RESUMO
BACKGROUND: Glycated Hb (HbA1c) has not been used for the diagnosis of gestational diabetes mellitus (GDM). Measurement of HbA1c levels is less complicated and more comfortable than glucose challenge test (GCT) for pregnant women. We studied HbA1c as a biomarker of GDM and as a screening test to avoid the use of GCT. METHODS: A prospective case-control study involves 745 pregnant women between 24th and 28th gestation week. HbA1c levels were measured and GDM was diagnosed according to Carpenter-Coustan criteria. Mean and SD were calculated for GCT value, HbA1c, age, and body mass index (BMI). A receiver operating characteristic (ROC) curve was plotted to evaluate the diagnostic performance of HbA1c test in diagnosing GDM. Cut-off points were calculated to rule out GDM and sensitivity (Se) and specificity (Sp) were also determined. A study of the implementation of HbA1c cut-offs was performed to avoid the GCT or to perform the confirmatory oral glucose tolerance test (OGTT). RESULTS: The area under the curve (AUC) was 0.67 (0.58-0.76). Using 4.6% HbA1c as a cut-off prevented false negatives but only decreased the number of GCTs performed by 7.2%. However, using 4.7% HbA1c resulted in one false negative (reduction of 15.0%). Finally, by selecting 4.8% HbA1c, we found two false negatives, but there were 25.9% who do not require a GCT. CONCLUSIONS: Adoption of HbA1c as a screening test for GDM may eliminate the need of GCT. Although the HbA1c test does not have sufficient Se and Sp to be used as the only diagnostic test, the use of a rule-out strategy in combination with the OGTT could be useful.