[Association between abnormal oral glucose tolerance test patterns in the second trimester and large for gestational age newborns].

To investigate the impact of abnormal patterns of 75 g oral glucose tolerance test (OGTT) in the second trimester on the risk of large for gestational age (LGA) newborn deliveries. General clinical data and OGTT results of 66 290 pregnant women who received regular prenatal care and delivered in Guangdong Maternal and Child Health Hospital from December 24, 2016 to July 26, 2022 were collected. According to the results of OGTT, the pregnant women were divided into 8 groups: normal blood glucose group (normal fasting blood glucose, 1-hour and 2-hour after oral glucose, 54 518 cases), gestational diabetes mellitus (GDM) 0 group (only abnormal fasting blood glucose, 1 430 cases), GDM 1 group (only abnormal blood glucose at 1-hour after oral glucose, 2 150 cases), GDM 2 group (only abnormal blood glucose at 2-hour after oral glucose, 3 736 cases), GDM 0+1 group (both fasting blood glucose and 1-hour after oral glucose were abnormal, 371 cases), GDM 0+2 group (both fasting blood glucose and 2-hour after oral glucose were abnormal, 280 cases), GDM 1+2 group (abnormal blood glucose at 1-hour and 2-hour after oral glucose, 2 981 cases) and GDM 0+1+2 group (abnormal fasting blood glucose, 1-hour and 2-hour after oral glucose, 824 cases). Multivariate logistic regression was used to analyze the effects of different abnormal OGTT patterns on LGA. In addition, the blood glucose measurements at the three time points of OGTT were combined and used as continuous variables in the receiver operating characteristic (ROC) curve to evaluate the predictive value of each blood glucose measurement mode for LGA and the area under the curve (AUC) was compared. (1) Multivariate logistic regression analysis showed that the risks of LGA were significantly increased in GDM 0 group (=1.76, 95%: 1.50-2.08; <0.001), GDM 0+1 group (=2.29, 95%: 1.72-3.04; <0.001), and GDM 0+1+2 group (=1.98, 95%: 1.61-2.43; <0.001). (2) ROC curve analysis showed that fasting blood glucose, 1-hour after oral glucose, 2-hour after oral glucose, fasting+1-hour after oral glucose, fasting+2-hour after oral glucose, 1-hour+2-hour after oral glucose, and fasting+1-hour+2-hour after oral glucose had certain predictive value for LGA (all <0.001). The AUC of fasting blood glucose measurement was higher than that of 2-hour blood glucose measurement in predicting LGA, and the difference was statistically significant (<0.05). There was no significant difference in the AUC between fasting blood glucose and other blood glucose measurement modes for predicting LGA (all >0.05). In the abnormal OGTT patterns, pregnant women with abnormal fasting blood glucose, abnormal fasting+1-hour after oral glucose, and abnormal fasting+1-hour+2-hour after oral glucose have an increased risk of LGA. Fasting blood glucose measurement is of great significance for the prediction of LGA, and could be used as an optimal indicator to evaluate the risk of LGA in clinical practice.