Abstract: Objective To observe the efficacy and safety of insulin in the treatment of gestational diabetes in different gestational age. Methods The data of 84 women with gestational diabetes were retrospectively collected, and divided into early treatment group (n=40) and late treatment group (n=44) based on first use of insulin in different gestational age. All patients received controlled diet (calorie intake no more than 35 kJ·kg^-1·d^-1, carbohydrate, fat and protein were distributed by 11:4:5). Insulin injection started with 4~6 units before meals and at bedtime by 4:2:3:1 depending on the levels of fast blood glucose and postprandial 2 h blood glucose, patients were divided into early treatment group (first injection before pregnant 28 week) and late treatment group (first injection group after pregnant 28 week). The data of fast blood glucose and postprandial 2 h blood glucose at the first and delivery, pregnancy outcomes and neonatal conditions were collected of 2 groups. Results During delivery, the fast blood glucose and postprandial 2 h blood glucose levels in the early treatment group were significantly lower than those in the late treatment group (P<0.05). In the early treatment group, the improvement of postprandial 2 h blood glucose level vs. fasting was significantly lower than that in the late treatment group (P<0.05). The number of cesarean section and fetal distress number between the two groups were not significantly different. The number of macrosomia of early treatment group was significantly less than the late treatment group, and the 5 min neonatal Apgar score of early treatment was significantly higher than late treatment group (P<0.05). Conclusion Treatment with insulin before pregnant 28 weeks can effectively control the blood glucose, improve the adverse uterus environment and pregnancy outcomes with safety.

Key words: gestational diabetes mellitus, giant baby, insulin