LAUSR

OBJECTIVES Predictors of gestational diabetes mellitus (GDM) recurrence (GDMR) was determined in southern Chinese women. MATERIAL AND METHODS A total of 366 women with GDM who had two consecutive singleton deliveries at our hospital between January 2014 and October 2020 were enrolled in the current study. We retrospectively compared the clinical characteristics, fasting plasma glucose level (FPG-1), and oral glucose tolerance test-1h-1 and -2h-1 (OGTT 1hr-1: 1-h post-load glucose level during the first pregnancy and OGTT 2hr-1: 2-h post-load glucose level during the first pregnancy) for the first pregnancy between patients in the GDMR group (n = 166) and the non-GDMR group (n = 210). RESULTS The incidence of GDMR in the study population was 44.15%. During the first pregnancy, women in the GDMR group had significantly higher OGTT 1h-1, OGTT 2h-1, and FPG-1 + OGTT 1h + 2h-1 compared to the non-GDMR group. When the threshold of the FPG-1 + OGTT 1h + 2h-1 level in the first pregnancy was > 23.6 mmol/L, the specificity for predicting GDMR was 0.85, the sensitivity was 0.45, and the area under the receiver operating characteristic curve (ROC-AUC) was 0.70, indicating a 70% probability of predicting GDMR in the next pregnancy. Logistic regression analysis showed that patients with a combined abnormal FPG-1 + OGTT 1h + 2 h-1 level had a 10-fold increased risk for GDMR in subsequent pregnancies than patients with normal indicators (OR: 10.542, 95% CI: 3.097-35.881; p < 0.0001). CONCLUSIONS The OGTT 1h-1 and OGTT 2h-1 are independent risk factors for GDMR in southern Chinese women. Women with an FPG-1 + OGTT 1h + 2h-1 threshold level > 23.6 mmol/L in the first pregnancy had a 10-fold greater probability of developing GDMR in the second pregnancy than women in the non-GDMR group.