The objective of the present study was to determine if any associations between reproductive experience and anthropometric or sub-clinical metabolic alterations of glucose metabolism exist. Sixty-seven women were recruited from the University of Brasilia Hospital and were evaluated at 12–18 months postpartum. Demographic, socio-economic, physical activity, anthropometric and health history (biochemical, reproductive) data were obtained. After a 12h overnight fast, a 2h oral glucose tolerance test was performed. Blood samples were collected at several points: at baseline, after intake of d-glucose solution (750g/l; 100ml) and every 30min thereafter. Blood glucose and lipids were measured by enzymic assays. Blood insulin was measured by RIA. In multiple regression analysis four dependent logarithmically transformed (logt) variables (increased area under the glucose curve (IAUGC), increased area under the insulin curve (IAUIC), insulin peak (IP), homeostasis model of assessment (HOMA)) were adjusted for parity, age, lactation index, BMI, percentage body fat (PBF), waist circumference, superior skinfold thickness sum:inferior skinfold thickness sum ratio and oral contraceptive use. PBF was positively associated with logt-IAUIC (P=0·004) and IP (P=0·006). However, the lactation index was negatively associated with logt-IAUIC (P=0·02) IAUGC and HOMA did not present significant associations. We conclude that during the postnatal period, independent of parity, body adiposity accumulation is associated with initial alterations in insulin secretion. Furthermore, independent of body adiposity, breast-feeding has a long-lasting protective effect on insulin response.