Associations Between Pre-Pregnancy Body Mass Index, Gestational Weight Gain and Preterm Birth: a Cohort Study in Wuhan, China

Background Preterm birth (PTB) is the leading cause of neonatal mortality and morbidity worldwide. Methods This cohort study was designed to investigate the associations between pre-pregnancy BMI, total gestational weight gain (GWG), and GWG during early pregnancy with PTB utilizing data of 83,096 Chinese women from the Maternal and Children Healthcare Information Tracking System of Wuhan, China. Results Women who were underweight, overweight or obese prior to pregnancy had an overall elevated risk of PTB, compared to their normal weight counterparts. Women with total GWG below the IOM recommendation had an increased risk of PTB compared to women who had GWG within the recommendation, whereas an increasing risk of PTB was observed asweekly early pregnancy GWGincreased. When stratied by subtypes of PTB, pre-pregnancy underweight was associated with higher risk of spontaneous PTB, and pre-pregnancy overweight /obese increased the risk of both spontaneous PTB and medically indicated PTB. Women with total GWG below the IOM recommendation had elevated risk for spontaneous PTB and PROM, and women with GWG above the recommendation had decreased risk for all three subtypes of PTB, whereas risk for the three subtypes of PTB increased along with increasing weekly GWG of early pregnancy. Conclusions should be considered in combination to reduce the risk of PTB, women should modify their weight gains during pregnancy according to the results.


Background
Preterm birth (PTB), de ned as a delivery of live born infant before 37 completed gestational weeks [1], is the leading cause of neonatal mortality and morbidity worldwide [2], and has been reported to be strongly associated with long-term health problems such as neurological disabilities and various chronic diseases [3,4]. During the recent decades, the burden of preterm birth is substantial and increasing [5].
Therefore, it is important to identify the potential modi able risk factors for prevention of preterm birth.
However, as a complex phenomenon, the etiology of preterm birth is not yet well understood to date.
Several previous studies have indicated that the maternal overweight/obesity is one potential modi able risk factor for PTB [6,7], and thus provided a target for intervention of PTB during pre-conception care.
Besides, as weight control is considered to be more feasible during pregnancy than before conception, there is increasing concern about the association of gestational weight gain (GWG) with PTB. However, conclusions of previous investigations have been inconsistent, as several studies reported an association between lower GWG and elevated risk of PTB [8], while some studies indicated that risk of PTB increased with higher GWG [9].
Most of the previous studies only evaluated the GWG by weight data throughout pregnancy, which may lead to biased associations because GWG differs by term and preterm birth [10], and is not linear throughout pregnancy [11]. GWG during early pregnancy was considered to be critical for embryogenesis and fetal growth [12], however, few studies have speci cally examined GWG early in pregnancy related to PTB. Furthermore, PTB is a heterogeneous condition, but few studies have examined whether associations between pre-pregnancy BMI, GWG and PTB differ by different subtypes of PTB. Besides, most of previous studies were conducted in developed countries. Commonly, above 30% of reproductive age women generally have a high body mass index (BMI ≥ 23.0 kg/m 2 ) in western United States [13], however,this number in developing countries was only 8.5% 14 , and was less investigated. Therefore, the magnitude and direction of the association of pre-pregnancy BMI, GWG and PTB has not been well studied, especially in developing countries. Hence, we conducted a retrospective cohort study with the records of 83,096 women in China to investigate the independent as well as joint association of pre-pregnancy BMI, total GWG and early pregnancy GWG with the risk of subtypes of PTB in singleton pregnancies.

Study Population
This is a retrospective cohort study conducted in Wuhan, China, using electronic medical record (EMR) data from the Maternal and Children Healthcare Information Tracking System of Wuhan, which is a large integrated healthcare system including the information of maternal demographic characteristics, medical history, antenatal examinations and delivery information from all of the 93 hospitals and 121 community health centers in Wuhan. Eligible participants in this study should met the following criteria: (1) women who delivered a live singleton newborn with no birth defects within 28-41 weeks' gestational age between June 1, 2015 and June 1, 2017; (2) lived in the urban area of Wuhan during pregnancy and (3), at least had two weight records during early pregnancy, once earlier than 9 weeks of gestation and another should not later than 20weeks of gestation. Women who were younger than 16 years or older than 50 years at delivery were excluded. Also, participants with unknown anthropometric data (i.e. maternal height, pre-pregnancy weight and weight at delivery) were excluded.
A total of 110,078 electronic medical records were conducted and 83,096 women met the eligibility criteria and were included in the study. 68,527 of them had records of at least two weight measurement during early pregnancy (8-20 weeks).

Assessment of study variables
Gestational age was calculated from the delivery date and the date of the last recorded normal menstrual period. Preterm delivery was de ned as a delivery between 28 weeks 0 days and 36 weeks 6 days of gestation. We excluded very preterm deliveries (<28 weeks gestation) as there were few in this cohort.
We additionally categorized preterm term subtype as either spontaneous preterm birth, premature rupture of membranes (PROM), or medically indicated preterm birth, based on the records of clinical diagnosis reported by the obstetrician at birth. Medically indicated preterm birth was de ned by either induction or caesarean section without uterine contractions or rupture of membranes prior to delivery. PROM was de ned as birth with premature rupture of membranes, and spontaneous preterm birth was identi ed as early onset of delivery and no identi able medical indication, without a PROM diagnosis.
Maternal weight at delivery was measured within 3 days before the delivery day, and GWG was calculated by subtracting maternal pre-pregnancy weight from the weight at delivery and then categorized using to the recommendations of the Institute of Medicine (IOM) (2009), according to maternal pre-pregnancy BMI. GWG within the IOM recommendations was de ned as 12.5-18 kg, 11.5-16 kg, 7-11.5 kg, and 5-9 kg respectively for underweight, normal weight, overweight, and obese women.
Gestational BMI gain was categorized as minimal (<5 kg/m 2 ), moderate (5-10 kg/m 2 ), and excessive (> 10 kg/m 2 ) based on evidence from a previous study [13]. Each one point increase in BMI is roughly equivalent to 2.5 kg in weight gain, using the Chinese national average for female weight and height at reproductive age (158 cm, 54 kg) [14].
We used average weekly weight gain between 8-20 weeks gestation to evaluate GWG during early pregnancy, which was calculated as the latest prenatal care weight before 20 weeks of gestation minus the rst prenatal care weight(before 9weeks) divided by gestation age of the latest prenatal care (up to 20 weeks of gestation) and classi ed as class I (<200 g/week), class II (200-400 g/week), class III (400-600 g/week), and class IV (> 600 g/week). [15] Statistical Analysis Unconditional logistic regression was conducted to calculate odds ratios (ORs), and 95% con dence intervals (CIs) to evaluate the association between PTB and maternal pre-pregnancy BMI, GWG and BMI gain during pregnancy. Models were adjusted for some potential confounders, including infant gender, birth weight, maternal age, parity education level and models that evaluated maternal pre-pregnancy BMI and BMI/weight gain during pregnancy were mutually adjusted.
Analyses were further strati ed by maternal pre-pregnancy BMI categories, and effect modi cations with these variables were evaluated by including the relevant cross-product terms in the regression models. Linear trends were tested using the Wald test. Statistical analyses were conducted with SAS, version 9.4, (SAS Institute, Inc., Cary, North Carolina) and P values < 0.05 was considered statistically signi cant. Table 1 presents the selected characteristics of women in the cohort. 3,983 out of 83,096(4.79%) women delivered a preterm birth infant. Among all the preterm births, 1,569 (39.39%) were spontaneous preterm births, 1,337 (33.57%) were premature rupture of membranes and 1,077 (27.04%) were medically indicated preterm births. Women aged over 30 years, multiparous women, women who gave birth to a male infant, and women who were overweight/ obese before pregnancy were more likely to have preterm birth. The mean total GWG among women who had preterm birth was 14.87±5.99kg, lower than that of women with term birth (17.57±6.94kg). <0.01).The adjusted OR for women who hada gestational BMI gain more than 10 kg/m 2 was 0.29(CI:0.25-0.33) compared with women who gained less than 5 kg/m 2 during pregnancy.

Results
A different trend was apparent for the association of early pregnancy GWG and PTB. As average GWGincreased during early pregnancy, an increasing risk of PTB was observed(p for trend < 0.01). Compared with women who gained less than 200 grams per week before 20 weeks of pregnancy, the risk of PTB was signi cantly higher among women who gained greater than 400 grams per week [Adjusted OR:1.50 (1.33-1.69)]. Notably, women who gained greater than 600 grams per week during early pregnancy had the highest risk of PTB with an adjusted OR of 2.23 (CI: 2.01-2.48).
Results for the associations of evaluated variables with subtypes of PTBare presented in Table 3 1.02, 1.36)], and women with GWG above the recommendation had decreased risk for all the subtypes of preterm birth. Adjusted odds for three subtypes of preterm birth decreased as gestational BMI gain increased ( p for trend < 0.01), whereas the risk for three subtypes of preterm birth increased along with increasing weekly GWG of early pregnancy (p for trend < 0.01).
Further, we examined the associations of gestational BMI gain and GWG during early pregnancy with PTB strati ed by pre-pregnancy BMI (See additional le 2, Table 4). Women with higher BMI gain during the whole pregnancy had a signi cantly decreased risk of PTB across all pre-pregnancy BMI categories (p for trend < 0.01). Conversely, among women who were underweight/normal weight prior to pregnancy, increasing riskof PTB was observed as average GWG during early pregnancy increased(p for trend < 0.01). In particular, women who were underweight before pregnancy and who had the highest average GWG during early pregnancy (≥600g/week) had the highest risk of PTB[Adjusted OR:4.61 (3.68-5.77)], whereas the corresponding ORs were still elevated but not as strong in women who had normal weight before pregnancy [Adjusted OR: 1.88 (1.66-2.14)]. However, no signi cantly association of the risk of PTB and GWG during early pregnancy was observed among women who were overweight/obese before pregnancy(p for trend=0.79). There was signi cant heterogeneity between pre-pregnancy BMI categories for the association of GWG during early pregnancy withPTB risk (P for heterogeneity<0.01).

Discussion
In this cohort study conducted among Chinese women, we found that maternal overweight/obesity prior to pregnancy was independently associated with an overall increased risk of preterm birth, which is in line with results from previous investigations [6,7]. When strati ed by subtypes, pre-pregnancy overweight/obesity was observed to signi cantly increase the risk of medically indicated preterm birth and spontaneous preterm birth in our study, but not the preterm caused by premature rupture of membrane.
The mechanisms linking pre-pregnancy overweight/obesity with risk of PTB is not well understood to date, but probably involves in ammatory, neuroendocrine and lifestyle factors [15]. Goldenberg and Culhane [16] have indicated that preterm birth is mediated by increased systemic in ammation due to a wide range of pre-pregnancy risk factors.Gestational diabetes, pre-eclampsia and obesity are also related to increased systemic in ammation, sometimes called the metabolic syndrome of pregnancy. Especially central obesity, which is more strongly related to insulin resistance than obesity per se, predisposes individuals to these diseases. [16,17], which are known contributors to medically indicated preterm birth. Another previous study also suggested that the association between maternal overweight /obesity and excess risk of medically indicated preterm birth may largely be due to obesity-related pregnancy disorders [18].
On the other end of the pre-pregnancy BMI spectrum, pre-pregnancy underweight has been reported to be associated with an increased risk of PTB [8,19], though fewer studies have separated different subtypes of preterm birth. In this study, we found that women who were underweight before pregnancy were at greater risk of overall PTB. However, results for subtypes of PTB showed that pre-pregnancy underweight was associated with higher risk of spontaneous PTB, while no signi cant association of medically indicated PTB with pre-pregnancy underweight was found in our study. This result is consistent with a previous study from Boston Birth Cohort [20].
In contrast to the more consistent evidence linking pre-pregnancy BMI with PTB, studies for association of gestational weight gain (GWG) and PTB have yielded inconsistent results. Although several previous studies have reported an association between lower GWG and increased risk of preterm birth [8], some studies indicated a positive association between excessive GWG and elevated risk of preterm birth [9]. In the present study, we found that low GWG ( below IOM recommendation) during the whole pregnancy was signi cantly associated with the increased risk of Spontaneous PTB and PROM, while excessive GWG (above IOM recommendation) was found to be associated with decreased risk for all types of PTB. As BMI is considered by some studies to be a better indicator of body fat than weight alone [13], we also classi ed gestational weight gain according to the net change of BMI, and similarly, excessive BMI gain during the whole pregnancy was shown to be related to a decreased risk of all types of PTB. Rebecca et al. found a similar conclusion in a systemic review and meta-analysis of 1.3 million pregnancies that gestation weight gain above recommendations was associated with lower risk of preterm birth [21]. According to a recent systematic review by McDonald et al [22]., women with high total GWG were observed to have lower risks of PTB, high weekly GWG was associated with increased PTB. This observation points to the need more study, although the effect may be due to the association of conditions such as preeclampsia, which is often accompanied by edema and signi cant short-term increased in weight.
However, most of the previous studies evaluating the association of GWG with PTB relied on only two weight measures: weight near conception and weight at delivery, which may have biased associations because GWG over gestational periods differs by term and preterm birth [10]. Furthermore, GWG during early pregnancy was considered to be critical for embryogenesis and fetal growth [12], however, few studies have speci cally examined GWG early in pregnancy related to PTB. Therefore, in this study, we also investigated the association of average GWG during early pregnancy and PTB, and our results indicated that high weekly GWG of early pregnancy was signi cantly associated with higher risk of all types of PTB. While GWG in the rst half of pregnancy is mainly the result of maternal tissues deposition and placental growth, gains from that point on until the end of pregnancy might be in uenced by accumulation of amniotic uid. As high GWG is associated with in ammatory up-regulation through increased production of adipokines by adipose tissue andaugmented systemic secretion of proin ammatory cytokines, which may contribute to the biological pathway of PTB.
In order to explore whether the pre-pregnancy BMI modi es the association between GWG and PTB risk, we strati ed the association by maternal pre-pregnancy BMI categories, and found a signi cant association between excessive BMI gain during the whole pregnancy and a decreased risk of PTB across all the pre-pregnancy BMI categories, whereas the high weekly GWG during early pregnancy is associated with a signi cantly elevated risk of PTB among women who are underweight or normal weight prior to pregnancy. Interestingly, the association between high weekly GWG of early pregnancy and PTB was strongest for women who were underweight before pregnancy. In contrast to our results, ndings of Sharma et al. do not support any signi cant association between GWG in the rst and second trimester and PTB among underweight and normal weight women [10]. Further prospective studies are needed to examine whether a causative relationship between pre-pregnancy BMI, GWG of different periods of pregnancy, and PTB exists, as well as the biological mechanisms underlying this relationship.
One of the strengths of this study is the large population-based cohort which allowed us to evaluate the role of both total GWG and early GWG in relation to risk of PTB as the women's anthropometric characteristics during early pregnancy were available.Also, we were able to identity subtypes of PTB in this study. To our knowledge, this is the rst study investigating the association of pre-pregnancy BMI, total GWG as well as early pregnancy GWG with different subtypes of PTB among Chinese women. A potential limitation of the study is that our data relies on a self-reported pre-pregnancy weight, which could be under estimated and the potential misclassi cation bias may exist. However, previous studies suggest that the resulting BMI category from self-reported data rarely alters, and the self-reported weight may be considered to be an acceptable substitute for actual measurements [23,24].

Conclusion
we conducted a population-based cohort study in China to explore the association of pre-pregnancy BMI, total GWG and early pregnancy GWG with the risk of PTB. We found that pre-pregnancy overweight and obesity were independently associated with the greater risk of medically indicated preterm birth and spontaneous preterm birth, whereas pre-pregnancy underweight was associated with higher risk of spontaneous PTB. High BMI gain during the whole pregnancy was shown to be related to a decreased risk of all types of PTB. In contrast, high weekly GWG of early pregnancy was signi cantly associated with higher risk of all types of PTB. Our results indicate that maternal underweight, overweight/obesity, total GWG, and GWG during early pregnancy should be considered in combination to reduce the risk of PTB.Further prospective studies are needed to examine the relationship of trimester-speci c GWG and PTB, as well as the underlying biological mechanisms. There are no con icts of interest among the authors.

Funding
There was no funding in this research.

Author Contribution
The rst author Yiyang Guo contributed to the study design, data analysis, and manuscript preparation. Chao Xiong and Aifen Zhou contributed to the data analysis and manuscript revision. Ronghua Hu and Rong Yang revised the manuscript. The corresponding authors Yukai Du contributed to the conception of this study and revised the manuscript.