BACKGROUND: Chronic complications in type 2 diabetic patients have microvascular and macrovascular components. Previous studies have shown that incidence of macrovascular complications correlates with the serum homocysteine levels, but the relationship is unclear. In addition, the connection between the microvascular complications and the serum homocysteine levels is still obscure and controversial. In this study, the relationship between the serum homocysteine levels and microvascular and macrovascular complications were evaluated in type 2 diabetic patients. METHODS: In 58 type 2 diabetic patients, the serum homocysteine levels, folic acid levels, Vit B12 levels, PAI-1 levels, the standard risk factors for macrovascular complications, the fasting serum glucose levels, the HbA1C levels, and the fasting insulin and C-peptide concentrations, the renal function tests, and the carotid intima-media thickness were measured and the relationship between them and the serum homocysteine level was analyzed according to the presence and absence of macrovascular and microvascular complications. RESULTS: 1) In type 2 diabetic patients, the mean serum homocysteine level was 9.9+/-.2 mol/L. The serum homocysteine level showed no relationship with the clinical and biochemical variables including the risk factors for atherosclerosis except the serum creatinine and creatinine clearance. 2) The maximum, minimum, and mean of the intima- media thickness of right carotid artery were 4.00+/-.20, 0.50+/-.04, 1.04+/-.62 mm, of left carotid artery were 3.54+/-.00, 0.31+/-.02, 1.03+/-.55 mm, and means were 3.77+/-.10, 0.44+/-.03, 1.03+/-.54 mm, and correlated with the serum homocysteine leve l (p=0.03), but only the serum LDL cholesterol level independently correlated with the intima-media thickness (p=0.04). 3) The serum homocysteine level (p=0.01) and intima-media thickness (p<0.01) was significantly higher in type 2 diabetic patients with macrovascular complications than without it. 4) The serum homocysteine level did not correlate with the incidence microvascular complications, but the intima-media thickness did correlate with diabetic nephropathy (p=0.03). CONCLUSIONS: The serum homocysteine level did not correlated with the incidence of diabetic microvascular complications. However, there was a small correlation with the risk factors of macrovascular complications. The intima- media thickness correlated with the incidence of macrovascular complications, and the relationship with diabetic nephropathy requires further study.