Gastrointestinal complaints and occult bleeding have been commonly described in marathon runners. We hypothesized that these complaints may arise from intestinal ischemia caused by the shunting of blood away from the splanchnic circulation during endurance racing followed by reperfusion injury. Studies in animal models have suggested prophylactic vitamin E supplementation may prevent this type of injury. We sought to determine if prerace vitamin E supplementation would prevent intestinal ischemia/reperfusion injury in humans. Forty subjects who planned to complete the 1996 Houston-Tennaco Marathon were randomized to receive vitamin E (1000 IU daily) or placebo (soya lecithin) for 2 wk before the race in a double-blinded trial. Inclusion criteria included no use of non-steroidal anti-inflammatory drugs (NSAIDs) within 24 d of the race or vitamin or mineral supplements containing vitamins C or E or selenium within 30 d of the race. Subjects were studied 2 wk before the race and immediately following the race. Blood was obtained for serum vitamin E and total lipid and salicylate concentrations. A solution of lactulose (5 g) and mannitol (2 g) was consumed and urine was collected for 6 h. Aliquots were assayed for lactulose and mannitol concentration. Stool samples were tested for occult blood and following the race subjects rated their nausea, abdominal pain, and cramping on a 1-5 scale. Twenty-six subjects (24 male, 2 female) completed the marathon. Finish times ranged between 2 h 43 min and 5 h 28 min. All subjects had heme-negative stool prerace and four developed heme-positive stool postrace, with no difference between vitamin E and placebo groups (Fisher's exact = 0.63). All had non-detectable salicylate concentrations pre- and postrace. Serum vitamin E concentration increased in botPP = 0.02 in the vitamin E group and 1.45 +/- 0.40 to 1.66 +/- 0.48 mg/dL in the placebo group, P = 0.02). However, the serum vitamin E: total lipid ratio increased significantly in the vitamin E-supplemented group (0.0022 +/- 0.0002 to 0.0051 +/- 0.0015, P = 0.02), but not in the placebo group (P = 0.25). Overall, the urinary lactulose:mannitol ratio increased from 0.03 +/- 0.02 to 0.06 +/- 0.08 postrace (P = 0.06) without difference between vitamin E or placebo groups. Intestinal permeability increased significantly more in those who developed occult bleeding. More subjects in the placebo group developed abdominal cramping (Fisher's exact = 0.04) and abdominal pain (Fisher's exact = 0.06), although there was no difference in severity between groups. There was no difference in the incidence of nausea and no diarrhea was reported by any subject. Intestinal permeability tends to increase and occult gastrointestinal bleeding occurs during endurance running, suggesting the occurrence of intestinal ischemia/reperfusion injury. Prerace supplementation with the antioxidant vitamin E had no effect on performance, intestinal injury, occult bleeding, or the severity of postrace gastrointestinal complaints. Vitamin E supplementation was associated with a decreased incidence of these complaints but had no effect on their severity.