Shobha H Mehta1, Michael Kruger, Robert J Sokol. 1. Department of Obstetrics, Gynecology and Women's Health, Henry Ford Health System, Detroit, MI, USA. smehta3@hfhs.org
Abstract
OBJECTIVE: Our objective is to determine if there is a relationship between diabetes during pregnancy and childhood obesity, in our inner-city, African-American population. METHODS: Pertinent child, neonatal and maternal pregnancy and delivery data were collected from mothers of children age 2-5 years old. Outcome variable definition was based on children's body mass index (BMI) subgroups; independent variable definition on birthweight subgroups based on customized growth percentiles. Covariates included pre and postnatal factors. Those covariates marginally related to diabetes (p < 0.2) by bivariate analyses, were allowed to compete in logistic regression, with p < 0.05 significant. RESULTS: Four hundred and ninety-three patients were enrolled, of which 35 (7.1%) had diabetes during pregnancy. Children of diabetic mothers were more likely to be obese at age 2-5 years than those of non-diabetics (p = 0.004). Five of 20 covariates had p < 0.2 in bivariate setting. Following stepwise logistic regression, diabetes and maternal prepregnancy BMI were significant determinants of childhood obesity. When large-for-gestational age (LGA) was added into the model, diabetes was no longer significant (p = 0.105); only LGA (p = 0.008) and maternal prepregnancy BMI (p = 0.032) were significantly associated with childhood obesity. CONCLUSIONS: In our inner-city, primarily African-American population, diabetes in pregnancy is significantly related to childhood obesity at age 2-5 years. Well-controlled diabetes during pregnancy that avoids macrosomia may lead to prevention of future childhood obesity as well.
OBJECTIVE: Our objective is to determine if there is a relationship between diabetes during pregnancy and childhood obesity, in our inner-city, African-American population. METHODS: Pertinent child, neonatal and maternal pregnancy and delivery data were collected from mothers of children age 2-5 years old. Outcome variable definition was based on children's body mass index (BMI) subgroups; independent variable definition on birthweight subgroups based on customized growth percentiles. Covariates included pre and postnatal factors. Those covariates marginally related to diabetes (p < 0.2) by bivariate analyses, were allowed to compete in logistic regression, with p < 0.05 significant. RESULTS: Four hundred and ninety-three patients were enrolled, of which 35 (7.1%) had diabetes during pregnancy. Children of diabetic mothers were more likely to be obese at age 2-5 years than those of non-diabetics (p = 0.004). Five of 20 covariates had p < 0.2 in bivariate setting. Following stepwise logistic regression, diabetes and maternal prepregnancy BMI were significant determinants of childhood obesity. When large-for-gestational age (LGA) was added into the model, diabetes was no longer significant (p = 0.105); only LGA (p = 0.008) and maternal prepregnancy BMI (p = 0.032) were significantly associated with childhood obesity. CONCLUSIONS: In our inner-city, primarily African-American population, diabetes in pregnancy is significantly related to childhood obesity at age 2-5 years. Well-controlled diabetes during pregnancy that avoids macrosomia may lead to prevention of future childhood obesity as well.
Authors: Shreyas V Kumbhare; Dhrati V V Patangia; Ravindra H Patil; Yogesh S Shouche; Nitinkumar P Patil Journal: J Biosci Date: 2019-06 Impact factor: 1.826