Associations of gestational weight gain with birth weight outcomes in early pregnancy weight loss women: Findings from the Beijing Birth Cohort Study

To explore the relationship between gestational weight gain (GWG) and birth weight outcomes and establish suggested GWG patterns in early pregnancy weight loss women.


| INTRODUC TI ON
Gestational weight gain (GWG) is directly associated with birth weight and the long-term health of the offspring. 1[4][5] The famous concept of Developmental Origins of Health and Disease (DOHaD) illustrated that poor nutrition during early pregnancy is a risk factor for chronic diseases in later years, 6 which emphasized the importance of early pregnancy management.
However, there is a lack of consensus on the relationships between early pregnancy weight change and pregnancy outcomes.Some studies declared that early pregnancy weight gain independently impacts maternal and neonatal complications and the long-term health of children. 7,8On the contrary, several studies have indicated that the association of early pregnancy weight gain with pregnancy outcomes was influenced by subsequent weight gain across pregnancy. 4,9,10 note, it is reported that about 50% to 70% of pregnant women suffer from nausea and vomiting during pregnancy (NVP).
The severe and persistent form of NVP, like dehydration, ketosis, and weight loss, is defined as hyperemesis gravidarum (HG). 11HG is the leading cause of severe weight loss in early pregnancy.[13][14] However, there is a lack of research on the association between mild and moderate weight loss in early pregnancy and pregnancy outcomes.Furthermore, it's unclear whether women need to catch up in weight gain in mid-late pregnancy after losing weight in early pregnancy and how to define their optimal weight gain patterns.Thus, we aimed to explore the association between GWG and birth weight outcomes in early pregnancy weight loss women.Meanwhile, we tried to establish suggested GWG patterns in the early pregnancy weight loss Chinese population based on GWG-associated perinatal and neonatal outcomes.

| Study design and participants
This study used data collected from the Beijing Birth Cohort Study and conducted a retrospective analysis.The Beijing Birth Cohort Study was a prospective cohort of pregnant women and their children in Beijing Obstetrics and Gynecology Hospital in China and was approved by the ethics committee of the Beijing Obstetrics and Gynecology Hospital (2018-KY-009-01).Informed consent documents must be signed by all participants before being included.For this study, singleton pregnant women aged 18-45 and delivered between January 2014 and December 2021 were included.Women with pre-existing chronic diseases such as diabetes and hypertension were excluded.We also excluded miscarriages, births before 28 weeks, post-term births, and births with missing baseline information, GWG, and perinatal outcomes.The flow chart of the study is shown in Figure 1.

| Measurements and outcomes
Clinical information and offspring anthropometric measurements were collected through medical record review.The prepregnancy body weight was either self-reported or replaced by the weight measured at the first prenatal visit around the 5-6 weeks.Prepregnancy body mass index (BMI) was calculated as prepregnancy weight divided by height squared (kg/m 2 ).According to guidelines for preventing and controlling of overweight and obesity in Chinese adults, 15 prepregnancy BMI was categorized into four groups: underweight (<18.5),normal weight (18.5-23.9),overweight (24.0-27.9),and obese (≥28.0).
Body weight was measured in the hospital at each prenatal visit.
Weight change in early pregnancy was calculated using weight measurements within 16 weeks, because the formal medical record during pregnancy should be registered at 16 weeks in our hospital.According to the Standard of recommendation for weight gain during the pregnancy period (WS/T 801-2022), 16 all participants were firstly divided into three groups based on the early pregnancy weight change.Weight loss in early pregnancy was defined as weight gain less than 0 kg, appropriate weight gain in early pregnancy was defined as weight gain within 2.1-2.6,1.8-2.9,1.7-2.5, and 0-1.8 kg for underweight, normal weight, overweight, and obese women, respectively, 16 the remaining were defined as inappropriate weight gain in early pregnancy.As for the early pregnancy weight loss women, women with apparent weight loss referred to whose weight change in early pregnancy divided by prepregnancy weight more than or equal to 5%, and women with mild weight loss referred to whose rate between 0% to 5%.Meanwhile, all early pregnancy weight loss women were divided into three groups based on their total GWG further.Total GWG within, below, or above the recommended threshold were defined as appropriate, inadequate, or excessive GWG, respectively.
Birth weight was classified as macrosomia, large for gestational age (LGA), low birth weight (LBW), and small for gestational age (SGA).LBW was defined as delivery after 37 weeks but the infant's birth weight was below 2500 g, to make it accurate, the term of "full-term LBW" would be used in this article instead.Macrosomia, SGA, and LGA were defined when the infant's birth weight was above 4000 g, below the 10th centile, and above the 90th centile of the standard growth curve respectively.The cutoff value for SGA and LGA were from the Chinese standards for newborn weight by the Capital Institute of Pediatrics. 17Patterns of GWG included the optimal range and rate of GWG, calculated by the interquartile range (IOR).The subgroup of optimal maternal and neonatal outcomes met the following criteria: (1) birth at ≥37 weeks of gestation, (2) infant birth weight between 10th-90th centile, and (3) no pregnancy complications (gestational hypertensive disorders and gestational diabetes mellitus).Weight gain was divided by the number of corresponding gestational weeks to calculate the GWG rate.

| Statistical analysis
SPSS 25.0 software (IBM, Armonk, NY, USA) was used to analyze all data, and the results are presented as means ± standard deviations (SD) and frequency (%).We compared sociodemographic characteristics in different groups by chi-squared test for categorical variables and analysis of variance (ANOVA) for continuous variables.Multivariable logistic regression analysis was applied to assess the effect of early pregnancy weight change on birth weight outcomes, after controlling for confounders, such as maternal age, prepregnancy BMI, parity, level of education, family history of diabetes and hypertension and so on.
Next, we used the same way to estimate the effect of GWG in early pregnancy weight loss women on birth weight outcomes.In addition, the analyses were repeated in women with apparent and mild weight loss in early pregnancy to ensure the robustness of our findings.
The optimal range and rate were constructed using the IQR of the total GWG and weekly weight gain among individuals in the optimal maternal and neonatal outcomes subgroup.An analysis of multivariable logistic regression was conducted to compare the birth weight outcomes in women whose weekly weight gain was above, below, and within the optimal rate.A P value less than 0.05 was considered statistically significant.

| Patient characteristics
Of the 20 688 women included in the study, the incidence of early pregnancy weight loss was 12.6% (2614/20688), compared to the incidence of early pregnancy non-weight loss which was 87.4% (18 074/20688).
Among non-weight loss women, 3313 women had appropriate weight gain in early pregnancy.Comparing the sociodemographic characteristics of 2614 weight loss women with 3313 appropriate weight gain women, the weight loss women showed a lower mean age and a higher F I G U R E 1 Flow chart of the participants included in the analysis.
Women aged 18-45 who were pregnant with a singleton pregnancy and delivered between January 2014 and December 2021, N=30987 Excluded (N=10299): Pre-existing chronic diseases such as diabetes and hypertension, N=219 2 Miscarriages, births (including stillbirths) before 28 weeks, post-term births more than or equal to 42 weeks, N=440 3 Missing data on baseline information, GWG and perinatal outcomes, N=9640 Eligible pregnancies, N=20688 Weight loss in early pregnancy, N=2614 Non-weight loss in early pregnancy, N=18074

Optimal maternal and neonatal subgroup, N=1121
Non-optimal maternal and neonatal subgroup, N=572 proportion of being underweight and overweight than women with appropriate weight gain in the same period.The subsequent weight gain was also significantly different (P = 0.045) (Table 1).

| The relationship between weight gain in early pregnancy and GWG and birth weight outcomes
Compared to women who gained appropriate weight in early pregnancy, the early pregnancy weight loss women had lower total weight gain value and incidence of excessive GWG, while they showed higher weight gain in mid-pregnancy and the incidence of appropriate GWG (Table 1).In addition, women who had weight loss in early pregnancy had a higher risk of SGA (OR = 1.43, 95% CI: 1.14-1.80,P = 0.002) and lower risk of macrosomia (OR = 0.73, 95% CI: 0.59-0.91,P = 0.005) and LGA (OR = 0.74, 95% CI: 0.64-0.86,P < 0.001).However, there was not a statistically significant difference in all kinds of birth weight outcomes after adjusting for total GWG and other confounders (P > 0.05) (Table 2).

TA B L E 1
Characteristic of baseline information and GWG in participant women according to the early pregnancy weight gain category.
Figure 2a).Women who gained inadequate GWG had a higher risk of SGA (OR = 1.79, 95% CI: 1.25-2.58,P = 0.002), the results were maintained in the subgroup of apparent weight loss women (P < 0.001) but not in the subgroup of mild weight loss women (P = 0.464, Figure 2b).

| Patterns of GWG among the early pregnancy weight loss and normal BMI women
The suggested GWG range was only calculated in normal-weight groups due to the limited sample sizes.About 71.1% (797/1121) and 92.3% (1035/1121) women in the optimal maternal and neonatal outcomes subgroup have returned to the prepregnancy weight up to 20 and 24 weeks of gestation, respectively.Thus, the optimal GWG range after returning to the prepregnancy weight was 8.5 to 13.0 kg, which was the IQR of weight gain between 20 weeks of gestation to delivery.The optimal weight gain range and rate in mid-late pregnancy were 11.0-16.0kg and 0.46 to 0.67 kg/week, which was the IQR of the weight gain range and rate between 16 weeks of gestation to delivery 3).

show the association of weekly weight gain categories
based on the current study with maternal and neonatal outcomes.

| DISCUSS ION
This study found that weight loss in early pregnancy had no significant adverse effects on birth weight outcomes independently.
However, if women lost weight in early pregnancy but had inappropriate GWG, the risk of adverse birth weight outcomes were increased.Furthermore, the suggested GWG range was 11.0-16.0kg and the optimal weight gain rate was 0.46 to 0.67 kg/week from 16 weeks of gestation to delivery for women with normal weight and weight loss in early pregnancy.Our study emphasizes the importance of focusing on the subsequent weight gain and total GWG among early pregnancy weight loss women.

| The relationship between the early pregnancy weight loss, GWG, and birth weight outcomes
In line with our study, several studies suggested that catch-up in weight gain in trimester compensates the effect of weight loss in the first trimester on neonatal birth weight.Thus, women who lost weight in the first trimester had the potential to improve or reverse perinatal outcomes through weight intervention and management in mid and late pregnancy.Naho et al. used data from the Japan Environment and Children's Study (JECS) and included a total of 91 313 singleton pregnant women with hyperemesis gravidarum.They found that women who lost weight ≥5% in the first trimester were more likely to develop SGA than women who gained >3%.However, when the weight loss women catch-up in weight gain later in the second trimester, they had infants with larger birth weights and a lower risk of SGA. [9]Furthermore, a study from the US found that achieving a reference trajectory in the second or third trimester normalized the risk of SGA for women weight loss in the first trimester. 4wever, a study from Norway performed on 892 women with hyperemesis gravidarum had different results.They showed that insufficient early weight gain predicted SGA independently, and weight change during the latter half of pregnancy and total GWG were not associated with birth weight. 12Of note, subjects in this study lost a mean of 4.4 kg absolute weight in early pregnancy, which was higher than the mean of 2.4 kg weight loss in our study.The higher absolute weight loss may explain the different results.
In addition, the different views may be associated with the great differences in the prepregnancy BMI distribution and recommended GWG values in different countries and regions.According to previous studies, the mean prepregnancy BMI of Japanese women was 21.1 kg/m 2 , and overweight and obese women accounted for 4%-10%. 9,18However, the proportion of overweight and obese before pregnancy in Norway women was nearly 40% and the mean prepregnancy BMI of women who had NVP was TA B L E 3 Gestational weight gain and weekly weight gain in early pregnancy weight loss women with normal BMI and optimal maternal and neonatal outcomes.24.4 kg/m 2 . 12which is higher than the level in Asia.According to the Japanese standard, women with normal prepregnancy BMI were recommended to gain 10.0-13.0kg during the pregnancy, 19 while it was 11.5-16.0kg according to the IOM standard used in Europe and the USA.In this study, the optimal range and rate were constructed using the IQR of GWG and weekly weight gain, the statistics-based approach, which is a command strategy for calculating optimal value and being validated in clinical studie. 21,22As for our results, the optimal weight gain range in the mid-late pregnancy is 11.0-16.0kg and the optimal weekly weight gain is 0.46 to 0.67 kg/week from 16 weeks to delivery for normal prepregnancy BMI women.They are increased by one-third compared to the Chinese recommendations of 8.0-14.0kg and 0.26 to 0.48 kg/week. 16This new Chinese standard of recommendation has been proven more suitable for Chinese people than the US National Academy of Medicine guidelines. 23In addition, this phenomenon of catch up weight gain in mid-late pregnancy corroborates with the previous study in our group, which showed that obese women with inadequate weight gain in the first trimester had paradoxically more rapid weekly weigh gain after 20 weeks. 24According to our optimal weight gain rates, women who had higher weight gain rate increased risk of macrosomia and LGA, which suggested that the early pregnancy weight loss women also need to be aware of excessive weight gain.Due to the low incidences of SGA and full-term LBW in our study, we did not find a relationship between lower weight gain rate and the risk of SGA and full-term LBW.

| Strengths and limitations
This is the first study describing the characteristic of subsequent weight gain and exploring the relationships between GWG and birth weight outcomes in early pregnancy weight loss women in a large Chinese population.Similar studies in the past were conducted on women with hyperemesis gravidarum and extreme weight loss in early pregnancy, our study first include both mild and apparent weight loss women for analysis.Meanwhile, we provide the suggested GWG patterns and optimal cutoff weight gain value derived from birth weight outcomes for the first time.
However, our study had certain limitations.First, we relied on self-reported prepregnancy weight data.Previous studies reported that women tend to underestimate their weight, 25 resulting in a misclassification of women in the different BMI groups and overestimating of weight gain throughout the pregnancy.Meanwhile, due to the schedule of prenatal visit in our hospital, body weight at 12 or 13 weeks was not available for most participants, so we took 16 weeks as the cutoff for early pregnancy intentionally and used weight at 16 weeks to calculate the weight change in early pregnancy instead.Second, we could not provide suggested GWG patterns in overweight and obese women due to the limited sample size.Third, the confounding factors such as diet and physical activity habits were unavailable in our study.

| CON CLUS ION
Weight loss in early pregnancy is not the independent risk factor of birth weight outcomes.GWG may offset the expected effect of weight loss in early pregnancy on birth weight outcomes.Compared with the recommended weight gain rate for general women, women with normal BMI and weight loss in early pregnancy suggest similar GWG ranges and a weight gain rate increased by no more than onethird in the subsequent pregnancy in order to get optimal maternal and neonatal outcomes.In the future, our results need to be con-

K
E Y W O R D S birth weight outcomes, catch-up in weight gain, early pregnancy, gestational weight gain, weight loss

3. 3 |
Figures 2 shows the relationships between different GWG categories among the early pregnancy weight loss women and birth Abbreviation: BMI, body mass index.

4. 2 |
Patterns of GWG among the early pregnancy weight and normal BMI women firmed in a larger and more diverse population.The medical staff should provide individualized weight interventions for early pregnancy weight loss women and pay more attention to their weight changes in mid-late pregnancy by monitoring their body weight during pregnancy regularly.
The relationship between weight gain in early pregnancy and birth weight outcomes.
Abbreviations: BMI, body mass index; GWG, gestational weight gain.TA B L E 2Note: Model 1: Multivariable logistic regression analysis was adjusted for maternal age and prepregnancy BMI.Model 2: Multivariable logistic regression analysis was adjusted for maternal age, prepregnancy BMI and GWG.Reference group: appropriate weight gain in early pregnancy group.Abbreviations: BMI, body mass index; full-term LBW, full-term low birth weight; GWG, gestational weight gain; LGA, large for gestational age; SGA, small for gestational age.

3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 aOR (95%CI)
0% to 5%.aOR, adjusted odds ratio; full-term LBW, full-term low birth weight; GDM, gestational diabetes mellitus; HDP, hypertensive disorders of pregnancy; LGA, large for gestational age; SGA, small for gestational age.Macrosomia: Multivariable logistic regression analysis was adjusted for maternal age, level of education, prepregnancy BMI, parity, family history of diabetes and hypertension, gestational age at delivery and GDM.LGA: Multivariable logistic regression analysis was adjusted for maternal age, level of education, pre-pregnancy BMI, parity, family history of diabetes and hypertension and GDM.Full-term LBW: Multivariable logistic regression analysis was adjusted for maternal age, level of education, prepregnancy BMI, parity, family history of diabetes and hypertension, gestational age at delivery and HDP.SGA: Multivariable logistic regression analysis was adjusted for maternal age, level of education, prepregnancy BMI, parity, family history of diabetes and hypertension and HDP.Reference group: appropriate GWG group；*P < 0.05; **P < 0.001.

Outcome Weekly weight gain categories among the early pregnancy weight loss and normal BMI women
Reference group: 25th-75th % group.Abbreviations: aOR, adjusted odds ratio; full-term LBW, full-term low birth weight; GDM, gestational diabetes mellitus; HDP, hypertensive disorders of pregnancy; LGA, large for gestational age; SGA, small for gestational age.Multivariable logistic regression analysis was adjusted for maternal age, level of education, prepregnancy BMI, parity, family history of diabetes and hypertension, and gestational age at delivery.Multivariable logistic regression analysis was adjusted for maternal age, level of education, prepregnancy BMI, parity, family history of diabetes and hypertension, GDM, and HDP.
a b c Multivariable logistic regression analysis was adjusted for maternal age, level of education, prepregnancy BMI, parity, family history of diabetes and hypertension, gestational age at delivery, GDM, and HDP.d Multivariable logistic regression analysis was adjusted for maternal age, level of education, prepregnancy BMI, parity, family history of diabetes and hypertension, gestational age at delivery, and GDM. e Multivariable logistic regression analysis was adjusted for maternal age, level of education, prepregnancy BMI, parity, family history of diabetes and hypertension, gestational age at delivery, and HDP.f Multivariable logistic regression analysis was adjusted for maternal age, level of education, prepregnancy BMI, parity, family history of diabetes and hypertension, and GDM.g Multivariable logistic regression analysis was adjusted for maternal age, level of education, prepregnancy BMI, parity, family history of diabetes and hypertension, and HDP.*P < 0.05.**P < 0.001.