Diagnostic Accuracy of Body Mass Index and Fasting Glucose for The Prediction of Gestational Diabetes Mellitus after Assisted Reproductive Technology

BACKGROUND: The aim of the present study was to determine the maternal pre-pregnancy body mass index (BMI), first-trimester fasting blood sugar (FBS), and the combination of (BMI+FBS) cut-points for at-risk pregnant women conceived by assisted reproductive technology (ART) to better predict the risk of developing gestational diabetes mel- litus (GDM) in infertile women. MATERIALS AND METHODS: In this nested case-control study, 270 singleton pregnant women consisted of 135 (GDM) and 135 (non-GDM) who conceived using ART were assessed. The diagnosis of GDM was confirmed by a one-step glucose tolerance test (O-GTT) using 75 g oral glucose. BMI was classified base on World Health Organization (WHO) criteria. The relationship between BMI, FBS, and BMI+FBS with the risk of GDM development was deter- mined by logistic regression and adjusted for confounding factors. Receiver operating characteristic (ROC) curve analysis was performed to assess the value of BMI, FBS, and BMI+FBS for the prediction of GDM. RESULTS: The GDM group had significantly higher age, BMI, family history of diabetes, and history of polycystic ovary syn- drome in comparison with the non-GDM group (P<0.05). Overweight and obese women had 3.27, and 5.14 folds increase in the odds of developing GDM, respectively. There was a 17% increase in the risk of developing GDM with each 1 mg/dl increase in fasting glucose level. The cut points for FBS 84.5 mg/dl (72.9% sensitivity, 74.4% specificity), BMI 25.4 kg/m(2) (68.9% sensitivity, 62.8% specificity), and BMI+FBS 111.2 (70.7% sensitivity, 80.6% specificity) was determined. CONCLUSION: The early screening and high-quality prenatal care should be recommended upon the co-occurrence of high FBS (≥84.5 mg/dl) in the first-trimester of the pregnancy and the BMI (≥25.4 kg/m(2)) in pre-pregnancy period in women undergone ART. The combination of BMI and FBS is considered a better prediction value.