In previous open studies, misoprostol and metronidazole reduced nonsteroidal anti-inflammatory drug-induced intestinal permeability changes and inflammation respectively. We assessed the effects of indomethacin treatment (50 mg three times a day) for one week with either coadministered metronidazole (400 mg twice a day, group 1, N = 9) or misoprostol (200 micrograms four times a day, group 2, N = 7) on intestinal permeability to [51Cr]EDTA and mannitol in healthy volunteers, using double-blind, placebo-controlled, randomized techniques. Given alone, neither metronidazole nor misoprostol affected [51Cr]EDTA permeation, whereas indomethacin alone increased it from 1.20 (0.40) [mean percent urinary recovery (SD) groups 1 and 2] to 2.43 (0.72), P < 0.002. Coadministered metronidazole (group 1) prevented this increase [1.10 (0.39) before, 1.55 (0.54) after, P > 0.05], whereas misoprostol (group 2) did not [1.31 (0.51) before, 3.26 (1.10) after, P = 0.005]. No drug regimen altered mannitol permeation. Indomethacin and misoprostol did not affect urinary recovery of intravenously administered probes. The results with metronidazole, if related to its antibacterial effects, support evidence from animal models that bacteria contribute to NSAID-induced intestinal damage. The previously reported reduction of indomethacin-induced increased permeability by misoprostol during a one-day study is not seen when the drugs are used in standard clinical doses for one week.