Muiltiple linear regression analysis between 2-h PG and different covariates in the entire study population and in groups with different glucose tolerance.

Multiple linear regression analysis was performed to investigate the determinants of 2-h postload glucose, after adjustment for gender, age, smoking and drinking, BMI, SBP, DBP, total cholesterol, HDL and LDL, triglycerides, log-CRP, e-GFR, FPG and HbA1c. Values were regression coefficient (β) ± standard error (SE). Definitions of NGT (normal glucose tolerance), IGT (impaired glucose tolerance) and 2-h NDM (2-h newly diagnosed diabetes) were in methods. SUA, serum uric acid; BMI, body mass index; SBP, systolic blood pressure; DBP, diastolic blood pressure; HDL, high density lipoprotein; LDL, low density lipoprotein; log-CRP, logarithmic transformed C-reactive protein; FPG, fasting plasma glucose; e-GFR, estimated glomerular filtration rate. International system of units (SI) conversion: plasma glucose 1 mg/dl = 1/18 mmol/l; SUA 1 mg/dl = 59.5 µmol/l.