Relationship between Birth Weight, Early Growth Rate, and Body Composition in 5- to 7-Year-Old Children

Predicting low birth weight risks in pregnant women in Brazil using machine learning algorithms: data from the Araraquara cohort study

BACKGROUND

Low birth weight (LBW) is a critical factor linked to neonatal morbidity and mortality. Early prediction is essential for timely interventions. This study aimed to develop and evaluate predictive models for LBW using machine learning algorithms, including Random Forest, XGBoost, Catboost, and LightGBM.

METHODS

We analyzed data from 1,579 pregnant women enrolled in the Araraquara Cohort, a population-based longitudinal study. Predictor variables included maternal sociodemographic, clinical, and behavioral factors. Four ML algorithms Random Forest, XGBoost, CatBoost, and LightGBM, were trained using an 80/20 train-test split and 10-fold cross-validation. To address class imbalance, the Synthetic Minority Over-sampling Technique (SMOTE) was applied. Model performance was assessed using metrics such as area under the receiver operating characteristic curve (AUROC), F1-score, and precision-recall. Variable importance was evaluated using Shapley values.

RESULTS

XGBoost demonstrated the best performance, achieving an AUROC of 0.94, followed by CatBoost (0.94), Random Forest (0.94), and LightGBM (0.94). Maternal gestational age was the most influential predictor, followed by marital status and prenatal care frequency. Behavioral factors, such as physical activity, also contributed to LBW risk. Shapley analysis provided interpretable insights into variable contributions, supporting the clinical applicability of the models.

CONCLUSION

Machine learning, combined with SMOTE, proved to be an effective approach for predicting LBW. XGBoost stood out as the most accurate model, but Catboost and Random Forest also provided solid results. These models can be applied to identify high-risk pregnancies, improving perinatal outcomes through early interventions.

Growth Status of Full-Term Infants with Different Sizes for Gestational Age During the First Year of Life

Objective

This study aimed to assess the growth of full-term infants with different sizes at birth and examine catch-up and catch-down growth in their first year.

Methods

This retrospective population-based cohort study was based on the Guangdong Provincial Women and Children Health Information System. 194797 full-term singleton live births were extracted. Measurements for weight and length were taken at birth, 6 months, and 12 months. The size-for-gestational age was categorized as small (SGA, <10th centile), appropriate (AGA, 10th-90th centiles), or large (LGA, >90th centile) based on the international newborn size for gestational age and sex INTERGROWTH-21st standards. Catch-up and catch- down growth were defined as a change in standard deviation in z-score greater than 0.67 in the growth curves.

Results

Of the 194797 full-term singletons, the average gestational age was 39.28 ± 1.03 weeks, and the average weight of the newborns was 3205 ± 383 grams. 15632 infants were identified as SGA (8.0%) and 12756 were LGA (6.5%). At 1 year of age, catch-up growth in weight was observed in 63.0% of SGA infants, 29.5% of AGA infants, and 5.4% of LGA infants. Conversely, catch-down growth occurred in 3.3% of SGA infants, 17.8% of AGA infants, and 54.7% of LGA infants. The proportions of catch-up growth in length for SGA, AGA, and LGA infants within the first year were 31.4%, 22.5%, and 17.1%, respectively. Catch-up or catch-down growth predominantly occurred before 6 months of age. However, from 6 to 12 months, there was no significant variation in WAZ among children with different birth sizes.

Conclusion

In their first year of life, full-term singleton live births tend towards regression to the mean in their postnatal weight and length. The average delay in the growth of LGA is compensated by an increase in it of the SGA. Early monitoring and intervention are crucial for optimizing growth in infants with different birth sizes.

Early adiposity rebound: predictors and outcomes

Adiposity rebound (AR) refers to the second rise of the body mass index (BMI) curve that usually occurs between six and eight years of age. AR timing has a significant impact on patients' health: early AR (EAR), usually before the age of five, is considered to be the earliest indicator of obesity and its related health conditions later in life. Many studies have evaluated factors that can be predictors of EAR, and identified low birth weight and gestational weight gain as novel predictors of EAR, highlighting the role of the intrauterine environment in the kinetics of adiposity. Furthermore, children with breastfeeding longer than 4 months have been found to be less likely to have an EAR, whereas children born to advanced-age mothers, high maternal BMI had a higher risk of having an EAR. Some differences were found in the timing of AR in boys and girls, with girls being more likely to have EAR. The aim of this review is to answer the following three questions: 1) Which are the prenatal and perinatal factors associated with increased risk of EAR? Is gender one of these? 2) Which are the outcomes of EAR in childhood and in adulthood? 3) Which measures can be taken in order to prevent premature AR?

Fetal overgrowth and weight trajectories during infancy and adiposity in early childhood

BACKGROUND

Large-for-gestational age (LGA), a marker of fetal overgrowth, has been linked to obesity in adulthood. Little is known about how infancy growth trajectories affect adiposity in early childhood in LGA.

METHODS

In the Shanghai Birth Cohort, we followed up 259 LGA (birth weight >90th percentile) and 1673 appropriate-for-gestational age (AGA, 10th-90th percentiles) children on body composition (by InBody 770) at age 4 years. Adiposity outcomes include body fat mass (BFM), percent body fat (PBF), body mass index (BMI), overweight/obesity, and high adiposity (PBF >85th percentile).

RESULTS

Three weight growth trajectories (low, mid, and high) during infancy (0-2 years) were identified in AGA and LGA subjects separately. BFM, PBF and BMI were progressively higher from low- to mid-to high-growth trajectories in both AGA and LGA children. Compared to the mid-growth trajectory, the high-growth trajectory was associated with greater increases in BFM and the odds of overweight/obesity or high adiposity in LGA than in AGA children (tests for interactions, all P < 0.05).

CONCLUSIONS

Weight trajectories during infancy affect adiposity in early childhood regardless of LGA or not. The study is the first to demonstrate that high-growth weight trajectory during infancy has a greater impact on adiposity in early childhood in LGA than in AGA subjects.

IMPACT

Large-for-gestational age (LGA), a marker of fetal overgrowth, has been linked to obesity in adulthood, but little is known about how weight trajectories during infancy affect adiposity during early childhood in LGA subjects. The study is the first to demonstrate a greater impact of high-growth weight trajectory during infancy (0-2 years) on adiposity in early childhood (at age 4 years) in subjects with fetal overgrowth (LGA) than in those with normal birth size (appropriate-for-gestational age). Weight trajectory monitoring may be a valuable tool in identifying high-risk LGA children for close follow-ups and interventions to decrease the risk of obesity.

Association between food environments and fetal growth in pregnant Brazilian women

INTRODUCTION

Birth weight is described as one of the main determinants of newborns' chances of survival. Among the associated causes, or risk factors, the mother's nutritional status strongly influences fetal growth and birth weight outcomes of the concept. This study evaluates the association between food deserts, small for gestational age (SGA), large for gestational age (LGA) and low birth weight (LBW) newborns.

DESIGN

This is a cross-sectional population study, resulting from individual data from the Live Birth Information System (SINASC), and commune data from mapping food deserts (CAISAN) in Brazil. The newborn's size was defined as follows: appropriate for gestational age (between 10 and 90th percentile), SGA (< 10th percentile), LGA (> 90th percentile), and low birth weight < 2,500 g. To characterize food environments, we used tertiles of the density of establishments which sell in natura and ultra-processed foods. Logistic regression modeling was conducted to investigate the associations of interest.

RESULTS

We analyzed 2,632,314 live births in Brazil in 2016, after appropriate adjustments, women living in municipalities with limited availability of fresh foods had a higher chance of having newborns with SGA [OR2nd tertile: 1.06 (1.05-1.07)] and LBW [OR2nd tertile: 1.11 (1.09-1.12)]. Conversely, municipalities with greater availability of ultra-processed foods had a higher chance of having newborns with SGA [OR3rd tertile: 1.04 (1.02-1.06)] and LBW [OR2nd tertile: 1.13 (1.11-1.16)]. Stratification by race showed that Black and Mixed/Brown women had a higher chance of having newborns with SGA [OR3rd tertile: 1.09 (1.01-1.18)] and [OR3rd tertile: 1.06 (1.04-1.09)], respectively, while Mixed-race women also had a higher chance of having newborns with LBW [OR3rd tertile: 1.17 (1.14-1.20)]. Indigenous women were associated with LGA [OR3rd tertile: 1.20 (1.01-1.45)].

CONCLUSION

The study found that living in areas with limited access to healthy foods was associated with an increased risk of SGA and low birth weight among newborns, particularly among Black and Mixed/Brown women. Therefore, urgent initiatives aimed at reducing social inequalities and mitigating the impact of poor food environments are needed in Brazil.

Association between food environments and fetal growth in pregnant Brazilian women

Introduction

Birth weight is described as one of the main determinants of newborns’ chances of survival. Among the associated causes, or risk factors, the mother’s nutritional status strongly influences fetal growth and birth weight outcomes of the concept. This study evaluates the association between food deserts, small for gestational age (SGA), large for gestational age (LGA) and low birth weight (LBW) newborns.

Design

This is a cross-sectional population study, resulting from individual data from the Live Birth Information System (SINASC), and commune data from mapping food deserts (CAISAN) in Brazil. The newborn’s size was defined as follows: appropriate for gestational age (between 10 and 90th percentile), SGA (< 10th percentile), LGA (> 90th percentile), and low birth weight < 2,500 g. To characterize food environments, we used tertiles of the density of establishments which sell in natura and ultra-processed foods. Logistic regression modeling was conducted to investigate the associations of interest.

Results

We analyzed 2,632,314 live births in Brazil in 2016, after appropriate adjustments, women living in municipalities with limited availability of fresh foods had a higher chance of having newborns with SGA [OR2nd tertile: 1.06 (1.05–1.07)] and LBW [OR2nd tertile: 1.11 (1.09–1.12)]. Conversely, municipalities with greater availability of ultra-processed foods had a higher chance of having newborns with SGA [OR3rd tertile: 1.04 (1.02–1.06)] and LBW [OR2nd tertile: 1.13 (1.11–1.16)]. Stratification by race showed that Black and Mixed/Brown women had a higher chance of having newborns with SGA [OR3rd tertile: 1.09 (1.01–1.18)] and [OR3rd tertile: 1.06 (1.04–1.09)], respectively, while Mixed-race women also had a higher chance of having newborns with LBW [OR3rd tertile: 1.17 (1.14–1.20)]. Indigenous women were associated with LGA [OR3rd tertile: 1.20 (1.01–1.45)].

Conclusion

The study found that living in areas with limited access to healthy foods was associated with an increased risk of SGA and low birth weight among newborns, particularly among Black and Mixed/Brown women. Therefore, urgent initiatives aimed at reducing social inequalities and mitigating the impact of poor food environments are needed in Brazil.

The association of intrauterine and postnatal growth patterns and nutritional status with toddler body composition

BACKGROUND

Growth patterns may be indicative of underlying changes in body composition. However, few studies have assessed the association of growth and body composition in poorly resourced regions experiencing the double-burden of malnutrition exists. Thus, the aims of this study were to investigate the association of intrauterine and postnatal growth patterns with infant body composition at 2 years in a middle-income country.

METHODS

Participants were from the International Atomic Energy Agency Multicentre Body Composition Reference study. Fat mass (FM), fat free mass (FFM), Fat mass index (FMI), fat free mass index (FFMI), and percentage fat mass (%FM) were measured in 113 infants (56 boys and 57 girls), from Soweto, South Africa, using deuterium dilution from 3 to 24 months. Birthweight categories were classified using the INTERGROWTH-21 standards as small (SGA), appropriate (AGA), and large-for gestational age (LGA). Stunting (> -2 SDS) was defined using the WHO child growth standards. Birthweight z-score, conditional relative weight and conditional length at 12 and 24 mo were regressed on body composition at 24 mo.

RESULTS

There were no sex differences in FM, FFM, FMI and FFMI between 3 and 24 mo. SGA and AGA both had significantly higher %FM than LGA at 12 mo. LGA had higher FM at 24 mo. Children with stunting had lower FM (Mean = 1.94, 95% CI; 1.63-2.31) and FFM (Mean = 5.91, 95% CI; 5.58-6.26) at 12 mo than non-stunting, while the reverse was true for FFMI (Mean = 13.3, 95% CI; 12.5-14.2) at 6 mo. Birthweight and conditionals explained over 70% of the variance in FM. CRW at both 12 and 24 mo was positively associated with FM and FMI. CRW at 12 mo was also positively associated with FMI, while CH at 24 mo was negatively associated with both FFMI and FMI in boys.

CONCLUSION

Both LGA and SGA were associated with higher body fat suggesting that both are disadvantaged nutritional states, likely to increase the risk of obesity. Growth patterns through infancy and toddler period (1-2 years) are indicative of body fat, while growth patterns beyond infancy are less indicative of fat-free mass.