BackgroundGestational diabetes mellitus (GDM) is a risk factor for the development of diabetes mellitus (DM). However, there is a low return rate for this screening, so it is important to search for earlier methods for evaluation after delivery, to increase the number of pregnant women screened, so you can start the treatment or prevention of these early comorbidities. To determine the accuracy of the 75 g 2-h oral glucose tolerance test (OGTT) performed between 48–72 h after delivery for the diagnosis of DM using the OGTT after 6 weeks as the gold standard criterion, and to identify the optimal cutoff points for this exam for diabetes screening after a pregnancy complicated by GDM.Methods82 women with previous GDM underwent an OGTT between 48–72 h postpartum and repeated the test 6 weeks after delivery.ResultsThe prevalence of DM and prediabetes based on the first OGTT was 3.7 and 32.9 %, respectively, and 8.5 and 20.7 %, respectively, at the second OGTT. For those with DM, the area under the curve (AUC) based on the fasting plasma glucose (FPG) was 0.77 (95 % CI 0.61–0.92), and based on 2-h OGTT was 0.82 (95 % CI 0.66–0.97). For patients with prediabetes, the AUC based on the FPG was 0.73 (95 % CI 0.59–0.86) and based on the 2-h OGTT was 0.74 (95 % CI 0.61–0.87). Using a FPG cutoff value of 78 mg/dl (4.3 mmol/L) and a 2-h OGTT cutoff value of 130 mg/dl (7.2 mmol/L) for DM, the specificity was 58.7 and 60 %, the sensitivity was 71.4 and 85.7 %, the positive predictive value was 13.9 and 16.7 and the negative predictive value was 95.7 and 97.9 %, respectively.ConclusionsOGTT performed early in postpartum is a useful tool for identifying women with previous GDM who must perform an OGTT 6 weeks after delivery.