Effect of the interval between birth and second dose of hepatitis B vaccine on perinatal transmission of hepatitis B virus

ABSTRACT

Preventing perinatal transmission is important for hepatitis B (HepB) elimination. We conducted a retrospective cohort study to assess the interval between HepB birth-dose (HepB-BD) to second-dose (HepB-SD) vaccination on perinatal transmission. Among 39,313 infants born to HepB s-antigen (HBsAg)-positive mothers from a Korean national database 38,411 (97.7%) had completed timely immunophylaxis with HepB-BD 41,572 (99.8%) with hepatitis B immune globulin, and 1027 (2.6%) were HBsAg-positive at ≥ 9 months. Maternal factors (i.e. HepB e-antigen status, age, or nationality) were associated with an increased risk of infection whereas short gestational length decreased it. The HepB-BD – HepB-SD interval (<8 vs. ≥8 weeks) did not alter the risk.

KEYWORDS:

• Hepatitis B vaccine
• mother-to-child transmission
• hepatitis B virus
• schedule
• interval
• immunoprophylaxis

Background

Infants born to mothers with chronic hepatitis B virus (HBV) infection can acquire HBV infection, which usually occurs at the time of delivery. In most countries, especially those with high HBV endemicity, this is the most important mode of HBV transmission. One of the World Health Organization (WHO) targets toward the elimination of chronic hepatitis B (HepB) is to reduce vertical transmission by achieving HepB surface antigen (HBsAg) seropositivity rates ≤0.1% in those 5-years-old and under by 2030.^1,2 Previous studies demonstrated that maternal hepatitis B virus e-antigen (HBeAg)-seropositivity and higher maternal viral load were the most significant risks for perinatal HBV transmission, with the risk of transmission as high as 90%.³ In order to decrease this risk, it is recommended to start post-exposure immunoprophylaxis (PEI) with hepatitis B immune globulin (HBIg) and a HepB birth-dose vaccine (HepB-BD) in infants born to HBV-infected mothers within 24 h of birth.⁴ Following PEI, at least two additional doses of the HepB vaccine are administered to complete the vaccination series.

To increase access to measures to prevent perinatal HBV transmission, the Korean Disease Control and Prevention Agency (KDCA) launched the Perinatal Hepatitis B Prevention Programme (PHBPP) in July 2002.⁵ The programme consists of: (1) HBIg and HepB-BD within 12 h of birth; (2) a second dose of HepB vaccine (HepB-second dose [SD]) at 1 month of age and a third dose of HepB vaccine (HepB third dose [TD]) at 6 months of age; and (3) a post-vaccination serologic test (PVST) for HBsAg and hepatitis B surface antibody (anti-HBs) between 9 and 15 months of age.

Although there is a recommended interval between doses for the currently available HepB-containing vaccines, data are insufficient to determine whether outcomes differ among infants with perinatal exposure to HBV when the interval between the birth dose and the next dose of HepB is 4 weeks or ≥8 weeks.⁴ Many countries recommend HepB-SD following an interval of either six weeks or two months after birth whereas some countries in Asia, including mainland China, Taiwan, and the Republic of Korea (ROK), recommend HepB-SD at one month (or 4 weeks) after birth.^6,7 In order to reduce the number of visits and increase timely vaccination coverage, most countries give the HepB vaccine doses after birth at the same time as Diphtheria-Tetanus-Pertussis-containing vaccines: either at 6, 10, and 14 weeks of age, or at 2, 4, and 6 months. Moreover, in low- and middle-income countries, vaccination sessions are not daily but periodic because many national immunization programmes are heavily reliant on outreach or mobile immunization activities.

We conducted a retrospective cohort study using a large database of infants born to HBV-infected mothers to compare the outcome of perinatal transmission by subgroup in relatively high numbers of eligible infants in the ROK between 2002 and 2013.⁵ Our aim was to compare perinatal transmission of HBV according to the time interval between HepB-BD and HepB-SD.

Methods

This study used an administrative de-identified dataset developed during a previous study.⁵ In summary, the KDCA receives data regarding PHBPP from Public Health Centres responsible for collecting relevant information from participating health facilities. The dataset contains information from both infants and mothers on HepB serostatus and services provided. The vaccination status of each infant was additionally retrieved from the Integrated Health Information System. Full details are described in the earlier publication.⁵ As the de-identified dataset was developed as part of a programme impact assessment and quality assurance, according to Korean legislation and rules, institutional review board approval was not required.

Inclusion criteria for this study were infants who: 1) were registered to PHBPP between July 2002 and December 2013; 2) were born to HBsAg-positive mothers; 3) received at least three doses of HepB vaccine including their birth dose; 4) had at least one PVST at 9 months of age or older; and 5) with known status for the following variables: maternal HBsAg and HBeAg serostatus, maternal age, maternal nationality, infant sex, gestational length, birthweight, delivery type, and timing of administration of HepB vaccines (HepB-BD, -SD and -TD), and HBIg, if administered. Infants were excluded from the study if any of the following applied: born at <24 weeks or >42 weeks of gestation; premature (gestational length <37 weeks) with a birthweight of <2000 g; or birthweight <500 g.

A descriptive analysis was conducted with independent variables (i.e., maternal age, maternal nationality, maternal HBeAg serostatus, sex of infants, gestational length, infant birth weight, delivery type, timing of HepB-BD, interval between HepB-BD and HepB-SD, interval between HepB-SD and HepB-TD, and administration of HBIg) against the result of infant HBsAg status, as determined at nine months of age or later as recommended by KDCA. The analysis was further stratified by the interval between HepB-BD and HepB-SD of <8 weeks and ≥8 weeks. To determine the association between the vaccination interval and perinatal transmission, univariate analyses were first done quantifying unadjusted odds ratios assessed by χ²-test. The association between the outcome and the exposure was assessed and those variables related with both were included in the multivariable analysis. These variables included maternal age, nationality of the mother, maternal HBeAg-serostatus, infant sex, gestational age, and delivery type. Infant birth weight, which is highly correlated with gestational age, was excluded in the analysis.

For the multivariate analysis, a logistic regression was done using infant HBsAg-serostatus as a dependent variable, timing of the second dose as the main independent variable and the aforementioned co-variates. In this analysis, infants with the interval between HepB-BD and HepB-SD <4 weeks or those without a history of HBIg at birth were excluded as the numbers included were too small to be compared.

Statistical analyses were conducted using R software version 3.5.1 (R Project for Statistical Computing, Vienna, Austria).

Results

From July 2002 to December 2013, 39313 infants registered with the PHBPP met the inclusion criteria and did not meet the exclusion criteria of this study. Among those, 1027 (2.6%) infants were HBsAg-positive when tested at 9 months of age or older (Table 1). Most of the study population received HepB-BD on the day of birth (38,411 [97.7%]), HepB-SD 4–<8 weeks after HepB-BD (38,327 [97.5%]), and HepB-TD ≥20 weeks after HepB-SD (38,762 [98.6%]). Almost all of the study population received HBIg (39,242 [99.8%]) – Day 0 (38,264 [97.5%]), Day 1 (453 [1.2%]), and Day 2+ (525 [1.3%]) (Range: [2, 724], Median: [6], IQR: [3─12]).

Table 1. Bivariate analysis on factors associated with immunoprophylaxis failure among infants born to HBV-infected mothers, stratified by the interval between HepB-BD and HepB-SD, Republic of Korea, 2002–2013.

	Infants, n				Interval HepB-BD – HepB-SD 4–<8 weeks				Interval HepB-BD – HepB-SD ≥8 weeks
		HBsAg-positive infants			Infants, n	HBsAg-positive infants			Infants, n	HBsAg-positive infants
		n	%	p-value		n	%	p-value		n	%	p-value
Overall	39,313	1,027	2.6		38,907	1,013	2.6		406	14	3.4
Maternal age
<25 years	2,451	135	5.5	<.001	2,415	133	5.5	<.001	36	2	5.6	.636
25–29 years	14,906	419	2.8		14,774	416	2.8		132	3	2.3
30–34 years	17,091	368	2.2		16,906	362	2.1		185	6	3.2
≥35 years	4,865	105	2.2		4,812	102	2.1		53	3	5.7
Maternal nationality
Korean	37,621	932	2.5	<.001	37,250	919	2.5	<.001	371	13	3.5	1.000
Non-Korean	1,692	95	5.6		1,657	94	5.7		35	1	2.9
Maternal HBeAg status
HBeAg-negative	25,106	195	0.8	<.001	24,854	192	0.8	<.001	252	3	1.2	.002
HBeAg-positive	14,207	832	5.9		14,053	821	5.8		154	11	7.1
Sex of infants
Female	19,234	486	2.5	.312	19,038	477	2.5	.254	196	9	4.6	.287
Male	20,079	541	2.7		19,869	536	2.7		210	5	2.4
Gestational length
24–36 weeks’ gestation	1,811	34	1.9	.048	1,789	33	1.8	.043	22	1	4.5	1.000
37–42 weeks’ gestation	37,502	993	2.6		37,118	980	2.6		384	13	3.4
Infant birth weight
<2,000 g	39,295	1025	2.6	.071	17	2	11.8	.068	1	0	0.0	1.000
≥2,000 g	18	2	11.1		38,890	1011	2.6		405	14	3.5
Delivery type
Vaginal delivery	14,198	338	2.4	.035	24,856	679	2.7	.044	259	10	3.9	.582
Caesarean section	25,115	689	2.7		14,051	334	2.4		147	4	2.7
Timing of HepB-BD
Administered at Day 0	38,411	1004	2.6	1.000	38,014	991	2.6	.956	397	13	3.3	.273
Administered at Day 1	429	11	2.6		424	10	2.4		5	1	20.0
Administered at Day 2+	473	12	2.5		469	12	2.6		4	0	0.0
Timing of HepB-SD after HepB-BD
<4 weeks	580	21	3.6	.315	N/A				N/A
4–<8 weeks	38,327	992	2.6
8–<12 weeks	301	11	3.7
≥12 weeks	105	3	2.9
Timing of HepB-TD after HepB-SD
<4 weeks	5	0	0.0	.228	2	0	0.0	.731	3	0	0.0	.364
4–<20 weeks	546	20	3.7		330	10	3.0		216	10	4.6
≥20 weeks	38,762	1007	2.6		38,575	1003	2.6		187	4	2.1
HBIg
No	71	2	2.8	1.000	67	2	3.0	1.000	4	0	0.0	1.000
Yes	39,242	1025	2.6		38,840	1011	2.6		402	14	3.5

Abbreviations: HepB-BD, hepatitis B birth-dose vaccine; HepB-SD, hepatitis B second-dose vaccine; HepB-TD, hepatitis B third-dose vaccine; HBeAg, hepatitis B e antigen; HBIg, hepatitis B immune globulin; HBsAg, hepatitis B surface antigen; N/A, not applicable.

Data excluded infants with interval of HepB-BD – HepB-SD <4 weeks and infants who were tested for HBsAg before 9 months after birth.

In the univariate analysis, infants born to HBsAg-positive mothers had a higher risk of testing positive for HBsAg at ≥9 months of age when the mother was HBeAg-positive (crude odds ratio [OR], 7.95; 95% confidence interval 6.79–9.30), non-Korean (2.34; 1.89–2.91), or had a maternal age of <25 years (2.64; 2.04–3.43 vs. ≥35 years) or 25–29 years (1.31; 1.06–1.63 vs. ≥35 years), or if the infant was born by vaginal delivery (1.16; 1.01–1.32) or had low birth weight (500 to <2000 g)(4.67; 1.07–20.32). The risk was lower when delivered early (24–36 weeks gestational length; 0.70; 0.50–0.99; p < .05 for all) (Table 2). An interval between HepB-BD and HepB-SD of 4–<8 weeks or ≥8 weeks (Range: [8.0, 526.4], Median: [9.4], IQR: [8.6–12.0]) did not alter risk of the infant of testing positive for HBsAg (0.75; 0.44–1.28).

Table 2. Univariable and multivariable analyses for factors associated with the risk of perinatal HBV infection among infants born to HBV-infected mothers, Republic of Korea, 2002–2013.

	Univariate analysis			Multivariate analysis
	Crude OR	(95% CI)	p-value	Adjusted OR	(95% CI)	p-value
Second dose schedule <8 weeks (vs. ≥8 weeks)	0.75	(0.44–1.28)	.290	0.79	(0.46–1.37)	.405
Maternal age (vs. ≥35 years)
<25 years	2.64	(2.04–3.43)	<.001	1.52	(1.14–2.02)	.004
25–29 years	1.31	(1.06–1.63)	.014	1.06	(0.85–1.32)	.591
30–34 years	1.00	(0.80–1.24)	.983	0.91	(0.73–1.14)	.413
Non-Korean mother (vs. Korean mother)	2.34	(1.89–2.91)	<.001	1.56	(1.22–2.01)	<.001
Maternal HBeAg-positive (vs. maternal HBeAg-negative)	7.95	(6.79–9.30)	<.001	7.66	(6.54–8.98)	<.001
Male infant (vs. female infant)	1.07	(0.94–1.21)	.298
Short gestational length (24–36 weeks)(vs. 37–42 weeks)	0.70	(0.50–0.99)	.046	0.68	(0.48–0.96)	.029
Low birth weight (500–<2,000 g) (vs. ≥2,000 g)	4.67	(1.07–20.32)	.040
Vaginal delivery (vs. Caesarean section)	1.16	(1.01–1.32)	.030	0.94	(0.82–1.08)	.382
Timing of BD (vs. Day 0)
Day 1	0.98	(0.54–1.79)	.949
Day 2+	0.97	(0.54–1.73)	.917
Timing of HBIg (vs. Day 0)
Day 1	0.93	(0.51–1.69)	.803
Day 2+	0.95	(0.54–1.64)	.842
Never	1.08	(0.26–4.41)	.916

Abbreviations: CI, confidence interval; HepB-BD, hepatitis B birth-dose vaccine; HepB-SD, hepatitis B second-dose vaccine; HepB-TD, hepatitis B third-dose vaccine; HBeAg, hepatitis B e antigen; HBIg, hepatitis B immune globulin; HBsAg, hepatitis B surface antigen; OR, odds ratio.

The results from the multivariate analysis were similar to those from the univariate analysis; positive maternal HBeAg (adjusted OR 7.66; 6.54–8.98), a non-Korean mother (1.56; 1.22–2.01) a maternal age of <25 years old vs. ≥35 years (1.52; 1.14–2.02) increased the risk of the infant testing positive for HBsAg at ≥9 months of age, whereas short gestational length (0.68; 0.48–0.96) decreased this risk. The timing of HepB-BD (Day 0 vs. Day 1 or Day 2+) or assessing this as a continuous variable, did not alter the risk of perinatal HepB transmission (data not shown). The interval between HepB-BD and HepB-SD of 4–<8 or ≥8 weeks again did not alter the risk of infection (0.79; 0.46–1.37). Similar results were seen for a 6-week interval between HepB-BD and HepB-SD with even a smaller difference (unadjusted OR 1.01; 0.71–1.48).

Discussion

This study demonstrated, using robust data from a large cohort, that there is low likelihood that an interval between HepB-BD and HepB-SD of ≥8 weeks is associated with higher perinatal HBV transmission compared with an interval of 4–<8 weeks, provided that HepB-BD and HBIg are given in timely fashion. During the time period investigated, there was still significant transmission in those born to HBeAg-positive mothers regardless of vaccination schedule, and despite the almost universal use of HBIg. This suggests that other interventions, such as maternal use of antiviral therapy, in addition to HBIg and HepB vaccination should be routinely considered for this population to further reduce the risk. We have discussed potential reasons for different risk factors and relevant public health implications in the separate article that analyzed the same dataset, aiming to provide a more comprehensive review.⁵

In a prior study,⁵ a similar pattern was observed in the relationship between gestational age and HBsAg-positivity of infants when stratified by HBeAg status of their mothers. No significant differences in transmission rates were observed among infants born to HBeAg-negative mothers across gestational age groups, but infants born to HBeAg-positive mothers showed increased transmission rates with longer gestation. These findings support intrauterine transmission as the likely cause of remaining HBsAg-positive cases, suggesting that shorter gestational length may reduce transmission rates.

The evidence for the optimal HepB-BD – HepB-SD interval has been scarce, with conflicting results between studies. Some evidence indicates that there is no effect on the timing of the second dose, which is further supported by the results found in this study. In a study conducted in the United States among infants born to HBsAg-positive mothers of Asian-American ethnicity receiving different types of HepB vaccines with HBIg, infants receiving the first 2 doses at 0 and 1 month had an incidence of HBV infection of 8.4% (95% CI 6.2–10.6%) after 18-month follow-up.⁸ This was comparable with an incidence of HBV infection of 6.9% (2.5–11.3%) in those who received the first 2 doses at 0 and 2 months with a univariate OR of 0.77 (0.39–1.51). In contrast, a study from Thailand suggested a HepB-BD – HepB-SD interval of over 10 weeks increased the risk of chronic HBV infection in infants without HBIg administration born to HBV-infected mothers by 3.74-times (0.97–14.39), but without statistical significance due to the small sample size.⁹

The current study has limitations. Although this is the largest analysis to date assessing infant PEI and subsequent HepB vaccination using a robust database, a small number of infants received a HepB-SD vaccine later than 8 weeks of age and even fewer who received it later than 12 weeks of age. Thus, this reduced the power of the study to determine significant differences at the longer intervals between HepB-BD and HepB-SD. Furthermore, registry data cannot identify if there was an intrinsic or practical reason for the small subset of infants who received PEI and subsequent vaccination outside of the recommended schedule. Also, as this was a retrospective analysis, it was not possible to obtain additional data than originally captured. For example, use of antiviral therapy, which reduces a risk of transmission, was not captured, although there is no reason to believe there would be a different use in one group over the other.¹⁰ Despite these limitations, to our knowledge, this is the largest study of its kind providing important public health information showing no differences with the timing of the HepB-SD in a context of very high timeliness of HBIg and HepB-BD.

In conclusion, our study suggests that a prolonged interval between HepB-BD and HepB-SD may not affect the failure rate of immunoprophylaxis in a setting of high uptake of HBIg, timely HepB-BD and a high completion rate of primary-series HepB vaccination. This finding can potentially help countries and healthcare professionals move toward a more unified pediatric schedule with the administration of HepB vaccines at the same time as other routine vaccines, which could improve timely vaccination and compliance. Further register-based studies may be needed to confirm this finding.

Author contributors

JCVZ and TUY conceived the study; TUY, JCVZ, HAP, and YJ designed the study; HAP did the data analysis; TUY, JCVZ, HAP, CWJ, and DK contributed to the interpretation of the data; TUY and JCVZ wrote the manuscript, which was critically reviewed and revised by HAP, CWJ, DK and YJ. All authors were involved in the approval of the final content before submission.

Acknowledgments

Editorial assistance with the preparation of the manuscript was provided by inScience Communications, Springer Healthcare Ltd, UK, and was funded by Sanofi.

Disclosure statement

JCVZ is a Sanofi employee and may hold shares and/or stock options in the company. TUY, HAP, CWJ, DK and YJ declare no potential conflicts of interest with respect to the research, authorship and publication of this manuscript.

Data availability statement

Study’s supporting data not public; available on request from TUY.

Additional information

Funding

Sanofi provided financial support for publication fees.