Introduction The main causes of reduced glucose levels during metformin therapy appear to be an increase in insulin action in peripheral tissues and reduced hepatic glucose output due to inhibition gluconeogenesis. Objective The purpose of the study was to establish the effect of metformin on fasting and postprandial insulin secretion. Method The study carried out was double blind, controlled, comparative, randomized, multicentric, including two groups of out-patient department (OPD) patients. 43 patients were administered metformin (Tefor ICN Canada), and 46 patients were given placebo. Patients enrolled in the study were newly diagnosed with diabetes mellitus (DM) type 2, glycaemia < 12 mmol/l, and had the Body Mass Index (BMI) > 30 kg/m2. Before treatment, blood biochemistry was done: fasting and postprandial glycaemia, glycosylated haemoglobin (HbA1c) value, fasting and postprandial insulinaemia, blood lipids (total cholesterol, total triglycerides, HDL cholesterol, and LDL cholesterol), and gamma glutaryl transferase (GGT) level. BMI was also established. After 42 days of treatment, fasting and postprandial insulinaemia were tested again. Analysis of the effects of therapy, and identification of co-variants for fasting and postprandial insulinaemia, were done by ANOVA two way and ANCOVA method. Results It was shown that metformin accompanied by diet, as compared to placebo accompanied by diet, lowered the fasting insulinaemia value during six weeks of therapy in obese patients with DM type 2 (24.392 mU/l vs. 25.667 mU/l), interacting both with BMI pre-therapy, and interacting with fasting insulinaemia pre-therapy (p<0.001). A significant effect of the interaction of covariants BMI and GGT was defined. As for the effect of therapy on postprandial insulinaemia, it was found that there was a high statistical significance of the effect of BMI interacting with initial values for postprandial insulinaemia before therapy, and interacting with therapy (p<0.01). Adjusted mean values for postprandial insulinaemia after therapy in the placebo group were lower as compared to the metformin group (44.807 mU/l vs. 47.114 mU/l). Conclusion It can be concluded that, as compared to placebo, metformin is more efficient in reducing insulin resistance in obese patients with DM type 2. In addition, as compared to placebo, metformin maintains more efficient productive insulin secretion, indicating that metformin protects the pancreas from beta cell depletion.