Improvement in Racial Disparities in Heart Transplantation following the Heart Allocation Policy Change

Objectives . Heart transplantation (HT) is a defnitive therapy for refractory heart failure, making it the gold-standard treatment for recipients with end-stage disease. Heart allocation policy (HAP) in the United States was changed on October 18th, 2018. Te aim of this study was to assess the efect of the new policy on racial disparities in heart transplantation (HT) outcomes. Methods . Te United Network for Organ Sharing (UNOS) registry was used to identify adult recipients undergoing isolated HT between 2010 and 2021. Recipients were stratifed into pre-HAP (January 2010 to September 2018) vs. post-HAP (October 2018 to September 2021). Recipient race was classifed as White, Black, Hispanic, or other. Te primary outcome was post-HT mortality. Cox proportional hazard models were used for risk-adjustment in evaluating the independent efect of race on post-HT mortality. Results . A total of 27,403 recipients underwent HTin 143 centers during study period. Te proportion of non-Whites undergoing HT increased in the post-HAP era: (pre-HAP: White 66.0%, Black 21.2%, Hispanic 8.2%, Other 4.6% versus post-HAP: White 62.5%, Black 23.2%, Hispanic 9.5%, Other 4.8%; p < 0 . 001). In risk-adjusted analysis, Black recipients were at higher risk of post-HTmortality in the pre-HAP era (HR 1.31, 95% CI 1.22–1.41; p < 0 . 001) but not in the post-HAP era (HR 1.12, 95% CI 0.03–1.34; p � 0 . 222) compared to White recipients. Other non-White recipients had comparable risk-adjusted post-HT mortality rates compared to White recipients both in the pre-HAP and post-HAP eras. Conclusions . Under the new heart allocation system, a higher percentage of recipients are non-White. In addition, racial disparities in HT outcomes have improved with Black recipients no longer having an increased risk-adjusted mortality following HT.


Introduction
Heart transplantation (HT) is a defnitive therapy for refractory heart failure, making it the gold-standard treatment for recipients with end-stage disease [1][2][3]. Minority recipients, and specifcally black recipients, have historically experienced higher mortality rates post-HT than white recipients [1,[4][5][6][7][8][9][10]. Prior research has demonstrated a center efect such that Black recipients are more likely to receive their HT care at worse performing centers, but this is likely not the only contributing factor to the mortality diference [6]. Te rates of referral and rates of the undergoing HT are traditionally lower in Black recipients than those in white recipients [5,[11][12][13][14] although listing of Black recipients for HT increased after the 2014 implementation of the Affordable Care Act Medicaid expansion [15]. An additional factor that negatively contributes to survival of Black recipients post-HT is that Black HT recipients have the highest rate of HLA antigen mismatch with their donor heart out of all racial groups [16].
Te heart allocation policy (HAP) was changed on October 18 th , 2018, in an attempt to address several issues with the preexisting policy. Changes enacted by this policy included the introduction of a 6-tier system in which the highest risk patients were prioritized and broader organ sharing for patients with the highest urgency statuses in an efort to reduce waitlist mortality and improve donor organ allocation. Te patients at highest risk as determined by this policy include those on extracorporeal membrane oxygenation (ECMO), those with a surgically implanted biventricular support device who are nondischargeable, and those with a mechanical circulatory support device (MCSD) with a concurrent life-threatening ventricular arrhythmia. Notably, patients with left ventricular assist devices (LVADs) who were able to be discharged were placed in lower risk tiers. Te policy also addressed specifc circumstances in which expedited wait list times were required. Before this change, candidates were classifed by risk based on a 3-tier system in which patients with univentricular assist devices-LVADs or RVADs-and intra-aortic balloon pumps were in the highest risk tier along with the patients on ECMO. Te HAP change allows greater prioritization of the highest risk patients by introducing expanded risk stratifcation [17]. Te purpose of this study was to evaluate the impact of the 2018 HAP change on racial disparities in HT outcomes.

Study Design.
Te United Network for Organ Sharing (UNOS) database is a prospectively maintained registry of all solid organ transplantations performed in the United States. Te UNOS database was queried for all adult (≥18 years) recipients who underwent isolated HTs between January 2010 and September 2021. Recipients that underwent HT in centers that performed less than 5 transplants per year were excluded. Recipients were stratifed by race with categories being White, Black, Hispanic, or Other. After recipients were identifed, they were stratifed into a pre-HAP group or a post-HAP group using October 18 th , 2018, as the cutof date. Tis study was deemed exempt from review by the Institutional Review Board at the Medical University of South Carolina.

Outcomes.
Te primary outcome in this study assessed all-cause post-HT mortality. Secondary outcomes included the rates of major postoperative complications (stroke, acute renal failure requiring dialysis, acute rejection, need for permanent pacemaker implantation, and post-HT length of hospital stay).  Table 1. Characteristics for post-HAP HT recipients stratifed by race are summarized in Table 2.

Survival following Isolated Heart Transplantation.
A multivariable Cox proportional hazards model for pre-HAP era total mortality following isolated HT is shown in Table 3. After risk adjustment, Black race was associated with an increased risk for all-year mortality compared to White race (HR 1.31, 95% CI 1.22-1.41, p < 0.001). A multivariable Cox proportional hazards model for post-HAP era all-year mortality following isolated HT is shown in Table 3. After risk adjustment, Black race was not associated with a signifcant increase in all-year mortality (HR 1.12, 95% CI, 0.93-1.34, p � 0.222).

Secondary Outcomes after Isolated Heart Transplantation.
Secondary outcomes after pre-HAP era HT stratifed by race are shown in Table 4, and secondary outcomes after post-HAP era HT stratifed by race are shown in Table 5. In the pre-HAP era, recipients of Hispanic race had a lower rate of pacemaker requirement post-HT (2.8% White, 2.0% Black, 0.7% Hispanic, and 2.6% Other; p � 0.004), and recipients categorized as Other sufered less acute rejection events (21.1% White, 26.0% Black, 22.1% Hispanic, and 13.5% Other; p < 0.001). Tere were no signifcant diferences in pre-HAP rates of acute renal failure requiring dialysis, stroke, or length of stay between racial groups. In the post-HAP era, recipients in the Other group had a higher rate of stroke post-HT (4.0% White, 3.1% Black, 2.8% Hispanic, and 5.5% Other; p � 0.04) but a lower rate of acute rejection events (19.8% White, 20.9% Black, 17.1% Hispanic, and    28.00 (6.24) 28. 15

Discussion
Historically, racial disparities in isolated HT outcomes have been noted for Black recipients [1,[4][5][6][7][8][9][10]. Previous literature suggests multiple potential causes of this fnding. Foremost, Black recipients are more likely to be treated at centers that have higher than average mortality rates following HT [6]. Second, Black recipients were less likely to be referred for initial evaluation and subsequent transplant, thus potentially making them higher risk at the time of transplant due to later presentation and more advanced heart failure [13]. Breathett et al. assessed incidence of transplant in early and late adopters of the Afordable Care Act (ACA) and found that the rate of HT for Black recipients increased in earlyadopting states [15]. It did not, however, narrow the gap in rates when compared to their White counterparts [15]. Last, many social determinants have been cited as potential infuences. Some of these factors include access to private health insurance, primary care preventative services, education, and Medicare or Medicaid coverage [12]. One variable to consider when evaluating mortality in a subset of recipients is the center in which their transplant occurred. High-volume centers tend to have lower mortality rates. Te centers and their staf are better prepared for complex cases and adverse events [10,18,19]. Kim et al. analyzed institutions pre-and post-HAP and found that low volume centers seem to have improved waitlist mortality and deterioration since the policy change, whereas intermediate and high-volume centers have not shown any signifcant diferences in outcomes [20]. Black recipients are more likely than their white counterparts to be transplanted at worse performing centers with higher-than-expected mortality rates. While the center status afects outcomes for minority recipients, controlling for this does not eliminate the disparity completely [6]. Te 2018 HAP change was enacted to address several issues in the preexisting US policy that had been in place since 2006. Chouairi et al. studied HT recipients from 2011 to 2020 stratifed by race and pre-and post-HAP eras. Teir analysis showed that the rates of HT increased for all groups post-HAP, but Black recipients were still less likely than White recipients to receive an HT in the post-HAP era. Trivedi et al. studied HT recipients from 1987-2020 stratifed by race and time period of transplant, although not by preand post-HAP. Teir analysis also showed that Black recipients were less likely to receive HTs than other racial groups, but they posited that post-HT survival in Black recipients has increased and is now comparable to post-HT survival in other racial groups including White recipients.
Limitations of the present study include that data are limited to the UNOS registry. In addition, we cannot capture unmeasured practice diferences between programs performing HT, although center volume which is an important center-level predictor was controlled for in multivariable Cox proportional hazards modeling [18,19]. Additionally, because data are limited to UNOS database, race and ethnicity codifcation is also limited to information found in the database. Our study used the list of races reported by UNOS that includes White, Black, Hispanic, Asian American Indian/Alaska Native, Native Hawaiian/other Pacifc Islander, multiracial, and unknown. Asian American Indian/Alaska Native, Native Hawaiian/other Pacifc Islander, and multiracial patients were classifed as "Other" in this study, and we did not have any unknowns. Another limitation of this study is that the UNOS registry does not contain information on recipient-specifc factors such as perioperative care that could potentially impact survival. We also could not assess for variables that may relate to access to care and earlier referral to the advanced heart failure specialists. Lastly, because the HAP occurred in 2018, it is possible that with longer follow-up in the post-HAP change era, outcomes will change. It is also important to note that the post-HAP period includes the beginning of the COVID-19 pandemic which could potentially afect the mortality in a nonuniform way among racial groups. More research is needed to discern the race-specifc impact of COVID-19 among patients awaiting heart transplantation.  Tis analysis of the UNOS registry determined that while pre-HAP Black recipients showed increased all-year mortality when compared with White recipients, post-HAP Black recipients did not. In addition, a higher proportion of HT recipients were non-White in the post-HAP era. While data should continue to be studied over the next several years, this analysis demonstrates that the 2018 HAP change is associated with a reduction in racial disparities in HT outcomes.

Conclusion
While continued observation is necessary, initial results suggest that the 2018 heart allocation policy change was successful in reducing racial disparities in heart transplantation outcomes. Our results show that Black patients do not face any signifcantly increased mortality as compared to White patients after this policy change.

Data Availability
Te data used in this study included all adult recipients of heart transplant between January 2010 and September 2021. Tese data were collected from the UNOS database and can be accessed here.