This study was designed to compare the effects of fluoxetine and imipramine on fasting blood glucose (FBG) in patients with major depressive disorder. Sixty nondiabetic patients with major depressive disorder (based on Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition criteria) entered this randomized, double-blind study. Patients did not receive any medication affecting serum FBG levels for at least 2 weeks before the initiation of the study. Patients were assigned to receive 20 to 40 mg/d of fluoxetine or 75 to 200 mg/d of imipramine for 8 weeks. Pregnant women and patients with diabetes mellitus and a history of any major heart disease were excluded from this study. Additionally, none of the patients should have received electroconvulsive therapy within 6 months before the initiation of the antidepressants. FBG levels were measured at the initiation, as well as 4 and 8 weeks after starting antidepressants. Nineteen patients in the fluoxetine and 24 patients in the imipramine groups completed the study. In the fluoxetine group, FBG level was decreased from 88.5 mg/dL (baseline) to 85.0 mg/dL at week 4 (P = 0.73), and to 79.8 mg/dL at week 8 (P < 0.001). On the other hand, in the imipramine group, FBG level was increased from 86.96 mg/dL (baseline) to 89.71 mg/dL at week 4 (P = 0.079), and to 96.90 mg/dL at week 8 (P < 0.001). This 8-week study showed that FBG levels may decrease in depressive patients receiving fluoxetine and may increase in those patients treated with imipramine. Therefore, it is suggested to measure and monitor FBG before initiation and during treatment with fluoxetine and imipramine.