AUTHOR=Cochran David M. , Jensen Elizabeth T. , Frazier Jean A. , Jalnapurkar Isha , Kim Sohye , Roell Kyle R. , Joseph Robert M. , Hooper Stephen R. , Santos Hudson P. , Kuban Karl C. K. , Fry Rebecca C. , O’Shea T. Michael 

TITLE=Association of prenatal modifiable risk factors with attention-deficit hyperactivity disorder outcomes at age 10 and 15 in an extremely low gestational age cohort

JOURNAL=Frontiers in Human Neuroscience

VOLUME=Volume 16 - 2022

YEAR=2022

URL=https://www.frontiersin.org/journals/human-neuroscience/articles/10.3389/fnhum.2022.911098

DOI=10.3389/fnhum.2022.911098

ISSN=1662-5161

ABSTRACT=BACKGROUND: The increased risk of developing Attention-Deficit Hyperactivity Disorder (ADHD) in extremely preterm infants is well-documented. Better understanding of perinatal risk factors, particularly those that are modifiable, can inform prevention efforts.  
METHODS: We examined data from the Extremely Low Gestational Age Newborns (ELGAN) Study participants were screened for ADHD at age 10 with the Child Symptom Inventory-4 (N = 734) and assessed at age 15 with a structured diagnostic interview (MINI-KID) to evaluate for the diagnosis of ADHD (N = 575). We studied associations of pre-pregnancy maternal body mass index (BMI), pregestational and/or gestational diabetes, maternal smoking during pregnancy (MSDP), and hypertensive disorders of pregnancy (HDP) with 10-year and 15-year ADHD outcomes. Relative risks were calculated using log-binomial regression models, adjusted for maternal age, maternal educational status, use of food stamps, public insurance status, marital status at birth, and family history of ADHD. We defined ADHD as a positive screen on the CSI-4 at age 10 and/or meeting DSM-5 criteria at age 15 on the MINI-KID. We evaluated the robustness of the associations to broadening or restricting the definition of ADHD. We limited the analysis to individuals with IQ ≥ 70 to decrease confounding by cognitive functioning. We evaluated interactions between BMI and diabetes status. We assessed for mediation of risk increase by alterations in inflammatory or neurotrophic protein levels in the first week of life.
RESULTS: Elevated maternal BMI and maternal diabetes were each associated with a 55-65% increase in risk of ADHD, with evidence of both additive and multiplicative interactions between the two exposures. MSDP and HDP were not associated with the risk of ADHD outcomes. There was no evidence of association between alterations in postnatal inflammatory or neurotrophic proteins and development of ADHD.
CONCLUSIONS: Contrary to previous population-based studies, MSDP and HDP did not predict ADHD outcomes in this extremely preterm cohort, but elevated maternal pre-pregnancy BMI and maternal diabetes were associated with increased risk of ADHD at age 10 and/or 15, with positive interaction between these exposures. There was no evidence of mediation of these risks through alterations in neonatal inflammatory or neurotrophic proteins.