The relationship between late-onset gestational diabetes mellitus [GDM] and oxidative stress is not well known, and the importance of the oxidant/antioxidant equilibrium in the clinical evolution and its complications require elucidation. The aim of the study was to evaluate the relationships between maternal levels of markers of oxidative stress in women with late-onset GDM that, potentially, may have considerable clinical implications in the pathogenesis and/or the evolution of GDM. Pregnant women (n = 78; 53 with GDM, 25 controls), between the 24th and 29th week of gestation, were enrolled. Both groups were analysed for demographic data, perinatal and obstetrics outcomes together with the levels of the marker's oxidative stress and antioxidant status. Control versus patient results in the univariate analysis were the following: pre-gestational body mass index [BMI] 23.31 ± 4.2 vs. 27.13 ± 4.6 kg/m(2) (P = 0.001); weeks at delivery 39.2 ± 3.05 vs. 38.9 ± 1.8 (P = 0.09); Caesarean delivery 12.5 vs. 43% (P = 0.004); macrosomia 4 vs. 9.4% (P = 0.6); lipoperoxides [LPO] 2.06 ± 1.00 vs. 3.14 ± 1.55 μmol/mg (P = 0.001); catalase 3.23 ± 1.41 vs. 2.52 ± 1.3 nmol/min/ml (P = 0.03); superoxide dismutase [SOD] 0.11 ± 0.04 vs. 0.08 ± 0.01 U/ml (P = 0.0003); glutathione peroxidase [GPX] 0.03 ± 0.006 vs. 0.025 ± 0.006 nmol/min/ml (P = 0.01); glutathione reductase [GSH] 0.004 ± 0.002 vs. 0.004 ± 0.004 nmol/min/ml (P = 0.9)]; and glutathione transferase [GST] 0.0025 ± 0.0012 vs. 0.0027 ± 0.00017 nmol/min/ml (P = 0.7). Multivariate analysis showed catalase might have a protective effect against GDM development and LPO seems to be a risk factor for the disease. These data suggest an increase in oxidative stress and a decrease in antioxidative defence in women with late-onset GDM and, as such, may have considerable clinical implications in the pathogenesis and/or the course of the pregnancy in these patients.