Hostname: page-component-848d4c4894-x24gv Total loading time: 0 Render date: 2024-05-20T18:09:11.934Z Has data issue: false hasContentIssue false

Preconception paternal/maternal BMI and risk of small/large for gestational age infant in over 4·7 million Chinese women aged 20–49 years: a population-based cohort study in China

Published online by Cambridge University Press:  21 February 2022

Tonglei Guo
Affiliation:
National Research Institute for Family Planning, No. 12, Dahuisi Road, Haidian District, 100081 Beijing, People’s Republic of China National Human Genetic Resource Center, No. 12, Dahuisi Road, Haidian District, 100081 Beijing, People’s Republic of China
Ying Yang*
Affiliation:
National Research Institute for Family Planning, No. 12, Dahuisi Road, Haidian District, 100081 Beijing, People’s Republic of China National Human Genetic Resource Center, No. 12, Dahuisi Road, Haidian District, 100081 Beijing, People’s Republic of China Graduate School of Peking Union Medical College, Building 18, No. 9, Dongdan Santiao, Dongcheng District, 100730 Beijing, People’s Republic of China
Jiajing Jia
Affiliation:
National Research Institute for Family Planning, No. 12, Dahuisi Road, Haidian District, 100081 Beijing, People’s Republic of China National Human Genetic Resource Center, No. 12, Dahuisi Road, Haidian District, 100081 Beijing, People’s Republic of China Graduate School of Peking Union Medical College, Building 18, No. 9, Dongdan Santiao, Dongcheng District, 100730 Beijing, People’s Republic of China
Yuzhi Deng
Affiliation:
National Research Institute for Family Planning, No. 12, Dahuisi Road, Haidian District, 100081 Beijing, People’s Republic of China National Human Genetic Resource Center, No. 12, Dahuisi Road, Haidian District, 100081 Beijing, People’s Republic of China Graduate School of Peking Union Medical College, Building 18, No. 9, Dongdan Santiao, Dongcheng District, 100730 Beijing, People’s Republic of China
Yuanyuan Wang
Affiliation:
National Research Institute for Family Planning, No. 12, Dahuisi Road, Haidian District, 100081 Beijing, People’s Republic of China National Human Genetic Resource Center, No. 12, Dahuisi Road, Haidian District, 100081 Beijing, People’s Republic of China
Ya Zhang
Affiliation:
National Research Institute for Family Planning, No. 12, Dahuisi Road, Haidian District, 100081 Beijing, People’s Republic of China National Human Genetic Resource Center, No. 12, Dahuisi Road, Haidian District, 100081 Beijing, People’s Republic of China
Hongguang Zhang
Affiliation:
National Research Institute for Family Planning, No. 12, Dahuisi Road, Haidian District, 100081 Beijing, People’s Republic of China National Human Genetic Resource Center, No. 12, Dahuisi Road, Haidian District, 100081 Beijing, People’s Republic of China
Yuan He
Affiliation:
National Research Institute for Family Planning, No. 12, Dahuisi Road, Haidian District, 100081 Beijing, People’s Republic of China National Human Genetic Resource Center, No. 12, Dahuisi Road, Haidian District, 100081 Beijing, People’s Republic of China Graduate School of Peking Union Medical College, Building 18, No. 9, Dongdan Santiao, Dongcheng District, 100730 Beijing, People’s Republic of China
Jun Zhao
Affiliation:
National Research Institute for Family Planning, No. 12, Dahuisi Road, Haidian District, 100081 Beijing, People’s Republic of China National Human Genetic Resource Center, No. 12, Dahuisi Road, Haidian District, 100081 Beijing, People’s Republic of China
Zuoqi Peng
Affiliation:
National Research Institute for Family Planning, No. 12, Dahuisi Road, Haidian District, 100081 Beijing, People’s Republic of China National Human Genetic Resource Center, No. 12, Dahuisi Road, Haidian District, 100081 Beijing, People’s Republic of China
Qiaomei Wang
Affiliation:
Department of Maternal and Child Health, National Health Commission of the PRC, No. 1, Xizhimenwai South Road, Xicheng District, 100044 Beijing, People’s Republic of China
Haiping Shen
Affiliation:
Department of Maternal and Child Health, National Health Commission of the PRC, No. 1, Xizhimenwai South Road, Xicheng District, 100044 Beijing, People’s Republic of China
Yiping Zhang
Affiliation:
Department of Maternal and Child Health, National Health Commission of the PRC, No. 1, Xizhimenwai South Road, Xicheng District, 100044 Beijing, People’s Republic of China
Donghai Yan
Affiliation:
Department of Maternal and Child Health, National Health Commission of the PRC, No. 1, Xizhimenwai South Road, Xicheng District, 100044 Beijing, People’s Republic of China
Xu Ma*
Affiliation:
National Research Institute for Family Planning, No. 12, Dahuisi Road, Haidian District, 100081 Beijing, People’s Republic of China National Human Genetic Resource Center, No. 12, Dahuisi Road, Haidian District, 100081 Beijing, People’s Republic of China Graduate School of Peking Union Medical College, Building 18, No. 9, Dongdan Santiao, Dongcheng District, 100730 Beijing, People’s Republic of China
*
* Corresponding authors: Ying Yang, email angela-yy65@hotmail.com; Xu Ma, email nfpcc_ma@163.com
* Corresponding authors: Ying Yang, email angela-yy65@hotmail.com; Xu Ma, email nfpcc_ma@163.com
Rights & Permissions [Opens in a new window]

Abstract

Evidence of couples’ BMI and its influence on birth weight is limited and contradictory. Therefore, this study aims to assess the association between couple’s preconception BMI and the risk of small for gestational age (SGA)/large for gestational age (LGA) infant, among over 4·7 million couples in a retrospective cohort study based on the National Free Pre-pregnancy Checkups Project between 1 December 2013 and 30 November 2016 in China. Among the live births, 256 718 (5·44 %) SGA events and 506 495 (10·73 %) LGA events were documented, respectively. After adjusting for confounders, underweight men had significantly higher risk (OR 1·17 (95 % CI 1·15, 1·19)) of SGA infants compared with men with normal BMI, while a significant and increased risk of LGA infants was obtained for overweight and obese men (OR 1·08 (95 % CI 1·06, 1·09); OR 1·19 (95 % CI 1·17, 1·20)), respectively. The restricted cubic spline result revealed a non-linear decreasing dose–response relationship of paternal BMI (less than 22·64) with SGA. Meanwhile, a non-linear increasing dose–response relationship of paternal BMI (more than 22·92) with LGA infants was observed. Moreover, similar results about the association between maternal preconception BMI and SGA/LGA infants were obtained. Abnormal preconception BMI in either women or men were associated with increased risk of SGA/LGA infants, respectively. Overall, couple’s abnormal weight before pregnancy may be an important preventable risk factor for SGA/LGA infants.

Type
Research Article
Copyright
© The Author(s), 2022. Published by Cambridge University Press on behalf of The Nutrition Society

Birth weight, as a major predictor of infant growth and survival, is associated with neonatal mortality and morbidity, cognitive ability and metabolic disease including obesity, hypertension as well as type 2 diabetes mellitus during childhood or adult life(Reference Edvardsson, Steinthorsdottir and Eliasdottir1Reference McGuire6). Previous studies found that a high birth weight increases the risk of obesity in later life and may act as a mediator between prenatal influences and risk of disease in adults(Reference Yu, Han and Zhu7). Magnus(Reference Magnus8) reported that more than 50 % of the total variance in birth weight result from variation of fetal genes, using the birth weights of offspring of twins. Besides fetal genes, multifactorial risk factors including socio-demographic factors (maternal age, ethnicity, educational level, economic status and paternal height), lifestyle (maternal diet, physical activity, smoking, drinking), nutritional status (preconception BMI, pregnancy weight gain, anaemia), chronic disease (hypertension, diabetes) and antenatal care are associated with birth weight which is caused by the interaction of intra-uterine, genetic and environmental factors(Reference Abera, Ejara and Gebremedhin9Reference McCowan, North and Kho16).

An increasing number of studies have identified the association between maternal BMI and infant birth weight including small for gestational age (SGA), large for gestational age (LGA), macrosomia or low birth weight, which suggests that higher maternal preconception BMI is associated with an increased risk of LGA, while maternal preconception underweight is associated with an increased risk of SGA(Reference Pan, Zhang and Wang17,Reference Liu, Dai and Dai18) . The mechanism of the effect of maternal preconception BMI on birth weight has been unclear. Sharp and colleagues reported that both maternal underweight and obesity affect the neonatal epigenome via an intra-uterine mechanism, and increased DNA methylation may mediate the associations of maternal underweight with lower offspring obesity and maternal obesity with greater offspring obesity(Reference Chen, Li and Chen19). Furthermore, studies have shown that insulin resistance which causes metabolic disorders and increases the availability of maternal nutrients to the fetus, leading to fetal growth acceleration, often occurs in women with high BMI or excessive weight gain during pregnancy(Reference Zhao, Xu and Wu20,Reference Sharp, Lawlor and Richmond21) .

However, less is known about the association between paternal BMI and risk of LGA/SGA, and the few previous studies have shown contradictory results. A study from Croatia found a significant association between paternal BMI and birth weight, while several other studies reported that paternal BMI was not directly associated with LGA/SGA(Reference Gauster, Hiden and Blaschitz22Reference Takagi, Iwama and Metoki26). However, paternal BMI in the few previous studies was almost collected from self-report rather than formal measurement. Previous evidence has shown that the mechanisms underlying a possible association between paternal BMI and SGA/LGA might involve affecting spermatozoa or genetic regulation, such as insulin-like growth factor-I and insulin-like growth factor-II, which are expressed from the copy of the paternal gene(Reference Griffiths, Dezateux and Cole27Reference Constancia, Hemberger and Hughes29).

Nowadays, consensus has been reached that low birth weight is not an adequate measure of a ‘small baby’. Accumulating evidence suggests that the variable ‘SGA’ is perceived as a more optimal measure than birth weight or low birth weight to assess fetal growth generally(Reference Wang, Zhou and Cheng30Reference Lefebvre and Samoilenko32). Therefore, this study aims to assess the association between preconception couple’s BMI and the risk of LGA/SGA, among over 4·7 million couples in a cohort study based on the National Free Pre-pregnancy Checkups Project (NFPCP) between 2013 and 2016.

Methods

Study population and design

Data for this national large population-based retrospective cohort study were extracted from the NFPCP, which is a national preconception free health service to provide free preconception health examinations and counselling for rural reproductive-aged couples throughout China. Since 2013, the NFPCP services were extended to both rural and urban married couples. The NFPCP has been supported by the National Health Commission and the Ministry of Finance of the People’s Republic of China. Detailed design, organisation and implementation of NFPCP have been described in previously published articles(Reference Savitz, Hertz-Picciotto and Poole33Reference Wang, Cao and Peng35).

In general, 5 709 510 Chinese women who were aged 20–49 years old at last menstrual period (LMP) participated in the NFPCP from 1 January 2013 to 31 December 2016 and successfully got pregnant and gave birth until 31 December 2017 were included. Then 134 258 women or their husbands with missing information on BMI data, 31 169 women who had multiparous pregnancy and 238 805 women who had other types of adverse pregnancy outcomes, such as fetal death, ectopic pregnancy, spontaneous abortion and medically induced abortion, 585 465 women with missing information on birth weight or gestational week were excluded. Finally, a total of 4 719 813 women with singleton pregnancy were included in the analysis after considering the exclusion criteria in Fig. 1.

Fig. 1. Flow chart of the study population. NFPCP, National Free Pre-pregnancy Checkups Project; LMP, last menstrual period.

This study was approved by the Institutional Research Review Board at the National Research Institute for Family Planning, Beijing, China. Written informed consent was obtained from all NFPCP participants. Our study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.

Data collection

The general NFPCP programme consists of the three stages: preconception health examinations, early pregnancy follow-up and pregnancy outcome follow-up. These stages were contained in the NFPCP to collect different types of data from all enrolled couples. Face-to-face interviews, medical examination and telephone interviews were conducted by qualified health staff using a standard questionnaire which includes a pre-pregnancy examination chart (both husband and wife) to collect baseline information about demographic characteristics and several follow-up charts including the information of lifestyle or medical examination during early pregnancy, pregnancy outcome and infant condition in local family planning service agencies or maternal and child service centres.

In the first stage (preconception health examinations), couples who met the fertility policy and planned to conceive within 6 months were advocated and encouraged by the domestic resident committee to participate in a preconception health examination. The basic information on family history, lifestyle, dietary nutrition, social-psychological factors, environmental poisons, physical examination and clinical examination was recorded in the pre-pregnancy examination chart for husband and wife, separately. The body weight and height of wives and their husbands wearing light, indoor clothing and no shoes were measured according to a standardised protocol. Next, BMI was calculated.

In the second stage (early pregnancy follow-up), the early pregnancy follow-up interview was conducted by trained healthcare staff using the telephone to obtain the conception status of women within 3 months after preconception health examinations. For the women who did not get pregnant at the first follow-up interview, repeated investigations were conducted subsequently every 3 months until up to 1 year after preconception examination. For pregnant women, information regarding the LMP, toxic or harmful substances exposure and any lifestyle changes in the first trimester of pregnancy was collected.

In the final stage (pregnancy outcome follow-up), women who had become pregnant during the early pregnancy follow-up were recontacted by trained research staff using telephone within 1 year to collect pregnancy outcome information. Adverse pregnancy outcomes including spontaneous abortion, low birth weight, induced labour, ectopic pregnancy, birth defects, preterm birth and stillbirth were documented in the follow-up chart. Besides, information regarding puerpera’s self-reported information on their delivery condition and fetal information such as the delivery date, birth weight and sex was self-reported and recorded during the interview. Women were encouraged to actively report their adverse pregnancy outcome events, if women had an abortion or other adverse pregnancy outcomes occurred in the early pregnancy period.

In the current study, the gestational week of women was initially calculated by the duration between the LMP and the time of delivery which were recorded in the early pregnancy follow-up and the pregnancy outcomes follow-up, respectively. Then, ultrasound examination adjusted gestational week was also collected in the pregnancy outcome follow-up survey. When the two gestational week records were different, the ultrasound examination adjusted gestational week was used.

Exposure and outcome

The body weight and height were measured with wives and their husbands wearing light, indoor clothing and no shoes, respectively. Next, maternal/paternal own preconception weight and height were used to calculate the maternal/paternal preconception BMI (calculated as the weight (kg) in kilograms divided by height (m) in meters squared), which was further classified into four groups: (1) underweight (< 18·5 kg/m2), (2) normal weight (18·5–23·9 kg/m2), (3) overweight (24·0–27·9 kg/m2) and (4) obese (≥ 28·0 kg/m2 or higher), respectively, according to the Chinese guidelines for the prevention and control of overweight and obesity in adults(Reference Yang, He and Li36). Couples were categorised into nine groups according to their BMI levels (maternal BMI: paternal BMI): (1) underweight (wife): underweight (husband); (2) underweight (wife): normal weight (husband); (3) underweight (wife): overweight and obese (husband); (4) normal weight (wife): underweight (husband); (5) normal weight (wife): normal weight (husband); (6) normal weight (wife): overweight and obese (husband); (7) overweight and obese (wife): underweight (husband); (8) overweight and obese (wife): normal weight (husband); (9) overweight and obese (wife): overweight and obese (husband).

SGA infant is defined as newborn birth weight by gestational age and sex below the 10th percentile (< 10th percentile). Appropriate for gestational age infant is defined as newborn birth weight by gestational age and sex between the 10th percentile and 90th percentile (10th to 90th percentile). LGA infant is identified as newborn birth weight by gestational age and sex beyond the 90th percentile (> 90th percentile) according to the Chinese national survey(Reference Zhang, Wang and Shen37).

Covariates

The ages of women were calculated as the difference between the date of birth and the first day of the LMP of women; these ages were categorised into various age groups (20–24·9, 25–29·9, 30–34·9, 35–39·9 and ≥ 40 years). Higher education was categorised as levels of education of senior high school, college or higher. Ethnicity was categorised as Han nationality and others (non-Han nationality). Diabetes mellitus was categorised as either self-reported diabetes or fasting blood glucose ≥ 7·0 mmol/l. Physicians conducted seated blood pressure measurement in the right arm of the seated women/men using an automated blood pressure monitor, on a single occasion after women/men rested for 10 min or more. Hypertension was defined as self-reported hypertension or systolic blood pressure ≥ 140 mmHg or diastolic blood pressure ≥ 90 mmHg. Smoking was defined as smoking before or during early pregnancy (women/men who smoked at least 1 cigarette per day at least 1 year at the time of baseline examination). Alcohol drinking was defined as drinking once per week on average at the time of examination (regardless of the type of drinking, such as white wine, liquor, beer, red wine and yellow rice wine, before or during early pregnancy). Passive smoking was defined as exposure to environmental tobacco smoke before or during early pregnancy. History of adverse pregnancy outcomes was defined as the history of preterm birth (live birth between 28 and 36 completed weeks of pregnancy), later fetal death (stillbirth after 28 weeks of gestation or newborns who died within 7 days after birth) or spontaneous abortion (pregnancy loss occurring before the 28th week of gestation) in previous pregnancies.

Statistical analysis

Continuous variables with normal distribution were expressed as mean values (standard deviations), and non-normally distributed variables were expressed as median (inter-quartile range). Categorical variables were expressed as numbers (percentages) for baseline characteristics.

The associations between the two categories of maternal/paternal preconception BMI levels (maternal/paternal preconception BMI levels, couple’s BMI levels) and the risk of SGA or LGA were examined, respectively. The Odds ratios (OR) and their corresponding 95 % confidence intervals (CI) of SGA/LGA associated with BMI status of couples (nine groups) and maternal/paternal preconception BMI levels (four groups) were estimated using age-adjusted and multivariate-adjusted multinomial logistic regression models, respectively, compared with appropriate for gestational age infant.

The OR and their corresponding 95 % CI were estimated by age-adjusted and multivariate-adjusted multinomial logistic regression models separately, using normal weight (18·5–23·9 kg/m2) as the reference group. Maternal and paternal age at LMP (20–24·9, 25–29·9, 30–34·9, 35–39·9 and ≥ 40 years old) as covariates were adjusted in all age-adjusted models. Covariates in the multivariate-adjusted regression models of BMI status of couples (nine groups) included maternal and paternal age at LMP (20–24·9, 25–29·9, 30–34·9, 35–39·9 and ≥ 40 years old), maternal and paternal height (continuous variables), maternal and paternal ethnic (Han, other ethnic), maternal and paternal education levels (senior high school, college or higher), parity (0, 1+), maternal and paternal area of residence (rural or urban), maternal hypertension (yes, no), maternal diabetes (yes, no), maternal and paternal alcohol drinking (yes, no), maternal and paternal smoking (yes, no), maternal and paternal passive smoking (yes, no) and history of adverse pregnancy (yes, no). Meanwhile, in the multivariate-adjusted regression models of maternal/paternal preconception BMI levels (four groups) we additionally adjusted for paternal or maternal BMI levels (< 18·5, 18·5–23·9, 24·0–27·9 and ≥ 28·0 kg/m2). Besides, the individual and couple models after regrouping BMI levels using WHO criteria (underweight < 18·5 kg/m2, normal weight 18·5–25·0 kg/m2, overweight 25·0–30·0 kg/m2 and obese 30·0 kg/m2 or higher) were rerun(Reference Mi, Zhang and Wang38).

Furthermore, the dose-response relationship of maternal/paternal preconception BMI levels and risk of SGA/LGA were assessed using restricted cubic spline, respectively, and five knots at the 5th, 25th, 50th, 75th and 95th percentiles of maternal/paternal preconception BMI levels were used in plotted smooth curves (the restricted cubic spline with three, four or five knots was separately fitted, and the models with the lowest Akaike information criterion as the best model were chosen), and Wald statistics was used to test the non-linearity of the dose-response(Reference Zhu, Zhang and Zhang39,40) . Covariates were the same as the multivariate-adjusted regression models of maternal/paternal preconception BMI levels. Besides, the models after regrouping SGA/LGA using INTERGROWTH-21st criteria were rerun(Reference Schoenaker, Simon and Chaturvedi41). Additionally, sensitivity analysis was conducted after excluding couples with missing data on baseline characteristics.

Statistical analysis was performed using R software (V.3.5.0; https://www.r-project.org/) with the analysis packages ‘epade (version 0.3.8)’, ‘forestplot (version 1.7.2)’, ‘rms (version 5.1–2)’, ‘ggplot2 (version 3.1.0)’, ‘reshape2 (version 1.4.3)’ and ‘speedglm (version 0.3–2)’. All statistical tests were two-sided, and P values < 0·05 were considered statistically significant.

Results

Among all the 4 719 813 enrolled couples, 14·85 % of women were overweight or obese (12·09 % were overweight, 2·76 % obese) and 13·53 % were underweight, while 33·53 % of men were overweight or obese (26·46 % were overweight, 7·07 % obese) and 4·10 % were underweight (Table 1). Women who smoked cigarettes, were exposed to second-hand smoke and have pre-existing diabetes or history of adverse pregnancy outcomes were more likely with abnormal BMI. Women with BMI ≥ 25 kg/m2 were more likely to be of Han nationality, aged and poor educated, from rural areas and have higher blood pressure level than those women with BMI within 18·5–24·9 kg/m2. Meanwhile, women with BMI < 18·5 kg/m2 were more likely to be parous and drink alcohol (online Supplementary Table S2). The husband with abnormal BMI was more likely to be of Han nationality, consumes alcohol, smoke cigarettes or exposed to second-hand smoking. Besides, the husband with BMI ≥ 25 kg/m2 was more likely to be older, living in the city and with more educational attainment (online Supplementary Table S3). Detailed descriptive characteristics of the study population by maternal/paternal preconception BMI are given in Table 1, online Supplementary Table 2 and Table 3, respectively.

Table 1. Characteristics of the total study population

(Numbers and percentages)

BMI, body mass index (calculated as the weight in kilograms divided by height in meters squared); LMP, last menstrual period; IQR, inter-quartile range; SGA, small for gestational age; LGA, large for gestational age; NA, missing data.

The final study population included 4 719 813 women (Fig. 1) with a median birth weight of 3350 g (inter-quartile range 3100–3600) and median gestational age at birth of 39·71 weeks (inter-quartile range 38·57–40·43); 2 453 312 (51·98 %) infants were male and 2 261 613 (47·92 %) were female. Among the live births, 256 718 (5·44 %) SGA events and 506 495 (10·73 %) LGA events were documented, respectively. The incidence of SGA (7·61 %) in the maternal preconception underweight group was significantly higher than that in normal weight (5·28 %), overweight (4·25 %) or obese (4·06 %) group, while the corresponding SGA incidence was 7·06, 5·64, 4·98 and 4·49 %, respectively, in paternal preconception underweight, normal, overweight or obese group. Meanwhile, the incidence of LGA was 8·70, 10·48, 13·46 and 15·26 % for underweight, normal weight, overweight and obese women, and the corresponding incidence for the husband was 9·32, 10·24, 11·47 and 13·09 %, respectively (Fig. 2).

Fig. 2. Associations between maternal/paternal preconception BMI and risk of small/large for gestational age infant (OR and 95 % CI). SGA, small for gestational age infant; AGA, appropriate for gestational age infant; LGA, large for gestational age infant; OR, odds ratios, BMI, body mass index (calculated as the weight in kilograms divided by height in meters squared). * Models were adjusted for maternal and paternal age at LMP, height, ethnic, education, area of residence, alcohol drinking, smoking, passive smoking and paternal BMI, maternal hypertension, diabetes, parity as well as history of adverse pregnancy. † Models were adjusted for maternal and paternal age at LMP, height, ethnic, education, area of residence, alcohol drinking, smoking, passive smoking and maternal BMI, hypertension, diabetes, parity as well as history of adverse pregnancy.

After adjusted for maternal and paternal age at LMP, height, ethnicity, education, area of residence, alcohol drinking, smoking, passive smoking and maternal BMI, hypertension, diabetes, parity as well as the history of adverse pregnancy, it is shown that husbands who were underweight had significantly higher risk (OR 1·17 95 % CI (1·15, 1·19)) of SGA compared with the husband with normal BMI, while overweight or obese husband had lower risks with multivariable-adjusted OR of 0·92 (95 % CI 0·90, 0·93) and 0·87 (95 % CI 0·85, 0·88), respectively. In addition, a significant and increased risk of LGA was observed for overweight and obese men (OR 1·08 (95 % CI 1·06–1·09); OR 1·19 (95 % CI 1·17, 1·20)) respectively, but the negative association was identified with LGA for underweight men (OR 0·94 (95 % CI 0·93, 0·96)) (Fig. 2). Reduced paternal BMI was associated with an increased risk of SGA when paternal BMI was less than 22·64 (P non-linear < 0·001). Meanwhile, increasing paternal BMI was associated with an increased risk of LGA when paternal BMI was more than 22·92 (P non-linear< 0·001). In our analysis, similar results about the association between maternal preconception BMI and SGA/LGA (BMI level was 20·71/20·96) were obtained and detailed multivariable-adjusted OR (95 % CI) and restricted cubic spline result were described in Fig. 4 and online Supplementary Table S4, respectively. Similar results were observed in the analysis according to WHO criteria of BMI (online Supplementary Table S7). Moreover, similar results were observed in the analysis according to INTERGROWTH-21st criteria of SGA/LGA (online Supplementary Table S9).

Fig. 3. Adjusted OR of small/large for gestational age infant according to BMI of couples. SGA, small for gestational age infant; AGA, appropriate for gestational age infant; LGA, large for gestational age infant; OR, odds ratios, BMI, body mass index (calculated as the weight in kilograms divided by height in meters squared). *† Multivariable-adjusted OR (95 % CI) was adjusted for maternal and paternal age at LMP, height, ethnic, education, area of residence, alcohol drinking, smoking, passive smoking and maternal hypertension, diabetes, parity as well as history of adverse pregnancy.

Further stratified analysis has shown that SGA infants rates were significantly higher among maternal underweight groups, compared with the reference group (couples with normal BMI), while overweight and obese women groups had significantly lower rates. Inversely, compared with couples with normal BMI, LGA infants rates were significantly lower among maternal underweight groups, while overweight and obese women groups had significantly higher rates of LGA infants (Fig. 3). Detailed multivariable-adjusted OR (95 % CI) were described in online Supplementary Table S5. Similar results were observed in the analysis according to WHO criteria of BMI (online Supplementary Table S8). In sensitive analysis, similar results were obtained in analysis by excluding couples with missing data on baseline characteristics (online Supplementary Table S10).

Fig. 4. Dose–response relationship between maternal/paternal preconception BMI and risk of small/large for gestational age infant. SGA, small for gestational age infant; LGA, large for gestational age infant; OR, odds ratios, BMI, BMI (calculated as the weight in kilograms divided by height in meters squared). Graphs show the multivariable-adjusted OR of associations between paternal/maternal preconception BMI and the risk of small for gestational age infant (A1, B1), large for gestational age infant (A2, B2), respectively. In the graph, black curves and shaded grey areas show predicted OR and 95 % CI, respectively. A1, A2: Maternal and paternal age at LMP, height, ethnic, education, area of residence, alcohol drinking, smoking, passive smoking and maternal BMI, hypertension, diabetes, parity as well as history of adverse pregnancy were used in the analysis as covariates. B1, B2: Maternal and paternal age at LMP, height, ethnic, education, area of residence, alcohol drinking, smoking, passive smoking, paternal BMI and maternal hypertension, diabetes, parity as well as history of adverse pregnancy were used in the analysis as covariates. The black lines represent the reference level.

Discussion

In our large nationwide population-based retrospective cohort study of over 4·7 million couples in China, maternal/paternal abnormal preconception BMI levels were associated with increased risk of adverse SGA/LGA infants, respectively. A statistically significant decreased risk of SGA was identified in females/males with the increase of their BMI levels, and the higher the preconception BMI level is, the lower the risk of SGA is. Inversely, an increasing trend of the relative risk of LGA was found with the increasing of paternal or maternal BMI levels; preconception overweight and obesity were associated with increased risks of LGA both in men and women. Furthermore, we found significant non-linear dose–response relationships that reduced paternal/maternal BMI were associated with an increased risk of SGA when paternal/maternal BMI was less than 22·64/20·71. Meanwhile, increasing paternal/maternal BMI was associated with an increased risk of LGA when paternal/maternal BMI was more than 22·92/20·96. After adjusted for couple’s covariates including maternal BMI, we still found that higher paternal preconception BMI was associated with an increased risk of having an LGA infant, while lower paternal preconception BMI was associated with higher risks of SGA. To our knowledge, this is the first largest comprehensive study to explore an association for couple’s preconception BMI with the risk of SGA/LGA in a Chinese population-based study. In the present study, similar results about the association between maternal preconception BMI and SGA/LGA to previous studies which have consistently shown that maternal BMI is positively associated with birth weight of infants were obtained(Reference Pan, Zhang and Wang17,Reference Liu, Dai and Dai18,42Reference Goldstein, Abell and Ranasinha44) . However, previous studies of paternal BMI and its influence on birth weight showed contradictory results(Reference Gauster, Hiden and Blaschitz22,Reference Miletic, Stoini and Mikulandra24Reference Takagi, Iwama and Metoki26,Reference Ma, Zhang and Zhang45Reference Campbell and McPherson47) . Although some studies have found that paternal body weight and height significantly correlated with the infant birth weight and length, a significant dose–response relationship was only identified in male infants(Reference Gauster, Hiden and Blaschitz22,Reference Miletic, Stoini and Mikulandra24,Reference Ma, Zhang and Zhang45) . Inversely, several other studies which were conducted in the third trimester of gestation or using paternal data collected from their wives found no association between paternal preconception BMI and SGA/LGA, when certain paternal and maternal risk factors such as maternal height, BMI or paternal height were adjusted(Reference Gil-Sanchez, Demmelmair and Parrilla23Reference Takagi, Iwama and Metoki26,Reference Derraik, Pasupathy and McCowan46,Reference Pomeroy, Wells and Cole48,Reference Shah49) . In the current study, a positive, significant and independent non-linear dose–response relationship of paternal preconception BMI (less than 22·64) with LGA risk was observed, whereas a negative non-linear dose–response relationship between paternal preconception BMI (more than 22·92) and SGA risk was shown in this large cohort. Moreover, paternal data directly collected prior to pregnancy were used in our study which may effectively avoid information bias as well as recalling bias.

In this study, compared with maternal/paternal normal BMI, a decreased risk of SGA for overweight and obese women/men was observed, but maternal/paternal overweight and obesity were associated with higher risks of LGA, respectively. Moreover, although maternal/paternal underweight was associated with lower risks of LGA, a significant and positive association of SGA was observed for underweight women/men, respectively. These findings imply that abnormal (both low and high) paternal/maternal BMI were associated with higher risks of SGA/LGA, respectively. Sufficient and consistent studies suggested that maternal pregnancy BMI and gestational weight gain have been associated with SGA/LGA(Reference Goldstein, Abell and Ranasinha44,Reference Stephenson, Heslehurst and Hall50) . We propose that couples’ well weight management in the preconception period might be crucial for maternal and fetal health which might prevent the occurrence of SGA/LGA infant. Till now, the underlying mechanisms through which abnormal paternal BMI contributes to SGA/LGA still remain unknown. Evidence indicated that paternal obesity has been shown to increase DNA methylation near the transcription start locus of ARFGAP3 gene of germ cells, which was associated with lower offspring birth, weight, increase the histone modification and modify the expression of sperm microRNA(Reference Noor, Cardenas and Rifas-Shiman51Reference Li, Shen and Hua54). Furthermore, previous studies reported that the major growth factors in the offspring gene, such as insulin-like growth factor-I, and insulin-like growth factor-II which are associated with placental, fetal growth including skeletal length, are expressed from the copy of the paternal gene, but part results of these studies were obtained from animal experiments(Reference Constancia, Hemberger and Hughes29Reference Lecomte, Maloney and Wang31,Reference Van Wyk and Smith55,Reference Fowden56) . All these studies indicated that the impact of paternal BMI on SGA/LGA might be resulted from affecting spermatozoa or genetic regulation. Besides, the mechanism of the effect of maternal preconception BMI on birth weight has also been unclear. A Norwegian Research reported that more than 50 % of the total population variance in birth weight result from fetal genes, and that less than 20 % result from variation in maternal genes. The rest variance (20–30 %) could be caused by random environmental effects(Reference Magnus8). A previous study has suggested that abnormal maternal BMI (both maternal underweight and obesity) affects the neonatal epigenome via an intra-uterine mechanism, and increased DNA methylation might mediate the associations of maternal underweight with lower offspring obesity and maternal obesity with greater offspring obesity(Reference Chen, Li and Chen19). Furthermore, studies have shown that insulin resistance which leads to metabolic disorders and increases the availability of maternal nutrients to the fetus, causing fetal growth acceleration, often occurs in women with high BMI(Reference Zhao, Xu and Wu20,Reference Sharp, Lawlor and Richmond21) . These studies suggested that epigenetics or other mechanism of couples’ BMI may have potential to change that risk of SGA/LGA.

However, further randomised controlled trials lab research or cohort studies in other populations are still needed to comprehensively and deeply clarify the relationships between paternal/maternal preconception BMI and SGA/LGA infant or other adverse pregnancy outcomes, including miscarriage, preterm birth and fetal death during the preconception period. These studies might provide more evidence to act against the overweight and obesity epidemic, which have become prevalent in women of reproductive age and cause a high disease burden in the present and future, as well as promote preconception health which is strongly associated with pregnancy outcome, and interventions during preconception period are urgently required to improve both maternal and child health(Reference Stephenson, Heslehurst and Hall50,Reference Van Wyk and Smith55) .

A previous study found that maternal BMI during 14–16 weeks of gestation had a dominant influence on LGA; our results were consistent with this research(Reference Derraik, Pasupathy and McCowan46). In our analyses, the maternal-offspring SGA/LGA association is stronger than the paternal-offspring SGA/LGA association. These findings indicate that maintaining an optimal BMI level for couples prior to pregnancy might be the most beneficial condition for pregnancy preparation, although paternal preconception BMI showed a weaker effect on the risk of SGA/LGA. Interestingly, we also found that the women whose husbands were overweight or obese were more likely to be overweight or obese (online Supplementary Table S6), which might be explained by couples having similar dietary or physical exercise models within one family unit or unfavourable lifestyles which result in obesity might affect each other in couples.

To our knowledge, this study is the largest comprehensive population-based cohort study to explore the association between the paternal and/or maternal preconception BMI and the risk of SGA/LGA in the study population of more than 4·7 million women, adjusting for numerous couple’s confounders and covariates. Another strength is that dose–response relationship curves between paternal preconception BMI and risk of SGA/LGA infants were firstly assessed in our study. In addition, the maternal and paternal preconception weight and height data were measured objectively in this study. However, this analysis has some limitations. First, data on couples’ socio-economic status, maternal gestational weight gain, and maternal complications, which have been associated with SGA/LGA infant, were not available in our study, so these factors could not be adjusted in our multivariable analysis. Another limitation is that data on infant birth weight were self-reported in this study, which could lead to inaccurate outcome calculation and the misclassification of SGA/LGA. Third, information on paternal dietary or physical exercise habits which might be shared with the mother was not collected in our study.

In conclusion, our study demonstrated that greater maternal/paternal preconception BMI was associated with an increased risk of having an LGA infant, while lower maternal/paternal preconception BMI was associated with higher risks of SGA, respectively. Although maternal-offspring SGA/LGA association is stronger than the paternal-offspring SGA/LGA association, with respect to the high prevalence of obesity worldwide, the importance of couple’s weight management (couple’s normal BMI) should not be neglected during preconception examination or counselling. Our findings may have implications for managing and counselling in pregnancies to avoid adverse pregnancy outcomes. Future research studies are needed to clarify whether and how preconception counselling and interventions for couples with abnormal preconception BMI can reduce the risk of SGA/LGA.

Acknowledgements

The authors thank health workers and countless participants throughout thirty-one provinces in the NFPCP for their considerable efforts and collaboration.

This work was supported by National Key Research and Development Program of China (No. 2016YFC1000307), CAMS Innovation Fund for Medical Sciences (No. 2018-I2M-1-004).

The corresponding author has full access to data in the study and takes responsibility for data integrity and the accuracy of data analysis. T. G. searched the literature, analysed the data, interpreted the results and drafted the manuscript. J. J. and Y. D. searched the literature and interpreted the results. Q. W., H. S. and D. Y. led the data collection and laboratory testing. Y. Z., D. Y., Y. W., H. Z., Z. P., J. Z., Y. H. and Y. Z., collected the data. Y. Y. and X. M. revised the manuscript.

There are no conflicts of interest.

Supplementary material

For supplementary materials referred to in this article, please visit https://doi.org/10.1017/S000711452200054X

References

Edvardsson, VO, Steinthorsdottir, SD, Eliasdottir, SB, et al. (2012) Birth weight and childhood blood pressure. Curr Hypertens Rep 14, 596602.CrossRefGoogle ScholarPubMed
Lynch, JL & Gibbs, BG (2017) Birth weight and early cognitive skills: can parenting offset the link? Matern Child Health J 21, 156167.CrossRefGoogle ScholarPubMed
Mugnier, A, Mila, H, Guiraud, F, et al. (2019) Birth weight as a risk factor for neonatal mortality: breed-specific approach to identify at-risk puppies. Prev Vet Med 171, 104746.CrossRefGoogle ScholarPubMed
Werneck, AO, Silva, D, Collings, PJ, et al. (2017) Birth weight, biological maturation and obesity in adolescents: a mediation analysis. J Dev Orig Health Dis 8, 502507.CrossRefGoogle ScholarPubMed
Yang, Y, Wang, Z, Mo, M, et al. (2018) The association of gestational diabetes mellitus with fetal birth weight. J Diabetes Complications 32, 635642.CrossRefGoogle ScholarPubMed
McGuire, SF (2017) Understanding the implications of birth weight. Nurs Womens Health 21, 4549.CrossRefGoogle ScholarPubMed
Yu, ZB, Han, SP, Zhu, GZ, et al. (2011) Birth weight and subsequent risk of obesity: a systematic review and meta-analysis. Obes Rev 12, 525542.CrossRefGoogle ScholarPubMed
Magnus, P (1984) Causes of variation in birth weight: a study of offspring of twins. Clin Genet. 25, 1524.CrossRefGoogle ScholarPubMed
Abera, Z, Ejara, D & Gebremedhin, S (2019) Nutritional and non-nutritional factors associated with low birth weight in Sawula Town, Gamo Gofa Zone, Southern Ethiopia. BMC Res Notes 12, 540.CrossRefGoogle ScholarPubMed
Spada, E, Chiossi, G, Coscia, A, et al. (2018) Effect of maternal age, height, BMI and ethnicity on birth weight: an Italian multicenter study. J Perinat Med 46, 10161021.CrossRefGoogle ScholarPubMed
Yadav, DK, Chaudhary, U & Shrestha, N (2011) Risk factors associated with low birth weight. J Nepal Health Res Counc 9, 159164.Google ScholarPubMed
Torres-Arreola, LP, Constantino-Casas, P, Flores-Hernandez, S, et al. (2005) Socioeconomic factors and low birth weight in Mexico. Bmc Public Health 5, 20.CrossRefGoogle ScholarPubMed
Slemming, W, Bello, B, Saloojee, H, et al. (2016) Maternal risk exposure during pregnancy and infant birth weight. Early Hum Dev 99, 3136.CrossRefGoogle ScholarPubMed
Aboye, W, Berhe, T, Birhane, T, et al. (2018) Prevalence and associated factors of low birth weight in Axum town, Tigray, North Ethiopia. BMC Res Notes 11, 684.CrossRefGoogle ScholarPubMed
Figueiredo, A, Gomes-Filho, IS, Batista, J, et al. (2019) Maternal anemia and birth weight: a prospective cohort study. PLOS ONE 14, e212817.CrossRefGoogle ScholarPubMed
McCowan, LM, North, RA, Kho, EM, et al. (2011) Paternal contribution to small for gestational age babies: a multicenter prospective study. Obesity 19, 10351039.CrossRefGoogle Scholar
Pan, Y, Zhang, S, Wang, Q, et al. (2016) Investigating the association between prepregnancy body mass index and adverse pregnancy outcomes: a large cohort study of 536 098 Chinese pregnant women in rural China. BMJ Open 6, e11227.CrossRefGoogle ScholarPubMed
Liu, Y, Dai, W, Dai, X, et al. (2012) Prepregnancy body mass index and gestational weight gain with the outcome of pregnancy: a 13-year study of 292 568 cases in China. Arch Gynecol Obstet 286, 905911.CrossRefGoogle ScholarPubMed
Chen, YH, Li, L, Chen, W, et al. (2019) Pre-pregnancy underweight and obesity are positively associated with small-for-gestational-age infants in a Chinese population. Sci Rep 9, 15544.CrossRefGoogle ScholarPubMed
Zhao, R, Xu, L, Wu, ML, et al. (2018) Maternal pre-pregnancy body mass index, gestational weight gain influence birth weight. Women Birth 31, e20e25.CrossRefGoogle ScholarPubMed
Sharp, GC, Lawlor, DA, Richmond, RC, et al. (2015) Maternal pre-pregnancy BMI and gestational weight gain, offspring DNA methylation and later offspring adiposity: findings from the Avon Longitudinal Study of Parents and Children. Int J Epidemiol 44, 12881304.CrossRefGoogle ScholarPubMed
Gauster, M, Hiden, U, Blaschitz, A, et al. (2007) Dysregulation of placental endothelial lipase and lipoprotein lipase in intrauterine growth-restricted pregnancies. J Clin Endocrinol Metab 92, 22562263.CrossRefGoogle ScholarPubMed
Gil-Sanchez, A, Demmelmair, H, Parrilla, JJ, et al. (2011) Mechanisms involved in the selective transfer of long chain polyunsaturated Fatty acids to the fetus. Front Genet 2, 57.CrossRefGoogle Scholar
Miletic, T, Stoini, E, Mikulandra, F, et al. (2007) Effect of parental anthropometric parameters on neonatal birth weight and birth length. Coll Antropol 31, 993997.Google ScholarPubMed
Knight, B, Shields, BM, Turner, M, et al. (2005) Evidence of genetic regulation of fetal longitudinal growth. Early Hum Dev 81, 823831.CrossRefGoogle ScholarPubMed
Takagi, K, Iwama, N, Metoki, H, et al. (2019) Paternal height has an impact on birth weight of their offspring in a Japanese population: the Japan Environment and Children’s Study. J Dev Orig Health Dis 10, 542554.CrossRefGoogle Scholar
Griffiths, LJ, Dezateux, C & Cole, TJ (2007) Differential parental weight and height contributions to offspring birthweight and weight gain in infancy. Int J Epidemiol 36, 104107.CrossRefGoogle ScholarPubMed
Albouy-Llaty, M, Thiebaugeorges, O, Goua, V, et al. (2011) Influence of fetal and parental factors on intrauterine growth measurements: results of the EDEN mother-child cohort. Ultrasound Obstet Gynecol 38, 673680.CrossRefGoogle ScholarPubMed
Constancia, M, Hemberger, M, Hughes, J, et al. (2002) Placental-specific IGF-II is a major modulator of placental and fetal growth. Nature 417, 945948.CrossRefGoogle Scholar
Wang, J, Zhou, J, Cheng, CM, et al. (2004) Evidence supporting dual, IGF-I-independent and IGF-I-dependent, roles for GH in promoting longitudinal bone growth. J Endocrinol 180, 247255.CrossRefGoogle ScholarPubMed
Lecomte, V, Maloney, CA, Wang, KW, et al. (2017) Effects of paternal obesity on growth and adiposity of male rat offspring. Am J Physiol Endocrinol Metab 312, E117E125.CrossRefGoogle ScholarPubMed
Lefebvre, G & Samoilenko, M (2017) On the use of the outcome variable ‘small for gestational age’ when gestational age is a potential mediator: a maternal asthma perspective. Bmc Med Res Methodol 17, 165.CrossRefGoogle ScholarPubMed
Savitz, DA, Hertz-Picciotto, I, Poole, C, et al. (2002) Epidemiologic measures of the course and outcome of pregnancy. Epidemiol Rev 24, 91101.CrossRefGoogle ScholarPubMed
Urquia, ML & Ray, JG (2012) Seven caveats on the use of low birthweight and related indicators in health research. J Epidemiol Community Health 66, 971975.CrossRefGoogle ScholarPubMed
Wang, Y, Cao, Z, Peng, Z, et al. (2015) Folic acid supplementation, preconception body mass index, and preterm delivery: findings from the preconception cohort data in a Chinese rural population. BMC Pregnancy Childbirth 15, 336.CrossRefGoogle Scholar
Yang, Y, He, Y, Li, Q, et al. (2015) Preconception blood pressure and risk of preterm birth: a large historical cohort study in a Chinese rural population. Fertil Steril 104, 124130.CrossRefGoogle Scholar
Zhang, S, Wang, Q & Shen, H (2015) Design of the national free proception health examination project in China. Zhonghua Yi Xue Za Zhi 95, 162165.Google ScholarPubMed
Mi, YJ, Zhang, B, Wang, HJ, et al. (2015) Prevalence and secular trends in obesity among Chinese adults, 1991–2011. Am J Prev Med 49, 661669.CrossRefGoogle ScholarPubMed
Zhu, L, Zhang, R, Zhang, S, et al. (2015) Chinese neonatal birth weight curve for different gestational age. Zhonghua Er Ke Za Zhi 53, 97103.Google ScholarPubMed
World Health Organization Global Database on Body Mass Index: BMI Classification (2020) http://apps.who.int/bmi/index.jsp?introPage=intro_3.html (accessed September 2020).Google Scholar
Schoenaker, DA, Simon, D, Chaturvedi, N, et al. (2014) Glycemic control and all-cause mortality risk in type 1 diabetes patients: the EURODIAB prospective complications study. J Clin Endocrinol Metab 99, 800807.CrossRefGoogle Scholar
EHF (2016) RMS: Regression Modeling Strategies. Berlin, Germany: Springer.Google Scholar
Villar, J, Cheikh, IL, Victora, CG, et al. (2014) International standards for newborn weight, length, and head circumference by gestational age and sex: the Newborn Cross-Sectional Study of the INTERGROWTH-21st Project. Lancet 384, 857868.CrossRefGoogle ScholarPubMed
Goldstein, RF, Abell, SK, Ranasinha, S, et al. (2017) Association of gestational weight gain with maternal and infant outcomes: a systematic review and meta-analysis. JAMA 317, 22072225.CrossRefGoogle ScholarPubMed
Ma, M, Zhang, W, Zhang, J, et al. (2020) Effect of paternal body mass index on neonatal outcomes of singletons after frozen-thawed embryo transfer cycles: analysis of 7908 singleton newborns. Fertil Steril 113, 12151223.CrossRefGoogle Scholar
Derraik, J, Pasupathy, D, McCowan, L, et al. (2019) Paternal contributions to large-for gestational-age term babies: findings from a multicenter prospective cohort study. J Dev Orig Health Dis 10, 529535.CrossRefGoogle ScholarPubMed
Campbell, JM & McPherson, NO (2019) Influence of increased paternal BMI on pregnancy and child health outcomes independent of maternal effects: a systematic review and meta-analysis. Obes Res Clin Pract 13, 511521.CrossRefGoogle ScholarPubMed
Pomeroy, E, Wells, JC, Cole, TJ, et al. (2015) Relationships of maternal and paternal anthropometry with neonatal body size, proportions and adiposity in an Australian cohort. Am J Phys Anthropol 156, 625636.CrossRefGoogle Scholar
Shah, PS (2010) Paternal factors and low birthweight, preterm, and small for gestational age births: a systematic review. Am J Obstet Gynecol 202, 103123.CrossRefGoogle ScholarPubMed
Stephenson, J, Heslehurst, N, Hall, J, et al. (2018) Before the beginning: nutrition and lifestyle in the preconception period and its importance for future health. Lancet 391, 18301841.CrossRefGoogle ScholarPubMed
Noor, N, Cardenas, A, Rifas-Shiman, SL, et al. (2019) Association of periconception paternal Body Mass Index with Persistent changes in DNA methylation of offspring in childhood. JAMA Netw Open 2, e1916777.CrossRefGoogle ScholarPubMed
Terashima, M, Barbour, S, Ren, J, et al. (2015) Effect of high fat diet on paternal sperm histone distribution and male offspring liver gene expression. Epigenetics-US 10, 861871.CrossRefGoogle ScholarPubMed
Zhao, H, Zhao, Y, Ren, Y, et al. (2017) Epigenetic regulation of an adverse metabolic phenotype in polycystic ovary syndrome: the impact of the leukocyte methylation of PPARGC1A promoter. Fertil Steril 107, 467474.CrossRefGoogle ScholarPubMed
Li, N, Shen, Q & Hua, J (2016) Epigenetic remodeling in male germline development. Stem Cells Int 2016, 3152173.CrossRefGoogle ScholarPubMed
Van Wyk, JJ & Smith, EP (1999) Insulin-like growth factors and skeletal growth: possibilities for therapeutic interventions. J Clin Endocrinol Metab 84, 43494354.CrossRefGoogle ScholarPubMed
Fowden, AL (2003) The insulin-like growth factors and feto-placental growth. Placenta 24, 803812.CrossRefGoogle ScholarPubMed
Black, RE, Victora, CG, Walker, SP, et al. (2013) Maternal and child undernutrition and overweight in low-income and middle-income countries. Lancet 382, 427451.CrossRefGoogle ScholarPubMed
Figure 0

Fig. 1. Flow chart of the study population. NFPCP, National Free Pre-pregnancy Checkups Project; LMP, last menstrual period.

Figure 1

Table 1. Characteristics of the total study population(Numbers and percentages)

Figure 2

Fig. 2. Associations between maternal/paternal preconception BMI and risk of small/large for gestational age infant (OR and 95 % CI). SGA, small for gestational age infant; AGA, appropriate for gestational age infant; LGA, large for gestational age infant; OR, odds ratios, BMI, body mass index (calculated as the weight in kilograms divided by height in meters squared). * Models were adjusted for maternal and paternal age at LMP, height, ethnic, education, area of residence, alcohol drinking, smoking, passive smoking and paternal BMI, maternal hypertension, diabetes, parity as well as history of adverse pregnancy. † Models were adjusted for maternal and paternal age at LMP, height, ethnic, education, area of residence, alcohol drinking, smoking, passive smoking and maternal BMI, hypertension, diabetes, parity as well as history of adverse pregnancy.

Figure 3

Fig. 3. Adjusted OR of small/large for gestational age infant according to BMI of couples. SGA, small for gestational age infant; AGA, appropriate for gestational age infant; LGA, large for gestational age infant; OR, odds ratios, BMI, body mass index (calculated as the weight in kilograms divided by height in meters squared). *† Multivariable-adjusted OR (95 % CI) was adjusted for maternal and paternal age at LMP, height, ethnic, education, area of residence, alcohol drinking, smoking, passive smoking and maternal hypertension, diabetes, parity as well as history of adverse pregnancy.

Figure 4

Fig. 4. Dose–response relationship between maternal/paternal preconception BMI and risk of small/large for gestational age infant. SGA, small for gestational age infant; LGA, large for gestational age infant; OR, odds ratios, BMI, BMI (calculated as the weight in kilograms divided by height in meters squared). Graphs show the multivariable-adjusted OR of associations between paternal/maternal preconception BMI and the risk of small for gestational age infant (A1, B1), large for gestational age infant (A2, B2), respectively. In the graph, black curves and shaded grey areas show predicted OR and 95 % CI, respectively. A1, A2: Maternal and paternal age at LMP, height, ethnic, education, area of residence, alcohol drinking, smoking, passive smoking and maternal BMI, hypertension, diabetes, parity as well as history of adverse pregnancy were used in the analysis as covariates. B1, B2: Maternal and paternal age at LMP, height, ethnic, education, area of residence, alcohol drinking, smoking, passive smoking, paternal BMI and maternal hypertension, diabetes, parity as well as history of adverse pregnancy were used in the analysis as covariates. The black lines represent the reference level.

Supplementary material: File

Guo et al. supplementary material

Guo et al. supplementary material

Download Guo et al. supplementary material(File)
File 129.9 KB