Abstract
Objective In this study, we aimed to evaluate the relationship between body fat percentage (BFP) and the risk of gestational diabetes mellitus (GDM). Methods We conducted a cohort study involving 336 singleton pregnant women attending an antenatal care clinic before 14 weeks of gestation. Body composition was measured during their first antenatal visit by using a multi-frequency segmental body composition analyzer. GDM was diagnosed by a 50-g glucose challenge test (GCT) and a 100-g oral glucose tolerance test (OGTT) during the first visit and repeated during 24-28 weeks of gestation. Rates of GDM were compared between women with BFP ≥30% and those with BFP <30%. The ability of BFP and body mass index (BMI) to diagnose GDM was assessed, as well as their correlation. Results Of the 296 women included in the analysis, 171 had BFP ≥30%, and 125 had BFP <30%. The prevalence of GDM was 17.9%. BFP correlated well with BMI (correlation coefficient: 0.956, p<0.001). BFP ≥30% and BMI ≥25 kg/m(2) significantly increased the risk of GDM (22.2% vs. 12%, p=0.023 and 26.4% vs. 14.4%, p=0.014, respectively). The sensitivity of BFP ≥30% and BMI ≥25 kg/m(2) for GDM diagnosis was 71.1% and 43.3%, respectively while the specificity was 45.3% and 73.7%, respectively. Both BFP and BMI had comparable efficacy in diagnosing GDM [areas under the receiver operating characteristic (ROC) curves (AUC) of 0.634 and 0.642, respectively]. Conclusions BFP ≥30% and BMI ≥25 kg/m(2) significantly increased the risk of GDM. BFP correlated well with BMI and had similar efficacy in diagnosing GDM.