BACKGROUND: Severe, refractory obsessive-compulsive disorder (OCD) is a disabling condition. Stimulation of the subthalamic nucleus, a procedure that is already validated for the treatment of movement disorders, has been proposed as a therapeutic option. METHODS: In this 10-month, crossover, double-blind, multicenter study assessing the efficacy and safety of stimulation of the subthalamic nucleus, we randomly assigned eight patients with highly refractory OCD to undergo active stimulation of the subthalamic nucleus followed by sham stimulation and eight to undergo sham stimulation followed by active stimulation. The primary outcome measure was the severity of OCD, as assessed by the Yale-Brown Obsessive Compulsive Scale (Y-BOCS), at the end of two 3-month periods. General psychopathologic findings, functioning, and tolerance were assessed with the use of standardized psychiatric scales, the Global Assessment of Functioning (GAF) scale, and neuropsychological tests. RESULTS: After active stimulation of the subthalamic nucleus, the Y-BOCS score (on a scale from 0 to 40, with lower scores indicating less severe symptoms) was significantly lower than the score after sham stimulation (mean [+/-SD], 19+/-8 vs. 28+/-7; P=0.01), and the GAF score (on a scale from 1 to 90, with higher scores indicating higher levels of functioning) was significantly higher (56+/-14 vs. 43+/-8, P=0.005). The ratings of neuropsychological measures, depression, and anxiety were not modified by stimulation. There were 15 serious adverse events overall, including 1 intracerebral hemorrhage and 2 infections; there were also 23 nonserious adverse events. CONCLUSIONS: These preliminary findings suggest that stimulation of the subthalamic nucleus may reduce the symptoms of severe forms of OCD but is associated with a substantial risk of serious adverse events. (ClinicalTrials.gov number, NCT00169377.)