Associations of Maternal Dietary Patterns during Pregnancy with Offspring Adiposity from Birth Until 54 Months of Age

Abstract

Most studies linking maternal diet with offspring adiposity have focused on single nutrients or foods, but a dietary pattern approach is more representative of the overall diet. We thus aimed to investigate the relations between maternal dietary patterns and offspring adiposity in a multi-ethnic Asian mother-offspring cohort in Singapore. We derived maternal dietary patterns using maternal dietary intake information at 26-28 weeks of gestation, of which associations with offspring body mass index (BMI), abdominal circumference (AC), subscapular skinfold (SS), and triceps skinfold (TS) were assessed using longitudinal data analysis (linear mixed effects (LME)) and multiple linear regression at ages 0, 3, 6, 9, 12, 15, 18, 24, 36, 48, and 54 months. Three dietary patterns were derived: (1) vegetables-fruit-and-white rice (VFR); (2) seafood-and-noodles (SfN); and (3) pasta-cheese-and-bread (PCB). In the LME model adjusting for potential confounders, each standard deviation (SD) increase in maternal VFR pattern score was associated with 0.09 mm lower offspring TS. Individual time-point analysis additionally revealed that higher VFR score was generally associated with lower postnatal offspring BMI z-score, TS, SS, and sum of skinfolds (SS + TS) at ages 18 months and older. Maternal adherence to a dietary pattern characterized by higher intakes of fruit and vegetables and lower intakes of fast food was associated with lower offspring adiposity.

Keywords: BMI; adiposity; children; developmental origins of health and diseases; dietary patterns; fruit; pregnancy; subscapular skinfold; triceps skinfold; vegetables.

Figure 1

Participant flow chart. BMI, body mass index; SS, subscapular skinfold; TS, triceps skinfold; SST, sum of skinfolds thickness; AC, abdominal circumference.

Figure 2

Predicted adjusted mean values with corresponding standard errors (from linear mixed effects models): (A) BMI z-score (n = 1048); (B) subscapular skinfold thickness (n = 1034); (C) triceps skinfold thickness (n = 1036); and (D) sum of skinfolds thickness (n = 1034) according to quartiles of maternal vegetables-fruit-and-white rice pattern score. The linear mixed effects models were adjusted for exact age at measurement, infant sex (except for BMI z-score), birth order, gestational age, duration of any breastfeeding, ethnicity, maternal age, height, pre-pregnancy BMI, weight gain until 26–28 weeks gestation, education level, gestational diabetes, energy intake, and scores of the other two dietary patterns (e.g., adjusting for SfN and PCB pattern scores for associations between VFR pattern and childhood adiposity). Median (range) of standardized dietary pattern scores: −1.09 (−3.26, −0.65) SD for Q1; −0.31 (−0.64, −0.02) SD for Q2; 0.28 (−0.02, 0.67) SD for Q3; 1.16 (0.68, 3.27) for Q4. Q, quartile.

Figure 2

Predicted adjusted mean values with corresponding standard errors (from linear mixed effects models): (A) BMI z-score (n = 1048); (B) subscapular skinfold thickness (n = 1034); (C) triceps skinfold thickness (n = 1036); and (D) sum of skinfolds thickness (n = 1034) according to quartiles of maternal vegetables-fruit-and-white rice pattern score. The linear mixed effects models were adjusted for exact age at measurement, infant sex (except for BMI z-score), birth order, gestational age, duration of any breastfeeding, ethnicity, maternal age, height, pre-pregnancy BMI, weight gain until 26–28 weeks gestation, education level, gestational diabetes, energy intake, and scores of the other two dietary patterns (e.g., adjusting for SfN and PCB pattern scores for associations between VFR pattern and childhood adiposity). Median (range) of standardized dietary pattern scores: −1.09 (−3.26, −0.65) SD for Q1; −0.31 (−0.64, −0.02) SD for Q2; 0.28 (−0.02, 0.67) SD for Q3; 1.16 (0.68, 3.27) for Q4. Q, quartile.