As many as half of obsessive-compulsive disorder (OCD) patients treated with an adequate trial of serotonin reuptake inhibitors (SRIs) fail to fully respond to treatment and continue to exhibit significant symptoms. Many studies have assessed the effectiveness of antipsychotic augmentation in SRI-refractory OCD. In this systematic review, we evaluate the efficacy of antipsychotic augmentation in treatment-refractory OCD. The electronic databases of PubMed, PsychINFO (1967-2005), Embase (1974-2000) and the Cochrane Central Register of Controlled Trials (CENTRAL, as of 2005, Issue 3) were searched for relevant double-blind trials using keywords 'antipsychotic agents' or 'neuroleptics' and 'obsessive-compulsive disorder'. Search results and analysis were limited to double-blind, randomized control trials involving the adult population. The proportion of subjects designated as treatment responders was defined by a greater than 35% reduction in Yale Brown Obsessive-Compulsive Scale (Y-BOCS) rating during the course of augmentation therapy. Nine studies involving 278 participants were included in the analysis. The meta-analysis of these studies demonstrated a significant absolute risk difference (ARD) in favor of antipsychotic augmentation of 0.22 (95% confidence interval (CI): 0.13, 0.31). The subgroup of OCD patients with comorbid tics have a particularly beneficial response to this intervention, ARD=0.43 (95% CI: 0.19, 0.68). There was also evidence suggesting OCD patients should be treated with at least 3 months of maximal-tolerated therapy of an SRI before initiating antipsychotic augmentation owing to the high rate of treatment response to continued SRI monotherapy (25.6%). Antipsychotic augmentation in SRI-refractory OCD is indicated in patients who have been treated for at least 3 months of maximal-tolerated therapy of an SRI. Unfortunately, only one-third of treatment-refractory OCD patients show a meaningful treatment response to antipsychotic augmentation. There is sufficient evidence in the published literature, demonstrating the efficacy of haloperidol and risperidone, and evidence regarding the efficacy of quetiapine and olanzapine is inconclusive. Patients with comorbid tics are likely to have a differential benefit to antipsychotic augmentation.