Concentrations of reactive oxygen species (ROS) increase during exercise secondary to increased oxygen uptake, xanthine oxidase activity, and immune system activation. Carbohydrate compared to placebo beverage ingestion is associated with an attenuated cortisol and catecholamine response. Catecholamines can undergo autooxidation to form ROS. We hypothesized that during intense exercise, ingestion of carbohydrate compared to placebo would diminish oxidative stress. Sixteen experienced marathoners ran on treadmills for 3 h at approximately 70% VO2(max) on two occasions while receiving carbohydrate or placebo beverages (11/h, double-blinded) in a randomized, counterbalanced order. Blood samples were collected before and immediately after exercise, snap frozen in liquid nitrogen, and stored at -80 degrees C until analysis. Plasma samples were analyzed for F2-isoprostanes (FIP) and lipid hydroperoxides (ROOH) as measures for lipid peroxidation, ferric reducing ability of plasma (FRAP) as a measure of plasma antioxidant potential and for cortisol. The pattern of change in cortisol was significantly different between carbohydrate and placebo conditions (P = 0.024), with post-exercise levels higher in the placebo condition. Under both carbohydrate and placebo conditions, significant increases in FIP, ROOH, and FRAP were measured, but the pattern of increase was not different (FIP, interaction effect, P = 0.472; ROOH, P = 0.572; FRAP, P = 0.668). Despite an attenuation in the cortisol response, carbohydrate compared to placebo ingestion does not counter the increase in oxidative stress or modulate plasma antioxidant potential in athletes running 3 h at 70% VO2(max).