Impact of the 2018 UNOS Heart Transplant Policy Changes on Patient Outcomes

Abstract

In 2018, the United Network for Organ Sharing implemented a 6-tier allocation policy to replace the prior 3-tier system. Given increasing listings of critically ill candidates for heart transplantation and lengthening waitlist times, the new policy aimed to better stratify candidates by waitlist mortality, shorten waiting times for high priority candidates, add objective criteria for common cardiac conditions, and further broaden sharing of donor hearts. There have been significant shifts in cardiac transplantation practices and patient outcomes following the implementation of the new policy, including changes in listing practices, waitlist time and mortality, transplant donor characteristics, post-transplantation outcomes, and mechanical circulatory support use. This review aims to highlight emerging trends in United States heart transplantation practice and outcomes following the implementation of the 2018 United Network for Organ Sharing heart allocation policy and to address areas for future modification.

CENTRAL ILLUSTRATION

Impact of the 2018 United Network for Organ Sharing Heart Transplant Policy Change

ACHD = adult congenital heart disease; ECMO = extracorporeal membrane oxygenation; HCM = hypertrophic cardiomyopathy; HT = heart transplantation; IABP = intra-aortic balloon pump; ICU = intensive care unit; LVAD = left ventricular assist device; MCS = mechanical circulatory support; RCM = restrictive cardiomyopathy.

The U.S. Organ Procurement and Transplantation Network (OPTN) operates through the United Network for Organ Sharing (UNOS) to develop and implement transplant allocation policy. The heart allocation system has developed over time in response to multiple factors including the advent of left ventricular assist devices (LVADs), patient outcomes, physician listing practices, and geographic variations in wait times. Medical urgency on the heart transplant waitlist is determined primarily by listing status, stratifying patients from most to least urgent.

To date, there have been 4 major iterations of the U.S. heart transplant allocation policy. The OPTN/UNOS first introduced a 2-tier heart allocation system in 1988, which was based upon urgency, waiting time, geography, and blood type.¹ This was expanded to a 3-tier system in 1998, and expanded again to include broader sharing in 2006, with a focus on reducing waitlist time and mortality for the most critically ill candidates. Despite these changes, the discrepancy between the number of donor hearts and heart transplant candidates continued to grow, with listings more than doubling since 2006. In particular, the number of critically ill patients awaiting transplant had more than tripled, with status 1A and 1B patients comprising more than 70% of those awaiting transplant in 2015.^2,3 Much of the issue with the 2006 system was that medical urgency was determined based on physician decisions about when to initiate therapies, such as inotropes and mechanical circulatory support, rather than an objective allocation score or specific hemodynamic criteria. Thus, in 2018, the OPTN/UNOS implemented a revised 6-tier allocation policy intended to better stratify candidates by waitlist mortality, shorten waiting times for high priority candidates, add objective criteria for common cardiac conditions, further broaden sharing of donor hearts, and collect clinical data for future development of a heart allocation score (Table 1). This review aims to highlight trends in U.S. heart transplantation (HT) practices and outcomes since the introduction of the 2018 OPTN/UNOS heart allocation policy and discuss potential measures to address shortcomings of the new system (Central Illustration).

TABLE 1.

Changes in the 2018 UNOS Heart Transplant Policy

Adult Heart Allocation Criteria
1999–2018	2018-Present
Status categories
Status 1A Admitted with TAH/IABP/ECMO LVAD with complications Continuous ventilation Continuous single or multiple inotropes requiring hemodynamic monitoring Dischargeable LVADs for 30 days	Status 1 ECMO (up to 7 daysa) Nondischargeable, surgically implanted, nonendovascular BiV support device MCS with life-threatening ventricular arrhythmia
	Status 2 Nondischargeable, surgically implanted, nonendovascular LVAD (up to 14 daysa) IABP (up to 14 days)a VT/VF, mechanical support not required MCS with device malfunction/mechanical failure TAH, BiVAD, RVAD, or VAD for single ventricle patients Percutaneous endovascular MCS (up to 14 daysa)
	Status 3 Dischargeable LVAD for discretionary 30 days Multiple inotropes or single high dose inotropes with continuous hemodynamic monitoring MCS with device infection, hemolysis, pump thrombosis, right heart failure, mucosal bleeding, or aortic insufficiency Temporary MCS after 14 days (7 days for ECMO) without reapproval
Status 1B All LVADs Continuous inotrope infusion	Status 4 Stable LVAD candidates not using 30-day discretionary period Inotropes without hemodynamic monitoring CHD, HCM/RCM, or amyloidosis IHD with intractable angina Retransplantation
Status 2 All other listed candidates	Status 5 Combined organ transplants (on waitlist for at least 1 other organ at same hospital)
	Status 6 All remaining active candidates
Geographic changes
Assigns donation service areas and concentric bands called “zones” to determine heart distribution Candidates are prioritized based on HLA sensitization Stratified by status and blood type	250 nautical mile fixed distance circle from donor hospital HLA sensitization exception removed Stratified by status and blood type

^aPrograms must justify need for continued support at end of this period.

BiV = biventricular; BiVAD = biventricular assist device; CHD = congenital heart disease; ECMO = extracorporeal membrane oxygenation; HCM = hypertrophic cardiomyopathy; HLA = human leukocyte antigen; IABP = intra-aortic balloon pump; IHD = ischemic heart disease; LVAD = left ventricular assist device; MCS = mechanical circulatory support; RCM = restrictive cardiomyopathy; RVAD = right ventricular assist device; TAH = total artificial heart; VAD = ventricular assist device; VF = ventricular fibrillation; VT = ventricular tachycardia; UNOS = United Network for Organ Sharing.

LISTING PRACTICES

URGENCY STATUS.

Before 2018, the majority of patients were listed as the highest priority statuses, statuses 1A and 1B.³ With the introduction of the 6-tier system targeting enhanced risk stratification, only the most critically ill patients who meet specific criteria are listed in the highest priority statuses (Table 1, Figure 1). Patients with prior status 1A indications are now stratified into statuses 1–3. In early analyses, up to 35% were listed as status 1–3 compared to 25% of the 1A cohort from the prior era.⁴ In the most recent Scientific Registry of Transplant Recipients (SRTR) report, 36% of all waiting list registrations in 2021 were adult status 1–3 at listing, although <5% of listings were status 1 and received the highest priority. The UNOS Benchmark Report noted that <1% of adult waitlist candidates were status 1 in a snapshot of the waiting list on March 31, 2022.

FIGURE 1. New Criteria for Highest Urgency Listings.

Flowchart depicting the new, specified criteria regarding cardiogenic shock, life-threatening ventricular arrhythmia, and mechanical circulatory support (MCS) complication qualifying for high-urgency listing. ALT = alanine transaminase; AST = aspartate transaminase; BiV = biventricular; CI = cardiac index; CPR = cardiopulmonary resuscitation; ECMO = extracorporeal membrane oxygenation; K = potassium; MAP = mean arterial pressure; Mg = magnesium; PCWP = pulmonary capillary wedge pressure; SBP = systolic blood pressure; SvO₂ = mixed venous oxygen saturation; VF = ventricular fibrillation; VT = ventricular tachycardia.

In a comparative analysis of the 2 eras, a larger proportion of transplantation procedures (78% vs 68%) were performed in the highest priority patients following the policy change.⁴ Baseline characteristics of candidates listed for transplantation in the new era are largely similar to the prior era, except that a greater proportion of candidates require intensive care unit (ICU)–level care (61.7% vs 31.0%; P < 0.001) or are supported by temporary mechanical circulatory support (MCS) or mechanical ventilation.^5,6 Venoarterial extracorporeal membrane oxygenation (VA-ECMO) and intra-aortic balloon pump (IABP) use, corresponding to status 1 and status 2 listings, respectively, have significantly increased following the policy change.^4,7 Overall, the number of candidates listed for transplantation remains similar between eras, and transplantation centers and organ procurement organizations with greater proportions of high-priority listings in the pre-change era remained more likely to have high-priority listings in the post-change era.⁸ Interestingly, an analysis that applied the new risk stratification rules to candidates from the pre-policy era demonstrated 17% more status 2 listings than anticipated, corresponding to the observed increase in the use of IABP.⁸ The odds of high-priority listing were 5 times greater than expected following the policy change despite similar candidate characteristics.⁸ Without a clear change in candidate characteristics nor overall number of listings, these phenomena may reflect a concerted change in clinical practice by transplant providers to meet the new high-priority criteria for patients that had lower urgency by previous era criteria.

EXCEPTION REQUESTS.

A specific target of the new allocation policy was to reduce the number of patients listed by exception by including populations not explicitly addressed in the prior policy (Table 1). Ventricular arrhythmias, loss of vascular access, congenital heart disease, and LVAD complication accounted for more than 70% of the 1A exception requests; more than 90% of these were approved by a regional review board (RRB) and 80% of these were transplanted.¹ Similarly, ventricular arrhythmias, congenital heart disease, retransplantation, coronary artery disease with refractory angina, and restrictive cardiomyopathy (RCM) were the most common 1B exceptions.¹ The 2018 policy change failed to achieve the goal of reducing the number of exception listings. In fact, status change by exception has increased significantly for high-risk patients in the new era from 3.5% previously to 15%.^8–10 Waitlist addition by exception ranges from 15% for status 4 candidates to more than 31% for status 2 candidates, and more than 30% of status 1 transplant recipients were stratified by exception.¹¹ The most common reasons for exception requests are restrictive (RCM) or hypertrophic cardiomyopathy (HCM), congenital heart disease, and cardiac allograft failure. Further, there is a clustering of exception requests in status 2, and these patients prioritized by exception have significantly higher transplantation rates despite having similar risk for death or delisting compared with patients listed by standard criteria.¹²

Although most exception requests were approved in the previous era, 94% of exception requests have been approved since 2018, suggesting that there are still opportunities to refine status criteria to align with clinical risk assessment.¹¹ A potential source for high exception approval is that requests are now evaluated by RRBs in a different OPTN region which may have the unintended consequence of minimizing the implications of policy adherence to local donor availability.

To address concerns about the number of exception requests in the new policy, the UNOS Heart Committee has issued guidance on the hemodynamic measurement requirements for inotrope-supported patients, an extension of status 4 time for patients receiving inotropes without hemodynamic monitoring, and criteria for status 2 and 3 listing in patients with RCM/HCM.¹³ Significant exception use remains a pitfall of the current policy. The impact of exception use, outside RRB use, guidance statements, and policy modifications all merit further evaluation and should be addressed with future policy iterations.

WAITLIST MORTALITY

During the prior era, the HT waitlist continued to grow and supplanted the number of donors. The result was longer waiting times, with almost half of the patients remaining on the waitlist for more than 1 year, and larger proportions of high-urgency candidates with higher waitlist mortality rate.¹⁴ A primary goal of the policy change was to ensure that candidates with the highest urgency were stratified accordingly to increase access to donor hearts and, in turn, reduce waitlist time and mortality. As previously noted, the new system more accurately stratifies patients by risk of waitlist mortality.⁴ Status 1 candidates demonstrated a rate of 139 deaths/100 patient-years, whereas status 6 candidates’ rates were as low as 3.9 deaths/100 patient-years.⁴ Comparatively, status 1A patients in the old system had waiting list mortality rates ranging from 4.8% to 35.7%, and all 3 urgency statuses had mortality rates <50 deaths/100 patient-years with similar rates in statuses 1B and 2.^4,15 Several studies have analyzed waitlist mortality in the new era. Despite a greater proportion of highest priority additions (statuses 1–3) to the waitlist in the new era, waitlist mortality is similar in some studies and even reduced in others both at 90 days (6.3% vs 5.0%; P = 0.02) and 1 year (13.3% vs 11.7%; P = 0.03).^4,16–18 Similar if not improved waitlist mortality despite highest priority listings in the new era are suggestive of the policy change functioning as intended, although further data will continue to assess waitlist outcomes.

WAITLIST TIME

Time spent on the transplant waitlist is a product of many factors. Poor clinical status with high urgency naturally warrants shorter time to transplantation, whereas clinically stable patients with less urgency may safely remain waitlisted for longer periods. Organ availability also plays a role, with regional variations in organ supply and number of candidates. In the new system, overall waitlist time has decreased from 112 days to 39 days (P < 0.001) per SRTR/UNOS data and similar findings from single-institutional data.^6,16,18,19 Critically ill patients requiring VA-ECMO, IABP, and other temporary mechanical support now have shorter time to transplantation.^7,18,20 In LVAD patients listed as bridge to transplantation (BTT), median waitlist time decreased from 139.5 days to 37 days (P < 0.001) in the new system, although the majority were higher than status 4 at time of transplantation, and stable status 4 LVAD patients had a lower incidence of transplantation at 360 days.²¹ Overall, data suggest that median waitlist time has reduced with the new policy, particularly for the highest-urgency candidates.

There remains concern in the heart transplantation community that the number of patients supported with temporary rather than durable MCS will have a negative impact on waitlist outcomes. However, comparisons of waitlist morbidity between eras are limited at present. High-urgency candidates will almost certainly benefit from shorter waiting times under the new policy, but many of these candidates are being supported on device strategies known to be associated with relatively high short- and intermediate-term risk or have experienced a significant complication on a durable MCS device. A deeper analysis of the time to transplantation vs morbidity tradeoffs is critical in the evaluation of policy effectiveness.

POST-TRANSPLANTATION OUTCOMES

Although the policy change was primarily intended to reduce waitlist mortality, it is vital to ensure that the change did not have unintended negative impact on post-transplantation outcomes. Compared to the prior system, transplant recipients are of similar age, body mass index, and ethnicity have similar serum creatinine and bilirubin concentrations and have similar prevalence of diabetes, prior malignancy, and cerebrovascular disease. However, a higher number of recipients have congenital heart disease or are receiving ICU-level care, MCS, mechanical ventilatory support, or intravenous antibiotics at time of transplantation.⁶ An early analysis using UNOS registry data in the new system suggested increased risk of death or re-transplantation at 90 and 180 days.²² The investigators hypothesized the results to be a reflection of transplantation occurring in higher acuity candidates — an intended goal of the policy change.²² However, sample size limitations coupled with informative censorship and ascertainment biases likely influenced the interpretation of this analysis.^23,24 Others have suggested a signal for increased mortality at 6 months, although methodological differences may account for these discrepancies.^19,25,26 One-year post-transplantation survival from several single-center and UNOS/SRTR databases have shown similar survival or improved survival rates (92.1% vs 87.5%; P < 0.001) despite higher acuity recipients as noted above.¹⁷ In keeping with this notion, the new prioritization system appears to be stratifying mortality risk, with status 1 candidates showing higher waitlist mortality than other statuses.⁴ Although the conclusions from the above outcomes analyses are not entirely concordant, they suggest that post-transplantation mortality in the short- and intermediate-term is similar despite higher acuity candidates receiving transplants.

Analysis of short-term post-transplantation morbidity shows a similar incidence of primary graft dysfunction, hyperacute rejection, vasoplegia, ventilatory time, and ICU or hospital length of stay compared to the prior era.¹⁶ However, increased rates of in-hospital stroke (3.7% vs 2.7%; P < 0.001) and renal failure requiring dialysis (17.1% vs 11.7%; P < 0.001) have been noted and are associated with pretransplantation ECMO and IABP support and longer ischemic times.^18,27 Before policy change, rejection and infection were common causes of early hospital readmission.²⁸ These outcomes must be assessed in the current policy era. One-year post-transplantation measures of ejection fraction, surgical wound infections, readmission rates, rejection, development of donor-specific antibodies, cardiac allograft vasculopathy, and retransplantation are similar between eras.¹⁶ However, longer-term outcomes associated with mortality, including renal failure, graft dysfunction, coronary vasculopathy, and quality of life, must be assessed.

With increased rates of transplantation among high-urgency candidates in the new era, those with lower urgency (status 6) are less likely to be transplanted and likely will have longer wait times or experience a change in clinical course that qualifies for a high priority. Analysis of UNOS data shows candidate characteristics and post-transplantation outcomes in lower-urgency patients are similar between eras with low-urgency patients being more frequently older, female, recipients of organs from higher-risk donors, having longer waitlist times, and representing a much smaller proportion of transplant recipients.²⁹ Nonetheless, post-transplantation survival has remained similar among the low-urgency candidates, as was the case in the prior system.²⁹

TEMPORARY MCS USE

Following the policy change, the most marked practice change was use of temporary MCS as a BTT. At the time of listing, the number of patients supported with VA-ECMO (1.8% vs 2.7%; P = 0.02), IABP (5.3% vs 10.3%; P < 0.01), and biventricular support (1.3% vs 2.1%; P = 0.02) all increased. Other estimates have demonstrated up to tripling of temporary MCS use.^6,18,30,31 However, clinical deterioration or death on the waitlist was significantly reduced in the new system for those listed with temporary MCS (HR: 0.53; P = 0.001).³¹ Use of temporary MCS for cardiogenic shock (CS) at U.S. transplantation centers increased from 25.4% to 42.6% in the post-change era (P = 0.04) despite similar baseline patient characteristics in both eras; temporary MCS use did not increase for other forms of CS (ie, post-myocardial infarction) or in nontransplantation centers.³² In addition, there is significant center-level variability in temporary MCS use, and there is a strong correlation between the use of temporary MCS and the overall number of transplantations performed—demonstrating an avenue of potential inequity in organ allocation skewed towards centers with higher use of temporary MCS regardless of clinical acuity.³³ Low-volume transplantation centers that did not increase MCS use saw declines in monthly transplantation volumes.¹⁹ In a survey of heart failure cardiologists and surgeons, 84% of respondents reported using more temporary MCS to achieve higher urgency status.³⁴ Still, other factors may play a role in influencing MCS use, such as the implementation of institutional multidisciplinary CS teams that likely use algorithmic deployment of advanced therapies and trends toward MCS for myocardial infarction–related CS.

ECMO.

In the prior era, use of VA-ECMO was a strong predictor of 1-year post-transplantation mortality.^35,36 In the new era, an initial analysis also showed pretransplantation ECMO support as an independent predictor of post-transplantation mortality.⁵ However, subsequent analyses of patients bridged to transplant with ECMO have shown a marked improvement in 6-month posttransplantation survival following the policy change (74.6% to 91.2%; P = 0.002). This change may be a reflection of the decreased time spent on ECMO support in the new era (3 days vs 7 days; P < 0.006), the increased incidence of transplantation (69% vs 24%; P < 0.001), and a commitment of teams to optimize the care and timing of transplantation in ECMO-supported patients.⁷ Interestingly, there appear to be no significant clinical, demographic, or hemodynamic differences between patients with ECMO at time of listing in either era.⁷ Previously, patients requiring VA-ECMO support had equal priority to other 1A indications. Following the policy change, increased transplantation and survival rates may reflect their proper stratification to highest priority with timely transplantation ameliorating increased waitlist mortality. Thus, the new status 1 criteria and its inclusion of VA-ECMO makes the clinical decision to initiate ECMO use in these critically-ill patients more compelling. This decision may even be advantageous if data continue to show shorter time to transplant, thus limiting time and potential complications with ECMO. Although improved transplantation and mortality rates are encouraging, assessment of ECMO-related complications is also necessary to comprehensively assess the impact of its increased use in the current policy era. Longer-term outcomes have yet to be reported and should continue to stratify this critically ill patient cohort to allow a deeper understanding of waitlist risk of morbidity and mortality.

IABP.

IABPs constitute the largest proportion of temporary MCS use in the policy-change era. Clinical characteristics and hemodynamic profiles of those with IABP support are similar in the 2 eras except for a modest reduction in cardiac index now (1.94 L/min/m² vs 2.04 L/min per m²; P < 0.001).²⁰ Under the new policy, IABP patients have a shorter median time to transplant (13 days vs 80.5 days; P < 0.001), higher incidence of transplant (HR: 2.15; 95% CI: 1.82–2.55), and reduced waitlist death or deterioration (HR: 0.55; P = 0.011).^20,30 Survival outcomes up to 1-year post-transplantation in those bridged with IABP are similar between the 2 eras.^20,30 There are higher rates of IABP exception requests in the new policy era, likely explained by policy language requiring transition to durable MCS after 14 days of IABP support unless there is a contraindication. Additionally, patients treated with temporary MCS must meet hemodynamic criteria for CS to qualify as a status 2 candidate (Figure 1); otherwise, a listing center may apply for an exception. Clerkin et al³⁷ have shown that IABP patients who did not meet the specified hemodynamic criteria for CS had an increased risk of death or delisting (HR: 2.28; 95% CI: 1.10–4.75) relative to the patients who met these criteria, although this aptly reflects more critical illness underlying the decision to use MCS (ie, cardiopulmonary resuscitation, systolic blood pressure <70 mm Hg, etc) (Figure 1). But overall, IABP patients granted an exception are not observed to have increased waitlist mortality or delisting nor 1-year post-transplantation mortality thus far when compared to IABP patients listed by standard criteria.^20,37 Morbidity outcomes such as kidney injury and dialysis, stroke, and pacemaker insertion in IABP patients appear to be similar, but the marked increase in pre-transplantation IABP use requires further evaluation to understand the waitlist and post-transplantation implications of this shift.³⁸

IMPELLA.

Impella use in patients undergoing transplantation has increased (1% vs 4%; P < 0.01), with shorter waiting times (median 12 days vs 45 days; P < 0.01), higher rates of transplantation (80% vs 56%; P < 0.01), lower waitlist mortality (13% vs 25%; P < 0.01), and similar post-transplantation survival in the post-policy change era.³⁹

DURABLE LVAD USE

Although temporary MCS is increasingly used to bridge directly to transplantation, durable LVAD has decreased significantly.^5,37 In one analysis, LVAD-supported transplant candidates had a lower frequency of transplantation within 1 year of listing compared to the old system (52% vs 61%; P < 0.001),⁴⁰ although other estimates show similar transplantation rates.⁴¹ Despite similar waitlist mortality in BTT LVAD patients, 1-year post-transplantation survival was lower following the policy change (83.4% vs 91.7%; P < 0.001).⁴¹ The majority of these transplantations occurred in status 2 and 3 patients with higher-risk donors and longer ischemic times, which also may have contributed to the change in survival. The higher status of these LVAD patients (as opposed to status 4) is suggestive of ventricular assist device (VAD) complication driving transplant, inherently increasing risk. Other analyses have shown similar post-transplantation survival rates between the 2 eras.^40,42 A recent analysis of ECMO and durable MCS bridging strategies has shown that the selected ECMO cohort had shorter time on mechanical ventilation, fewer ICU and hospital days, and were less likely to be discharged to inpatient rehabilitation despite similar baseline patient characteristics. Those bridged with durable MCS commonly experienced complications and were listed for transplantation within 3 months and transplanted at a median of 155 days after device implantation.⁴³

With improvements in both LVAD and transplantation outcomes, clinicians and advanced heart failure patients are faced with challenging decisions about the optimal strategy for life-sustaining therapies. Stable LVAD patients have a low mortality risk with improvements in quality of life and submaximal exercise performance. However, long-term survival on LVAD is inferior to transplantation and LVAD complications increase morbidity and mortality risks with both continued device support and transplantation. A direct pathway to HT that circumvents an antecedent cardiac operation may be a desirable option for some with the recognition that mean post-transplantation survival is 12 years.⁴⁴ As a result, many believe the approach that offers the longest total survival time is combining long-term durable LVAD support followed by transplantation. The extent to which transplant allocation policy can support this goal is unclear and not well-aligned with the mandate to reduce waitlist mortality.

DONOR CHARACTERISTICS

As expected with changes in zone restrictions for heart allocation to the highest urgency candidates (Table 1), distance between donor and recipient centers has significantly increased in the new era.^4,5,7,19 In turn, average ischemic times have increased as well, from 3.0 to 3.4 hours.^4,19 Although ischemic times typically remain <4 hours, outcomes may potentially be affected with the marginally longer times in the new era. In one single-center analysis, incidence of primary graft dysfunction increased from 5.4% to 18.7% (P = 0.005) and, subsequently, 30-day mortality numerically increased from 5.4% to 9.9% (P = 0.257), which the investigators attributed to longer ischemic times, although longer-term outcomes were unchanged.⁴⁵ Another analysis of the UNOS database has shown that ischemic times >6 hours are more common in the new era in (22.1% vs 18.5%; P < 0.001), which was independently associated with decreased 1-year survival (90.9% vs 87.5%; P = 0.004).⁴⁶ Additionally, increased donor age has been noted to be independently associated with lower survival (HR: 1.09; 95% CI: 1.05–1.12) in the new era, although donor age appears numerically similar compared to the prior era.^4,5,7 The association between age and survival is undoubtedly multifactorial and involves a complex interaction between donor characteristics, less graft tolerance to increased ischemic times, possible propensity for graft dysfunction or vasculopathy, and a higher-acuity recipient population. The significant increase in distance traveled and ischemic times may confer long-term impacts in the new era and continued monitoring of long-term outcomes is necessary.

FINANCIAL IMPACT

There are limited data evaluating the financial impact of the policy change. Initial predictions suggested decreased cost-effectiveness related to increased MCS use, increased procurement radius, and longer ischemic times.⁴⁷ Procurement-related transportation costs have increased since the policy change, and Nationwide Inpatient Sample database analysis shows increased cost for index HT hospitalization related to MCS use despite similar lengths of stay.^48,49 Nonetheless, shorter waiting periods, particularly for high-priority patients who would be expected to use inpatient and critical care services, coupled with a reduction in the use of durable LVADs as a BTT may counterbalance the rising costs of transplantation given recent advances in organ procurement approaches and longer travel distances. At this point, impacts of the policy change and their financial implications are speculative, multifactorial, and require further analysis.

SPECIFIC CARDIOMYOPATHIES AND ADULT CONGENITAL HEART DISEASE

The majority of patients with HCM, RCM, and congenital heart disease qualify as status 4 at the time of listing (Table 1).⁵⁰ However, these patients were frequently transplanted under exception at a higher status in both eras.¹² Patients tended to be status 2 at time of transplantation—typically via use of IABP—although the OPTN/UNOS guidance statement for exception use in HCM/RCM patients provides criteria for status 2 listing without IABP as well.^13,50 Waitlist time is shorter and rates of transplantation are significantly higher in the new era for these patients, whereas waitlist and post-transplantation survival are similar for congenital heart disease patients.^50,51 HCM/RCM patients did show improved waitlist survival in the new era in one study, with reduction in waitlist time and higher rate of transplantation.⁵² These patients had significantly lower filling pressures and similar cardiac index, but higher use of temporary MCS, although illness severity may not necessarily be captured by these indices in HCM/RCM.⁵² In a new system aimed to better stratify mortality risk of HCM/RCM patients who may not meet traditional hemodynamic parameters for high-urgency listing, the impact appears marginal, and it remains unclear whether this small impact is a direct result of the allocation policy or practice changes.

MULTIORGAN TRANSPLANTATION

Before the 2018 policy, there were no specific allocation rules related to multiorgan transplantation that included the heart. Considering the severity of illness in this vulnerable cohort, all multiorgan candidates were allowed to be listed as status 5 with the expectation that they would require status upgrades as their cardiac condition worsened. Changes in listing patterns for multiorgan transplantation have been similar to heart alone with increased use of VA-ECMO (5.2% vs 0.9%; P < 0.001), IABP (24.4% vs 9.3%; P < 0.001), and ICU-level care (47.8% vs 29.8%; P < 0.001) in the post-policy era.⁵³

Heart-liver candidates are now most commonly listed in status 4 and status 2 (34.3% and 22.5%, respectively), with only 0.5% listed as status 1; in the previous era, 23% and 27% of these patients were listed as status 1A and 1B, respectively.⁵⁴ Similar trends for heart-lung and heart-kidney candidates have not been reported. Heart-liver candidates on the waitlist show reduced waitlist mortality in the new era, but comparable data for other multiorgan candidates are not available.⁵⁴ Increased MCS use (and high proportion of status 2 listings seen in heart-liver listings) likely reflects a similar intentional practice change as seen with isolated heart listings, which may have implications in the equitable distribution of non-heart organs and organ-specific outcomes. MCS may be used in place of inotropes to elevate priority for multiorgan candidates, after which grafts from a donor which previously may have benefited multiple potential recipients are now diverted to a single multiorgan recipient. Additionally, the choice to use MCS in heart-liver recipients, in particular, may come at the cost of increased risk of complication such as bleeding and/thrombosis.

Since the policy change, there has been no significant change in post-transplantation survival for combined heart-lung and heart-liver recipients. Heart-kidney recipients are more likely to require post-transplantation dialysis and have a higher mortality rate in the new era.⁵³ At this time, the key drivers of these outcomes are undefined but may be related to longer ischemic times and/or recipient factors.

FUTURE DIRECTIONS

Although the 2018 heart policy revision has accomplished a majority of its intended goals, opportunities remain to improve candidate prioritization and donor heart allocation. The high rate of exception requests suggests that either the policy does not adequately account for illness severity or that criteria developed for comorbidities are excessively stringent and/or misaligned with clinical practice. A deeper understanding and categorization of exception requests will be required to understand and correct this deficiency. For example, the heart committee identified that exceptions are most frequent in status 2 listings, typically caused by omission of hemodynamic data altogether or nonqualifying hemodynamic values.⁵⁵ HCM/RCM candidates for whom specific guidance has already been introduced were being listed under exception, suggesting a lack of familiarity with the policy as another source of exception use.⁵⁵ A more uncomfortable hypothesis is that programs are continuing to submit exception requests for patients with risk profiles misaligned with higher statuses because they know the approval rate is high. During policy development the heart committee heard the concerns of the transplantation community regarding minimizing potential to “game the system,” which was the rationale for interjecting more objectivity into the definitions of VAD complications, ventricular arrhythmias, and CS. Further investigation into additional cohorts whose illness severity is not being captured well will be necessary to provide guidance for RRBs or amend urgency criteria.

The predictable shifts in BTT mechanically assisted circulation from durable LVAD to temporary devices require further investigation. Specifically, the tradeoffs of complication burden, time on support, clinical stability, and cost should be evaluated. The 2018 policy change has shown that a large cohort of patients who were previously treated with durable LVAD can be safely bridged to transplant with a less-invasive temporary support strategy as long as the waiting time is shortened.

Prioritizing patients with significant presensitization remains a difficult issue without sound solution. Highly sensitized patients experience decreased access to immunocompatible grafts and thus longer wait times and worse waitlist and post-transplantation outcomes.⁵⁶ Although sensitization risk parameters are submitted at time of listing, sensitization is not directly addressed by the current policy. As a result of the precision required to immunologically match a highly sensitized patient to a suitable donor, these candidates would be expected to benefit most from access to a greater number of potential donors. The Canadian system includes a category 4S for highly sensitized candidates (calculated panel-reactive antibody >80%) that receives second highest priority as well as nationwide access to organs, although benefits of such a strategy have not yet been shown. An even broader geography for sensitized patients in the United States may mitigate the issue, although not without increasing ischemic times and cost. Further, prioritizing donors for lower-risk, presensitized candidates over higher-status patients would require careful policy consideration. More experience with emerging desensitization strategies may provide benefit outside of the purview of the allocation policy. Further study will be required to identify optimal strategies for sensitized patients.

The OPTN lung transplant committee recently approved a continuous distribution model, based on a composite allocation score of weighted categories, which is set to take effect in early 2023. Other UNOS committees are currently discussing a similar model. A continuous distribution framework has the potential to address some of the shortcomings in current policy. A sliding scale category based on time with LVAD may provide a pathway for stable BTT LVAD patients to increase priority for transplant, particularly before development of complications. A weighting based on age, race, blood type, sensitization, and location could further reduce inequities. The efficacy of the current criteria in stratifying waitlist outcomes may allow these to provide an initial framework for the scoring system. To address MCS usage, the committee may consider specification of policy language to contain more stringent requirements for MCS usage to qualify for high-urgency listing, such as failure of dual inotropes or specific criteria for demonstration of contraindication to durable MCS. A recently developed risk calculator derived from MOMENTUM 3 (Multicenter Study of MagLev Technology in Patients Undergoing Mechanical Circulatory Support Therapy with HeartMate 3) data predicts 1- and 2-year survival after HeartMate 3 LVAD implantation based on objective criteria and has the potential to address the latter area.⁵⁷ Future opportunities to address shortcomings while capitalizing on successes of the current policy are cause for optimism as the heart failure community moves forward.

CONCLUSIONS

Since the implementation of the 2018 OPTN/UNOS heart allocation policy focused on improving risk stratification of critically ill patients to the highest priority and improving waitlist outcomes, several trends have emerged in heart transplantation practice patterns (Figure 2). Although patient demographics and clinical characteristics remain similar, more high-priority listings have been observed. There has been a marked shift in the use of MCS devices away from durable LVADs to direct bridge with temporary MCS. In turn, high-priority patients spend less time on the waitlist, receive the transplant much sooner, and have similar if not improved waitlist and post-transplantation mortality compared with the prior era. Patients with congenital heart disease, HCM, and RCM have seen similar trends. Patients with BTT LVAD complications have shorter time to transplantation, although those with stable LVAD courses may spend more time on the waitlist. The overall findings of improved waitlist outcomes and stable mortality at face value reflect a positive impact of the allocation change, although the current policy is not without shortcomings. Future studies will continue to identify and assess areas of intervention as we refine transplantation policy and optimize the care of advanced heart failure patients.

FIGURE 2. Changes Observed Following Implementation of the 2018 UNOS Heart Transplant Policy.

Diagram summarizing overall trends and changes observed after implementation of the 2018 UNOS heart transplant allocation policy. BTT = bridge to transplantation; HT = heart transplantation; IABP = intra-aortic balloon pump; ICU = intensive care unit; LVAD = left ventricular assist device; UNOS = United Network for Organ Sharing; other abbreviations as in Figure 1.

HIGHLIGHTS.

• The 2018 United Network for Organ Sharing policy introduced a 6-tier allocation system to address shortcomings of the prior policy.

• Since implementation, significant changes in waitlist outcomes and advanced heart failure practice patterns have emerged.

• Further study is necessary to understand the long-term impacts of the observed changes in

• practice.

ABBREVIATIONS AND ACRONYMS

CS

cardiogenic shock

HCM

hypertrophic cardiomyopathy

HT

heart transplantation

IABP

intra-aortic balloon pump

LVAD

left ventricular assist device

MCS

mechanical circulatory support

OPTN

U.S. Organ Procurement and Transplantation Network

RCM

restrictive cardiomyopathy

RRB

regional review board

SRTR

Scientific Registry of Transplant Recipients

UNOS

United Network for Organ Sharing

VA-ECMO

venoarterial extracorporeal membrane oxygenation