Association between plasma homocysteine and microalbuminuria in persons without hypertension, diabetes mellitus, and cardiovascular disease

Background Microalbuminuria and plasma homocysteine levels are both considered to be markers of endothelial dysfunction and shown to be predictors of cardiovascular disease (CVD) in epidemiological studies. However, previous studies examining the association between plasma homocysteine and microalbuminuria have suggested that this association is explained by confounding factors such as preexisting CVD, diabetes, hypertension, and reduced kidney function. Methods We examined the association between plasma homocysteine and microalbuminuria in a representative sample of US men and women aged C20 years, who were free of diabetes mellitus, hypertension, and CVD and who participated in the Third National Health and Nutrition Examination Survey (n = 3,948). Microalbuminuria was defined as urinary albumin-to-creatinine ratio C30 mg/g. Results Plasma homocysteine levels were positively associated with microalbuminuria in men but not in women (p-interaction \0.0001) after adjusting for age, race/ethnicity, smoking, drinking, body mass index, physical activity, glomerular filtration rate, blood pressure, highdensity lipoprotein cholesterol, glycated hemoglobin, serum folate, serum vitamin B12, and C-reactive protein. In men, the multivariable odds ratio (OR) [95% confidence interval (CI)] of microalbuminuria comparing the highest to the lowest quartile of homocysteine was 5.17 (2.00–13.36); (p-trend = 0.005). Further, men in the highest quartile of homocysteine and age C60 years had [12-fold odds of microalbuminuria compared with men in the lowest homocysteine quartile and age \60 years. In contrast, in women, the multivariable OR (95% CI) comparing the highest to the lowest quartile of homocysteine was 0.96 (0.86–1.07); (p-trend = 0.41). Conclusions Among relatively healthy adults, plasma homocysteine levels are associated with microalbuminuria only in men.