Seroprevalence, Importance of Rapid Testing and Factors Associated with HBsAg Carriage in Pregnancy in Three Referral Maternity Hospitals in Mono (Benin), in 2020 ()
1. Introduction
Viral hepatitis is an infectious disease that causes a greater number of deaths worldwide compared to mortality due to tuberculosis or the human immunodeficiency virus (HIV). Indeed, in 2019, they caused around 820,000 deaths. These deaths are mainly due to cirrhosis or hepatocellular carcinoma [1] . According to the latest epidemiological estimates from the World Health Organization (WHO), 296 million people were living with chronic hepatitis B in 2019 and 1.5 million new infections are detected each year [1] [2] . Mother-to-child transmission of the hepatitis B virus (HBV) remains a major cause of the maintenance of HBV endemic in the world despite the availability of an effective vaccine against this virus for more than 25 years [3] . In order to reduce maternal and fetal morbidity linked to infection with the hepatitis B and C viruses, the World Health Organization (WHO) proposes systematic screening for hepatitis B (HBsAg) in course of pregnancy. But HBV screening tests are not always accessible to all pregnant women in all localities in Benin due to lack of financial resources. The rapid screening test for hepatitis B and C viruses could be an opportunity to reduce the prevalence, and improve the diagnosis and management of these infections in pregnancy. It should be remembered that there are data on the prevalence of HBV in pregnant women in certain departments of Benin such as Littoral and Borgou. However, there is still no data on the subject in the Mono. For these reasons, this study was conducted to assess the impact of a free rapid diagnostic orientation test for hepatitis B virus infection on seroprevalence and factors associated with HBsAg carriage by pregnant women in reference maternities in western Benin (Mono). This will then make it possible to generalize it to the whole country and would contribute to having a real idea of HBV and HCV infections during pregnancy.
2. Methods
This was a descriptive cross-sectional study with prospective data collection. It took place from October 5 to December 31, 2020 (3 months) in the prenatal consultation services (PCS) of the maternities of the Departmental Hospital Center (DHC) of Mono, the zone hospital (ZH) of Lokossa and the municipal health center of Athiémé. The study population consisted of pregnant women received in PCS.
The inclusion criteria were: pregnant woman, prenatal follow-up carried out in one of the maternities hosting this study and having given her consent to participate in the study.
Exclusion criteria were breach of consent and invalid rapid HBV tests (no red bars appear in both windows C and T or in window C).
Sampling was non-probability with systematic listing of all eligible pregnant women according to the inclusion criteria. The sample size N was determined by the Schwartz formula:
,
With
N: the calculated sample size;
ε = 1.96 (value at 5% risk);
i = 5% (margin of error);
p = baseline prevalence of HBV infection in pregnant women;
q = 1 – p.
Determination of N: considering the prevalence of HBV which was 14% at the University Clinic of Gynecology-Obstetrics (CUGO) in 2017 by Kpossou et al. [4] .
. Thus 201 pregnant women constituted the representative population for this study.
The HBV rapid test (Micropoint®) used in our study was made in China in April 2014 by Micropoint. The sensitivity and specificity displayed by the leaflet was 98.4% with an expiry date of April 2022.
The dependent variable was HBsAg carriage defined by a positive rapid screening test and confirmed by the Enzyme Linked Immuno-Sorbent Assay serology (ELISA Intec HBsAg®). The independent variables were the sociodemographic and clinical characteristics of pregnant women carrying viral hepatitis B and the therapeutic and prognostic aspects of this infection on pregnancy.
The pregnant women were received in turn in a room where they were alone with the interviewer. After the information on the subject, the consent of the pregnant woman was collected before carrying out the pre-test counseling. Then, the hepatitis B screening test was performed with the RDOT and the result was immediately communicated to each participant before the post-test counselling. In pregnant women screened for HBsAg positive, a search for HBV-HCV and HBV-HIV co-infection was also carried out during confirmation by ELISA test. In addition, cases detected positive for HBsAg were referred for specialized care by a hepato-gastroenterologist.
Data collection was done using KoBoCollect on the basis of an elaborate and validated survey form. The questionnaire used was designed by the authors themselves. It contained 85 questions including general information, gynecological and obstetric history, medicosurgical history, social survey, clinical signs, information about current pregnancy and serological markers (see Appendix). Data collection was carried out by four (04) investigators (medical students) including the principal investigator in collaboration with the midwives of the various maternities concerned.
Data processing and analysis were done with EPI INFO software version 7.2.1.0. The graphs and tables were designed with EXCEL 2013, the quantitative variables were presented in mean, median, standard deviation then the qualitative variables in frequencies or percentages. The chi2 test was used to determine the associations between the different variables, with a significance level of 5%.
The study was carried out in strict compliance with ethical standards and rules of ethics. Thus we have received written authorization from the local Ethics Committee for Biomedical Research of the University of Parakou (Ref: 0483/CLERB-UP/P/SP/R/SA) and from the Departmental Director of Health of the Mono. Then we explained the purpose, nature and scope of the study, the procedure for recruiting pregnant women and collecting data from doctors, gynecologists and midwives. Data collection was carried out under conditions respecting basic measures in terms of confidentiality of the information obtained and the results of the tests that were carried out. The tests were anonymous. The questionnaires and the samples were identified only by a code that will not allow the identification of the pregnant woman who agreed to participate in the study. This number was used to link the various pieces of information.
In addition, it should be noted that in this study, women who had never given birth were considered nulliparous. The primiparous are women who have had a single birth before this study, the pauciparous are women who gave birth at least three (02) times before the study. Multiparas are considered by women who have given birth four (04) times. Grand multiparas are pregnant women whose number of deliveries is greater than or equal to five (05).
3. Results
3.1. Sociodemographic Characteristics
v Sample size
We studied 201 pregnant women in three referral maternity hospitals.
v Age
The average age was 27.7 ± 5.4 years with extremes of 15 and 45 years. The modal age group was 25 to 29 years old.
v Education level
A good proportion (81.6%, n = 164) of pregnant women surveyed were educated and 50.8% had at least secondary education (Figure 1).
v Occupation
In this study population, 88.6% (n = 178) of the pregnant women surveyed were non-civil servants and therefore had no fixed monthly income (Table 1). Other occupations were: secretary, nurse’s assistant, computer scientist, pharmacy assistant, laboratory biologist, missionary and photographer.
v Marital status
Life as a couple in a monogamous regime was the most represented with a proportion of 75.6% (n = 152). It is followed by life as a couple in a polygamous regime, i.e. a proportion of 20.9% (n = 42). Then come single people with a proportion of 3% (n = 6) and finally only one pregnant woman was widowed in this study.
![]()
Figure 1. Distribution of pregnant women attending the Lokossa Athiémé reference maternity units in 2020, by level of education.
![]()
Table 1. Distribution of pregnant women attended in the Lokossa-Athiémé reference maternity units in 2020 according to their occupation.
3.2. Clinical Features
v Gyneco-obstetric history
➢ Parity
The pauciparous were the most represented in our series in a proportion of 38.8% (n = 78). They were followed by nulliparous 24.9% (n = 50). Primiparous represented the third strongest proportion 24.4% (n = 49). Multiparous and grand multiparous are the least represented with a respective proportion of 11.4% (n = 23) and 0.5% (n = 1).
➢ Number of living children
In this series, 73.6% (n = 148) of pregnant women had at least one living child versus 26.4% (n = 53) who had none.
➢ Other gyneco-obstetric history
More than half 52.7% (n = 106) had no gyneco-obstetric history among pregnant women. Spontaneous miscarriages were noted in 22.9% (n = 46) of pregnant women surveyed. Caesarean section, hypertension on pregnancy, induced miscarriages, threats of premature delivery were mentioned respectively by 17.4% (n = 35), 4.5% (n = 9), 2% (n = 4) and 0.5% (n = 1).
v Medical and surgical history
Morbid antecedents were observed in 5% (n = 10) of pregnant women surveyed. These were sickle cell anemia (1.5%, n = 3), hypertension (2.5%, n = 5), hemophilia (0.5%, n = 1) and others (0.5%, n = 1). Almost all pregnant women and their partners did not know their HBV serological status in a respective proportion of 98% (n = 99) and 92% (n = 185) and HCV with a respective proportion of 100% (n = 201) and 91.5% (n = 184). Most of the pregnant 99% (n = 199) were therefore not vaccinated against the hepatitis B virus.
v Social survey
➢ Number of known sexual partners in life
More than half of pregnant women 59.2% (n = 119) had known at least two sexual partners (Figure 2).
➢ Other risk factors for HBV
Acts leading to skin invasion such as piercings and scarifications were the main other risk factors found (Table 2).
v Clinical signs
The most common symptoms encountered during this study were physical asthenia (51.2% n = 103), anorexia (23.4%, n = 47), fever (10%, n = 20), arthralgia (8.5%, n = 17), jaundice (2%, n = 4) and purpura (0.5% n = 1). The other symptoms were represented by headaches, abdominal pain, dyspnoea, undocumented epigastralgia: and were present in 5.9% (n = 12) of pregnant women.
➢ Number of antenatal consultations (ANC) carried out
In the study, only 46.3% (n = 93) of pregnant women had performed at least three ANC during their pregnancy. The majority (53.7%, n = 108) performed less than three ANC.
![]()
Figure 2. Distribution of pregnant women monitored in the Lokossa Athiémé reference maternity units in 2020 according to the number of sexual partners.
![]()
Table 2. Distribution of pregnant women monitored in the Lokossa-Athiémé reference maternity units in 2020 according to other risk factors.
3.3. Paraclinic Features
Only 11.9% of the respondents carried out hepatitis B serology as part of the prenatal follow-up, before inclusion in the study. The other prenatal check-ups carried out by pregnant women were: Rhesus blood grouping (36.3%, n = 73), hemoglobin electrophoresis (14.9%, n = 30), syphilitic serology (13.4%, n = 27), obstetric ultrasound (54.2%, n = 109) and HIV serology (100%, n = 201).
3.4. Seroprevalence of Hepatitis B and Co-Infections with HCV and HIV in the Reference Maternity Wards of Lokossa-Athiémé in 2020
v Seroprevalence of hepatite B in pregnancy
The seroprevalence of HBsAg in pregnancy in this study was 5.5% (11/201). The pregnant women who tested positive for RDOT Ag HBs were all confirmed by the ELISA test.
v Co-infection HBV-HCV et HBV-HIV
No HBV-HCV co-infection was detected. Similarly, no HBV-HIV co-infection was recorded during this study.
3.5. Factors Associated with Hepatitis B Virus Infections
Sociodemographic factors such as age, educational level, and occupation were not associated with HBV infection. Thus, of HBV-positive pregnant women, 45.4% (n = 5) were housewives, 36.4% (n = 4) were resellers, 18.2% (n = 2) were craft workers. But these results obtained do not show any significant association (p = 0.36) between the profession and the carriage of the HBs antigen during pregnancy in this study.
History such as parity, history of miscarriage and threat of childbirth were also not associated with HBV infection in this study (Table 3).
Among the pregnant women screened positive for HBV, four were aware of their previous HBV serological status, only one of whom was following traditional treatment. The other three were not on any treatment. None of them had any particular medical and surgical history and had never been vaccinated against HBV. Ignorance of serological status was associated with HBV infection during pregnancy (p = 0.0001) (Table 3).
The number of lifetime sexual partners (p = 0.01) and the practice of excision (p = 0.0001) were associated with HBV infection during pregnancy (Table 3).
The results obtained showed no significant influence of hepatitis B on pregnancy in this study.
![]()
![]()
Table 3. Factors associated with HBsAg carriage during pregnancy.
4. Discussion
In this study, the seroprevalence of viral hepatitis B in pregnant women was 5.5%. Among blood donors, it was estimated at 6.3% in 2011 according to surveys conducted by Kodjoh et al., which classified the Mono department in the WHO intermediate zone (prevalence between 2% - 7%) [5] . This result is similar to those of Kouakou et al. [6] who noted in a referral hospital in Ivory Coast in 2017 a seroprevalence of 5.4% and Fiacre et al. [7] who reported a prevalence of 6.1% in Cameroon in 2019. In 2017 Alassan et al. [8] found a hepatitis B seroprevalence of 14.02% among pregnant women in Parakou in northern Benin. Similarly, Kpossou et al. [4] reported 14% HBV on pregnancy in 2017 in Cotonou. This difference could be explained by the fact that Kpossou et al. and Alassan et al. had carried out their study respectively in the departments of Littoral and Borgou which are classified among the high prevalence areas of the WHO (prevalence > 8) according to Kodjoh et al. in 2013 [9] .
The average age was 27.7 ± 5.4 years with the extremes of 15 and 45 years in the study. This result is similar to that of De Paschale et al. [10] who reported an average age of 26.2 years with extremes ranging from 15 to 41 years in northern Benin in 2014. Similarly, in Morocco, Sbiti et al. [11] found in 2016 a mean age of 28 ± 6 years with extremes ranging from 17 to 43 years. In Mali, in the obstetrics gynecology department of the CHU Gabriel Touré in 2020, an average age of pregnant women of 26.9 ± 5.6 years with extremes of 18 and 43 years had been found by Traoré [12] . In Benin, at the University Clinic of Gynecology and Obstetrics (CUGO) of the CNHU-HKM in 2017, an average age of 30 ± 5 years with extremes of 30 to 35 years was noted [4] . This age difference could be explained by the fact that women marry at a young age in Mono according to the 2015 social dashboard [13] .
Zeba et al. [14] in 2009 in Ouagadougou found a predominance of educated pregnant women in the proportions of 91.7%. This study also found a predominance of pregnant women attending school (81.4%). This result could be explained by the fact that the study was carried out in urban or peri-urban communes where young women are more likely to be educated.
The majority of pregnant women in the study were resellers or traders (38.8%), workers or artisans (30.3%) and housewives (16%). Ousmane et al. [15] found in 2016 in Niger that the majority of pregnant women were housewives (85%) and civil servants (12%). Similar studies carried out in Cotonou found mostly civil servants [4] [10] . This difference is explained by the fact that our study took place in an agricultural area where women are more involved in trade and crafts, whereas those mentioned were carried out in the city of Benin’s economic capital, where fixed monthly income activities predominate (public and private sectors).
Among the pregnant women in this study, 96.5% were married. This result is identical to those of Ejeta et al. [16] in Ethiopia in 2014 who reported 96.4% of pregnant women married. De Paschale et al. [13] in 2014 in Tanguiéta found a similar proportion (95%). These high proportions of pregnant women in a couple could be due to the fact that in African culture pregnancy is frowned upon outside of couple’s life.
Some studies have shown an influence of hepatitis B on pregnancy. This is the case of Sirilert et al. [17] in Thailand in 2012 who reported that there is a slightly increased risk of premature birth and low birth weight linked to the carriage of HBsAg during pregnancy. Cai et al. [18] in 2015 in China noted that chronic HBV infection during pregnancy may increase the risk of intrahepatic cholestasis of pregnancy and premature rupture of membranes. This is not the case in the study. Most studies of the sub-region have come to the same result as ours [6] [10] [17] . The discordance of the results could be explained by the fact that most of the studies which show an association between viral hepatitis B and pregnancy are carried out in Asian countries and the correspondence with other populations must remain cautious.
The 11 pregnant women who tested positive for RDOT underwent an ELISA test which confirmed the presence of HBsAg in all cases. In the study by Kpossou et al. on 23 pregnant women screened positive by the rapid diagnostic orientation test, the confirmation ELISA serology was also positive for the 23 cases [4] . The same is true for Alassan et al. where all the cases positive to the rapid diagnostic test were confirmed by the ELISA method [8] . It can therefore be concluded that RDOT has an important advantage in the detection of a considerable number of cases of viral hepatitis B in pregnancy. However, from a comparative study between rapid tests and ELISA (the reference method) carried out in Nigeria in 2019 by Abulude et al. [19] , it appears that ELISA was slightly superior because 02 negative cases with the rapid test were positive by ELISA. We can therefore remember that, when the RDOT is negative, the pregnant woman must be informed of the limits of the test and therefore of the possibility of carrying out a reference examination from a blood sample.
The most affected age group was 30 to 34 years old and age was not associated with HBsAg carriage during pregnancy in this study. Fouelifack et al. [20] in Cameroon in 2016 reported that pregnant women aged 30 to 34 were significantly more infected with HBV (OR = 2; 95% CI: 1.82 - 2.05). Similarly Luuse et al. [21] found in Ghana in 2016 that the most affected age group was 15 - 20 years with a statistically significant association (p < 0.001). The young age observed in these studies and ours could be explained by the fact that the transmission of HBV in our regions remains mainly perinatal.
The pregnant women most exposed to carrying were resellers (n = 4), craft workers (n = 2). All these socio-demographic factors were not associated with HBV infection during pregnancy in this series. A study carried out in northern Benin in 2017 noted no significant association between sociodemographic factors [age (p = 0.909)], marital status (p = 0.880), level of education (p = 0.749), socio-professional status (p = 0.329), monthly income (p = 0.160)] and HBsAg carriage [8] . This is not the case with Bigot et al. [22] in 1989 in Cotonou (Benin) who reported in their series that women positive for HBsAg came from low socioeconomic strata. Sidibe et al. [23] found a high percentage of multigravidae (76%) and multiparous (75.1%) in HBV-positive pregnant women. A study carried out in southern Benin in 2017 also noted the dominance of multigravidae (90%) and multiparous (86.6%) [4] . Furthermore, these studies did not show any association between parity and HBsAg carriage. Similar observations were made in our study, with the predominance of paucigravidae (n = 4), multigravidae (n = 4) and pauciparous (n = 6); moreover, gestality and parity were not significantly associated with HBV. Ngaira et al. in Kenya [24] in 2014 noted that multiparous were the most exposed with a significant association between parity and HBsAg carriage during pregnancy.
Ignorance by pregnant women of their previous HBV serological status could lead to neglect of means of prevention, thus exposing them to the risk of contamination. This is clearly found in our study with only 4/201 (2%) of pregnant women knowing that they are HBsAg carriers and 84.6% of pregnant women surveyed unaware of their previous serological status. Unawareness of prior HBV serological status was significantly (p = 0.0001) associated with HBsAg carriage during pregnancy in our study. The same result was found in Benin in the University Clinic of Gynecology and Obstetrics of the National and University Hospital Hubert Koutoukou Maga in 2017 [4] , knowledge of a previously positive HBV serology was associated with carrying of HBsAg in pregnant women in this study. Thus, among pregnant women with no knowledge of their previous HBV serological status, 10.3% (n = 13) had a positive HBV test and 89.2% (n = 107) pregnant women were negative for HBV with a p value = 0.01.
Vaccination of women of childbearing age is one of the major pillars of the prevention of vertical transmission of HBV. In this study, only 02 pregnant women, i.e. a proportion of 1%, were vaccinated against the hepatitis B virus and none of the HBV positive pregnant women was. This result is similar to that reported in Morocco in 2014 with a rate of 2.4% of pregnant women having been vaccinated [11] . The same observation was made by Ousmane et al. [15] in 2016 in Niger. They reported that none of the women surveyed had been vaccinated against the hepatitis B virus. It appears that new vaccination strategies must be introduced in order to improve HBV vaccination coverage in Africa in general and in Benin in particular.
In this study, 4% of our pregnant women with several sexual partners had positive serology and this result was significant. Yohanes et al. [25] reported a much higher rate of 16.7% (p = 0.04) of pregnant women having known several sexual partners. Fouelifack et al. [20] found that multiple sexual partners were significantly associated with hepatitis B infection (OR = 11.47; 95% CI: 5.11 - 25.74). This result is logical given that multiple sexual partnerships promote the dissemination of HBV and it is urgent to take measures to raise awareness about the transmission of hepatitis B. However, this conclusion remains to be put into perspective because the transmission of HBV in our country is especially perinatal.
The practice of female circumcision is one of the African practices carried out under precarious aseptic conditions. In the study, a pregnant woman who tested positive for RDOT Ag HBs had been excised. This result was significantly associated with HBsAg carriage during pregnancy. Murad et al. [26] came to the same result, reporting a significant association of excision with HBsAg seropositivity (OR = 3.3, 95% CI: 1.1 - 10.2; p = 0.03). Alassan et al. [8] , on the other hand, had not noted any significant association between excision and HBsAg positivity.
We found no association between HBsAg carriage and manicure or pedicure with common material (p = 0.48), tattoos (p = 0.94), scarification (p = 0.79), piercings (p = 0.8), acupuncture (p = 0.96), treatments (p = 0.61) and digestive endoscopies (p = 0.96). It is the same finding in Ousmane et al. [15] who had noted no significant association between the carriage of HBsAg during pregnancy with dental care, tattoos and piercings. Some authors have found associations between these risk factors and HBsAg carriage in pregnant women. This is the case of Mve Koh [27] in Cameroon in 2018 who reported an association between dental care [OR: 2.4, 95% CI (2.26 - 4.96), p = 0.01]; scarifications [OR 7.4 95% CI (3.77 - 14.58) p < 0.01] and HBsAg carriage during pregnancy. The discrepancy in the results could be explained by the fact that these traditional practices are increasingly carried out under better hygienic conditions.
This study provides useful information for the fight against hepatitis B in Benin. Currently, screening for hepatitis B is not free for pregnant women in all maternity hospitals in Benin. However, since 2005, vaccination against HBV has been applied to all infants as part of the expanded vaccination program and since 2020 a dose within 24 hours of birth has been administered to newborns throughout Benin.
The limitation of this work is not having systematically carried out the ELISA test at the same time as the TROD for the screening of HBs Ag in order to prove the reliability of the use of the TROD. This limitation is essentially linked to the fact that the study was not funded.
5. Conclusion
Screening for viral hepatitis B during pregnancy is not often carried out in maternity wards in western Benin. It was easy, thanks to the rapid diagnostic orientation test which had made it possible to find an intermediate prevalence of HBsAg in pregnant women according to the WHO classification. The factors associated with the carriage of HBsAg during pregnancy were multiple sexual partnerships, excision and ignorance of the serological status prior to HBV. This study did not show that there is a link between obstetric complications and the carriage of the HBS antigen. Setting up free HBV screening by RDOT in pregnant women in all maternity hospitals in Benin would be useful to contribute to the eradication of HBV in Benin.
Authors’ Contributions
All authors participated in the active writing and editing of the article. All authors read and approved the final version of the manuscript.
Appendix
QUESTIONNAIRE
“VIRAL HEPATITIS B AND C IN PREGNANCY IN THREE REFERENCE MATERNITIES IN THE LOKOSSA-ATHIEME HEALTH ZONE IN 2020”
File number: [___] [___] [___] Date:
CONSENT FORM (For pregnant women)
I have read and understood the information about the purpose of the study on viral hepatitis B and C in pregnancy in the Lokossa-Athiémé reference hospitals: diagnosis and treatment.
I had the opportunity to ask any questions to the members of the research team. The answers were provided to me in language that is understandable to me.
I’ve understood that:
The disadvantages and advantages of my participation in this study
My participation in the study is voluntary and I can withdraw at any time.
The researchers involved in this study will have access to the data which concerns me in compliance with the strictest confidentiality
Research files could be inspected by the Parakou ethics committee to ensure the smooth running of the study.
It was clearly explained to me and I understood that my consent does not relieve the research organizers of their responsibility and I retain all my rights as guaranteed by law.
I freely and informedly consent to participate in this investigation.
Date:
Signature