Objective To analyze the effect of gestational diabetes mellitus (GDM) on maternal and neonatal outcomes in twin pregnancies. Methods A retrospective cohort study was conducted on 772 cases of dichorionic twin pregnancies delivered in Beijing Haidian Maternal and Child Health Hospital from January 2013 to October 2019. The general clinical characteristics and the maternal and neonatal outcomes were collected, including age, parity, gestational weeks, mode of conception, delivery mode and the maternal and neonatal complications. According to the results of 75 g oral glucose tolerance test (OGTT) on gestational age 24-28 weeks, the subjects were divided into GDM group (157 cases) and normal glucose tolerance (NGT) group (615 cases). The glycated hemoglobin A1c (HbA1c) was tested among 137 GDM women. The GDM women were classified into two groups: the uncontrolled GDM group with HbA1c≥5.5% (25 cases) and well-controlled GDM group with HbA1c<5.5% (112 cases). The t test, rank sum test, χ2 test and Fisher exact probability method were used for comparison between groups. Results The age of GDM group was statistically significant older than that of NGT group (P<0.001). There were more women aged over 35 years in GDM group than in NGT group [28.0% (44/157) vs. 17.1% (105/615), P=0.002]. Women with GDM were significantly more obese before pregnancy than the NGT women (P=0.002). Whereas there was no significant difference in gestational weeks, cesarean section rate, premature birth rate between the two groups. There was no difference between GDM and NGT groups in the incidence of fetal growth restriction, hypertensive disorders of pregnancy, premature rupture of membranes, placental abruption, postpartum hemorrhage, neonatal low birthweight, neonatal respiratory distress, neonatal malformation or dysplasia neonatal death or admission to neonatal ward (P> 0.05). Compared with the NGT group, the well-controlled GDM group and the uncontrolled GDM group, there was no significant difference in gestational weeks, cesarean section rate, premature birth rate, and incidence of fetal growth restriction, hypertensive disorders of pregnancy, premature rupture of membranes, placental abruption, postpartum hemorrhage (P>0.05). However, the incidence of premature birth, fetal growth restriction and premature rupture of membranes increased in the uncontrolled GDM group, without statistical significance (P>0.05), and no significant difference in the neonatal outcomes among them (P>0.05). Conclusion In this study, GDM did not increase the maternal and neonatal adverse outcomes in women with twin pregnancies. [ABSTRACT FROM AUTHOR]