Notice: Wiley Online Library will be unavailable on Saturday 27th February from 09:00-14:00 GMT / 04:00-09:00 EST / 17:00-22:00 SGT for essential maintenance. Apologies for the inconvenience.
Fourth Department of Internal Medicine, Aristotle University Medical School, Hippokration General Hospital of Thessaloniki, Thessaloniki, Greece
Senior Lecturer of Internal Medicine, Fourth Department of Internal Medicine, Aristotle University Medical School, Hippokration General Hospital of Thessaloniki, 49 Konstantinopoleos Street, 54642, Thessaloniki, Greece
If you can't find a tool you're looking for, please click the link at the top of the page to "Go to old article view". Alternatively, view our Knowledge Base articles for additional help. Your feedback is important to us, so please let us know if you have comments or ideas for improvement.
Despite the dramatic improvements in the treatment of patients with hepatitis B virus (HBV) over the last decade, chronic HBV infections are still associated with increased morbidity and mortality rates.1 Long-term treatment with nucleos(t)ide analogues (NUCs) results in stabilization and even improvement for most patients with decompensated HBV cirrhosis and thus significantly reduces the need for liver transplantation (LT) for HBV-induced chronic liver failure.2, 3 However, hepatocellular carcinoma may still develop in patients with cirrhosis after several years of treatment-induced virological remission4 and is currently the most common cause of LT in patients with chronic HBV.
In the early 1990s, the introduction of long-term hepatitis B immunoglobulin (HBIG) use significantly reduced the rate of posttransplant HBV recurrence and improved the prognosis of patients; LT became practical for many patients with HBV-related end-stage liver disease.5 In the late 1990s, the use of lamivudine (LAM) further improved the outcomes of transplant patients with HBV.6 Subsequently, newer NUCs were used in the pretransplant and posttransplant periods; the main aims were the treatment of HBIG failure and a reduction in the need for expensive HBIG preparations.6 Currently, combinations of HBIG and NUCs (mostly LAM) are considered the standard of care for prophylaxis against HBV recurrence after LT.2 The efficacy of these combinations has been shown to be superior to the efficacy of HBIG or NUCs alone. A recent meta-analysis of 6 studies showed that in comparison with HBIG alone, the combination of HBIG and LAM resulted in 12-, 12-, and 5-fold reductions in HBV recurrence, HBV-related death, and all-cause posttransplant mortality rates, respectively.7 Another systematic review showed that the combination of HBIG and LAM was superior in preventing the reappearance of hepatitis B surface antigen (HBsAg) and HBV DNA and in reducing overall and HBV-related mortality rates in comparison with HBIG alone; this combination was also superior in preventing the reappearance of only HBsAg in comparison with LAM alone.8
Although these meta-analyses showed the superiority of combined HBIG-LAM prophylaxis versus HBIG or LAM monoprophylaxis,6 there is still controversy about the optimal protocol: HBIG has several limitations, and LAM, the first and most commonly used NUC in transplant patients, is associated with high rates of viral resistance when it is used for the long-term treatment of chronic hepatitis B.9 In addition, among the newer NUCs, adefovir dipivoxil (ADV) has already been evaluated in several studies, but its effectiveness and optimal use in combination with HBIG and/or LAM are not well documented in the LT setting.
In this review, we systematically evaluate all the available data so that we can accurately assess not only the effectiveness of several combinations of HBIG and NUCs as prophylaxis against HBV recurrence after LT but also treatment-related factors affecting the effectiveness. In particular, we document the rates of HBV recurrence with respect to various HBIG factors (dosage, mode, frequency, and duration), the use of different NUCs, and the HBV DNA status at the time of LT so that we can identify the optimal post-LT prophylaxis.
ADV, adefovir dipivoxil; anti-HBs, antibody to hepatitis B surface antigen; bDNA, branched DNA; HBeAg, hepatitis B e antigen; HBIG, hepatitis B immunoglobulin; HBsAg, hepatitis B surface antigen; HBV, hepatitis B virus; IM, intramuscular(ly); IV, intravenous(ly); LAM, lamivudine; LT, liver transplantation; NA, not available; NUC, nucleos(t)ide analogue; PC, prospective cohort study; PCR, polymerase chain reaction; RC, retrospective cohort study; RCT, randomized controlled trial; TMA, transcription-mediated amplification.
MATERIALS AND METHODS
Data Sources and Searches
We searched MEDLINE and PubMed to identify all medical publications including the terms hepatitis B, recurrence, and liver transplantation (January 1998 to June 2010). In addition, we performed a manual search of all relevant review articles and the original studies as well as abstracts from major hepatology and LT congresses during the last 2 years.
All studies published in English were included if they fulfilled all of the following criteria:
1They were randomized trials or observational cohort studies.
2The patients were adults who underwent LT for HBV-related liver disease.
3The patients received HBIG and NUCs as prophylaxis against post-LT HBV recurrence, and there was no restriction on the duration of HBIG use.
4Data on the incidence of post-LT HBV recurrence and antiviral prophylaxis were available.
5HBV recurrence was defined as the reappearance of HBsAg, HBV DNA, or both after LT.
The literature search was performed by 1 reviewer (E.C.); after screening the titles and abstracts, he determined which studies could potentially be included. Each study on this list of preselected publications was evaluated by 2 independent reviewers (E.C. and G.V.P.) to determine whether it fulfilled all the inclusion criteria.
Data Extraction and Quality Assessment
The extraction of data from the selected publications was performed by 2 researchers (E.C. and J.G.) according to a predefined form. Any disagreements about data extraction were arbitrated by discussions with 2 other researchers (E.A. and G.V.P.). The data that were extracted from the selected studies included the country, the center or centers, the date of publication, the study period, the sample size, the HBIG administration protocol, the NUC prophylaxis, the follow-up period, and the number of patients with HBV recurrence. Randomized controlled trials (RCTs), prospective cohort studies (PCs), and retrospective cohort studies (RCs) were considered to provide high-quality, intermediate-quality, and low-quality evidence, respectively.
Data Synthesis and Analysis
We used a descriptive approach for summarizing the study characteristics and outcomes (HBV recurrence in patients receiving both HBIG and NUCs as prophylaxis). All data were analyzed with the SPSS statistical package (version 14.0, SPSS, Inc., Chicago, IL). Quantitative variables were expressed as means and standard deviations (or as medians and ranges if their distributions were skewed). A chi-square test was used for the identification of factors that were significantly associated with HBV recurrence. The significance testing was 2-sided, and significance was set at P < 0.05.
In all, 630 articles were initially identified in the literature search, but only 56 of these articles10-65 evaluated the rate of post-LT HBV recurrence in patients receiving HBIG and NUCs. Because 15 studies were published from 7 centers [2 from Turkey,40, 56 2 from Germany,18, 57 7 from 3 centers in the United States,36, 37, 39, 58-61 2 from Korea19, 62 and 2 from China54, 63] with overlapping study periods, only the latest study from each center was included in the analysis. In addition, 2 studies35, 64 from the same Chinese center had overlapping study periods, but we decided to include the oldest one because more details on antiviral prophylaxis were provided. Finally, newer antiviral agents such as entecavir and tenofovir were evaluated in only 2 studies55, 65 with small cohorts, and few data were available; therefore, patients who received the newer antiviral agents were not analyzed further (Fig. 1). Finally, 46 studies10-55 using HBIG and NUCs (LAM and/or ADV) were included in our analysis (Table 1). There were 3 RCTs,19, 41, 49 18 PCs,10, 12-15, 21, 24, 26, 28, 32, 33, 42-45, 52, 54, 55 and 25 RCs11, 16-18, 20, 22, 23, 25, 27, 29-31, 34-40, 46-48, 50, 51, 53 (Table 1). Ten of the 46 studies13, 15, 17, 20-22, 35, 39, 45, 46 compared HBIG monoprophylaxis and HBIG-NUC combinations, whereas data comparing NUCs and HBIG-NUC combinations were extracted from 9 studies.23, 25, 26, 29, 32, 36, 41, 49, 55
Table 1. Published Studies of HBIG and NUCs (lamivudine and/or adefovir) in Combination for the Prevention of HBV Recurrence After LT for HBV-Related Liver Disease
Mean Follow-Up (Months)
HBV DNA Assay (Lower Limit of Detection)
HBV Recurrence [n (%)]
NOTE: RCTs, PCs, and RCs were considered to provide high-quality, intermediate-quality, and low-quality evidence, respectively.
In 1 patient, recurrence was associated with no compliance with LAM.
In 1 patient, recurrence was associated with no compliance with HBIG.
In 2 patients, recurrence was observed after HBIG discontinuation.
In 12 patients, recurrence was observed after LAM discontinuation.
In 2 patients, recurrence was observed after HBIG discontinuation.
Recurrence occurred because of a lack of compliance with antiviral prophylaxis.
Recurrence occurred in 2 patients because of a lack of compliance or HBIG discontinuation.
Recurrence occurred in 2 patients because of poor compliance.
Group 1: High-dosage IV HBIG (10,000 IU in the anhepatic phase and for the next 6 days and monthly thereafter)
LAM with or without ADV
PCR (200 copies/mL)
Group 2: Low-dosage IV HBIG (3000-6000 IU monthly or 10,000 IU every 2-6 months)
Group 3: IM HBIG (1000-1500 IU every 1-2 months)
Group 4: Finite HBIG (HBIG discontinuation after various periods)
Characteristics of Post-LT Prophylaxis
In all, 2162 patients received HBIG and NUCs (LAM and/or ADV) as prophylaxis against post-LT HBV recurrence. In particular, only LAM was used in 1910 patients (88.3%),10-24, 27-30, 33-37, 39-42, 44, 47-49, 51, 52, 54, 55 and ADV was used in 154 patients (7.1%); both ADV and LAM were used in 117 of these 154 patients.25-27, 30-33, 42-44, 50, 51, 55 Eighty-eight patients (4.1%)45, 46, 53 received HBIG with either LAM or ADV (no more data were available), and 10 patients (0.5%)38 received HBIG and LAM and continued with a combination of LAM and ADV after 6 months. Patients who received HBIG and ADV with or without LAM were considered to be treated with HBIG and ADV and were analyzed together (unless stated otherwise).
We evaluated data on the mode and duration of HBIG administration as well as the dosage of HBIG during the anhepatic phase, the first week (days 0-7) after LT, and the following 3 weeks (days 8-30). Data for the mode of HBIG administration were available for 2059 patients (ie, all patients except those in 3 studies32, 50, 51 and in a subgroup of another study55): 810 always received intravenous (IV) HBIG, 870 always received intramuscular (IM) HBIG, and 379 received IV HBIG for the first period after LT and IM HBIG thereafter.10-31, 33-49, 52-55 Data for the duration of HBIG administration were available from all but 1 study32: HBIG prophylaxis was given for a finite duration (median = 180 days, range = 6-726 days) after LT to 242 of 2128 patients (11%).19, 26, 31, 37, 38, 47, 50, 55
In the anhepatic phase, HBIG was given IV in 1095 (57%), IM in 622 (32%) or both IV and IM in 209 (11%) patients; moreover, high dosages (≥10,000 IU) and low dosages (<10,000 IU) were given to 890 patients (46%) and 1036 patients (54%), respectively (data were available for 1926 patients).10-18, 20-22, 24-26, 28-31, 33-49, 51-54 During the first week after LT, IV HBIG was administered to 1081 of 1971 patients (55%), IM HBIG was administered to 632 patients (32%), and IV HBIG and IM HBIG were administered to 258 patients (13%),10-15, 17, 19-26, 28-31, 33-44, 46-55 while high dosage of HBIG (≥10,000 UY/day) was given in 533 (28%) of 1891 patients with available data).10, 11, 17, 19-22, 24, 26, 28, 31, 35-38, 42, 52, 55 Finally, from days 8 to 30,10-13, 15-18, 20-22, 24, 25, 27-30, 33-36, 38, 40-49, 52-54 IV HBIG was given to 843 patients (45%), IM HBIG was given to 733 patients (39%), and IV/IM were given to 292 patients (16%; data were available for 1868 patients); the median HBIG dosage was 2000 IU (range = 400-25,000 IU). In addition, from days 8 to 30, HBIG was given at a stable dosage to 1224 of 1876 patients (65%) and according to serum antibody to hepatitis B surface antigen (anti-HBs) levels to 652 patients (35%; median dosage = 100 IU/L, range = 50-500 IU/L).10-13, 15-18, 20-22, 24-30, 33-36, 38, 40-49, 52-54 HBIG was also given at a high dosage (≥10,000 IU at any time interval) to 427 of 1912 patients (22%) and at a low dosage (<10,000 IU at any time interval) to 1485 patients (78%).10-15, 17-21, 24, 25, 27-31, 33-36, 38, 40-42, 44, 46-49, 51-54 In 3 studies,45, 46, 55 details on HBIG administration were not provided separately for combinations of HBIG with different NUCs (ie, LAM or ADV with or without LAM), whereas the NUC therapy changed 24 weeks after LT in 1 study.38 Therefore, for subgroup analyses, the mode of HBIG administration was known for 1866 patients; the duration of HBIG administration was known for 1914 patients; and the HBIG dosage (≥10,000 or <10,000 IU) during the anhepatic phase, the first week after LT, and the following 3 weeks was known for 1837, 1822, and 1739 patients, respectively.
Administration of NUCs
LAM was given at a daily dosage of 100 to 300 mg [100-150 mg for 91% of the patients (1730/1910)], and ADV was given at a daily dosage of 5 to 10 mg (the dosage was adjusted to the renal function). LAM had been initiated before LT in 70% of the patients who received HBIG and LAM after LT, whereas ADV had been initiated before LT in 73% of the patients who received HBIG and ADV after LT. According to the available data, 646 patients had detectable HBV DNA at the time of LT,11, 13, 15-17, 19, 21, 22, 24, 26-31, 33-35, 37, 40, 42, 42, 47,, 49, 54, 55 and 372 patients (40%) were hepatitis B e antigen (HBeAg)-positive (data on the HBeAg status were available for 938 patients).13-15, 17, 19, 21, 22, 27, 28, 31, 33-37, 41, 43,, 46, 52, 54 YMDD mutants were present before LT in 73% of the tested patients (103/142) who received HBIG and ADV26, 27, 30-33, 42-44, 55 and in 2.5% of the tested patients (17/669) who received HBIG and LAM after LT (P < 0.001).
HBV Recurrence With Combined HBIG-NUC Prophylaxis
HBV recurrence was detected in 6.6% of the patients (143/2162) who received combined HBIG-NUC prophylaxis during a median follow-up of 21 months (range = 6-83 months). Twenty-three of the 143 patients with HBV recurrence reported HBIG discontinuation (4),27, 36 LAM discontinuation (12),35 or poor compliance with combined HBIG-NUC prophylaxis (7)19, 24, 37, 50, 51; data regarding the dosage of HBIG during the anhepatic phase and the first week after LT were not available (NA) for 2 of the 23 patients (9%). Hence, the rate of HBV recurrence was 5.6% (120/2139; Table 1) in patients continuing to receive HBIG and NUC prophylaxis 11 months (median; range = 1-40 months) after LT.
The rate of HBV recurrence was not associated with the characteristics of HBIG administration, except perhaps for the HBIG dosage during the first week after LT. In particular, HBV recurrence rates did not differ between patients who received a high HBIG dosage and patients who received a low HBIG dosage during the first week after LT [6.4% (34/533) versus 6.4% (87/1358), P = 0.90]. However, when we excluded 21 patients suffering from HBV recurrence who discontinued their HBIG-NUC prophylaxis or were poorly compliant (and for whom data on HBIG dosage during the first week were available), we found that HBV recurrence was less frequent in patients who received a high HBIG dosage versus patients who received a low HBIG dosage during the first week after LT [3.5% (18/517) versus 6.1% (82/1353), P = 0.035; Fig. 2].
HBV recurrence developed significantly more frequently in patients receiving HBIG and LAM [7.1% (136/1910)] versus patients receiving HBIG and ADV [3.2% (5/154), P = 0.05]. When only patients who continued to receive HBIG-NUC prophylaxis were taken into account, the combination of HBIG and ADV was again found to be superior to the combination of HBIG and LAM [2.0% (3/152) versus 6.1% (115/1889), P = 0.024; Fig. 2]. Patients who received HBIG and LAM and patients who received HBIG and ADV had similar follow-up periods (30 ± 16 versus 32 ± 17 months, P = 0.78), but the HBIG-LAM patients more frequently received HBIG indefinitely [90% (1600/1773) versus 57% (51/89), P < 0.001] and at high dosages [22% (374/1690) versus 0% (0/49), P < 0.001] and stable dosages [64% (1043/1627) versus 14% (8/56), P < 0.001] from days 8 to 30 after LT (Table 2). In addition, the rate of detectable HBV DNA at the time of LT was lower for patients who received HBIG and LAM [40% (503/1263)] versus patients who received HBIG and ADV [70% (37/53), P < 0.001].
Table 2. HBIG Administration in Patients Receiving HBIG and LAM or HBIG and ADV for the Prevention of HBV Recurrence After LT for HBV-Related Liver Disease
HBIG and LAM
HBIG and ADV
HBIG in the anhepatic phase [n/N (%)]
HBIG during the first week after LT [n/N (%)]
High dosage (≥10,000 IU/day)
HBIG during days 8-30 after LT [n/N (%)]
High dosage (≥10,000 IU/day)
Indefinite HBIG administration [n/N (%)]
Data on HBV recurrence after LT with respect to the HBV DNA status at the time of LT were available for 1200 of the 2162 patients (55.5%).11, 13, 15-17, 19-22, 24, 26, 27, 29-31, 34-38, 40-43, 45, 47, 49, 54, 55 As expected, the HBV DNA status at the time of LT significantly affected the probability of HBV recurrence, which was observed in 8.5% of the patients (41/483) with detectable HBV DNA at the time of LT and in only 4.0% of the patients (29/717) with undetectable HBV DNA (P = 0.002; Fig. 2). Except for indefinite HBIG prophylaxis, which was more frequent in patients with detectable HBV DNA at the time of LT [67% (334/499) versus 5% (33/602), P < 0.001], HBIG administration was similar for the 2 groups. In addition, patients with detectable HBV DNA at the time of LT received HBIG and ADV more frequently after LT [6.8% (37/540) versus 2.1% (16/776), P < 0.001]. Among the patients with detectable HBV DNA at the time of LT, high dosages of HBIG were significantly more frequent for those who received indefinite HBIG prophylaxis versus those who received finite HBIG prophylaxis [13% (35/279) versus 6% (8/141), P = 0.02] from days 8 to 30 after LT. In contrast, among the patients with undetectable HBV DNA at the time of LT, high dosages of HBIG were significantly less frequent for those who received indefinite HBIG prophylaxis versus those who received finite HBIG prophylaxis during the first week after LT [28% (137/491) versus 100% (33/33), P < 0.001] and during the following 3 weeks [16% (71/432) versus 30% (10/33), P < 0.001].
Data on HBV recurrence after LT with respect to the HBeAg status at the time of LT were available for 320 of 938 patients (34%).13-15, 17, 19, 21, 22, 27, 28, 31, 33-37, 41,, 43, 46, 52, 54 There were no significant differences in the HBV recurrence rates of HBeAg-positive patients [4.2% (6/143)] and HBeAg-negative patients [1.7% (3/177), P = 0.19]. Finally, there were no significant differences in the HBV recurrence rates from studies published in recent years [2005-2010; 5.8% (105/1810)] and studies published earlier [1998-2004; 4.6% (15/327), P = 0.43].
In the HBIG-LAM prophylaxis subgroup,10-24, 27-30, 33-37,, 39-42, 44, 47-49, 51, 52, 54, 55 the HBIG dosage during the first week after LT was the only factor associated with HBV recurrence. In particular, patients who received a high HBIG dosage (≥10,000 IU/day) instead of a lower HBIG dosage during the first week after LT experienced HBV recurrence significantly less frequently [3.2% (14/440) versus 6.5% (80/1233), P = 0.016]. Moreover, the rate of HBV recurrence was significantly higher for patients with detectable HBV DNA at the time of LT versus patients with undetectable HBV DNA [9.4% (32/340) versus 4.1% (25/615), P = 0.001].
For the HBIG-ADV prophylaxis subgroup,25-27, 30-33,, 42-44, 50, 51, 55 no definitive conclusions could be drawn about the possible impact of the HBIG dosage and the HBV DNA status at the time of LT on the risk of HBV recurrence because details about HBIG administration and the HBV DNA status were NA for 2 studies with cases of HBV recurrence.32, 55 The addition of LAM to the combination of HBIG and ADV did not seem to affect the rate of post-LT HBV recurrence, which occurred in 3 of 115 patients (2.6%) receiving HBIG, ADV, and LAM and in none of 37 patients receiving HBIG and ADV (P = 0.34; Fig. 3).
HBV Recurrence With HBIG and NUCs Versus HBIG Alone
As previously reported, in the 10 studies including patients who received HBIG monoprophylaxis for a median of 31 months (range = 12-98 months),13, 15, 17, 20-22, 35, 39, 45, 46 the rate of posttransplant HBV recurrence was significantly higher in patients who received HBIG monoprophylaxis versus patients who received combined HBIG-NUC prophylaxis [26.2% (68/260) versus 6.6% (143/2162), P < 0.001; Fig. 3]. At the same time, patients who received HBIG monoprophylaxis received IV HBIG significantly more frequently from days 8 to 30 after LT [82% (207/252) versus 45% (843/1868), P < 0.001] and less frequently had detectable HBV DNA at the time of LT [31% (73/235) versus 40% (483/1200), P = 0.03].
HBV Recurrence With HBIG and NUCs Versus NUCs Alone
In the 9 studies including patients who received only NUC prophylaxis for a median of 22 months (range = 12-85 months),23, 25, 26, 29, 32, 36, 41, 49, 55 posttransplant HBV recurrence was generally observed significantly more frequently with just NUCs versus HBIG and NUCs [19.0% (36/189) versus 6.6% (143/2162), P < 0.001; Fig. 3], although there was no significance difference in the proportions of patients with detectable HBV DNA at the time of LT [35% (44/124) versus 40% (483/1200), P = 0.35]. However, in patients receiving NUC monoprophylaxis, the type of NUC may have some effect on the risk of HBV recurrence. HBV recurrence was observed in 25.4% of the patients (36/142) who received LAM monoprophylaxis and in none of the 47 patients who received ADV prophylaxis without HBIG (with or without LAM) for a median of 16 months (range = 6-21 months, P < 0.001). Similarly, with respect to HBV recurrence, LAM monoprophylaxis was inferior to combined HBIG-NUC prophylaxis (P = 0.002), but HBIG-free ADV prophylaxis was not (P > 0.50).
In the last decade, HBIG-NUC combinations have been used in transplant patients with HBV in attempts to improve the efficacy of posttransplant prophylaxis and reduce costs. In 3 recent meta-analyses7, 8, 66 (317, 1335, and 551 patients, respectively), the combination of HBIG and LAM was found to achieve significantly lower HBV recurrence rates in comparison with HBIG or LAM monoprophylaxis. These findings have been confirmed and generalized in this systematic review, which includes 2162 patients who received HBIG, LAM, and ADV alone or in combination with LAM. In particular, our data show that HBV recurrence occurred in 6.6% of the patients who received HBIG and NUCs, in 26.2% of the patients who received HBIG monoprophylaxis (P < 0.001), and in 19.0% of the patients who received just NUCs (LAM or ADV with or without LAM; P < 0.001). We have also confirmed the importance of the HBV DNA status at the time of LT as a risk factor for HBV recurrence after LT even in patients receiving combined HBIG-NUC prophylaxis. In particular, HBV recurrence developed in 8.5% of the patients with detectable HBV DNA at the time of LT and in 4.0% of the patients with undetectable HBV DNA (P = 0.002), although the latter patients received indefinite HBIG prophylaxis (5% versus 67%, P < 0.001) and combined HBIG-ADV prophylaxis (2.1% versus 6.8%, P < 0.001) significantly less frequently. However, several practical questions remained unanswered; these questions mainly concern the ideal duration and dosage of HBIG administration in transplant patients with HBV who receive combined HBIG-NUC prophylaxis.6 In addition, although agents with resistance profiles better than LAM's profile (eg, ADV) have been used,6 their impact on HBV recurrence after LT has not yet been adequately evaluated.
One of the most important findings of our study is that the combination of HBIG and ADV with or without LAM is more effective than the combination of HBIG and LAM for the prevention of HBV recurrence, which developed in 2% to 3% and 6% to 7% of patients, respectively (P < 0.05). In addition, this difference remained significant when we excluded the 2 studies37, 49 with selected groups of patients who received combinations of HBIG and LAM [2.0% (3/152) versus 6.2% (115/1851), P = 0.024]. Although the numbers of patients with HBV recurrence who received HBIG and ADV with or without LAM were relatively small and the statistical difference with respect to the patients treated with HBIG and LAM was not very strong, patients who received HBIG and ADV with or without LAM significantly more frequently had factors favoring HBV recurrence. For example, the detection of HBV DNA at the time of LT was more frequent (70% versus 40%, P < 0.001), and indefinite HBIG use was less frequent (57% versus 90%, P < 0.001)]. This finding may not be unexpected because the use of ADV with or without LAM is clearly more effective than the use of LAM alone in transplant or nontransplant patients with HBV because these patients can develop LAM-resistant strains during LAM therapy in the pretransplant period.32 Thus, the better efficacy of combined HBIG-ADV prophylaxis with or without LAM may be related to the better long-term efficacy of antiviral agents (ADV with or without LAM versus LAM alone) against both wild-type and LAM-resistant HBV strains.67 Although the number of patients who received HBIG and ADV (without LAM) was too small for definite conclusions, none of the 37 patients who received this prophylaxis regimen developed HBV recurrence.
The need for indefinite prophylaxis with a high HBIG dosage substantially increases the costs of managing transplant patients with HBV. Our data show that the HBIG dosage during the first month (days 8-30) and indefinite HBIG use had no significant impact on the rates of HBV recurrence with combined HBIG-NUC prophylaxis. Among the parameters of HBIG use that were evaluated in this review, only a high dosage during the first week after LT was found to be significantly associated with HBV recurrence. In fact, with HBIG-NUC prophylaxis, the HBV recurrence rates were 3.5% with high HBIG dosages (≥10,000 IU/day) and 6.1% with lower HBIG dosages during the first week after LT (P = 0.035); the 2 subgroups were similar for all other characteristics of combined prophylaxis, including the time to recurrence after LT (11 ± 12 versus 11.5 ± 10 months). The importance of the HBIG dosage during the first week after LT may be related to the high risk of HBV recurrence during the first postoperative period; at this time, the immunosuppression levels are highest, and the immunosuppression regimen includes frequent courses of corticosteroid pulse therapy for the management of acute rejection episodes. The latter has been shown to be another significant factor for HBV recurrence.35
A limitation of our review is that we were not able to analyze the likelihood of heterogeneity in immunosuppression because of the limited available data. The use of immunosuppression, which is an important factor for HBV recurrence, may have decreased in recent years in most centers, particularly during the maintenance phase; this might have had a favorable effect on the HBV recurrence rates in the studies with HBIG and ADV, which were conducted more recently than most of the studies with HBIG and LAM. However, there was no association between HBV recurrence and the time of publication [1998-2004 versus 2005-2010 or other periods (data not shown)]. In addition, the potential effect of immunosuppression changes over time might be difficult to evaluate because the enrollment periods of the studies differed greatly (median = 6 years, range = 2-20 years), and the immunosuppression protocol was modified over time in several studies. Another limitation of our review is the heterogeneity of the HBV DNA assays, which had different detection limits (Table 1); this may have affected the detectability of HBV DNA at the time of LT as well as the diagnosis of HBV recurrence after LT.
We found no association between the HBeAg status before LT and HBV recurrence after LT, but the available data were rather limited. Similarly, although coinfections with hepatitis C and/or hepatitis D and acute liver failure due to HBV have been associated with a lower risk of HBV recurrence, further analysis was impossible because of the small numbers of patients in these subgroups (24 patients with hepatitis C virus, 48 patients with hepatitis D virus, and 108 patients with acute liver failure) and the small number of studies (5)24, 27, 30, 42, 55 that evaluated their impact on HBV recurrence.
Although the number of patients who received HBIG-free ADV prophylaxis (with or without LAM) was relatively small (n = 47) and the post-LT follow-up was short (median = 16 months), this approach seems superior to LAM monoprophylaxis and perhaps is not inferior to combined HBIG-NUC prophylaxis (Fig. 3). In fact, no case of HBV recurrence was observed in 31 patients receiving ADV prophylaxis with or without LAM in 2 studies26, 32 or in 16 patients who received this prophylaxis after the discontinuation of HBIG 1 year post-LT in another study.49 However, more data for larger numbers of patients with longer follow-up periods are required before definite conclusions about the efficacy of HBIG-free ADV prophylaxis can be drawn.
Interestingly, 2 factors that were found to affect the rates of HBV recurrence with HBIG and NUCs, a high HBIG dosage during the first week after LT and the HBV DNA status at the time of LT, were found to be valid only in patients who received combined HBIG-LAM prophylaxis and not in patients who received HBIG and ADV. Thus, it is possible that a more effective prophylactic approach (eg, HBIG and ADV versus HBIG and LAM) may neutralize the negative effects of other factors such as viremia at the time of LT or the initial HBIG dosage. In the future, the use of newer, more potent NUCs such as entecavir and tenofovir may further increase the efficacy of post-LT prophylaxis in transplant patients with HBV and lead to substantial reductions in the HBIG dosage or even eliminate the need for HBIG, at least after the initial postoperative period.
In conclusion, although the management of transplant patients with HBV is crucial for satisfactory long-term outcomes, strong comparative data on post-LT HBV prophylaxis are lacking, so the optimal approach with the best cost-benefit ratio remains unclear. Our review shows that the combination of HBIG and NUCs is an effective prophylactic approach with very low recurrence rates, particularly in patients with undetectable HBV DNA at the time of LT. This is currently a common practice because patients with HBV cirrhosis should undergo transplantation only after the clearance of serum HBV DNA by oral antiviral therapy. However, the risk of HBV recurrence is only approximately 2 times higher (8.5% versus 4.0%), and this policy could prevent LT for some patients who desperately need it (particularly those with hepatocellular carcinoma). Our findings favor the use of HBIG and ADV (with or without LAM) versus the combination of HBIG and LAM as prophylaxis against HBV recurrence, particularly in high-risk patients (eg, those with detectable HBV DNA at the time of LT). Patients receiving HBIG-LAM prophylaxis should receive a high HBIG dosage (≥10,000 IU IV per day) for the first 7 days after LT; however, the administration of HBIG does not seem to have an effect on the efficacy of combined HBIG-ADV prophylaxis. Further studies using newer NUCs with greater potency and better resistance profiles (entecavir and tenofovir) are needed to evaluate their safety and efficacy in regimens with no HBIG at all or after a short initial period (eg, 1-4 weeks) post-LT. We hope that the results of 2 ongoing clinical trials (NCT00059267 and NCT01046799 at http://www.clinicaltrials.gov) will elucidate some of these issues. If NUC monoprophylaxis is shown to be effective in clinical studies, then drug compliance and eventually effectiveness in clinical practice should be assessed. In particular, drug compliance during long-term NUC therapy may be a very important issue for transplant patients with HBV who feel healthy but have a lifelong risk of HBV recurrence.