Racial disparities in liver transplantation for hepatitis B: To be or not to be

Several studies have documented significant racial disparities in access to and outcomes of liver transplantation in the United States. A study of United Network for Organ Sharing (UNOS) data from 1994 to 1998 by Reid et al.1 found underrepresentation of African Americans, 18 to 70 years old, on the liver recipient wait list relative to their proportion in the general US population as well as their higher age-adjusted mortality from cirrhosis. In addition, African Americans patients had more severe liver disease when they were placed on the wait list, were more likely to be removed from the list because of death or becoming too ill, and were less likely to undergo transplantation within 4 years of being placed on the wait list. African Americans patients had a shorter waiting time for orthotopic liver transplantation (OLT), which was consistent with more severe and advanced liver disease at listing. These results suggested that African Americans face barriers to referral for liver transplantation that are likely to include decreased access to care related to higher poverty rates and lower rates of private health insurance in comparison with Caucasians or whites. However, racial disparities in health status and health care quality persist for many disorders after adjustments for education level, household income, and health payer status. Disparities in post–liver transplantation outcomes have also been documented in Asian patients. Nguyen and Thuluvath2 found lower 2-year graft and patient survival rates after transplantation in Asian and African American patients with respect to Caucasians between 1988 and 1996 that differed by liver disease etiology. African Americans with hepatitis C virus (HCV), cryptogenic cirrhosis, and acute liver failure had worse outcomes than whites; in contrast, Asians with hepatitis B virus (HBV), primary biliary cirrhosis, acute liver failure, and cryptogenic cirrhosis had lower survival rates than whites.

HBIG, hepatitis B immune globulin; HBV, hepatitis B virus; HCC, hepatocellular carcinoma; HCV, hepatitis C virus; MELD, Model for End-Stage Liver Disease; OLT, orthotopic liver transplantation; UNOS, United Network for Organ Sharing.

A better understanding of the sources of disparities in access and outcomes of liver transplantation is necessary for the development of effective strategies to eliminate them and improve the health status of patients with end stage liver disease residents. A recent report by Moylan et al.3 showed that some disparities between African Americans and Caucasians in outcomes on the liver transplantation wait list before 2002 can be attributed to UNOS organ allocation policy. The authors compared the outcomes in African Americans and white patients who were placed on the liver transplantation list before (1996–2000) and after (2002–2006) implementation of the Model for End-Stage Liver Disease (MELD) liver donor allocation policy.3 Moylan et al. confirmed the disparities reported by Reid et al.1 among patients placed on the wait list in the prior MELD era (1996–2000). In contrast, the odds of dying or becoming too sick for liver transplantation and the odds of receiving liver transplantation within 3 years under MELD did not differ by race. The MELD allocation policy also corrected a disparity between Asian-Pacific Islanders and Caucasians in the probability of receiving liver transplantation for primary hepatocellular carcinoma (HCC).4 Yet, racial disparities in liver transplantation persist. African Americans with cirrhosis who were transplanted between 2002 and 2006 had a higher median MELD score, which indicates ongoing barriers to referral for liver transplantation in the MELD era. African Americans transplanted between 2002 and 2006 also continued to experience lower graft and patient survival rates than Caucasians.5 Compared to Caucasians, African American patients were more likely to undergo transplantation for HCV and acute liver failure and to have a higher MELD score at transplantation and were less likely to undergo live donor transplantation. African American patients with HCV had a significantly lower survival rate that was abolished by adjustments for the higher MELD scores and older age of African Americans. Graft survival was lower in African Americans who underwent transplantation for most indications. Still, African Americans were more likely to have recurrent disease, particularly HCV, and allograft rejection as causes for graft loss than whites, but were less likely to undergo repeat liver transplantation. These results suggest that earlier referral and better therapy to prevent and treat recurrent HCV and allograft rejection might improve the outcomes of liver transplantation for African Americans. Hispanics were also more likely to undergo liver transplantation for HCV from 2002 to 2006 but remarkably had better posttransplant patient and graft survival rates than Caucasians.5

Prior to the use of hepatitis B immune globulin (HBIG) treatment to prevent recurrent infection, both patient survival and graft survival following liver transplantation for HBV were inferior to survivals following liver transplantation for all other indications except HCC. Consequently, HBV was deemed to be a contraindication to liver transplantation at some centers. The lower posttransplant survival in HBV-positive recipients has been completely abolished over the past 2 decades by HBIG monotherapy starting in the late 1980s and early 1990s, and HBIG combined with oral nucleoside treatment in the late 1990s. Indeed, a recent study of UNOS data by Kim et al.6 found no differences in outcomes of liver transplantation for HBV between Asians and Caucasians transplanted in the United States during the era of HBIG and oral nucleoside therapy (1997–2001). African American patients had a lower survival rate and African American race was associated with lower patient survival in a proportional hazards regression model. The basis for this difference was not apparent. Less is known about racial differences in organ allocation and the outcomes of liver transplantation for HBV infections during the MELD era.

In this issue of Liver Transplantation, Bzowej et al.7 and the National Institutes of Health HBV OLT study group present the results of a combined retrospective-prospective observational study of the outcomes of HBV infection among Caucasian, Asian, and African American patients who were placed on the UNOS liver transplantation recipient wait list between 1996 and 2005 (n = 274) and who had liver transplantation performed between 2001 and 2007 (n = 170). Most patients (∼80%) in the current study received transplantation under the MELD allocation policy. Patients were followed for a mean of 22 months after listing and for 31 months after transplantation. End-stage cirrhosis was the indication for transplantation in 62%, HCC was the indication in 30%, and acute liver failure was the indication in 7%. In comparison with Asians and African Americans, end-stage liver cirrhosis was a more common indication for listing among Caucasians; HCC was more common among Asian patients, and acute liver failure was more prevalent among African Americans. The races did not differ with respect to other variables at listing or at transplantation. There were no racial differences in the probability of liver transplantation up to 5 years after listing. All patients with acute liver failure received liver transplantation within 8 days of listing, and patients with HCC were transplanted at a higher rate than patients with end-stage liver disease. Only the liver transplant indication and MELD score were associated with the probability of liver transplantation in a Cox regression model. In addition, there were no significant racial differences in HCC incidence or liver-related mortality while patients were on the wait list. Furthermore, there were no differences in posttransplant survival rates between the 3 races. Patient survival rates at 1 and 5 years ranged from 90% to 94% and 85% to 94%, respectively. Recurrent HCC was the only variable associated with posttransplant mortality in a multivariable model. Recipient race was not a significant variable in the probability of liver transplantation.

The availability of HBV serological tests, serum HBV DNA and resistance data, guidelines for HBV treatment before and after transplantation, and information on HBV recurrence after transplantation were unique features of the study. Sixty percent were receiving antiviral therapy with lamivudine and/or adefovir dipivoxil at the time of listing. The duration of treatment and the specific regimens used prior to listing are not discussed, but most treated patients appeared to have incomplete HBV DNA suppression. Serum HBV DNA was detected in 60% at listing and at the time of liver transplantation. The mean titer was 10,000 HBV DNA copies/mL at listing, and 60% had more than 10⁵ copies per milliliter at the time of transplantation. This probably explains the high HBV breakthrough and genotypic resistance rates detected before transplantation. Ninety-seven percent of the patients were treated with HBIG plus an oral agent after transplantation. Despite a high rate of active HBV replication prior to transplantation, only 8% of the patients experienced recurrent HBV infection after transplantation. This is consistent with prior studies showing lower recurrence rates in patients who receive a combination of HBIG and an oral agent instead of monotherapy.8 Consequently, recurrent HBV infection was not an important factor in patient and graft survival. For unclear reasons, recurrent HBV was significantly higher in Caucasian patients, especially at 2 and 4 years after transplantation. In a multivariable analysis, HBV recurrence was associated with hepatitis B e antigen status at listing but not with race, pretransplant serum HBV DNA concentration, transplantation after the approval of adefovir by the Food and Drug Administration, or transplant center. Other potential causes for HBV recurrence such as the use of dual therapy versus monotherapy with HBIG or oral nucleoside, variability in patient compliance with the HBIG/oral therapy, and HBIG serum dynamics were not examined. Also, HCC recurred in only 8% with no racial difference, despite the observation that more than 50% of tumors exceeded the Milan criteria at the time of transplantation. Factors associated with HCC recurrence such as the tumor stage at listing and transplantation were not analyzed.

The results are consistent with a previous study by Moylan et al.3 of patients on the UNOS liver transplant wait list who underwent transplantation according to the MELD liver donor allocation policy. MELD allocation policy selected HBV patients for liver transplantation on the basis of the severity of liver disease and the risk of liver-related mortality. Analogously to Kim et al.,6 Bzowej et al.7 also found similar survival rates in Caucasians and Asians who underwent liver transplantation for HBV between 1997 and 2001 when treatment with HBIG combined with lamivudine was available. However, the finding that African Americans had posttransplant survival similar to that of Asian and Caucasian patients contradicts Kim's results in a study of 1200 patients. This difference between the studies can probably not be explained by differences in organ allocation policy in the United States before and after 2002.5 On the other hand, only 23 African Americans patients were listed and 17 African Americans were transplanted in the study by Bzowej et al.; this sample was too small for dependable conclusions. Thus, the results concerning African American patients must be received with caution. The study has several additional shortcomings. The partial retrospective study design is a potential source of selection bias. Enrollment required informed consent from the patients. Therefore, patients with HBV who denied consent during the prospective period (2001–2006) and patients who were removed from the wait list because they became too ill or because of HCC progression or death during the retrospective period (1996–2001) were excluded. In addition, the follow-up period was relatively short; less than 50% of the patients were followed for more than 2 years either on the wait list or after transplantation. Longer term follow-up and inclusion of more African American patients will be necessary to determine whether these short-term trends in pretransplant and posttransplant outcomes, such as a new HCC diagnosis on the wait list, recurrent HBV and HCC, and survival, will be maintained.

The conclusion that “Caucasians, Asian Americans and African Americans with hepatitis B can be managed similarly in the transplant setting” seems intuitive. However, aside from the MELD allocation policy and HBV treatment guidelines, the medical management and health care that patients received were not described. The observational study design makes it difficult to identify other variables and medical management that led to excellent long-term outcomes, such as pretransplant management and the selection of patients with acute liver failure and HCC. The results indicate that Asians and Caucasians “can expect similar outcomes on the wait list as well as after transplantation,” but the OLT study group cohort includes too few African Americans to adequately address disparities between African Americans and Caucasians. More study is necessary to determine whether the disparities between African Americans and Caucasians in outcomes of liver transplantation for HBV identified in 1997–2001 persist in the most recent era.