A 6-week double-blind placebo-controlled trial was carried out to examine the efficacy and tolerability of moclobemide, a monoamine oxidase type A selective and reversible inhibitor, in the treatment of bulimia nervosa. Patients were admitted to the study even if they were unable to adhere to a tyramine-free diet. Fifty-two normal-weight women (age range 18-40 years) suffering from bulimia nervosa (DSM-IV criteria) completed the trial. Particular emphasis was placed on evaluating the incidence of hypertension and other side-effects in chronically treated patients. At the usual antidepressant dose of 600 mg, moclobemide was not significantly superior to placebo in the reducing the weekly number of binge eating episodes or in improving several measures of eating attitudes and behaviour (BITE, EDI, TFEQ) in normal-weight bulimia nervosa. The dropout rate was relatively low (29%), and the side-effects were limited and equally distributed between the two treatment groups. No patient experienced a hypertensive crisis during the study and no serious side-effect was detected. The study indicates that moclobemide 600 mg pro die is not efficacious in bulimia nervosa, but it can be safely administered, even to young subjects, at a very high risk of consuming large amounts of tyramine-rich foods without dietary restrictions.