Advanced Maternal Age and Risk of Large-for-Gestational-Age Infants: Mediating Roles of Prenatal Body Mass Index and Specific Lipid Profiles

Abstract

Background Advanced maternal age (AMA) is associated with adverse pregnancy outcomes, but the mechanisms linking AMA to fetal overgrowth remain poorly quantified. This retrospective cohort study aimed to estimate the potential statistical mediating roles of prenatal BMI and blood lipids in the association between AMA and large-for-gestational-age (LGA) infants. Methods In this retrospective cohort study, 11,504 singleton pregnancies (28–41 weeks) in Changzhou, China, were analyzed. Women were classified as AMA (≥35 years, n=1,330) or non‑AMA (n=10,174). Multivariable regression and mediation analyses assessed associations between AMA, lipid profiles, BMI, and LGA risk. Results AMA pregnancies showed significantly higher LGA rates than non-AMA pregnancies (24.44% vs 14.37%, P<0.001). After adjustment, AMA independently increased LGA risk (adjusted OR: 1.37, 95% CI: 1.18–1.60, P<0.001). AMA was associated with higher prenatal BMI (β=0.37 kg/m2), elevated triglycerides (β=0.27 mmol/L), and lower low density lipoprotein-cholesterol (LDL-C) (β=-0.07 mmol/L) (all P<0.001). These metabolic factors independently predicted LGA risk: BMI (adjusted OR: 1.20 per 1 kg/m2 increase), triglycerides (adjusted OR: 1.13 per 1 mmol/L increase), and LDL-C (adjusted OR: 1.32 per 1 mmol/L decrease) (all P<0.001). Exploratory mediation analysis estimated that prenatal BMI statistically accounted for 19.8% of the total AMA‑LGA effect, while triglycerides and LDL‑C accounted for 5.1% and 2.8%, respectively. Conclusion AMA is independently associated with increased LGA risk, and this association is statistically partially mediated by higher BMI, elevated triglycerides, and lower LDL‑C. Given the retrospective observational design and the single‑timepoint measurement of lipids at delivery, these findings should be interpreted as statistical decompositions of associations rather than as causal. Nevertheless, they suggest that BMI and lipid profiles may be useful for risk stratification in older pregnant women, though prospective validation is warranted.