$$\rightleftharpoonup{xx}$$
$$\longleftharp{xx}$$,
$$\longrightharp{xx}$$,
Ethical approval for this retrospective research was granted by the Institutional Review Board (IRB) of Renji Hospital, School of Medicine, Shanghai Jiao Tong University. All procedures were executed in strict adherence to the guidelines outlined in the Declaration of Helsinki. Because the analysis exclusively utilized de-identified historical records, the review board explicitly exempted the study from requiring written informed consent. Patient privacy and data security were rigorously maintained throughout the investigation. The stepwise methodology employed to screen and classify the final analytical cohort from the primary transplant registry is detailed in Figure 1. The reagents and the equipment used are listed in the Table of Materials.
1. Study design and patient population
The analytical cohort was established by screening individuals who underwent liver transplantation between January 1, 2010, and July 31, 2024. Ultimately, 150 recipients were enrolled based on the following specific inclusion parameters: an age range of 18– 74 years, a confirmed LT procedure, and postoperative HBsAg quantification greater than 1 IU/mL. To guarantee that the de novo cases represented purely community-acquired or recipient-originated infections rather than graft-transmitted diseases, any recipient of an organ from an anti-HBc-positive donor was strictly excluded, along with individuals lacking continuous follow-up records. The finalized cohort was subsequently divided according to pre-surgical serology: the DNH cohort (n = 36) comprised initially HBsAg-negative patients who acquired a novel infection postoperatively, whereas the recurrent cohort (n = 114) consisted of individuals with pre-existing chronic HBV whose HBsAg and/or HBV DNA re-emerged post-transplantation.
2. Data collection, laboratory assays, and definitions
All patients were regularly followed up in the outpatient clinic after liver transplantation. The standardized postoperative surveillance protocol included intensive monitoring (e.g., weekly during the first month, monthly for the first 6 months, and every 3 – 6 months thereafter). A comprehensive review of medical records was conducted to collect data on demographic characteristics, primary liver disease (including indications for transplantation and baseline severity assessed via the Model for End-Stage Liver Disease [MELD] and FIB-4 scores), post-transplant immunosuppression regimens, time to HBsAg seroconversion, and final treatment outcomes.
To address perioperative viral management, standard antiviral prophylaxis was administered in accordance with institutional guidelines. All patients in the recurrent cohort (pre-transplant HBsAg-positive) received standard perioperative prophylaxis consisting of nucleos(t)ide analogs (NAs) initiated before or at the time of transplant, supplemented with an intraoperative anhepatic phase administration of hepatitis B immunoglobulin (HBIG). Postoperatively, these patients were maintained on NA therapy with or without low-dose HBIG based on individual risk stratification. Conversely, as patients in the DNH cohort were HBsAg-negative pre-transplantation and received grafts from anti-HBc-negative donors, they did not undergo routine HBV-specific antiviral prophylaxis before the serological confirmation of the de novo infection.
For serological testing, peripheral venous blood samples (5 mL) were collected, centrifuged at approximately 1,500 x g for 10 min to obtain serum, and stored at -70°C. Levels of HBsAg, hepatitis B surface antibody (anti-HBs), hepatitis B e antigen (HBeAg), hepatitis B e antibody (anti-HBe), and anti-HBc were quantitatively detected using generic/commercial enzyme-linked immunosorbent assay kits on an automated electrochemiluminescence immunoanalyzer. The established normal reference thresholds were: HBsAg < 1 IU/mL, anti-HBs & lt; 10 mIU/mL, HBeAg & lt; 1 IU/mL, anti-HBe >1 IU/mL, and anti-HBc & gt; 1 IU/mL. To mitigate the risk of overlooking occult HBV infections, which standard chemiluminescence immunoassays might miss, routine surveillance incorporated highly sensitive real-time polymerase chain reaction (PCR) testing for HBV DNA alongside standard serological panels. HBV DNA levels were quantified using commercial PCR assays with a lower limit of detection of 20 IU/mL. All viral load values were standardized and logarithmically transformed (log10 IU/mL) for statistical consistency. In line with contemporary treatment concepts and the global pursuit of viral hepatitis elimination by 203010,11, “ advantageous population ” was defined for finite-duration NA therapy as patients meeting the following criteria: HBsAg & lt; 100 IU/mL, HBeAg seroconversion (anti-HBe positivity), HBV DNA & lt; 10 IU/mL, and alanine aminotransferase (ALT) within the normal upper limit.
3. Statistical analysis
All statistical analyses were performed using R software and IBM SPSS Statistics. Categorical variables were presented as counts and percentages and compared between the DNH and recurrent groups using the Chi-square test or Fisher's exact test when expected cell counts were less than 5. Continuous variables were assessed for normality using the Shapiro-Wilk test. Recognizing the heavily skewed, non-normal distribution of clinical parameters (e.g., time to onset, peak biochemical markers), all continuous data are expressed as median [interquartile range, IQR] and compared using the non-parametric Mann-Whitney U test.
To rigorously evaluate the primary outcome of functional cure (defined as sustained HBsAg loss), time-to-event survival analyses were conducted using the Kaplan-Meier method, with differences between strata assessed via the log-rank test. To identify independent predictors of functional cure, a multivariate Cox proportional-hazards regression model was constructed to calculate adjusted hazard ratios (aHRs).
To ensure mathematical stability and model convergence, the final multivariate model incorporated core clinical variables, including age, infection type (DNH vs. recurrent), and HBIG combination therapy. Potential baseline confounders with quasi-complete separation, such as sex and initial HBeAg status, were evaluated in univariate analyses but excluded from the final multivariate model to maintain the robustness of the hazard ratio estimates. A two-tailed P-value of less than 0.05 was considered statistically significant for all tests.