Objective: Our aim was identify pregnant women with impaired gestational glucose tolerance (IGGT) at risk for more severe gluco-metabolic alterations who could require subsequent insulin therapy during pregnancy. Methods: We studied 78 pregnant women with IGGT after a 100 g glucose oral tolerance test. Patients were divided into two groups based on a cut-off value of a 15% reduction from normal values of the glucose load. Sixty-three patients had at least one value above this cutoff point, while 15 had all residual values below the cut off. All patients were put on a diet and glycemia reassessed: those who showed pre-prandial blood glucose higher than or equal to 90 mg/dL and/or 2-h post-prandial higher than or equal to 120 mg/dL underwent insulin theraoy. We matched the presence of at least one residual value in the oral glucose tolerance test above the limit used with the subsequent need for insulin treatment. Results: The presence of at least one residual value above the “-l5%” cutoff in the glucose tolerance test was associated with high risk (positive predictive value 79%), whereas normality of the residual values indicated low risk (negative predictive value 80%), of insulin need during the rest of pregnancy. Conclusions: Pregnant women with IGGT definitely do not have a normal metabolic condition, sometimes even requiring diet and insulin treatment. From our results, the need for more accurate monitoring and insulin treatment may be predicted by simply looking at the residual values in the glucose tolerance test.