The aim of this study was to assess the effect of the long-term different diabetic therapies on the plasma leptin level in type 2 diabetic subjects. We measured plasma leptin, body fat and fasting plasma insulin in 96 type 2 diabetic male subjects. They had received the same treatment regimen for more than one year (3.5 +/- 2.3 years, mean +/- SD) and were weight-stable over the previous three months. The distribution was as follows: diet control group: 32, oral hypoglycemic agent (OHA) group: 32, and insulin group: 32. The results showed that the plasma leptin level of the different therapy groups was all positively correlated with body fat. The fasting insulin levels were significantly higher (p < 0.0001) in the insulin group than those in the other two groups. The fasting insulin of the OHA group was also greater than that of the diet group but was not statistically significant. The leptin concentrations were significantly higher in the insulin group (p < 0.001) and OHA group (p = 0.0082) than that in the diet group. The leptin concentrations of insulin group were also significantly higher (p = 0.0021) than that of the OHA group. Stepwise multiple regression analysis revealed that the significant differences in the leptin level of whole group was mainly affected by fasting insulin (p < 0.0001), followed by fat percentage (p = 0.001), fat distribution (p = 0.009) and fasting sugar (p = 0.02), whereas there was no association of leptin with age, height, glycosylated hemoglobin A1c, lipid, or blood pressure. We concluded that long-term different diabetic therapies may affect the plasma leptin level, which is mediated mainly by insulin changes. This insulin effect is independent of body fat and may be superior to the fat effect on plasma leptin in the type 2 diabetic patients.