ORLANDO – Longer-term data are providing more evidence of a possible link between maternal diabetes and autism spectrum disorder in their children.
Anny Xiang, PhD, and coathors with Kaiser Permanente of Southern California sought to further understand the possible effect of maternal T1D on offspring’s development of autism spectrum disorder (ASD) by expanding the cohort and timeline of their earlier work (JAMA. 2015;313(14):1425-1434).
A total of 621 children were exposed in utero to T1D , 9,453 to T2D, 11,922 to gestational diabetes diagnosed by 26 weeks, and 24,505 to gestational diabetes diagnosed after 26 weeks.
Across the cohort, 1.3% of children were diagnosed with autism spectrum disorder (ASD). The rate was barely different, at 1.5%, for those whose mothers developed gestational diabetes after 26 weeks. But rates of ASD were higher – 3.1%, 2.5%, 2.1% – among those whose mothers had T1D, T2D, and gestational diabetes that developed at 26 weeks or earlier, respectively. The findings were adjusted for co-founders such as birth year, age at delivery, eduction level and income, Dr. Xiang said at the annual scientific sessions of the American Diabetes Association.
Compared to offspring of mothers without diabetes, ASD was more common in the children of mothers with T1D (adjusted HR=2.36, 95% CI, 1.36-4.12) mothers with type 2 diabetes (AHR= 1.45, 95% CI, 1.24-1.70) and gestational diabetes mellitus that developed by 26 weeks gestation (1.30, 95% CI, 1.12-1.51).
The numbers remained similar after they were adjusted for smoking during pregnancy and prepregnancy BMI, statistics which were available for about 36% of the subjects, according to the findings which were published simultaneously in JAMA (June 23, 2018. doi:10.1001/jama.2018.7614).
Possible explanations for the link between ASD and maternal diabetes include maternal glycemic control, prematurity, and levels of neonatal hypoglycemia, Dr. Xiang said.
The results do not take into account any paternal risks for offspring developing ASD, which also includes diabetes, Dr. Xiang said, noting that two previous studies linked diabetes in fathers to ASD, although to a lesser extent than diabetes in mothers. (Epidemiology. 2010 Nov;21(6):805-8; Pediatrics. 2009 Aug;124(2):687-94)
The study also doesn’t take breastfeeding into account, Dr. Xiang noted. A 2016 study found that women with T2D were less likely to breastfeed (J Matern Fetal Neonatal Med. 2016;29(15):2513-8), and some research has suggested that breastfeeding may be protective against the development of ASD in children (Nutrition 2012;28(7-8):e27-32).
In addition, the study doesn’t track maternal glucose levels over time.
Session co-chair Peter Damm, MD, professor of obstetrics at the University of Copenhagen, said in an interview that he is impressed by the study. He cautioned, however, that it does not prove a connection.“This not a proof, but it seems likely, or like a possibility,” he said.
One possible explanation for a diabetes/ASD connection is the fact that the fetal brain is evolving throughout pregnancy unlike other body organs, which simply grow after developing in the first trimester, he said. As a result, glucose levels may affect the brain’s development in a unique way compared to other organs.