Limited studies have assessed longitudinal growth among children exposed to different types of diabetes in-utero. We assessed BMI from birth to age 10 years in offspring of mothers who had pre-existing type 1 (T1D) or type 2 (T2D), gestational diabetes mellitus (GDM) or no diabetes during pregnancy. GDM was further divided into GDM managed with (GDM with meds) or without (GDM no meds) diabetes medication. More than 3.5 million BMI measurements were obtained from medical records of 218,227 singleton children born at 28-44 weeks gestation from 2008 to 2015. Relationships between BMI, age and diabetes exposure were modelled using random effects mixed models with quadratic B splines adjusted for maternal age, race/ethnicity, education, household income, pre-pregnancy BMI, gestational weight gain, maternal comorbidity, and child’s sex. At birth, the adjusted BMI had small differences among groups except for T1D which had the highest BMI [diff=0.4; 95%CI 0.2-0.6 kg/m2 relative to no diabetes]. After birth, the adjusted BMIs were comparable between T1D and T2D exposed. At about 48 months, the growth pattern started separating with highest BMI in T1D and T2D groups, followed by GDM with meds, GDM no meds and no diabetes groups (Figure). Thus, despite standard diabetes management, there exists a hierarchical BMI pattern in offspring exposed to different types of diabetes even after adjusting for important covariates, starting as early as age 4.


M.A. Sidell: None. T. Chow: None. M.P. Martinez: None. C. Koebnick: None. K.A. Page: None. T.A. Buchanan: None. A. Xiang: None.

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at