Outcomes in liver transplant recipients with hepatitis B virus: Resistance and recurrence patterns from a large transplant center over the last decade



Hepatitis B virus (HBV) recurrence following liver transplantation (LTx) has been controllable primarily with the use of hepatitis B immune globulin (HBIg) and lamivudine (LAM). However, HBV resistance to LAM and/or HBIg has become an increasing problem prompting the use of newer antiviral agents. The purpose of our study was to investigate the association between therapy, HBV breakthrough, and allograft / patient survival in HBV-positive liver transplant recipients. We performed a retrospective review of the medical records of patients that were transplanted for HBV from June 1994 to May 2003. A total of 92 patients, positive for either hepatitis B surface antigen (HBsAg) or HBV deoxyribonucleic acid (DNA) pretransplant, received LAM monotherapy or HBIg (6 months) plus LAM therapy post–liver transplant. HBV breakthrough post-LTx was noted in 14 patients. All patients had detectable HBV DNA prior to liver transplantation; none of the patients that were HBV DNA negative prior to transplant had detectable HBV DNA posttransplant. Of these 14, 9 patients (64%) were switched from LAM to adefovir dipivoxil (ADF) and 5 patients (36%) to tenofovir disoproxil fumarate (TNV). In conclusion, pre-LTx HBV viremia should be considered in planning post-LTx prophylaxis. Trials to evaluate oral antiviral agents in combination with or without HBIg therapy are needed. (Liver Transpl 2004;10:1372–1378.)

The introduction of protocols that include hepatitis B immunoglobulin (HBIg) and / or lamivudine (LAM) to prevent hepatitis B virus (HBV) recurrence post–liver transplantation (LTx) has improved the survival of patients transplanted for HBV-related liver disease.1–4 However, posttransplant prophylaxis for prevention and / or suppression to prevent the recurrence of HBV post–liver transplant has not been standardized and tends to vary widely from 1 transplant center to another.5–11

Some investigators have suggested that a combination between HBIg and 1 or more nucleoside or nucleotide analogs may be the best prophylactic therapy for the prevention of recurrent HBV after LTx.8 HBIg, however, is very expensive and can be difficult to obtain due to limited availability. In addition, it is not entirely devoid of side effects. Because of these considerations, optimal duration of treatment remains controversial.8, 12 Cost-effectiveness models for optimizing HBIg utilization have yet to be developed. Moreover, the development of HBV viral resistance to LAM or to HBIg therapy as a result of inadequate maintenance of HBV surface antibody levels has become increasingly common.

New treatment options such as tenofovir disoproxil fumarate (TNV) and adefovir dipivoxil (ADF) have provided enhanced prevention of recurrent HBV post–liver transplant. In view of these new nucleoside / nucleotide antiviral agents, we undertook a review of results using these novel agents to determine the frequency of HBV resistance and the treatment outcomes of patients experiencing HBV breakthrough due to LAM and or HBIg resistance post–liver transplant.


HBV, hepatitis B virus; LTx, liver transplantation; HBIg, hepatitis B immunoglobulin; LAM, lamivudine; ADF, adefovir dipivoxil; TNV, tenofovir disoproxil fumarate; HBsAg, hepatitis B surface antigen; DNA, deoxyribonucleic acid.

Patients and Methods

We reviewed the medical records of patients positive for either hepatitis B surface antigen (HBsAg) and / or HBV deoxyribonucleic acid (DNA) who had received a liver transplant at the University of Miami / Jackson Memorial Hospital's Liver Transplant Center from June 1994 to May 2003. Patient demographics, information pertaining to the pre- and posttransplant phases, and survival data were collected. Extracted data included pre- and post-LTx HBV prophylaxis, HBV therapy resistance, and viral recurrence or breakthrough.

Hepatitis B Assays

Pretransplant testing for all recipients included HBsAg levels, HBsAb levels, and HBV core antibody levels (enzyme immunoassay; Abbott Laboratories, Abbott Park, IL). If the HBsAg was positive, hepatitis B e antigen and hepatitis B e antibody testing were performed, as was a qualitative HBV DNA by polymerase chain reaction (Sharp signal; Digene, Gaithersburg, MD). Patients qualitatively positive for serum HBV DNA were tested for quantitative HBV DNA titers by DNA hybridization (Hybrid capture, Digene; sensitivity: .5 pg/mL, or 141,500 copies/mL). Posttransplantation serologic testing was performed every 1 to 6 months.

Treatment Resistance

LAM resistance was defined as qualitatively negative HBV DNA on therapy with subsequent HBV DNA positivity associated with an increase in serum alanine aminotransferase to >2 times normal. Criteria for HBV recurrence were based on the presentation of abnormal liver function tests combined with the presence of either a positive HBV DNA level or positive immunoperoxidase stain for HBsAg, and / or the detection of DNA by PCR.

Study Groups

The liver transplant recipients who received HBV antiviral prophylaxis post–liver transplant were divided into 4 subgroups according to their pretransplant viral replication status in combination with their posttransplant antiviral therapy.

  • Group I HBV DNA − pre-LTx (LAM therapy)

  • Group II HBV DNA − pre-LTx (HBIg and LAM therapy)

  • Group III HBV DNA+ pre-LTx (LAM therapy)

  • Group IV HBV DNA+ pre-LTx (HBIg and LAM therapy)

All patients were HBsAg+ prior to transplant with the exception of 2 patients in Group IV who were HBsAg– but HBV DNA+.

Immunosuppression Protocol

Post–liver transplant immune suppression protocol at our center consisted of combined tacrolimus and methylprednisolone therapy. Tacrolimus was administered at the time of LTx in a dose of 0.05 mg/kg orally and continued postoperatively with dose adjustments to maintain whole blood 12-hour trough levels between 15 and 20 ng/mL for the 1st postoperative week, then 10 ng/mL for the 1st 3 months, followed by 8 ng/mL thereafter for the 1st year of therapy. Methylprednisolone was initiated with a 1-gm intravenous bolus immediately after the reperfusion of the hepatic graft and then tapered to 200 mg/day (day 1), 160 mg/day (day 2), 120 mg/day (day 3), 80 mg/day (day 4), 40 mg/day (day 5), 20 mg/day (day 6), and then reduced to zero within 3–6 months post-LTx. Patients were observed for occurrence of acute rejection, evidence of allograft dysfunction, and the occurrence of opportunistic infection or malignancies.

HBIg and LAM Dosing

Standard protocol for prevention of posttransplant hepatitis B consisted of 10,000 units of HBIg intravenously intraoperatively daily for 7 days, biweekly for 1 month, and then monthly until month 6. Serum HBV antibody levels during HBIg therapy were not routinely collected unless the transplant recipient demonstrated an elevation in their alanine aminotransferase level or other evidence of HBV breakthrough. LAM was initiated between 1 and 34 months (median: 2 months) at an oral dose of 150 mg/day prior to transplantation, and continued indefinitely.

Novel Antiviral Therapies for Hepatitis B

Treatment protocols for HBV therapy resistance included standard therapy with LAM, as well as the newer antiviral therapy, including TNV, ADF, lobucavir, and famciclovir. Patients initiating TNV, ADF, lobucavir, or famciclovir were placed into study protocols relevant to that therapy. Therefore, LAM was discontinued if patients were started on ADF, as per the ADF study protocol, while if TNV was initiated, LAM was continued as combination therapy.

Statistical Analysis

The database of all patients that qualified for the study within the allotted time period was collected using Microsoft Excel (Microsoft, Redmond, WA) and formed the basis for the final analyses. Data were analyzed by STATISTICA program (StatSoft, Tulsa, OK). Analysis of variance was used to compare means among treatment groups and to compare results pre- and posttreatment. In all cases, 2-tailed P values < .05 were considered significant. The Institutional Review Board of the University of Miami approved this study.



During the period from June 1994 to May 2003, 92 patients with HBsAg (HBsAg+) received antiviral therapy to prevent recurrence of hepatitis B post–liver transplant. Of these 92 patients, 79 (86%) were male and 13 (14%) were female. Detailed patient demographics and time from LTx are displayed in Table 1. The average donor age was 50.5 years (range 5-70 years).

Table 1. Patient Characteristics
Subgroups*NGender‡Age in years mean (range)Ethnicity‡Follow-up months median (range)Patient survival n (%)Graft survival n (%)HBV breakthrough posttransplant n (median months)
  • Abbreviations: M, male; F, female; C, Caucasian; AA, African American; A, Asian; H, Hispanic; O, Other; NS, not significant.

  • *

    Groups of patients were characterized as follows: I, HBV DNA-negative pre-LTx (LAM therapy); II, HBV DNA-negative pre-LTx (HBIg and LAM therapy); III, HBV DNA-positive pre-LTx (LAM therapy); IV, HBV DNA-positive pre-LTx (HBIg and LAM therapy).

  • N is number of patients in each subgroup.

I33M = 2950 (22–60)C = 946 (1–102)27 (82)27/38 (71)0
  F = 4 AA = 2    
    A = 7    
    H = 14    
    O = 1    
II18M = 1550 (31–69)C = 870 (5–104)14 (78)14/21 (67)0
  F = 3 AA = 0    
    A = 5    
    H = 3    
    O = 2    
III18M = 1657 (44–68)C = 664 (13–88)17 (94)17/19 (89)9 (12)
  F = 2 AA = 1    
    A = 1    
    H = 5    
    O = 5    
IV23M = 1960 (34–66)C = 928 (1–82)19 (83)19/25 (76)5 (20)
  F = 4 AA = 1    
    A = 4    
    H = 7    
    O = 2    
P value     NS, log-rank testNS, logistic regression0.10, Fisher's exact test

A total of 43 transplant recipients had HBV as the sole infection. Other reasons for LTx included: hepatocellular carcinoma (n = 18), hepatitis C (n = 8), alcoholic liver disease (n = 4), combined hepatitis C and alcohol abuse (n = 3), combined hepatitis D and hepatocellular carcinoma (n = 3), cholangiocarcinoma (n = 1), hepatitis D (n = 3), combined hepatitis C and hepatitis D (n = 3), hepatitis C and human immunodeficiency virus (n = 2), autoimmune hepatitis (n = 2), hemochromatosis (n = 1), and Budd Chiari syndrome (n = 1).

Patients were categorized by their pre–liver transplant HBV status and their posttransplant antiviral treatment as follows:

  • Group I (n = 33) HBV DNA– pre-LTx and received LAM alone.

  • Group II (n = 18) HBV DNA– pre-LTx and received HBIg + LAM.

  • Group III (n = 18) HBV DNA+ pre-LTx and received LAM alone.

  • Group IV (n = 23) HBV DNA+ pre-LTx and received HBIg + LAM.

The median length of both patient and graft follow-up post–liver transplant was 42 months (range 1-92 months).

Viral Recurrence

Groups I and II (HBV DNA–)

No patient within Groups I or II experienced evidence of HBV viral recurrence post–liver transplant.

Groups III and IV (HBV DNA+)

In contrast, 14 / 41 (34%) patients in Groups III and IV experienced HBV recurrence despite antiviral therapy post-LTx. The median time for detection of abnormal alanine aminotransferase and HBV recurrence was 15.5 months (range 3-30 months). Nine of 18 patients (50%) in Group III that had been treated with LAM alone had HBV viral recurrence, while Group IV (treated with a combination of HBIg and LAM) had 5 / 23 (22%) patients develop detectable HBV DNA. The actuarial rates of recurrence at 1 and 3 years for Group III were 17% and 50%, respectively, and for Group IV the rates were 13% and 22%, respectively (P = .0004, log-rank test). Of the 9 patients with breakthrough in Group III, 5 were placed on ADF and 4 on TNV. For the breakthrough patients in Group IV, 1 was given TNV and 4 received ADF.

One patient presented with LAM resistance and HBV viremia (6 million copies/mL) at the time of transplant and was started on HBIg in combination with TNV posttransplant day 1 and was nondetectable for HBV by week 12. This patient was coinfected with HBV, HCV, and human immunodeficiency virus and had received LAM for several months prior to transplant for his HBV and human immunodeficiency virus. Tables 2 and 3 display additional information regarding patients in Groups III and IV who experienced HBV viral breakthrough, both at the time of the event and after follow-up on treatment.

Table 2. Characteristics of Patients With Hepatitis B Breakthrough at Time of Event
CaseGroupGenderAgeRaceComorbidityTime to BT (months)HBV DNA* (mEq/mL)HBsAgHBeAganti-HBeALT (U/L)RxHBV DNA neg (months)§
  • Abbreviations: III, HBV DNA (+) Pre-LTx (LAM therapy); IV, HBV DNA (+) Pre-LTx (HBIg and LAM therapy); M, male; F, female; G, Greek; AA, African American; C, Caucasian; H, Hispanic; A, Asian; ALD, alcoholic liver disease; HCV, hepatitis C virus; HIV, human immunodeficiency virus-1; BT, breakthrough of lamivudine-resistant HBV.

  • *

    Results are not comparable. DNA was measured by both branch-DNA and polymerase chain reaction techniques. Mil denotes 106.

  • HBV surface antigen (HBsAg) and e antigen (HBeAg) and antibody (anti-HBe) are denoted as positive (+) or negative (−).

  • Rxs adjunct treatment of adefovir (ADF) or tenofavir (TNV).

  • §

    Time in months to attain undetectable HBV DNA.

  • **

    Thereafter positive.

3IIIM68CNone18860 Mil++45ADF13
6IIIM60GHCC259.6 Mil++502TNV5
10IVM55HNone1715 Mil++112ADF36
13IVM47CNone218.2 Mil++483ADF13**
Table 3. Characteristics of Patients With Hepatitis B Breakthrough After Follow-up on Treatment
CaseGroup*RxTime on Rx (months)HBV DNA (mEq/mL)anti-HBs§HBsAg§HBeAg§anti-HBe§ALT (U/L)
  • *

    III, HBV DNA (+) Pre-LTx (LAM therapy); IV, HBV DNA (+) Pre-LTx (HBIg and LAM therapy).

  • Rxs adjunct treatment of adefovir (ADF) or tenofavir (TNV).

  • HBV DNA measurements after follow-up on treatment. Results are not comparable. DNA was measured by both branch-DNA and polymerase chain reaction techniques. K denotes 103. ND = undetectable.

  • §

    HBV surface antibody (anti-HBs), surface antigen (HBsAg), and e antigen (HBeAg) and antibody.

  • This patient was deceased 20 months after viral recurrence.

4IIIADF36300 ++18
5IIIADF5610,191 ++54
7IIITNV12ND ++35
10IVADF3012 K+++17
13IVADF6140 KNA++33
14IVADF21850 KNA++22

Allograft and Patient Survivals

Overall, there were 15 patient deaths within all 4 groups. These were due to sepsis (n = 4), recurrent cancer (n = 4), cardiopulmonary failure (n = 4), graft vs. host disease (n = 2), and HBV recurrence (n = 1). Proportions of patient survival over time for each group are shown in Figure 1.

Figure 1.

Kaplan-Meier survival curves by patient group. Months of survival post-LTx are plotted along the x-axis, with the proportion surviving plotted along the y-axis. Groups of patients were characterized as follows: I, HBV DNA– pre-LTx (LAM therapy); II, HBV DNA– pre-LTx (HBIg and LAM therapy); III, I, HBV DNA+ pre-LTx (LAM therapy); IV, HBV DNA+ pre-LTx (HBIg and LAM therapy).

Groups I and II

A total of 27 of the 33 patients in Group I (82%) were alive at the end of the study period. There were 5 retransplants required in this group, yielding an allograft survival rate of 27 / 38 (71%). The death rate was .003 per 1,000 months of follow-up. Group II had 14 / 18 (78%) patient survivals, also a death rate of .003 per 1,000 months of follow-up. Three patients required retransplantation, yielding an allograft survival rate of 14 / 21 (67%).

Groups III and IV

Broken down by group, patient and allograft survival rates were 94% and 89%, respectively, in Group III and 83% and 76%, respectively, in Group IV. Death rates per 1,000 months of follow-up were .002 for Group III and .01 for Group IV (P = not significant; Cox regression). The 1 death related to HBV recurrence was a 48-year-old Asian male in Group IV. This patient had originally responded to posttransplant antiviral LAM therapy, but developed LAM resistance at month 15. He survived 20 months with the addition of famciclovir to his original regimen. He then developed progressive, fibrosing cholestatic hepatitis that culminated in graft failure, and he received a short course of ADF just before his death.

Addition of Novel Antiviral Agents

ADF (10 mg/day orally) or TNV (300 mg/day orally) was added 1 to 66 months after HBV DNA breakthrough and continued for 2 to 20 months (median = 4 months) in 14 patients. No adverse reactions were noted in the patients treated with TNV or ADF. The average times to undetectable HBV DNA between TNV- or ADF-treated patients were 260 and 315 days, respectively.

Hepatitis B e Antigen Serological Conversion

Serological conversion from hepatitis B e antigen (+) to hepatitis B e antibody (+) following treatment with either ADF or TNV occurred in 8 (50%) of 16 patients. Loss of surface antigen was seen in 22% of the patients that were treated for HBV-LAM resistance.

Antiviral Therapy–Related Adverse Events

There were no documented antiviral drug related side effects among these patients.


Posttransplant prophylaxis with HBIg therapy, with or without a nucleotide analog such as LAM or ADF, is the current standard of care for most transplant centers for all forms of HBV transplantation.4, 11, 13–19 The advent of HBIg therapy has resulted in successful prevention of recurrent HBV post–liver transplant, improving both patient and allograft survival.20 However, there is disagreement regarding the timing, dosing, and duration of HBIg treatment in transplant centers throughout the world. In addition, there remains a lack of consensus regarding what constitutes target serum HBV surface antibody levels.19 Some centers do not monitor serum HBV surface antibody values, while others advocate the routine observation of levels and recommend HBIg dosing to HBV surface antibody concentrations as high as 450 IU/mL.15 Other challenges with HBIg therapy include significant cost and occasionally limited availability.

LAM is an oral nucleoside analog that provides relatively effective suppression of HBV in the post–liver transplant period.21 However, LAM therapy must be continued indefinitely in transplant recipients, putting patients at risk of the development of the YMDD mutation that results in LAM resistance. HBV viral recurrence due to the development of LAM resistance will frequently result in disease progression with allograft failure and death. Newer antiviral agents, such as TNV and ADF, have resulted in a decrease in mortality among our HBV transplant recipient population. We sought to determine whether antiviral therapy with either of these agents was associated with an improvement in the ability to treat HBV reemergence due to putative LAM resistance.

In this retrospective analysis of nearly a decade of patients transplanted for HBV-associated liver disease, patients who were HBV DNA– prior to liver transplant had no recurrence of HBV post–liver transplant. We found 82% patient and 71% allograft survivals in the patients negative for HBV pre–liver transplant who were treated with LAM alone (Group I). Similarly, survival rates were 78% (patient) and 67% (allograft) in HBV DNA– patients treated with a combination of HBIg and LAM (Group II). It appears that no additional advantage was conferred by use of combination HBIg and LAM therapy compared to the success of LAM monotherapy in this patient population.

In contrast, we found that those patients who had detectable HBV DNA viral titers pre–liver transplant were at risk of HBV viral recurrence post–liver transplant. This is consistent with work from the early 1990s that demonstrated an increased risk of HBV recurrence in patients who were HBV DNA+ pretransplant.22 In our study, 14 out of the 41 patients (34%) with detectable HBV DNA pre–liver transplant experienced HBV recurrence despite posttransplant antiviral therapy. The median time for detection of abnormal alanine aminotransferase and HBV recurrence was 15.5 months (range 3-30 months). Broken down by treatment type, actuarial HBV recurrence rates at 1 and 3 years in Group III (treated with LAM monotherapy) were significantly higher than the 1- and 3-year rates in Group IV (treated with a combination of HBIg and LAM) (P = .004; log-rank test). The combination of HBIg followed by LAM treatment appeared to provide a decreased risk of HBV recurrence in comparison to LAM monotherapy, suggesting that all patients within these 2 subgroups should be treated with HBIg + LAM. These results are consistent with prior literature,23, 24 although even stronger results with combination therapy have been observed recently.25, 26 Moreover, as breakthrough generally occurred after discontinuation of combination LAM / HBIg therapy, our results suggest that maintenance of HBIg may be helpful in preventing breakthrough. It should be noted, however, that the observed differences in recurrence rates may have been due to the availability of ADF and TNV with resultant successful control of HBV viral breakthrough in this group. Other centers have also reported similar success with these agents, resulting in improved survival in HBV infected transplant recipients.21, 27 Additionally, ADF may be more ideally suited for these 2 groups of patients, as previously reported in LAM-resistant patients by Buti and Esteban28 However, we did not detect any patient or allograft survival advantage with combination treatment. It should further be noted that breakthrough in these groups was presumed to be due to LAM resistance, although since we did not sequence the virus, patient nonadherence to treatment could not be ruled out as a possible factor.

In conclusion, our results suggest that antiviral therapy to prevent recurrent hepatitis B post–liver transplant should be tailored to HBV status pretransplant. These findings have increased our awareness of the importance of HBIg, especially in certain groups at high risk for HBV recurrence. Future, prospective studies will be needed to determine if preemptive therapy with ADF and TNV provide an improved long-term solution to treatment resistance.