The use of longitudinal hemoglobin A1c values to predict adverse obstetric and neonatal outcomes in pregnancies complicated by pregestational diabetes.

BACKGROUND: Although an elevated early pregnancy hemoglobin A1c has been associated with both spontaneous abortion and congenital anomalies, it is unclear whether A1c assessment is of value beyond the first trimester in pregnancies complicated by pregestational diabetes. OBJECTIVE: We sought to investigate the prognostic ability of longitudinal A1c assessment to predict obstetric and neonatal adverse outcomes based on degree of glycemic control in early and late pregnancy. MATERIALS AND METHODS: This was a retrospective cohort study of all pregnancies complicated by pregestational diabetes from January 2012 to December 2016 at The Ohio State University Wexner Medical Center with both an early A1c (<20 weeks' gestation) and late A1c (>26 weeks' gestation) available for analysis. Patients were categorized by good (early and late A1c <6.5%), improved (early A1c >6.5% and late A1c <6.5%) and poor (late A1c >6.5%) glycemic control. A multivariate regression model was used to calculate adjusted odds ratios (aOR) for each identified obstetric and neonatal outcome, controlling for maternal age, body mass index, race/ethnicity, type of diabetes, and gestational age at delivery compared to good control as the referent group. RESULTS: A total of 341 patients met inclusion criteria during the study period. The median A1c values improved from early to late gestation in the good (5.7% [interquartile range [IQR], 5.4-6.1%] versus 5.4%; [IQR 5.2-5.7%]), improved (7.5% [IQR, 6.7-8.5] versus 5.9% [IQR, 5.6-6.1%]) and poor (8.3% [IQR, 7.1-9.6%] versus 7.3% [IQR, 6.8-7.9%]) glycemic control groups. There were no statistically significant differences in the rate of adverse outcomes between the good and improved groups except for an increased rate of neonatal intensive care unit admissions in the improved group (aOR, 3.7; confidence interval [CI], 1.9-7.3). In contrast, the poor control group had an increased rate of shoulder dystocia (aOR, 6.8; CI, 1.4-34.0), preterm delivery (aOR, 3.9; CI, 2.1-7.3), neonatal intensive care unit admission (aOR, 2.8; CI, 1.4-5.3), respiratory distress syndrome (aOR, 3.0; CI, 1.1-8.0), hypoglycemia (aOR, 3.2; CI, 1.5-6.9), large for gestational age weight at birth (aOR, 2.7; CI, 1.5-4.9), neonatal length of stay >4 days (aOR, 3.1; CI, 1.6-6.0) and preeclampsia (aOR, 2.4; CI, 1.2-4.6). There were no differences in rates of cesarean delivery, umbilical artery pH <7.1, or Apgar score <7 at 5 minutes after regression analysis. CONCLUSION: Antenatal hemoglobin A1c values are useful for objective risk stratification of patients with pregestational diabetes. Strict glycemic control throughout pregnancy with a late pregnancy A1c target of <6.5% leads to reduced rates of obstetric and neonatal adverse outcomes independent of early pregnancy glucose control.