Objective: The aim of the study was to investigate the efficacy and tolerability of long-term acarbose therapy in type 2 diabetic patients. Study design: In this double-blind, single-centre group comparison, patients were randomised to receive either acarbose or matching placebo, in addition to their regular antidiabetic therapy, over a period of 78 weeks. Eligibility for inclusion in the efficacy evaluation included a study duration of ≥510 days. Methods: The primary efficacy parameter was the change in glycosylated haemoglobin (HbA1) from baseline to end of study. Secondary variables included changes in blood glucose and lipid parameters, as well as signs of retinopathy and nephropathy. Patients: A total of 139 patients were assessed for safety and 88 patients (44 in each treatment group) were included in the efficacy analysis. Patients were generally overweight and the majority had previously been treated with sulphonylureas. Results: Acarbose significantly improved fasting and 1-hour postprandial blood glucose levels compared with placebo (p = 0.039 and 0.009), and improvements in HbA1 with acarbose versus placebo fell just short of significance (p = 0.057). There were no differences between treatments in changes in microvascular complications, but blood pressure improved with acarbose treatment. Two patients in the acarbose group experienced elevated liver enzyme levels. Generally, acarbose had a good safety profile and was well tolerated. Conclusion: Long-term treatment with acarbose was safe and efficacious in patients with type 2 diabetes mellitus that was insufficiently controlled by other oral antidiabetics.