Because of the organ shortage, liver transplantation (LT) teams have been forced to use grafts from donors that previously would have been rejected. An example is the use of grafts from donors who are hepatitis B surface antigen (HBsAg)–negative but positive for antibody to hepatitis B core antigen (anti-HBc). In the early 1990s, it was first recognized that the use of these grafts in LT was associated with a risk of transmission of hepatitis B virus (HBV) infection.1 Because of the concern about the potential detrimental effect on survival, a quite generalized opinion in the transplant community for years was that these grafts should be avoided or used only under very exceptional circumstances.2 Fortunately, substantial improvements have been made in the last years, and anti-HBc(+) grafts are no longer rejected. The prevalence of anti-HBc(+) liver donors is about 6% to 12% in most Western countries3–6 but is much higher (∼ 50%–60%) in areas in which HBV infection is endemic.7
Most improvements have resulted from the recognition of the risk factors associated with HBV reactivation in recipients of anti-HBc(+) grafts. Indeed, the wide variation in the rate of donor-related HBV infection (17%–92%) in the reported series is not only explained by methodological reasons or by differences in the prevalence of HBV infection in the general population. It is now well established that the recipient's HBV serological status at the time of LT is the most important factor determining the risk of HBV reactivation.8–10 In the absence of antiviral prophylaxis, the risk is highest (∼75%–80%) in HBV-naïve recipients, that is, those who are both antibody to hepatitis B surface antigen–negative [anti-HBs(−)] and anti-HBc(−), intermediate (∼15%–20%) in subjects who are only anti-HBc(+) or only anti-HBs(+), and lowest (∼0%–5%) in patients who are both anti-HBs(+) and anti-HBc(+). Being only anti-HBs(+) at the time of LT, as a result of either vaccination or natural immunity, decreases but does not completely eliminate the risk of HBV reactivation. This lack of complete protection may be due to insufficient anti-HBs titers at the time of LT, loss of anti-HBs titers, or a lack of complete immunological protection from humoral immunity alone. The anti-HBs titers that are considered protective after transplantation are still unclear. A recent study from Taiwan11 suggested that a pre-LT anti-HBs titer > 200 IU/mL may be sufficient to prevent HBV infection in pediatric recipients of anti-HBc(+) grafts, but this requires further confirmation. In addition to pre-LT anti-HBs titers, the evolution of post-LT serum anti-HBs levels over time may also be important. Loss of anti-HBs is not an uncommon event in this scenario, as shown by Arslan et al.12 in a series of patients who had undergone HBV vaccination before LT. In that study, the prevalence of anti-HBs before LT was 36% versus 11.6% and 8% at 1 and 2 years after LT, respectively. A patient who developed HBV reactivation following a progressive decline in postvaccination anti-HBs titers has been recently reported.13
Based on these data, strategies have been developed to allow the use of anti-HBc(+) grafts and, at the same time, minimize the risk of HBV transmission. The best approach is the selective allocation of anti-HBc(+) grafts to low-risk non-HBV recipients, as defined by their pre-LT serological status, or to patients undergoing LT for HBV-related cirrhosis that are given appropriate prophylaxis against HBV recurrence. However, this selective allocation is not always feasible, particularly in urgent cases and in areas with a high rate of HBV infection in the general population. Therefore, the need for appropriate prophylaxis is now the rule rather than the exception.
Strategies to prevent donor-transmitted HBV infection mostly derive from those used to prevent recurrent infection in patients transplanted for hepatitis B and include hepatitis B immunoglobulin (HBIG), lamivudine, and both in combination. Overall, even though many series were small and heterogeneous and had short follow-up, the reported strategies have proven to be very effective. However, these strategies are mostly empirical, and no consensus has been reached so far. As a result, there are still many unanswered questions. First, it is unclear whether all recipients of anti-HBc(+) grafts need prophylaxis. In this regard, there are enough data in the literature to support the use of the recipient's HBV profile at LT as a reliable tool to guide the need for prophylaxis. Accordingly, all naïve recipients should receive some form of prophylaxis because they have a very high risk of HBV reactivation.3, 14 However, the need for prophylaxis in other groups of patients is not as clear. Because recipients who are both anti-HBs(+) and anti-HBc(+) have a very low risk of HBV reactivation, regular monitoring of HBV markers every 6 to 12 months after LT rather than antiviral prophylaxis might be an appropriate strategy in this subgroup of patients. On the other hand, given the intermediate risk of HBV reactivation for recipients who are only anti-HBc(+) or only anti-HBs(+), closer monitoring of HBV markers (ie, every 3–6 months) could be more cost-effective than active prophylaxis, but well-designed studies should be performed in the future to better address this issue. In addition, among recipients with isolated anti-HBs(+), serial monitoring of post-LT anti-HBs titers is important. A progressive decline in these titers may be an indication for adding specific prophylaxis.
The second unanswered question is related to the type of prophylaxis. Currently, there is not a standard prophylaxis regimen for recipients of anti-HBc(+) grafts. Nucleoside analogues, in particular lamivudine, have been the drugs most frequently used in this setting.15 Lamivudine monotherapy has provided almost complete protection against HBV reactivation if the patient is compliant.5, 7, 16 It must be noted, though, that a significant proportion of the recipients who received lamivudine were anti-HBs(+) and/or anti-HBc(+) and thus had a low risk of HBV reactivation. In addition, long-term efficacy data are lacking. There have been some recent reports of lamivudine resistance in recipients of anti-HBc(+) grafts.17, 18 Therefore, it is likely that drugs with better resistance profiles will be increasingly used and eventually replace lamivudine. Actually, adefovir and entecavir are starting to be used as first-line therapy by some centers.15
HBIG has also been used, alone or in combination with lamivudine, in recipients of anti-HBc(+) grafts. The very high efficacy of HBIG when given in combination with lamivudine is not surprising given the almost complete lack of recurrence observed with this regimen in patients undergoing LT for HBV-related liver disease and the minimal viral burden of anti-HBc(+) grafts.5, 19 A recent international survey demonstrated that HBIG continues to be used in about 60% of centers, always in combination with nucleoside analogues.15 However, the amount administered, the mode of delivery (intramuscular versus intravenous), and the duration of therapy are very variable. HBIG monotherapy has been very effective in the short term,4, 20 although long-term benefits are not so clear. An increased rate of late HBV reactivation has been recently reported because of either noncompliance or the emergence of surface gene escape mutants.21, 22
Some centers have recently suggested the use of selective prophylactic strategies in which the need and type of prophylaxis are based on the presence of HBV DNA in the serum or liver of the anti-HBc(+) donor.5, 23, 24 The rationale for these approaches is that detection of donor HBV DNA in the serum or liver by polymerase chain reaction (PCR) confirms the presence of occult HBV infection25 and, therefore, the risk of HBV transmission. Assessment of donor HBV DNA in serum appears not to be helpful because it is rarely detectable by PCR in anti-HBc(+) donors known to transmit hepatitis B. In contrast, HBV DNA is detectable by PCR in the liver for about 80% of donors who have transmitted HBV to the recipient.4, 5, 21, 26 Nery et al.5 described a patient with undetectable HBV DNA in the donor liver who developed hepatitis B 14 months after discontinuation of lamivudine. Therefore, the analysis of HBV DNA in liver tissue to dictate the need for prophylaxis remains controversial. Before this strategy is adopted, further standardization of the PCR technique and of the types of liver extracts used (frozen or formalin liver tissue) is required.25 In the mean time, a more conservative and perhaps more effective strategy is to consider at the time of LT not only donor HBV DNA but also the recipient's HBV serological status.5
The ideal duration of prophylaxis in recipients of anti-HBc–positive grafts is also unclear. In studies in which HBV markers were assessed at regular intervals after LT, hepatitis B developed most often within the first 1 to 2 years, but additional cases occurred later after several years of follow-up. Loss et al.27 recently showed that among 8 lamivudine-compliant patients who were initially HBV DNA–positive in the donor liver, 7 became HBV DNA–negative by PCR in follow-up biopsies. However, it is known that loss of HBV DNA in the allograft does not equal viral eradication. In the study of Loss et al., 1 donor liver with negative HBV DNA at the time of grafting became weakly positive 2 months after LT despite lamivudine prophylaxis. Therefore, until better tools are available to identify patients in whom prophylaxis can be safely discontinued, continued antiviral prophylaxis is recommended. An alternative to long-term prophylaxis is post-LT HBV vaccination, which has been mostly explored in pediatric recipients of anti-HBc(+) grafts. The results of this strategy are not completely convincing as follow-up has been generally short and most children had vaccine-induced or natural immunity to HBV prior to LT.28–30 In addition, as with HBIG, the emergence of escape mutants is a concern.30
LT recipients who develop de novo hepatitis B rarely clear HBsAg spontaneously, and most become chronic carriers. Although generally considered a relatively benign disease, there are reports that de novo HBV infection can present as severe acute hepatitis or evolve into an aggressive form of chronic hepatitis with rapid development of cirrhosis.31, 32 Surprisingly, few series in the literature have specifically evaluated the efficacy of nucleoside analogues in the treatment of hepatitis B reactivation, with most of the reported experience being with lamivudine.21, 33 A high rate of HBsAg seroconversion has been reported when lamivudine is administered promptly after the appearance of HBsAg.21
Surprisingly, few studies to date have analyzed the impact of using anti-HBc(+) grafts on patient/graft survival in non-HBV recipients and whether antiviral prophylaxis improves outcome.3, 14 In a multicenter study performed before lamivudine was available, Dickson et al.14 reported that, although hepatitis B was benign in most cases, 4-year patient survival was significantly lower in recipients of grafts from anti-HBc(+) donors (56% versus 76%). After adjusting for recipient age, gender, diagnosis, urgency of LT, and transplantation center, the mortality rate was 2.4 times higher for recipients of anti-HBc(+) grafts. In contrast, in a study performed at our center,3 we showed that 4-year patient survival was not significantly different between recipients of anti-HBc(+) and anti-HBc(−) grafts (68% versus 76%). However, most of our patients were treated with lamivudine at a median of 205 days after the detection of HBsAg. In this issue of Liver Transplantation, Yu et al.34 evaluate the impact of using anti-HBc(+) grafts on patient/graft survival and whether outcome is related to prior exposure to HBV, as defined by the recipient's pre-LT HBV serological status. To overcome limitations due to sample size, Yu et al. analyzed data reported to the United Network for Organ Sharing (UNOS) between April 1, 1994 and January 18, 2006. The study population consisted of 35,620 adults who underwent primary LT for indications other than hepatitis B, of whom 1270 (3.5%) received an anti-HBc(+) graft.
This study showed several important findings. First, the proportion of anti-HBc(+) donors has increased in the United States from 1.8% in 1994 to 6% in 2006. This most likely reflects the fact that transplant teams have changed their attitude toward these donors, in part because of the current organ shortage but also because of the perception that their use is not associated with an increased risk of graft loss or death. In a survey conducted in 2001, only 57% of US programs would accept anti-HBc(+) grafts for HBV-naïve recipients.35 However, a similar survey conducted in 2007 showed that most centers were willing to accept anti-HBc(+) grafts for HBV-naïve recipients.15 The larger proportion of anti-HBc(+) grafts in recent years may also be a reflection of the increase in donor age. We previously showed that the prevalence of anti-HBc(+) donors increased with age and that nearly 30% of donors over 60 years were anti-HBc(+).3
Another interesting finding of the study by Yu et al.34 is that recipients of anti-HBc(+) grafts were more likely to be anti-HBc(+) whether they were anti-HBs(+) (25.5% versus 9.1%) or not (17.6% versus 8.5%) in comparison with recipients of anti-HBc(−) grafts. A reasonable explanation for this finding is that donor-recipient matching was attempted by transplant teams trying to allocate anti-HBc(+) grafts to patients with positive HBV serological markers. Despite these efforts, the matching rate was only 53.4%. Ideally, this rate should be much higher, provided that there are appropriate recipients on the waiting list. Donor-recipient matching can be quite difficult to achieve in the Model for End-Stage Liver Disease era because allocation does not take donor characteristics into account. In the study by Yu et al., the rate of matching did not significantly change in recent years; this may have resulted from factors other than the availability of appropriate recipients on the waiting list. An alternative explanation is that with the availability of very effective drugs for prophylaxis, matching is no longer viewed as necessary as before.
In the study by Yu et al.,34 recipients of anti-HBc(+) grafts had significantly worse patient survival than those who received anti-HBc(−) grafts. However, after adjustments for donor and recipient variables known to affect outcome, patient survival was not different between the 2 groups. The authors suggest that older donor age, a higher proportion of black race donors, and a higher proportion of recipients with hepatitis C virus infection and hepatocellular carcinoma in the anti-HBc(+) donor group likely accounted for the worse survival in the univariate analysis. In our study, although donor age was higher in recipients of anti-HBc(+) grafts, the severity of liver disease and the proportion of recipients with hepatitis C virus infection and hepatocellular carcinoma were similar between the 2 groups.3 In the study by Dickson et al.,14 no information was provided regarding these aspects. In addition, important predictors of post-LT survival such as donor age and ischemia time were not taken into account in the multivariate analysis.
Unfortunately, the UNOS database has significant inherent limitations. In particular, it is not possible to determine the number of recipients of anti-HBc(+) grafts who developed HBV infection and the number of deaths due to hepatitis B. The lack of difference in adjusted survival could be explained by effective prevention of HBV reactivation by donor-recipient matching or, alternatively, by the use of effective drugs not only to prevent but also to treat HBV infection. Results of a subanalysis of the study indirectly support the benefits of donor-recipient matching. Use of anti-HBc(+) grafts appears to affects recipients without anti-HBs more negatively than those with anti-HBs.
The study by Yu et al.34 provides limited data about the benefits of HBIG or lamivudine prophylaxis in recipients of anti-HBc(+) grafts. These variables have been available in the UNOS database only since 2004. Among the 374 recipients of anti-HBc(+) grafts from 2004 through January 2006, only 92 (24.5%) received either HBIG or lamivudine. In addition, their use was not associated with the recipient anti-HBc/anti-HBs status. Another important limitation of the UNOS database is that it does not identify whether lamivudine was used for prophylaxis or to treat established hepatitis B. Nevertheless, the fact that only 24% of recipients of anti-HBc(+) grafts received HBIG or lamivudine is quite surprising. Because about 47% of recipients were HBV-naïve and this proportion did not change over time, it is clear that a significant proportion of the naïve recipients in this series were not given any prophylaxis. In addition, the fact that the use of HBIG or lamivudine was not associated with the recipient anti-HBc/anti-HBs status is likewise surprising but further underlines the lack of consensus regarding which patients should receive prophylaxis.
Finally, Yu et al.34 showed that use of anti-HBc(+) grafts had no apparent impact on the survival of patients transplanted for hepatitis B. Although this is in agreement with data reported by Joya-Vazquez et al.,36 in that study, HBV recurrence was 2.5 times more frequent and the time to reinfection was significantly shorter (2.9 versus 6.4 years) in the anti-HBc(+) donor group. These surprising findings could not be assessed in the series by Yu et al. and therefore need to be confirmed in further studies.
In summary, despite a number of limitations, the key message of the study by Yu et al.34 is that recipients of anti-HBc(+) grafts have survival rates similar to those of recipients of anti-HBc(−) grafts when adjustments are made for other predictors of post-LT survival. A second important message is that these grafts may be particularly safe to use for recipients who are anti-HBs(+) before LT. This is encouraging news for both transplant physicians and their patients in the current era of critical organ shortage. The challenge of how to manage recipients of anti-HBc(+) grafts is certainly not over, and areas of uncertainty persist. However, in the end, there are now enough reasons to conclude that anti-HBc(+) grafts should no longer be considered marginal organs.