Folate deciency in early pregnancy increases the risk of low birth weight: survey of neonates in eastern China

Low birth weight (LBW) is a major cause of fetal mortality and morbidity. This study aims to assess the relationship between maternal serum folate levels and LBW in early pregnancy in eastern China. We conducted a retrospective study including 124 newborns with LBW (born ≥ 1500g to < 2500 g) and 393 normal birth weight neonates (NBW). The maternal methylenetetrahydrofolate reductase (MTHFR) C677T polymorphism genotyping was performed by the gene chip hybrid method. The concentrations of serum homocysteine (hcy), folate, vitamin B12 and vitamin D were measured before 18 weeks of pregnancy. Multivariable logistic regression model was used to identify the predictors of LBW.

MTHF irreversible (5). Its C677T variants make the amino acid changed from alanine to valine, which results in a reduced enzymatic activity. Evidence by Frosst et al (6). showed that compared with MTHFR 677CC genotype, MTHFR 677CT genotype retained nearly 65% of the enzymatic activity, and MTHFR 677TT genotype retained only approximately 30%. It was proved that the mutation of MTHFR C677T can lead to the decreasing of the folate utilization and hyper-homocysteinemia (HHcy) (7). Hcy is considered to damage the vascular endothelium, destroy the coagulation and brinolytic system, ultimately lead to adverse birth outcomes (8,9). Statistics show folate de ciency is ubiquity in a quarter of pregnant women worldwide (10). Folate de ciency has been linked to neural tube defects (NTD) (11,12), whereas its impact on other severe pregnancy outcomes is inconclusive. Some studies provide support to maternal MTHFR 677TT genotypes, together with maternal circulating folate and homocysteine (hcy) concentrations are independent risk factors for small gestational age (SGA) (13)(14)(15). Other observational studies found no direct association between low folate levels and pregnancy complications (16)(17)(18). To date, the role of folate intake to prevent NTD is veri ed (19,20). Most relevant studies adopted self-statement of multivitamin supplementation, which may be imprecise, so serum folate concentration during pregnancy would be more ideal. Moreover, few studies focused on folate de ciency in early pregnancy in developing countries. In consideration of rural districts, folate de ciency was de ned as < 6.8 nmol/ml and vitamin B12 de ciency as < 150 pmol/ml, which are recommended by WHO (21). As majority of the women have multivitamin supplement during the rst three-mouths pregnancy in urban area of China, the cut-offs of micronutrient (Vitamin B12 and folate) are inappropriate.
Given above elements, we collected a cohort of hospital-delivered neonates to explore the relationship between maternal folate status and LBW in eastern Chinese pregnant women before 18 weeks of gestation.

Study population
Hospital-delivered newborns were enrolled between April 2018 and October 2020 in the Obstetrics and Gynecology Hospital of Fudan University, including a total of 124 newborns with LBW and 393 NBW.
Epidemiological data were collected for pregnant women covered maternal age at delivery (<35, ≥35 years), pre-pregnancy body mass index (BMI), maternal education level (no education or primary education, secondary education, higher education), gravidity and parity.
The BMI was calculated by weight before pregnancy /(height 2 ) and grouped into four categories according the China guidelines (underweight, <18.5, normal weight, 18.5-23.9, overweight, 24.0-27.9; obesity, ≥28.0). Preeclampsia was de ned as PRH (SBP≥140 mmHg or DBP≥90 mmHg) with proteinuria (24h urinary protein level of >0.3 g or urine dipstick protein level≥+). GDM was diagnosed by oral glucose tolerance test (fasting plasma glucose≥5.1 mmol/L, 1h plasma glucose≥10.0 mmol/L, or 2h plasma glucose≥8.5 mmol/L). The PTB was de ned to deliver within 37 weeks of gestation. Gestational age of serum sampling was determined by the result of ultrasound. Serum biochemical parameters were measured within 18 weeks of gestation. Gestational age, delivery mode (eutocia vs. cesarean), gender and birth weight of newborns were obtained from medical records.
Pregnant women were excluded if they were under 18 years old, smoking or drinking, multiple pregnancy, embryo transfer, in vitro fertilization, and microbial infection. Neonates with birth defects, spontaneous abortion, birth weight <1500 g or gestational age <30 weeks were also excluded in this study.

Detection of MTHFR C677T polymorphism
Genomic DNA was extracted from the whole blood of the pregnant women using a column extraction kit. The MTHFR C677T polymorphism was detected by gene chip hybrid analysis following the manufacturer's instructions. The polymerase chain reaction (PCR), hybridization, gene array detection and analysis were conducted using the BaiO genotype detecting gene array kit and equipment (BaiO Technology Corp., Shanghai). The MTHFR C677T was genotyped as wild (CC), heterozygous (CT) and homozygous (TT) gene type, respectively.

Measurement of biochemical parameters
The biochemical parameters were measured before the 18th week of pregnancy. Plasma hcy was determined using Hitach 7600 automatic chemistry analyzer (Hitach Diagnostics Ltd.). Serum folate, vitamin B12 and vitamin D concentrations were measured using an Architect i2000 Analyzer (Abbott). The cut-off values for distribution of metabolic parameters considered in this study were hyper hcy ≥8μmol/L, folate de ciency <32.5 nmol/L, vitamin B12 de ciency <280 pmol/L, and vitamin D de ciency <30 nmol/L, respectively.

Statistical analysis
Statistical analyses were performed using the SPSS version 18.0 software (SPSS Inc., Chicago, IL, USA). Variable selection in multivariable modeling was based on clinical and statistical signi cance.
Continuous variables were expressed as mean ± standard deviation (SD) and categorical variables were indicated as number and percent (n, %). Independent-sample t test or chi-square test was used to evaluate the difference in clinical characteristics. A two-sided p-value less than 0.05 was considered to be statistically signi cant. Multivariable logistic regression analysis was used to determine the adjusted odds ratios (AOR) with 95% con dence intervals (CI) for the associations between folate de ciency and LBW, adjusted for maternal age, pre-pregnancy BMI, MTHFR 677T, PE, delivery mode, parity, and gestational age.

Socio-demographic Characteristics
A cord of 517 pregnant women with neonate were included. Of these, 23.98% (124/517) were LBW babies. The baseline characteristics are summarized in Table 1. There was a large difference in birthweight between LBW (2164.90 ± 227.53 g) and NBW (3350.71 ± 415.75 g) neonates. Gestational ages were also lower in LBW newborns (249.72 ± 15.89) than NBW (274.06 ± 9.84). The proportion of underweight, MTHFR T allele, folate de ciency, primipara, cesarean and preeclampsia of LBW were higher than NBW group. No other differences in maternal or neonatal characteristics were observed, including maternal age (p = 0.674), education level (p = 0.937), GDM (p = 0.352) and gender (p = 0.094). Besides, there were no difference with the proportion of pregnant women being de cient in vitamin B12 (23.4% Vs 24.4%) and vitamin D (28.2% Vs 21.6%) and HHcy (29.8% Vs 25.7%).    (Table 3). The proportion of LBW was about 23.98% because we oversampled LBW at enrolment. As a result, the likelihood of delivering a LBW newborn would be less as compared to pregnant women in rural areas (26,28).
A majority of studies principally emphasized the association of maternal folic acid supplementation and fetal birth weight (33)(34)(35). Our study focused on serum folate levels before 18 weeks of pregnancy and evaluated the relationship with LBW. In our individuals, serum folate concentrations were generally high, with a mean value of 37.71 nmol/l. Beside the previously known factors (prematurity, preeclampsia) (36,37), the results showed that folate de ciency (< 32.5 nmol/l) before 18 weeks gestation was an independent risk factor for LBW in eastern Chinese pregnant women. This is more accurate than studying the relationship between folic acid supplementation and LBW. Similarly, our result indicated that the maternal preeclampsia in late pregnancy was found to be an independent factor contributing to LBW. The probability of LBW was 6-folds higher among mothers having preeclampsia than those not having PE.
This nding is in line with studies conducted in Ethiopia and Nepal (29,38). This might due to the decreased oxygen and nutrient perfusion to the placenta(30).
Our nding indicated that MTHFR 677T and folate de ciency were associated with LBW, not including vitamin B12 and D de ciency or elevated Hcy. Seemingly our ndings were in disagreement with other results, which show that elevated Hcy and lowered folate levels during pregnancy are associated with LBW (39,40). In fact, Hcy levels were signi cantly higher in individuals having 677T allele of MTHFR, while folate level was distinctly lower. However, following the binary logistic regression analysis, it turns out that folate de ciency but not MTHFR 677T was independent risk factor. This result might due to the different cut-off value of serum parameter levels. Here we de ne it as lower range of the parameter spectrum. Besides, no statistically signi cant difference was found in the education level of our individuals. However, a retrospective analysis of the Nigeria population found maternal characteristic (education) was a predictor of LBW, including a cohort of 9,244 live births (41). Another case-control study in Blacks and Whites reported maternal educational attainment was inversely associated with LBW overall, including 2,922 births to Black (n = 2,146) and White (n = 776) mothers) (42). In this study, the rate of secondary school and higher education graduates were 94.97%. We thought that the increased awareness of adequate multivitamin nutrition intake was associated with increased education levels. These two ndings were contrary to our results, which may be due to the overall high level of education and low sample size of low education level pregnant women in our area.
It was reported that the risk of LBW in women with pre-pregnancy low BMI was signi cantly increased. One birth cohort survey of 3401 neonates in Pomerania found that compared to women with normal prepregnancy BMI, underweight women had an increased chance of premature labour and low birth weight (AOR = 1.73; 95% CI: 1.29-2.31)(43). Our result was consistent with this phenomenon (AOR = 5.15; 95% CI: 2.40-11.05). In this research, we also found another interesting phenomenon. The likelihood of LBW was higher in primipara compared with multipara, suggesting multipara as a protective impact on LBW. This might be due to the fact that multipara women might have a better preparation folic acid supplementation and timely detection and treatment of folate statue. As a result, the likelihood of having a low birth weight baby would be less as compared to primipara. This nding supports that timely supervision, multivitamin supplement, and folate determination are important during early pregnancy.
There were several certain strengths and limitations in this study. We did not recruit babies with birth asphyxia or antenatal infection, delivery before 30 weeks gestation and/or < 1500 g birthweight. It allows us to con dently exclude any clinically severe conditions. In additions, the data were collected from medical records and avoided recall bias of self-report. In view of the fact that almost all mothers of our cases added folic acid or multivitamins, supplementation before pregnancy was not investigated, which could modify the level of plasma parameters. Further studies are warranted to examine other factors. The data was collected in eastern China so the results cannot generalizable to other zones.

Conclusions
Our study con rms a positive association between adequate serum folate concentrations before 18th pregnancy and increased birth weight. Based on our ndings, maternal adequate folate status (> 32.5 nmol/L) can reduce the risk of LBW. Further investigations with high-quality prospective data are required to explore the mechanisms. Clinicians should pay more attention to folate de ciency pregnant women and strengthen the supervision of early pregnancy, which has importance for newborn health.
Abbreviations LBW: Low birth weight; NBW: normal birth weight neonates; MTHFR: methylenetetrahydrofolate reductase; Hcy: homocysteine; PTB: preterm birth; SGA: small gestational age; NTD: neural tube defects; BMI: body mass index; PCR: polymerase chain reaction; AOR: adjusted odds ratios; CI: con dence interval Declarations Ethics approval and consent to participate The current retrospective study protocol was approved by the Medicine Ethics Committee of Obstetrics and Gynecology Hospital of Fudan University (No. 2020-54). All women written informed consent in this study.

Consent for publication
Not applicable Availability of data and materials The datasets used and analysed during the current study are available from the corresponding author on reasonable request YC analyzed and interpreted the data and was a major contributor in writing the manuscript. CY critically reviewed and modi ed the manuscript. All authors read and approved the nal manuscript.