Differential effect of pre-pregnancy low BMI on fetal macrosomia: a population-based cohort study

Pregnant women at 15 to 20 weeks of gestation (n=139,472) were recruited to this study during their first prenatal visit in Qingdao from August 1, 2018, to June 30, 2020. Data were registered in the Women and Children’s Health Care Center System, which was established in 2018 under the charge of Qingdao Women and Children’s Hospital. This registry provides comprehensive information covering results of regular health examinations, pre-pregnancy, and delivery details. Pregnant women and their detailed information at each gestation stage were identified and extracted using their ID number. Exclusion criteria included pregnant women with pre-existing diabetes (n=433), missing data for pre-pregnancy weight/height (n=348), termination/abortion before 24–28 gestational weeks (n=448), loss to follow-up before 24–28 gestational weeks (n=3492, no 75 g oral glucose tolerance test (OGTT) screening (n=14,629), termination/abortion (n=444), loss to follow=up (n=11,942), multiple births (n=1172), and dysmorphia (n=776) or missing key variables (n=20) (Fig. 1). A total of 105,768 singleton births were eligible for final analysis. Participation in the study was voluntary, and written informed consent was obtained from each study subject. This study was approved by the Institutional Review Board of Qingdao Women and Children’s Hospital Ethics (No. 002-2018-FEKY).

Detailed information of demographic characteristics, folic acid supplements, parity, alcohol drinking, tobacco smoking, and occupational physical activity were collected using standardized questionnaires. Meanwhile, responses from participants were checked by trained interviewers to improve the validity of the self-reported data. The primary outcome was macrosomia. Macrosomia was defined as an infant with birth weight ≥ 4000 g [19] and were further classified into three subgroups: grade 1: 4000–4499 g, grade 2: 4500–4999 g; grade 3: ≥ 5000 g [20]. Maternal pre-pregnancy BMI was calculated from self-reported values of height and weight before conception according to subjects’ answers to the questionnaire. Maternal pre-pregnancy BMI was categorized as low BMI (BMI<18.5 kg/m²), normal (BMI: 18.5–24.9 kg/m²), overweight (BMI: 25.0–29.9 kg/m²), and obese (BMI ≥30.0 kg/m²) based on the World Health Organization (WHO) criteria [21]. GDM was diagnosed in accordance with the International Association of Diabetes and Pregnancy Study Group recommendation (IADPSG) using 75 g 2-h OGTT: a fasting glucose ≥5.1 mmol/L, or a 1-h result ≥10.0 mmol/L, or a 2-h result ≥ 8.5 mmol/L [22]. Self-reported occupational physical activities were categorized into 3 levels: (1) light (mostly sitting for office work, e.g., secretary), (2) moderate (standing and walking, e.g., sale assistant, craftspeople), and (3) active (walking and lifting, heavy manual labor, e.g., industrial or farm worker) [23]. Maternal smoking was defined as mothers who smoked at least one cigarette per day and kept smoking for over 3 months before conception or kept smoking during gestation. Due to the very low prevalence rate of maternal smoking (before gestation 0.66% [n=693], during gestation 0.17% [n=176]), this factor was not included in the analysis. Paternal smoking was defined as fathers who smoked at least one cigarette per day and kept smoking for over 3 months. Maternal drinking was defined as mothers who drank alcohol regularly before or during gestation; paternal drinking was defined as fathers who drank alcohol regularly before conception.

Demographic characteristics, lifestyle, and environmental exposures of the study population are summarized in Table 1. A total of 105,768 live births were included in the final analysis where 12,335 (11.66%) macrosomic infants were identified. A higher proportion of macrosomia was observed in male fetuses, multiparous, maternal age of 35 years and above, mothers with lower educational level (high school and below), maternal BMI above 24.9 kg/m², and current and previous GDM. Besides, women without anemia and preterm birth more often had macrosomia, while other factors such as paternal smoking, paternal and maternal drinking, and maternal thyroid disease might not contribute to substantial differences in macrosomia prevalence. Due to the very low prevalence rate of maternal smoking (before gestation 0.66% [n=693], during gestation 0.17% [n=176]), this factor was not included in the analysis.

Albeit multiparous mothers (51.53%) had an increased risk of giving birth to a macrosomia infant (OR=1.24, 95% CI 1.19–1.28), the interactive effect between pre-pregnancy BMI and maternal age on macrosomia was statistically significant in both nullipara (P_adjusted=0.0265) and multipara (P_adjusted=0.0356). We further evaluated the association between macrosomia and pre-pregnancy BMI stratifying by maternal age, parity, and fetal genders (Table 2). A significantly increased risk of macrosomia is in mothers who were overweight or obese in both nullipara and multipara (adjusted P values all <0.05). Maternal low BMI (for nullipara: adjusted OR [aOR]= 0.47, 95% CI 0.42–0.53; for multipara, aOR=0.45, 95% CI 0.39–0.52) was inversely associated with macrosomia, and such association was consistent across all maternal ages and in both fetal genders. Specifically, the effect of pre-pregnancy BMI on macrosomia varied with maternal age in both nullipara and multipara. Nulliparous women aged 35 years and above were most benefited from a lower pre-pregnancy BMI against fetal macrosomia (aOR=0.16, 95% CI 0.05–0.49) compared with women of other age groups, whereas multiparous mothers younger than 25 years of age who were low BMI before pregnancy had the lowest risk for macrosomia (aOR=0.17, 95% CI 0.05–0.55). The inverse relationship between low BMI and macrosomia also holds true regarding GDM status in nullipara (non-GDM: aOR= 0.47, 95% CI 0.42–0.53; GDM: aOR= 0.48, 95% CI 0.33–0.70). As for multipara, a lower BMI did not appear to be a protective factor against macrosomia considering GDM status (previous GDM: aOR= 0.81, 95% CI 0.43–1.52; previous and current GDM: aOR= 0.41, 95% CI 0.16–1.04).

To further determine the relationship between pre-pregnancy BMI and fetal macrosomia, other maternal factors including parity, maternal age, and GDM status were jointly evaluated in Fig. 2. In nullipara without GDM, a lower BMI was inversely associated with macrosomia across all age groups (Fig. 2A). In particular, elder mothers were more benefited from a lower BMI (<18.5 kg/m²) as the adjusted odds ratios for macrosomia gradually declined from 0.64 (95% CI 0.51–0.80) in the 20-to-24-year-old group to 0.43 (95% CI 0.36–0.52) in the 25-to-29-year-old group to 0.40 (95% CI 0.29–0.53) in the 30-to-34-year-old group and hit the bottom of only 0.19 (95% CI 0.06–0.60) among those aged 35 years and above. A similar pattern could also be observed in nulliparous mothers with GDM, but the magnitude of the effect was less pronounced (Fig. 2B). Among multiparous mothers without GDM (neither current nor previous GDM), low BMI was also a protective factor for macrosomia, but contrary to nulliparous mother, the protection was more significant in the younger age group (20-to-24-year-old group: aOR=0.21, 95% CI 0.06–0.68; 25-to-29-year-old group: aOR=0.45, 95% CI 0.33–0.60; 30-to-34-year-old group: aOR=0.44, 95% CI 0.34–0.56; ≥35-year-old group: aOR=0.45 95% CI 0.29–0.70) (Fig. 2C). In multipara with current and/or previous GDM, the protective effect of low BMI only reached a significant level among those aged 30–34 years (aOR=0.47, 95% CI 0.28–0.81) but not the other age groups (Fig. 2D). Among pregnant women who had current and/or previous GDM, none of nullipara aged 35 years with a BMI below 18.5 kg/m² nor multipara aged 20-to-24 years with a BMI below 18.5 kg/m² or above 30.0 kg/m² gave birth to a macrosomic neonate. Hence, these groups of subjects were not shown in Fig. 2.