We prospectively studied 262 women with prior gestational diabetes mellitus (GDM) and 66 control women to compare their prevalence of metabolic syndrome and its relationship with insulin secretion and sensitivity. A 75-g oral glucose tolerance test was scheduled 5 years after delivery along with lipid profile, anthropometrics, and blood pressure measurement. Metabolic syndrome was defined according to the National Cholesterol Education Program 2001, and insulin sensitivity and secretion were estimated with the homeostasis model assessment. Women with prior GDM had similar insulin sensitivity and lower insulin secretion than control women. In comparison with control women, women with prior GDM had higher blood pressure, waist circumference, very low-density lipoprotein cholesterol, and oral glucose tolerance test blood glucose values but, with the exception of fasting hyperglycemia, did not have an increased prevalence of metabolic syndrome or its components. The multivariate prediction of metabolic syndrome and its components was similar with age and current homeostasis model assessment-insulin secretion and resistance indexes or with age, obesity, and GDM. The main predictor was current insulin resistance in the first case and obesity in the second, obesity being the best predictor overall. We conclude that in our population and at midterm follow-up, women with prior GDM have a decreased insulin secretion and display a higher prevalence of fasting hyperglycemia but not the full-blown picture of metabolic syndrome. Obesity, a surrogate index of insulin resistance, is the best predictor of metabolic syndrome at follow-up.