No clear-cut evidence for cadmium-induced renal tubular dysfunction among over 10,000 women in the Japanese general population: a nationwide large-scale survey

Objective: To examine whether environmental exposure to cadmium has been inducing kidney dysfunction among middle-aged women in the general population in Japan. Methods: This study was conducted in 2000 and 2001. Morning spot urine samples were collected from 10,753 women (mostly aged 35 to 60 years) in ten prefectures all over Japan (thus about 1,000 women per site). Urine samples were analyzed for cadmium (Cd-U), calcium (Ca-U), magnesium (Mg-U), zinc (Zn-U), alpha(1)- and beta(2)-microglobulins (alpha(1)- and beta(2)-MG-U). The urinary analyte concentrations were corrected for creatinine (cr) concentration (i.e., Ucr). The data thus obtained were subjected to statistical evaluation by chi-square test, ANOVA, multiple comparison test, and simple regression analysis (SRA) as well as multiple regression analysis (MRA) including logistic regression analysis (LRA). Log-normal distribution was assumed for Cd-Ucr, alpha(1)-MG-Ucr and beta(2)-MG-Ucr, whereas normal distribution was considered for age, Ca-Ucr, Mg-Ucr and Zn-Ucr. Results: Geometric mean values (GM) of Cd-Ucr were distributed unevenly, depending on the sampling areas, with a grand GM of 1.3 mug/g cr, the highest (3.2 mug/g cr) and lowest GM values(0.8 mug/g cr) being significantly different from GMs of other areas. Correlation matrix analysis with subjects of all ages showed that log alpha(1)-MG-Ucr and 109 beta(2)-MG-Ucr correlated significantly (r=0.272 and 0.202, respectively) with log Cd-Ucr, but they correlated also with age (r=0.280 and 0.213, respectively). The same analysis with the two selected age groups (41-50 and 51-60 years), however, failed to show close correlation of alpha(1)-MG-Ucr and log beta(2)-MG-Ucr with Cd-Ucr. Both MRA and LRA indicated that age was a confounding factor in the evaluation of the effect of Cd-U on the two MG levels. Whereas the LRA with the all-age group showed a positive influence of log Cd-Ucr on log alpha(1)-MG-Ucr and 109 beta(2)-MG-Ucr, such effect disappeared in parallel with the disappearance of age effects when LRA was conducted with the two selected age groups. An exceptional observation was the influence of log Cd-Ucr on log alpha(1)-MG-Ucr (but not on log beta(2)-MG-Ucr) in LRA when a cut-off value of 5.00 mg for alpha(1)-MG-U/g cr was applied. Comparison between the low and high Cd-U groups showed that both alpha(1)-MG-Ucr and beta(2)-MG-Ucr were higher in the high Cd-U groups, but prevalence of cases with alpha(1)-MG-Ucr and beta(2)-MG-Ucr in excess of the cutoff values did not differ between the two groups except when a cut-off value of 5.00 mg/g cr was employed for alpha(1)-MG-U. Conclusions: In over-all evaluation, no clear-cut evidence was obtained in the present study to show that environmental exposure to Cd has induced tubule dysfunction among middle-aged women in the general population in Japan. It might be the case, however, that an increase in alpha(1)-MG-U was associated with Cd exposure. In this sense, it is apparently desirable from public health viewpoints to make further efforts to reduce the intensity of the general population's exposure to environmental Cd.