In a prospective, randomized, double-blind therapeutic trial, 191 patients with non-alcoholic steatohepatitis were treated for 8 weeks daily b.i.d. orally either with betaine glucuronate combined with diethanolamine glucuronate and nicotinamide ascorbate (Ietepar) (96 patients) or with undistinguishable placebo capsules (95 patients). The verum treatment effectively reduced by 25% hepatic steatosis (p < 0.01) and by 6% hepatomegaly (p < 0.05), while placebo did not significantly reduce the disorders. Verum was also more effective than placebo on discomfort in abdominal upper right quadrant. The global efficacy of treatment was rated by the doctor "very good" or "good" in 48% of verum treated patients and only in 17% after placcbo (P of difference = 9 x 10(-6)). 52% of patients self-rated efficacy as "very good" or "good" after verum and only 34% after placebo (P of difference = 0.017). The verum treatment provoked a significant reduction of the increased liver transaminases (ALT, AST and gamma-GT) while placebo was ineffective. Adverse events were recorded in 10% of verum-treated patients and in 7% under placebo (no significant difference). In both groups the adverse events were mild and transient, did not require treatment discontinuation and were undistinguishable from common symptoms of liver disorders. In conclusion, the 8-week treatment with betaine glucuronate combined with diethanolamine glucuronate and nicotinamide ascorbate was found effective in non-alcoholic steatohepatitis, a disorder for which the hitherto pharmacological interventions were poorly and inconsistently effective.