Maternal dietary patterns as predictors of neonatal body composition in Ethiopia: the IABC birth cohort study

Abstract

Background: Malnutrition during pregnancy is associated with adverse birth outcomes, but the importance of maternal diet during pregnancy for neonatal body composition remains inconclusive. This study investigated the role of maternal diet during pregnancy for neonatal body composition in the Ethiopian iABC birth cohort.

Methods: The data stemmed from the first visit at birth comprising 644 mother-child pairs. Shortly after delivery, the diet of the last week of pregnancy was assessed by a non-quantitative and non-validated 18-items food frequency questionnaire. Multiple imputation was used to handle missing data. Twin births and implausible values were excluded from analysis (n = 92). The Dietary Diversity Score (0-9 points) was constructed and exploratory dietary patterns were derived via principal component analysis. Neonatal fat mass and fat-free mass were assessed by air-displacement plethysmography. The associations of maternal Dietary Diversity Score and exploratory dietary patterns with gestational age, neonatal anthropometric measures and body composition were investigated using multiple-adjusted linear regression analysis.

Results: In this cohort (n = 552), mean ± standard deviation (SD) mother's age was 24.1 ± 4.6 years and the median maternal Dietary Diversity Score was 6 (interquartile range = 5-7). An 'Animal-source food pattern' and a 'Vegetarian food pattern' were identified. The mean ± SD birth weight was 3096 ± 363 g and gestational age was 39.0 ± 1.0 weeks. Maternal adherence to the Animal-source food pattern, but not Vegetarian food pattern, was related to birth weight [79.5 g (95% confidence interval (CI): -14.6, 173.6)]. In the adjusted model, adherence to the Animal-source food pattern was associated with higher neonatal fat-free mass [53.1 g (95% CI: -20.3, 126.6)], while neonates of women with high compared to low adherence to Dietary Diversity Score and Vegetarian food pattern had higher fat mass [19.4 g (95% CI: -7.4, 46.2) and 33.5 g (95% CI: 2.8, 64.1), respectively].

Conclusions: In this Ethiopian population, maternal diet during pregnancy was associated with neonatal body composition. The analysis of body composition adds important detail to the evaluation of maternal dietary habits for the newborn constitution.

Keywords: Body composition; Dietary diversity; Dietary patterns; Ethiopia; Neonate; Pregnancy; iABC.

Fig. 1

Dietary patterns derived by principal component analysis and rotated factor loadings in 552 pregnant Ethiopian women. Analytical sample (N total = 552) excluded twin births (n = 18) and implausible values for Fat-free mass index and Fat mass index (n = 74)

Fig. 2

Associations of maternal adherence to the Dietary Diversity Score (♦), Animal-source food pattern () or Vegetarian food pattern (▲) with neonatal absolute (A) fat-free mass (g) and (B) fat mass (g) among 552 Ethiopian mother-child-pairs. Multiple-adjusted mean differences (β), 95% confidence intervals and p values were calculated by linear regression and adjusted for age (mother), sex (child), gestational age, parity, mode of delivery, delivery complications, supplementation and medication during pregnancy, number of antenatal care visits, diseases, Dietary Diversity Score (exploratory dietary patterns only), socioeconomic variables (possession of consumer durables, access to electricity and private piped water, mother’s and father’s occupation and education). Analytical sample (N total = 552) excluded twin births (n = 18) and implausible values for Fat-free mass index and Fat mass index (n = 74). T = tertile