The objective of the study was to demonstrate whether the N-methyl-D-aspartate antagonist, amantadine, can safely ameliorate tardive dyskinesia (TD) without deteriorating the mental state of the patients. Twenty-two TD patients, with a mean age 52 years, participated in the study. A double-blind, placebo-controlled, crossover design was used. Patients were randomly assigned to receive either amantadine or placebo for 2 weeks followed by a washout period of 4 days. Subsequently,the groups were crossed over, and the procedure repeated. Participants received amantadine (100 mg) or placebo. Tardive dyskinesia was assessed by means of the Abnormal Involuntary Movements Scale(AIMS). The primary efficacy end point was changes in AIMS score, at baseline and at the end of the 2-week treatment period. After amantadine treatment, patients exhibited a reduced average score of total AIMS (from 13.5 before treatment to 10.5 after treatment,P = 0.000), of facial and oral AIMS (from 5.5 before treatment to 4.2 after treatment, P = 0.002), of extremity AIMS (from 4.18 before treatment to 2.8 after treatment, P = 0.000), and of severity AIMS (from 2.04 before treatment to 1.54 after treatment, P = 0.002). With amantadine,the average total AIMS reduction was 21.81%. On the contrary with placebo treatment, no reduction was noted.In addition, amantadine administration exhibited a positive effect,which was statistically significant for incapacitation and Clinical Global Impression score. Amantadine did not alter any of the cognitive measures used in this study, as Mini-Mental State Examination, distress, and Brief Psychiatric Rating Scale. Amantadine may be an effective and safe treatment for TD. The severity of TD movements in patients receiving amantadine improved significantly more than in those receiving placebo, as measured by the AIMS score.