BACKGROUND
We performed this study to investigate the effect of metformin on glycemia, insulin secretion and body weight in patients with non-insulin-dependent diabetes melltus(NIDDM) who could not aehieve satisfactory glycemic control by sulfonylurea or diet therapy. METHODS: A total of 167 patients with NIDDM were included in this study. At baseline the patients underwent anthropometry and a 75g oral glucose tolerance test. Jn addition, levels of hemoglobin Alc (HbAlc), setum lipids, fasting and postprandial 2hr glucose were measured. Metformin was given in an initial dose of 500mg twice daily and increased by 500mg every month as long as the fasting blood sugar(FBS) concentration exceeded 7.8mmol/L and the side effects were tolerable. After 3 rnonths of metformin therapy we defined a responder as a patient who experienced a FBS of under 7.8 mmol/L or a HbAlc of under 7%. Patients who failed to respond to metformin monotherapy were excluded in the study. Anthrapometric changes and results of a 75 g oral glucose tolerance test were reevaluated in the responder group after 6 months of metformin treatment. RESULTS: I) The overall response rate to metformin mono-therapy was 55.6%(79/142) in the study population. 2) There were significant changes in body weight (64.4+/-8.2 vs 62.9+/-8.4 kg, p(0.01) and body mass index(25.3+/-2.3 vs 24.6+/-2.3kg/m, p<0.01) during metformin treatment. 3) There were significant decreases in the fasting, postprandial 2hr serum glucose(10.1+/-2.8 vs 7.9+1.6, 15,2+/-5.0 vs 12.2+/-3.9 mmol/L, p 0.01) and HbAlc levels(8.4+/-1.7 vs 6.5+/-0.9%, p<0.05) after 6 months of metformin treatment. 4) There were significant decreases in the levels of AUC[g](59.2+/-15.5 vs 49.4+/-9.4mmol L-1. Min-1, p =C0.01) without changes of AUC[I] and AUC[I]/ AUC[g] ratio (558.0+486.0 vs 536.4+374.4 pmol.L-1. Min-1, p=0.71, 11.7+/-13.0 vs 11.8+/-10.0, p=0.89). 5) The incidence of side effects was 25% in the study population, but most of them were mild and fade away with continuous use of metformin, CONCLUSION: Metforrnin monotherapy improved glycemic control in NlDDM patients who failed to respond to diet or sulfonylurea therapy and may be a useful hypoglycemic agent for the treatment of NIBDM.