The Future Direction of the Adult Heart Allocation System in the United States

Transcription

1 American Journal of Transplantation 2015; 15: Wiley Periodicals Inc. Special Article C Copyright 2014 The American Society of Transplantation and the American Society of Transplant Surgeons doi: /ajt The Future Direction of the Adult Heart Allocation System in the United States D. M. Meyer 1, *, J. G. Rogers 2, L. B. Edwards 3, E. R. Callahan 3, S. A. Webber 4, M. R. Johnson 5, J. D. Vega 6, M. J. Zucker 7 and J. C. Cleveland Jr 8 1 Department of Cardiovascular and Thoracic Surgery, University of Texas Southwestern Medical Center, Dallas, TX 2 Department of Medicine, Duke University, Durham, NC 3 United Network for Organ Sharing, Richmond, VA 4 Department of Pediatrics, Vanderbilt University School of Medicine, Nashville, TN 5 Department of Medicine, University of Wisconsin School of Medicine, Madison, WI 6 Department of Surgery, Emory University School of Medicine, Atlanta, GA 7 Department of Medicine, Newark Beth Israel Medical Center, Newark, NJ 8 Department of Surgery, University of Colorado School of Medicine, Denver, CO Corresponding author: Dan M. Meyer, Danm.meyer@UTSouthwestern.edu Ensuring equitable and fair organ allocation is a central charge of the United Network for Organ Sharing (UNOS) as the Organ Procurement and Transplantation Network (OPTN) through its contract with the Department of Health and Human Services (DHHS). The OPTN/UNOS Board initiated a reassessment of the current allocation system. This paper describes the efforts of the OPTN/UNOS Heart Subcommittee, acting on behalf of the OPTN/UNOS Thoracic Organ Transplantation Committee, to modify the current allocation system. The Subcommittee assessed the limitations of the current three-tiered system, outcomes of patients with status exceptions, emerging ventricular assist device (VAD) population, options for improved geographic sharing and status of potentially disenfranchised groups. They analyzed waiting list and posttransplant mortality rates of a contemporary cohort of patient groups at risk, in collaboration with the Scientific Registry of Transplant Recipients to develop a proposed multi-tiered allocation scheme. This proposal provides a framework for simulation modeling to project whether candidates would have better waitlist survival in the revised allocation system, and whether posttransplant survival would remain stable. The tiers are subject to change, based on further analysis by the Heart Subcommittee and will lead to the development of a more effective and equitable heart allocation system. Abbreviations: CHD, congenital heart disease; DHHS, Department of Health and Human Services; HAS, heart allocation score; LVAD, left ventricular assist device; OPTN, Organ Procurement and Transplantation Network; RRB, regional review board; SRTR, Scientific Registry of Transplant Recipients; TAH, total artificial heart; TSAM, thoracic simulation allocation modeling; UNOS, United Network for Organ Sharing; VAD, ventricular assist device; VF, ventricular fibrillation; VT, ventricular tachycardia Received 06 May 2014, revised publication 22 July 2014 Introduction and accepted for The United Network for Organ Sharing (UNOS) serves as the Organ Procurement and Transplantation Network (OPTN) through its contract with the Department of Health and Human Services (DHHS) and is charged with ensuring fair and equitable allocation of organs in the United States. Recent discussions involving the OPTN/UNOS Board have led to a reassessment of the current adult heart allocation scheme. The impetus to review the heart allocation system stems from a combination of factors, including: (1) the increase in candidates awaiting transplantation without a corresponding increase of available donors; (2) higher than desirable waiting list mortality rates in the highest acuity (Status 1A) patients; and (3) the changing landscape of the waitlisted patients due to increased utilization of ventricular assist devices (VADs). The goals of the revision are increasing transplantation rates for adult candidates with the highest waiting list mortality rates and achieving the greatest survival benefit for heart transplant recipients. The Heart Subcommittee, acting on behalf of the Thoracic Committee, has initiated efforts to modify the current system. These efforts also involved input from the two federal contractors, the OPTN and the SRTR. The status of these proceedings and the future plans for a modified heart allocation system will be reviewed. History of the Heart Allocation System and Policy Evolution In 1984, the US Congress passed the National Organ Transplant Act, creating the OPTN and organ procurement organizations (OPOs) (1). DHHS issued the Final Rule in 44

2 Heart Allocation in the United States Table 1: Heart allocation scheme for adult candidates from 1999 to present Adult heart allocation Adult heart allocation 2005 to present Local Status 1A Local Status 1A Zone A Status 1A Zone A Status 1A Zone B Status 1A Local Zone A Zone B Status 1A Zone B Zone A Zone C Status 1A Zone B Zone D Status 1A Zone C Status 1A Zone D Status 1A Zone E 1 Status 1A 1 The Board approved the addition of Zone E in June , which is still the primary regulation governing the OPTN/UNOS in developing and implementing organ allocation policies (2). DHHS stressed the importance of access to available organs to ensure equitable regional distribution, and utility to consider transplant candidates severity of illness. In 1988, the OPTN/UNOS introduced a two-tiered heart allocation system based upon urgency, waiting time, geography and blood type. The urgency-based system and the concentric geographic zones remain in place today, with only two major policy changes since its inception. In June 1998, a three-tiered system was created, and in November 2005, broader sharing of donor hearts was initiated (Table 1). This change led to a substantial decrease in the rate of waiting list mortality for Status 1A and Status 1B candidates (Table 2). In June 2012, the disparity in mortality rates for Status 1A and 1B candidates, in conjunction with the almost tripling of the number of Status 1A and 1B listings since the 2005 policy change, led the Thoracic Committee to re-assess the current system with respect to allocation of organs to the most critically ill candidates, and to ensure fairness in the geographic distribution of organs. From this date, it was the charge of the Heart Subcommittee to focus on the heart allocation system. The subcommittee, which consists of members of the Thoracic Committee, includes regional and at-large representatives from heart failure cardiology, surgery, OPO administrators, patient advocacy representative and transplant coordinators. Meetings were held via webbased teleconference on a monthly basis, and in person biannually. Areas requiring further specialized discussion received input from other UNOS committees such as the OPO and histocompatibility committees. Rationale for Change/Goals of the Modified System In response to the UNOS Board s charge to the Thoracic Committee, the Heart Subcommittee first assessed whether the current allocation system was in need of change (3). Despite the positive impact prior changes to the allocation system over the last decade have had on the waiting list mortality rates without negatively affecting posttransplant survival rates (Figure 1), the Heart Subcommittee determined waiting list mortality remained high for certain groups and recognized the need to provide more access to transplantation to the most urgent candidates without compromising their posttransplant survival. Table 2: Waiting list mortality/removal for too ill in adult heart candidates Waiting era: 7/12/03 7/11/06 (prepolicy) Waiting era: 7/12/06 7/11/09 (postpolicy era 1) Waiting era: 7/12/09 1/11/13 (postpolicy era 2) Status # patients ever waiting Death þ too ill removals/100 PY # patients ever waiting Death þ too ill removals/100 PY # patients ever waiting Death þ too ill removals/100 PY 1A B All statuses combined Methods: The mortality/too ill rates were computed as the ratio of the total number of deaths while listed or within 7 days of waiting list removal and waiting list removals for too ill to transplant in comparison to the number of patient-years waiting. As patients frequently are inactivated shortly before removal from the waiting list for death, time spent at inactive status and the event (death or too ill removal) were attributed to the preceding active status. Deaths reported to the OPTN were supplemented with the Social Security Death Master file. Results: For Status 1A candidates, the mortality/too ill rate declined from deaths/too ill removals per 100 patient-years (PY) to 54.6 deaths/too ill removals per 100 PY. However, Status 1A candidates still have a threefold higher rate compared to candidates (55 vs. 17 deaths/too ill removals per 100 PY in the most recent era). American Journal of Transplantation 2015; 15:

3 Meyer et al Survival (%) Pre-Policy: 7/12/04-5/11/06 (N=1532) Post-Policy Era 2: 7/12/08-5/11/11 (N=5013) Pre-Policy vs. Post-Policy Era 1: p = 0.99 Pre-Policy vs. Post-Policy Era 2: p = Time(days) Post-Policy Era 1: 7/12/06-5/11/08 (N=3029) Figure 1: Posttransplant survival for adult recipients: all statuses combined. Moreover, there is a public perception of inadequate sharing of organs across the country (4). With input from other interested parties throughout the country, including a November 2013 Forum of US Heart Allocation Policy of which members of the Thoracic Committee served as conference Co-chairs and Advisors, the Subcommittee identified particular issues with the current adult heart allocation policy that deserved careful assessment: The limitations of the three-tiered urgency-based system. The need for status exceptions. The status of VAD patients awaiting heart transplantation. Options for improved zonal or geographic sharing. Defining and prioritizing highly sensitized heart candidates. Limitations of the three-tiered urgency-based system The current urgency-based system does not distinguish between Status 1A and 1B candidates waiting within their respective statuses. Within a status in a geographic zone, hearts are allocated to candidates in order of decreasing time spent at that status, in addition to blood typematching. Four options were considered for improving the current system: (1) leaving the current three-tiered system in place with no further changes; (2) altering the three-tiered system by changing the qualifying criteria for each status; (3) expanding the number of tiers; and (4) developing a formal heart allocation score (HAS). With input from the OPTN and the SRTR that addressed issues of data availability to guide candidate risk stratification, cost and timing of system implementation, the Subcommittee elected to pursue expansion of the number of tiers in an attempt to develop a more granular allocation system that enhanced risk stratification. While there was some support for the HAS, subcommittee members believed the HAS would take too long to develop and implement, and would be too inflexible to modify during a time of rapidly changing heart support technology. Moreover, the data currently collected by the OPTN limited the ability of the Subcommittee to develop a data-driven HAS at this time. Incomplete information may lead to inaccurate estimates of impact on waiting list mortality, creating an HAS that performs differently than intended. Though reducing waiting list mortality rates is the primary goal of the revised allocation system, the Subcommittee also considered posttransplant survival. Assessing transplant benefit helps avoid organ wastage, which could occur in a purely urgency-based allocation system. Efforts to reduce mortality by combining posttransplant survival in the allocation scheme has been recently reported by Smits et al from Eurotransplant (5). The Subcommittee reviewed waiting list mortality and posttransplant survival for transplant candidates and recipients; stratified by status, criteria and sub-criteria (Appendix A). Figure 2 shows waiting list mortality substantially greater for candidates listed as Status 1A as compared to and. Posttransplant mortality rates are more similar across statuses. This review also assessed potentially underserved populations, such as adults with congenital heart disease (CHD), re-transplantation, restrictive cardiomyopathy and hypertrophic cardiomyopathy. Observed waiting list mortality at 6 months was similar for all diagnoses except retransplant, while posttransplant mortality was highest for the re-transplant and CHD diagnoses (Figure 3). Amyloidosis and restrictive cardiomyopathy were associated with the lowest posttransplant mortalities. However, specific types of amyloid were not collected by the OPTN. Review of this information, helped the Subcommittee identify specific patient populations within Status 1A and 1B that may deserve a more urgent status. The need for status exceptions The Heart Subcommittee also analyzed the Status 1A and 1B exception requests to ensure that candidates who did Post-transplant mortality Wai ng list mortality 0% 2% 4% 6% 8% 10% 12% % died within 6 months Status 1A Figure 2: Six-month mortality stratified by status at listing (for waiting list mortality rates) and status at transplant (for posttransplant mortality rates) for adult heart listings between 2010 and 2011, and adult heart transplants between 2010 and ALL 46 American Journal of Transplantation 2015; 15: 44 54

4 Heart Allocation in the United States % died within 6 months 14% 12% 10% 8% 6% 4% 2% 0% Wai ng list mortality Post-transplant mortality not fit into current status criteria were not disenfranchised by the future allocation system. All Status 1A and 1B exception requests submitted between July 2009 and June 2011 were reviewed and categorized (Figures 4 and 5). The common diagnoses representing the Status 1A and 1B exception requests were ventricular tachycardia and fibrillation (VT/VF), CHD and the no intravenous (IV) access for inotropes/pulmonary artery catheter contraindicated candidates. Of the 605 Status 1A exception requests examined, 563 (93%) were approved by the regional review board (RRB), and nearly 80% of the Status 1A candidates and 75% of the candidates whose requests were approved were transplanted, though not necessarily while listed with an approved exception. This analysis helped the ALL Amyloidosis Congenital CAD Dilated CM Hypertrophic CM Restric ve CM Retransplant Figure 3: Six-month waitlist mortality and posttransplant mortality stratified by diagnosis for adult heart listings between 2010 and 2011, and adult heart transplants between 2010 and Subcommittee determine which additional patient populations require placement in the modified allocation tiers. The status of VAD patients awaiting transplantation The increased use of left VADs (LVADs) has contributed to the larger numbers of Status 1A and 1B patients listed over the last 5 years (Figure 6), while the waitlist mortality for stable VAD patients is lower than many other candidates registered under other Status 1A criteria (6). Much of the allocation debate revolves around the increased use of LVADs and the appropriate status designation of stable continuous flow LVAD patients, which has changed the outcomes of patients with end-state heart failure (7). The studies by Dardas et al (8) and Pinney (9) showed that stable LVAD patients have a low risk of adverse events, defined as death or waitlist removal for being too ill to transplant (1% cumulative hazard) when compared to Status 1A patients on dual inotropes (6% cumulative hazard), and far lower than Status 1A patients on paracorporeal VADs (15% cumulative hazard). However, waitlist mortality increases and the hazard of an adverse event for the stable LVAD group changes significantly in the setting of LVAD complications (6% cumulative hazard). Similarly, there is a small but significant decrement in posttransplant survival for those patients who have had VAD complications (Figures 7 and 8). Therefore, the Heart Subcommittee decided upon a lower tier still within the higher acuity tier range would be an acceptable compromise to the debate over allocation status in the stable LVAD population. Additionally, due to the increased mortality for patients with LVAD complications, the Subcommittee debated whether VAD complications, 11% Unable to tolerate inotropes, 8% Other: Miscellaneous, 8% Hypertrophic CM, 4% Congenital, 14% Retransplant, 5% Other, 8% Restrictive CM, 3% No IV access for inotropes/pac contraindicated, 18% CAD refractory angina, 1% Amyloidosis, 1% VT/VF, 27% (N=640) Figure 4: Categorization of Status 1A exception requests submitted between July 2009 and June American Journal of Transplantation 2015; 15:

5 Meyer et al CAD refractory angina, 11% Restrictive CM, 7% Unable to tolerate inotropes, 6% VAD complications, 6% Amyloidosis, 4% Retransplant, 12% Hypertrophic CM, 5% Other, 7% Other: Miscellaneous, 2% Congenital, 14% No IV access for inotropes/pac contraindicated, 1% VT/VF, 32% (N=310) Figure 5: Categorization of exception requests submitted between July 2009 and June to continue or change the duration of the Status 1A time available for LVAD patients. In a contemporary cohort, approximately 35% of adult heart recipients had an LVAD at transplant, compared to just over 20% for transplants performed 5 6 years ago. The Subcommittee questioned whether different Status 1A time intervals for LVAD candidates would increase transplant rates while not adversely affecting other candidates on the waiting list. The SRTR used thoracic simulation allocation modeling (TSAM) (10) to model the potential effect of changing the amount of Status 1A time available to VAD candidates (Appendix B). The TSAM did not show a difference in the 2,750 Status 1A 2,500 2,250 Candidates (N) 2,000 1,750 1,500 1,250 1, /5/2006 1/5/2007 7/5/2007 1/5/2008 7/5/2008 1/5/2009 7/5/2009 1/5/2010 7/5/2010 1/5/2011 7/5/2011 1/5/2012 7/5/2012 1/5/2013 7/5/2013 Snapshot date Figure 6: Heart candidate status distribution between July 2006 and July American Journal of Transplantation 2015; 15: 44 54

6 Heart Allocation in the United States VAD < 30 days, TAH, IABP, ECMO (N=799) IV inotropes + PAC (N=694) Device complica ons (N=500) Excep on (N=179) complications justify a more urgent status listing in the modified allocation scheme. The use of uniform definitions will improve data quality that will inform future allocation policy decisions. Survival (%) Time (days) Figure 7: Posttransplant survival for Status 1A adult heart transplants performed between 2010 and 2011 stratified by criteria (p ¼ ). Note: There were eight recipients in criteria c (ventilator) at the time of transplant. Survival (%) 100 LVAD < 30 days (N=491) BiVAD < 30 days (N=77) TAH (N=47) IABP (N=164) ECMO (N=17) Device complica ons (N=500) IV inotropes + PAC (N=694) Excep on (N=179) Time (days) Figure 8: Posttransplant survival for Status 1A adult heart transplants performed between 2010 and 2011 stratified by criteria and sub-criteria for Status 1A criteria A (p ¼ ; p ¼ after excluding TAH and ECMO). Note: There were three recipients with RVAD < 30 days at time of transplant (criteria A, sub-criteria (i), RVAD). number of waitlist deaths or transplant rates between 30-, 45-, 60-, and 90-day time allowances (unpublished data, presented by SRTR research staff to UNOS Heart Subcommittee via conference call on August 27, 2013). Thus, extending the elective Status 1A times for LVAD patients awaiting heart transplantation did not appear to have an impact on either the waiting list mortality or likelihood of transplantation. The Subcommittee has also tried to better define the device complications that could be used to justify Status 1A registrations. They developed a guidance document, defining VAD complications of sufficient severity to justify RRB approval of a Status 1A registration (11). The Subcommittee may use these definitions when determining which device American Journal of Transplantation 2015; 15: Options for improved zonal or geographic sharing In November 2012, the Board charged all of the organspecific committees to address geographic disparity of access to transplantation by identifying a metric for fairness in geographic allocation. Currently, the time a candidate spends waiting for transplant and the number of transplants performed on higher urgency candidates varies across geographic regions recently ranging from 37% to 86% by region (Figure 9). Options to increase regional access to potential donors include offering hearts to Status 1A candidates across a broader geographic area prior to allocating the hearts to and 2 candidates as well as eliminating local zone allocation altogether. Another potential scheme is based on the number of donors projected to be available in a certain geographic area. Modeling geographic heart allocation is expected to occur following completion of the overall tiered allocation structure. Defining and prioritizing highly sensitized candidates Management of highly sensitized patients in the allocation algorithm is another area of significant discussion as definitions are critical and must be consistent across heart transplant programs. OPTN data to guide these deliberations is limited because of the variability in HLA laboratory practice and its impact on accuracy and reporting. The Subcommittee has reviewed allocation in other systems. The Canadian system considers a patient with a calculated panel reactive antibody of 80% to be sensitized, and these patients qualify for the second highest priority in their tiered system. The Subcommittee will consult with the OPTN/ UNOS Histocompatibility Committee to attempt to answer these questions and develop an equitable solution for the highly sensitized patients awaiting heart transplantation. % of transplants Region Status 1A Figure 9: The proportion of adult heart recipients undergoing heart transplant between 2010 and 2012 as Status 1A, by region. 49

7 Meyer et al Implementation of the Modified Allocation System The analysis of the considerations detailed above resulted in a straw man model (Table 3) which serves as a starting point for SRTR s TSAM analysis. For the purposes of the TSAM, SRTR will place candidates in the most urgent tier for which they qualify. The TSAM will help to project whether candidates would have better waitlist survival in the revised allocation system, and whether posttransplant survival rates would remain stable. The modeling with TSAM will be an iterative process. The first model will be based on the current allocation rules as closely as possible, with each subsequent model reflecting only slight changes to the rules. This will enable the Subcommittee to better analyze how specific proposed policy changes would impact allocation and outcomes. The tiers that make up the straw man are therefore subject to change, based on further analysis by the Subcommittee. The implementation of the modified heart allocation system will depend upon the results of the modeling being performed by the SRTR. After the modified system is deemed acceptable by the Thoracic Committee, the proposal will be distributed for public comment. After the public comment period closes and the comments are reviewed, the Thoracic Committee will determine if further public comment is necessary. After addressing the comments and making revisions where appropriate, the proposal will be presented to Table 3: Proposed Straw Man tiers for heart allocation model Tier Proposed straw man tiers Corresponding criteria in OPTN/UNOS policy as of September 1, i. ECMO i. Status 1A(a)(iv) ii. Mechanical ventilation ii. Status 1A(c) iii. Nondischargeable VAD iii. Subset of Status 1A(a)(i) and subset of (aa) iv. Mechanical circulatory support with life-threatening iv. Status 1A(b)(iv) ventricular arrhythmia 2 i. Intra-aortic balloon pump i. Status 1A(a)(iii) ii. Ventricular tachycardia/ventricular fibrillation, mechanical ii. Subset of Status 1A exceptions support not required iii. Mechanical circulatory support with device malfunction/ iii. Status 1A(b)(iii) mechanical failure iv. Total artificial heart iv. Status 1A(a)(ii) v. Dischargeable BiVAD or RVAD v. Subset of Status 1A(a)(i) and subset of (aa) 3 i. LVAD for up to 30 days i. Subset of Status 1A(a)(i) ii. Status 1A exception ii. Status 1A(e) iii. Multiple inotropes or single high-dose inotropes with iii. Status 1A(d) continuous hemodynamic monitoring iv. Mechanical circulatory support with device infection iv. Status 1A(b)(ii) v. Mechanical circulatory support with thromboembolism v. Status 1A(b)(i) vi. Mechanical circulatory support with device-related vi. Status 1A(b)(v) complications other than infection, thromboembolism, device malfunction/mechanical failure or life-threatening ventricular arrhythmia 4 i. Diagnosis of congenital heart disease (CHD) with: i. NA a. Unrepaired/incompletely repaired complex CHD, usually with cyanosis b. Repaired CHD with two ventricles (e.g., TOF, TOGV) c. Single ventricle repaired with Fontan or modifications ii. Diagnosis of ischemic heart disease with intractable angina ii. NA iii. Diagnosis of hypertrophic cardiomyopathy iii. NA iv. Diagnosis of restrictive cardiomyopathy iv. NA v. Diagnosis of amyloidosis v. NA vi. Stable LVAD candidates after 30 days vi. Subset of (aa) vii. Inotropes without hemodynamic monitoring vii. (bb) viii. Retransplant viii. NA ix. exceptions ix. ¼ exception 5 Approved combined organ transplants: heart lung; heart Not applicable liver; heart kidney 1 6 All remaining active candidates 7 Inactive/not transplantable Inactive 1 Tier five candidates may ascend to a higher acuity tier if indicated based upon cardiac status. 50 American Journal of Transplantation 2015; 15: 44 54

8 Heart Allocation in the United States the OPTN/UNOS Board. If approved, the final phase of project will involve programming of the revised allocation system, with implementation of the modified system dependent upon the available information technology resources. Limitations of the Proposed Modified System Based upon data available to the Thoracic Committee, there are invariably areas where knowledge deficits exist. Examples include the outcomes of patients implanted with a total artificial heart (TAH) or with short-term mechanical circulatory support devices (12,13). The small numbers of implants relative to other devices and therapies make any meaningful data analysis regarding waiting list mortality and posttransplant survival very limited. Moreover, as modeling is a key aspect of the proposed allocation system, whether the proposed changes produce a significantly more efficient or equitable allocation of organs cannot be proven until the system is implemented and results assessed. Summary OPTN/UNOS efforts over the past three decades to reduce waiting list mortality and improve posttransplant survival have been successful. However, with steady growth in the number of heart transplant candidates, the OPTN/UNOS Board mandated the Thoracic Committee to evaluate opportunities for broader, more equitable organ sharing of donor hearts. The Thoracic Committee has begun work on a new heart allocation system that expands the number of allocation statuses, accounts for severity of illness, considers geographic variations in heart allocation, incorporates the emerging population of VAD patients, with and without associated complications, includes previously disenfranchised groups of patients, and focuses on candidates underlying physiology rather than their diagnoses. These modifications will lead to the development of a more effective and equitable heart allocation system. Acknowledgments This work was performed in part for the OPTN/UNOS Thoracic Organ Transplantation Committee and was supported in part by Health Resources and Services Administration contract C. The content is the responsibility of the authors alone and does not necessarily reflect the views or policies of the Department of Health and Human Services, nor does mention of trade names, commercial products, or organizations imply endorsement by the US Government. The authors would also like to acknowledge the contributions made by the SRTR in the analyses that support the ongoing development of the heart allocation tiers. Disclosure The authors of this manuscript have conflicts of interest to disclose as described by the American Journal of Transplantation. D. M. Meyer Institutional PI for Heartware Endurance Trial, Clinical Events Committee REVIVE-IT Trial and Sunshine Heart, Inc. M. R. Johnson Consultant: CareMark, Clinical Events Committee: Transmedics, Advisory Board meeting attendee: XDX. J. D. Vega Institutional research support from Thoratec and Heartware. J. C. Cleveland Jr Institutional PI for Heartware Endurance Trial. M. J. Zucker Research support and/or advisory board for Thoratec, Heartware, and Circulite, and consultant for Sunshine Heart, Hemostemix Corporation and Isis Pharmaceutical. References 1. National Organ Procurement Act, U.S.C. x OPTN/UNOS Policy 3.7: Allocation of Thoracic Organs, September 1, Stevenson LW. The urgent priority for transplantation is to trim the waiting list. J Heart Lung Transplant 2013; 32: Smits JM, devries E, DePauw M, et al. Is it time for a cardiac allocation score? First results from the Eurotransplant pilot study on a survival benefit-based heart allocation. J Heart Lung Transplant 2013; 32: Moazami N, Sun B, Feldman D. Stable patients on left ventricular assist device support have a disproportionate advantage: Time to re-evaluate the current UNOS policy. J Heart Lung Transplant 2011; 30: Wever-Pinzon O, Drakos SG, Kfoury AG, et al. Morbidity and mortality in heart transplant candidates supported with mechanical circulatory support: Is reappraisal of the current United Network for Organ Sharing thoracic organ allocation policy justified? Circulation 2013; 127: Dardas T, Mokadam NA, Pagani F, Aaronson K, Levy WC. Transplant registrants with implanted left ventricular assist devices have insufficient risk to justify elective Organ Procurement and Transplantation Network status 1A time. J Am Coll Cardiol 2012; 60: Pinney SP. Timing isn t everything: Donor heart allocation in the present LVAD era. J Am Coll Cardiol 2012; 60: Schaubel DE, Dykstra DM, Murray S, et al. Analytical approaches for transplant research, Am J Transplant 2005; 5: OPTN/UNOS. Guidance regarding adult heart Status 1A(b) devicerelated complications. June 25, Available at: transplant.hrsa.gov/sharedcontentdocuments/guidance_adult_ Heart_Status_Device_Related_Complications.pdf. Accessed June 25, Kirsch ME, Nguyen A, Mastroianni C, et al. SynCardia temporary total artificial heart as bridge to transplantation: Current results at la pitie hospital. Ann Thorac Surg 2013; 95: Prashant NM, Bartlomiej Z, Aron FP, et al. CentriMag 1 short-term ventricular assist as a bridge to solution in patients with advanced heart failure: Use beyond 30 days. Eur J Cardiothorac Surg 2013; 44: e310 e315. American Journal of Transplantation 2015; 15:

9 Meyer et al Appendix A: Status Criteria for Adult Heart Candidates (OPTN Policy Implemented as of September 1, 2013) Adult candidate status Each candidate awaiting heart transplantation receives a status code corresponding to the candidate s medical urgency for transplant. A heart transplant candidate at least 18 years of age at the time of listing receives a status code as follows: Status Status 1A Definition A candidate listed as Status 1A is admitted to the listing transplant center hospital (with the exception for a 1A(b) candidate) and has at least one of the following devices or therapies in place: (a) Mechanical circulatory support for acute hemodynamic decompensation that includes at least one of the following: (i) left and/or right ventricular assist device implanted candidates listed under this criterion, may be listed for 30 days at any point after being implanted as Status 1A once the treating physician determines that they are clinically stable. Admittance to the listing transplant center hospital is not required. (ii) Total artificial heart; (iii) Intra-aortic balloon pump; or (iv) Extracorporeal membrane oxygenator (ECMO). Qualification for Status 1A under criterion 1A(a)(ii), (iii), or (iv) is valid for 14 days and must be recertified by an attending physician every 14 days from the date of the candidate s initial listing as Status 1A to extend the Status 1A listing. (b) Mechanical circulatory support with objective medical evidence of significant device-related complications such as thromboembolism, device infection, mechanical failure, or life-threatening ventricular arrhythmias. A transplant center can report a complication not listed here. The report of an other complication will result in a review by the respective heart regional review board. (Candidate sensitization is not an appropriate device-related complication for qualification as Status 1A under this criterion. The applicability of sensitization to thoracic organ allocation is specified by Policy (Exception for Sensitized Candidates).) (i) Admittance to the listing center transplant hospital is not required. Qualification for Status 1A under this criterion is valid for 14 days and must be recertified by an attending physician every 14 days from the date of the candidate s initial listing as Status 1A to extend the Status 1A listing. (c) Continuous mechanical ventilation. Qualification for Status 1A under this criterion is valid for 14 days and must be recertified by an attending physician every 14 days from the date of the candidate s initial listing as Status 1A to extend the Status 1A listing. (d) Continuous infusion of a single high-dose intravenous inotrope or multiple intravenous inotropes, in addition to continuous hemodynamic monitoring of left ventricular filling pressures. Qualification for Status 1A under this criterion is valid for 7 days and may be renewed for an additional 7 days for each occurrence of a Status 1A listing under this criterion for the same candidate. The OPTN contractor shall maintain in the heart status justification form in UNet SM a list of the specific inotropes and doses approved by the Board of Directors to be compliant with this criterion. Status 1A by Exception A candidate who does not meet criterion (a), (b), (c), or (d) may nevertheless be Status 1A upon application by his or her transplant physician. The transplant physician must justify to the applicable Regional Review Board why the candidate is considered, using acceptable medical criteria, to have an urgency and potential for benefit as other candidates in Status 1A. The justification must be for a candidate admitted to his or her listing transplant center hospital and must include a rationale for incorporating the exceptional case as part of Status 1A. Timing of the review of these cases, whether prospective or retrospective, will be left to the discretion of each Regional Review Board. A candidate s listing under this exceptional provision is valid for 14 days. Any further extension of the Status 1A listing by exception requires prospective review and approval by a majority of the Regional Review Board Members. If Regional Review Board approval is not given, the candidate s transplant physician may list the candidate as Status 1A, subject to automatic referral to the Thoracic Organ Transplantation Committee. A report of the decision of the Regional Review Board and the basis for it shall be forwarded for review by the Thoracic Organ Transplantation Committee. The Thoracic Organ Transplantation Committee may refer the case to the Membership and Professional Standards Committee. Submission of Status 1A Justification Form 52 American Journal of Transplantation 2015; 15: 44 54

10 Heart Allocation in the United States Adult candidate status (Continued) Status Definition A completed Heart Status 1A Justification Form must be submitted in UNet SM in order to list a candidate as Status 1A, or extend his or her listing as Status 1A in accordance with the criteria listed above. When a candidate s time at Status 1A expires, the candidate will automatically be classified as. The attending physician must classify the candidate as or 7 if the candidate s medical condition does not qualify for Status 1A or. A candidate listed as has at least one of the following devices or therapies in place: (aa) left and/or right ventricular assist device implanted; or (bb) continuous infusion of intravenous inotropes. by Exception A candidate who does not meet the criteria for may nevertheless be listed as upon application by his or her transplant physician. The transplant physician must justify to the applicable Regional Review Board why the candidate is considered, using acceptable medical criteria, to have an urgency and potential for benefit as other candidates. The justification must include a rationale for incorporating the exceptional case as part of. A report of the decision of the Regional Review Board and the basis for it shall be forwarded for review by the Thoracic Organ Transplantation Committee. The Thoracic Organ Transplantation Committee may refer the case to the Membership and Professional Standards Committee. Submission of Justification Form A completed Heart Justification Form must be submitted to UNet SM in order to list a candidate as. A candidate who does not meet the criteria for Status 1A or 1B is listed as. Status 7 A candidate listed as Status 7 is considered temporarily unsuitable to receive a thoracic organ transplant. Appendix B: Methodology of the Thoracic Simulated Allocation Model (TSAM) The Thoracic Simulated Allocation Model (TSAM) is a program that has been used routinely by SRTR to assess policy proposals. 1 The TSAM simulates the arrival of donated organs and new candidates on the waiting list over a 2-year period; checks compatibility of organs with candidates on the waiting list at the time an organ becomes available; creates ordered lists of compatible candidates (candidates with higher priority for receiving the organ over candidates with lower priority in each ordered list); simulates candidate acceptance or refusal of organ offers using a logistic regression model based on organ acceptance behavior in July 2009 June 2011; calculates number of transplants and number of organs discarded; and uses Cox proportional hazard models to project patient survival for each transplant. The TSAM repeats this process 10 times for each set of allocation rules, each time randomly permuting the order of donor arrivals and generating new random numbers to determine organ offer acceptance. Since the same donors and candidates are used in each of the simulations, and they are the actual donors and candidates from July 1, 2009 June 30, 2011 and not independent samples, statistical tests of comparisons are 1 Scientific Registry of Transplant Recipients. Thoracic Simulation Allocation Model (TSAM) User Guide. Version 4. Available at: Accessed June 10, American Journal of Transplantation 2015; 15: not possible. Instead, the average and the minimum maximum range of results for the 10 iterations are described for each set of allocation rules. Of note, this range reflects variability of the simulation modeling, not variability in actual organ allocations. Fixed inputs to the TSAM include donor lists, candidate lists and candidate status changes. Randomness is introduced to the system by permuting donor organ arrival dates and times, and through probability models, including organ acceptance models, posttransplant survival models and waiting list survival models. Organ acceptance is predicted using logistic regression models with candidate and donor covariates. Separate organ acceptance models are used for heart, lung and heart-lung acceptance. Posttransplant survival is predicted using Cox proportional hazard regression models with recipient and donor covariates. Waiting list survival models are used to extend status update histories for candidates who receive a transplant in actuality by matching the transplanted candidate to a similar waitlisted candidate and borrowing the additional history. If this borrowing process were not done, a candidate who was transplanted in real life could never receive a transplant in TSAM at a later date than the real-life transplant. All candidates should be able to receive organ offers until removal from the waitlist, death while waiting or end of the observation period. Candidates are matched based on predicted waiting list mortality, which is modeled using Cox proportional hazard regression with candidate factors. 53

11 Meyer et al Finally, allocation rules may be altered to compare a proposed allocation scheme to that currently in use. While the TSAM is a helpful tool for exploring proposed changes to the allocation system, it has several important limitations. TSAM assumes acceptance behavior does not change in response to simulated policy changes; and moreover, it is based on historical acceptance behavior. TSAM assumes that all organ offers follow the stated allocation rules, and does not allow for exceptions or expedited placements. TSAM models are limited by the available data during the cohort period. All prediction models include uncertainty; TSAM relies on 13 separate models. 54 American Journal of Transplantation 2015; 15: 44 54