OBJECTIVETo assess factors predisposing to proteinuria in patients with type 2 diabetes mellitus (DM).

METHODSA case-control study was conducted on two groups of patients with type 2 diabetes without proteinuria (urine protein < 300 mg/24 h, n = 106) and with proteinuria (urine protein >or= 500 mg/24 h, n = 106). The two groups were matched by age (+/- 3 years), sex, race and place of residence. Information on these subjects including demography, history of disease, family history of diseases, lifestyle and behavioral variables, were obtained through questionnaire. Variables including blood pressure, fasting blood glucose (FBG), quantity of protein in 24-h urine were measured. Cox's proportional hazards regression analysis was then performed.

RESULTSFactors that were independently associated with the development of proteinuria, would include occupation, diabetic duration, glycemic control, hypertension, duration of hypertension and daily intake of vegetable after diagnosis of DM. Their corresponding odds ratios (OR) with 95% confidence intervals (CI) were 2.429 (1.299 - 4.542), 1.078 (1.029 - 1.131), 2.316 (1.341 - 3.998), 1.810 (1.059 - 3.092), 1.043 (1.012 - 1.074) and 0.551 (0.334 - 0.907), respectively, while the presence of proteinuria was not associated with family history of DM, family history of hypertension, certain dietary habits and smoking habit. Multivariate logistic regression model was then fitted with three variables, including diabetic duration, glycemic control and hypertension.

CONCLUSIONThe prevalence of proteinuria in labor workers was higher than that in intellectuals. Longer history of diabetic duration, poor glycemic control, hypertension and longer duration of hypertension all independently contributed to the development of proteinuria. The risk of proteinuria decreased with increasing daily intake of vegetables after the patients being diagnosed as having DM.