Background: To determine the value of repeated random blood glucose (R-B-glucose) measurements alone or in combination with traditional risk factors [family history of diabetes, obesity, prior large-for-gestational-age (LGA) infant or prior gestational diabetes mellitus (GDM)] to predict the outcome of the oral glucose tolerance test (OGTT).
Methods: A prospective population-based study was undertaken in a Swedish county. All pregnant nondiabetic women (n = 4918) visiting the maternal health care clinics over a 2-year period were offered a 75-g OGTT in gestational weeks 28-32. Traditional risk factors and values of repeated R-B-glucose measurements were registered, as well as the results of the OGTT, in terms of fasting B-glucose and 2-h B-glucose.
Results: A total of 3616 women (73.5%) had an OGTT. Of these, 1.7% had GDM, 1.3% impaired glucose tolerance (IGT) and 0.4% diabetes mellitus (DM). An R-B-glucose cut-off level > or =8.0 mmol/L as the only indicator for an OGTT was optimal for detecting GDM with regard to sensitivity (47.5%) and specificity (97.0%). It has the same sensitivity for detecting GDM as using traditional risk factors, but reduces the need to carry out the OGTT from 15.8% to 3.8% of the population. Combined with prior LGA infant or prior GDM as indications for the OGTT in the present study, all women with DM and 44.7% of those with IGT will be identified. Only 7.3% of the population will have to take the OGTT.
Conclusion: A random B-glucose level > or = 8.0 mmol/L prior LGA infant or prior GDM as an indicator for taking the OGTT is a simple and effective first step in a two-step screening model for GDM.