Objective:To study the risk factors for abnormal glucose metabolism in pregnant women with a history of gestational diabetes mellitus (GDM).Methods:A retrospective analysis was performed on pregnant women who had two consecutive deliveries and were was complicated by GDM in the previous pregnancy at the First Affiliated Hospital of Sun Yat-sen University from January 2011 to May 2019. Clinical data of both pregnancies were collected, including general information, fasting blood glucose in early pregnancy and 75 g oral glucose tolerance test (OGTT) results, glycosylated hemoglobin A1c and blood lipid profile at 24-28 gestational weeks. The incidence and risk factors of abnormal glucose metabolism in these cases during the present pregnancy were analyzed. Analysis of variance, Kruskal-Wallis test, SNK- q or LSD- t-test, and Chi-square test were used for data analysis. Single-factor logistic regression analysis was used to analyze the high-risk factors, and multifactor logistic regression analysis was performed to fit the model. Variable collinearity diagnosis was performed using the coldiag2 command. Results:(1) A total of 455 cases were enrolled in the study. According to the fasting glucose level in the first trimester and the OGTT results in the present pregnancy, they were divided into three groups: normal OGTT group ( n=240), GDM group ( n=189), and pre-gestational diabetes mellitus group (PGDM, n=26). The incidence of abnormal glucose metabolism in these patients during the present pregnancy was 47.2% (215/455). (2) Those with a history of GDM had higher pre-pregnancy weight, lower weight gain, higher cesarean section rate, smaller gestational age at delivery, and higher neonatal birth weight in the present pregnancy than those in the previous pregnancy [(55.6±8.5) vs (53.3±7.9) kg, t=-4.059; (11.2±4.2) vs (12.5±4.4) kg, t=4.435; 47.9% (218/455) vs 33.0% (150/455), χ2=20.481; (38.6±1.3) vs (38.8±1.3) weeks, t=2.288; (3 177±463) and (3 114±460) g, t=-2.044; all P<0.05]. (3) In the PGDM group, the 2-h plasma glucose level after 75 g OGTT was higher than that in the previous pregnancy [(11.4±1.1) vs (9.9±1.7) mmol/L, t=-3.299, P=0.002]. (4) In the present pregnancy, the PGDM group had the highest fasting blood glucose in early pregnancy, followed by the GDM group and the normal OGTT group [4.6 mmol/L (4.2-7.6 mmol/L), 4.3 mmol/L (4.0-4.6 mmol/L) and 4.1 mmol/L (3.8-4.4 mmol/L), χ2=34.498, P<0.001]. The PGDM group had the least postpartum weight retention, followed by the normal OGTT group and the GDM group [(1.2±3.9), (1.6±3.9), and (2.6±4.9) kg, F=3.086, P<0.05]. (5) Multivariate logistic regression analysis showed postpartum weight retention and the 1-h and 2-h plasma glucose levels after 75 g OGTT in the previous pregnancy were independent risk factors for abnormal glucose metabolism in pregnant women with a history of GDM (postpartum weight retention: OR=1.054, 95% CI: 1.005-1.106; 1-h plasma glucose: OR=1.284, 95% CI: 1.087-1.516; 2-h plasma glucose: OR=1.272, 95% CI: 1.071-1.511). Conclusions:The incidence of abnormal glucose metabolism is higher in subsequent pregnancy in women with GDM history, which may be related to various factors, such as postpartum weight retention and plasma glucose after 75 g OGTT in the previous pregnancy.