BACKGROUND Body mass index (BMI) is a predictor of chronic kidney disease (CKD) in the general population. Here, we evaluated the association between BMI, body composition and CKD according to age and gender.

METHODS We examined 75,127 adults between 2005 and 2008. Odds ratios for CKD were calculated between two age groups (a younger group, aged <65 years, and an elderly group, aged > or =65 years) using BMI, lean mass index (LMI, lean mass divided by the square of height), fat mass index (FMI, fat mass divided by the square of height), waist circumference and the lean mass-to-fat mass ratio (LFR).

RESULTS The elderly group comprised 8.9% of the population, and 7.4% of all subjects had CKD. By multivariate analysis, BMI was positively correlated with CKD, but FMI was negatively correlated in the younger group. LMI was not associated with CKD in the younger group. In the elderly group, BMI, FMI and waist circumference were positively correlated with CKD. However, LMI and LFR were negative predictors of CKD. In female subjects, the correlation trends in the younger and the elderly group were opposite.

CONCLUSIONS The associations between BMI, body composition and CKD differed according to age and gender stratifications.