Epidemiological characteristics of de novo hepatitis B infection in liver transplant recipients—An experience from a tertiary care centre in Qatar

The emergence of hepatitis B surface antigen in a patient with previously negative hepatitis B virus (HBV) serology post‐orthotropic liver transplant (OTLX) is known as de novo hepatitis B (DNHB). As there are no data on patients with DNHB available from Qatar, we aim to do a pioneer study indexing their clinical profile and epidemiology of patients with DNHB in Qatar.


| INTRODUC TI ON
The appearance of hepatitis B surface antigen (HBsAg) in a patient with previously negative hepatitis B serology post-orthotropic liver transplantation (OTLX) is known as de novo hepatitis B (DNHB).
The etiology of DNHB may be a transfusion of blood product, infection in donor liver or occult pre-transplant infection in the recipient. 1 Internationally, there is an increasing demand and burden of shortage for donor livers, which is more pronounced in the middle eastern countries due to the high prevalence of hepatitis B and hepatitis C. Hence, liver from hepatitis B core antibody (HBcAb)-positive donors is being increasingly used, which, however, has the risk of higher hepatitis B virus (HBV) reactivation post-OTLX due to immunosuppressive therapy. 2 Many international studies have shown that HBcAb-positive grafts can be donated safely, and adequate antiviral prophylaxis decreases post-OTLX reactivation of HBV significantly. [3][4][5] Risk of DNHB post-OTLX has been reported as highly variable, ranging from 16% to 88% in international studies 6,7 and 5% to 7% in studies from the middle east region. 8   Post-operative visits were monthly during the first 6 months and then every 2 months. Hepatitis B serology, HBV DNA, and liver imaging were done every 6 months during the first year, then yearly, or earlier if there is a significant rise in transaminase levels. Biopsy of the transplanted liver was done annually or with changes in serum transaminase levels. The DNHB patients in this study were defined as those who were tested negative to HBsAg before transplantation but positive to the same at any time after the procedure, provided other possible modes were ruled out. DNHB patients were evaluated further with HBV DNA titer, elastography, and liver biopsy to assess the grade of hepatitis and stage of fibrosis. Our treatment protocol was entecavir 0.5 mg once daily. Hepatitis B immunoglobulin (HBIG) was not used in any of the cases. Prospectively maintained records in the Liver Transplant Clinic were reviewed retrospectively for demographic, transplant, treatment, and follow-up related information of DNHB cases.

| Statistical analysis
The normally distributed data and results were reported with mean and SD; the remaining results were reported with median and interquartile range. Categorical data were summarized using frequencies and percentages. Preliminary analyses were conducted to examine the distribution of the data variables using the Kolmogorov-Smirnov test. Associations between two or more qualitative variables were assessed using the Chi-square (χ 2 ) test, Fisher exact, or Yatescorrected chi-square tests as appropriate. Quantitative data and outcome measures between the two independent groups were analyzed using unpaired "t" test (or Mann-Whitney U test for skewed data). Survival functions were estimated with the Kaplan-Meier survival curve method. Additionally, the mean duration of followup with 1-and 5-year survival proportions was also calculated, and for those who did not survive, the cause of mortality was recorded.
This was further classified into liver-related and unrelated causes of mortality and included in statistical analysis. Pictorial presentations of the key results were made using appropriate statistical graphs. All P values presented were two-tailed, and P < .05 were considered as statistically significant. All statistical analyses were done using statistical packages SPSS 22.0 (SPSS Inc) and Epi-info (Centers for Disease Control and Prevention) software.

| RE SULTS
One hundred fifty-nine patients from Qatar underwent liver transplantation during 1986-2019. Among them, 65% were males, and   to the Scheuer score, with a mean elastography score of 6.7 ± 1.9.
The complete epidemiological profile of infected cases is given in Table 1, and a complete serological profile before and after transplant is given in Table 2. A flowchart describing the outcome in these patients is shown in Figure 1.
Most of the patients were treated with entecavir or tenofovir. As  Table 1). These patients were continued on the same drugs as long as they were stable. DNHB cases had a mortality rate of 23.5% Attributable mortality to liver-related causes was 75% of all-cause mortality in both infected and non-infected groups. Various factors that might affect the mortality rate among DNHB cases were statistically analyzed and presented in Table 3.
Comparative statistical findings indicate that both the mean/ median AST and ALT were significantly higher in patients who died compared to those who survived (P < .05) (Figure 3). Similarly, mortality was found to be considerably higher in patients with a severe degree of fibrosis compared to mild-to-moderate degrees of fibrosis (P < .05). Other characteristics such as age, gender, nationality, blood group, time from OTLX to the detection of DNHB, and antiviral treatment did not show any significant association

TA B L E 2 (Continued)
with mortality. Moreover, younger age and female gender were positively associated with higher proportions of survival; however, these differences were not statistically significant (P > .05), as shown in Table 3. A scatter diagram describing the linear relationship of total bilirubin, AST, and ALT with the time of onset of infection is shown in Figure 4, and a ROC curve to determine optimal cutoff values for bilirubin, AST, and ALT in predicting mortality is shown in Figure 5. This group represents the patient with DNHB who died during the period of follow-up. c Chi-square Fisher exact test was used for 2 × 2 tables and for tables more than 2 × 2, Yates-corrected chi-square test was applied in case of small cell frequencies (50% or more cells have expected frequencies < 5), whereas quantitative outcome measures were compared by using t test or Mann-Whitney U test (for skewed data) as appropriate to compute respective statistical P-value.

| D ISCUSS I ON
It is crucial to expand the donor pool. DNHB infections usually develop after LT using HBcAb-positive grafts, especially in patients with no prior exposure to HBV. The risk of transmission of hepatitis B to a liver transplant recipient is between 60% and 80% when the donor is HBcAb-positive. Hence, HBcAb-positive donors have not been preferred. 5 However, recent reports suggest that the use of HBcAb-positive grafts may not be independently associated with poor outcomes. 4 The majority of DNHB cases in our study are expected to have received HBcAb-positive donor livers, but donor information was not available. The outcome of patients with DNHB was not inferior compared to those without DNHB.
International data recommend HBsAb titer more than 1000 IU/L before transplantation. 13 In our cohort, none of the patients had this level. It is recommended to maintain HBsAb titer more than 100 IU/L post-transplantation. 14 94% of the patients in our study had loss of HBsAb post-transplantation, and their titer was less than 10 IU/L. None of our patients received HBIG or nucleoside analogues (NA) as prophylaxis post-transplantation since donor data were not available. Liver transplantation (LT) from HBcAb-positive donors is being increasingly used due to the shortage of organs. In these patients, the risk of HBV reactivation is high after LT due to immunosuppressive therapy. In a study by Cholongitas et al, HBV recurrence was found to be 11% in HBsAg-positive LT patients who received HBcAb-positive grafts compared to HBcAb-negative grafts, but overall survival was the same in both groups. They also noted that without prophylaxis, HBV reactivation was 48% in naïve patients. 15 In the early days, prophylaxis for recurrent HBV infection was given to HBsAg-positive patients using monotherapy with HBIG or lamivudine (3TC). This caused significant improvement of graft survival after LT, but the re-infection rates continued to be 30%-40%. 16 Also, 3TC monotherapy resulted in the development of HBV reverse transcriptase mutations that lead to antiviral drug resistance. 17 Combination therapies of HBIG with NA were successful in controlling HBV infection by reducing the HBV recurrence rate to less than 5%. DNHB infection rates in HBsAg-negative patients were reduced to 19%, 2.6%, and 2.8% using HBIG, 3TC, and combination, respectively. 18 HBsAg seroclearance was observed in 35.2% (6/17) of DNHB patients in our cohort. Two each received treatment with entecavir and tenofovir monotherapy, one with their combination and one with lamivudine monotherapy. A meta-analysis by Zheng et al showed that entecavir was the best prophylactic option for reducing the risk of HBV recurrence when compared against 5 other regimens (entecavir, tenofovir, adefovir, lamivudine, lamivudine plus tenofovir, and lamivudine plus adefovir). 19 Fung et al concurred that long-term entecavir monotherapy resulted in a durable HBsAg seroclearance rate of 92%, undetectable HBV DNA rate of 100% at 8 years, and excellent long-term survival of 85% at 9 years. 20

| Limitations
The determination of potential sources of HBV infection is of the utmost importance. Three possible routes (blood transfusions, recipient sources, and environmental factors) have been analyzed in this study. However, the non-availability of donor data made the evaluation of potential sources incomplete. This was not a controlled prospective work. Being retrospective in nature, follow-up and treatment strategies varied among patients. Our small number of patients limits the statistical power afforded by our data. Large, prospective, multicenter studies with long-term follow-up are required to provide statistically significant conclusions.

| CON CLUS ION
Orthotropic liver transplant in centers selecting donors liberally without screening for HBV poses risk of DNHB. However, the 5-year survival of those with DNHB is comparable to those without DNHB. As there is a considerable demand for donor livers worldwide, patients can still be referred to such centers. However, it is prudent that accu-

CO N FLI C T S O F I NTE R E S T
The authors declare no conflicts of interest.

AUTH O R CO NTR I B UTI O N S
Arun Prabhakaran Nair involved in study design and concept, data entry, data analysis, manuscript preparation, and literature review.
Sreethish Sasi (Corresponding author, Study guarantor) involved in data entry, data analysis, manuscript preparation, and literature review. Muna S. Al-Maslamani, Sulieman Abu Jarir, and Samar A.
Hashim prepared manuscript and reviewed and edited the manuscript. Prem Chandra designed the study, analyzed the data, and prepared statistical tools. Moutaz Derbala (Mentor) involved in study design and concept, manuscript preparation, and manuscript review and editing.

F I G U R E 5
Receiver operating characteristic (ROC) curve to determine an optimum cutoff value for bilirubin, AST, and ALT in predictive mortality. Statistical Method-ROC: The ROC curve was calculated using significant predictors AST, ALT, and total bilirubin to derive best suitable cutoff values and to assess model discrimination and predictive accuracy. The ROC curves provide a comprehensive and visually attractive way to summarize the accuracy of predictions. The ROC curve shows the trade-off between sensitivity and specificity and is a better method to detect the performance of a developed test, which classifies subjects into two categories such as survival and mortality. Result Interpretation: The discriminative ability of the model including predictors AST, ALT, and total bilirubin was found to be good with an area under the ROC curve value of 0.846 (AST), 0.846 (ALT) and 0.644 (total bilirubin). The cutoff score point of ≥67 had sensitivity 75% and specificity 76.9% (for AST), cutoff score point of ≥83 had sensitivity 75% and specificity 76.9% (for ALT), and at a cutoff score point of ≥12 (total bilirubin) had sensitivity 75% and 53.8%, respectively, as shown in the figure. ALT, alanine aminotransferase; AST, aspartate aminotransferase