This paper summarizes literature data concerning the action of acarbose, an alpha-glucosidase inhibitor, on the concentrations of plasma lipids. Clinical trials in which acarbose has been used in the treatment of non-insulin-dependent diabetics have sometimes shown that it reduces serum triglycerides while it has little or no effect on serum cholesterol levels. The results of a randomized double-blind placebo-controlled study lasting 24 weeks are discussed in more detail. Under the controlled conditions, the effects of acarbose treatment on fasting concentrations of cholesterol, HDL-cholesterol, and triglycerides did not reach statistical significance for the entire patient group. However, in the highest tertile of initial cholesterol concentrations acarbose treatment led to significant lowering of the cholesterol concentration and of the total-to-HDL-cholesterol ratio. The most important benefits of acarbose were observed after a test meal given on day 0 and on week 24 of treatment. The triglyceride increment 1 h postprandial was significantly lowered. This was associated by a significant decrease of the insulin increment. Reduction of hyperinsulinaemia appears to be the mechanism by which acarbose treatment can improve plasma lipid concentrations.