those with blood type O or B to 45% for those with blood type AB (8). Heart–lung combinations and intestine–liver combinations fall under different allocation provisions. For heart–lung, when a candidate is eligible for a heart, the lung from the same donor is allocated to that candidate before lung-only candidates. If the candidate is in greater need of the lung (vs. heart), then the heart is allocated to the heart–lung candidate from the same donor only if no isolated heart candidate with medical urgency (1A status) is eligible to receive the heart. In intestine–liver allocation, the liver is first offered via the liver match run and then may be offered to the combined liver– intestine candidates in the intestine match run (7). Thus, MOT allocation varies substantially by organ combination. Each allocation strategy makes different accommodations to equity and utility. This heterogeneity shows that an ethics analysis of MOT practices must address the existing allocation systems and the clinical nuances of specific organ combinations.
Problems of Equity With Organ Allocation to Multi-Organ Transplant Recipients Although multiple definitions of fairness—or equity—have been proposed, organ allocation via MOT poses problems American Journal of Transplantation 2014; 14: 21–26
Ethics of Multi-Organ Transplantation
within all of them. Drawing on concepts articulated by the philosopher John Rawls, two main principles may be applied to promoting distributive justice for transplant candidates (9). First, the Equality Principle states that individuals who can derive similar benefit from an organ ought to have equivalent access to it. Allocation based on a lottery, or less optimally, a first-come and first-served system, might satisfy this principle. Instead, current procedures allow MOT recipients to ‘‘jump ahead in line’’ and receive higher priority for their additional organs than others who may receive similar or greater benefit. Rawls’ Difference Principle allows for inequalities in allocation only if the imbalances would benefit the least advantaged, or ‘‘worst off’’ (9). The challenge for an equity analysis is that many factors contribute to making one ‘‘poorly off’’—for example, mortality risk, morbidities, the length of time waiting for organs or sensitization. The Fair Innings approach suggests that the younger a transplant candidate is, the worse off that candidate is (having enjoyed fewer years of healthy life). Thus, it is unclear that MOT candidates are ‘‘worse off’’ in terms of need for their additional organs compared to all other candidates for those organs. For example, although adult MOT recipients are among the worst off as measured by overall illness acuity, children with end-stage organ disease who may never reach adulthood may be even more ‘‘worse off.’’ As a second example, highly sensitized kidney transplant candidates have extremely reduced access to transplant because of incompatibility with most donors. As a third example, patients with prolonged waiting times for a single kidney face elevated risks of mortality and reduced quality of life. Yet candidates in these three examples often have lower priority for that kidney than an SLK or SPK candidate. Another inequitable element of current allocation is the absence of standard criteria for MOT eligibility. Using the SLK example, no minimal criteria for the severity of kidney dysfunction needs to be satisfied to qualify. As a result, transplant centers differ greatly in how they determine SLK listing eligibility (10) and the percentage of liver transplant candidates who receive SLKs varies from 4% to 12% across US regions (Figure 2). Variation in SLK practices may lead to heterogeneous access to kidneys, such that patients listed in regions with high rates of MOT have greater diversions from their organ supply.
Problems of Utility With Organ Allocation to Multi-Organ Transplant Recipients This variation in MOT practice and the absence of minimal listing criteria create the possibility of inefficient organ allocation (as measured by survival benefit derived from the organs), while making it difficult for practitioners to interpret outcome data and construct valid risk-benefit assessments American Journal of Transplantation 2014; 14: 21–26
Figure 2: Geographic variation in simultaneous liver–kidney (SLK) transplant as a percentage of all liver transplants. Donor service areas are geographic areas of the United States served by local organizations that procure organs donated for transplantation.
of MOT for most patients. The lack of minimal listing criteria for MOT enables physicians to list some patients for MOT whose outcome may only be improved incrementally by the nonprimary organ—and diverts that nonprimary organ from other candidates who may die waiting (11–13). For example, it is common for MOT candidates to receive a kidney transplant even before receiving chronic dialysis, because of the concern that complications of transplant surgery may cause the patient’s chronic kidney disease to proceed to ESRD. However, unlike candidates for a kidneyalone transplant who must demonstrate an estimated glomerular filtration rate (eGFR) 20 mL/min/1.73 m2, a MOT candidate does not need to demonstrate any level of eGFR to obtain a kidney. Unfortunately, the lack of high-quality data about the added value of MOT versus single-organ transplant has undermined consensus about what should constitute minimal clinical criteria. No randomized controlled trials of singleorgan transplant versus MOT have been conducted. Most studies have been retrospective and commonly based on data from the Organ Procurement and Transplantation Network (OPTN), with the prominent limitation of unmeasured confounding, because MOT candidates differ in meaningful ways from single-organ recipients (13–16). The other valuable data source has been single-center studies with the problem of limited generalizability (17,18). In the case of SLK, for example, an analysis of OPTN data by Locke et al (11) suggested that due to an elevated risk of ESRD after liver transplant alone, SLK should be considered for liver transplant candidates with >12 weeks of dialysis. An analysis by Ruebner et al (12) of nondialyzed liver transplant candidates suggested that diabetic patients with a median eGFR
C
Copyright 2013 The American Society of Transplantation and the American Society of Transplant Surgeons doi: 10.1111/ajt.12557
Personal Viewpoint
Revisiting Multi-Organ Transplantation in the Setting of Scarcity P. P. Reese1,2,3,*, R. M. Veatch4, P. L. Abt5 and S. Amaral2,6 1
Renal-Electrolyte and Hypertension Division, Perelman School of Medicine at the University of Pennsylvania, Philadelphia, PA 2 Department of Biostatistics and Epidemiology, University of Pennsylvania, Philadelphia, PA 3 Leonard Davis Institute of Health Economics, University of Pennsylvania, Philadelphia, PA 4 Kennedy Institute of Ethics, Georgetown University, Washington, DC 5 Department of Surgery, Penn Transplant Institute, Hospital of the University of Pennsylvania, Philadelphia, PA 6 Division of Nephrology, The Children’s Hospital of Philadelphia, Philadelphia, PA Corresponding author: Peter P. Reese, [email protected]
In the setting of organ scarcity, the ethics of multiorgan transplantation (MOT) deserve new examination. MOT offers substantial benefits to certain recipients, including avoiding serial surgeries. However, MOT candidates in the United States commonly receive priority for their nonprimary organ over many individuals who need that organ, which may undermine equity. The absence of standard criteria for MOT eligibility also enables large and unfair regional variation in MOT, such as simultaneous liver–kidney transplantation. Unfortunately, MOT may also undermine utility (optimal patient and graft survival) in circumstances where providing multiple organs to one person fails to achieve the greater collective benefit attained by providing transplants to multiple people. Policy reforms should include the adoption of minimal clinical criteria for MOT candidacy with the attendant goal of decreasing regional variation in MOT. In the future, these minimal criteria can be revised to accommodate new research about which patients derive the most benefit from MOT. Incentives to perform MOT should also be reduced, such as by including MOT outcomes in center-specific reports. These reforms run the risk that the transplant community could be perceived as abandoning MOT candidates, but offer an opportunity to align transplant practice and ethical principles. Keywords: Allocation, ethics, multi-organ transplantation
Abbreviations: eGFR, estimated glomerular filtration rate; ESRD, end-stage renal disease; HRSA, Health Resources and Services Administration; MELD, Model for End-Stage Liver Disease score; MOT, multi-organ transplantation; OPTN, Organ Procurement and Transplantation Network; SLK, simultaneous liver–kidney transplantation; SPK, simultaneous pancreas–kidney transplantation; STA, simultaneous thoracic and abdominal transplantation; US, United States Received 07 June 2013, revised and accepted for publication 23 October 2013
Background Over the last 50 years, clinical innovations have brought transplantation to diverse patients who were historically considered too high risk—including multi-organ transplantation (MOT) recipients (who receive two or more organs during the same surgery). Unfortunately, innovations in expanding the organ pool have not been equally successful, leading to ever-lengthening waiting lists and waiting times. In the setting of organ scarcity, MOT is an arena that presents opportunities for reframing key ethical concepts around allocation reform. The practice of MOT has become common, but striking heterogeneity in patient selection and listing practices across the United States reveals problems of equity and utility. In simultaneous liver–kidney (SLK) or simultaneous pancreas–kidney (SPK) transplantation, when the MOT candidate receives priority for the kidney over kidney-alone candidates, MOT may undermine equity. While MOT provides substantial clinical advantages to many recipients, the advantages provided by the additional organs must be weighed against the claims and likely outcomes of other candidates for those organs. Reforms should include a reduction in incentives to perform MOT, innovative research to better establish minimal clinical criteria for MOT eligibility and a transparent public process for determining allocation priority for MOT. Reforms to organ allocation policy and practice should be responsive both to general ethical principles as well as the specific clinical circumstances for different types of MOT. 21
Reese et al
Variation in MOT Allocation and Ethical Implications The most common MOT is SPK (>19 000 performed in the United States from 1988 to 2013), followed by SLK (>5000 performed). Most pancreas allografts are allocated through the kidney–pancreas waiting list, determined primarily by waiting time, with extra priority for sensitization (1). When one of the required organs is a liver, heart or lung, the MOT candidate is registered on the individual waiting list for the primary organ. If the candidate is allocated the primary organ, the additional required organ from the same donor is sequestered by that candidate. In these diverse clinical scenarios, MOT candidates commonly receive higher priority for the nonprimary organ (usually a kidney) than all other patients waiting for a single transplant with the same organ. These allocation systems raise serious equity concerns, particularly when those patients ‘‘passed over’’ for the nonprimary organ have strong ethical claims such as pediatric status or long waiting time. Pancreas transplantation improves quality of life and reduces the risk of complications for candidates with Type 1 diabetes and severe hypoglycemic episodes (2,3). In SPK transplantation, the pancreas improves kidney allograft survival (vs. kidney transplant alone), and possibly improves patient survival (4,5). SPK candidates receive higher priority for the transplant than kidney-alone candidates with similar or greater waiting time. This allocation priority is available even to some SPK candidates with Type 2 diabetes, for whom the benefits of pancreas transplant are uncertain (6). Liver and heart allocation are based on the ‘‘sickest first’’ principle. Liver transplant candidates are classified as status 1A or 1B based on medical urgency, then ranked according to their Model for End-Stage Liver Disease (MELD) score, which reflects mortality risk on the waiting list (7). For heart transplantation, candidates are classified as status 1A or 1B based on ventilator needs, circulatory status, inotropic support and life expectancy without transplant. For both liver and heart allocation, waiting time is considered within categories of similar medical urgency. Lung transplant allocation strives to maximize overall utility by integrating estimates of pre- and posttransplant survival. Lung candidates receive a Lung Allocation Score; higher scores indicate greater expected survival benefit from transplantation and lead to higher priority (7). In contrast, kidney allocation has historically been based primarily on waiting times, while also giving priority to zeroantigen mismatches and subgroups such as sensitized patients. Notably, unlike patients with liver or lung failure, patients with end-stage renal disease (ESRD) can often survive for prolonged periods without transplantation because of chronic dialysis. Consequently, kidney-alone candidates may be perceived as having less medical urgency than subjects with primary heart, liver or lung failure and some degree of renal insufficiency. In MOT allocation, the kidney is 22
Figure 1: Distribution of kidney donor profile index (KDPI) scores among recipients of kidney-alone and multi-organ transplant recipients. Figure generated using Organ Procurement and Transplantation Network data and represents a cohort of adult (18 years) solid organ transplant recipients who received a deceased donor kidney transplant from February 2002 to April 2013. p < 0.001 for each comparison of kidney-alone recipients to each other multi-organ transplant group.
always relegated to nonprimary organ status. In the year 2011, 548 kidneys (5% of all transplanted deceased donor kidneys) were allocated to MOT recipients. As shown in Figure 1, these kidneys are among the highest quality organs procured. Meanwhile, the percentage of adult patients receiving a deceased donor kidney transplant within 3 years has been declining since 1991. Among kidney-alone candidates, the percentage of individuals who will receive a kidney transplant by 3 years varies from 17% to 18% for those with blood type O or B to 45% for those with blood type AB (8). Heart–lung combinations and intestine–liver combinations fall under different allocation provisions. For heart–lung, when a candidate is eligible for a heart, the lung from the same donor is allocated to that candidate before lung-only candidates. If the candidate is in greater need of the lung (vs. heart), then the heart is allocated to the heart–lung candidate from the same donor only if no isolated heart candidate with medical urgency (1A status) is eligible to receive the heart. In intestine–liver allocation, the liver is first offered via the liver match run and then may be offered to the combined liver– intestine candidates in the intestine match run (7). Thus, MOT allocation varies substantially by organ combination. Each allocation strategy makes different accommodations to equity and utility. This heterogeneity shows that an ethics analysis of MOT practices must address the existing allocation systems and the clinical nuances of specific organ combinations.
Problems of Equity With Organ Allocation to Multi-Organ Transplant Recipients Although multiple definitions of fairness—or equity—have been proposed, organ allocation via MOT poses problems American Journal of Transplantation 2014; 14: 21–26
Ethics of Multi-Organ Transplantation
within all of them. Drawing on concepts articulated by the philosopher John Rawls, two main principles may be applied to promoting distributive justice for transplant candidates (9). First, the Equality Principle states that individuals who can derive similar benefit from an organ ought to have equivalent access to it. Allocation based on a lottery, or less optimally, a first-come and first-served system, might satisfy this principle. Instead, current procedures allow MOT recipients to ‘‘jump ahead in line’’ and receive higher priority for their additional organs than others who may receive similar or greater benefit. Rawls’ Difference Principle allows for inequalities in allocation only if the imbalances would benefit the least advantaged, or ‘‘worst off’’ (9). The challenge for an equity analysis is that many factors contribute to making one ‘‘poorly off’’—for example, mortality risk, morbidities, the length of time waiting for organs or sensitization. The Fair Innings approach suggests that the younger a transplant candidate is, the worse off that candidate is (having enjoyed fewer years of healthy life). Thus, it is unclear that MOT candidates are ‘‘worse off’’ in terms of need for their additional organs compared to all other candidates for those organs. For example, although adult MOT recipients are among the worst off as measured by overall illness acuity, children with end-stage organ disease who may never reach adulthood may be even more ‘‘worse off.’’ As a second example, highly sensitized kidney transplant candidates have extremely reduced access to transplant because of incompatibility with most donors. As a third example, patients with prolonged waiting times for a single kidney face elevated risks of mortality and reduced quality of life. Yet candidates in these three examples often have lower priority for that kidney than an SLK or SPK candidate. Another inequitable element of current allocation is the absence of standard criteria for MOT eligibility. Using the SLK example, no minimal criteria for the severity of kidney dysfunction needs to be satisfied to qualify. As a result, transplant centers differ greatly in how they determine SLK listing eligibility (10) and the percentage of liver transplant candidates who receive SLKs varies from 4% to 12% across US regions (Figure 2). Variation in SLK practices may lead to heterogeneous access to kidneys, such that patients listed in regions with high rates of MOT have greater diversions from their organ supply.
Problems of Utility With Organ Allocation to Multi-Organ Transplant Recipients This variation in MOT practice and the absence of minimal listing criteria create the possibility of inefficient organ allocation (as measured by survival benefit derived from the organs), while making it difficult for practitioners to interpret outcome data and construct valid risk-benefit assessments American Journal of Transplantation 2014; 14: 21–26
Figure 2: Geographic variation in simultaneous liver–kidney (SLK) transplant as a percentage of all liver transplants. Donor service areas are geographic areas of the United States served by local organizations that procure organs donated for transplantation.
of MOT for most patients. The lack of minimal listing criteria for MOT enables physicians to list some patients for MOT whose outcome may only be improved incrementally by the nonprimary organ—and diverts that nonprimary organ from other candidates who may die waiting (11–13). For example, it is common for MOT candidates to receive a kidney transplant even before receiving chronic dialysis, because of the concern that complications of transplant surgery may cause the patient’s chronic kidney disease to proceed to ESRD. However, unlike candidates for a kidneyalone transplant who must demonstrate an estimated glomerular filtration rate (eGFR) 20 mL/min/1.73 m2, a MOT candidate does not need to demonstrate any level of eGFR to obtain a kidney. Unfortunately, the lack of high-quality data about the added value of MOT versus single-organ transplant has undermined consensus about what should constitute minimal clinical criteria. No randomized controlled trials of singleorgan transplant versus MOT have been conducted. Most studies have been retrospective and commonly based on data from the Organ Procurement and Transplantation Network (OPTN), with the prominent limitation of unmeasured confounding, because MOT candidates differ in meaningful ways from single-organ recipients (13–16). The other valuable data source has been single-center studies with the problem of limited generalizability (17,18). In the case of SLK, for example, an analysis of OPTN data by Locke et al (11) suggested that due to an elevated risk of ESRD after liver transplant alone, SLK should be considered for liver transplant candidates with >12 weeks of dialysis. An analysis by Ruebner et al (12) of nondialyzed liver transplant candidates suggested that diabetic patients with a median eGFR
