Sero-prevalence of Hepatitis B virus infection and associated factors among pregnant women attending antenatal care service in health institutions in Gedeo Zone, Southern Ethiopia

Background: Hepatitis B virus infection is a major public health problem worldwide which is a major cause of morbidity and mortality. This study aimed to assess the prevalence of hepatitis B virus infection and associated factors among pregnant mothers in Gedio Zone, southern Ethiopia. Methods: Institutional based cross-sectional study was conducted in governmental and private health facilities in Gedeo zone from January to April 2019. The study participants were selected using stratied random sampling techniques. Eugene strip test was used to determine hepatitis B virus infection among pregnant mothers. The status of HIV was collected from the records. Other variables were collected from the mothers using interviewer administered questionnaires. Logistic regression was used for the analysis. Adjusted Odds Ratios and their 95% Condence Interval were calculated to determine association between HBsAg Sero-positivity and various factors. A p-value less than 0.05 were considered as signicant. The data was analyzed using the SPSS version 25 statistical software. Results: Prevalence of hepatitis B virus among pregnant mothers was 9.2% in Gedio Zone. Previous birth at health institution [AOR=4.4, 95% CI: 1.7, 11.2], blood transfusions [AOR=4.4, 95% CI: 1.8, 10.5], previous history of Hospital admission [AOR=3.3, 95% CI: 1.5, 7.5], ear piercing practice [AOR=5.7, 95% CI: 1.1, 29.0], current Gestational age [AOR=3.6, 95% CI: 1.2, 11.2], and HIV status of the mother [AOR=6.1, 95% CI: 1.3, 30.0] had statistical signicant association with HBsAg Sero-positivity. Conclusions: Hepatitis B virus infection was found to have higher endemicity (9.2%) in the Gedio Zone. History of blood transfusion, hospital admissions, ear piercing, being HIV positive, gestational age and institutional delivery were signicant predictors for HBsAg sero-positivity. Early initiation care creation risks of child


Introduction
Hepatitis B virus (HBV) is a hepatotropic deoxyribonucleic acid (DNA) virus which occurs through immune-mediated killing of infected liver cells [1]. It is a major blood-borne and sexually transmitted infectious agent, and poses a serious global public health problem which is approximately 100 times more contagious than human immunode ciency virus (HIV) and is found in diverse populations and subpopulations [2][3]. HBV is transmitted mainly through parenteral or mucosal exposure to infected blood and body uids, such as secretions or saliva, unsafe sexual intercourse; transfusion of HBV infected blood and blood products, usually either by a vertical or horizontal route early in life in highly endemic areas, resulting in a high rate of chronic infections [4.6].
Mother-to child transmission (MTCT) is one of the main transmission pathway and is responsible for approximately one-half of chronic hepatitis B (CHB) infection worldwide [7]. It is also associated with a high risk of maternal complications and has effects on both the mother (such as preeclampsia, placenta praevia, preterm delivery, placental separation, ante partum hemorrhage, preterm labor, increased incidence of intra-ventricular hemorrhage, gestational diabetes mellitus) and child (leads to fetal and neonatal hepatitis and higher risk of developing chronic liver disease and cancer) [8][9][10].
Pregnant mothers who have been tested positive for both hepatitis B surface antigen (HBsAg) and hepatitis B e antigen (HBeAg) have 70-90% risk of transmitting infection to their newborn infants and about 10-40% risk if they test positive for only HBsAg. In endemic areas, where carrier rates are greater than 5%, perinatal transmission is common; especially when HBV infected mothers is also HBeAg positive [11][12][13]. Global ly, 350 million people are chronic carriers. Of these, one million of them are expected to suffer serious illness and death from cirrhosis and hepatocellular carcinoma (HCC) and about 600,000 people die annually from acute or chronic complications of hepatitis B infection [14][15][16].
Africa has the second largest number of chronic carriers after Asia, and is considered a region of high endemicity (≥8%) [16]. Although it is di cult to assess the exact burden of HBV in Africa, the Seroprevalence of hepatitis B surface antigen (HBsAg) has been estimated to be in the range from 6% to 20% [17][18]. A higher prevalence 9.7%-16.6% was observed in other developing countries [19][20][21][22], while the prevalence of HBV in Ethiopia among pregnant women, have shown moderate endemicity, with the prevalence of HB sAg positivity ranging from 2.3%-7.9% [23][24][25][26][27][28][29]. Different factors such as having a history of blood transfusion, history of use of sharp materials, having multiple sexual partners, ear pricing, history of abortion, Place of delivery, practice of female Genital mutilation, history of tooth extraction cesarean section and tattooing were some of the major risk factors associated with HBVs Ag seropositivity in previous studies in similar settings [23][24][25][26][27][28][29][30][31][32][33][34].
Since HBV infected pregnant women are at risk of infecting their babies, knowing magnitude of HBV status and its risk factors in the area is very important in preventing mother-to-child transmission and reducing the burden of the disease. However, in resource constrained settings such as in Ethiopia, laboratory diagnosis of HBV infection is not part of routine care in ANC of all health facilities which makes the detection of pregnant mothers with HBV di cult, which intern also make the intervention very di cult [35]. Observational community studies of serological markers of HBV infection have an important role in identifying population endemicity, and possible routes of transmission which could help in the development of appropriate control measures. Therefore this study intends to ll the limited information gap re garding the prevalence and associated factors of HBV among pregnant women in the southern part of Ethiopia speci cally in Gedio Zone.

Design and study sites
This study was institutional based cross-sectional study conducted in selected health facilities in Gedio Zone Southern Nations and nationalities regional state, Ethiopia. The Zone has 1 referral hospital, 3 primary hospitals (Bule, Gedeb and Yerga Chefe), 38 health centres, 146 health posts, 4 NGO clinics and Sample size determination and procedure The sample size was determined using single population proportion formula. The sample size was calculated based on the following assumptions: 95% con dence interval (CI), Hepatitis B virus prevalence rate 3.5% [28], degree of precision of 1.75% and none-response rate of 10%. Finally, the calculated sample size was 479. The study participants were recruited using Strati ed random sampling method from selected health institutions as indicated in (Figure 1).

Data collection
Data was collected from January to March 2019. The socio demographic characteristics, such as age, residence, employment status, level of education, and marital status were collected using interviewer administered questionnaire by trained health professionals. HBV infection was determined using Eugene strip test.

Laboratory analysis
Three milliliters of venous blood sample was collected using was collected with ethylene diamine tetra acetate (EDTA) anti-coagulated tube following standard operating procedure (SOP) by trained laboratory professional. Two supervisors controlled data collection process. The tubes were labelled and processed at the time of collection at Dilla University Teaching Hospital central Medical laboratory. Serum was separated by centrifugation at 3000 rpm for 10 min. Each serum was subjected to HBsAg antibody rapid test. (Manufacturer: Shangai Eugene Biotech co., Ltd. Shangai. China. Email: info@eugenebio.com.)

Data quality assurance
To ensure quality of data, the questionnaire was prepared in English language, translated to Amharic and Gediofa and translated back to English by other person who can speak all the three languages. Further the questionnaire was pre-tested and eld editing was done to ensure completeness and correctness of data. Pre-test was conducted on 5% of total sample size in Bule primary hospital which was not included in this study. Collected data was checked daily for consistency and accuracy. Standardized procedures were strictly followed during blood sample collection, storage. 40 Two supervisors controlled data collection process. Testing for surface antigen was performed by senior laboratory technologist. Known positive and negative samples were run to control the quality of HBsAg kit as external quality assurance.

Data processing and analysis
Data was entered into Epi-Data version 3.1 and transferred to SPSS version 25 for analysis. Binary and multivariable logistic regression analyses were used to determine the association between explanatory variables and the outcome variable using odds ratio at 95% CI. Predictor variables with P-value <0.25 in the bivariate analysis were candidates for the multivariable logistic regression model. Adjusted Odds Ratios and their 95% Con dence Interval were calculated to determine association between HBsAg Sero-positivity and various factors. A p-value less than 0.05 were considered as signi cant. The Hosmer-Lemeshow test was used to check the overall model tness.

Socio-demographic characteristics
As shown in (Table 1), majority of the mothers interviewed 411(85.8%) were married and more than two third of the respondents 185(38.6%) have primary education. More than half of the participants 245(51.1%) were housewives while 265(55.3%) of the mothers were rural residents. Age between 26 and 30 years was the dominant maternal age group with 151(31.5%) while mothers older than 35 were only 23(4.8%).

Obstetric factors
Regarding the obstetrics history of the mothers, 121(25.3%) had no previous birth history, while the rest 358(74.7%) reported that they have multiple pregnancies. Out of 479 pregnant women those who are on the rst trimester were 99(20.7), and almost half of the pregnant women 245(50.5%) are on the second trimester while the rest are on the third trimester. It was assessed that 32(6.7%) of the pregnant women had a history of abortion in the past. Among 479 respondents those mothers history of home delivery was 103(21.5%) and history of institutional delivery had 255(52.2%).

Prevalence of Hepatitis B virus among ANC following pregnant mothers
The prevalence of HBV among pregnant mothers, who are on ANC follow-up in Gedio Zone, was 9.2% during study period.

Bivariate and Multivariable Analysis
During the bivariate analysis history of blood transfusion, previous history of Hospital Admission, previous history of abortion, Gestational age, previous place of birth, age of the pregnant Women, and HIV status of the pregnant women have a signi cant association with HBV status of the pregnant women. As indicated candidate in (table 1) variables with P value <0.25 were candidates for the multivariate logistic regression analysis. In the nal analysis women with previous history of blood transfusion, previous history of hospital admission, previous place of birth, ear piercing practice, Current gestational age and HIV status of the pregnant women were signi cant factors that determine, Seroprevalence of Hepatitis B virus of pregnant women in Gedeo zone in the multivariable logistic regression.

Discussion
To indicate the prevalence and endemicity of HBV active infection in the general population, surface antigen (HBsAg) is used as the main marker in a particular geographical area 43 . In this study area, the overall Sero-prevalence of HBsAg among pregnant women was 9.2 % which is considered as high endemicity area, (with sero-positivity ≥8% sero positive), based on WHO classi cation criteria of endemicity of HBV infection [16].
The prevalence was highest across the country to the researchers' knowledge. Previous studies in different regions of Ethiopia reported lower proportions, from the lowest prevalence (3.7 %) in Jimma town to the highest (8.4%) in Dire Dawa [23][24][25][26][27][28][29]. Yet, the nding was lower than ndings reported by other developing countries, with 8% in Mali [19], 9.7% in Cameroon [20], 10.8% in Yemen [21], 11.8 % in Uganda [22], and 16.6% in Nigeria [23]. The observed discrepancies in the magnitude of HBV prevalence across different geographical location might be due to variation in socio-demographic characteristics of the study population such as socio-cultural, environmental and tribal practices. Moreover, the variation in diagnosis methodologies, level of awareness, and the quality and access to antenatal care service provision might add to the difference.
Blood transfusion is a well-established risk factor for HBsAg. Pregnant women in Gedio Zone, with previous history of blood transfusion were 3.3 [AOR=3.3, 95% CI: 1.5, 7.5] times more likely to be infected with HBV than women who had no history of blood transfusion. The result is in accordance with the previous studies [28,[30][31].  [21,30]. Since Ethiopia is categorized as a country with high HIV burden [39] and high HBV endemic area [39] the possibility HBV/HIV co-infection is very much anticipated. Moreover, it was reported that co-infection of HIV/HBV could greatly facilitates HBV replication and reactivation leading to higher HBV-DNA levels and a reduced spontaneous clearance of the virus [40].
Compared to women in the third trimester, pregnant women in the rst trimester were more likely to be infected with HBV. This nding may have an implication on the early initiation of ANC service for pregnant mothers. Since, community transmission is possible through different routes such as ear piercing practice another prognostic factor which apparent in this study, early initiation of ANC service is critical to identifying the risk of HBV infection to the mother as well as to the baby to prevent mother to child transmission and to manage complications earlier [36]. Earlier screening of pregnant mothers during their antenatal care visits to health facilities for HBV has a signi cant relevance for identifying the risk of HBV infection.

Conclusions
Higher prevalence (9.2%) of HBV infection was detected in the study area. Our study illustrated that Seroprevalence of HBV infection was signi cantly associated with HIV positive status, having a history of blood transfusion, previous institutional delivery, hospital admission, being in the rst trimester and ear piercing practice were independent predictor of HBV infection in ANC following pregnant women in Gedeo Zone southern Ethiopia.
To reduce the Sero-prevalence of HBV infection in Gedio Zone, health education on the risk of HIV, home delivery, unsafe ear piercing practices and safety issues during admission to hospitals is needed at Zonal level and health facility level. At national level, screening all pregnant women for hepatitis B virus should be made as part of routine antenatal care service.