Alimentary Pharmacology and Therapeutics

Systematic review with meta-analysis: the risk of mother-to-child

transmission of hepatitis B virus infection in sub-Saharan
Africa
E. Keane*,†,1, A. L. Funk‡,1 & Y. Shimakawa‡

*Ecole Pasteur/CNAM de Sante SUMMARY

Publique, Paris, France.
†
Mater Misercicordiae University
Hospital, Dublin, Ireland.
Background
‡
Unite d’Epidemiologie des Maladies The risk of mother-to-child transmission of hepatitis B virus (HBV) has
Emergentes, Institut Pasteur, Paris, been quoted as 70–90% among women positive for hepatitis B surface anti-
France. gen (HBsAg) and e antigen (HBeAg), and 5–30% among HBsAg-positive
HBeAg-negative women. These risks are derived from Asia; little is known
Correspondence to: about sub-Saharan Africa.
Dr Y. Shimakawa, Unite
d’Epidemiologie des Maladies Aim
Emergentes, Institut Pasteur, 25 rue
To determine the risk of mother-to-child transmission in sub-Saharan
du Doctour Roux, Paris, France.
E-mail: yusuke.shimakawa@gmail.com
Africa, according to maternal HBeAg and type of prophylaxis.

1
These authors contributed equally to
this article.
Publication data
Submitted 11 April 2016
First decision 28 May 2016
Resubmitted 15 August 2016
Accepted 17 August 2016
EV Pub Online 15 September 2016
As part of AP&T’s peer-review process, a
technical check of this meta-analysis
was performed by Dr Y. Yuan. The
Handling Editor for this article was
Professor Geoffrey Dusheiko, and it was
accepted for publication after full
peer-review.
Methods
We searched Medline, Global Health, Embase, African Journals Online and
African Index Medicus. We included observational or interventional studies
that enrolled infants of HBV-infected women, and that tested for HBsAg or
HBV DNA between 3 and 12 months of age.
Results
Fifteen articles from 11 African countries were included. Among HBeAg-posi-
tive women, the pooled risk was 38.3% (95% CI: 7.0–74.4%) without prophy-
laxis, which was significantly lower than the lower bound of 70–90% risk in the
literature (P = 0.007). Among HBeAg-negative women, the pooled risk was
4.8% (95% CI: 0.1–13.3%) without prophylaxis, which lays within the lower
range of the 5–30% risk in Asia. By extrapolating the pooled transmission risks
to the number of births to infectious mothers, an estimated 1% of newborns
(n = 367 250) are annually infected with HBV at birth in sub-Saharan Africa.

Conclusions
Compared to Asia, the risk of mother-to-child transmission is low in sub-Saharan
Africa. However, the annual number of infants perinatally infected with HBV is
twice the number of incident paediatric HIV infections in sub-Saharan Africa
(n = 190 000). This highlights the importance of preventing mother-to-child
transmission of HBV in sub-Saharan Africa, which has been long neglected.

Aliment Pharmacol Ther 2016; 44: 1005–1017

ª 2016 John Wiley & Sons Ltd 1005

doi:10.1111/apt.13795
E. Keane et al.
INTRODUCTION
Hepatitis B virus (HBV) infection is a serious public
health problem, especially in sub-Saharan Africa where
the prevalence of hepatitis B surface antigen (HBsAg)
exceeds 8%.1 In 2010, there were an estimated 61 000
deaths in sub-Saharan Africa due to cirrhosis or hepa-
tocellular carcinoma (HCC) secondary to chronic HBV
infection.2 Of those who have established chronic HBV
infection in this region, the majority were infected
before adolescence either from mother-to-child trans-
mission or through horizontal transmission between
children.3 Maternal transmission mainly occurs perina-
tally through contact with infectious maternal fluids in
the birth canal; trans-placental intrauterine transmis-
sion and transmission through breastfeeding are
uncommon.4
Prevention of mother-to-child transmission is impor-
tant for many reasons. First, this mode of transmission
increases the risk of chronicity following acute infec-
tion; 80–90%, 20–30% and <10% of individuals
infected at birth, early childhood and adolescence/
adulthood, respectively, develop chronic infection.5 Sec-
ond, mother-to-child transmission may increase the
risk of liver disease, including HCC, in those who
become chronic carriers6–8; in West Africa, mother-to-
child transmission was associated with a five times
higher risk of developing significant liver fibrosis when
compared to horizontal transmission.9 Third, the
WHO has recently developed a global strategy to elim-
inate hepatitis B by 2030, and prevention of mother-
to-child transmission became a core intervention in
achieving this goal.10
The risk of mother-to-child transmission from HBV-
infected women (represented by positive HBsAg) is lar-
gely determined by the presence of hepatitis B e antigen
(HBeAg), a sero-marker of high viral replication, as well
as by HBV DNA levels. Globally, the risk of mother-to-
child transmission in the absence of prophylaxis is
quoted as 70–90% among women positive for both
HBsAg and HBeAg and 5–30% among women positive
for HBsAg but negative for HBeAg.4, 11–13 However,
these figures are derived mostly from Asian studies,14–16
and it is unclear whether these transmission risks apply
to sub-Saharan Africa.
The objectives of this systematic review are to: (i) esti-
mate the risk of mother-to-child transmission from
HBeAg-positive mothers and HBeAg-negative mothers
in sub-Saharan Africa according to the type of preventive
measures taken; (ii) compare the risk of mother-to-child
transmission in the absence of preventive measures to
1006
Asian estimates and (iii) estimate the transmission risk
from mothers with occult HBV infection. We also
extrapolated the estimated transmission risks to number
of births to estimate the number of infants infected at
birth each year in sub-Saharan Africa.
MATERIALS AND METHODS
This systematic review followed a protocol developed in
advance (Data S1), and was reported according to the
PRISMA guidelines.17
Eligibility criteria
We included observational studies (cross-sectional or
cohort) or intervention studies to prevent mother-
to-child transmission that had been conducted in sub-
Saharan Africa and that enrolled infants of mothers
positive for HBsAg or HBV DNA. Mother-to-child
transmission was defined by either HBsAg positivity or
HBV DNA positivity in infants between 3 and 12 months
of age.15, 16 We excluded articles if infant HBsAg status
was only assessed before 3 months as it takes up to
60 days for HBsAg to be detected following exposure.18
We only included studies in which child outcomes
could be stratified by the infectivity of mothers (presence
of HBeAg or detection/quantification of HBV DNA).
Maternal HBV markers were categorised according to
whether they were assessed in the pre-natal, at delivery or
post-partum period.
Search strategy
We searched OVID Medline, OVID Global Health,
ELSEVIER Embase, African Journals Online (AJOL) and
African Index Medicus up to July 2016. The following
search terms and their variations were used: hepatitis B,
mother-to-child transmission and sub-Saharan Africa. A
manual search through bibliographies was also con-
ducted. No language restrictions were applied. The full
search strategy is reported in Data S2.
Study selection
Two authors (EK and ALF) independently screened titles
and abstracts of all papers identified by the electronic
searches. Potentially eligible papers were obtained as a
full-text publication, and independently reviewed for
their eligibility. Disagreements were resolved by a third
reviewer (YS).
Data extraction
Data extraction was independently carried out by two
authors (EK and YS) using a pre-piloted standardised
Aliment Pharmacol Ther 2016; 44: 1005–1017
ª 2016 John Wiley & Sons Ltd
 Systematic review with meta-analysis: perinatal transmission of HBV in Africa
form (Data S3). Extracted information included study
design, settings, any preventive measures [hepatitis B
vaccine, hepatitis B immunoglobulin (HBIG), anti-viral
therapy], type of serological/virological assays, maternal
information (median/mean age, HBV genotypes, HIV
co-infection, obstetric complications, mode of delivery),
infant information (type of HBV marker assessed and
age at assessment), and outcomes (the number of eligi-
ble children with HBsAg-positive or HBV DNA-posi-
tive mothers, the number followed up, the number
infected) separated by maternal HBeAg status. When-
ever possible, the outcomes were further stratified by
maternal HBV DNA status. Nine corresponding
authors were contacted for further information, five
replied and three provided numerical data that were
not presented in the published papers. Where studies
did not mention whether the hepatitis B vaccine was
administered or not, we ascertained, via the WHO/
UNICEF report, whether the vaccine was integrated in
the national programme in this country at that time.
The included studies were also evaluated for the risk
of bias (Data S4).19
Statistical analysis
The principle summary measures were the risk of
mother-to-child transmission without any preventive
intervention in infants of mothers positive for both
HBsAg and HBeAg, and in infants of mothers positive
for HBsAg but negative for HBeAg. This was defined
as the proportion of the number of infants positive
for HBsAg and/or HBV DNA divided by the total
number assessed. The transmission risk from mothers
with occult HBV infection (defined as negative HBsAg
and detectable HBV DNA) was also estimated. If a
study reported children’s outcomes at ≥2 time points
between 3 and 12 months of their age, the latest time
point was used. To pool the risks, meta-analysis was
carried out using ‘metaprop’ command20 with STATA
13.1 (STATA Corporation, College Station, TX, USA).
After the variance of the proportions was stabilised
using Freeman-Tukey double arcsine transformation,21
estimates were pooled using the DerSimonian-Laird
random-effects model.22 The confidence intervals for
the individual studies were presented with the score-
test while those for the pooled estimates were pre-
sented with the Wald test.20 The percentage of total
variation between studies due to heterogeneity was
assessed using the I2 statistic.23 Using the Z test, the
pooled African estimates without any prophylaxis were
Aliment Pharmacol Ther 2016; 44: 1005–1017
ª 2016 John Wiley & Sons Ltd
compared with the lower boundary of the transmission
risk frequently quoted in the literature: 70% for
HBeAg-positive mothers and 5% for HBeAg-negative
mothers.4, 11–13
Subgroup analyses were performed to explore the
sources of between-study heterogeneity in the risk of
mother-to-child transmission without any prevention
from HBeAg-positive mothers and HBeAg-negative
mothers. The following potential sources of heterogeneity
were defined a priori and a test of heterogeneity was car-
ried out using the ‘metaprop’ command20: study region;
type of serological assay [moderately sensitive assays:
radioimmunoassay (RIA), haemagglutination assay (HA)
and immunochromatography (IC); or highly sensitivity
assays: enzyme immunoassay (EIA) and chemilumines-
cent immunoassay (CIA)]; sample size (N ≤ 10, 11–30,
>30); maternal virological status (HBV DNA levels,
genotypes, HIV co-infection); and infant information
[type of HBV marker assessed and age at assessment
(<6 months or ≥6 months)]. A multivariable meta-
regression was not conducted due to the small number
of studies included.
The effect of each preventive measure was assessed by
comparing the pooled transmission risks between those
with and without the intervention, after stratification by
maternal HBeAg status.
Finally, we estimated the number of infants infected
with HBV at birth each year in sub-Saharan Africa. We
first obtained the number of births to infectious mothers
by extrapolating the prevalence of HBsAg in each coun-
try1 and prevalence of HBeAg in women of child-bearing
age in SSA24 to the number of births in each country as
per the UN Population Division25 for the period of
2010–2015 as below.
ðNo. of birth to HBsAg-positive HBeAg-positive mothersÞ
¼ ðannual number of birthsÞ  ðprevalence of HBsAgÞ
 ðprevalence of HBeAgÞ
ðNo. of birth to HBsAg-positive HBeAg-negative mothersÞ
¼ ðannual number of birthsÞ  ðprevalence of HBsAgÞ
 ð1  prevalence of HBeAgÞ
Then, we extrapolated the estimated transmission risks
by type of prevention (no prophylaxis, timely birth dose
vaccine or delayed vaccine) to the number of births to
HBeAg-positive and HBeAg-negative mothers, after
taking account of country-specific hepatitis B vaccine
coverages in 2014 as below.26
1007
E. Keane et al.

ðNo. of infants infected with HBV at birthÞ ¼ ðNo. of birth to HBsAg-positive HBeAg-positive
mothersÞ  fðvaccine coverageÞ  ðtransmission risk from HBeAg-pos mothers with vaccineÞ
þ ð1  vaccine coverageÞ  ðtransmission risk from HBeAg-pos mothers without prophylaxisÞg
þ ðNo. of birth to HBsAg-positive HBeAg-negative mothersÞ  fðvaccine coverageÞ
ðtransmission risk from HBeAg-neg mothers with vaccineÞ þ ð1  vaccine coverageÞ
ðtransmission risk from HBeAg-neg mothers without prophylaxisÞg

Seven countries (Botswana, Cabo Verde, Djibouti, The
Gambia, Namibia, Nigeria and Sao Tome and Principe)
have already integrated birth dose vaccine into their
national programmes, and thus we used the coverage of
birth dose vaccine and the estimated risk of transmission
when the birth dose was administered. For the rest of
the countries, we applied the coverage of three doses of
hepatitis B vaccine and the estimated transmission risk
when the vaccine was delayed for >1 week.
RESULTS
The search strategy identified 4274 papers. After exclud-
ing 1452 articles in duplicate, 2822 were screened, and
69 papers were assessed in full text. Finally, 15 papers
were included in this review (Figure 1). Reasons for
exclusion are presented in Figure 1 and Table S1.
After excluding two studies that did not report the
number of women screened,27, 28 a total of 14 239
women were screened for HBsAg of whom 1143 tested
positive (pooled prevalence of 8.5%, 95% CI: 7.3–9.9%).
HBeAg was also tested in 951 HBsAg-positive women,
and 171 were positive (pooled prevalence of 16.2%, 95%
CI: 10.7–22.6%). Of 1222 infants of HBsAg-positive
mothers, 919 were finally assessed for HBV markers
and 745 infants, of whom 127 were born to HBeAg-
positive mothers and 618 to HBeAg-negative mothers,
provided at least one data point between 3 and
12 months of age.
Characteristics of the included studies
The majority of the included studies were conducted in
West Africa (n = 9),27–35 followed by East (n = 5)36–40
and Southern Africa (n = 1)41 (Table 1). There were 13
observational cohort studies27, 28, 30–33, 35–41 and two
nonrandomised controlled trials:29, 34 in Senegal, Yvonnet
et al. assessed the efficacy of the hepatitis B birth dose
vaccine (<24 h after birth) vs. no intervention29; and in
Ivory Coast, Ekra et al. assessed the efficacy of starting the
hepatitis B vaccine within 24 h of birth vs. within
6–8 weeks.34 The latter provided two-independent cohorts
in our analysis (birth dose cohort, 6 weeks cohort), while
in the former study the information on the control group
was insufficient and we could only include the interven-
tion group in the analysis. Of the 13 observational studies,
six were conducted without any prophylactic mea-
sures.30, 32, 33, 36–38 Two Gambian studies provided hep-
atitis B vaccine with varying schedules (median: 3 weeks,
95% received at >1 week after birth).28, 31 One Nigerian
study provided three different prophylactic regimens: hep-
atitis B vaccine and HBIG within 24 h (n = 16); hepatitis
B vaccine within 24 h without HBIG (n = 12); and hep-
atitis B vaccine at 2–7 days without HBIG (n = 11).27
Four studies exclusively recruited women co-infected with
HIV; three of these provided maternal anti-retroviral regi-
mens including lamivudine35, 40, 41 or tenofovir41 during
pregnancy, with a median duration of 2,40 341 and
4 months35 before the delivery. The fourth study of co-
infected mothers did not provide any pre-natal maternal
treatment, however, lamivudine was given to both moth-
ers and infants for the first 7 days after delivery.39 In all
studies involving co-infected mothers, the infants were
given hepatitis B vaccine after 6–8 weeks of birth as part
of the standard of care. Apart from the Nigerian study,
none of the studies provided HBIG to infants.
Maternal HBV status was determined prenatally
(n = 7),27, 28, 33–35, 39, 41 at delivery (n = 7)29, 30, 32, 36–38, 40
and post-partum (n = 2).31, 34 Maternal HBeAg was
assessed in all the included studies, however, HBV DNA
was examined only in six studies.32, 33, 36, 39–41 HBV
genotype in mothers was only reported in one study from
Malawi, for which all (100%) were genotype A.39 Women
with pre-term labour were excluded in one study,37 and
mode of child delivery was described in two studies.27, 40
Two studies employed HBV DNA as a marker of child
infection39, 41 while the rest used HBsAg. The risk of bias
in the included studies is summarised in Table S2.
Risk of mother-to-child transmission without
preventive measures
The pooled risk of mother-to-child transmission, without
any preventive measures, in infants of HBeAg-positive
mothers was 38.3% (total sample size n = 34, 95% CI:

1008 Aliment Pharmacol Ther 2016; 44: 1005–1017

ª 2016 John Wiley & Sons Ltd
 Systematic review with meta-analysis: perinatal transmission of HBV in Africa

Articles identified (N = 4274) through:

Medline (n = 1665)
Global Health (n = 833)
Embase (n = 1558)
African Journals Online (n = 70)
African Index Medicus (n = 148)

Duplicates removed (N = 1452)

Articles screened on the basis of title and abstract

(N = 2822)

Excluded by screening (N = 2755)

Manually added (N = 2)

Full-text articles assessed for eligibility (N = 69):

English language (n = 59)
French language (n = 9)
German language (n = 1)

Excluded articles (N = 54):

- Child HBV status not reported (n = 12)
- Maternal HBV status not reported
(n = 5)
- Maternal HBeAg or HBV DNA not
reported (n = 16)
- Children assessed at the age of >12
months (n = 6)
- Children assessed below the age of 3
months (n = 7)
- Child outcome is not presented
separately by maternal HBeAg or HBV
DNA status (n = 5)
- Review (n = 3)

Included articles (N = 15, a total of 17 cohorts):

No preventive measures (6 cohorts)
Vaccine starting at >1 week (3 cohorts)
Vaccine starting after 24 hours but before 7
days (1 cohort)
Vaccine starting at birth (3 cohorts)
Prenatal maternal therapy & vaccine
starting at 6 weeks (3 cohorts)
Figure 1 | Flow diagram of
Post-natal maternal/infantile therapy &
study selection. vaccine starting at 6 weeks (1 cohort)

7.0–74.4%) (Figure 2). This was significantly lower than strong evidence among HBeAg-negative mothers
the 70–90% risk that has been repeatedly quoted in the (I2 = 71.5%, P = 0.007).
literature (P = 0.007). The pooled risk in infants of
HBeAg-negative mothers was 4.8% (total sample size Subgroup analysis
n = 184, 95% CI: 0.1–13.3%) (Figure 3). This was not Of the eight potential sources of heterogeneity, maternal
different from the lower boundary of the 5–30% risk HBV DNA, genotype and HIV co-infection were not
presented in the literature (P = 0.2). There was good evi- assessed since only few reported on these variables. Simi-
dence of between-study heterogeneity among studies of larly, infant information was not evaluated because the
HBeAg-positive mothers (I2 = 50.6%, P = 0.09), and type of HBV marker and age at assessment did not vary

Aliment Pharmacol Ther 2016; 44: 1005–1017 1009

ª 2016 John Wiley & Sons Ltd
E. Keane et al.

Table 1 | Characteristics of the included studies

Assay methods Maternal factors Preventive measures Child factors
Sample
size Limit of Anti-viral Age at
Author, used HBsAg/ detection Mean/ When HBV therapy Child HBV
year, for this HBeAg for HBV median %HIV status during HBV assessment
reference Country Design analysis assay DNA age* (+) assessed Vaccine pregnancy marker (months)

Yvonnet, Senegal In 11 RIA/RIA N/A 25 N/R Delivery <24 h No HBsAg 3

198229 (15–41)
Marinier, Senegal Co 77 RIA/RIA N/A N/R N/R Delivery No No HBsAg 12
198530
Greenfield, Kenya Co 51 RIA/RIA 10 pg/mL N/R N/R Delivery No No HBsAg 9
198636
Tsega, Ethiopia Co 20 RIA/RIA N/A 27.7 N/R Delivery No No HBsAg 12
198837 (20–38)
Hall, 198931 Gambia Co 94 HA/RIA N/A N/R N/R Post-natal >1 week No HBsAg 12
Grathwohl, Cameroon Co 1 EIA/EIA 1000 N/R N/R Delivery No No HBsAg 6
199232 copies/
mL
Roingeard, Senegal Co 21 EIA/EIA 10 pg/mL 27  6 N/R Pre-natal No No HBsAg 6
199333
Menendez, Tanzania Co 48 EIA/EIA N/A 24.5  6.2 9.7% Delivery No No HBsAg 8
199938
Ekra, Ivory In 295 EIA/EIA N/A 26.1 (IQR: N/R Birth dose vaccine cohort
200834 Coast 14–53) Pre-natal <24 h No HBsAg 9
6 week vaccine cohort
Post-natal 6 weeks No HBsAg 9
Ilboudo, Burkina Co 14 IC/IC N/A 28.1  4.3 100% Pre-natal 8 weeks Yes HBsAg 4
201035 Faso
Onakewhor, Nigeria Co 39 IC/IC N/A 30  15 3.0% Pre-natal <1 week† No HBsAg 9
201327
Chasela, Malawi Co 42 CIA/EIA 32 IU/mL 25 (IQR: 100% Pre-natal 6 weeks No‡ HBV DNA 12
201439 22–29)
Hoffmann, South Co 10 CIA/CIA 20 IU/mL 29 (IQR: 100% Pre-natal 6 weeks Yes HBV DNA 12
201441 Africa 26–31)
Shimakawa, Gambia Co 9 HA/RIA N/A 25 (IQR: N/R Before >1 week No HBsAg 12
201428 23–30) pregnancy
Pirillo, Malawi Co 13 EIA/CIA 13 IU/mL 27 (IQR: 100% Delivery 6 weeks Yes HBsAg 12
201540 23–30)

CIA, chemiluminescent immunoassay; Co, cohort study; HA, haemagglutination assay; EIA, enzyme immunoassay;
IC, immunochromatography; In, interventional study; N/A, not applicable; N/R, not reported; RIA, radioimmunoassay.
* Presented as  s.d., or range, or IQR.
† Sixteen neonates received both hepatitis B vaccine and HBIG within 24 h of birth, 12 received only hepatitis B vaccine within
24 h of birth, and 11 received only hepatitis B vaccine at 2–7 days.
‡ The study provided lamivudine to both mothers and infants for 1 week, postnatally.

across the studies. Sample size was found to be a
source of heterogeneity in the risk of mother-to-child
transmission without prophylaxis from HBeAg-positive
mothers (Table 2 and Figure S1). The risk was lower in
studies with larger sample sizes (N > 10: 20.7%, 95%
CI: 7.1–38.1%) than those that included few infants
(N ≤ 10: 88.4%, 95% CI: 33.4–100%, P = 0.009). The
transmission risk did not differ according to the sero-
logical assay methods or the ‘region’ in HBeAg-positive
mothers. In infants of HBeAg-negative mothers, those
in East Africa tended to have higher risk (7.5%, 95%
CI: 1.7–15.9%) than in West Africa (0.5%, 95% CI:
0.0–4.3%, P = 0.05) (Figure S2).
Effect of preventive measures in infants of HBeAg-
positive mothers
Among the infants of HBeAg-positive mothers, the
pooled risk of mother-to-child transmission in those
who received hepatitis B vaccine at >1 week (36.6%, 95%
CI: 8.3–69.9%) and those with a timely birth dose vac-
cine (32.4%, 95% CI: 10.2–58.3%) did not differ from
the transmission risk without any prevention (38.3%,
95% CI: 7.0–74.4%, P = 0.9 and 0.8 respectively; Fig-
ure 2). In infants of women co-infected with HIV, the
pooled risk was higher when their mothers had received
anti-viral therapy during pregnancy (47.0%, 95% CI:
21.8–72.8%) than when both mothers and infants

1010 Aliment Pharmacol Ther 2016; 44: 1005–1017

ª 2016 John Wiley & Sons Ltd
 Systematic review with meta-analysis: perinatal transmission of HBV in Africa

No.
Author, infected/No.
year assessed ES (95% CI)

No intervention
Marinier, 1985 3/15 0.20 (0.07, 0.45)
Greenfield, 1986 1/2 0.50 (0.09, 0.91)
Grathwohl, 1992 1/1 1.00 (0.21, 1.00)
Roingeard, 1993 2/2 1.00 (0.34, 1.00)
Menendez, 1999 3/14 0.21 (0.08, 0.48)
Subtotal (I^2 = 50.59%, P = 0.09) 0.38 (0.07, 0.74)

Vaccine starting at >1 week

Hall, 1989 2/12 0.17 (0.05, 0.45)
Ekra, 2008 10/17 0.59 (0.36, 0.78)
Shimakawa, 2014 1/3 0.33 (0.06, 0.79)
Subtotal (I^2 = 60.76%, P = 0.08) 0.37 (0.08, 0.70)

Vaccine starting at birth (<24 h)

Yvonnet, 1982 1/2 0.50 (0.09, 0.91)
Ekra, 2008 9/24 0.38 (0.21, 0.57)
Onakewhor, 2013 0/1 0.00 (0.00, 0.79)
Subtotal (I^2 = 0.00%, P = 0.75) 0.32 (0.10, 0.58)

Antiviral therapy during pregnancy & vaccine starting at 6 weeks

Ilboudo, 2010 3/5 0.60 (0.23, 0.88)
Hoffmann, 2014 3/6 0.50 (0.19, 0.81)
Pirillo, 2015 2/6 0.33 (0.10, 0.70)
Subtotal (I^2 = 0.00%, P = 0.72) 0.47 (0.22, 0.73)
Figure 2 | Pooled estimates of Antiviral therapy/prophylaxis after birth & vaccine starting at 6 weeks
risk of mother-to-child Chasela, 2014 5/17 0.29 (0.13, 0.53)
transmission from HBeAg-
positive mothers by type of 0 .5 1
preventive measure.
Transmission risk from HBeAg-positive mothers by type of intervention

received anti-viral agents in the first week after birth The pooled risk was 0.7% (total sample size n = 11, 95%
(29.4%, 95% CI: 13.3–53.1%, P = 0.002). CI: 0.0–24.1%; Figure 4).

Effect of preventive measures in infants of HBeAg-
negative mothers
Compared to infants of HBeAg-negative mothers with-
out any prophylaxis, the pooled mother-to-child trans-
mission risk was significantly lower in those who
received hepatitis B vaccine at >1 week (0.0%, 95% CI:
0.0–0.0%, P = 0.01) and those with a timely birth dose
vaccine (0.0%, 95% CI: 0.0–0.0%, P = 0.02; Figure 3).
Among infants of women co-infected with HIV, the
pooled risk was higher in the studies of pre-natal mater-
nal anti-viral therapy (5.1%, 95% CI: 0.0–29.8%) than in
immediate post-natal anti-viral therapy in mothers and
infants (0.0%, 95% CI: 0.0–13.3%), although the differ-
ence was not significant (P = 0.1).
Risk of mother-to-child transmission from mothers
with occult HBV
Two studies assessed the risk of mother-to-child trans-
mission from mothers with occult HBV infection. In
both studies, the mothers were also infected with HIV.
Number of infants infected at birth each year
Of 177 115 000 infants born in sub-Saharan Africa over
the period 2010–2015, an estimated 1 836 250 (1.0%)
are perinatally infected with HBV; annually, 367 250
infants are infected at birth in sub-Saharan Africa. Esti-
mates by country are presented in Table S3.
DISCUSSION
In this meta-analysis, we found that: (i) in the absence
of any preventive measures the pooled transmission risk
from HBeAg-positive mothers in sub-Saharan Africa was
significantly lower than those reported in Asia (38.3% vs.
70-90%); (ii) the transmission risk from HBeAg-negative
mothers in sub-Saharan Africa was similar to Asian esti-
mates (4.8% vs. 5–30%); (iii) the administration of hep-
atitis B vaccine at birth and at >1 week was associated
with a reduction in transmission risk from HBeAg-nega-
tive mothers but not HBeAg-positive mothers and (iv)
the transmission risk from HBeAg-positive pregnant
women co-infected with HIV was substantial even when

Aliment Pharmacol Ther 2016; 44: 1005–1017 1011

ª 2016 John Wiley & Sons Ltd
E. Keane et al.

No.
Author, infected/No.
year assessed ES (95% CI)

No intervention
Marinier, 1985 0/62 0.00 (0.00, 0.06)
Greenfield, 1986 7/49 0.14 (0.07, 0.27)
Tsega, 1988 1/20 0.05 (0.01, 0.24)
Roingeard, 1993 2/19 0.11 (0.03, 0.31)
Menendez, 1999 1/34 0.03 (0.01, 0.15)
Subtotal (I^2 = 71.47%, P = 0.01) 0.05 (0.00, 0.13)

Vaccine starting at >1 week

Hall, 1989 0/82 0.00 (0.00, 0.04)
Ekra, 2008 0/112 0.00 (0.00, 0.03)
Shimakawa, 2014 0/6 0.00 (0.00, 0.39)
Subtotal (I^2 = 0.00%, P = 0.77) 0.00 (0.00, 0.00)

Vaccine starting at 2-7 days

Onakewhor, 2013 0/11 0.00 (0.00, 0.26)

Vaccine starting at birth (<24 h)

Yvonnet, 1982 0/9 0.00 (0.00, 0.30)
Ekra, 2008 0/142 0.00 (0.00, 0.03)
Onakewhor, 2013 0/11 0.00 (0.00, 0.26)
Subtotal (I^2 = 0.00%, P = 0.64) 0.00 (0.00, 0.00)
Vaccine & immunoglobulin starting at birth (<24 h)
Onakewhor, 2013 0/16 0.00 (0.00, 0.19)

Antiviral therapy during pregnancy & vaccine starting at 6 weeks

Ilboudo, 2010 0/9 0.00 (0.00, 0.30)
Hoffmann, 2014 0/4 0.00 (0.00, 0.49)
Pirillo, 2015 2/7 0.29 (0.08, 0.64)
Subtotal (I^2 = 39.54%, P = 0.19) 0.05 (0.00, 0.30)
Figure 3 | Pooled estimates of
Antiviral therapy/prophylaxis after birth & vaccine starting at 6 weeks
risk of mother-to-child
Chasela, 2014 0/25 0.00 (0.00, 0.13)
transmission from HBeAg-
negative mothers by type of
0 .5 1 preventive measure.
Transmission risk from HBeAg-negative mothers by type of intervention

Table 2 | Potential sources of heterogeneity in the estimates of mother-to-child transmission risk (without any
intervention) by maternal HBeAg status

Risk in infants of HBeAg-positive mothers Risk in infants of HBeAg-negative mothers

Variables Pooled estimate 95% CI P-value Pooled estimate 95% CI P-value

Assay methods
RIA/HA/IC 19.4% 11.1–46.6 0.2 4.4% 0.0–19.1 0.9
EIA/CIA 67.0% 0.3–100 5.0% 0.3–13.4
Sample size
≤10 88.4% 33.4–100 0.009 N/A 0.5
11–30 20.7% 7.1–38.1 7.4% 0.6–18.8
>30 N/A 3.9% 0.0–16.3
Region
West Africa 66.4% 0.0–100 0.3 0.5% 0.0–4.3 0.05
East Africa 21.2% 1.5–49.6 7.5% 1.7–15.9

mothers received anti-viral therapy during pregnancy or
when infants received anti-viral prophylaxis.
Despite a similarly high prevalence of HBsAg between
Asia and sub-Saharan Africa, the major mode of trans-
mission differs. In the pre-vaccine era, an estimated 40%
of chronic carriers were perinatally infected in Asia while
in sub-Saharan Africa the vast majority (90%) were
infected horizontally during childhood and only a
minority (10%) had acquired through mother-to-child
transmission.42 Indeed, we estimated here that 1% of

1012 Aliment Pharmacol Ther 2016; 44: 1005–1017

ª 2016 John Wiley & Sons Ltd
 Systematic review with meta-analysis: perinatal transmission of HBV in Africa

No.
Author, infected/No.
year assessed ES (95% CI)

Occult HBV

Chasela, 2014 0/9 0.00 (0.00, 0.30)

Hoffmann, 2014 1/2 0.50 (0.09, 0.91)

Figure 4 | Pooled estimates of Subtotal (I^2 = 0.00%, P = .) 0.01 (0.00, 0.24)

risk of mother-to-child
transmission from mothers 0 .5 1
with occult HBV infection.
Transmission risk from mothers with occult HBV infection

Table 3 | Risk of mother-to-child transmission by maternal HBeAg and type of preventive measure in Asia and in
sub-Saharan Africa

Transmission risk from HBsAg- Transmission risk from HBsAg-

positive HBeAg-positive mothers positive HBeAg-negative mothers

Sub-Saharan Sub-Saharan References for

Asia Africa* Asia Africa* Asian estimates
14–16, 48
No prophylaxis 70–90% 38% 5–30% 5%
Timely birth dose vaccine alone 20% 32% <0.5% 0% 48, 49

Timely birth dose vaccine & HBIG 5–10% No data <0.5% 0% 48–50

Timely birth dose vaccine & HBIG & <2%† No data No data No data 51, 52

anti-viral therapy during pregnancy

* Estimates from the current analysis.

† Using tenofovir as anti-viral therapy.

newborns in sub-Saharan Africa are infected at birth,
whereas this has been reported as 3–5% in Asia.12
The incidence of mother-to-child transmission in an
area is determined by three factors: the prevalence of
infectious mothers, the risk of transmission from an
infectious mother to her infant, and the coverage of
preventive services. The lower incidence of mother-to-
child transmission in sub-Saharan Africa has generally
been attributed to a lower prevalence of HBeAg in
HBsAg-positive African mothers.3 However, we found
that the transmission risk from HBeAg-positive mothers
in the absence of prophylaxis was significantly lower in
sub-Saharan Africa (38.3%) than in Asia (70–90%), and
the risk from HBeAg-negative mothers lays within the
lower range of the Asian estimates (4.8% vs. 5–30%).
Although the number of studies in this review was rela-
tively small, studies with larger sample sizes showed
even smaller transmission risk from HBeAg-positive
women (20.7%), supporting a robustness of this geo-
graphical difference in the risk of mother-to-child
transmission.
There are several explanations for this apparent geo-
graphical difference in transmission risk. First, among
mothers positive for HBeAg, HBV DNA levels may be
lower in sub-Saharan Africa than in Asia. However, previ-
ous African studies do not support this; median HBV
DNA levels in HBeAg-positive pregnant women
(106–8 IU/mL)43, 44 were similar to those in Asia.45 Sec-
ond, circulating HBV genotypes differ between Africa and
Asia, and certain genotypes are known to increase the risk
of mother-to-child transmission (e.g. genotype C com-
pared to B in Asia).46 Although genotype was not exam-
ined except for one study, we found a significant difference
in transmission risk from HBeAg-negative mothers
between West and East Africa. This may be related with a
variation in predominant genotype: E in West and A in
East Africa.9 Third, obstetric complications are known to
be associated with a transmission from infectious moth-
ers.47 These complications may be more likely to result in
child deaths in sub-Saharan Africa due to insufficient
neonatal care, which may lead to a reduction in the num-
ber of HBV-infected infants aged at 3–12 months.

Aliment Pharmacol Ther 2016; 44: 1005–1017 1013

ª 2016 John Wiley & Sons Ltd
E. Keane et al.
In Asia, the efficacy of the birth dose vaccine without
HBIG compared to no prophylaxis has been firmly estab-
lished; the risk reduces from 70–90% to 20% in infants of
HBeAg-positive mothers, and from 5–30% to <0.5% in
those born to HBeAg-negative mothers (Table 3).48, 49
However, we found that transmission risk from HBeAg-
positive mothers in sub-Saharan Africa was similar
between the three varied vaccination groups: no vaccine
(38.3%), delayed vaccine at >1 week (36.6%), and timely
birth dose vaccine (32.4%). The only controlled trial for
the prevention of mother-to-child transmission in Africa
(which was included in our pooled analysis) showed a
substantial transmission risk from HBeAg-positive moth-
ers in the timely birth dose group (37.5%).34 We do not
know whether this worrisome finding represents a gen-
uine lack of efficacy of timely birth dose vaccine in Africa,
or if this is related to the quality of vaccines that were used
in this particular study (e.g. loss of potency due to inad-
vertent freezing of the vaccines within the cold chain).
Until further data on the efficacy of timely birth dose vac-
cine becomes available in Africa, it is safer to assume that
the birth dose vaccine alone may not be sufficient to pre-
vent transmission from HBeAg-positive mothers in sub-
Saharan Africa.
In contrast, among infants born to HBeAg-negative
mothers, the pooled transmission risk was reduced from
4.8% without prophylaxis to zero in those who received
hepatitis B vaccine at birth or at >1 week after birth.
One possible explanation for this reduction is that the
5% of infants infected in the absence of any intervention
might have been contaminated after birth through early
horizontal transmission. As this review included infants
aged up to 12 months, it is possible that some early hor-
izontal transmission has been captured; this horizontal
transmission might have been eliminated when a vaccine
was given at birth or even later.
In Asia, additional administration of HBIG to the
birth dose vaccine has been shown to further reduce
the transmission risk from HBeAg-positive mothers;
from 20% with only the birth dose vaccine, to 5–10%
when this was combined with HBIG (Table 3).48, 50
Moreover, to eliminate the risk of immunoprophylaxis
failure, anti-viral therapy has been administered to
highly viraemic women during pregnancy in addition
to hepatitis B vaccine and HBIG at birth, with a pro-
ven efficacy.51, 52 In contrast, in sub-Saharan Africa,
only one study27 provided HBIG, and anti-viral treat-
ment was only administered to pregnant women co-
infected with HIV, primarily to prevent mother-to-child
transmission of HIV.
1014
Our study provided a unique opportunity to estimate
the risk of HBV mother-to-child transmission from HIV
co-infected women treated with anti-virals during preg-
nancy but without neonatal immunoprophylaxis, and
found a substantial transmission risk (47.0% from
HBeAg-positive, and 5.1% from HBeAg-negative moth-
ers). We could not assess the efficacy of this strategy
(anti-virals during pregnancy without birth dose vac-
cine), because there was no control group of infants born
to HIV co-infected mothers without any prophylaxis.
Another study of HIV co-infected mothers did not pro-
vide any anti-viral therapy during pregnancy, but admin-
istered anti-retrovirals, including lamivudine, to both
mothers and infants in the first week after delivery to
prevent mother-to-child transmission of HIV through
breastfeeding.39 Interestingly, the risk of HBV transmis-
sion in this study was significantly lower than the studies
which gave anti-virals during pregnancy, suggesting that
anti-viral prophylaxis to neonates, a strategy similar to
pre-exposure prophylaxis of infants against HIV, might
be another option to prevent mother-to-child transmis-
sion of HBV.
There are few limitations in our study. First, we
could not examine the transmission risk by maternal
HBV DNA levels, which are more accurate than
HBeAg to predict the risk of mother-to-child transmis-
sion.53 This is because only few included studies
assessed maternal HBV DNA levels, and these studies
used different assay methods with a varying limit of
detection. Nevertheless, our estimates based on HBeAg
are still highly useful in sub-Saharan Africa as the
access to HBV DNA assays is often restricted to a
sophisticated laboratory in large urban centres.54 More-
over, our analysis clearly demonstrated that maternal
HBeAg is a perfectly sensitive marker (sensitiv-
ity = 100%) to predict HBV transmission to their chil-
dren despite timely administration of birth dose
vaccine. Second, the efficacy of preventive measures
should be best estimated by pooling a risk ratio from
randomised controlled trials. However, we assessed the
efficacy of timely birth dose and delayed hepatitis B
vaccine by comparing pooled transmission risk of vacci-
nated cohorts with that of nonvaccinated cohorts,
because none of the included studies had a control
group except one.34 Care must be taken to interpret
these results because of their vulnerability to confound-
ing. Finally, there were two papers that did not men-
tion whether or not there was administration of
hepatitis B vaccine in the study cohort and were con-
sidered as ‘received vaccination’ based on the national
Aliment Pharmacol Ther 2016; 44: 1005–1017
ª 2016 John Wiley & Sons Ltd
 Systematic review with meta-analysis: perinatal transmission of HBV in Africa

programme: Ilboudo et al.35 and Pirillo et al.40 We SUPPORTING INFORMATION

assumed that all infants in these studies received hep-
atitis B vaccine based on the high coverage of hepatitis
B vaccine in these countries reported by the WHO/
UNICEF during the study period: 89–92% in Burkina
Faso and 91–97% in Malawi.26
We estimated that, annually, 367 250 infants in sub-
Saharan Africa are infected with HBV at birth, which is
approximately twice the number of children newly
infected with HIV in this region (n = 190 000).55 This
highlights the importance of preventing mother-to-child
transmission of HBV in sub-Saharan Africa. Our system-
atic review also delineated a lack of high quality African
studies that assessed the efficacy of measures to prevent
mother-to-child transmission of HBV (Table 3). The
combination of numerous interventions, as evidenced by
many Asian studies, is unlikely to be readily applicable
to sub-Saharan Africa due to differences in the standard
of care. In Asia, most countries implement universal
birth dose vaccine with occasional availability of HBIG
and/or anti-viral treatment during pregnancy, while in
sub-Saharan Africa, the current coverage of the birth
dose vaccine is only 10%,26 HBV screening is rarely per-
formed at antenatal care, HBV DNA assays are often
unavailable and access to HBIG or anti-viral therapy for
HBV mono-infected individuals is severely limited.54
Preventive options adapted to the African context, for
example, antenatal screening for HBsAg and HBeAg,
and selective birth dose vaccine and nucleoside analogue
therapy for pregnant women, need to be urgently devel-
oped and validated.56
Additional Supporting Information may be found in the
online version of this article:
Data S1. Protocol for the systematic review and meta-
analysis.
Data S2. Search strategy.
Data S3. Data extraction information.
Data S4. Framework for assessing the risk of bias in
individual studies.
Figure S1. Subgroup analyses for the risk of mother-
to-child transmission from HBeAg-positive mothers.
Figure S2. Subgroup analyses for the risk of mother-
to-child transmission from HBeAg-negative mothers.
Table S1. Excluded papers and reasons.
Table S2. Description of risk of bias.
Table S3. Number of infants perinatally infected each
year in sub-Saharan Africa.
AUTHORSHIP
Guarantor of the article: Y. Shimakawa.
Author contributions: EK and YS designed the study. EK and ALF
independently conducted the review. EK and YS independently
extracted the data. ALF and YS conducted the meta-analysis.
All authors wrote the paper and approved the final version.
ACKNOWLEDGEMENTS
We greatly thank Pr. Arnaud Fontanet for his critical review and
helpful comments, and Jan Hellert and Simeon Carstens for their
help in translating articles.
Declaration of personal interests: None.
Declaration of funding interests: This work was supported by the
NeoVac (Neonatal Vaccination against Hepatitis B in Africa) project
(funded by the Total Foundation). ALF was supported by the Pas-
teur-Paris University (PPU) International PhD Program.

REFERENCES
1. Schweitzer A, Horn J, Mikolajczyk RT,
Krause G, Ott JJ. Estimations of
worldwide prevalence of chronic
hepatitis B virus infection: a
systematic review of data published
between 1965 and 2013. Lancet 2015;
6736: 1–10.
2. Cowie BC, MacLachlan JH. The global
burden of liver disease attributable to
hepatitis B, hepatitis C, and alcohol:
increasing mortality, differing causes.
Hepatology 2013; 58: 218A–9A.
3. Kiire CF. The epidemiology and
prophylaxis of hepatitis B in sub-
Saharan Africa: a view from tropical
and subtropical Africa. Gut 1996; 38
(suppl. 2): S5–12.
4. Shepard CW, Simard EP, Finelli L,
Fiore AE, Bell BP. Hepatitis B virus
infection: epidemiology and vaccination.
Epidemiol Rev 2006; 28: 112–25.
5. Hyams KC. Risks of chronicity
following acute hepatitis B virus
infection: a review. Clin Infect Dis 1995;
20: 992–1000.
6. Chang M-H. Natural history and
clinical management of chronic
hepatitis B virus infection in children.
Hepatol Int 2008; 2(suppl. 1): S28–36.
7. Shimakawa Y, Yan H-J, Tsuchiya N,
Bottomley C, Hall AJ. Association of
early age at establishment of chronic
hepatitis B infection with persistent
viral replication, liver cirrhosis and
hepatocellular carcinoma: a systematic
review. PLoS ONE 2013; 8: e69430.
8. Shimakawa Y, Lemoine M, Bottomley
C, et al. Birth order and risk of
hepatocellular carcinoma in chronic
carriers of hepatitis B virus: a case-
control study in The Gambia. Liver Int
2015; 35: 2318–26.
9. Shimakawa Y, Lemoine M, Njai HF,
et al. Natural history of chronic HBV
infection in West Africa: a longitudinal
population-based study from The
Gambia. Gut 2015; doi:10.1136/gutjnl-
2015-309892 [Epub ahead of print].
10. WHO. Combating Hepatitis B and C to
Reach Elimination by 2030. Geneva,
Switzerland: WHO, 2016.
11. Mast EE, Margolis HS, Fiore AE, et al.
A comprehensive immunization
strategy to eliminate transmission of
hepatitis B virus infection in the United
States: recommendations of the
Advisory Committee on Immunization

Aliment Pharmacol Ther 2016; 44: 1005–1017 1015

ª 2016 John Wiley & Sons Ltd
E. Keane et al.
Practices (ACIP) part 1: immunization
of infants, children, and adolescents.
MMWR Recomm Rep 2005; 16: 1–31.
12. WHO. Preventing Mother-to-Child
Transmission of Hepatitis B: Operational
Field Guidelines for Delivery of the Birth
Dose of Hepatitis B Vaccine. Geneva,
Switzerland: WHO, 2006.
13. Visvanathan K, Dusheiko G, Giles M,
et al. Managing HBV in pregnancy.
Prevention, prophylaxis, treatment and
follow-up: position paper produced by
Australian, UK and New Zealand key
opinion leaders. Gut 2016; 65: 340–50.
14. Okada K, Kamiyama I, Inomata M,
Imai M, Miyakawa Y, Mayumi M. E
antigen and anti-E in the serum of
asymptomatic carrier mothers as
indicators of positive and negative
transmission of hepatitis B virus to
their infants. N Engl J Med 1976; 294:
746–9.
15. Beasley RP, Trepo C, Stevens CE,
Szmuness W. The e antigen and
vertical transmission of hepatitis B
surface antigen. Am J Epidemiol 1977;
105: 94–8.
16. Stevens CE, Neurath RA, Beasley RP,
Szmuness W. HBeAg and anti-hbe
detection by radioimmunoassay:
correlation with vertical transmission of
hepatitis B virus in Taiwan. J Med Virol
1979; 3: 237–41.
17. Liberati A, Altman DG, Tetzlaff J, et al.
The PRISMA statement for reporting
systematic reviews and meta-analyses of
studies that evaluate health care
interventions: explanation and
elaboration. J Clin Epidemiol 2009; 62:
e1–34.
18. Weinbaum CM, Williams I, Mast EE,
et al. Recommendations for
identification and public health
management of persons with chronic
hepatitis B virus infection. MMWR
Recomm Rep 2008; 8: 1–20.
19. Altman DG. Systematic reviews of
evaluations of prognostic variables. In:
Egger M, Smith GD, Altman DG, eds.
Systematic Reviews in Health Care:
Meta-Analysis in Context, 2nd ed.
London: BMJ Publishing Group, 2001;
228–47.
20. Nyaga VN, Arbyn M, Aerts M.
Metaprop : a Stata command to
perform meta-analysis of binomial data.
Arch Public Health 2014; 72: 1–10.
21. Freeman MF, Tukey JW.
Transformations related to the angular
and the square root. Ann Math Stat
1950; 21: 607–11.
22. DerSimonian R, Laird N. Meta-analysis
in clinical trials. Control Clin Trials
1986; 7: 177–88.
23. Higgins JPT, Thompson SG, Deeks JJ,
Altman DG. Measuring inconsistency in
meta-analyses. BMJ 2003; 327: 557–60.
1016
24. Ott JJ, Stevens GA, Wiersma ST. The
risk of perinatal hepatitis B virus
transmission: hepatitis B e antigen
(HBeAg) prevalence estimates for all
world regions. BMC Infect Dis 2012; 12:
131.
25. United Nations, Population Division,
Department of Economic and Social
Affairs. World Population Prospects:
The 2015 Revision, 2015. Available at:
http://esa.un.org/unpd/wpp/Download/
Standard/Fertility/ (accessed 22 January
2016).
26. WHO & UNICEF. WHO-UNICEF
estimates of vaccine coverage, 2015.
Available at: http://apps.who.int/
immunization_monitoring/
globalsummary/timeseries/
tswucoveragehepb3.html (accessed 22
January 2016).
27. Onakewhor J, Charurat M, Matthew O,
Osagie E, Asemota M, Omoigberale A.
Immunologic pattern of hepatitis B
infection among exposed and non-
exposed babies in a PMTCT program
in low resource setting: does every
exposed newborn require 200 IU of
hepatitis B immunoglobulin? J Vaccines
Vaccin 2013; 4: 207.
28. Shimakawa Y, Bottomley C, Njie R,
Mendy M. The association between
maternal hepatitis B e antigen status, as
a proxy for perinatal transmission, and
the risk of hepatitis B e antigenaemia in
Gambian children. BMC Public Health
2014; 14: 532.
29. Yvonnet B, Digoutte P, Denis F, Correa
P. [Prevention of mother-to-child
transmission of hepatitis B virus by
vaccination of the newborn]. Med Afr
Noire 1982; 29: 721–31.
30. Marinier E, Barrois V, Larouze B, et al.
Lack of perinatal transmission of
hepatitis B virus infection in Senegal,
West Africa. J Pediatr 1985; 106: 843–9.
31. The Gambia Hepatitis Study Group.
Hepatitis B vaccine in the expanded
programme of immunisation: the
gambian experience. Lancet 1989; 1:
1057–60.
32. Grathwohl J, Ndumbe P, Leke R, Uy A,
Gerlich WH, Repp R. [Risk of perinatal
hepatitis B virus transmission in
mothers with low viral DNA levels
detectable by the polymerase chain
reaction]. Monatsschr Kinderheilkd
1992; 140: 366–8.
33. Roingeard P, Diouf A, Sankale JL, et al.
Perinatal transmission of hepatitis B
virus in Senegal, west Africa. Viral
Immunol 1993; 6: 65–73.
34. Ekra D, Herbinger KH, Konate S, et al.
A non-randomized vaccine effectiveness
trial of accelerated infant hepatitis B
immunization schedules with a first
dose at birth or age 6 weeks in Cote
d’Ivoire. Vaccine 2008; 26: 2753–61.
35. Ilboudo D, Simpore J, Ouermi D, et al.
Towards the complete eradication of
mother-to-child HIV/HBV coinfection
at Saint Camille Medical Centre in
Burkina Faso, Africa. Brazilian J Infect
Dis 2010; 14: 219–24.
36. Greenfield C, Osidiana V, Karayiannis
P, et al. Perinatal transmission of
hepatitis B virus in Kenya: its relation
to the presence of serum HBV-DNA
and anti-HBe in the mother. J Med
Virol 1986; 19: 135–42.
37. Tsega E, Tsega M, Mengesha B,
Nordenfelt E, Hansson BG, Lindberg J.
Transmission of hepatitis B virus
infection in Ethiopia with emphasis on
the importance of vertical transmission.
Int J Epidemiol 1988; 17: 874–9.
38. Menendez C, Sanchez-Tapias JM,
Kahigwa E, et al. Prevalence and
mother-to-infant transmission of
hepatitis viruses B, C, and E in
Southern Tanzania. J Med Virol 1999;
58: 215–20.
39. Chasela CS, Kourtis AP, Wall P, et al.
Hepatitis B virus infection among HIV-
infected pregnant women in Malawi
and transmission to infants. J Hepatol
2014; 60: 508–14.
40. Pirillo MF, Scarcella P, Andreotti M,
et al. Hepatitis B virus mother-to-child
transmission among HIV-infected
women receiving lamivudine-containing
antiretroviral regimens during
pregnancy and breastfeeding. J Viral
Hepat 2015; 22: 289–96.
41. Hoffmann CJ, Mashabela F, Cohn S,
et al. Maternal hepatitis B and infant
infection among pregnant women living
with HIV in South Africa. J Int AIDS
Soc 2014; 17: 18871. eCollection 2014.
42. Edmunds WJ, Medley GF, Nokes DJ,
O’Callaghan CJ, Whittle HC, Hall AJ.
Epidemiological patterns of hepatitis B
virus (HBV) in highly endemic areas.
Epidemiol Infect 1996; 117: 313–25.
43. Andersson MI, Maponga TG, Ijaz S,
et al. The epidemiology of hepatitis B
virus infection in HIV-infected and
HIV-uninfected pregnant women in the
Western Cape, South Africa. Vaccine
2013; 31: 5579–84.
44. Ducancelle A, Abgueguen P, Birguel J,
et al. High endemicity and low
molecular diversity of hepatitis B virus
infections in pregnant women in a rural
district of North Cameroon. PLoS ONE
2013; 8: 10–2.
45. Wen W-H, Chang M-H, Zhao L-L,
et al. Mother-to-infant transmission of
hepatitis B virus infection: significance
of maternal viral load and strategies for
intervention. J Hepatol 2013; 59: 24–30.
46. Wen WH, Chen HL, Ni YH, et al.
Secular trend of the viral genotype
distribution in children with chronic
hepatitis B virus infection after
Aliment Pharmacol Ther 2016; 44: 1005–1017
ª 2016 John Wiley & Sons Ltd
 Systematic review with meta-analysis: perinatal transmission of HBV in Africa
universal infant immunization.
Hepatology 2011; 53: 429–36.
47. Xu D-Z, Yan Y-P, Choi BCK, et al.
Risk factors and mechanism of
transplacental transmission of hepatitis
B virus: a case-control study. J Med
Virol 2002; 67: 20–6.
48. Lee C, Gong Y, Brok J, Boxall EH,
Gluud C. Hepatitis B immunisation for
newborn infants of hepatitis B surface
antigen-positive mothers. Cochrane
Database Syst Rev 2006; (2): CD004790.
49. Machaira M, Papaevangelou V,
Vouloumanou EK, Tansarli GS, Falagas
ME. Hepatitis B vaccine alone or with
hepatitis B immunoglobulin in neonates
of HBsAg+/HBeAg- mothers: a
systematic review and meta-analysis. J
Aliment Pharmacol Ther 2016; 44: 1005–1017
ª 2016 John Wiley & Sons Ltd
Antimicrob Chemother 2015; 70: 396–
404.
50. Chen H-L, Lin L-H, Hu F-C, et al.
Effects of maternal screening and
universal immunization to prevent
mother-to-infant transmission of HBV.
Gastroenterology 2012; 142: 773–81. e2.
51. Chen HL, Lee CN, Chang CH, et al.
Efficacy of maternal tenofovir disoproxil
fumarate in interrupting mother-to-
infant transmission of hepatitis B virus.
Hepatology 2015; 62: 375–86.
52. Pan CQ, Duan Z, Dai E, et al.
Tenofovir to prevent hepatitis B
transmission in mothers with high viral
load. N Engl J Med 2016; 374: 2324–34.
53. Burk RD, Hwang L-Y, Ho GYF,
Shafritz DA, Beasley RP. Outcome of
perinatal hepatitis B virus exposure is
dependent on maternal virus load. J
Infect Dis 1994; 170: 1418–23.
54. Lemoine M, Eholie S, Lacombe K.
Reducing the neglected burden of viral
hepatitis in Africa: strategies for a
global approach. J Hepatol 2015; 62:
469–76.
55. UNAIDS. Fact Sheet 2015, 2015.
Available at: http://www.unaids.org/
sites/default/files/media_asset/
20150901_FactSheet_2015_en.pdf
(accessed 1 April 2016).
56. Shimakawa Y, Toure-Kane C, Mendy
M, Thursz M, Lemoine M. Mother-to-
child transmission of hepatitis B in sub-
Saharan Africa. Lancet Infect Dis 2016;
16: 19–20.
1017