BACKGROUND: Studies have demonstrated that poststroke depression(PSD) may be related with the disequilibrium between noradrenaline and 5-hydroxytryptamine (5-HT) caused by cerebral injury. The injured regions involve noradrenergic and 5-hydroxytryptaminergic neurons as well as conduction pathway.The levels of noradrenaline and 5-HT would be decreased.OBJECTIVE: To observe the effect of fluoxetine on preventing against PSD and recovery of neurologic function, and analyze the relationship of fluoxetine and the 5-HT level.DESIGN: A randomized controlled clinical trial.SETTING: Department of Neurology, First Hospital Affiliated to Soochow University.PARTICIPANTS: Ninety consecutive patients, 47 female and 43 male, were recruited who admitted to hospital for recent stroke in the Department of Neurology, First Affiliated Hospital of Soochow University between September 2003 and February 2005. Subjects were aged (64±7) years, ranging from 47 to 79 years old. They all met the diagnosis criteria of various cerebrovascular diseases formulated in the 4th National Cerebrovascular Disease Conference and confirmed as stroke by skull CT or MRI; The time from onset to tentative administration was less than 7 days; The patients had clear consciousness, without obvious language disorder. They were randomized into treatment group (n =48) and placebo group (n =42).METHODS: ①All the patients were given routine treatment according to treatment guideline of cerebrovascular disease after admission. Patients in the treatment group and placebo group received 20 mg/d fluoxetine and placebo (component: vitamin C) for 8 weeks, respectively. ② Neurologic deficit was assessed according to 24-item Hamilton Rating Scale for Depression (HAMD) and Activity of Daily Living Scale (ADL) before and at 2,4 and 8 weeks after test, separately; Meanwhile, the levels of platelet 5-HT and plasma 5-HT were determined. Grading criteria of HAMD intergral depression: non-depression ＜ 8 points;mild depression 8 - 20 points; moderate depression 21 - 35 points; severe depression ＞ 35 points. ADL was assessed with Barthel index score (full mark 100 points). Higher points indicated better incidence and smaller dependence. Neurologic deficit score was made according to scoring criteria of neurologic deficit formulated in 1995 4th National Cerebrovascular Disease Conference: a score of 0 - 15 indicated a mild focal neurologic deficit, a score of 16 - 30 a moderate focal neurologic deficit, and a score of 31 - 45 a severe focal deficit.MAIN OUTCOME MEASURES: Scores of HAMD, ADL and neurologic deficit, and levels of plasma and platelet 5-HT of patients from 2 groups before, 2,4 and 8 weeks after test.RESULRS: Seventy-three of 90 randomized patients participated in the final analysis. In the treatment group,11 patients dropped out due to insufficient clinical response (n =4), somatic side effects (n =2), intervening medical illness (n =1), hypomania (n =3), and other reasons (n =2). In the placebo group, 6 patients existed due to insufficient clinical response (n =2), somatic side effects (n =1) and other reasons (n =3). ① Before treatment, there were no significant differences in scores of HAMD, DAL and neurologic deficit in patients between two groups (P ＞ 0.05). After 8 weeks of treatment, the scores of HAMD, DAL and neurologic deficit in the treatment group were significantly different from those in the placebo group (12.6±5.3 vs. 16.3±3.7; 8.6±6.4 vs. 11.2±6.4; 60.4±12.5 vs. 52.3 ±13.5, P ＜ 0.01). ② After 8 weeks of treatment, platelet 5-HT level of patients in the treatment group was significantly lower than that in the placebo group [(325.3 ±110.5) mg/L vs. (653.6±138.4) mg/L, P ＜ 0.05], while there were no significant differences in plasma 5-HT between two groups (P ＞ 0.05).CONCLUSION: Early fluoxetine treatment obviously retards PSD. The increase of platelet 5-HT level promotes the recovery of neurologic function.