Thirty two patients with active Crohn's disease were included in a controlled randomised trial to determine the efficacy and safety of polymeric enteral nutrition compared with steroids, to achieve and maintain clinical remission. The polymeric diet was administered through a fine bore nasogastric tube by continuous, pump assisted infusion (2800 (SEM 120) kcal/day). The steroid group received 1 mg/kg/day of prednisone. Both treatments were effective in inducing clinical remission: 15 of the 17 patients given steroids and 12 of the 15 patients assigned to the polymeric diet went into clinical remission (defined by a Van Hees index < 120) within four weeks of treatment. The percentage reduction of the Van Hees index was 34.8 (4.9)% for steroids and 32.3 (5)% for enteral nutrition (mean difference 2.5%; 95% CI--11.8% to +16.8%). Mean time elapsed to achieve remission was similar in both groups (2.0 (1) v 2.4 (1.2) weeks). Tolerance of the enteral diet was excellent. Four patients in the steroid group had mild complications attributable to this treatment. Ten patients (66.6%) in the steroid group and five (41.6%) in the enteral nutrition group relapsed within a year of discharge, but no differences were found in the cumulative probability of relapse during the follow up period. These results suggest that polymeric enteral nutrition is as safe and effective as steroids in inducing short term remission in active Crohn's disease.