BACKGROUND: Neuroleptic agents are effective in treating Tourette syndrome (TS). Side effects such as acute myodystonia and oculogyric crisis as well as the risk of delayed dyskinesia have led to search for alternative therapy.OBJECTIVE: To explore the effects of metoclopramide in controlling the symptoms of TS and improving the intelligence and memory function without causing extrapyramidal side effects.DESIGN: Randomized controlled trial and blind method.SETTING: Neurological Department of Xiangya Hospital Affiliated to Central South University; Institute of Traditional Chinese Medicine of Hunan Province; Neurological Department of the First People's Hospital of Nanning City; Public Health College of Central South University.PARTICIPANTS: Fifty patients with TS came from Tic Specialty Clinic,Neurological Department of Xiangya Hospital, between January 2000 and June 2001. Patients and their family members agreed toaccept our medical treatment and reexamination on fixed dates. There were 38 males and 12 females aged 5-21 years with the mean age of (10.4±4.2) years. All of them were divided into metoclopramide group and haloperidol group with 25 in each group.METHODS: Metoclopramide and haloperidol were white tablets bottled without tag (100 tablets of white color medicine in each bottle). The two types of medicine in each bottle were marked Ⅰ or Ⅱ, respectively. The blind method (patients, their parents and evaluation blinded) was used to test the real biochemical nature of the medicine. Medicine was given by a specific doctor on regular days (2, 4, 8, 16 weeks) to patients on their revisit. The daily dose of metoclopramide for body weight was 0.5-2 mg/(kg· d) and haloperidol was 0.05 mg/(kg· d) and was assigned 3-4 times (principle of dose individuality). In addition, TS patients with severe tics in metoclopramide group would receive metoclopramide 10 mg by intramuscular injection twice a day, at the initial treatment for 2-4 days. ①Before and after treatment, TS severity of patients at weeks 2, 4, 8 and 16 was evaluated by YGTSS (the total score was 100 points; the higher the score, the more severe the condition and overall impairment). ② The clinical curative effect of the medicines was evaluated with YGTSS deduction rate before and after treatment. ③ Asberg side-effect rating score: The lower the score, the fewer and milder the side effects. ④ The cognition function was evaluated with WISC score before treatment and after 3 months of treatment. WISC consisted of language and performance components with 11 subtests, while memory scale category had 10 subtests (memory score lower than the standard indicated problem in memory function).MAIN OUTCOME MEASURES: ① Comparison of YGTSS in patients before and after treatment; ② comparison of efficacy in patients of the two groups; ③ comparison of intelligence quotient and memory quotient of patients before and after treatment; ④ adverse events and side effects.RESULTS: Totally 50 patients entered the final analysis. ① Comparison of Tourette' s syndrome in patients before and after treatment: YGTSS score in the two groups decreased significantly after treatment. The comparison of the differences suggested that metoclopramide had better curative effect than haloperidol. In metoclopramide group, YGTSS score was reduced steadily at weeks 2, 4, 8 and 16. In haloperidol group, YGTSS score was also reduced steadily at weeks 2, 8 and 16, but it was similar at week 2 and week 4. ② Comparison of efficacy in patients of the two groups: The total clinical curative effect rate was 92% (23/25) in both groups withoutany difference (P ＞ 0.05). ③ Comparison of intelligence quotient beforeand after treatment: In both metoclopramide group and haloperidol group itwas higher after treatment than before treatment (93.0±15.1, 87.0±14.6; 93.2±17.0, 87.3±13.6, t=3.43, 2.3, P ＜ 0.01). ④ Comparison of memory quo tient of patients before and after treatment: It was higher only in metoclo pramide group after treatment than before treatment (87.8±12.8, 75.8±15.5, t=3.30, P ＜ 0.01). ⑤ Score of Asberg side effects: It was lower in metoclo pramide group than in haloperidol group [(1.00±0.76), (3.24±1.40) points, t=7.05, P ＜ 0.01]. CONCLUSION: The results suggest that metoclopramide is effective in con trolling the symptoms of TS and improving intelligence and memory function. Metoclopramide causes mild side effects, but no extrapyramidal side effects.