Begin typing your search above and press return to search.
Volume: 16 Issue: 6 December 2018

FULL TEXT

ARTICLE
Improvements in Outcomes for Ethnic Minorities During the Share 35 Era Are Not Due to Decreased Rates of Early Graft Loss

Objectives: Our aim was to investigate the effects of the Share 35 policy on outcomes in ethnic minorities and recipients who experienced early graft failure.

Materials and Methods: We analyzed donor and recipient data from the United Network for Organ Sharing database before (June 6, 2011 to June 18, 2013) and after (June 18, 2013 to June 30, 2015) implementation of Share 35. Graft and patient survival outcomes were compared.

Results: There were significant differences in 1- and 2-year graft and patient survival rates between ethnicities pre-Share 35 (P = .03, P < .001, P = .01, P < .001, respectively). There were no significant differences in 1- and 2-year graft and patient survival between ethnicities post-Share 35 (P = .268, P = .09, P = .343, P = .087, respectively). There were no differences in early graft failure rates pre- and post-Share 35 at 7 days (2.1% vs 2.0; P = .71) and 30 days (4.0% vs 3.8%; P = .47) after transplant, with a decreased early graft failure rate shown at 90 days after transplant (6.8% vs 5.8%; P = .003). When analyzed separately, the low Model for End-Stage Liver Disease (score of < 35) and the high Model for End-Stage Liver Disease recipients (score of ≥ 35) both exhibited reduced early graft failure rates post-Share 35 (6.1% vs 5.3% and 10.8% vs 7.8%, respectively; P < .05).

Conclusions: Share 35 was associated with a short-term reduction in ethnic disparities. Most ethnic groups experienced improved survival in the Share 35 era. Share 35 was not associated with an increase in early graft failure and is an efficacious policy with regard to short-term outcomes.


Key words : Ethnic, Liver, Organ transplant wait list

Introduction

Share 35 was initiated by the United Network for Organ Sharing (UNOS) on June 18, 2013, due to pressure to reduce wait list mortality and geographic disparities. Under Share 35, deceased donor livers are shared regionally with candidates who have a Model for End-stage Liver Disease (MELD) score of > 35 before they are offered to local candidates with MELD scores of < 35.1

The effects of this policy on wait list mortality have been well-studied.1-3 Rates have decreased across all MELD scores by 8% and by 30% in candidates with MELD scores of > 30.1 Furthermore, most reprioritized candidates (those with MELD < 35) were not disadvantaged.4 Despite these positive outcomes, certain types of recipients have been understudied in the Share 35 era. There are no reports regarding how Share 35 affects outcomes in ethnic minorities and recipients with early graft failure (EGF).

Thus, our aim was to investigate whether the implementation of the Share 35 policy resulted in any improvements in outcomes in these cohorts. Using data from the Scientific Registry for Transplant Recipients, we evaluated and compared patient and graft outcomes 2 years before the implementation of Share 35 (June 6, 2011 to June18, 2013) versus 2 years after (June 18, 2013 to June 30, 2015). We hypothesized that the implementation of Share 35 would be associated with improvements in outcomes in these understudied recipients.

Materials and Methods

Data source, study population, and variables
Data containing liver transplant donor and recipient (n = 19 719) information from June 18, 2011 through June 18, 2015 were extracted from the UNOS database. Transplants performed before June 15, 2013 (n = 9482) were combined into the pre-Share 35 cohort, whereas the remaining cases (n = 10 237) were included in the post-Share 35 cohort. Recipients of living-donor organs and simultaneous kidney, pancreas, intestine, lung, or heart transplants were excluded. Details regarding the recipient and donor demographics were queried. Recipient characteristics obtained included age, sex, race, body mass index, transplant indication, MELD, wait list time, dialysis, status (intensive care unit vs hospitalized vs home), and mechanical ventilatory needs at time of transplant. Donor characteristics included age, sex, race, body mass index, creatinine, donor risk index (DRI), split liver vs whole liver, percent micro- and macrosteatosis in donor liver, bilirubin, donation after cardiac death, and expanded criteria donor. Additional perioperative details were queried, including human leukocyte antigen (HLA) mismatch level, organ share status (local vs regional vs national), warm ischemia time, and cold ischemia time.

Statistical analyses
Descriptive statistics of clinical and demographic characteristics were summarized using one-way analysis of variance and t test for continuous variables and chi-square test for categorical variables. For continuous variables, compliance with the normality assumption was tested using Shapiro-Wilks diagnostic test, and Kruskal-Wallis rank test was performed when the normality assumption was violated. Survival curves were obtained using the Kaplan-Meier product limit method.

The rate of EGF due to any cause was measured at 7, 30, and 90 days posttransplant. Causes of EGF were further analyzed and classified into primary graft dysfunction/nonfunction, vascular thrombosis (hepatic artery, portal vein), acute and chronic rejection, viral hepatitis infection, biliary complications, patient death, or others. For the ethnicity analyses, patient death and graft failure were measured at 3 months, 1 year, and 2 years posttransplant.

In the survival analysis, patient death and graft failure during the respective outcome assessment period were the primary endpoints. Patients who did not experience any of the endpoints were censored on June 18, 2011 for the pre-Share 35 cohort and on June 30, 2015 for the post-Share 35 cohort.

Multivariate Cox regression analyses were conducted to investigate adjusted risk factors. These included donor and recipient demographics, peri- and postoperative factors, factors linked to organ quality, and organ share status.

P values of less than .05 were considered signi-ficant. Statistical analysis was performed using STATA (version 13, StataCorp, College Station, TX, USA).

Results

Ethnic minorities
Pre- and post-Share 35 recipient demographic data are shown in Table 1. Pre- and post-Share 35 donor characteristics are listed in Table 2. Notably, there was a significant increase in the number of donations from donors after cardiac death from pre- to post-Share 35 (456 vs 634; P < .005) (Table 2). Furthermore, the number of organs shared regionally significantly increased from pre- to post-Share 35 (1887 vs 3146; P < .005) (Table 2).

Regarding graft survival, there were no significant differences in 3-month survival between ethnicities in the pre- or post-Share 35 groups (P = .416 and P = .812, respectively) (Table 3). There were significant differences in 1-year graft survival between ethnicities pre-Share 35 (P = .03), but no significant differences post-Share 35 (P = .268)

(Table 3). There were significant differences in 2-year graft survival rates between ethnicities pre-Share 35 (P < .001), but no significant differences in 2-year graft survival between ethnicities post-Share 35 (P = .09) (Table 3).

No significant differences were seen among the different ethnic groups for 3-month patient survival rates in both the pre- and post-Share 35 cohorts (P = .662 and P = .732, respectively) (Table 4). Regarding 1-year patient survival, rates among the ethnic groups were significantly different pre-Share 35 (P = .01) but not significantly different post-Share 35 (P = .343) (Table 4). Similarly, 2-year patient survival rates between ethnicities were significantly different pre-Share 35 (P < .001) but not post-Share 35 (P = .087) (Table 4).

Early graft failure
Table 5 summarizes the significant differences in demographics of the EGF recipients pre- and post-Share 35. Overall, no significant differences were shown between the 2 cohorts regarding sex, race, body mass index, and transplant indication. Significant donor characteristics can be found in Table 6. Notably, the average DRI was increased post-Share 35 (1.76 to 1.78; P < .001).

We found no differences in EGF rates at the 2 earlier time points pre- and post-Share 35 (2.1% vs 2.0% at 7 days, P = .71; 4.0% vs 3.8% at 30 days, P = .47; Table 7). At 90 days after liver transplant, EGF rates were noted to be decreased in the post-Share 35 era compared with those shown pre-Share 35 (5.8% vs 6.8%; P = .003).
Major causes of EGF included primary nonfunction, vascular thrombosis (ie, hepatic artery, portal vein), acute and chronic rejection, biliary complications, viral hepatitis, and other unspecified causes. Further analysis of our cohorts revealed an overall increase in rates of primary graft nonfunction post-Share 35 (0.04% vs 0.33%; P < .001) and a decrease in rates of vascular thrombosis (0.76% vs 0.43%; P = .003) (Table 7).

Early graft failure rates pre- and post-Share 35 were compared separately in the low MELD (< 35) and the high MELD (≥ 35) cohorts (Table 7). Before the Share 35 policy, 6.1% of the low MELD group experienced EGF at 90 days after liver transplant compared with 5.3% after implementation of Share 35 (P = .02). Similarly, the EGF rate was also decreased when we compared the high MELD recipients at pre- versus post-Share 35 (10.8% vs 7.8%; P = .003). Figures 1A and 1B show the Kaplan-Meier analyses of graft survival before and after implementation of Share 35 in the low and high MELD recipients.

When the analysis was separated based on recipient MELD scores, we found that mortality 90 days after liver transplant was decreased for patients with MELD ≥ 35 from 8.12% in the pre-Share 35 era to 6.34% in the post-Share 35 era (P = .045, Table 7), although the cumulative survival rates were found to be no different between the 2 periods using Kaplan-Meier analysis (Figure 1, C and D).

Discussion

Ethnic disparities have plagued liver transplantation along with other types of solid-organ transplantation for many years.5-9,10 Historically, African American (AA) recipients have shown inferior graft and patient survival rates compared with White, Hispanic, Asian, and other recipients both before and during the MELD era.5-7,11 Many have tried to elucidate reasons why AA patients typically experience inferior graft and patient outcomes after liver transplant. It has been propounded that AA patients are more likely to be more seriously ill at the time of liver transplant, are more likely to experience rejection and infection, and are more likely to have public insurance.5,6,8,12 Reports vary regarding outcomes in White and Hispanic recipients. Some have demonstrated that White recipients experience survival advantages over Hispanic recipients,5,9 whereas others have reported that Hispanic recipients experience superior outcomes compared with White recipients.6,10 Data are scarce regarding biologic, disease-related, and socioeconomic factors that are associated with survival rates in Hispanic patients after liver transplant. However, analyses from Hispanic recipients of kidney transplants have suggested that, despite higher rates of diabetes, lower average incomes and lower average levels of education, Hispanic patients tend to experience superior outcomes versus White patients.13 This so-called “Hispanic paradox” has been attributed to possible genetic advantages, better social support, or the “salmon bias theory,” which suggests that older Hispanic recipients return to their birthplace at the end of life and are therefore not counted in studies of mortality.13 Finally, Asian recipients typically experience the best outcomes after liver transplant, displaying superior graft and patient survival rates compared with White, Hispanic, AA, and other ethnic groups.7,10 It has been proposed that lower average body mass index and levels of access to liver transplant that are above their proportional representation on wait lists are among the reasons for improved outcomes in Asian populations.10

Our results for 1-year and 2-year graft and patient survival rates before the implementation of Share 35 correlate with the above-mentioned findings. Our data showed significant differences in graft and patient survival rates between ethnicities, with Asian patients generally experiencing the best outcomes followed by Whites and Hispanics, with AA patients having the worst outcomes. However, after the implementation of Share 35, the expected ethnic disparities in liver transplant outcomes were reduced. There were no significant differences in either 1- or 2-year graft and in patient survival rates between ethnicities. Overall, our analysis demonstrated that, after the implementation of Share 35, previously observed ethnic disparities were eliminated, suggesting that, at least in the short-term, the policy has unintentionally but successfully helped equalize outcomes between ethnicities.

Our data also suggested that the reduction in overall ethnic disparities was not due to one ethnic group experiencing better graft and patient survival rates post-Share 35 at the expense of another. Regarding graft survival, all ethnicities showed increases in rates of survival from pre- to post-Share 35 at both 1 and 2 years except Asian recipients who experienced a decrease of less than 1%. Furthermore, these increases were relatively large. Graft survival rates 2 years after implementation of Share 35 were 5% to 10% higher in all ethnicities. A similar trend was observed in patient survival with all groups except recipients categorized as other, who demonstrated either an increase or no change in survival rates from pre- to post-Share 35 at 1 and 2 years. Similar to the trend seen in graft survival rates, patient survival rates 2 years after implementation of Share 35 were as much as 10% higher than pre-Share 35 rates. These results correlate with those of Chow and associates4 since they suggest that those candidates who are reprioritized under Share 35 are not disadvantaged.

One possible explanation for these results is that recipients who were historically sicker at the time of liver transplant are now able to receive organs earlier since they have priority over local candidates with better MELD scores. It is possible that Share 35 has led to improved outcomes in AA recipients since it has been demonstrated that AA patients tend to be more seriously ill at the time of liver transplant. Additionally, it is possible that the transplant community in the Share 35 era is better equipped to handle the care of ethnic minorities.

This analysis led us to closely examine EGF, which is an infrequent but potentially devastating outcome after liver transplant as it can result in a prolonged intensive care unit course or hospital stay, the need for retransplant, and increased post-transplant mortality. Although there is no consensus on the definition of EGF, it is most commonly described as the loss of allograft function (synthetic and metabolic) within 90 days of transplant from any cause, including primary graft dysfunction/nonfunction, vascular thrombosis (portal vein, hepatic artery), acute and chronic rejection, biliary complications, or viral hepatitis. Because of previous suggestions that Share 35 led to an unintended decrease in organ discard rate, we questioned whether this may imply an increased use of suboptimal donor organs for transplant procedures that in turn cause an increase in EGF. In addition, we wanted to assess whether the allocation of “better” organs to ethnic minorities (who tend to be more ill at time of transplant) could explain the improvements in ethnic disparities. We specifically wanted to determine whether EGF rates would increase in the low MELD (< 35) cohort and decrease in the high MELD (> 35) cohort after Share 35.

Our analysis revealed that the average DRI was increased from 1.76 to 1.78 after implementation of Share 35 (P < .001); however, due to the large sample size, this difference may not be clinically significant. Despite this increase in DRI, and in accordance with other reports showing overall unchanged outcomes after liver transplant in the Share 35 era,1,14 the results of our study indicated that the prevalence of EGF was largely unaffected or slightly improved by the Share 35 allocation policy. The decrease in EGF was seen in both low and high MELD groups in the Share 35 era. Thus, our data do not suggest that superior donor organs are being preferentially offered to recipients with high MELD scores (> 35). Furthermore, the improvements in ethnic disparities were not related to a decrease in EGF in ethnic minorities. Taken together, our findings suggested that perhaps before implementation of Share 35 many usable donor organs were unnecessarily discarded.

Analysis of causes of EGF revealed a significant increase in those due to secondary to primary nonfunction. Risk factors for primary nonfunction include donor age, steatosis, and warm and cold ischemic time.15 However, we found these factors not to be significantly different in our study cohorts pre- and post-Share 35 to explain the increased rates of primary nonfunction. In fact, we are wary that our analysis may be flawed by reporting bias inherent to the nature of this type of study. We also note that the rates of vascular thrombosis, biliary complication, viral hepatitis, and recurrent disease were signi-ficantly decreased after implementation of Share 35, thus together contributing to the decreased overall EGF rate after the policy change.

The main strengths of our analysis included the relatively large sample size from the nationwide UNOS database, which effectively captures outcomes from pre- and post-Share 35. At the same time, our study period was not so long such that surgical techniques and principles of perioperative manage-ment had changed drastically to bias the results. To our knowledge, this is the first analysis to investigate outcomes in understudied groups in the pre-Share 35 era. One limitation of the study is its retrospective nature.

The overall goal of the Share 35 allocation policy was to improve allocation of deceased donor allografts and reduce wait list mortality.1 Our analysis suggests that Share 35 could have unintentionally, yet effectively, reduced short-term ethnic disparities in liver transplant outcomes. Simultaneously, there was a slight decrease in the prevalence of EGF after 90 days, suggesting that Share 35 has not resulted in an increased use of suboptimal allografts. These results seem to confirm an equalization of outcomes among ethnicities not caused by worse outcomes in another group or by a higher rate of EGF in another group. Although these results are promising, future studies will be required to determine whether these trends continue at 5 and 10 years after liver transplant.


References:

  1. Massie AB, Chow EK, Wickliffe CE, et al. Early changes in liver distribution following implementation of Share 35. Am J Transplant. 2015;15(3):659-667.
    CrossRef - PubMed
  2. Freeman RB, Jr. Survival benefit: quality versus quantity and trade-offs in developing new renal allocation systems. Am J Transplant. 2007;7(5):1043-1046.
    CrossRef - PubMed
  3. Feng S, O'Grady J. Share 35: a liver in time saves lives? Am J Transplant. 2015;15(3):581-582.
    CrossRef - PubMed
  4. Chow EK, Massie AB, Luo X, et al. Waitlist Outcomes of Liver Transplant Candidates Who Were Reprioritized Under Share 35. Am J Transplant. 2017;17(2):512-518.
    CrossRef - PubMed
  5. Nair S, Eustace J, Thuluvath PJ. Effect of race on outcome of orthotopic liver transplantation: a cohort study. Lancet. 2002;359(9303):287-293.
    CrossRef - PubMed
  6. Ananthakrishnan AN, Saeian K. Racial differences in liver transplantation outcomes in the MELD era. Am J Gastroenterol. 2008;103(4):901-910.
    CrossRef - PubMed
  7. Thuluvath PJ, Guidinger MK, Fung JJ, Johnson LB, Rayhill SC, Pelletier SJ. Liver transplantation in the United States, 1999-2008. Am J Transplant. 2010;10(4 Pt 2):1003-1019.
    CrossRef - PubMed
  8. Mathur AK, Schaubel DE, Zhang H, Guidinger MK, Merion RM. Disparities in liver transplantation: the association between donor quality and recipient race/ethnicity and sex. Transplantation. 2014;97(8):862-869.
    CrossRef - PubMed
  9. Quillin RC, 3rd, Wilson GC, Wima K, et al. Independent effect of black recipient race on short-term outcomes after liver transplantation. Surgery. 2015;157(4):774-784.
    CrossRef - PubMed
  10. Fan PY, Ashby VB, Fuller DS, et al. Access and outcomes among minority transplant patients, 1999-2008, with a focus on determinants of kidney graft survival. Am J Transplant. 2010;10(4 Pt 2):1090-1107.
    CrossRef - PubMed
  11. Neff GW, Kemmer N, Kaiser T, Zacharias V, Majoras N, Safdar K. Outcomes in adult and pediatric liver transplantation among various ethnic groups. Transplant Proc. 2007;39(10):3204-3206.
    CrossRef - PubMed
  12. Yoo HY, Thuluvath PJ. Outcome of liver transplantation in adult recipients: influence of neighborhood income, education, and insurance. Liver Transpl. 2004;10(2):235-243.
    CrossRef - PubMed
  13. Fischer MJ, Go AS, Lora CM, et al. CKD in Hispanics: Baseline characteristics from the CRIC (Chronic Renal Insufficiency Cohort) and Hispanic-CRIC Studies. Am J Kidney Dis. 2011;58(2):214-227.
    CrossRef - PubMed
  14. Edwards EB, Harper AM, Hirose R, Mulligan DC. The impact of broader regional sharing of livers: 2-year results of "Share 35". Liver Transpl. 2016;22(4):399-409.
    CrossRef - PubMed
  15. Chen XB, Xu MQ. Primary graft dysfunction after liver transplantation. Hepatobiliary Pancreat Dis Int. 2014;13(2):125-137.
    CrossRef - PubMed


Volume : 16
Issue : 6
Pages : 714 - 720
DOI : 10.6002/ect.2017.0047


PDF VIEW [156] KB.

From the 1College of Medicine and Life Sciences, University of Toledo, Toledo, Ohio, USA; the 2Maricopa Medical Center, Phoenix, Arizona; the 3School of Policy, Government, and International Affairs, George Mason University, Arlington, Virginia, USA
Acknowledgements: J. Brooks, T. Wong, and A. DeLeonibus wrote the manuscript; N. Koizumi provided statistical analysis and review; E. Neglia provided statistical analysis; J. Ortiz provided study design and review. The authors declare no conflicts of interest and have no funding to disclose.
Corresponding author: Joseph Brooks, 3000 Arlington Ave., Toledo, Ohio 43614, USA
Phone: +1 513 600 3049
E-mail: Joseph.brooks@rockets.utoledo.edu