Associations of mid-pregnancy HbA1c with gestational diabetes and risk of adverse pregnancy outcomes in high-risk Taiwanese women

Background The objective of this study was to investigate the associations among the mid-pregnancy glycated hemoglobin A1c (HbA1c) level, gestational diabetes (GDM), and risk of adverse pregnancy outcomes in women without overt diabetes and with positive 50-g, 1-h glucose challenge test (GCT) results (140 mg/dL or greater). Methods This prospective study enrolled 1,989 pregnant Taiwanese women. A two-step approach, including a 50-g, 1-h GCT and 100-g, 3-h oral glucose tolerance test (OGTT), was employed for the diagnosis of GDM at weeks 23-32. The mid-pregnancy HbA1c level was measured at the time the OGTT was performed. A receiver operating characteristic (ROC) curve was used to determine the relationship between the mid-pregnancy HbA1c level and GDM. Multiple logistic regression models were implemented to assess the relationships between the mid-pregnancy HbA1c level and adverse pregnancy outcomes. Results An ROC curve demonstrated that the optimal mid-pregnancy HbA1c cut-off point to predict GDM, as diagnosed by the Carpenter-Coustan criteria using a two-step approach, was 5.7%. The area under the ROC curve of the mid-pregnancy HbA1c level for GDM was 0.70. Compared with the levels of 4.5-4.9%, higher mid-pregnancy HbA1c levels (5.0-5.4, 5.5-5.9, 6.0-6.4, 6.5-6.9, and > 7.0%) were significantly associated with increased risks of gestational hypertension or preeclampsia, preterm delivery, admission to the neonatal intensive care unit, low birth weight, and macrosomia (the odds ratio [OR] ranges were 1.20-9.98, 1.31-5.16, 0.88-3.15, 0.89-4.10, and 2.22-27.86, respectively). Conclusions The mid-pregnancy HbA1c level was associated with various adverse pregnancy outcomes in high-risk Taiwanese women. However, it lacked adequate sensitivity and specificity to replace the two-step approach in the diagnosis of GDM. The current study comprised a single-center prospective study; thus, additional, randomized control design studies are required.