The use of HbA1c as an aid in the diagnosis of gestational diabetes mellitus.

The ability of HbA1c was evaluated in the diagnosis of gestational diabetes mellitus (GDM) and for the implications of its use in clinical practice. A total of 339 pregnant women with an estimated gestational age of 24-28 weeks were evaluated for GDM using an oral glucose tolerance test (OGTT) based on International Association of Diabetes and Pregnancy Study Group (IADPSG) criteria. A ROC curve was drawn to determine the sensitivity and specificity of HbA1c in detecting GDM. An HbA1c cut-off value of ≥ 5.2% had sensitivity of 64.15% and specificity of 67.48% in diagnosing GDM, and a positive predictive value of 26.77%. Using two cut-offs in the 'rule-in and rule-out' algorithm, 94.4% of the GDM cases would have been detected but 33% (43 of the 130) women would had been misclassified. HbA1c is not suitable to use alone for the diagnosis of GDM and is not useful in decreasing the need for OGT testing.