The recommendation of universal screening for gestational diabetes (GDM) by oral glucose tolerance test (OGTT) in pregnancy may pose a resource challenge for healthcare systems. In light of growing recognition that women who develop GDM have chronic metabolic dysfunction that predates their pregnancy, we hypothesized that pre-gravid measurement of A1c and glucose may provide the capacity to rule out GDM in low-risk women and thereby reduce the overall OGTT burden in pregnancy. To test this hypothesis, we identified all women in Ontario, Canada, with singleton live-birth pregnancies between January 2008 and December 2015 (n=1,001,006, of which 47,180 had GDM), and created derivation and validation cohorts (each n=500,503). In the derivation cohort, pre-gravid A1c (OR=9.72, 95%CI 8.76-10.79), fasting glucose (OR=3.32, 95%CI 3.13-3.51) and random glucose (OR=1.70, 95%CI 1.65-1.75) each predicted subsequent GDM. Area-under-receiver-operating-characteristic curve (AROC) was higher for A1c (0.684) than for fasting glucose (0.647) or random glucose (0.625). Test characteristics revealed that the negative predictive value (NPV) for ruling out GDM was optimized (NPV=98.7%) at pre-gravid A1c ≤4.3%. However, when evaluated in the validation cohort, this pre-gravid A1c threshold would only reduce the need for antepartum OGTT in 0.11% of women. Conversely, if test sensitivity for predicting GDM were set at 95.1% in the derivation cohort, the resultant threshold of pre-gravid A1c ≤5.0% would reduce the need for antepartum OGTT in 7.2% of women but at the cost of missing 2.5% of GDM diagnoses in the validation cohort.
In conclusion, pre-gravid A1c and glucose levels can predict a woman’s future risk of GDM. However, their test characteristics are clinically not acceptable for reducing the need for GDM screening by OGTT in pregnancy.
R. Retnakaran: Consultant; Self; Eli Lilly and Company, Novo Nordisk Inc., Sanofi. Research Support; Self; Boehringer Ingelheim International GmbH, Novo Nordisk Inc. B.R. Shah: None.