Gestational diabetes: An alternative, patient-friendly approach for using the diagnostic 100-g OGTT in high-risk populations

Objective: To simplify the 100-g oral glucose tolerance test (OGTT) for making a diagnosis of gestational diabetes mellitus (GDM). Methods: During a 6-year period, 6,801 pregnant women underwent the 3-h, 100-g OGTT in a high-risk, multiethnic population. All these women had either a positive glucose challenge test/ clinical history or were part of a universal screening protocol. The area (AUC) under the receiver-operating characteristic curve was used to compare the performance of the fasting plasma glucose (FPG), 1-h, 2-h and 3-h glucose values for GDM diagnosis. Ideal cut-offs of the best performing glucose value alone and in combination were used to find the best predictor for GDM with minimal loss of sensitivity and specificity. Results: Using the American Diabetes Association criteria, 2,127 (31.3%) women were classified as having GDM. The 2-h value was the best performing (AUC 0.933) glucose value; at an optimal cut-off of ≥8.6 mmol/1, the sensitivity was 83.6% with a specificity of 92.8%. It did not achieve sufficient collective specificity and sensitivity at various thresholds to be useful. However, in combination with FPG (either FPG ≥5.3 mmol/1 or 2-h ≥8.6 mmol/1) or (either FPG ≥5.6 mmol/1 or 2-h ≥8.6 mmol/1), the sensitivity/specificity were 98.5%/ 84.7% and 92.5%/89.3%, respectively. Conclusion: An abbreviated 100-g OGTT, using only the FPG and 2-h glucose value, may be a useful alternative in high-risk populations to make a diagnosis of GDM by being cost-effective and patient-friendly. © Springer-Verlag 2005.