By continuing to browse this site you agree to us using cookies as described in About Cookies
Notice: Wiley Online Library will be unavailable on Saturday 7th Oct from 03.00 EDT / 08:00 BST / 12:30 IST / 15.00 SGT to 08.00 EDT / 13.00 BST / 17:30 IST / 20.00 SGT and Sunday 8th Oct from 03.00 EDT / 08:00 BST / 12:30 IST / 15.00 SGT to 06.00 EDT / 11.00 BST / 15:30 IST / 18.00 SGT for essential maintenance. Apologies for the inconvenience.
Potential conflict of interest: Nothing to report.
Genotypes B and C are the major hepatitis B virus (HBV) genotypes in Taiwan, and genotype C is associated with more severe liver disease than genotype B. Whether the implementation of the hepatitis B immunization program has affected the secular trend of the HBV genotype distribution remains unknown. We thus investigated the HBV genotypes in hepatitis B surface antigen (HBsAg)–carrier children born before the implementation of the universal infant immunization program and in those born afterward. One hundred seven children who were infected with HBV despite appropriate immunization were enrolled as immunized cases with HBV breakthrough infection. Each case was matched with two unimmunized HBsAg carriers according to the age at enrollment. HBV genotypes were determined with molecular methods. Compared with unimmunized HBsAg carriers, more immunized children had HBsAg-positive mothers (65.9% versus 100%, P< 0.001) and were infected with genotype C (16.4% versus 42.1%, P< 0.001). Among the children born to HBsAg-positive mothers, the mothers' and children's HBV genotypes were highly concordant in both unimmunized [κ = 0.97, 95% confidence interval (CI) = 0.90-1.00] and immunized children (κ = 0.97, 95% CI = 0.92-1.00). After adjustments for gender, maternal age, and delivery mode, immunized HBsAg-carrier children born to HBsAg-positive mothers had a higher likelihood of genotype C infection than unimmunized children (odds ratio = 3.03, 95% CI = 1.62-5.65, P = 0.001). However, the increased genotype C to genotype B ratio was not seen in the HBsAg-carrier mother pool in the postimmunization era. Conclusion: In the postimmunization era, most HBV breakthrough infections are due to maternal transmission, and immunized children born to genotype C mothers may have a higher rate of breakthrough infection than those born to genotype B mothers. (HEPATOLOGY 2011;53:429-436.)
Hepatitis B virus (HBV) is a significant cause of chronic hepatitis, liver cirrhosis, and hepatocellular carcinoma (HCC) worldwide.1 In Taiwan, an area hyperendemic for HBV infection,2 the disease is usually acquired perinatally or in early childhood.3, 4 Since the launch of the universal infant immunization program against HBV in July 1984, the seropositive rate of hepatitis B surface antigen (HBsAg),5 the incidence/mortality rate of fulminant hepatitis,6, 7 and the incidence rate of HCC8 in Taiwanese children have all substantially declined. However, the current immunization strategy cannot completely eradicate the transmission of HBV. Approximately 10% of infants born to HBsAg-positive and hepatitis B e antigen (HBeAg)–positive mothers are still infected and suffer from chronic hepatitis B.9, 10
In addition, although the overall incidence rate of childhood HCC has declined, HBsAg-carrier children born after the implementation of the immunization program bear a higher risk of developing HCC than those born before the program (risk ratio = 2.3-4.5).11 The factors contributing to HBV breakthrough infection and the subsequent development of HCC in these carrier children remain largely unknown. It is, therefore, important to compare the clinical and virological characteristics of HBsAg-carrier children born before the implementation of hepatitis B immunization program and those born afterward.
At least eight HBV genotypes (A-H) have been identified worldwide on the basis of a divergence of 8% or more of the entire nucleotide sequences.12-15 Before the immunization era, HBV genotype B was the most prevalent genotype in Taiwan and accounted for approximately 80% of HBV strains. Genotype C accounted for the remaining 20%, and the other genotypes were very rare.16 In addition, in comparison with genotype B, genotype C is associated with a delayed HBeAg seroconversion, a higher viral load, and more severe liver diseases such as cirrhosis and HCC.17-21 Because a higher maternal viral load leads to a higher likelihood of HBV breakthrough infection in infants,22-25 we hypothesized that the HBV genotype associated with a delayed clearance of HBeAg and a higher viral load would result in a higher rate of breakthrough infection. Thus the distribution of HBV genotypes may change in the immunization era. In this study, the secular trend of the HBV genotype distribution was investigated in Taiwanese HBsAg-carrier children born before the implementation of the hepatitis B immunization program and in those born afterward. In addition, because perinatal transmission is an important route of HBV spread in Taiwan,3 HBV genotypes of HBsAg-positive mothers were also examined.
Abbreviations: CI, confidence interval; HBeAg, hepatitis B e antigen; HBIG, hepatitis B immunoglobulin; HBsAg, hepatitis B surface antigen; HBV, hepatitis B virus; HCC, hepatocellular carcinoma; PCR, polymerase chain reaction; SD, standard deviation; ULN, upper limit of normal.
Patients and Methods
Universal Hepatitis B Immunization Program
In Taiwan, the hepatitis B immunization program beginning at birth was implemented on July 1, 1984. After the program was launched, hepatitis B serological tests were compulsory for all pregnant women so infants born to HBsAg-positive mothers could be identified. Initially, the program covered only newborns of HBsAg-positive mothers; it was extended to all newborns after July 1986. Before July 1992, four doses of a plasma-derived vaccine (Hevac B, Pasteur-Merieux, Lyon, France) or its equivalent (Lifeguard hepatitis B vaccine, Hsin-Chu, Taiwan) were given at 0, 1, 2, and 12 months of age. After July 1992, three doses of the recombinant vaccine H-B-Vax II (5 μg/0.5 mL; Merck Sharp & Dohme, Rahway, NJ) or Engerix-B (20 μg/mL; SmithKline Beecham, Rixensart, Belgium) were administered (within the first week of birth, at 1 month of age, and at 6 months of age). For newborns of HBeAg-positive mothers or HBsAg-positive mothers with a high titer of HBsAg (reciprocal titer >1:2560 as confirmed by reverse passive hemagglutination testing), 0.5 mL (100 IU) of hepatitis B immunoglobulin (HBIG) was administered within 24 hours of birth.10, 26 For newborns of HBsAg-positive but HBeAg-negative mothers, the administration of HBIG was optional.
Children With Chronic HBV Infection
Four hundred seventy-one children who were 15 years of age or younger and had been diagnosed with chronic HBV infection (i.e., they were HBsAg-seropositive for at least 6 months) were recruited. These carrier children were enrolled from (1) the outpatient clinic of National Taiwan University Hospital in a prospective study beginning approximately 3 decades ago, (2) a prospective screening program for carrier children of HBsAg-seropositive mothers, or (3) five cross-sectional studies of a hepatitis B vaccination efficacy survey in 1984, 1989, 1994, 1999, and 2004.19 Informed consent was obtained from the children's parents or guardians. The institutional review board of National Taiwan University Hospital approved the protocol for this study.
Immunized Cases With HBV Breakthrough Infection.
To identify cases with HBV breakthrough infection despite immunization beginning at birth, we reviewed the hepatitis B immunization histories of the 471 children with chronic HBV infection. The immunization histories were obtained by the transcription of information from the health booklet distributed to each newborn by the Department of Health of Taiwan. We also checked the birth and/or immunization records kept by National Taiwan University Hospital if the subjects had been born or immunized there. Those with an unknown maternal HBsAg status, unclear hepatitis B immunization histories, or inappropriate hepatitis B immunization were excluded from the analysis. Finally, 107 of the 471 children with chronic HBV infection were confirmed to have received appropriate hepatitis B immunization at infancy and thus were designated as immunized cases with HBV breakthrough infection.
Age-Matched, Unimmunized HBsAg Carriers.
After checking the hepatitis B immunization histories, we identified 337 HBsAg-carrier children who did not receive any hepatitis B immunization. According to the age at enrollment, we divided the immunized and unimmunized HBsAg-carrier children into three groups: 0 to 5 years, 6 to 10 years, and 11 to 15 years. Each immunized case with HBV breakthrough infection was matched with two randomly selected unimmunized carriers from the corresponding age group. The flow of the participants is shown in Fig. 1.
The maternal HBsAg status was checked at the time of the children's enrollment. For children born after the implementation of the immunization program, the prenatal maternal HBsAg/HBeAg status was extracted from records. Serum samples of the HBsAg-positive mothers were collected after their children's enrollment and were stored for HBV genotyping at a later date.
In addition, to determine the maternal HBV genotype distribution in the general population in the postimmunization era, HBV genotypes of another 136 HBsAg-positive women who delivered babies from April 2007 to March 2009 at National Taiwan University Hospital were also examined at the time of delivery.
HBV Serology and Serum Aminotransferase Levels
The hepatitis B serological markers (HBsAg, antibody against HBsAg, antibody against hepatitis B core antigen, HBeAg, and antibody against HBeAg) were determined by enzyme immunoassays (Abbott Laboratories, North Chicago, IL). The serum alanine aminotransferase levels were measured with an autoanalyzer (model 7450, Hitachi, Tokyo, Japan).
The serum samples of HBsAg-carrier children were collected upon each subject's enrollment and were stored at −80°C for HBV genotype analysis at a later date. The children's HBV genotypes were determined by polymerase chain reaction (PCR) with genotype-specific primers in the regions of the pre-S1 and S genes.27 The procedures are described in detail in our previous publication.19 Maternal HBV genotypes were determined by real-time PCR with subsequent melting curve analysis based on a single nucleotide polymorphism in the X gene as previously described.28
When the HBV genotype identified in a given child was inconsistent with that identified in his or her mother, direct sequencing of the S gene from both the child and the mother was performed to confirm the genotyping results. The S genes were amplified with the primers 5′-ctg-ctg-gtg-gct-cca-gtt-cag-3′ (sense, position 57-74) and 5′-taa-cct-ttc-ata-cag-ttt-ctt-aac-acc-cag-aaa-ac-3′ (antisense, position 977-1011). The PCR products were sequenced on both strands with the BigDye Terminator V3.1 cycle sequencing kit (Applied Biosystems, Foster City, CA) with the same primers used for PCR. The sequencing products were analyzed with an ABI 3730xl DNA analyzer (Applied Biosystems). The obtained sequences were aligned with GenBank sequences corresponding to HBV genotypes B and C. The GenBank accession numbers were as follows: AB191369, AB191351, DM059403, CS388974, and CS409744 for genotype B and AB113875-AB113879, AB191350, AB191352, AB191353, AB191357, AB191359, AB191362, AB191365, AB191368, AB191380, and AB191388 for genotype C. BLASTN 2.2.22 was used to align sequences and determine genotype identity by means of sequence similarity.29, 30
Statistical analyses were performed with Stata 8.2 software (Stata Corp., College Station, TX) and SAS 9.1.3 software (SAS Institute, Inc., Cary, NC). Two-sided P values ≤ 0.05 were considered statistically significant. Continuous data were presented as means and standard deviations (SDs), whereas categorical data were summarized as frequencies and percentages. In univariate analysis, the group differences were examined with the two-sample t test or the Wilcoxon rank-sum test for continuous variables and with the chi-square test or Fisher's exact test for categorical variables. The agreement between the mothers' and children's HBV genotypes was analyzed with McNemar's test and the κ statistic with the 95% confidence interval (CI). To assess the effect of immunization on HBV genotype distributions in children born to HBsAg-positive mothers, a multivariate logistic regression analysis that included gender, maternal age, delivery mode, and immunization as predicting variables was performed.
Demographic and Clinical Data
The demographic and clinical characteristics of 107 immunized cases with HBV breakthrough infection and 214 age-matched, unimmunized HBsAg carriers are shown in Table 1. In comparison with unimmunized carriers, more immunized cases with HBV breakthrough infection were born to HBsAg-positive mothers (65.9% versus 100%, P< 0.001).
Table 1. Clinical Characteristics of Unimmunized and Immunized HBsAg-Carrier Children
Age-Matched, Unimmunized HBsAg Carriers (n = 214)
Immunized Cases With HBV Breakthrough Infection (n = 107)
Abbreviation: SD, standard deviation; ULN, upper limit of normal.
Five immunized cases with HBV breakthrough infection who were born before July 1, 1984 were the participants in the pilot study of HBIG plus hepatitis B vaccine for the prevention of perinatally transmitted HBV.
For immunized cases with HBV breakthrough infection, maternal HBsAg was checked before the birth of the children; for unimmunized carriers, maternal HBsAg was checked at the time of the children's enrollment.
HBV Genotype Distribution, Immunization, and Maternal HBsAg Status
The HBV genotype distributions in 107 immunized cases with HBV breakthrough infection were as follows: genotype B, 61 (57%); genotype C, 45 (42.1%); and mixed genotypes (B and C co-infection), 1 (0.9%). In contrast, the HBV genotype distributions in 214 age-matched, unimmunized HBsAg carriers were as follows: genotype B, 177 (82.7%); genotype C, 35 (16.4%); and mixed genotypes, 2 (0.9%). The prevalence of genotype C was significantly higher in immunized cases with HBV breakthrough infection versus age-matched, unimmunized carriers (42.1% versus 16.4%, P< 0.001).
Among unimmunized children, those born to HBsAg-positive mothers (n = 141) and those born to HBsAg-negative mothers (n = 73) had similar HBV genotype distributions (P = 0.93). Figure 2 depicts the HBV genotype distributions in HBsAg-carrier children stratified by immunization and maternal HBsAg status.
All the mothers of immunized cases with HBV breakthrough infection were HBsAg-positive (Table 1). As for the 214 unimmunized carriers, maternal HBsAg-seropositive rates were comparable among children with different HBV genotypes, and the rates were 65.5% for genotype B infection, 68.6% for genotype C infection, and 50% for infection with mixed genotypes (B and C; P = 0.93).
Mothers' and Children's HBV Genotypes
Maternal blood samples for HBV genotyping were available for 82 of 107 immunized cases with HBV breakthrough infection and for 91 of 141 unimmunized HBsAg carriers whose mothers were HBsAg-positive. Those mothers with HBV viral loads lower than 103 copies/mL were excluded because of the limitations of the genotyping method.28 Table 2 shows the correlation of HBV genotypes in children and their HBsAg-positive mothers. A high degree of agreement was found between mothers' and children's HBV genotypes in both the unimmunized group (κ = 0.97, 95% CI = 0.90-1.00) and the immunized group (κ = 0.97, 95% CI = 0.92-1.00).
Table 2. Comparison of HBV Genotypes Between HBsAg-Carrier Children and Corresponding HBsAg-Positive Mothers
Unimmunized HBsAg Carriers With HBsAg-Positive Mothers (n = 72)†
Immunized Cases With HBV Breakthrough Infection (n = 74)†
B (n = 51)
C (n = 21)
B (n = 43)
C (n = 31)
Maternal blood was collected after each child's enrollment.
The list excludes 69 unimmunized and 33 immunized HBsAg-carrier children for whom the maternal blood was unavailable for HBV genotyping or the mother's HBV genotype could not be determined.
Secular Trend of the HBV Genotype Distribution
Because the hepatitis B immunization program launched on July 1, 1984 was a national program, most immunized cases with HBV breakthrough infection and unimmunized HBsAg carriers were in different birth cohorts (Table 1). Figure 3 shows the HBV genotype distributions in consecutive birth cohorts in unimmunized and immunized HBsAg-carrier children. For unimmunized children, HBV genotype distributions were comparable among different birth cohort groups (P = 0.391). Similarly, the genotype distributions in immunized HBsAg-carrier children of different birth cohorts were also comparable (P = 0.250).
With respect to the maternal HBV genotype distribution in the general population in the postimmunization era, among the 136 HBsAg-positive mothers delivering babies in 2007-2009, 95 had a viral load higher than 103 copies/mL, and the HBV genotype B-infected. Among the 95 mothers, 81.1% (77/95) were genotype B–infected, and 18.9% (18/95) were genotype C–infected; the genotype distribution was comparable to that in the unimmunized HBsAg-carrier children (P = 0.558).
Effect of Immunization on the HBV Genotype Distribution
Because of the high concordance between mothers' and children's HBV genotypes, we assumed that all children born to HBsAg-positive mothers acquired the virus from their mothers. Table 3 lists the clinical characteristics of HBsAg-carrier children born to HBsAg-positive mothers and stratified by their HBV genotypes. By using multivariate logistic regression analysis, we found that immunized HBsAg-carrier children born to HBsAg-positive mothers had a higher likelihood of genotype C infection than unimmunized children after adjustments for gender, maternal age, and delivery mode (odds ratio = 3.03, 95% CI = 1.62-5.65, P = 0.001; Table 4).
Table 3. Clinical Characteristics of HBsAg-Carrier Children Born to HBsAg-Positive Mothers
In this study, the HBV genotypes identified in both immunized and unimmunized children were highly consistent with the corresponding HBsAg-positive mothers. In addition, approximately two-thirds of unimmunized HBsAg-carrier children were born to HBsAg-positive mothers, whereas all immunized cases with HBV breakthrough infection were born to HBsAg-positive mothers, regardless of the HBV genotypes. These findings suggest that mother-to-child transmission plays an important role in HBV spread in Taiwan, particularly for immunized cases with HBV breakthrough infection. Our data further provide evidence supporting the idea that both genotypes B and C can be transmitted by maternal and horizontal routes; this is somewhat different from recent speculation that genotype C is most responsible for perinatal transmission and that other genotypes (A, B, D, and F) are mainly horizontally transmitted.21 Such speculation was raised because of the finding that HBeAg seroconversion in patients with genotype C occurred decades later than HBeAg seroconversion in patients with other genotypes.21 Genotype C–infected women were thus considered more likely to be HBeAg-positive during their childbearing years and infected their offspring at birth. Overall, whether different HBV genotypes have different transmission routes remains controversial, and further studies are needed to clarify this interesting and important issue.
It is known that universal immunization with hepatitis B vaccines and HBIG beginning at birth can result in a dramatic reduction of perinatal transmission of HBV.9 However, HBV breakthrough infection does happen on special occasions, and our data show that almost all immunized children with breakthrough infection contracted the virus from their carrier mothers. In contrast, only two-thirds of unimmunized HBsAg carriers acquired their infection from their mothers. These facts suggest that the current universal infant immunization program not only decreases perinatal infection but also reduces horizontal HBV infection in children. Nevertheless, HBV breakthrough infection through maternal transmission remains a challenge for the global control of HBV infection.31 Further studies to identify risk factors associated with perinatal/maternal infection despite complete immunization are required to implement a better prevention strategy for these high-risk infants.
The major finding of this study is an increased ratio of genotype C to genotype B in immunized children with HBV breakthrough infection in comparison with unimmunized HBsAg carriers. However, the increased genotype C to genotype B ratio was not seen in HBsAg-carrier mothers who delivered babies in 2007-2009 (i.e., the postimmunization era). This finding in mothers in the postimmunization era excludes the possibility that the changing HBV genotype distribution in children is related to a change in the maternal HBV genotype distribution. These lines of evidence imply that the changing HBV genotype distribution in immunized children with HBV breakthrough infection may be linked to the immunization program itself rather than a shift of genotypes in consecutive birth cohorts.
Perinatal transmission of HBV is related to the maternal viral load22-25 and the mode of delivery.32 However, in this study, maternal viral loads at the time of delivery were not available because we did not enroll the HBsAg-carrier children before or at birth. In addition to the gender of the children and the delivery mode, we used the maternal age, which was related to HBeAg seropositivity and viral loads, as a predicting variable in the multivariate logistic regression model to assess the effect of immunization on the HBV genotype distribution in HBsAg-carrier children born to carrier mothers. We did not investigate the details of the feeding practices because the breastfeeding of infants of chronic HBV carriers poses no additional risk for the transmission of HBV with appropriate immunoprophylaxis.33 After adjustments for other factors, immunized HBsAg-carrier children born to carrier mothers have a higher likelihood of genotype C infection than unimmunized children. Because the maternal HBV genotype distribution remained unchanged after the implementation of the immunization program, these data indicate that the rate of HBV breakthrough infection in immunized children born to genotype C mothers is higher than the rate in those born to genotype B mothers. A possible explanation is that immunization raises the threshold of the maternal viral load causing perinatal infection; thus, HBV genotype C, which is associated with higher viral loads, became predominant after the implementation of the immunization program. Because genotype C patients are known to exhibit more frequent hepatitis flares and are at greater risk of developing cirrhosis and HCC than genotype B patients,17-21 immunized children with HBV breakthrough infection (as observed in our cohort) may have a more progressive disease course that likely requires more intensive follow-up and active medical intervention in comparison with traditional, unimmunized HBsAg-carrier children. Although HBV genotype C prevails in eastern and southeastern Asia and the Pacific islands, it is not uncommon in immigrants from these areas in the United States, Europe, Australia, and New Zealand.34 In a globalizing world in which international migration and transition are frequent, this important finding is applicable not only in Taiwan but also in the rest of the world.
In summary, our results provide evidence that both HBV genotypes B and C can be transmitted from maternal and horizontal origins and that maternal transmission is responsible for most breakthrough infections in immunized HBsAg carriers. The secular trend toward an increased prevalence of genotype C indicates that immunized children born to genotype C mothers may have a higher rate of breakthrough infection than those born to genotype B mothers. These children with HBV breakthrough infection need careful, long-term follow-up, and the current immunization strategy may need modification to further eliminate the perinatal transmission of HBV.
The authors thank Pei-Jer Chen and Shiou-Hwei Yeh for their professional opinion and technical support. They also thank Hui-Chuan Lee, Pei-Lin Tsai, and Shih-Ting Chiu for their help with subject follow-up and administrative work. Laboratory work performed by Cheng-Lun Chiang and De-Shiuan Su is highly appreciated.