Individuals with type 2 diabetes mellitus (n = 105; age 36-71 years) on diet therapy alone, and with quite good glycaemic control (mean HbA1c approximately 7.0%) were randomized to receive acarbose (100 mg three times daily) or placebo for 16 weeks, and changes in clinical and metabolic parameters indicative of Syndrome X were monitored. Fasting levels of glucose, glycosylated haemoglobin (HbA1c), true insulin, proinsulin, fibrinogen and lipids were measured four times weekly, and glucose, insulin, proinsulin and triglyceride responses to a standardized 1.6 MJ breakfast were determined at 0, 1 and 2 h post meal. Analysis was on an intention-to-treat basis. Fasting levels of glucose (P < 0.0001), triglycerides (P = 0.03) and HbA1c (P = 0.003) were reduced by acarbose over the 16 weeks of treatment. The mean change in HbA1c from week 0 to 16 differed by 0.4% (P = 0.003) between the two groups. Insulin (P = 0.06), proinsulin (P = 0.07) and glucose (P < 0.0001) responses to the standard meal were reduced. These data show that acarbose reduces fasting glucose and triglyceride levels, lowers HbA1c and limits the glycaemic and insulin response to food in individuals with type 2 diabetes mellitus with Syndrome X. Pharmacological agents that improve the metabolic environment and reduce insulin resistance have the potential to limit the progression of atherogenesis associated with type 2 diabetes mellitus.