IntroductionAccumulating evidence has shown that maternal metabolic conditions, such as pre-pregnancy overweight, diabetes mellitus, and hypertensive disorders of pregnancy (HDP) are potential risk factors of autism spectrum disorder (ASD). However, it remains unclear how these maternal conditions lead to neurodevelopmental outcomes in the offspring, including autistic symptoms. Leptin, an adipokine that has pro-inflammatory effects and affects fetal neurodevelopment, is a candidate mediator of the association between maternal metabolic factors and an increased risk of ASD. However, whether prenatal exposure to leptin mediates the association between maternal metabolic conditions and autistic symptoms in children has not been investigated yet.MethodsThis study investigated the associations between mothers' metabolic conditions (pre-pregnancy overweight, diabetes mellitus during or before pregnancy, and HDP), leptin concentrations in umbilical cord serum, and autistic symptoms among 762 children from an ongoing cohort study, using generalized structural equation modeling. We used the Social Responsive Scale, Second Edition (SRS-2) at 8–9 years old to calculate total T-scores. Additionally, we used the T-scores for two subdomains: Social Communication and Interaction (SCI) and Restricted Interests and Repetitive Behavior (RRB).ResultsUmbilical cord leptin levels were associated with pre-pregnancy overweight [coefficient = 1.297, 95% confidence interval (CI) 1.081–1.556, p = 0.005] and diabetes mellitus (coefficient = 1.574, 95% CI 1.206–2.055, p = 0.001). Furthermore, leptin levels were significantly associated with SRS-2 total T-scores (coefficient = 1.002, 95% CI 1.000–1.004, p = 0.023), SCI scores (coefficient = 1.002, 95% CI 1.000–1.004, p = 0.020), and RRB scores (coefficient = 1.001, 95% CI 1.000–1.003, p = 0.044) in children. Associations between maternal metabolic factors and autistic symptoms were not significant.DiscussionThe present study uncovered an association between cord leptin levels and autistic symptoms in children, while maternal metabolic conditions did not have an evident direct influence on the outcome. These results imply that prenatal pro-inflammatory environments affected by maternal metabolic conditions may contribute to the development of autistic symptoms in children. The findings warrant further investigation into the role of leptin in the development of autistic symptoms.