BACKGROUND: Data indicate that the levels of serum homocysteine in depressive patients are higher than those in normal subjects. OBJECTIVE: To investigate the levels of serum homocysteine in patients with major depressive disorder, to determine whether serum homocysteine levels differ with sex, family history, or drug treatment, and to compare depressive patients with normal subjects. DESIGN: Non-randomized concurrent control trial.SETTING: Mental Heath Center of Shandong Province.PARTICIPANTS: Forty in-patients (23 males and 17 females, 18-63 years old) with major depressive disorder were selected from the Mental Health Center of Shandong Province from January to October 2006. All selected patients met the depressive diagnostic standard of Chinese Classification of Mental Disorder (3rd Edition, CCMD-3), and total scores evaluated by the 17-item Hamilton Rating Scale for Depression (HRSD) were ≥ 20. Meanwhile, 36 healthy subjects (20 males and 16 females, 18-60 years old) were enrolled as controls; their total 17-item HRSD scores were ≤ 7. All selected subjects provided consent, and the study was approved by the local ethics committee. METHODS: Fasting venous blood (3 mL) was drawn in both groups at 8:00 in the ming. The levels of serum homocysteine were determined by a fluorescence polarization immunoassay (FPIA). The 17-item HRSD was also compiled from the patients when entering groups. The higher the scores were, the more severe the depression was. Enumeration data for both groups were compared by Chi-square test, measurement data were compared by t-test, and correlations were detected using Pearson and Spearman correlation analysis.MAIN OUTCOME MEASURES: ① Levels of serum homocysteine; ② incidence of hyperhomocysteinemia (Hhcy); ③ correlation between HRSD17 scores and levels of serum homocysteine in depressive patients.RESULTS: Forty depressive patients and 36 control subjects were included in the final analysis without any loss of participants. ① Levels of serum homocysteine and Hhcy detection rate: the levels of serum homocysteine in the depressive patients were significantly higher than those in the control group (t = 4.377, P=0.000). Hhcy detection rates were 42% (17/40) and 10% (4/36) in depressive group and control group, respectively. There was a significant difference between two groups (χ2 = 10.912, P = 0.001). In the depressive group, there were no differences in serum homocysteine levels between males and females, before and after treatment, or between patients with positive or negative family histories of depression (t = 0.217-0.520, P > 0.05). ② Correlation analysis: the HRSD17 scores in the depressive group were positively correlated with levels of serum homocysteine (r = 0.724, P = 0.000). CONCLUSION: ① The increase in serum homocysteine levels may play an important role in the pathogenic mechanism of depressive disorder. ② The higher the levels of serum homocysteine are, the more severe the depressive disorder is. ③ There are no significant differences in serum homocysteine levels between patients of different sex or family history, or before and after drug treatment, among depressive patients.