Possible effects of environmental cadmium exposure on kidney function in the Japanese general population

Objectives: To examine whether the current level of environmental exposure to cadmium (Cd) is associated with kidney dysfunction among general populat ions in Japan. Methods: A nationwide survey was conducted in Japan from 1991 to 1997 at 30 survey sites (with no known environmental heavy metal pollution), by the c ollection of 24-h food-duplicate samples, peripheral blood specimens and mo rning spot urine samples. In practice, 607 non-smoking adult women provided these samples. After being wet-ashed, the samples were analyzed for Cd in food duplicates (Cd-F), in blood (Cd-B) and urine (Cd-U) by inductively-cou pled plasma mass spectrometry (ICP-MS). Urine samples were also analyzed fo r alpha(1)-microglobulin (alpha(1)-MG), beta(2)-microglobulin (beta(2)-MG) and retinol-binding protein (RBP), creatinine (cr) and specific gravity. Po ssible tubular dysfunction in association with Cd exposure was examined by simple, multiple and logistic regression analyses, and comparison among thr ee different Cd-dose groups. To minimize the confounding effects of aging, 367 women from 41 to 60 years old were selected and subjected to the same s tatistical analyses. Results: The analysis of a whole population of 607 women showed that alpha( 1)-MG and possibly beta(2)-MG increased as a function of Cd-F, Cd-B and Cd- U. When the analysis was repeated with the selected population of 367 women aged 41-60, the Cd dose-dependent changes in alpha(1)-MG and beta(2)-MG be came less evident. The distribution of the selected population with alpha(1 )-MG above two low cut-off values of >4.9 and >8.4 mg/g cr or with beta(2)- MG above the lowest cut-off value of >400 mu g/g cr, was biased toward the group with higher Cd-Ucr, but such bias was not significant for both alpha( 1)-MG and beta(2)-MG when higher cut-off values were employed. No bias was detected with REP. Logistic regression analysis with alpha(1)-MG, beta(2)-M G and REP (with cut-off values given above) in combination with age, Cd-F, Cd-B and Cd-Ucr gave essentially the same results. Conclusions: The evidence for kidney dysfunction was of borderline signific ance in the present study population for which geometric mean Cd-F, Cd-B an d Cd-U were 24.7 mu g/day, 1.76 mu g/l, and 3.94 mu g/g cr, respectively. T he findings might suggest at the same time that the safety margin is small for the Japanese general population regarding environmental Cd exposure.