The effect of hepatitis B virus on the risk of pregnancy outcomes: a systematic review and meta-analysis of cohort studies

A double-stranded DNA virus belonging to the hepadnaviridae family, hepatitis B virus (HBV). The virus is enveloped and has a core with a viral DNA genome of approximately 3200 base pairs. In patient blood, the virus was initially identified as “Australia antigen,” subsequently known as hepatitis B surface antigen (HBsAg). Later on, as a marker for those at high risk of transmitting the disease, the hepatitis B e antigen (HBeAg) was discovered. The virus penetrates a hepatocyte, delivers its genome to the nucleus, and changes the relaxed circular DNA into covalently closed circular DNA (cccDNA) [1‐3]. Significant human morbidity and death are brought on by HBV infection, mostly as a result of the effects of chronic infection. According to recent estimates of chronically infected people ranging from 240 to 350 million, more than two billion people have ever been infected with HBV [4‐6]. Around 0.5–1.2 million people die annually on average [7]. There are three geographic regions where the prevalence of HBV infection is highest: East Asia and Africa (> 8%), the Mediterranean region (2–8%), and Eastern Europe (2%) [8]. More than half of the 350 million HBV carriers worldwide acquire the virus during pregnancy; rates of mother-to-child transmission also differ dramatically depending on the mother’s hepatitis B e antigen (HBeAg) status [9, 10]. Some mechanisms explain how HBV infection affects pregnancy. Reduced CD8 T cells and increased viral activity are caused by inhibiting the Th1 immune response and simulating the Th2 immune response, impairing the immunological response to HBV [11]. Due to the cross-reaction, increased regulatory T cells, and malfunctioning CD8 T cells, the exposure of the fetus to HBeAg may generate fetal T helper cell tolerance to HBeAg and HBcAg, which rasie HBV DNA levels during pregnancy [12, 13]. Studies have shown an increased risk of both newborn and maternal morbidity associated with HBV infection, including fetal distress, gestational diabetes mellitus, preterm delivery, and meconium peritonitis[14‐18]. Also, antepartum hemorrhage causing placental abruption and placenta previa can increase. A lower Apgar score is the only perinatal complication [14, 18]. However, there isn’t much research on the mechanisms underlying these results [19].

The effect of HBV on neonatal and maternal outcomes can create a basis for more accurate clinical decision-making. By identifying the specific connection between HBV and pregnancy outcomes, clinicians and specialists can reduce the impacts and enhance the quality of life for HBV patients. The research may aid in updating clinical guidelines and improving the care of HBV patients, which may aid in the early detection and prevention of pregnancy outcomes. Thus, by combining cohort data, we aimed to systematically review and meta-analyze the relationship between HBV and pregnancy outcomes.

The current study was a systematic review and meta-analysis, which was conducted to determine the exact relationship between the presence of infection and the chronic HBV and the occurrence of maternal outcomes such as pre-eclampsia, gestational diabetes, abortion, premature birth, infant death, and other related results. The primary studies in this meta-analysis were prospective or retrospective cohort studies. All the steps of this study were developed and carried out based on the structure of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guideline [20].

At first, the search strategy aimed at retrieving cohort studies published in the target field in international databases such as the Cumulated Index to Nursing and Allied Health Literature (CINAHL), SPORT Discuss via the EBSCO interface, PubMed (Medline), Scopus, Web of Science, and Embase. Then the retrieved articles were screened in Endnote version 8. Keywords were included: “Hepatitis B”, “Hepatitis B Virus”, “Pregnancy Outcome”, “Maternal Morbidity”, “Maternal Death”, “Pre-Eclampsia”, “Premature Birth”, and “Gestational Diabetes”, along with their synonyms in the Mesh database. To perform the screening, first, the duplicates of retrieved articles in the software were removed, and then, in the first step, screening based on the title; in the second step, screening based on the subject; and in the final step, screening based on the full text of the articles was done. And after these steps, the final articles were selected. The time frame for searching international databases and screening was from January 1990 to February 2023. Also, all the screening steps were done independently by authors (MA and MA), and disputes were resolved by the third researcher (YM), who was an expert. To carry out a comprehensive and detailed search, or in other words, to carry out gray literature to complete the search strategy, the first ten pages of Google Scholar along with a manual search (checking the sources and references of the final selected studies) were also done by the authors. The inclusion criteria and the final selection of studies in this meta-analysis were based on the PECOT structure, the specifics of which are detailed in Table 1. The studies that had all the characteristics listed in Table 2 were included in the present meta-analysis. After the screening, to extract the desired information from these articles based on the purpose of the study, a checklist prepared with the opinion of experts was used. Checklist elements included authors’ names, type of study, year of publication, total sample size, country of study, type of population, age, and effect size reported in the studies. Two authors (MA and MA/KZ) carried out data extraction independently, and the third researcher (YM), an expert, resolved disputes.

In this meta-analysis, after searching and retrieving all articles, 1390 articles in the PubMed database, 1002 articles in the Scopus database, and 779 articles in other relevant databases, including the Cumulated Index to Nursing and Allied Health Literature (CINAHL), SPORT Discuss via the EBSCO interface, Web of Science, and Embase, were retrieved. After removing the duplicates that included 1850 articles, 1321 articles were screened based on the title. In this stage, 902 articles were removed based on the title, and 419 articles entered the screening stage based on the abstract and then the full text. Finally, a total of 384 articles were removed in these steps. Thirty-five cohort studies were selected for meta-analysis and the present study (Table 2) (Fig. 1). The main point was that all selected cohort studies considered chronic HBV and examined its association with the occurrence of pregnancy outcomes.

The purpose of the present meta-analysis was to determine the relationship between HBV infection during pregnancy and the occurrence of pregnancy outcomes such as preeclampsia, premature birth, gestational diabetes, abortion, eclampsia, and hypertension during pregnancy. The results showed that the presence of HBV infection could increase the risk of adverse pregnancy outcomes such as preeclampsia, premature birth, gestational diabetes, abortion, eclampsia, and hypertension.

Preeclampsia can be considered one of the most common pregnancy complications in the last months of pregnancy (usually from the 20th week of pregnancy to 7 days after delivery), which is observed in 5% of pregnant women. Preeclampsia has no specific symptoms and is dangerous for the fetus and the mother. The contraction of blood vessels causes this disease and, as a result, leads to an increase in blood pressure and a decrease in blood flow in fetal organs such as the liver, kidney, and brain. This reduction in blood flow in the uterus leads to problems for the fetus, such as reduced growth, reduced amniotic fluid, etc. The current meta-analysis showed that the presence of HBV infection could increase the risk of preeclampsia in pregnant women by 10% compared to pregnant women without HBV infection. Meta-analysis studies are conducted to determine the more accurate and error-free effect and relationship between two important factors. These studies can control many possible errors in the relationship between exposure and the desired outcome. The important point in studying the relationship between the presence of HBV infection and the occurrence of preeclampsia in pregnant women was the difference in the pathophysiology and epidemiology of preeclampsia and HBV infection in early studies conducted in the world. In addition, the statistical population and method of care for each of these conditions differed in these studies. Other infections, such as parasitic, viral, and bacterial infections, may have a significant impact on the incidence and prevalence of pre-eclampsia in pregnant women, and the presence of other blood-borne and sexually transmitted infections, particularly hepatitis C and HIV/AIDS, may also increase susceptibility to HBV infection. As a result, it is likely that the presence of these conditions and other infections predisposes women to pre-eclampsia, with HBV acting only as an aggravating or enabling factor [23, 25, 28, 55, 56]. This factor was one of the main reasons for the high heterogeneity in the analysis of the relationship between the presence of HBV infection and the occurrence of preeclampsia.

Subgroup analyses in determining the relationship between HBV infection and the occurrence of preeclampsia in pregnant women showed that the risk of infection is higher in Asian women. This result confirmed the difference in the effect of HBV on the occurrence of preeclampsia in different geographical regions, because in other regions such as Europe or America, the relationship between HBV infection and the occurrence of preeclampsia was an inverse or protective relationship. The main reason for this is the difference in culture, the way services related to prenatal care before, during, and after pregnancy are provided and received, and, most importantly, the prevalence of other infections that are effective in Asian countries or other locations. Finally, there are plausible reasons in clinical texts for the association between HBV and the development of pre-eclampsia. In general, HBV infection has been associated with an increased risk of atherosclerosis in pregnant women [56‐60]. Furthermore, preeclampsia (marked by maternal endothelial dysfunction) may result in an imbalance of angiogenic, anti-angiogenic, and proangiogenic substances such as vascular endothelial growth factor [55, 61‐65]. According to previous research, there is a considerable link between HBV infection and insulin resistance, thrombocytopenia, obesity, and kidney damage or proteinuria [66‐68]. The interplay of these illnesses and HBV could explain the link between HBV and preeclampsia.

In the present meta-analysis, results showed that the presence of HBV infection increases the risk of gestational diabetes in pregnant women by 16% compared to women without HBV infection. Also, the risk of developing gestational diabetes in Asian women with HBV infection was lower than that of American women. This difference can be attributed to the importance of the issue of pregnancy and the difference in receiving services related to gestational diabetes screening in the Asian region and Asian countries. In addition, the different diet (especially in Southeast Asian regions) and the way of doing physical activity can be considered other reasons for this difference [69, 70]. Another point that can justify this relationship between Asian pregnant women is the difference in the prevalence of HBV infection in different regions, especially in different Asian and American countries [71, 72]. Of course, the various studies conducted in this field and the results of this meta-analysis confirm the fact that to more accurately determine the relationship between HBV infection and the occurrence of gestational diabetes, there is a need to conduct more studies taking into account the prevalence of HBV, the presence of chronic disease, and the background another is genetic factors and environmental factors such as nutritional and non-nutritive behaviors (smoking, alcohol, unprotected sex, etc.) [37, 73, 74]. An important factor that needs to be investigated in this connection is the body mass index of pregnant women, which can disrupt the relationship between HBV infection and the occurrence of gestational diabetes as an important confounding factor. In addition, the presence of other factors, such as high blood pressure, can also be one of the other factors influencing the relationship between HBV infection and the occurrence of gestational diabetes. These causes can be related to the occurrence of pre-eclampsia, metabolic syndrome, and then GDM [18, 28, 75, 76]. The current meta-analysis used a search technique that lasted through February 2023, and all retrieved studies were examined and screened. The inclusion of a defined selection of cohort studies, as well as the consideration of a specific time span for analysis and reporting, distinguishes this meta-analysis from review studies. Because the goal of this meta-analysis was to look at the relationship between hepatitis B infection and pregnancy outcomes, cohort studies were deemed the best primary research design for determining the relationship without taking into account interventions. This was one of the most significant distinctions between this meta-analysis and earlier research. In contrast, previous meta-analyses did not include all pregnancy outcomes. The outcomes of gestational diabetes and preterm birth were not included in the study by Karamati et al. [77], and the outcomes of gestational diabetes, pre-eclampsia, preterm birth, and miscarriage were not included in the analysis and review by Oliviera et al. [78] These outcomes were evaluated and analyzed in the current meta-analysis.

The important point of this meta-analysis was to estimate the effect size with high accuracy. Although the degree of heterogeneity in the estimated effect size was high, this degree of heterogeneity was indicative of statistical heterogeneity as determined by the I square index and Cochrane’s Q test. The important point was the absence of clinical heterogeneity or its presence at an acceptable level, which is interpreted by the estimated confidence intervals. All the confidence intervals obtained for the desired relationships in the present meta-analysis were narrow, which indicated high precision in the analysis. On the other hand, the narrow confidence interval is one of most important items in clinical interpretation and clinical justification of communication. For example, to determine the relationship between HBV infection and gestational diabetes, the estimated effect size was 1.16 with a confidence interval of 1.13 to 1.18. Despite the heterogeneity rate of 92.7%, the calculated confidence interval is very narrow, which indicates the existence of a sufficient sample and number of studies to determine the relationship and confirms the accuracy of the calculated relationship. Another strength of the present meta-analysis was the subgroup analysis based on the variables of geographic regions, gestational age, and the age of the pregnant mother, which to some extent, shows the role of confounding and other influencing variables in the relationship between HBV infection and pregnancy outcomes. The results of this meta-analysis can be very effective in developing or updating clinical guidelines.

One of the weaknesses or limitations of the present meta-analysis is the failure to perform subgroup analyses based on important variables such as body mass index, the presence of other infections (HIV/AIDS, hepatitis C infection), other underlying diseases (hypertension, genetics) pointed out that due to the lack of reporting of primary studies in their results, they were not included in the results of the present meta-analysis. We suggest that future studies be conducted to determine the relationship between HBV infection and the occurrence of pregnancy outcomes by considering these variables.

This meta-analysis found that hepatitis B infection during pregnancy may be associated with an increased risk of gestational diabetes, preterm delivery, pre-eclampsia, and eclampsia. However, confirmation of this association, as well as the specific biological pathways involved in the association between HBV infection and pregnancy outcomes, requires further investigation. As a result, it is crucial to enhance programs and healthcare services for women in society, focusing on the promotion of screening, care, and treatment programs for infectious diseases, particularly HBV. These efforts should be implemented across various communities, with particular emphasis on developing societies.