There is only limited available evidence of a relationship between total plasma homocysteine (tHcy) levels and type 2 diabetes in hypertensive subjects.A total of 5,935 Chinese essentially hypertensive subjects were recruited by cluster sampling from 60 communities. The cases had diabetes, whereas the controls did not. Anthropometric indices and biochemical parameters were assessed using standard procedures. A multivariable analysis was performed to analyze the association of tHcy and type 2 diabetes susceptibility in hypertensive subjects.The 5,241 controls (women/men: 2,716/2,625) and 594 cases (women/men: 291/303) were recruited consecutively. The level of tHcy was dose- dependently associated with type 2 diabetes in the hypertensive women subjects. After controlling for corresponding confounding factors, a significant trend was only noted in the women subjects, with odds ratios per 5 μmol/L tHcy of 1.11 (95% confidence interval (CI), 1.07-1.16) in the crude model, 1.05 (95% CI, 1.01-1.11) in model 1, and 1.07 (95% CI, 1.02-1.13) in model 2. However, no significant result was found for levels of tHcy>or=15 μmol/L vs <15 μmol/L in the men, women and all hypertensive subjects.When the level of tHcy was divided into quartiles, tHcy was positively associated with type 2 diabetes in hypertensive women subjects. However, when the level of tHcy was separated into hyperhomocysteinemic (>or=15 μmol/L) and normal (<15 μmol/L), no significant results were observed.