Tubular and glomerular kidney effects in the Chinese general population with low environmental cadmium exposure

Chemosphere 147 (2016) 3e8

Contents lists available at ScienceDirect

Chemosphere journal homepage: www.elsevier.com/locate/chemosphere

Tubular and glomerular kidney effects in the Chinese general population with low environmental cadmium exposure Dongyue Wang a, 1, Hong Sun b, *, 1, Ye Wu a, Zhengyuan Zhou a, Zhen Ding b, Xiaodong Chen b, Yan Xu b, ** a b

Changshu Center for Disease Prevention and Control, Suzhou 510000, Jiangsu, China Jiangsu Provincial Center for Disease Prevention and Control, Nanjing 210009, Jiangsu, China

h i g h l i g h t s
Cadmium in urine and blood were positively associated with tubular biomarkers even in children.
N-acetyl-b-D-glucosaminidase in urine was a very sensitive biomarker of cadmium exposure.
Cadmium in urine or blood was not associated with a decreased glomerular filtration rate.

a r t i c l e i n f o

a b s t r a c t

Article history: Received 30 July 2015 Received in revised form 2 November 2015 Accepted 5 November 2015 Available online 2 January 2016

Cadmium (Cd), a well-known nephrotoxic agent, has received a great deal of attention from the Chinese public because of reports of its presence in rice. But very few studies have assessed the renal risk of Cd exposure in children. In this cross-sectional study, we aimed to determine whether biologic measures of Cd exposure were associated with biomarkers of early kidney damage in children, adolescents and adults. A total of 1235 subjects (2e86.8 years old) participated in this study and provided samples of blood and urine. As a result, the median urinary Cd level was 0.38 mg g1 creatinine in adult men and 0.42 mg g1 creatinine in adult women, similar to reference values observed in the United States (median: 0.32e0.40 mg L1 in adults). Multiple linear regressions showed Cd in urine to be significantly positively associated with effects on renal tubule biomarkers (as indicated by increased levels of N-acetyl-b-Dglucosaminidase and b2-microglobulin) after adjusting for age, body mass index, blood lead, and urinary density, in all age groups including children. We also found positive associations between blood Cd and renal tubule biomarkers in children. In conclusion, adverse tubular renal effects might have occurred at the current low Cd levels in the study population, including children. These findings are particularly relevant assessing health risks associated with low environmental exposures to Cd. © 2015 Elsevier Ltd. All rights reserved.

Handling Editor: A. Gies Keywords: Cadmium Tubular effects Glomerular effects General population Children

1. Introduction Chronic kidney disease has become an important public health problem in China (Zhang et al., 2012). Cadmium (Cd) is a nephrotoxic metal which is widespread in the environment (Jarup et al., 2000), and is regarded as a possible risk associated with chronic kidney disease (Zhang et al., 2012). Various biomarkers have been

* Corresponding author. ** Corresponding author. E-mail addresses: [email protected] (H. Sun), [email protected] (Y. Xu). 1 Hong Sun and Dongyue Wang contributed equally to this manuscript. http://dx.doi.org/10.1016/j.chemosphere.2015.11.069 0045-6535/© 2015 Elsevier Ltd. All rights reserved.

used to assess nephrotoxicity among Cd-exposed populations. Elevations in enzymes primarily of renal tubular origin, such as Nacetyl-b-D-glucosaminidase (NAG), have been observed at low levels of environmental Cd exposure in the United States (Noonan et al., 2002). Increases in these enzymes have been associated with chemical-induced renal tubular damage (Goren et al., 1987). Elevations in the excretion of low-molecular-weight proteins, such as b2-microglobulin (BMG), have been used as indicators of damage to the tubular protein absorption capability (Noonan et al., 2002). Low-molecular-weight proteinuria among exposed workers with >10 mg urinary Cd g1 creatinine was irreversible and exacerbated the age-related decline in the glomerular filtration rate (GFR) (Roels et al., 1989). A combination of these biomarkers, which could reflect

4

D. Wang et al. / Chemosphere 147 (2016) 3e8

both tubular and glomerular kidney effects, has been used in a previous study (Akesson et al., 2005). Most studies on the effect of low Cd exposure have focused on adults and especially on older adults (Akesson et al., 2005; Weaver et al., 2014), when the accumulation of Cd in the kidney had reached its peak. Many recent studies have reported Cd exposure in children (Alomary et al., 2013) and even in infants (Sun et al., 2014). However, very few studies assess the renal risk of Cd exposure in children, who are known to absorb metals more readily than adults and are particularly sensitive for biological and developmental reasons (Fels et al., 1998). Furthermore, presently there is no margin between the Cd exposure level in children and the concentration considered to increase the risk of renal tubular damage. Taken together, it was necessary to include children in the study population (de Burbure et al., 2006). The aim of the present investigation was to study the association between low-level Cd concentrations in the body and a series of markers of tubular and glomerular function in a wide age spectrum of the general population, to assess whether the current level of Cd exposure may be of public health concern. The study was conducted in an area without significant industrial Cd pollution. 2. Materials and methods 2.1. Study site and populations This study was part of Jiangsu Metal and Health Survey (JMHS), performed in Changshu City, located in east Jiangsu Province in China, which covers a total area of 1264 Km2 with a residential population of 1.5 million at the end of 2013. We selected an urban community, Changshu City, and five surrounding rural communities as our study area. For each community, 210 subjects were invited to participate in the study, with ages classified as <10, <18, <30, <40, <50, <60, <70 and >70 years. For each age group, we invited 23 to 24 subjects, evenly divided between men and women. The study population in Changshu City comprised local residents aged 2 to 86.8 who had resided at their current address for at least 2 years. A total of 1235 subjects provided signed informed consent and participated this study. The sample collection was conducted from May 2013 to January 2014. Morning urinary samples (20 ml) were collected at 7 ame8 am and stored in polyacrylamide bottles. Venous blood samples were obtained from the arm of each child in the residential community by registered physicians. Blood samples (~2 mL) were drawn into vacutainers containing K2 EDTA (BD, Franklin Lakes, NJ) and shipped in dry ice to the laboratory of the Jiangsu Provincial Center for Disease Prevention and Control (CDC) for analysis. The weight and height of each participant with light clothing and no shoes or hats were recorded. We designed a structured questionnaire to capture the metal exposure and associated risk factors. The interview questions were divided into three parts: part 1, demographic information and family socioeconomic levels; part 2, food, nutrition, and living habits; and part 3, house and living environment. This study design was approved by the institutional review board of the Jiangsu Provincial CDC. All participants provided written, informed consent. 2.2. Urinary Cd and blood Cd and lead The blood samples were obtained from the antecubital vein and collected into vacutainer tubes (2 mL purple top EDTA tubes) for Cd and lead (Pb) measurements (Pb levels were determined in order to find out whether Pb in the investigated blood samples can induce toxic renal effects, besides Cd). The concentrations of Cd and Pb in

the collected blood samples and of Cd in the urine samples were measured with an inductively coupled plasma mass spectrometer (ICP-MS; Thermo fisher X-series 2, Houston, TX, USA) using a previously described operating method (Sun et al., 2014). The limit of detection was 0.025 mg L1 for blood Cd, 0.75 mg L1 for blood lead and 0.02 mg L1 for urinary Cd. A total of 25 (25/1235) and 1 (1/ 1235) participants had blood Cd and lead levels below the limit of detection. For those participants, we assigned a level equal to the limit of detection divided by the square root of 2. For internal quality assurance and control, the Seronorm™ Trace Elements Whole Blood Level-1 (SERO, Norway) standard reference materials were used. The observed values for each element were within the certified range. The recovery was 92% for BeCd, and 85% for UeCd. The relative standard deviation (RSD) for both BeCd and UeCd were less than 5%. Quality control and assurance procedures included standard reference samples and duplicate detection. The error was less than 8% and the relative standard deviation was within 5% for all samples. 2.3. Measurement of kidney outcomes Urine samples were analyzed for two biomarkers of tubular damage (NAG and BMG) and for creatinine. Analysis was conducted by the Bio-chemical Laboratory of Changshu CDC, China. We determined NAG in urine by a colorimetric method and determined BMG by a latex enhanced immune-turbidimetric method. We obtained both assay kits from Gcell (Jiuqiang Co. Beijing China) and used an auto-analyzer (model 7180; Hitachi, Tokyo, Japan). We measured serum and urinary creatinine with the kinetic Jaffe method using an auto-analyzer (model 7180; Hitachi, Tokyo, Japan). We used the Modification of Diet in Renal Disease (MDRD) study equation to estimate the glomerular filtration rate (eGFR) as an indicator of glomerular function: eGFR (milliliters per minute per 1.73 m2) ¼ 175  (serum creatinine)1.234  (age)0.179 ( 0.79 if the individual was female). The number of participants with eGFR levels <60 mL min1 per 1.73 m2 was relatively small (12 men and 31 women). For this study, we categorized participants as having reduced eGFR if their eGFR levels were below the 25th percentile using sex-specific cutoffs (<71.43 mL min1 per 1.73 m2 for men and <64.83 mL min1 per 1.73 m2 for women). 2.4. Statistical analysis All biological parameters are reported as the median and interquartile range (IQR) and were log-transformed to approximate normal distribution. Urinary Cd (UeCd) and other biomarkers are expressed per liter of urine and per gram of creatinine. We categorized the subjects into three groups: children (age <12), adolescents (age 12e18) and adults (age  18). The adult group was further divided into six groups by age (as <30, <40, <50, <60, <70, <80, and >80). The level of UeCd in each age group was compared by ANOVA analysis, with covariables including gender, age, blood Cd (BeCd), blood lead (BePb) and urinary creatinine (U-creatinine). The log UeCd was compared across gender with log U-creatinine as a covariable by ANOVA analysis followed by an Lsmeans post hoc test. Variations of urinary NAG (U-NAG) and urinary BMG (U-BMG) with UeCd were modeled using the natural cubic spline function, and we used the TRANSREG procedure of SAS to draw the cubic spline line. The models were run by stratifying the population according to gender. We assessed the relationship between log U-NAG or U-BMG and a set of independent variablesdlog UeCd, sex, age, log BeCd, log BePb, log U-creatinine, body mass index (BMI), family incomedin SAS using a Pearson's correlation and multiple linear regression model. At first, we used Pearson's correlation to select variables

D. Wang et al. / Chemosphere 147 (2016) 3e8

associated with log U-NAG or U-BMG. Then, we applied a stepwise forward selection to select covariables with a significance of 0.25 for a variable to enter and 0.10 to stay in the model. The condition index >10 was used as the indicator of the multi-colinearity. The underlying statistical assumptions about the homoscedasticity and normality of the errors were verified visually with regression residuals. Independence of the residuals was assessed by the DurbineWatson test. We used logistic regression to examine odds ratios of reduced eGFR associated with increased UeCd or BeCd levels (tertiles) in adults. Statistical models were progressively adjusted for age (continuous), gender (female/male), education (did not complete high school, high school diploma, higher education), BMI (continuous), smoking status (never/former/current), living region (urban/ rural), blood lead (log transformed), blood Cd (log transformed), and hypertension and diabetes status. 3. Results 3.1. Characteristics of subjects The population characteristic, exposure variables and kidney effect markers are summarized in Table 1, grouped by children, adolescents and adults. The age of the studied population ranged from 2.0 to 86.8 years. The average UeCd levels ranged from 0.12 mg L1 (in boys) to 0.41 mg L1 (in adult males) and from 0.14 mg g1 creatinine (in girls) to 0.42 mg g1 creatinine (in adult females). Average U-creatinine was significantly higher in adolescents than in children or adults (P < 0.01). In adults, the average Ucreatinine was much higher in men than in women. The overall BeCd was highest in adults and lowest in children; this distribution pattern was similar with UeCd. The BMG detected in males, expressed in mg L1 and mg g1 creatinine, was significantly higher than that detected in females (P < 0.01). 3.2. Association between excretion of UeCd and urinary protein With a natural cubic spline fitting model, Fig. 1 shows that UNAG increased with UeCd in both genders, and BMG showed a similar pattern of change (Fig. 1B). Further analysis in Table 2 shows

5

that excretion of UeCd (mg L1) was positively associated with UNAG (IU L1) in each age group, with variable correlation coefficients from 0.16 in children to 0.34 in adolescents, after adjusting with U-creatinine, BeCd, age and gender. U-creatinine was potently associated with U-NAG in all models for each age group. BeCd was weakly associated with U-NAG in children and adults, and this correlation was different among adult men and women. Excretion of UeCd was also associated with BMG (mg L1) in children and adult females (Table 2), and BeCd was positively associated with BMG in children. 3.3. BeCd, UeCd levels and eGFR Table 3 shows that the odds ratios for reduced eGFR (<71.43 mL min1 per 1.73 m2 in men and <64.83 mL min1 per 1.73 m2 in women) significantly increased between the lowest tertile of BeCd and second tertile in men (Model 1); however, after adjustment for sociodemographic factors, chronic kidney disease risk factors, blood lead levels, and urinary Cd levels, the odds ratios for reduced eGFR comparing the highest versus the lowest tertiles of BeCd were 1.37 (95% CI: 0.64, 2.94) in men and 0.91 (0.47, 1.74) in women (Table 3, Model 2). No significant differences were observed. Similar results of the odds ratios for reduced eGFR between UeCd tertiles are also shown in Table 4. The fully adjusted odds ratios for reduced eGFR comparing the highest versus lowest UeCd tertiles were 0.58 (0.29, 1.13) in men and 0.59 (0.31, 1.12) in women (Table 4). 4. Discussion The UeCd levels we reported here were much lower than other reports in China (Cui et al., 2005; Zhang et al., 2014) but were very similar to those reported in the United States (median: 0.32e0.40 mg L1 in adults) (USCDC, 2009) and in Canada (median: 0.24e0.39 mg L1 in adults) (Canada, 2010). These UeCd levels were only half of those observed in South Korea (median: 0.66 mg L1 for male, 0.73 mg L1 for female) (Lee et al., 2012) and were much lower than those in the general population in Japan (GM: 1.3 mg g1 creatinine) (Suwazono et al., 2011). These comparisons indicate

Table 1 The characteristics of subjects by age group. Children (<12)

Population characteristic N Age (years)a BMI(kg (m2)1) Smoke (never/former/current) Diabetes mellitus n (%) Hypertension n (%) Exposure variablesb U-Creatinine (g L1) UeCd (mg L1) UeCd (mg g1 creatinine) BeCd (mg L1) BePb (mg L1) Kidney effect markersb U-NAG (IU L1) U-NAG(IU g1 creatinine) U-BMG (mg L1) U-BMG(mg g1 creatinine) eGFR (mL min1 per 1.73 m2) a b

Adolescents (12e18)

Adults (>¼18)

Boys

Girls

Boys

Girls

Men

Women

74 7.4 (4.1e11.7) 16.6 ± 3.4 74/0/0 0 0

92 7.9 (2e11.7) 16.7 ± 5.0 92/0/0 0 0

83 14.3 (12.0e17.8) 20.2 ± 3.8 82/0/1 0 0

90 14.4 (12.1e17.8) 18.7 ± 2.7 90/0/0 0 1 (1.1%)

445 47.6 (18.1e84.8) 23.4 ± 3.5 169/46/230 21 (4.7%) 77 (17.3%)

451 48.8 (18.1e86.8) 23.1 ± 3.8 444/0/7 23 (5.1%) 72 (16.0%)

0.77 (0.53e1.06) 0.12 (0.06e0.19) 0.15 (0.09e0.24) 0.20 (0.11e0.30) 37.41 (27.88e61.48)

0.78 (0.55e1.06) 0.12 (0.07e0.19) 0.14 (0.09e0.26) 0.19 (0.09e0.34) 39.83 (28.49e58.88)

1.4 (0.89e1.86) 0.19 (0.12e0.33) 0.16 (0.10e0.26) 0.27 (0.17e0.42) 36.35 (29.59e52.18)

1.27 (0.90e1.77) 0.19 (0.09e0.34) 0.15 (0.09e0.23) 0.30 (0.18e0.51) 30.62 (21.56e54.78)

1.11 (0.74e1.50) 0.41 (0.23e0.67) 0.38 (0.21e0.65) 1.34 (0.38e2.88) 44.96 (32.54e70.78)

0.78 (0.56e1.11) 0.33 (0.18e0.60) 0.42 (0.23e0.70) 0.49 (0.31e0.92) 39.55 (26.30e65.46)

7.48 9.45 0.12 0.15

7.04 8.78 0.11 0.13

9.27 7.40 0.18 0.13

7.77 6.24 0.11 0.08

10.43 (2.50, 64.60) 10.31 (1.46, 199.09) 0.33 (0.01, 15.52) 0.37 (0.00, 31.35) 84.89 (19.30, 204.34)

8.12 (1.40, 35.20) 10.09 (2.16, 48.63) 0.18 (0.01, 8.26) 0.28 (0.00, 20.90) 78.56 (19.32, 239.60)

(1.50, (3.11, (0.01, (0.00,

32.00) 37.28) 0.52) 0.59)

Values are arithmetic mean (minimum, maximum) Values are median (P25, P75).

(1.40,21.00) (2.64, 30.46) (0.01, 0.72) (0.00, 0.51)

(2.10, (2.21, (0.01, (0.00,

29.20) 32.35) 1.33) 0.88)

(1.60, 30.10) (1.84, 48.98) (0.01,15.52) (0.00, 0.33)

6

D. Wang et al. / Chemosphere 147 (2016) 3e8

Fig. 1. Association of individual urinary cadmium (UeCd) with urinary N-acetyl-b-D-glucosaminidase (U-NAG) in mg L1 (A) and urinary b2-microglobin (U-BMG) in mg L1 (B), according to gender, in both participants (n ¼ 1235). The data were fitted using natural cubic splines.

that our study population had low environmental Cd exposure. We also found the Pb exposure was not high in the study population (Table 1). We measured blood lead as a possible confounding factor to the renal effects of Cd, but statistical analysis showed that blood lead had no effect on the renal effect (Table 2). Fig. 1 and Table 2 show a very close linear correlation between U-NAG/U-BMG and UeCd in both genders, which indicate that Cd exposure might affect renal function at different levels, since UNAG and U-BMG are biomarkers that represent different renal tubular dysfunctions. NAG, an enzyme localized in the lysosomes of the tubular cells, is a sensitive marker of leakage from damaged tubular cells (Moriguchi et al., 2009). It is well known that the Cdmetallothionein complex is reabsorbed by renal tubules cell and then rapidly degraded in lysosomes, and the released Cd is retained in the tubules cell and then rapidly bound to the metallothionein

synthesized in the kidney. The damage of renal tubular cells induced by Cd is probably due to the unbound Cd in the cells (Nawrot et al., 2010), and then causes leakage of NAG. This can reasonably explain the correlation between UeCd and U-NAG. More importantly, even in children with an average UeCd of 0.12 mg L1, the association was still significant between UeCd/ BeCd and U-NAG. If there is a causality relationship between UeCd and U-NAG as suggested by Akesson et al. (2005), our results implicate that lower levels of Cd that could possibly cause renal damage. BMG, a low-molecular weight protein, is a valid marker of the tubular reabsorption (Nordberg, 2010). We also observed the association between UeCd and U-BMG in children and female adults (Table 2). The presence of BMG in urine indicated compromised tubular reabsorption function. Thus, Cd might also be associated with decreased tubular reabsorption in these subpopulation

D. Wang et al. / Chemosphere 147 (2016) 3e8

7

Table 2 Factors associated with U-NAG and U-BMG in different age groups and gender. Age group

U-NAG (IU L1)

n

Variablea Children

U-BMG (mg L1)

b

p

D R2 0.49 0.39 0.06 0.04

166

Adolescents

U-creatinine UeCd BeCd

0.64 0.16 0.12

0.00** 0.00** 0.00**

U-creatinine UeCd Gender

0.52 0.34 0.17

0.00** 0.00** 0.01*

U-creatinine UeCd Age Gender

0.50 0.19 0.01 0.08

0.00** 0.00** 0.00** 0.01*

U-creatinine Age UeCd BeCd

0.50 0.01 0.19 0.05

0.00** 0.00** 0.00** 0.02*

U-creatinine UeCd Age BeCd

0.50 0.21 0.01 0.04

0.00** 0.00** 0.00** 0.03*

173

Adults All

0.46 0.37 0.07 0.02 0.38 0.27 0.07 0.03 0.01 0.30 0.20 0.08 0.02 0.01 0.38 0.27 0.08 0.02 0.00

896

Male

445

Female

451

Variablea

b

p

DR2

U-creatinine Age BeCd UeCd

1.35 0.13 0.17 0.38

0.00** 0.00** 0.02* 0.02*

U-creatinine UeCd

0.97 0.30

0.00** 0.09

UeCd U-creatinine Age Gender

0.21 0.73 0.02 0.3

0.01* 0.00** 0.00** 0.00**

U-creatinine Age

0.76 0.02

0.00** 0.00**

0.14 0.01 0.06 0.07 0.01 0.07 0.03 0.04

UeCd U-creatinine Age

0.32 0.74 0.02

0.00** 0.00** 0.00**

0.17 0.09 0.05 0.04

0.41 0.28 0.09 0.02 0.02 0.25 0.24 0.01

*P < 0.05, **P < 0.01. a Regression model used the following variables: log UeCd, sex, age, log BeCd, log BePb, log U-Creatinine, body mass index (BMI), and the detailed method was described in the Method section. Only significant factors were remained in the models.

Table 3 Odds ratios (95% CIs) for reduced eGFRa by blood cadmium level tertile. Blood cadmium level Men <0.50 0.50e2.30 >2.30 Women <0.35 0.35e0.69 >0.69

n

Cases/noncases

Model 1b

Model 2c

145 147 150

29/116 47/100 34/116

1 (reference) 1.88 (1.10,3.21)* 1.17 (0.67,2.05)

1 (reference) 1.52 (0.79,2.90) 1.37 (0.64,2.94)

146 146 150

34/112 43/103 34/116

1 (reference) 1.38 (0.82,2.32) 0.97 (0.56,1.66)

1 (reference) 1.50 (0.82,2.74) 0.91 (0.47,1.74)

Tertiles are based on distribution of blood cadmium in each gender. *P < 0.05. a eGFR<71.43 mL min1 per 1.73 m2 for adult man and eGFR< 64.83 mL min1 per 1.73 m2 for women. b Not adjusted (crude). c Adjusted for age, education, smoking status (never, former, current), blood lead (log mg L1), urinary cadmium.

Table 4 Odds ratios (95% CIs) for reduced eGFRa by urinary cadmium level tertile. Urinary cadmium Men <0.29 0.29e0.54 >0.54 Women <0.21 0.21e0.48 >0.48

n

Cases/noncases

Model 1b

Model 2c

145 146 151

38/107 35/111 37/114

1 (reference) 0.89 (0.71,1.77) 0.91 (0.54,1.54)

1 (reference) 0.70 (0.38,1.30) 0.58 (0.29,1.13)

145 146 151

43/102 38/108 40/111

1 (reference) 0.82 (0.67,1.54) 0.86 (0.52,1.42)

1 (reference) 0.65 (0.31,1.36) 0.59 (0.31,1.12)

Tertiles are based on distribution of blood cadmium in each gender. *P < 0.05. a eGFR<71.43 mL min1 per 1.73 m2 for adult man and eGFR< 64.83 mL min1 per 1.73 m2 for women. b Not adjusted (crude). c Adjusted for age, education, smoking status (never, former, current), blood lead and cadmium (log mg L1), urinary creatinine (log mg L1).

groups. As a possible reason, the Cd in tubular fluid might depress the reabsorption of BMG by tubular cell. According to the results on NAG and BMG levels, it can be concluded that Cd was associated

with renal tubular dysfunction in our study population, including children as well. We found that UeCd was a stable risk factor associated with UNAG and U-BMG in most of the age groups, while BeCd only showed effects in children with a low UeCd level. These results also indicated a much closer correlation of UeCd than BeCd with UNAG and U-BMG, which were consistent with a previous report (Ikeda et al., 2011). The possible reason was that kidney Cd might be the fundamental factor associated with U-NAG and U-BMG, and kidney Cd was associated with both BeCd and UeCd. Thus, BeCd levels were stable in the population aging 30 years, while the UeCd were increasing during a lifetime. On the other hand, UeCd was low and BeCd was increasing in children, so that kidney Cd was primarily affected by BeCd and BeCd influenced the levels of U-NAG and U-BMG. These results suggest that BeCd should be also considered when using UeCd as a biomarker of Cd in children. A recent study reported a high prevalence of chronic kidney disease in China (Zhang et al., 2012), and Cd was suspected as an important environmental pollutant associated with renal dysfunction (Nordberg et al., 2009). In the present study, we did not

8

D. Wang et al. / Chemosphere 147 (2016) 3e8

find an association between Cd in urine or blood and reduced eGFR indicating no correlation between glomerular damage and Cd exposure in the study population. Our result was very similar to the Korean ones, although Korean report showed that the highest tertiles of blood Cd might be a risk factor in women (Hwangbo et al., 2011). However, our results were not consistent with a previous report based on a national health and nutrition examination survey (Navas-Acien et al., 2009) that used different cut-off value for defining reduced eGFR. The relatively smaller sample size in the current study might have also limited our ability to find the association. This study has several methodological advantages, including the relatively large sample size covering a broad age spectrum, high participation rate, sample analysis with high analytical accuracy, inclusion of both BeCd and UeCd, and several different outcomes of renal effects. There are some limitations to our study. First, a single measure of random urinary Cd can vary a great deal, so an individual's urinary Cd level might not be an accurate reflection of body burden (Akerstrom et al., 2013). Further study will determine the association of Cd in 24 h urine with Cd body burden. Second, as a common problem in the interpretation of data from a crosssectional study, the exposure is measured at the same time as the effects, which may not be the etiologically relevant period. This may be problematic for BeCd because it largely reflects recent exposure. Third, we use urinary creatinine to adjust the dilution of urine in the regression model, but creatinine levels vary with age, sex and body mass (Noonan et al., 2002), and we also observe the increase of U-creatinine in adolescent (Table 1), so we analyze the data by age groups to diminish the effect of varied U-creatinine in different age stages. Forth, in the case of adults, concluding that glomerular damage is not increased by Cd would require more data (specific biomarkers such as micro-albumin) to overcome the limitations of calculations relying on serum creatinine only. In conclusion, this study demonstrated that even exposed to low levels of Cd, possible tubular effects were already observed in general population including children. NAG and BMG in urine were sensitive biomarkers for tubular dysfunction. The obtained eGFR indicate no glomerular kidney effects at the same Cd exposure levels. These findings may have important public health implications for the evaluation of environmental Cd exposure and selection of the sensitive biomarkers, and are particularly relevant for epidemiological studies of health risks associated with low environmental exposures to Cd. Acknowledgments: The authors are grateful for the valued help of the staff at Changshu CDC who diligently worked throughout the sampling collection process and measurement. The study was supported by National Health Research Special Funds (No. 201002001) and health research funds from Jiangsu Provincial Department of Health (No. LJ201129). References Akerstrom, M., Sallsten, G., Lundh, T., Barregard, L., 2013. Associations between urinary excretion of cadmium and proteins in a nonsmoking population: renal toxicity or normal physiology? Environ. Health Perspect. 121, 187e191. Akesson, A., Lundh, T., Vahter, M., Bjellerup, P., Lidfeldt, J., Nerbrand, C., Samsioe, G., Stromberg, U., Skerfving, S., 2005. Tubular and glomerular kidney effects in Swedish women with low environmental cadmium exposure. Environ. Health Perspect. 113, 1627e1631. Alomary, A., Al-Momani, I.F., Obeidat, S.M., Massadeh, A.M., 2013. Levels of lead, cadmium, copper, iron, and zinc in deciduous teeth of children living in Irbid, Jordan by ICP-OES: some factors affecting their concentrations. Environ. Monit.

Assess. 185, 3283e3295. Canada, H., 2010. Report on Human Biomonitoring of Environmental Chemicals in Canada: Results of the Canadian Health Measures Survey Cycle 1 (2007e2009). Cui, Y., Zhu, Y.G., Zhai, R., Huang, Y., Qiu, Y., Liang, J., 2005. Exposure to metal mixtures and human health impacts in a contaminated area in Nanning, China. Environ. Int. 31, 784e790. de Burbure, C., Buchet, J.P., Leroyer, A., Nisse, C., Haguenoer, J.M., Mutti, A., Smerhovsky, Z., Cikrt, M., Trzcinka-Ochocka, M., Razniewska, G., Jakubowski, M., Bernard, A., 2006. Renal and neurologic effects of cadmium, lead, mercury, and arsenic in children: evidence of early effects and multiple interactions at environmental exposure levels. Environ. Health Perspect. 114, 584e590. Fels, L.M., Wunsch, M., Baranowski, J., Norska-Borowka, I., Price, R.G., Taylor, S.A., Patel, S., De Broe, M., Elsevier, M.M., Lauwerys, R., Roels, H., Bernard, A., Mutti, A., Gelpi, E., Rosello, J., Stolte, H., 1998. Adverse effects of chronic low level lead exposure on kidney functionea risk group study in children. Nephrol. Dial. Transpl.: Off. Publ. Eur. Dial. Transpl. Assoc. e Eur. Ren. Assoc. 13, 2248e2256. Goren, M.P., Wright, R.K., Horowitz, M.E., Crom, W.R., Meyer, W.H., 1987. Urinary Nacetyl-beta-D-glucosaminidase and serum creatinine concentrations predict impaired excretion of methotrexate. J. Clin. Oncol. 5, 804e810. Hwangbo, Y., Weaver, V.M., Tellez-Plaza, M., Guallar, E., Lee, B.K., Navas-Acien, A., 2011. Blood cadmium and estimated glomerular filtration rate in Korean adults. Environ. Health Perspect. 119, 1800e1805. Ikeda, M., Ohashi, F., Fukui, Y., Sakuragi, S., Moriguchi, J., 2011. Closer correlation of cadmium in urine than that of cadmium in blood with tubular dysfunction markers in urine among general women populations in Japan. Int. Arch. Occup. Environ. Health 84, 121e129. Jarup, L., Hellstrom, L., Alfven, T., Carlsson, M.D., Grubb, A., Persson, B., Pettersson, C., Spang, G., Schutz, A., Elinder, C.G., 2000. Low level exposure to cadmium and early kidney damage: the OSCAR study. Occup. Environ. Med. 57, 668e672. Lee, J.W., Lee, C.K., Moon, C.S., Choi, I.J., Lee, K.J., Yi, S.-M., Jang, B.-K., Yoon, B.J., Kim, D.S., Peak, D., Sul, D., Oh, E., Im, H., Kang, H.S., Kim, J., Lee, J.-T., Kim, K., Park, K.L., Ahn, R., Park, S.H., Kim, S.C., Park, C.-H., Lee, J.H., 2012. Korea National Survey for Environmental Pollutants in the Human Body 2008: heavy metals in the blood or urine of the Korean population. Int. J. Hyg. Environ. Health 215, 449e457. Moriguchi, J., Inoue, Y., Kamiyama, S., Horiguchi, M., Murata, K., Sakuragi, S., Fukui, Y., Ohashi, F., Ikeda, M., 2009. N-Acetyl-b-d-glucosaminidase (NAG) as the most sensitive marker of tubular dysfunction for monitoring residents in non-polluted areas. Toxicol. Lett. 190, 1e8. Navas-Acien, A., Tellez-Plaza, M., Guallar, E., Muntner, P., Silbergeld, E., Jaar, B., Weaver, V., 2009. Blood cadmium and lead and chronic kidney disease in US adults: a joint analysis. Am. J. Epidemiol. 170, 1156e1164. Nawrot, T.S., Staessen, J.A., Roels, H.A., Munters, E., Cuypers, A., Richart, T., Ruttens, A., Smeets, K., Clijsters, H., Vangronsveld, J., 2010. Cadmium exposure in the population: from health risks to strategies of prevention. Biometals 23, 769e782. Noonan, C.W., Sarasua, S.M., Campagna, D., Kathman, S.J., Lybarger, J.A., Mueller, P.W., 2002. Effects of exposure to low levels of environmental cadmium on renal biomarkers. Environ. Health Perspect. 110, 151e155. Nordberg, G.F., 2010. Biomarkers of exposure, effects and susceptibility in humans and their application in studies of interactions among metals in China. Toxicol. Lett. 192, 45e49. Nordberg, G.F., Jin, T., Wu, X., Lu, J., Chen, L., Lei, L., Hong, F., Nordberg, M., 2009. Prevalence of kidney dysfunction in humans e relationship to cadmium dose, metallothionein, immunological and metabolic factors. Biochimie 91, 1282e1285. Roels, H.A., Lauwerys, R.R., Buchet, J.P., Bernard, A.M., Vos, A., Oversteyns, M., 1989. Health significance of cadmium induced renal dysfunction: a five year follow up. Br. J. Ind. Med. 46, 755e764. Sun, H., Chen, W., Wang, D., Jin, Y., Chen, X., Xu, Y., 2014. The effects of prenatal exposure to low-level cadmium, lead and selenium on birth outcomes. Chemosphere 108, 33e39. Suwazono, Y., Nogawa, K., Uetani, M., Kido, T., Nakagawa, H., 2011. Reassessment of the threshold of urinary cadmium by using hybrid approach in a cadmium nonpolluted area in Japan. Int. J. Hyg. Environ. Health 214, 175e178. USCDC, 2009. Fourth national report on human exposure to environmental chemicals. Atlanta (GA). Weaver, V.M., Vargas, G.G., Silbergeld, E.K., Rothenberg, S.J., Fadrowski, J.J., RubioAndrade, M., Parsons, P.J., Steuerwald, A.J., Navas-Acien, A., Guallar, E., 2014. Impact of urine concentration adjustment method on associations between urine metals and estimated glomerular filtration rates (eGFR) in adolescents. Environ. Res. 132, 226e232. Zhang, L., Wang, F., Wang, L., Wang, W., Liu, B., Liu, J., Chen, M., He, Q., Liao, Y., Yu, X., Chen, N., Zhang, J.E., Hu, Z., Liu, F., Hong, D., Ma, L., Liu, H., Zhou, X., Chen, J., Pan, L., Chen, W., Wang, W., Li, X., Wang, H., 2012. Prevalence of chronic kidney disease in China: a cross-sectional survey. Lancet 379, 815e822. Zhang, W.-L., Du, Y., Zhai, M.-M., Shang, Q., 2014. Cadmium exposure and its health effects: a 19-year follow-up study of a polluted area in China. Sci. Total Environ. 470, 224e228.