Correlation of body composition in early pregnancy on gestational diabetes mellitus under different body weights before pregnancy.
Level 3 - non-randomized controlled study
Retrospective cohort study
PubMed 36387861 · doi:10.3389/fendo.2022.916883
What was done
A retrospective cohort study analyzed 6,698 singleton pregnant women without pre-existing diabetes who received regular perinatal care at a single hospital in Zhengzhou, China, in 2021. In the first trimester (<14 weeks), bioelectrical impedance assessment (BIA) was used to measure body composition indicators, including body fat mass, lean mass, and waist-hip ratio. Logarithmic binomial regression and generalized linear models assessed the risk of developing gestational diabetes mellitus (GDM) across pre-pregnancy BMI categories (<24 vs. ≥24 kg/m²), and receiver operating characteristic (ROC) curves evaluated predictive accuracy.
What was found
GDM developed in 1,109 of 6,698 women (10.97% in BMI <24 kg/m² vs. 25.70% in BMI ≥24 kg/m²). All initial BIA metrics were significantly higher in the GDM group (P < 0.05). After fully adjusting for confounders, waist-hip ratio remained positively associated with GDM onset (up to 4.562 times the risk). First-trimester gestational weight gain had the highest predictive accuracy for GDM with an area under the curve (AUC) of 0.795 (cutoff: 1.415 kg), while body fat mass had an AUC of 0.663.
Why it matters
Early pregnancy body composition metrics, particularly waist-hip ratio and first-trimester weight gain, provide non-invasive clinical markers to identify women at higher risk for GDM across diverse pre-pregnancy BMI categories.
Limits
The study is retrospective and limited to a single medical center in China, restricting generalizability. BIA estimates in early pregnancy may be influenced by rapid shifts in maternal hydration, and dietary or lifestyle confounders were not described.
Cited by
- supports Having higher muscle mass during pregnancy is protective against developing gestational diabetes.