Perinatal and maternal factors associated with Autism Spectrum Disorder.

Abstract

Autism spectrum disorder (ASD) is a heterogeneous neurodevelopmental condition influenced by both genetic and environmental factors. Prenatal and perinatal exposures have been implicated in ASD etiology, but their influence may vary across clinical subgroups, including subgroups defined by co-occurring intellectual disability (ID). We conducted a population-based case-control study in southern Sweden including all children diagnosed with ASD before the age of 9, and whose mothers were born in Sweden. Diagnoses were confirmed through detailed medical record review, and information on ASD severity, ID status, and familial ASD were collected for subgroup analyses. A total of 996 ASD cases and 9,960 age- and sex-matched controls were identified from a regional perinatal database. Multivariable logistic regression estimated adjusted odds ratios (aORs) and 95% confidence intervals (CIs) for maternal, obstetric, and neonatal factors. Higher maternal body mass index (BMI) in early pregnancy was associated with increased likelihood of ASD (adjusted odds ratio [aOR] 1.41-1.76 for overweight and obesity compared with normal weight), with broadly similar associations observed across ASD subgroups defined by severity, intellectual disability, and familial ASD. Maternal smoking in early pregnancy (aOR 1.49, 95% CI 1.22-1.82) and both elective (aOR 1.22, 95% CI 1.01-1.48) and emergency cesarean delivery (aOR 1.40, 95% CI 1.15-1.81) were also associated with higher odds of ASD, with generally stronger associations in children without intellectual disability and in those with less severe ASD. Subgroup-specific associations were observed for maternal epilepsy and gestational diabetes, while prematurity showed weaker associations than anticipated and was mainly observed in severe ASD and non-familial cases. Low Apgar scores at 5 minutes showed no consistent association with ASD. Multiple maternal and perinatal factors were associated with ASD in this large Swedish cohort. Stratified analyses by ASD severity, ID status, and familial ASD revealed both shared and subgroup-specific association patterns, underscoring the value of considering ASD heterogeneity in studies of neurodevelopmental variation.