Background: Since 2002, there has been contrasting evidence indicating an increased risk of adverse cerebrovascular outcomes in people treated with the two most commonly prescribed atypical antipsychotics, risperidone and olanzapine. Objective: The objective of this systematic review is to examine the evidence of the correlation between antipsychotic drug use and the risk of cerebrovascular diseases. Method: We used the Medline computer database to search the relevant papers published in English from January 1966 to October 2005. We used the following key words: antipsychotic agents, transient ischaemic attack, cerebrovascular accident, stroke, vascular diseases, risk factors and mortality. Articles that investigated the relationship between antipsychotic drug exposure and subsequent cerebrovascular morbidity and mortality were collected and reviewed. Results: We found one meta-analysis based on a systematic review of randomised controlled trials (RCTs), 4 pooled analyses of RCTs, 4 database analyses, 1 case-control study and 1 cross-sectional study. A total of 11 RCTs suggested that 48 out of 2187 (2.2%) subjects exposed to antipsychotics experienced cerebrovascular adverse events, compared with 10 out of 1190 (0.8%) placebo-treated subjects. In terms of risks, the relative risk was statistically significant for risperidone (3.2, 95% Confidence Interval (CI) 1.4 to 7.2) but not for olanzapine 1.8 (95% Confidence Interval (CI) 0.5 to 6.3). In general, data from large observational administrative databases suggested no increased risk of cerebrovascular events with the atypical antipsychotics, compared with typical antipsychotics, although the evidence seems still inconclusive. Conclusion: So far, randomised and non-randomised studies provided contrasting findings on the risk of cerebrovascular accidents in subjects exposed to antipsychotic drugs. Individual patient data meta-analyses, carried out by independent organisations, are urgently needed to systematically summarise factors associated with cerebrovascular morbidity and mortality in individuals treated with first and second-generation antipsychotic drug.