Background: The International Association of Diabetes in Pregnancy Study Groups (IADPSG) has recommended diagnostic thresholds1 for gestational diabetes (GDM) based largely on the Hyperglycaemia and Adverse Pregnancy Outcome (HAPO) study, which found a continuous association between maternal glycaemia and adverse outcomes2 . These guidelines have been recommended by ADIPS in Australia, but not yet widely implemented.
Objective: To determine whether patients meeting the IADPSG criteria for GDM (any of fasting ≥5.1mmol/L, 1-hour ≥10.0mmol/L, 2-hour ≥8.5mmol/L), but who were untreated due to not fulfilling the 1991 “old” ADIPS criteria (fasting ≥5.5mmol/L and/or 2-hour ≥8.0mmol/L), have worse obstetric outcomes than patients not meeting either criteria.
Methods: We extracted results of all 75g oral glucose tolerance tests (OGTTs) performed on pregnant women at 24-32 weeks gestation at Mater Pathology (Brisbane, Australia) between 1998 and 2012. Importantly, testing practices varied with time, and results from outside laboratories were not included, therefore the GDM prevalence is not generalisable. OGTTs were classified as meeting the old ADIPS criteria, the IADPSG criteria, both, or neither. This data was linked to demographic and outcome data from the Mater Mothers’ Hospital obstetric and neonatal database (previously described3 ). Bivariate analysis was undertaken to compare demographic and pregnancy characteristics and outcomes. Multivariate analysis further explored the relationship between GDM categories and outcomes, with correction for potential confounders including BMI.
Results: 1737 OGTTs (of 9209 available) were linked to complete records. The frequency of GDM using the old ADIPS criteria and the IADPSG criteria was 12.3% and 21.9%, respectively. Eleven percent of patients met IADPSG but not old ADIPS criteria. On multivariate analysis this latter group showed an increased risk of macrosomia (birthweight > 4000g) (OR 1.63, 95% CI 1.14-2.34), and large for gestational age (LGA) infants (OR 1.77, 95% CI 1.24-2.54). Although the unadjusted OR for neonatal hypoglycaemia was significantly increased, OR 2.60 (95% CI 1.00-6.71), this was attenuated on multivariate analysis (OR 2.62, 95% CI 0.99-6.90).
Conclusion: Offspring of women who met criteria for GDM by the IADPSG but not old ADIPS criteria (and who were therefore untreated for GDM) showed an increased risk of macrosomia and LGA infants.