These authors contributed equally to this work.
Article first published online: 28 DEC 2012
Copyright © 2012 American Association for the Study of Liver Diseases
Volume 57, Issue 1, pages 37–45, January 2013
How to Cite
Wu, T.-W., Lin, H. H. and Wang, L.-Y. (2013), Chronic hepatitis B infection in adolescents who received primary infantile vaccination. Hepatology, 57: 37–45. doi: 10.1002/hep.25988
Potential conflict of interest: Nothing to report.
- Issue published online: 7 JAN 2013
- Article first published online: 28 DEC 2012
- Accepted manuscript online: 1 AUG 2012 07:35AM EST
- Manuscript Accepted: 17 JUL 2012
- Manuscript Received: 25 APR 2012
- National Science Council, Taiwan. Grant Number: NSC97-2314-B-715-001-MY3
- Mackay Medical College, Taiwan. Grant Number: 981B03, 991A01
Hepatitis B virus (HBV) infection is a global health issue. Universal infantile hepatitis B (HB) vaccination is very efficacious. However, HBV infections among those immunized subjects have been reported. The long-term efficacy of postnatal passive-active HB vaccination in high-risk subjects is not well explored. A total of 8,733 senior high school students who were born after July 1987 were assayed for hepatitis B surface antigen (HBsAg) and antibodies to HBsAg (anti-HBs). The overall HBsAg and anti-HBs-positive rates were 1.9% and 48.3%, respectively. The HBsAg-positive rate was 15% in HB immunoglobulin (HBIG) recipients (adjusted odds ratio [OR]: 15.63; 95% confidence interval [CI]: 10.99-22.22). Among students who did not receive HBIG, there was a significantly negative association between HB vaccination dosage and HBsAg-positive rate (P for trend = 0.011). Adjusted ORs for those who received 4, 3, and 1 to 2 doses were 1.00, 1.52 (95% CI: 0.91-2.53), and 2.85 (95% CI: 1.39-5.81), respectively. Among HBIG recipients, the HBsAg-positive rate was significantly higher in subjects with maternal hepatitis B e antigen (HBeAg) positivity and who received HBIG off-schedule. A booster dose of HB vaccination was administered to 1974 HBsAg- and anti-HBs-negative subjects. Prebooster and a postbooster blood samples were drawn for anti-HBs quantification. The proportions of postbooster anti-HBs titer <10 mIU/mL was 27.9%. Subjects with prebooster anti-HBs titers of 1.0-9.9 mIU/mL had significantly higher postbooster anti-HBs titers than those with prebooster anti-HBs titers of <1.0 mIU/mL (P < 0.0001). Conclusion: Having maternal HBeAg positivity is the most important determinant for HBsAg positivity in adolescents who received postnatal passive-active HB vaccination 15 years before. A significant proportion of complete vaccinees may have lost their immunological memories against HBsAg. (HEPATOLOGY 2013)
Hepatitis B virus (HBV) infection is a global issue, affecting two billion people in the world, with 360 million chronic carriers of hepatitis B surface antigen (HBsAg).1 The sequalae of chronic HBV infection, including hepatic failure, liver cirrhosis, and hepatocellular carcinoma, shorten lives and impose great economic burdens on society.
Taiwan has been an endemic area of HBV infection, with an HBV infection rate of 95% and a 15%-20% HBsAg carrier rate in the general population.2 Vertical transmission is the main cause of persistent HBV infection in Taiwan3; fortunately, it can be blocked by passive-active vaccination after birth.4-6 To control HBV infection, a hepatitis B (HB) vaccination program was launched in Taiwan in 1984, starting with newborns of highly infectious mothers, and expanded to all newborns in 1986.7
The remarkable effectiveness of universal infantile HB vaccination program (UIHBVP) is well documented.8-12 In Taiwan, the HBV infection and carrier rates of children born after the program has declined dramatically.10-12 The incidence of infantile fulminant hepatitis13 and childhood hepatocellular carcinoma has also declined.14, 15
However, the efficacy of the postnatal passive-active HB vaccination is not 100%. HBV chronic infections in immunized infants and children has been reported in Taiwan16-19 and other countries.20-23 Immunoprophylaxis failure occurred in a few high-risk infants even though HB immunoglobulin (HBIG) was administered immediately after delivery.4, 17, 18, 22, 23 The HBsAg-positive rates in high-risk infants age 0.5 to 2 years were 2%-9% after postnatal passive-active HB vaccination.17-23
Several studies have shown that the protective effects of HB vaccination can last for at least 10 years.24-28 However, the long-term protective effects of HB vaccination in high-risk subjects are not well explored. Additionally, it seems that the risks of HBV infection are increased dramatically during adolescence. The cumulative 10-year risk of HBV infection in postnatal passive-active HB vaccinated subjects was less than 15%.24, 25 On the contrary, more than 30% of these high-risk subjects were anti-HBc positive at 15 years of age.26
Here we report our recent study on HBV chronic infection among senior high school students who were born after the implementation of UIHBVP. We found that more than 15% of vaccinated students born to highly infectious mothers became HBsAg carriers. We also found that approximately one-sixth of complete vaccinees may have no or low immunological memories against HBsAg.
Subjects and Methods
Universal Infantile HB Vaccination Program (UIHBVP).
In July 1984, the Government of Taiwan initiated an HB mass vaccination program aiming primarily at interrupting mother-to-infant vertical transmission. High-risk infants were inoculated freely with a single dose of 145 IU of HBIG (Abbott Laboratories, Cutter, or Green Cross, Taiwan) within 24 hours after delivery. A 5 μg dose of a plasma-derived HBV vaccine (HavacB; Pasteur-Merieux, Lyon, France, or its equivalent derivative, Lifeguard hepatitis B vaccine, Hsin-Chu, Taiwan) was given to the infants at the first week of age. Two additional doses of HB vaccine were given at ages of 5 and 9 weeks. A booster dose of HB vaccination was given at 12 months of age.7 Recombinant HB vaccine was not available in Taiwan until July 1990. After 1 November 1992, the plasma-derived vaccine was replaced by recombinant yeast vaccines H-B-Vax II (Merck Sharp & Dohame, Rahway, NJ) and Engerix-B (SmithKline Beecham, Rixensart, Belgium).10
In July 1986, this program was extended to all neonates, then to preschool children in July 1987, to primary-school children in 1988, to middle-school children in 1989, and to adults in 1990.10 Since 1991, the vaccination log-sheet of first grade children were checked and unvaccinated or incompletely vaccinated children were given catch-up HB vaccination.10
Study Area. Hualien County is located in east Taiwan.
It has an area of 4,629 km2 and a population of 336,733 citizens in January 2012.29 Approximately 90% of the land area of Hualien County consists of mountains, in which the Central Mountain Chain are in the west and the East Coast Mountains are in the east. The population density for Hualien County and the whole Taiwan area is 72.8 and 641.9 persons per km2, respectively.29 Hualien has one city and nine rural and three mountainous townships. Fifty-six percent of the total population of Hualien County reside in Hualien City and Jian Township, both of which are located in northern Hualien. The prevalence rate of HBsAg in subjects born before 1978 was 17.0%,11 which was similar to that of the whole islands.2 There are seven public and six private senior high schools in Hualien County and the total number of freshmen is ∼4,000 annually.
There were two parts to the present study, including HBV seroepidemiology and HB revaccination study. The procedures of study subject recruitment of the seroepidemiological study were similar to our previous study.12 In brief, an invitation letter coupled with a short questionnaire and informed consent form was delivered to 10,648 freshmen of seven public senior high schools during October 2003 to May 2008. Consenting students were asked to provide their vaccination record and assayed for serum HBsAg and antibodies to HBsAg (anti-HBs).
Students with seronegativity for both anti-HBs and HBsAg were further invited to participate in the HB revaccination study. The study protocols were similar to our previous study, with little modification.30 In brief, a booster dose of 20 μg HB vaccine (Engerix-B, GlaxoSmithKline Biologicals, Rixensart, Belgium) was administered to each participant. One prebooster and one postbooster blood sample was obtained at the date and 4 to 6 weeks after the booster dose for anti-HBs quantification. All students were tested for serum antibody against HBV core antigen (anti-HBc). The study protocols were approved by the Institutional Review Board of Tzu Chi Medical Center and Taipei Medical University.
Records of Primary Infantile HB Vaccination.
Since 1983, the Department of Health of Taiwan implemented vaccination registration. A yellow vaccination log-sheet was provided to each woman who gave birth recently. The health institutions were obligated to register the dates and associated information on the yellow log-sheet. In the study, consenting students were asked to provide a copy of their vaccination record. To increase the completeness of the vaccination data, we also reviewed and abstracted HB vaccination records from 13 local health stations and major healthcare institutions in Hualien County. For those who had two sources of HB vaccination data, the copies that were provided by the study subjects were used as the primary source. Records of maternal HBeAg status for HBIG recipients were also reviewed and abstracted.
Definitions of Attributes of and Responses to HB Vaccination.
All the study subjects were born between July 1987 and July 1991. During this period, only high-risk infants who were born to mothers with HBeAg positivity or reciprocal serum titers of HBsAg higher than 2,560 were administered a single dose of HBIG and the recommended dates of HB vaccination were the 1st, 5th, and 9th weeks after delivery.30 In the present study, infantile and early childhood HB vaccination was defined as those administered before the age of 3 years. Completion of HB vaccination was defined as subjects who received at least 3 doses of HB vaccine before 12 months of age. On-schedule HB vaccination was defined as follows: (1) the first dose was administered at the first week after delivery; (2) the second dose was administered at 4 to 6 weeks of age; and (3) the third dose was administered at 8 to 10 weeks of age.
The response rate to a booster dose of HB vaccine was defined as the proportion of booster recipients whose postbooster titer was ≥10 mIU/mL.30, 31 The overall anti-HBs seropositive rate, designated as PRT, after a booster dose of HB vaccine was approximated by the following formula:
Where PR1 is the anti-HBs seropositive rate before booster and PR2 is the response rate in the booster recipients.
HBsAg and anti-HBs serostatus in blood samples obtained for the seroepidemiological study were detected by microparticle enzyme immunoassays (MEIA) with commercial kits AxSYN HBsAg (V2) (Abbott Diagnostics, North Chicago, IL) and anti-HBs (Abbott Diagnostics), respectively. Anti-HBc serostatus was also detected by MEIA with the commercial kit AxSYN CORE (Abbott Diagnostics). Serum anti-HBs titers in pre- and postbooster blood samples were assayed by a quantitative method with the commercial kit AxSYN AUSAB (Abbott Diagnostics). The detection limit of this quantitative method was 1.0 mIU/mL.
Pearson's chi2 test and Fisher's exact test were used to assess whether significant differences exist in the positive rate of HBsAg among students with different attributes. P 0.05 or less was regarded as statistically significant. Multivariate logistic regression models were used to estimate the adjusted odds ratio (aOR) and the 95% confidence interval (95% CI) of being HBsAg-positive for potential determinants.
A total of 8,813 (83%) freshmen were recruited. Eighty students who were born before July 1987 or did not have blood drawn were excluded, leaving a total of 8,733 individuals in the seroepidemiological study. The mean age with standard deviation at enrollment was 15.6 (0.4) years; 52.8% were male. The seropositive rate of HBsAg and anti-HBs were 1.9% and 48.3%, respectively.
Histories of infantile HB vaccination were not available for 819 (9.4%) subjects. Approximately 87% of subjects had official documentation of receiving ≥3 doses of HB vaccination before 3 years of age. Three hundred and eighty-one (4.8%) subjects had documented HBIG injection. Records of maternal HBeAg status of HBIG recipients were left blank in 211 (55.4%) subjects. Among 7,914 subjects who had official documentation of HB vaccination, 6,804 of them had data from two different sources. The consistency in the dates of HB vaccination was 94.5%.
To assess the effects of HB vaccination on HBV chronic infection, we restricted analyses to 7,914 subjects who had official documentation of HB vaccination. Table 1 shows that there was a significantly negative association between HB vaccination dosage and HBsAg-positive rate (P for trend = 0.0085). As compared with those who received four doses of HB vaccination, adjusted ORs of HBsAg positivity were 1.40 (95% CI: 0.92-2.14) and 2.38 (95% CI: 1.74-4.84) for those who received 3 and 1 to 2 doses, respectively. The HBsAg-positive rate was significantly higher in HBIG recipients than that of subjects without HBIG injection (1.3% versus 15.4%; adjusted OR = 15.63, 10.99-22.22).
|HBsAg+||Crude OR||Adjusted OR|
|Total||n||(%)||OR||95% CI||aOR*||95% CI|
|HB vaccination during infancy and early childhood†|
Table 2 shows that among 7,537 students who did not receive HBIG, there was a significantly negative association between HB vaccination dosage and HBsAg positive rate (P for trend = 0.011). As compared with those who received four doses of HB vaccination, the adjusted ORs were 1.52 (95% CI: 0.91-2.53) and 2.85 (95% CI: 1.39-5.81) for those who received 3 and 1 to 2 doses, respectively. Among 377 HBIG recipients, no significant association was observed.
|Without HBIG||With HBIG|
|Total||n||(%)||aOR*||95% CI||Total||N||(%)||aOR*||95% CI|
|HB vaccination during infancy and early childhood†|
Table 3 shows the HBsAg-positive rates and timing of HBIG injection and maternal HBeAg serostatus among 373 passive-active HB vaccination recipients. The HBsAg-positive rates were nonsignificantly lower for subjects who received HBIG injection at the date of birth than that of the next date after delivery (14.6% versus 19.4%, one-sided P = 0.14). Among 162 subjects with documented maternal HBeAg status, the HBsAg-positive rate was significantly higher for HBeAg seropositivity than that of seronegativity (17.8% versus 11.1%, one-sided P = 0.014).
|Date of HBIG dosing||0.60|
|Date of birth||294||43||(14.6)|
|The next date||72||14||(19.4)|
|After the next date||6||1||(16.7)|
|Maternal HBeAg status||0.60|
|HBIG at date of birth||101||15||(14.9)†|
|HBIG at the next date||31||9||(29.0) †|
|HBIG at the other dates||3||0||(0.0)|
Among 135 students with maternal HBeAg seropositivity, 101 and 31 of them received HBIG injection at the date of birth and the next day after delivery, respectively. The HBsAg-positive rate was significantly lower for those dosed at the date of birth than that of those receiving a dose on the next day (14.9% versus 29.0%, one-sided P = 0.032).
Among 4,364 HBsAg- and anti-HBs-negative students, 2,057 of them were recruited in the HB revaccination study. There was no significant difference in the HB vaccination histories between responders and nonresponders. Twenty-two participants were anti-HBc-positive and another 35 participants with prebooster anti-HBs titers ≥10 mIU/mL were excluded. Additionally, 26 participants who failed to provide postbooster blood samples were further excluded, leaving a total of 1,974 participants in the revaccination study.
The proportions of postbooster anti-HBs titers <10, 10-99, and ≥100 mIU/mL were 27.9, 29.7 and 42.4%, respectively (Table 4). There was a significant association between pre- and postbooster anti-HBs titers (P < 0.0001). Among 1,252 booster recipients who had prebooster anti-HBs titers of <1.0 mIU/mL, the postbooster anti-HBs titers <10, 10-99, and ≥100 mIU/mL were 40.6, 31.6, and 27.8%, respectively. The corresponding figures in 722 booster recipients who had prebooster anti-HBs titers of 1.0-9.9 mIU/mL were 5.8, 26.6, and 67.6%, respectively.
|Postbooster anti-HBs levels (mIU/mL)|
|Prebooster anti-HBs levels (mIU/mL)||n||<10||10-99||100-999||≥1,000|
The anti-HBs seropositive rate before booster was 48.3% (PR1) and the response rate in the booster recipients was 72% (PR2). The overall anti-HBs seropositive rate after a booster dose of HB vaccine was estimated to be 84.3%.
In the present study we found that the HBsAg seropositive rate in high-risk subjects who received HBIG injection 15 years ago was as high as 15%. The HBsAg seropositive rates were much higher in subjects with maternal HBeAg-positivity and received HBIG injection off-schedule. To the best of our knowledge, there is no research reporting the cumulative risk of 10 years or more of HBV chronic infection in subjects who received an HBIG injection in infancy.
It has been suggested that mother-to-infant transmission was the primary route of HBV infection in Taiwan.3 Approximately 90% of infants of HBeAg-positive mothers were persistently positive for HBsAg.4, 5 Although postnatal passive-active HB immunization was proven to be very efficacious, nonetheless persistent HBV infection in immunized subjects were observed in Taiwan16-18 and other countries.20-23 Moreover, a prospective study enrolled 665 high-risk newborns who were immunized with HBIG within 24 hours after delivery and subsequently received ≥3 doses of HB vaccination.19 Sixteen of the cohort members were positive for HBsAg within 1 hour after delivery and remained HBsAg-positive at 6 months of age.19 This evidence indicated that immunoprophylaxis failure occurred in a few high-risk infants even though HBIG was administered immediately after delivery. In Taiwan, only high-risk infants whom were born to women with HBeAg positivity or high levels of HBsAg were eligible for HBIG injection. It is reasonable to hypothesize that the observed high HBsAg seropositive rate in HBIG recipients in the present study was associated with high maternal HBV infectiousness.
It was recommended that passive-active HB vaccination should be administered as soon as possible after birth.1 In the present study, around 25% of all HBIG recipients and 22% of HBeAg positive recipients did not received HBIG at the date of birth. Statistically, those who did not receive HBIG injection at the date of delivery had a higher HBsAg seropositive rate. Consequently, off-schedule HBIG injection could possibly contribute to the observed significantly higher HBsAg-positive rate in the present study.
HBV transmission during early childhood is another major determinant of countries with HBsAg hyperendemicity.1 Herd immunity induced by mass or nation-wide vaccination programs is well documented.32-34 The HBsAg seropositive rates in newborns with maternal HBeAg-positivity and who had received passive-active HB immunization were ∼4%-6% at the age of 1 year.4, 16, 17 Among infants who were born to women with low or unknown levels of HBsAg, the HBsAg-positive rate at 15 months of age were 2.5% and 5.5% for subjects who had received HB vaccines on- and off-schedule, respectively.16 A more recent population study of passive-active HB-immunized infants reported that 15 (4.2%) of 357 subjects with maternal HBeAg-positivity were HBsAg-positive at 2 years of age.18 Among infants who were born by HBeAg-negative carriers and noncarriers, the corresponding figures were 0.8% and 0.2%, respectively.18 This evidence suggested that an HB vaccination program significantly decreased the risk of HBV infection during infancy and early childhood in the low- and high-risk populations in Taiwan. All subjects in the present study were born after the implementation of UIHBVP, therefore, the attributable risk for HBV infection during infancy and early childhood in HBsAg chronic infection should not be too high.
HBV reactivation or infection in later life, especially in those who have low or no immunological memories against HBsAg, is the major determinant of HBV chronic infection in low HBsAg endemic countries1 and should be taken into consideration. HBV reactivation in transplant recipients is well documented.35-37 A long-term prospective study of HB vaccination and risk of HBV infection in infants born to HBeAg-positive mothers showed that the natural infection rate of HBV before 3 years of age was 12%.17 Additionally, during the 5-year follow-up period, none of the anti-HBc-positive infants became HBsAg-positive.17 In another prospective study of 805 infants who were uninfected and anti-HBs-positive by 1 year of age, 52 of them became HBV-infected at the age of 1 to 5 years. An additional 61 subjects became HBV-infected at the age of 5 to 10 years. The cumulative 10-year risk of HBV infection was ∼15%.24 Additionally, among 113 HBV-infected subjects, four became chronically infected.24 In a more recent study, the anti-HBc and HBsAg-positive rates were 12% and 0%, respectively, in children aged 5 to 10 years.25 This evidence indicated that even in high-risk children the cumulative 10-year risk of HBV reactivation or infection is moderate and the risk of becoming an HBV carrier is low. These findings were supported by the evidence that the protective effects of HB vaccination can last for at least 10 years.27, 28 However, whether protective effects of HB vaccination can persist for much longer in high-risk subjects as well as the risk of becoming an HBV carrier in subjects who contracted HBV infection after 10 years of age are both far from being investigation.
In the present study, the overall anti-HBs seropositive rate after a booster dose of HB vaccine was estimated to be 84.3%. It was suggested that an anti-HBs titer of ≥10 mIU/mL is necessary for the prevention of HBV infection.38 Therefore, about one-sixth of subjects who had received infantile HB vaccination may have no or low immunological memories against HBsAg at 15 years of age and subsequently had elevated risk of HBV chronic infection when exposed to HBV.
A recent study assayed the HBV markers in a cohort of 78 adolescents with maternal HBeAg positivity who had developed protective levels of anti-HBs during infancy. The anti-HBc-positive rate of the cohort member was 33.3% at 15 years of age.26 One child became HBsAg-positive at the age 7 to 15 years old.26 On the other hand, the cumulative risks of HBV infection in high-risk subjects younger than 10 years of age were less than 15%.24, 25 However, whether the risks of HBV infection or reactivation in high-risk subjects increase significantly at 10 years or more after infantile HB vaccination needs further investigation.
In the present work, we found that a significant proportion of adolescents who were born to high-risk mothers and received postnatal passive-active HB immunization were HBsAg-positive. It seems that HBV infection or reactivation in adolescents with low or no immunological memories against HBsAg and mother-to-infant HBV transmission were the primary determinants. HBsAg chronic carriers were at significantly higher risks of endstage liver diseases.39, 40 Our findings indicate that more efforts should be devoted to further reduce HBV transmission. Booster HB vaccination in subjects aged 15 years or more should be considered, especially in subjects born to HBsAg-positive mothers or who had a high risk of HBV exposure. Additionally, it is now routine practice to use antiretroviral therapy in pregnant women with human immunodeficiency virus infection to prevent mother-to-infant transmission.41, 42 Preliminary studies have shown that lamivudine43, 44 and telbivudine45 during pregnancy may reduce vertical transmission of HBV. The routine use of anti-HBV treatment during pregnancy for those with high viral load awaits further study to confirm efficacy and safety.
There were concerns that needed to be addressed. Undetermined accuracy of the HB vaccination histories could result in information bias. Before the implementation of the National Health Insurance Program in March 1995 in Taiwan, the main provider of primary healthcare was the local health station. Hualien County is an area with limited healthcare resources. It was reasonable to hypothesize that most of the infants received vaccination in nearby local health stations and their vaccination data were recorded immediately in the yellow log-sheet by well-trained health professionals. Our estimates were comparable to previous research. The population-based study, which selected a random sample of 1,500 children born September 1986 to August 1987 from the whole Taiwan area, found that 87.2% received ≥3 doses of HB vaccination.46 In the present study, 4.8% of study subjects had documentation of HBIG injection, which is also similar to previous research.14, 18 More important, among subjects who had vaccination data from two different sources, the consistency in dates of HB vaccination was as high as 94.5%. We abstracted records of HB revaccination without knowing the subjects' HBsAg serostatus. Consequently, information bias should be minimal and nondifferential in the present study.
Additionally, less than half of the HBsAg- and anti-HBs-negative students were recruited in the HB revaccination study and less than 45% of the HBIG recipients had known maternal HBeAg status. We found that there was no significant difference in the HB vaccination histories between responders and nonresponders of the HB revaccination study (HB vaccination ≥3 dosages: 82.8% versus 84.4%; complete rate: 75.3% versus 77.3%) and between HBIG recipients with known and unknown maternal HBeAg status (HB vaccination ≥3 dosages: 97.5% versus 98.1%; complete rate: 89.8% versus 90.7%). It is therefore believed that bias associated with nonresponse was minimal in the HB revaccination study.
In conclusion, having a high-risk mother is one of the most important determinants for HBsAg positivity at adolescence. A significant proportion of adolescents who had received primary infantile HB vaccination may have lost their immunological memories against HBsAg. Routine use of anti-HBV treatment during pregnancy might be an option, if safety and efficacy are proven by further large-scale studies. Booster HB vaccination in subjects aged 15 years or more should be considered, especially in subjects born to HBsAg-positive mothers or who have a high risk of HBV exposure.
We thank the staffs of the health stations in Hualien County and the Center of Disease Control, Department of Health, Taiwan, for valuable contributions.
- 1World Health Organization (WHO). Hepatitis B vaccine. WHO Weekly Epidemiological Record 2009; 84: 405-420.
- 2Epidemiological study on hepatitis B virus infection in Taiwan. Chin J Gastroenterol 1984; 1: 1-9., , , , , , et al.
- 29Ministry of Interior. Statistical Yearbook of Interior. Republic of China: Ministry of Interior, Executive Yuan; 2011. Available at: http://statis.moi.gov.tw/micst/stmain.jsp?sys=100. Accessed 20 April 2012.
- 33Vaccine herd effect. Scand J Infect Dis 2011; 43: 683-689., , .