Study on the Relationship Between Hepatitis B Virus Infection in Pregnant Women and Adverse Pregnancy Outcomes

Objective: Aim to the relationship between adverse pregnant outcomes with chronic hepatitis B virus (HBV) infection in pregnant women. Simultaneously, assess the incidence of adverse pregnancy outcomes (APO) among different serum HBV status in pregnant women. Method: From 2017 to 2019, we studied HBsAg (+) pregnant women and HBsAg (-) who gave birth at our hospital in Guangzhou City, China. We compared of the incidence of pregnant women with HBsAg(+) or HBsAg(-). Further, among HBsAg(+) pregnant women, We compared of the incidence of pregnant women with HBeAg(+) group or HBeAg(-) group, high HBV DNA loads (HBV DNA ≥ 2×10^5IU/mL) group or low HBV DNA loads (HBV DNA (cid:0) 2×10^5IU/mL) group, respectively. Finally, multivariate logistic regression analysis was used to evaluate the independent association between HBV infection and the risk of developing APO. Result: First, Our research Indicates that the rates of gestational diabetes mellitus (GDM), intrahepatic cholestasis of pregnancy (ICP), premature rupture of membrane (PROM), Fetal distress (FD), Oligohydramnios, Premature delivery (PD), Low birth weight (LBW), Meconium contamination (MC), Neonatal hyperbilirubinemia(NH) in HBsAg(+) group were higher than those in HBsAg(-) group (P<0.05). Second, among 711 HBsAg(+) pregnant women, the rates of GDM and ICP in high loads of HBV DNA were higher than those in low loads of HBV DNA group (P<0.05). Similarly, The rates of ICP in HBeAg(+) group were higher than those in HBeAg(-) group. Further, through multivariable logistical regression model analysis, we observed maternal HBsAg carrier (OR, 6.758; 95% CI, 2.358-19.369) had an independent risk for ICP. Similarly, HBsAg carrier(OR, 1.101; 95% CI, 1.066-1.137) ,advanced age (OR, 1.407; 95% CI,1.016-1.137) and abortion(OR,1.446; 95% CI, 1.062-1.969) had independent risk for GDM. Conclusions: Chronic HBV infection can increase the rate of host adverse pregnancy outcomes (APO). The maternal viral load and HBeAg status were signicantly associated with the incidence of GDM and ICP. Maternal HBsAg carrier had an independent risk for GDM and ICP.


Introduction
Chronic hepatitis B (CHB) remains an important public health problem, with approximately 240 million HBV-infected individuals worldwide [1,2] . Among those with CHB infection, approximately 15-40% will further develop more harmful complications such as cirrhosis, liver failure or even hepatoma [3] . According to the World Health Organization (WHO) statistical report, China is one of the major endemic areas for CHB infection, where the prevalence of CHB in individuals under 60 years old is 7.2% [4] . Many studies show that the infection rate of hepatitis B virus is at a high level in Chinese fertile women, around 6.7-8% [5,6] . Previous studies on CHB infection in pregnant women mostly focused on vertical mother-to-child transmission (vMTCT), and viral load was considered to be the biggest risk factor affecting vMTCT. However, there were few studies on whether CHB infection had an impact on the occurrence of APO [7,8] .
Besides the impact of CHB on vMTCT, existing studies have shown that there is a correlation between pregnancy complicated with HBV infection and the occurrence of APO. HBV infection increases the risk of complications such as PD and GDM [9,10,11] .
But, some scholars do not support this view [12,13] . Therefore, whether pregnancy with CHB infection increases the risk of APO is full of controversy.
In addition, existing studies on pregnant patients with HBV infection and APO are insu cient. First, most studies analyze APO from a single aspect (HBsAg positive, HBeAg positive, or DNA viral load); Secondly, in terms of DNA viral load analysis, the lower limit of clinical detection (100IU/ml) is mostly used as the grouping basis, lacking clinically common indicators with high viral load (over 2×10^5IU/mL). Furthermore, because the in uencing factors of APO are complex and diverse, most studies have not further evaluated the other related in uencing factors except HBV infection for APO. Finally, most of the studies were based on methods such as case-control studies and retrospective cohort studies, lacking prospective clinical observation studies.
Considering the above de ciencies, this article conducted a prospective hospital-based cohort study. The objective was to further con rm the in uence of HBsAg, viral load and HBeAg in early pregnancy on APO. To explore the risk factors for APO; To further guide clinical management of pregnant women with HBV infection, and to provide ideas and basis for other related studies.

Study Design and Participant Population
From January 2017 through December 2019, after signing the informed consent, 740 individuals were included in each group of HBsAg(+) and HBsAg(-), and there were 29 HBsAg (+) pregnant women who were lost to follow-up after 6 weeks postpartum, while 65 HBsAg (-) pregnant women who dropped out or lost to follow-up. Finally, we recruited a total of 1386 pregnant women who gave birth at our hospital in Guangzhou, China. Including 711 HBsAg-positive and 675 HBsAg-negative mothers were studied. 151 of the 711 HBsAg-positive women also had high loads of HBV DNA (over 2 10 5 IU/mL), 189 of the 711 HBsAgpositive women had HBeAg-positive. The clinical records of the two groups were retrieved, including age, prenatal weight, parity, history of abortion, newborn sex. From 14 weeks of pregnancy to postpartum week 6, All the mothers were followed.

Statistical Analyses
The baseline characteristics of the patients were reported with the use of descriptive statistics, which includes percentages. In univariate analyses, categorical data were compared by chi-square tests was used to assess the homogeneity of the odds ratios (ORs) and 95% con dence intervals (CIs). logistic regression analysis was used to analyze the association between HBsAg positivity and ICP or GDM. P values of less than 0.05 were assessd to be of statistical signi cance. All analyses was performed using the SPSS version 22.0 software (IBM, NY, USA).

Results
We compared the incidence of adverse pregnancy outcomes under different conditions ( Fig. 1), the speci c results are as follow.
The existing research results show that there is a correlation between HBV infection in pregnancy with APO. HBV infection increases the incidence of PD, GDM, LBW [9,10,11] , which is consistent with our ndings. The reason why CHB infection increases the incidence of some APO may be related to the effect of HBV virus on liver function of inactivating enzymes and hormones. During pregnancy, women produce more endogenous hormones, which will put a heavier burden on the liver. The virus damages hepatocytes, which leads to a relatively high level of estrogen. High level of estrogen will lead to APO [14] . Furthermore, when the placenta and fetal membranes are infected by HBV, the chorionic vessels will change accordingly, causing the blood circulation of the placenta to drop. Reduced intrauterine blood oxygen supply will also increase the risk of APO [15] .
It is noteworthy that further analysis in this study found that the incidence of GDM and ICP in HBsAg(+) pregnant women with high viral load (2×10^5IU/ml) and HBeAg(+) were higher than their control group. This may be related to the maternal excessive in ammatory response. Existing studies have shown that maternal excess in ammation increases the risk of complications during pregnancy [16,17] . HBeAg is a marker of active HBV replication [18] . There was a strong in ammatory response in HBV infected patients with HBeAg (+) or high load of HBV DNA [19,20] . HBV DNA load is an important marker to predict the course of severe complications from HBV immune tolerance [19,21] . Chronic in ammation caused by HBV is associated with insulin resistance. In HBV infected patients, the insulin resistance level is signi cantly higher than the normal population [22] . In addition, HBsAg and HBV DNA were found in the pancreas of patients infected with HBV. These suggest that HBV may cause damage to pancreatic tissue, leading to insu cient insulin secretion [23] .
Logistic model was established to analyze the in uencing factors of APO and it was found that CHB infection could be an independent risk factor for ICP. Many studies also showed that the risk of ICP in pregnant women was higher when HBeAg was positive [24,25] . We think this may be related to the downgrading of the expression of NTCP(sodium taurocholate cotransporting polypeptide). Human NTCP has been identi ed as a functional receptor for HBV. HBV can mediate the infection through the speci c binding of surface antigen [26,27.28] . Meanwhile, NTCP is responsible for the transmembrane transport of sodium and bile acids in liver cells, and is responsible for about 80% of bile acid reuptake [29] . NTCP can transport bile acids to hepatocytes in the enterohepatic circulation and play an important role in the hepatoenteric circulation of cholic acid to maintain the dynamic balance of bile acids. Some studies have suggested that defects in NTCP may lead to intractable hyperbile acidemia [30,31] . In patients with CHB, hepatocytes are constantly destroyed and multiplied. In proliferative hepatocytes, the NTCP expression on cell membrane is decreasing [32] . Moreover, existing research suggest that, ICP is related to PGE2(prostaglandin E2), which will affects the function of natural killer cells [33,34] ; Mutations in genes associated with drug resistance(such as ABCB 11 [35] ,ABCC 2 [36] ,ABCB 4 [37,38] ,NR1H4 [39] ).
To sum up, pregnancy with HBV infection is a serious threat to maternal and child health. It is necessary to pay attention to the health education of pregnant women, the HVB DNA in early pregnancy and the regular examination of liver function during pregnancy(especially the related examination of gestational diabetes and cholestasis). Consulting about potential risks as well as focusing on antenatal surveillance for APO in HBV-infected pregnant women may be necessary.