OBJECTIVE: Alcohol consumption has been reported to have both beneficial and harmful effects on stroke occurrence. Several studies have demonstrated a significant association with heavy drinking, but the evidence linking light-to-moderate alcohol consumption still remains unclear. This study, using a systematic review of the published literature, aimed to explore the dose-response relationship between alcohol and stroke, the effect of irregular drinking and the effect of beverage types on the risk of stroke. METHODS: A structured search for English-language published literature since 1966 was made using several electronic databases. This was supplemented using a hand search of references in review articles and additional searches on key authors. From the 153 eligible articles, 41 studies were selected according to study design, categorization of the exposure and outcome measures. FINDINGS: An association between recent alcohol use and stroke was consistently reported. There was also some evidence for a linear positive association for haemorrhagic stroke and alcohol consumption. Inconsistent results emerged on the J-shaped relationship between alcohol and ischaemic stroke, and the association between alcohol and non-fatal or fatal stroke combined. The importance of the pattern of drinking was also demonstrated, indicating a higher risk for irregular drinkers. CONCLUSIONS: There is insufficient evidence to conclude that light-to-moderate alcohol drinking and wine intake have beneficial effects on stroke occurrence. On the contrary, findings from this review suggest the opportunity for a primary prevention regarding heavy drinking and binge drinking. More information regarding the risk of stroke associated with irregular alcohol drinking, and the joint effects of alcohol with other risk factors, would clarify the complex interaction between alcohol and stroke.