Several reports have indicated that LTs using HBsAg(+) grafts show outcomes equivalent to those using normal grafts. [9] The development of NAs with low resistance rates and HBIG has contributed to the expansion of the donor pool using HBsAg(+) grafts. [19–21] Furthermore, research has shown that HBsAg(+) grafts may be safely donated to recipients without a history of HBV infection who have been vaccinated against HBV. [22]
Since the 1983 nationwide vaccination program decreased the vertical transmission of HBV in Korea, [23] the overall prevalence of HBsAg carriage has remained below 3% since 2008 (range, 2–7%). [24] However, the prevalence was still high in older age groups in the 2019 data. [7] Therefore, the donor pool in Korea may be expanded by using HBsAg(+) grafts.
We compared the characteristics and post-LT outcomes of three grafts grouped by serology: HBsAg(+) [S(+), graft from donors with HBV replication], HBsAg(−)/HBcAb(+) [C(+), grafts from donors with a history of HBV infection but in the resolving stage], and HBsAg(−)/HBcAb(−) [SC(−), grafts from donors without a history of HBV infection]. We found no differences in the survival rates among the S(+), C(+), and SC(−) groups following PSM. This indicated that each graft showed a statistically imilar outcome in similar clinical situations. Thus, when HBIG and NA were used concurrently, the transplantation outcome of HBsAg(+) grafts was comparable to that of HBsAg(−)/HBcAb(+) or HBsAg(−)/HBcAb(−) grafts. Additionally, serum levels of liver enzymes in the HBsAg(+) group returned to normal within one year after surgery and remained steady until the 5-year follow-up. This suggests that the graft should be used carefully based on the recipient’s condition.
HBV reactivation is conventionally defined based on classical criteria, which include an increase in both the HBV DNA titer and a positive seroconversion of HBsAg. However, as recipients with HBsAg(+) grafts are in a persistent HBsAg(+) state, HBV recurrence should be characterized by an increase in the serum HBV DNA titer. [25] Accurate data analysis of HBV seroconversion and DNA replication after LT was not possible due to the paucity of data on HBV DNA titers in the KOTRY database.
Hepatitis D virus (HDV) coinfection suppresses HBV recurrence in LT patients. [26] However, other studies have asserted that HDV recipients have poor prognoses because of the progression of HDV-induced cirrhosis. [27, 28] In addition, HDV infection may contribute to HCC recurrence and de novo progression. [26, 28, 29] When employing grafts with HBV and HDV infection, the status of both the graft and recipient should be evaluated since it may alter the LT outcome; however, we did not analyze these data because they were not included in the dataset.
Within 2–3 years after surgery, significant events affecting the postoperative course, including patient and graft loss, graft rejection, and complications, were observed in the HBsAg(+) group. Subsequently, the incidence rate declined sharply. Infections were most frequent 2–4 years after transplantation. [30] HBV seroconversion and recurrence occurred within 3 years post-transplantation. HCC recurred within 3 years in the HBsAg(+) group. However, HCC recurrence was consistently observed during the 5-year follow-up period in the HBsAg(−) group. HCC recurrence is most common 14–16 months postoperatively [31], but late recurrence may occur even 5 years after transplantation. [32, 34] Therefore, further studies, including long-term follow-up of HCC recurrence, are necessary. Sabb et al. reported that HBV recurrence does not directly affect transplantation outcomes. [30] However, several studies have shown a close association between HBV infection and HCC recurrence. [35–37] In our study, HBV infection in donors and recipients did not influence HCC recurrence with concurrent use of HBIG and NA.
This study had several limitations. The small sample size (n = 20) of this study hindered a clear interpretation of the results and matching. For instance, we included both living- and deceased-donor patients during matching because the accuracy was lower when matching S(+) patients with only deceased-donor LT patients. The short follow-up duration (< 5 years) limited the long-term safety evaluations of these grafts. Questionable representativeness owing to the low registration rate in the KOTRY database (approximately 60% of all LTs conducted nationwide) might reduce the reliability of the results. The fact that it was impossible to further investigate the lost data due to the encryption of patient information was also a factor that hindered the integrity of the results. The interpretation of HBV serology in donors and recipients may not be clear as the current KOTRY database does not include HBV titers. Accurate evaluation of HBV recurrence and seroconversion is difficult because of the ambiguity of HBV-positive seroconversion and HBV recurrence notation methods in the KOTRY database. The serum levels of HBsAg and HBV DNA were monitored and recorded in the database.
Nevertheless, the strength of this study is that it included the largest dataset from a national database covering the largest comparative group. This demonstrated that post-LT outcomes, including patient and graft survival, HCC recurrence, graft rejection, hepatitis C virus recurrence, and various complications, were not significantly different between LTs using HBsAg(+) grafts. Moreover, the recovery of hepatic and renal function after transplantation did not differ from that of HBsAg(−) grafts. Comparative analysis performed after PMS provided a more accurate evaluation of HBsAg(+) transplantation outcomes. This study demonstrates the effects of HBV infection on transplantation outcomes by comparing the survival and transplantation outcomes of donors and recipients with various serological statuses.
In conclusion, HBsAg(+) liver grafts should not always be discarded and can be safely transplanted into HBsAg(+) patients in the era of NA with HBIG. Based on PSM, the use of HBsAg(+) liver grafts did not seem to increase postoperative morbidity and mortality.