Abstract
BACKGROUND: Iron, an essential micronutrient, plays a critical role in fetal neurodevelopment. Animal studies have demonstrated that maternal iron-deficient diets during pregnancy induce permanent structural and functional alterations in offspring brains. Dietary iron exists in two forms: heme iron (found in animal-derived foods), which exhibits high bioavailability (15-35%), and non-heme iron (predominantly from plant-based sources), with lower bioavailability (1-20%). Existing epidemiological studies have reported inconsistent findings regarding maternal iron intake and offspring neurodevelopment, and few have examined the effects of different iron forms. METHODS: This study was conducted in the Jiangsu Birth Cohort, a prospective cohort tracking pregnant women throughout gestation and following up their children. Dietary intake, including heme and non-heme iron, was assessed via a semi-quantitative food frequency questionnaire in early, middle and late pregnancy. The total iron intake was defined as the sum of iron intake from diet and supplements (specific iron supplements and multivitamin/ mineral supplements). Infant neurodevelopment was evaluated at 12 months of age using the Bayley Scales of Infant and Toddler Development, Third Edition screening test. This assessment covered cognitive, receptive communication, expressive communication, fine motor, and gross motor domains. Each domain was scored according to standardized criteria and categorized as "non-optimal" or "optimal" based on age-specific cut-off points. Poisson regression and generalized estimating equations were employed to analyze the associations between maternal iron intake and neurodevelopment of offspring. Furthermore, maternal (demographic, lifestyle, and clinical) and infant (birth and feeding) characteristics that might confound the associations were adjusted in the analysis. RESULTS: The final analytical cohort comprised 3,750 pregnant women and their offspring. Null associations were observed between total iron intake through pregnancy and infant neurodevelopment. Following log-transformation and energy adjustment, each one-unit increase in maternal heme iron intake was associated with a 35% reduced risk of non-optimal cognitive development in infants after the adjustment for potential confounders (RR = 0.65, 95% CI 0.45-0.93). Particularly, trimester-specific analysis demonstrated that maternal heme iron intake in the third trimester was significantly associated with non-optimal cognition development (RR = 0.67, 95% CI 0.52-0.85). Infants of mothers in the highest tertile of heme iron intake (> 3.29 mg/d) during late pregnancy exhibited a 28% lower risk of non-optimal cognition compared to those in the lowest tertile (< 2.22 mg/d) (RR = 0.72, 95% CI 0.56-0.93). While no association was observed for non-heme iron or iron supplements. CONCLUSIONS: Maternal heme iron intake, particularly in late pregnancy, may contribute to optimal infant neurodevelopment. These findings emphasize the importance of evaluating the distinct roles of maternal heme and non-heme iron intakes on neurodevelopment.