Patients with obsessive-compulsive disorder (OCD) have been shown to be preferentially responsive to serotonin (5-HT) uptake-inhibiting antidepressants including clomipramine, fluoxetine, fluvoxamine, and sertraline. The nontricyclic antidepressant, trazodone, also possesses serotonin reuptake inhibiting properties and has been reported to be efficacious in OCD in several case reports and open trials. In order to investigate trazodone's potential antiobsessive efficacy in a controlled fashion, 21 patients with OCD were entered into a double-blind, parallel design comparison of trazodone and placebo. There were no significant differences in baseline rating scores of OCD or depressive symptoms between those who entered the trazodone phase (N = 13) versus those who entered the placebo phase (N = 8). As measured by standardized OCD and depression rating scales, there was no significant difference in OCD or depressive symptoms in the 17 patients who completed 10 weeks of trazodone (N = 11, mean daily dose, 235 +/- 10 mg) versus 10 weeks of placebo (N = 6) administration. In comparison to clomipramine and fluoxetine treatment which we have found to be associated with greater than 95% reduction in platelet 5-HT concentration, there was only a 26% mean reduction in platelet 5-HT concentration after 10 weeks of trazodone administration. These results indicate that trazodone lacks substantial antiobsessive effects and is associated with only modest reductions in platelet 5-HT concentrations.