Hepatitis B surface antigen positive on pregnancy outcomes and BMI z-score in offspring: a population-based study, China, 2011-2018

Background: Hepatitis B surface antigen (HBsAg) positive leads to pregnancy outcomes and effects on offspring BMI z-score remains unclear in China. We are aim to explore if there is an association among HBsAg positive, BMI z-score, and pregnancy outcomes. Methods: We extracted the characteristic information and pregnancy outcomes of pregnant women, and BMI z-score from 1 to 3 years in offspring from the Medical Birth Registry of Xiamen that registered between 1 March 2011 and 30 March 2018. Results: Pregnant women with HBsAg positive had a higher risk of gestational diabetes mellitus (GDM) (1.128, 1.032-1.232), and cesarean delivery (1.116, 1.026-1.213). Interestingly, there is no difference between HBsAg positive and offspring BMI z-score (all P > 0.05). Additionally, the GDM women with HBsAg positive had higher incidence of low-for-gestation age (LGA), preterm birth, and cesarean delivery (22.95%, 9.48%, and 47.88%, respectively) (all P < 0.001). Conclusions: HBsAg positive may be a risk factor of GDM and cesarean delivery. However, HBsAg positive has no effects on offspring BMI z-score. Therefore, women should perform GDM and HBsAg screening before or during pregnancy. As well, a larger follow-up population study should be performed to confirm the association between HBsAg positive and offspring BMI z-score. 25.04%), P no for family of family history P 0.05). We indicated that the birth weight of pregnant women in GDM-HBsAg+ or NGDM-HBsAg+ group was lighter than GDM-HBsAg- or NGDM-HBsAg- group (3,165.5532.2 or 3,176.8453.6 vs. 3,195.5508.9 or 3,184.3460.8, P = 0.313). Whereafter, the proportion of LGA infants in GDM-HBsAg+ group (22.95%) or GDM-HBsAg- (22.70%) group was higher than NGDM-HBsAg+ (16.85%), and NGDM-HBsAg- (15.87%) group, P < 0.001. In addition, the proportion of preterm birth in GDM-HBsAg+ group and GDM-HBsAg- group was larger than NGDM-HBsAg+, and NGDM-HBsAg- group (9.48% and 7.34% vs. 5.32% and 5.18%, P < 0.001). Moreover, the percentage of stillbirth in GDM-HBsAg+ group and GDM-HBsAg-group was less than NGDM-HBsAg+ and NGDM-HBsAg- group (2.44% and 3.41% vs. 4.34% and 4.24%, P = 0.036). Last but not least, the percentage of cesarean delivery in GDM-HBsAg+ group was highest (47.88%) compared with GDM-HBsAg- group (43.76%), NGDM-HBsAg+ group (36.05%),


Introduction
Hepatitis B virus infection is a significant public health problem in the world, leading to social burden worldwide and high mortality [ 1 ]. It is reported that 240 million people are infected with hepatitis B virus, causing six hundred thousand deaths each year [ 2 ]. The largest burden of hepatitis B virus exists in China, with 74.6 million people infected [ 3 ]. An estimated 3.87-9.98% pregnancy women show hepatitis surface B antigen (HBsAg) positive [ 4 ]. Moreover, the rate of preterm birth is second largest in China, where 1.17 million babies are born before 37 weeks of gestation. Additionally, incidence of GDM is increasing in the world as obesity becomes more common [ 5 ]. Gestational diabetes mellitus (GDM) is characterized as impaired glucose intolerance with first recognition during pregnancy [ 6 ]. Study showed that 2 to 9 percent of pregnancies suffer from GDM, and is linked to substantial rates of perinatal or maternal complications [ 7 ]. Furthermore, GDM is related to serious adverse outcomes for maternal and their infants with huge health care burden. The adverse outcomes include birth trauma, fetal macrosomia, caesarean delivery, stillbirth, and preeclampsia [ 8 ] .
The relationship between hepatitis B virus infection and diabetes mellitus (DM) is still unclear [ 9 ]. However, the liver plays an important part in glucose metabolism [ 10 ]. Several studies investigated the specific relationship between HBsAg positive and GDM in pregnant women. Nevertheless, the results are distinct and even contradictory. Researches show that the association between HBsAg positive and GDM is not identified during pregnancy in Asian and American [ 4 , 11 ]. However, another study reported that the HBsAg positive increased the risk of GDM in Hong Kong people [ 12 ]. Hepatitis B virus infection is a major social and economic burden in developing country, especially, in China. However, HBsAg positive in pregnant women is an acknowledged issue that leading to some adverse pregnant outcomes remains unclear.
However, there are few studies researching the influence of pregnant women not only with GDM but also with HBsAg positive on pregnant outcomes. Meanwhile, there are no researches of associations between HBsAg positive and offspring BMI z-score. Therefore, we performed this study to explore the association among HBsAg positive, offspring BMI zscore, and pregnant outcomes.

Patients
Pregnant women with or without HBsAg positive were included in this study. We extracted information of pregnant women and their offspring who registered the Medical Birth Registry of Xiamen between 1 March 2011 and 30 March 2018, which related to Xiamen citizen health information system. Each citizen has a unique identification number at birth in Xiamen, China.

Registered data
The included information of the Medical Birth Registry of Xiamen as following: 1) maternal characteristics: age, weight, height, body mass index (BMI), obstetric history, education, family history of hypertension and DM, insulin treatment, systolic blood pressure, diastolic blood pressure, fasting plasma glucose (FPG) and fetus frequency, oral glucose tolerance test (OGTT); 2) birth outcomes: preterm birth, stillbirth, macrosomia, low birth weight, large-for-gestational age (LGA) infant, small-for-gestational age infant, cesarean delivery, and birth weight; 3) delivery characteristics: GDM, HBsAg, and gestational weight gain.
Meanwhile, BMI was counted by weight (kg) divided by height (m) squared, which divided into four groups ( 18.5 kg/m 2 ; 18.5-24.9 kg/m 2 ; 25.0-27.9 kg/m 2 ; 28 kg/m 2 ). Maternal age was segmented into five groups ( 25 years old; 25-29 years old; 30-34 years old; 35-39 years old; 40 years old). Frequency of delivery was divided into two groups (1 times; 2 times). The level of education of maternal was segmented into two groups ( 9-year compulsory education; 9-year compulsory education). OGTT performed at three different time points (0 h, 1 h, and 2 h). Besides, the MBRX for children begins with children's birth, including information from newborns to preschool (date of birth, sex, gestational week of birth, weight, Apgar score, family history of diseases, feeding modalities, etc).
Diagnosis criteria of pregnancy outcomes GDM diagnosis criteria were referred to the 2014 National Health and Family Planning Commission of the People's Republic of China criteria. When 75 g OGTT outcomes met or exceeded following plasma glucose value, the pregnancy can diagnosed with GDM. 0 h, 5.1 mmol/L; 1 h, 10.0 mmol/L; 2 h, 8.5 mmol/L. A 75 g OGTT was performed between the 24th and 28th weeks of gestation for all pregnancy women who did not previously know to suffer from diabetes. The test results were still valid even after 28 weeks. Macrosomia was diagnosed with birth weight more than 4000g. LGA was ascertained by birth weigh more than 90 percentile for gestational age. Small-for-gestation age referred to birth weight less than the 10th percentile for gestational age. What's more, the WHO weight percentile calculator (3,542 437g) was used to count for babies born at 24 to 41 weeks' gestation.
Preterm birth was defined as giving birth earlier than 37 weeks of pregnancy.

Statistical analyses
Statistical analyses were conducted with SPSS 18.0 (SPSS Inc, Chicago, IL, USA). The level of significance for all tests were two-tailed and P < 0.05. Continuous variables were showed as Mean SD that compared via Student t test or one-way ANOVA. Discontinuous variables were expressed as n (%) and compared by Pearson's Chi-square (2) test.
Multivariable logistic regression was used for multivariate analyses on account of models containing the factors to assess the association among HBsAg positive during pregnancy, GDM, and pregnancy outcomes.

Ethics statements
This study was approved by the Review Broad of the First Affiliated Hospital of Xiamen University, and informed consent was not required given the retrospective nature of this study. This study was conducted in accordance with the Helsinki declaration.

Results
Characteristics of pregnant women with or without HBsAg positive 28.94.4, P < 0.001). Furthermore, the number of pregnant women with different age indicated that pregnant women were older than 30 years old in HBsAg positive was larger than negative (42.04% vs. 37.91%, P < 0.001). In addition, compared with HBsAg negative, the FPG in pregnant women with HBsAg positive was lower (4.70.5 vs. 4.80.5, P < 0.001). Moreover, the incidence of OGTT value was abnormal for pregnant women with HBsAg positive were higher than negative (19.98% vs. 17.75%, P < 0.001). As well, the frequency of fetus more than 2 times, the chance of pregnant women with HBsAg positive was greater than negative (63.10% vs. 60.05%, P 0.001). Although the proportion of pregnant women with HBsAg positive accepted education was less than 9 years was larger than negative (28.03% vs. 25.04%), there was no significance (P = 0.054). Besides, compared with HBsAg negative, there were no significances existed in positive group for BMI, family history of DM, and family history of hypertension (all P > 0.05).
Pregnant outcomes of pregnant women with or without GDM and HBsAg status  Table 2).

Effects of HBsAg status on pregnant outcomes
According to univariable logistical regression analysis, we found that the HBsAg positive was risk factor for GDM (OR, 1.157; 95% CI, 1.063-1.259) and caesarean delivery (OR, 1.131; 95% CI, 1.047-1.222), all P < 0.01 (Table 3). What's more, a multiple variable logistical regression was also used to ensure independent risk factors for the pregnancy outcomes. Results showed that the HBsAg positive was independent risk factor for GDM (OR, 1.128; 95% CI, 1.032-1.232; P < 0.001) based on adjusted variables including age, BMI, antibiotic, and tocolytic agent. As well, the cesarean delivery (OR, 1.116; 95% CI, 1.026-1.213; P = 0.011) was also an independent risk factor after adjusting age, BMI, fetus frequency, insulin, GDM, and antibiotic variables.
Associations between HBsAg positive and offspring BMI z-score from 1 to 3 years We compared the offspring BMI z-score according to maternal HBsAg status in different age that indicated there is no association between maternal HBsAg positive and offspring BMI z-score. Table 4 shows the comparison of children's BMI Z-score from 1 to 3 years of age according to HBsAg status in pregnancy. After adjustment for offspring sex, maternal age, education, and infant feeding, and maternal gestational weight gain (multivariable-adjusted Model 2), offspring exposed to HBsAg positive had no significant mean values of BMI Z-score at 1, 2, and 3 years of age, in comparison with those unexposed to GDM(all P > 0.05). When introducing maternal pre-pregnancy BMI into the model (multivariable-adjusted Model 3), the differences were still not statistically significant among children of mothers with or without HBsAg positive.

Discussion
This study is the first research to investigate the association among HBsAg status, offspring BMI z-score, and pregnancy outcomes in China. We found that the GDM women with HBsAg positive who were older than normal pregnancy or non-GDM women with HBsAg negative. The result was consistent with the study, which also showed that the hepatitis B virus infected women were more likely to be older in age [ 13 ]. Moreover, the research also expressed that number of abnormal blood glucose in pregnancy women with HBsAg positive was higher than negative. A large-sample crosssectional research revealed that compared with patients with HBsAg negative, the patients with HBsAg positive were more likely to develop DM [ 14 ]. In light of the aforementioned outcomes, hepatitis B virus infection might be a potential risk factor for DM. The abnormal blood glucose existed in pregnancy women with HBsAg positive could be explained by several mechanisms. At first, the liver was a key role for regulating glucose homeostasis. Liver damaged by hepatitis B virus might cause a glycometabolism disorder, and inflammatory activities might lead to defective glucose homeostasis. In addition, some studies identified hepatitis B virus infection in pancreas.
Hepatitis B virus replications happened in extrahepatic parts, like pancreas, was responsible for causing DM and -cell damage. Secondly, insulin resistance could also be associated with the pathogenesis of hepatogenous diabetes [ 15 ]. In the present study, the higher percent for HBsAg positive with abnormal blood glucose might be illustrated by above mechanisms.
In our analysis, we discovered that the BMI of GDM women with HBsAg positive and negative groups were higher than of those without GDM. Compared with normal-weight women, the overweight pregnant women were at the risk of developing GDM[ resistance, leading to hyperglycemia. Pregnancy women were obesity was a key risk factor for GDM, with a study clarified the OR of developing GDM to be 2.6 for obese women [ 17 ]. What's more, another research revealed overweight-pregnant women developed GDM was 1.7 times compared with normal weight. A meta-analysis declared developing GDM was incremental 3.6 times in obesity, and 8.6 times in the severely obesity compared with normal-weight women[ 18 ]. This was in accord with our result for the weight of GDM women was heavier than non-GDM women. Meanwhile, we found that the percentage of fetus frequency was more than 2 times was largest in GDM women with HBsAg positive compared with other three groups. At present, there are not significant evidences to figure out this finding. We proposed that the multipara with HBsAg positive increased risk for GDM.
The study also suggested that GDM women with HBsAg positive had higher incidence of LGA and stillbirth for infants compared with other groups. For untreated GDM women, their infants were at high risk of LGA, which might be related to delivery injury, respiratory It was showed that HBsAg positive was an independent risk factor for GDM after adjusting multivariable in this analysis (OR, 1.124; 95% CI, 1.029-1.228). A possible mechanism for women during pregnancy with HBsAg positive developed GDM was that hepatitis B virus infection could cause insulin resistance, potentially in that tumor necrosis factor alpha. Secondly, this study only analyzed the association between HBsAg positive and maternal outcomes, the HBeAg was not examine. Thirdly, there is less follow-up data of offspring BMI z-score. Our further research should focus on the effect of HBeAg on maternal outcomes and expand the follow-up contents and sizes.

Conclusions
Our research supports that HBsAg positive is a risk factor of GDM and caesarean delivery.
In addition, the GDM itself also has an effect on pregnant outcomes, such as preterm birth, LGA infant, and cesarean delivery. It is suggests that early perform hepatitis B virus and diabetes mellitus screening before or during pregnancy is needed that could improve preterm birth and caesarean delivery risk to reduce child mortality. Meanwhile, large and randomized controlled trials are needed to investigate the effect of HBeAg on the pregnancy outcomes and HBsAg positive on offspring BMI z-score.

Ethics statements
This study was approved by the Review Broad of the First Affiliated Hospital of Xiamen University, and informed consent was not required given the retrospective nature of this study. This study was conducted in accordance with the Helsinki declaration.

Conflict of interests
The authors declare that they have no interests.