Blood glucose is often elevated in acute stroke, and higher admission glucose levels are associated with larger lesions, greater mortality and poorer functional outcome. In patients treated with thrombolysis, hyperglycemia is associated with an increased risk of hemorrhagic transformation of infarcts. For a number of years, tight glycemic control has been regarded as beneficial in critically illness, but recent research has been unable to support this notion. The only completed randomized study on glucose-lowering therapy in stroke has failed to demonstrate effect, and concerns relating to the risk of inducing potentially harmful hypoglycemia has been raised. Still, basic and observational research is overwhelmingly in support of a causal relationship between blood glucose and stroke outcome and further research on glucose-lowering therapy in acute stroke is highly warranted.