The association between total urinary arsenic concentration and renal dysfunction in a community-based population from central Taiwan

Other uses, including reproduction and distribution, or selling or

licensing copies, or posting to personal, institutional or third party

websites are prohibited.

In most cases authors are permitted to post their version of the

article (e.g. in Word or Tex form) to their personal website or

institutional repository. Authors requiring further information

regarding Elsevier’s archiving and manuscript policies are

encouraged to visit:

The association between total urinary arsenic concentration and renal

dysfunction in a community-based population from central Taiwan

Jein-Wen Chen

, Hsiao-Yen Chen

, Wan-Fen Li

, Saou-Hsing Liou

, Chien-Jen Chen

, Jhuo-Han Wu

,

Shu-Li Wang

a_{Division of Environmental Health and Occupational Medicine, National Health Research Institutes (NHRI), Miaoli 350, Taiwan} b_{Genomic Research Center, Academia Sinica, Taipei, Taiwan}

c

Graduate Institute of Epidemiology, College of Public Health, National Taiwan University, Taipei, Taiwan d

Institute of Environmental Medicine, College of Public Health, China Medical University Hospital, Taichung, Taiwan

a r t i c l e

i n f o

Article history: Received 9 August 2010

Received in revised form 25 January 2011 Accepted 28 February 2011

Available online 1 April 2011 Keywords:

Arsenic Renal dysfunction Type 2 diabetes mellitus Glomerular filtration rate Environmental epidemiology

a b s t r a c t

Arsenic (As) is an important environmental toxicant that can cause cancer and cardiovascular disease, but the relationship between As exposure and renal dysfunction is not clear. The aim of this study is to exam-ine the association between As exposure and renal dysfunction in a community-based population in cen-tral Taiwan. One thousand and forty-three subjects were recruited between 2002 and 2005. The risk for type 2 diabetes was increased by 2-fold (p < 0.05) in subjects with total urinary As (U-As) > 75lg g1

cre-atinine as compared with subjects whose U-As was 635lg g1_{creatinine after the adjustment for}

poten-tial confounders. The adjusted odds ratio for an abnormal b2 microglobulin (B2MG > 0.154 mg L1_{) was}

significantly higher in subjects with U-As > 35lg g1_{creatinine as compared with the reference group}

adjusted for age, sex, living area, cigarette smoking, diabetes, and hypertension. The risk for abnormal B2MG and estimated glomerular filtration rate (eGFR < 90 mL min1_{(1.73 m}2₎1_{) was both increased}

around 2-fold (p < 0.05) in subjects with U-As > 75lg g1_{creatinine as compared with those with}

U-As 6 35lg g1_{creatinine adjusted for all the risk factors plus lead (Pb), cadmium and nickel. The}

prev-alence of abnormal B2MG was 4.82 times higher in subjects with both over the median levels of U-As (85.1lg L1_{) and urinary Pb (18.9lg L}1_{) as compared to both lower than the median (p < 0.001). These}

results indicate that U-As might relate to renal dysfunction even other important risk factors were taken into account. Follow-up studies for causal inference are warranted.

Ó 2011 Elsevier Ltd. All rights reserved.

1. Introduction

Chronic arsenic (As) poisoning in the general population has been widely reported in many areas of the world and may cause life-threatening diseases (Yoshidaet al., 2004; Baig et al., 2009). As exposure commonly occurs through drinking water contami-nated with high amounts of inorganic forms of As (Smith et al., 2000; Kazi et al., 2009; Arain et al., 2009a). Environmental contam-ination with As and other metals is well documented in Taiwan. This study focused on Changhua County in central Taiwan because of its high As level. The study area had the highest contaminated level of As in groundwater based on a national survey in Taiwan between 2002 and 2005 (Taiwan-EPA, 2003–2006). Electroplating had seriously polluted agricultural soil over the last two decades

through the discharge of wastewater into irrigation ditches (Chang et al., 2006).

Exposure to inorganic As was found to be associated with increased rates of various chronic diseases, including diabetes (Navas-Acien et al., 2006), hypertension (Chen et al., 1995; Navas-Acien et al., 2005), metabolic syndrome (Wang et al., 2007), respiratory disorders (Arain et al., 2009a,b), and other macrovascular (Wang et al., 2003) and microvascular (Chiou et al., 2005) diseases related to diabetes. The prevalence of type 2 diabetes and renal disease were significantly increased in areas where there was increased exposure to As (Wang et al., 2003). However, little is understood regarding renal dysfunction caused by exposure to As as a function of diabetic status.Nordberg et al. (2005) assessed increased levels of urinary b2-microglobulin (B2MG), N-acetyl-h-glucosaminidase (NAG), and albumin (ALB) as markers of renal dysfunction, which was increased in response to As exposure (Nordberg et al., 2005). In a community-based study of high As levels in soil from Guizhou province in China, and the levels of B2MG, NAG, and ALB in residents from the polluted area were significantly higher than in residents from the

0045-6535/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.chemosphere.2011.02.091

⇑Corresponding author at: Division of Environmental Health and Occupational Medicine, National Health Research Institutes (NHRI), Miaoli 350, Taiwan. Tel.: +886 37 246 166x36509; fax: +886 37 587 406.

E-mail address:[email protected](S.-L. Wang).

Contents lists available atScienceDirect

Chemosphere

non-polluted areas (p < 0.01) (Hong et al., 2004). A recent report by Wang et al. (2009)indicated that chronic As exposure had a signif-icant adverse impact on kidney function of residents from the Xngjing Autonomous Region (Wang et al., 2009).

The chronic kidney disease (CKD) with an estimated glomerular filtration rate (eGFR) < 60 mL min1_{(1.73 m}2₎1_{in Taiwan is 11.9%.}

However, only 3.5% of patients were aware of their disorder (Wen et al., 2008). The present study aimed to examine the association between total urinary arsenic (U-As) exposure and renal dysfunc-tion in a community-based populadysfunc-tion. Total U-As is an established biomarker of inorganic As exposure in the absence of seafood in-take (Hughes, 2006; Navas-Acien et al., 2008). Renal function was evaluated by both B2MG and eGFR. GFR is commonly thought to be a good index of renal dysfunction (Levey et al., 1999) but is dif-ficult to measure in clinical practice. Therefore, an eGFR derived from serum creatinine level like most clinicians was used. The uri-nary levels of other metals that are known to be related to renal dysfunction such as lead (Pb), cadmium (Cd), and nickel (Ni) were also measured and their effects were considered in multivariable analyses.

2. Materials and methods 2.1. Study population

We have reported in our previous study about the study areas and subjects (Chang et al., 2006). Population census data obtained from the Ministry of Interior, Taiwan, were utilized to select the study population. Eligible subjects were aged 35–64 years and re-sided in their present household for at least 5 consecutive years at the time of recruitment. The density of metal-related factories was 4.2 factories km2 _{(2855 factories (686.95 km}2₎1_{) in Changhua}

County situated in central Taiwan (IDBMOEA, 2002). Fourteen townships were selected as study area during the period of 2002–2005 (Fig. 1). Of the 1086 stratified random sample by 3 age bands (35–44, 45–54, 55–64 years old) and 2 genders, 28 did not provide 5 ml sufficient amount of urine for As analysis, and

another 15 were rejected because their occupational and life-style records were incomplete. Thus, the final study group used for data analysis contained 1043 subjects.

2.2. Data and specimen collections

Prior to data collection, informed consent was obtained from each participant. The study protocol was approved by the Human Subjects Review Board of National Health Research Institutes in Taiwan. Well-trained interviewers used a structured questionnaire to record the data on demographic factors, residential and occupa-tional history from participants at home or a health care unit. Sub-jects were asked not to eat seafood for three days before the test, and venous blood was collected after P10-h fast. Subjects received health examinations at local health care units, where fasting blood and urine samples were collected. Subjects also received a general health checkup, including blood pressure measurements, from a general/internist physician. Urine and plasma specimens were stored at 4 °C immediately after collection and then be analyzed within 3 h. Fasting plasma glucose and lipids were quantified on the same day in the central laboratory of Changhua Christian Hos-pital, a medical center that has served local residents for several years.

2.3. Sample analysis

U-As concentrations were measured by an ELAN 6100 induc-tively coupled plasma-mass spectrometry (ICP-MS, Perkin Elmer, Shelton, CT, USA). Using ICP-MS normally generate better sensitivity with lower limit of detection than atomic absorption spectrometry (Montaser, 1998). A standard reference material, MEDICHEM Metal-le U Level 1 (Daignostica and Verfahrensentwicklung Kringserasse, Steinenbronn, Germany), was used for quality control evaluation. The method of detection limit (MDL) of As, Pb, Cd, and Ni were 0.045, 0.038, 0.063, and 0.066

l

3.2% (for As), 2.9% (for Pb), 5.8% (for Cd), and 4.7% (for Ni) of samples that had concentrations below the MDL. When the concentrations were below MDL, samples were assigned a value of 50% of the MDL which is a conventional approach and also adopted by NHANES III (The Third National Health and Nutrition Examination Survey) (Paschal et al., 2000). Fasting plasma samples were analyzed in the central laboratory of Changhua Christian Hospital for concentra-tions of glucose, triglycerides, cholesterol, and low- and high-density lipoproteins using a Beckmen SYNCHRON LX20 System (Beckmen Coulter, CA, USA). Urinary B2MG and urinary creatinine levels were determined in the same medical laboratory. B2MG lev-els were measured by MEIA (microparticle enzyme immunoassay) and creatinine levels were measured by the Modified Jaffe Method (CREA-HR-1, Wako Pure Chemical Industries, Osaka, Japan). 2.4. Indices of renal function

The B2MG, creatinine concentrations and eGFR were used to as-sess renal function. The formula developed by the ‘‘Modification of Diet in Renal Disease (MDRD) Study Group’’ is widely used to esti-mate GFR from serum creatinine measurements for systematically evaluating patients with chronic renal disease (Levey et al., 1999). The eGFR equation was:

eGFR ¼ 186  serum creatinine1:154

 age0:203_{ð0:742 if femaleÞ ðLevey et al:;1999Þ ð1Þ}

In clinical definition, chronic kidney disease was defined as kid-ney damage or GFR < 60 mL min1_{(1.73 m}2₎1 _{for 3 months or}

more (Levey et al., 2005); however, present study did not had more than one GFR estimate in different times due to the limitation of

Fig. 1. The location of study areas. A–N were the towns recruited subjects, in which A: Changhua City, B: Hemei, C: Lugang, D: Sioushuei, E: Huatan, F: Hsicou, G: Ershuei, H: Fangyuan, I: Dacheng, J: Erhlin, K: Chutang, L: Tienwei, M: Beidou, and N: Fenyuan Town.

cross-sectional study. In order to detect early signs of renal dys-function associated with As, 90 mL min1_{(1.73 m}2₎1_{was chosen}

as another cut-off value. Abnormal B2MG was defined as B2MG > 0.154 mg L1_{according to the criteria of the medical}

labo-ratory of Changhua Christian Hospital. In addition, diabetes melli-tus (DM) was defined as a fasting plasma glucose concentration P126 mg dL1_{or treatment with diabetic therapy. Hypertension}

was defined as a diastolic blood pressure P90 or systolic blood pressure P140 mm Hg, or on anti-hypertension therapy.

2.5. Statistical analyses

The data of the U-As, urinary Pb (U-Pb), Cd (U-Cd), and Ni (U-Ni) concentrations were not normally distributed and skewed to the right. Thus, geometric means (GM) along with 95% confidence intervals were calculated to prevent over-estimation from using arithmetic means. When the log transformed data were normally distributed, Student’s t-test or analysis of variance (ANOVA) was utilized to test difference among the groups. For transformed data that were not normally distributed, nonparametric methods were employed. If concentrations of metals below the MDL, the concen-trations were recorded to 1/2  MDL in the data analyses (Paschal et al., 2000). Associations between categorical variables were tested using a Chi-square test. Kendall’s tau-c trend test was used for trend analysis of demographic factors, lifestyles, and renal dys-function biomarkers by U-As. The American Conference of Govern-mental Industrial Hygienists (ACGIH) recommendation for the biological exposure index of U-As concentration is 35

l

(ACGIH, 1996). The cut-off of U-As level suggested by ACGIH was used to examine the association between U-As levels and renal function indices. Univariate and multivariate logistic regression analyses were performed to predict the presence of renal dysfunc-tion and diabetes based on U-As level. For smoking status, people smoke at least 1 cigarette per day more than six months was

defined as cigarette smoker. When fasting plasma glucose P126 mg dL1_{or on diabetes therapy, it was defined as diabetes,}

and if systolic blood pressure P90 or diastolic blood pressure P140 mm Hg or on anti-hypertension therapy was defined as hypertension. All analyses were performed using SPSS version 15.0.

3. Results

Table 1shows the demographic factors and related biochemical variables by tertiles of U-As concentration in the study population. B2MG, U-Pb, and U-Cd levels increased along with increased ter-tiles of U-As concentrations. Smoking status was also found posi-tively associated with U-As level. While, eGFR decreased with increasing U-As concentration.Table 2shows the demographic fac-tors and related biochemical variables according to B2MG concen-tration and eGFR. Blood pressure, fasting plasma glucose, and serum creatinine were significantly higher in the groups with abnormal renal function indices (B2MG > 0.154 mg L1_{and eGFR <}

60 mL min1_{(1.73 m}2₎1_{) as compared with the normal group}

after adjustment for age. A significant association was also found between renal function indices (B2MG and eGFR) and diabetes or hypertension status.

There is no normal range of U-As level recommended for general populations. Therefore, we used the biological exposure index of U-As (35

l

following analyses.Table 3shows the results of multiple logistic regression analyses for diabetes by groups of various U-As levels, with subjects whose levels were 635

l

group. The adjusted odds ratios (ORs) for diabetes mellitus (DM) showed an increasing trend according to As levels using ACGIH cut-off points (Table 3). The ORs for DM were 2.08 and 2.22 for U-As levels of >75–200 and >200

l

compared with U-As levels 635

l

Table 1

Demographic factors and related biochemical variables by tertiles of urinary arsenic (U-As) concentration. Characteristics (N = 1043) U-As (lg L1 ) p-Valuea <61.66 (n = 341) 61.66–112.2 (n = 371) >112.2 (n = 331) Continuous variables Age (y)b 51.2 ± 8.5 50.4 ± 8.6 50.7 ± 8.4 0.37 Body mass index (kg m2

)b

24.8 ± 3.5 25.3 ± 3.5 24.9 ± 3.3 0.08 Fasting plasma glucose (mg dL1

)b

102.0 ± 33.8 102.6 ± 31.5 102.4 ± 31.7 0.97 Systolic blood pressure (mm Hg)b _{130.6 ± 21.2 130.7 ± 22.0 130.9 ± 21.1 0.97}

b2 microglobulin (mg L1₎c _{0.2 (0.08–0.41) 0.3 (0.02–0.7) 0.5 (0.03–0.9) 0.04}*,e

Serum creatinine (mg dL1₎b

0.9 ± 0.7 0.9 ± 0.2 1.0 ± 0.6 0.11 Estimated glomerular filtration rate (eGFR, mL min1

(1.73m2 )1 )b,d 79.4 ± 15.2 77.4 ± 12.9 73.6 ± 11.9 0.02*,e Urinary lead (lg L1 )e 17.1 (15.9–18.5) 17.7 (16.3–19.2) 23.6 (21.5–25.8) 0.002**,e,f Urinary cadmium (lg L1 )e 0.9 (0.8–1.0) 0.9 (0.8–1.1) 1.1 (1.0–1.3) 0.007**,e,f Urinary nickel (lg L1 )e 4.3 (3.9–4.7) 4.3 (3.9–4.7) 4.5 (4.2–4.9) 0.33 Percentage (n) Categorical variables Male (%) 37.2% (127) 47.2% (175) 56.5% (187) <0.001*** Cigarette smoker (%)g 14.4% (49) 20.5% (76) 29.6% (98) <0.001*** Diabetes (%)h 7.9% (27) 8.6% (32) 8.5% (28) 0.10 Hypertension (%)i 41.3% (141) 41.2% (153) 42.9% (142) 0.22 a_{ANOVA with Scheffe post hoc test for continuous variables, Kendall’s tau-c trend test for categorical variables.}

b_{Mean ± SD or geometric mean ± geometric SD.} c

Geometric mean (95% confidence interval). d

Estimated glomerular filtration rate (eGFR, mL min1_{(1.73 m}2

)1_{) = 186  (serum creatinine (mg dL}1₎₎1.154_{ age (y)}0.203_{(0.742 if female).} e

1st tertile vs. 3rd tertile. f

2nd tertile vs. 3rd tertile. g

Cigarette smoker: at least 1 cigarette per day more than 6 months. h

diabetes: fasting plasma glucose P126 mg dL1

or on diabetes therapy. i

hypertension: systolic and diastolic blood pressure P90 or 140 mm Hg, respectively, or on anti-hypertension therapy. *

p < 0.05. **_{p < 0.01.} *** _{p < 0.001.}

for age, sex, living area, cigarette smoking, hypertension, Pb, Cd, and Ni.Table 4shows the results of multiple logistic regression analysis using three different regression models for an abnormal B2MG (>0.154 mg L1_{) and eGFR (<60 and <90 mL min}1_{(1.73 m}2₎1_are

both shown) based on U-As levels. The ORs for an abnormal B2MG excretion were 1.69, 2.11, and 2.04 for U-As levels of >35–75, >75– 200, and >200

l

groups with B2MG excretions 635

l

adjust-ment for age, sex, living area, cigarette smoking, diabetes, hyperten-sion status, Pb, Cd, and Ni. When considering a similar approach by non-adjusted U-As concentration (

l

as previous ones. Subjects with U-As levels >75

l

had a significantly higher OR of eGFR < 90 mL min1_{(1.73 m}2

)1

than those with U-As levels 635

l

eGFR < 60 mL min1_{(1.73 m}2₎1_{was considered as the cut-off value}

for renal dysfunction, subjects with U-As levels >200

l

creati-nine had a significantly higher OR of renal dysfunction than those with U-As levels 635

l

preva-lence of abnormal B2MG was 4.82 times higher in subjects with both

over the median levels of U-As (85.1

l

_l

as compared to both lower than the median (p < 0.001). 4. Discussion

The association between U-As and renal dysfunction found in this study is consistent with the results of a cross-sectional study in an area of high exposure to As in Guizhou, China (Nordberg et al., 2005). They reported the combined exposure of individuals to inorganic As and Cd resulted in a higher prevalence of renal dys-function than the separate exposure to either metal.Hsueh et al. (2009)reported that total U-As was associated significantly with CKD in a dose–response relationship, especially with a total As le-vel than 20.7 compared with 11.8

l

U-As is normally used as biomarker of current As exposure, the present study focused on total U-As levels with a consideration of other metals such as Pb, Cd, and Ni. However, multiple logistic regression analysis with the urinary levels of other metals entered as covariates showed that only As was significantly associated with

Table 3

Odds ratios (ORs) for diabetes mellitus according to American Conference of Governmental Industrial Hygienists (ACGIH) cut-off points of urinary arsenic (U-As) concentration (n = 1043).

U-As (lg g1

creatinine) With diabetes mellitus [n (percentage)]a,e

ORsb ORsc ORsd,e 635 (n = 133) 6 (4.5%) 1 1 1 >35–75 (n = 315) 26 (8.3%) 2.13 (0.86–3.39) 1.97 (0.69–2.67) 1.95 (0.56–2.66) >75–200 (n = 484) 45 (9.3%) 2.30 (1.32–4.89)* 2.11 (1.31–4.16)* 2.08 (1.05–3.69)* >200 (n = 111) 10 (9.8%) 2.77 (1.42–5.27)q** _{2.34 (1.38–4.67)}* _{2.22 (1.21–4.09)}* a

Diabetes mellitus: fasting plasma glucose =126 mg dL1_{or on diabetes therapy.} b

Adjusted by age, sex, living area, and cigarette smoking. c

Adjusted by age, sex, living area, cigarette smoking, and hypertension. d

Adjusted by age, sex, living area, cigarette smoking, hypertension, lead, cadmium, and nickel. e

Trend test for increasing diabetes prevalence with U-As level, p = 0.07 by X2

, and for ORsd

, p = 0.09 by logistic regression. *_{p < 0.05.}

** p < 0.01. Table 2

Demographic factors and related biochemical variables according to renal dysfunction indices (b2 microglobulin (B2MG) and estimated glomerular filtration rate (eGFR)). Characteristics (N = 1043) B2MG p-Valuea,b eGFR p-Valuea,b (>0.154 mg L1 , n = 249) (50.154 mg L1 , n = 794) (<60 mL min1 (1.73 m2 )1 , n = 94) (=60 mL min1 (1.73 m2 )1 , n = 949) Continuous variables Age (y)c _{53.2 ± 8.1 50.0 ± 8.4 <0.001}*** _{56.4 ± 7.6 50.2 ± 8.4 <0.001}***

Body mass index (kg m2₎c

24.9 ± 3.4 25.0 ± 3.5 0.53 26.2 ± 3.3 24.9 ± 3.4 0.01*

Systolic blood pressure (mm Hg)c

138.4 ± 24.6 128.4 ± 19.7 0.01* _{140.4 ± 22.7 129.8 ± 21.3 0.04}*

Fasting plasma glucose (mg dL1 )c 107.5 ± 38.1 100.7 ± 30.1 0.01** _{113.8 ± 46.4 101.2 ± 30.4 0.03}* Serum creatinine (mg dL1 )c 1.1 ± 0.1 1.0 ± 0.2 0.008** _{1.6 ± 1.6 0.9 ± 0.2 0.01}** Urinary arsenic (lg L1 )d 79.4 (75.7–83.6) 69.2 (64.3–93.0) 0.01** 81.3 (77.6–85.1) 69.5 (64.2–93.8) 0.19 Urinary cadmium (lg L1₎d _{1.0 (0.9–1.1) 1.0 (0.9–1.1) 0.81 1.0 (0.9–1.1) 0.9 (0.8–1.1) 0.31} Urinary lead (lg L1₎d _{19.5 (18.5–20.5) 16.0 (14.5–17.5) 0.008}** _{19.9 (18.0–20.0) 16.2 (14.3–18.4) 0.10} Urinary nickel (lg L1₎d 4.5 (4.2–4.7) 4.0 (3.6–4.4) 0.06 4.3 (4.1–4.5) 4.7 (3.9–5.5) 0.38 Percentage (%) Percentage (%) Categorical variables Male (%) 63.5 44.2 <0.001*** 57.4 46.0 0.02* Cigarette smoker (%) 28.9 19.0 0.001** _{24.5 21.1 0.26} Diabetes (%)e 13.3 8.4 0.019* _{23.4 8.2 <0.001}*** Hypertension (%)f 56.2 37.3 <0.001*** _{60.6 37.1 <0.001}*** a

Student’s t-test for normal distributed and Mann–Whitney test for non-normal distributed continuous variables, chi-square test for categorical variables. b

Age-adjustment. c

Mean ± SD or geometric mean ± geometric SD. d

Geometric mean (95% confidence interval). e

Diabetes: fasting plasma glucose P126 mg dL1

or on diabetes therapy.

f _{Hypertension: systolic and diastolic blood pressure P90 or 140 mm Hg, respectively, or on anti-hypertension therapy.} *_{p < 0.05.}

** _{p < 0.01.} ***_{p < 0.001.}

renal dysfunction. In the present study, renal tubular dysfunction, indicated by an elevated excretion of B2MG, occurred in subjects with high levels of U-As. Recently, Buchet et al. (2003)reported the renal effects of co-exposure to As and Cd in a Chinese popula-tion. In this study, U-As was claimed as a significant predictor of renal tubular and glomerular dysfunction (Buchet et al., 2003). In the present study, among the renal-toxic metals (such as As, Pb, Cd, and Ni) it was found that total U-As showed significant associ-ations with B2MG and eGFR after adjustment for all potential con-founding variables including Pb, Cd, and Ni. In a governmental report (Taiwan-EPA, 2003–2006), several metals in agriculture soil of study areas exceed the criteria of Taiwan, including As, Pb, Ni, and Hg due to seriously polluted by electroplating and metal-related factories. From questionnaire, around 13% of subjects living near metal-related factories (within 50 m), in which 34% of the fac-tories were electroplating and lead-acid batteries factory. It is likely that residents in this area were exposed to multi-elements. In fact, urinary metal measurement showed a high correlation be-tween U-As with other metals of Pb, Cd, and Ni. We have evaluated the effects of other metals by similar statistical approaches and found that Cd and Ni did not show significant ORs with diabetes,

abnormal B2MG, and eGFR. It was found only Pb showed signifi-cant ORs for abnormal B2MG in the upper tertile as compared with lower tertile (p < 0.05, data not shown but inSupplemental Table A and Table B). InTable 2, B2MG was also found significantly differ-ent between lower and upper groups of U-Pb, indicating that Pb exposure may also have adverse effects on renal tubular function. Thus, in the present study, the contribution of Pb to the observed renal dysfunction cannot be ruled out. When evaluating the effect of renal tubular or glomerular dysfunction by U-As, to consider the effect by U-Pb is needed.

For U-Cd the correlation was not significant, this may be be-cause the study areas were not highly contaminated with Cd; thus, the subjects may have ingested low levels of Cd with little influ-ence on renal function. U-Cd concentration in present study was 0.96

l

_l

area in China (Hong et al., 2004; Nordberg et al., 2005) with U-Cd concentrations 0.86 and 2.16

l

exposed areas, respectively. The U-Cd level of present study was higher than that of the subjects in NHANES III (0.48

l

creati-nine or 0.57

l

were living within contaminated areas in central Taiwan, and their U-As level was significantly associated with renal dysfunction.

Evidence of renal toxicity by As exposure (kidney cancer ex-cepted) was not frequently available from animal models using rel-atively high dose of As. A previously study reported that, depending on the serum marker used (serum creatinine or cystatin C), the increase in GFR ranged from 7% to 11% in children who had blood Pb levels in the upper quartile of the population (>55

l

mean blood Pb, 78.4

l

an experimental study focused on (Courtois et al., 2003), the initial mechanism may well depend on Pb-induced production of reactive oxygen species that upregulate cyclooxygenase (COX-2) expres-sion in the vascular smooth muscle. Oxidative stress has been iden-tified as an important mechanism in As-induced decreased kidney function through accumulation of As in kidney tissue, and reduced glutathione in treated animal (Sinha et al., 2008). Wang et al. (2009) reported that kidney is susceptible to secondary damage resulting from disease such as diabetes and hypertension (Wang et al., 2009). Further studies are needed.

U-As levels tend to vary greatly between studies. A study carried out in Bangladesh (Lindberg et al., 2008) found that the respective 5th and 95th percentiles of U-As concentration were 20 and

0 2 4 6 8 Low U-As, Low U-Pb Low U-As, High U-Pb High U-As, Low U-Pb High U-As, High U-Pb n=262 n=261 n=259 n=261 p=0.05 p=0.05 p<0.001 p=0.66 p=0.01 p=0.009 Odds ratios

Fig. 2. Logistic regression of abnormal b2-microglobulin (B2MG P 0.154 mg L1 ) by median of urinary arsenic (U-As: 85.13lg L1

) and urinary lead (U-Pb: 18.9lg L1

). The model was adjusted for age, sex, living area, cigarette smoking, diabetes, hypertension, cadmium, and nickel.

Table 4

Odds ratios (ORs) for renal dysfunction (elevated b2 microglobulin (B2MG) or reduced estimated glomerular filtration rate (eGFR)) according to American Conference of Governmental Industrial Hygienists (ACGIH) cut-off points of urinary arsenic (U-As) concentration (n = 1043).

U-As (lg g1_{creatinine) B2MG > 0.154 mg L}1_{[n (percentage)]}d _ORsa _ORsb _ORsc,d

635 (n = 133) 14 (10.5%) 1 1 1

>35–75 (n = 315) 66 (21.0%) 2.61 (1.61–6.36)** _{2.29 (1.13–5.55)}** _{1.69 (0.94–3.64)}

>75–200 (n = 484) 137 (28.3%) 3.82 (1.66–6.62)** _{3.44 (1.55–6.11)}** _{2.11 (1.23–4.98)}*

>200 (n = 111) 32 (28.8%) 3.28(1.78–6.56)*** _{3.24(1.56–6.24)}** _{2.04(1.11–4.37)}*

eGFR < 60 mL min1_{(1.73 m}2₎1_{[n (percentage)]}

635 (n = 133) 6 (4.5%) 1 1 1

>35–75 (n = 315) 31 (9.8%) 1.34 (0.78–2.23) 1.15 (0.63–1.89) 1.11 (0.56–1.80) >75–200 (n = 484) 48 (9.9%) 0.91 (0.51–1.79) 0.76 (0.44–1.86) 0.68 (0.42–1.33) >200 (n = 111) 10 (9.0%) 2.39 (1.11–5.44)* _{2.12 (1.08–5.45)}* _{1.98 (0.95–4.99)}*

eGFR < 90 mL min1_{(1.73 m}2₎1_{[n (percentage)]}

635 (n = 133) 98 (73.9%) 1 1 1

>35–75 (n = 315) 267 (84.8%) 1.80 (0.66–2.39) 1.67 (0.59–1.98) 1.45 (0.49–1.88) >75–200 (n = 484) 403 (83.3%) 2.50 (1.32–3.89)* _{2.21 (1.21–3.86)}* _{2.15 (1.06–3.78)}*

>200 (n = 111) 96 (86.5%) 3.08 (1.42–7.27)** _{2.31 (1.18–4.67)}* _{2.16 (1.11–3.49)}*

a

Adjusted for age, sex, living area, and cigarette smoking.

b_{Adjusted for age, sex, living area, cigarette smoking, diabetes, and hypertension.}

c _{Adjusted for age, sex, living area, cigarette smoking, diabetes, hypertension, lead, cadmium, and nickel.}

d_{Trend test for increasing rate of abnormal b2 microglobulin with U-As level, p = 0.05 by X}2_{, and for ORs}c_{, p = 0.05 by logistic regression.}

*_{p < 0.05.} ** _{p < 0.01.} ***_{p < 0.001.}

453

l

_l

med-ian in present study.Table 5shows the results of epidemiologic stud-ies in several countrstud-ies compared with the present study. The total U-As concentrations varied widely in different countries. The U-As concentrations were >100

l

to a high concentration of As (such as in Chile, China, and Bangla-desh). However, the U-As concentrations of healthy individuals that were not exposed to high As levels were lower (Denmark, UK, German, and US). The higher U-As levels found in men in this study (Table 5) were also reported for residents in the arseniasis endemic area in southwestern Taiwan and in the general population in a Danish study (Kristiansen et al., 1997). However, some studies shown inTable 5show an opposite result, with U-As levels higher in females than males. Further investigations are required to deter-mine the possible confounding variables that affect U-As concentra-tion. Important issues that need evaluation are whether different lifestyles result in different levels of exposure. Many epidemiologic studies on As poisoning from the consumption of As-contaminated water have been reported worldwide (Yoshida et al., 2004; Arain et al., 2009a). However, there are relatively few reports on the dose–response relationship between As exposure and As-related adverse health effects, because it is often difficult to evaluate individual levels of exposure to As.

A limitation of the present study is the use of total U-As level to reflect toxic form of inorganic As exposure. Compared to the time-consuming As species analysis, total U-As can be measured in a short period of time for a large number of samples, which is essen-tial for many study. However, this measure does not account for variation in As uptake and metabolism between individuals. In spite of this shortcoming, total U-As is still considered a reasonable biomarker for inorganic As exposure. A significant dose–response relationship between total U-As and eGFR in non-diabetic subjects (data not shown) was also found. The present study may be limited also to single time recruitment, thus it is difficult to establish a causal relationship between U-As concentration and renal dysfunc-tion. A future follow-up study of renal disease incidence according to the As exposure would be helpful to establish the temporality for causal inference.

5. Conclusions

In the present study, the rate of abnormal B2MG increased with the U-As concentrations in a dose-dependent manner. Renal dys-function rates significantly increased when the U-As rose above 75

l

and glomerular function (eGFR < 90 mL min1_{(1.73 m}2₎1_{). A}

Table 5

Total urinary arsenic levels from different countries.

Country Exposure status Sample size

Arsenic levels in urine References Denmark A group of unexposed Danish cohort Male: 89 Male: 15.3 ± 12.4lg L1 _{Kristiansen et al.}

(1997)

Female: 93 Female:11.3 ± 10.4lg L1 UK Healthy unexposed British subjects 23 12.3 (0.9–1080)lg L1

White and Sabbioni (1998)

German The population of German Environmental Survey II (1990–1992)

4001 6.3 (6.08–6.51)a_l

g L1

Seifert et al. (2000)

Chile Subjects who had long-term exposure to very high levels of arsenic in drinking water (735– 762lg L1

)

Fathers: 11 Father: 464 (184–1026)lg L1 _{Chung et al. (2002)}

Mothers: 11 Mother: 466 (161–773)lg L1 Sons: 14 Sons: 566 (55–1320)lg L1 Daughters: 8 Daughter: 427 (260–736)lg L1 UK Healthy volunteers Asian: 21 Somali Black-African population: 7.2lg g1

creatinine

Brima et al. (2006)

Somali: 22 Asian: 20.6lg g1 creatinine White: 20 White groups: 24.5lg g1

creatinine US US general population 788 7.1 (3.6–13.9)lg L1b

Navas-Acien et al. (2008)

China Subjects with low-arsenic and high-arsenic exposure through food intake

High: 10 High: 58.3 (56.2–60.5)lg g1_{creatinine Li et al. (2008)} Low: 35 Low: 23.4 (20.9–25.8)lg g1_creatinine

China Subjects in a high-arsenic-exposed area and control study site

Exposed subject: 113

Exposed subjects: 192.2 ± 22lg g1

creatinine Wang et al. (2009)

Control site: 30 Control site: 63.6 ± 5.9lg g1_creatinine Bangladesh Infants 7 months-of-age who were born in an

area of high-arsenic-contaminated tube wells

1799 8 weeks gestation: 81 (37–207)b_l

g L1

Tofail et al. (2009)

30 weeks gestation: 84 (42–230)b_l

g L1 Bangladesh The study subjects were 10,402 residents, who

had consumed contaminated well water

Male: 4138 Male: 131.7 ± 137.8lg L1_{; 229.8 ± 249.0lg g}1 creatinine Heck et al. (2009) Female: 6264 Female: 132.3 ± 152.2lg L1 ; 297.6 ± 289.4lg g1 creatinine Taiwan (present study)

Subjects live in an area where electroplating has seriously polluted agricultural soils through the discharge of wastewater into irrigation ditches Male: 489 Male: 91.2 (85.1–97.7)lg L1_{; 80.1 (72.8–87.3)} lg g1_creatinine Female: 554 Female: 74.1 (69.2–77.6)lg L1 ; 88.6 (80.8–96.3) lg g1 creatinine a _{Geometric mean (95% confidence interval).}

b

Median (interquartile range).

significantly higher OR of DM in subjects with a U-As P 75

l

creatinine was found as compared to the reference group with a U-As 6 35

l

4.82 times higher in subjects with both over the median levels of U-As (85.1

l

_l

lower than the median (p 6 0.001). These results indicate that U-As might relate to renal dysfunction, and the contribution of Pb to the observed renal dysfunction cannot be ruled out. Further studies are needed by using follow-up approach or experiment of renal tubular damage by As.

Funding sources

This study was supported by Grants EO097-SP02, NSC 98-2314-B-400-001-MY3 and EO098-SP02 from the Department of Health of the Republic of China.

Acknowledgements

We are highly indebted to the Changhua Christian Hospital, Changhua Public Health Bureau, and the public health nurses for their support. We thank Drs. Chi-Pang Wen, Dennis P.H. Hsieh., and Louis W. Chang for the kind supervision and valuable com-ments to this study. We also thank Yueh-Ching Wang for her assis-tance in data collection and management.

Appendix A. Supplementary material

Supplementary data associated with this article can be found, in the online version, atdoi:10.1016/j.chemosphere.2011.02.091. References

ACGIH, 1996. Threshold Limit Values for Chemicals and Physical Agents and Biological Exposure Indices for 1995–1996. American Conference of Governmental Industrial Hygientists, Cincinnati, OH.

Arain, M.B., Kazi, T.G., Baig, J.A., Jamali, M.K., Afridi, H.I., Shah, A.Q., Jalbani, N., Sarfraz, R.A., 2009a. Determination of arsenic levels in lake water, sediment, and foodstuff from selected area of Sindh, Pakistan: estimation of daily dietary intake. Food Chem. Toxicol. 47, 242–248.

Arain, M.B., Kazi, T.G., Baig, J.A., Jamali, M.K., Afridi, H.I., Jalbani, N., Sarfraz, R.A., Shah, A.Q., Kandhro, G.A., 2009b. Respiratory effects in people exposed to arsenic via the drinking water and tobacco smoking in southern part of Pakistan. Sci. Total Environ. 407, 5524–5530.

Baig, J.A., Kazi, T.G., Arain, M.B., Afridi, H.I., Kandhro, G.A., Sarfraz, R.A., Jamal, M.K., Shah, A.Q., 2009. Evaluation of arsenic and other physico-chemical parameters of surface and ground water of Jamshoro, Pakistan. J. Hazard. Mater. 166, 662– 669.

Buchet, J.P., Heilier, J.F., Bernard, A., Lison, D., Jin, T., Wu, X., Kong, Q., Nordberg, G., 2003. Urinary protein excretion in humans exposed to arsenic and cadmium. Int. Arch. Occup. Environ. Health 76, 111–120.

Brima, E.I., Haris, P.I., Jenkins, R.O., Polya, D.A., Gault, A.G., Harrington, C.F., 2006. Understanding arsenic metabolism through a comparative study of arsenic levels in the urine, hair and fingernails of healthy volunteers from three unexposed ethnic groups in the United Kingdom. Toxicol. Appl. Pharmacol. 216, 122–130.

Chang, F.H., Wang, H.J., Wang, S.L., Wang, Y.C., Hsieh, D.P., Chang, L.W., Ko, Y.C., 2006. Survey of urinary nickel in residents of areas with a high density of electroplating factories. Chemosphere 65, 1723–1730.

Chen, C.J., Hsueh, Y.M., Lai, M.S., Shyu, M.P., Chen, S.Y., Wu, M.M., Kuo, T.L., Tai, T.Y., 1995. Increased prevalence of hypertension and long-term arsenic exposure. Hypertension 25, 53–60.

Chiou, J.M., Wang, S.L., Chen, C.J., Deng, C.R., Lin, W., Tai, T.Y., 2005. Arsenic ingestion and increased microvascular disease risk: observations from the south-western arseniasis-endemic area in Taiwan. Int. J. Epidemiol. 34, 936–943.

Chung, J.S., Kalman, D.A., Moore, L.E., Kosnett, M.J., Arroyo, A.P., Beeris, M., Mazumder, D.N., Hernandez, A.L., Smith, A.H., 2002. Family correlations of arsenic methylation patterns in children and parents exposed to high concentrations of arsenic in drinking water. Environ. Health Perspect. 110, 729–733.

Courtois, E., Marques, M., Barrientos, A., Casado, S., Lopez-Farre, A., 2003. Lead-induced downregulation of soluble guanylate cyclase in isolated rat aortic segments mediated by reactive oxygen species and cyclooxygenase-2. J. Am. Soc. Nephrol. 14, 1464–1470.

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, 584–590. Heck, J.E., Nieves, J.W., Chen, Y., Parvez, F., Brandt-Rauf, P.W., Graziano, J.H.,

Slavkovich, V., Howe, G.R., Ahsan, H., 2009. Dietary intake of methionine, cysteine, and protein and urinary arsenic excretion in Bangladesh. Environ. Health Perspect. 117, 99–104.

Hong, F., Jin, T., Zhang, A., 2004. Risk assessment on renal dysfunction caused by co-exposure to arsenic and cadmium using benchmark dose calculation in a Chinese population. Biometals 17, 573–580.

Hsueh, Y.M., Chung, C.J., Shiue, H.S., Chen, J.B., Chiang, S.S., Yang, M.H., Tai, C.W., Su, C.T., 2009. Urinary arsenic species and CKD in a Taiwanese population: a case-control study. Am. J. Kidney Dis. 54, 859–870.

Hughes, M.F., 2006. Biomarkers of exposure: a case study with inorganic arsenic. Environ. Health Perspect. 114, 1790–1796.

IDBMOEA (International Development Bureau Ministry of Economic Affairs), 2002. Handbook of the Integrated Pollution Prevention Techniques for Metal Surface Treatment Industries – Electroplanting Industry. Industrial Development Bureau Ministry of Economic Affairs, Executive Yuan, Republic of China, Taipei.

Kazi, T.G., Arain, M.B., Baig, J.A., Jamali, M.K., Afridi, H.I., Jalbani, N., Sarfraz, R.A., Shah, A.Q., Niaz, A., 2009. The correlation of arsenic levels in drinking water with the biological samples of skin disorders. Sci. Total Environ. 407, 1010– 1026.

Kristiansen, J., Christensen, J.M., Iversen, B.S., Sabbioni, E., 1997. Toxic trace element reference levels in blood and urine: influence of gender and lifestyle factors. Sci. Total Environ. 204, 147–160.

Levey, A.S., Bosch, J.P., Lewis, J.B., Greene, T., Rogers, N., Roth, D., 1999. A more accurate method to estimate glomerular filtration rate from serum creatinine: a new prediction equation. Modification of diet in renal disease study group. Ann. Intern. Med. 130, 461–470.

Levey, A.S., Eckardt, K.U., Tsukamoto, Y., Levin, A., Coresh, J., Rossert, J., De Zeeuw, D., Hostetter, T.H., Lameire, N., Eknoyan, G., 2005. Definition and classification of chronic kidney disease: a position statement from kidney disease: improving global outcomes (KDIGO). Kidney Int. 67, 2089–2100.

Li, X., Pi, J., Li, B., Xu, Y., Jin, Y., Sun, G., 2008. Urinary arsenic speciation and its correlation with 8-OHdG in Chinese residents exposed to arsenic through coal burning. Bull. Environ. Contam. Toxicol. 81, 406–411.

Lindberg, A.L., Ekstrom, E.C., Nermell, B., Rahman, M., Lonnerdal, B., Persson, L.A., Vahter, M., 2008. Gender and age differences in the metabolism of inorganic arsenic in a highly exposed population in Bangladesh. Environ. Res. 106, 110– 120.

Montaser, A., 1998. Inductively Coupled Plasma Mass Spectrometry. George Washington University, Washington DC, USA, Wiley-VCH.

Navas-Acien, A., Sharrett, A.R., Silbergeld, E.K., Schwartz, B.S., Nachman, K.E., Burke, T.A., Guallar, E., 2005. Arsenic exposure and cardiovascular disease: a systematic review of the epidemiologic evidence. Am. J. Epidemiol. 162, 1037–1049. Navas-Acien, A., Silbergeld, E.K., Pastor-Barriuso, R., Guallar, E., 2008. Arsenic

exposure and prevalence of type 2 diabetes in US adults. JAMA 300, 814–822. Navas-Acien, A., Silbergeld, E.K., Streeter, R.A., Clark, J.M., Burke, T.A., Guallar, E., 2006. Arsenic exposure and type 2 diabetes: a systematic review of the experimental and epidemiological evidence. Environ. Health Perspect. 114, 641–648.

Nordberg, G.F., Jin, T., Hong, F., Zhang, A., Buchet, J.P., Bernard, A., 2005. Biomarkers of cadmium and arsenic interactions. Toxicol. Appl. Pharmacol. 206, 191– 197.

Paschal, D.C., Burt, V., Caudill, S.P., Gunter, E.W., Pirkle, J.L., Sampson, E.J., Miller, D.T., Jackson, R.J., 2000. Exposure of the U.S. population aged 6 years and older to cadmium: 1988–1994. Arch. Environ. Contam. Toxicol. 38, 377–383. Seifert, B., Becker, K., Hoffmann, K., Krause, C., Schulz, C., 2000. The german

environmental survey 1990/1992 (GerES II): a representative population study. J. Expo. Anal. Environ. Epidemiol. 10, 103–114.

Sinha, M., Manna, P., Sil, P.C., 2008. Arjunolic acid attenuates arsenic-induced nephrotoxicity. Pathophysiol 15, 147–156.

Smith, A.H., Arroyo, A.P., Mazumder, D.N., Kosnett, M.J., Hernandez, A.L., Beeris, M., Smith, M.M., Moore, L.E., 2000. Arsenic-induced skin lesions among Atacameno people in Northern Chile despite good nutrition and centuries of exposure. Environ. Health Perspect. 108, 617–620.

Taiwan-EPA, 2003–2006. The Annual Report of Environmental Water Quality Monitoring: Groundwater Quality Section. Environmental Protection Administration, Executive Yuan, Taipei, Taiwan.

Tofail, F., Vahter, M., Hamadani, J.D., Nermell, B., Huda, S.N., Yunus, M., Rahman, M., Grantham-McGregor, S.M., 2009. Effect of arsenic exposure during pregnancy on infant development at 7 months in rural Matlab, Bangladesh. Environ. Health Perspect. 117, 288–293.

Wang, J.P., Wang, S.L., Lin, Q., Zhang, L., Huang, D., Ng, J.C., 2009. Association of arsenic and kidney dysfunction in people with diabetes and validation of its effects in rats. Environ. Int. 35, 507–511.

Wang, S.L., Chang, F.H., Liou, S.H., Wang, H.J., Li, W.F., Hsieh, D.P., 2007. Inorganic arsenic exposure and its relation to metabolic syndrome in an industrial area of Taiwan. Environ. Int. 33, 805–811.

Wang, S.L., Chiou, J.M., Chen, C.J., Tseng, C.H., Chou, W.L., Wang, C.C., Wu, T.N., Chang, L.W., 2003. Prevalence of non-insulin-dependent diabetes mellitus and

related vascular diseases in southwestern arseniasis-endemic and nonendemic areas in Taiwan. Environ. Health Perspect. 111, 155–159.

Wen, C.P., Cheng, T.Y., Tsai, M.K., Chang, Y.C., Chan, H.T., Tsai, S.P., Chiang, P.H., Hsu, C.C., Sung, P.K., Hsu, Y.H., Wen, S.F., 2008. All-cause mortality attributable to chronic kidney disease: a prospective cohort study based on 462 293 adults in Taiwan. Lancet 371, 2173–2182.

White, M.A., Sabbioni, E., 1998. Trace element reference values in tissues from inhabitants of the European Union. X: a study of 13 elements in blood and urine of a United Kingdom population. Sci. Total Environ. 216, 253–270.

Yoshida, T., Yamauchi, H., Fan Sun, G., 2004. Chronic health effects in people exposed to arsenic via the drinking water: dose–response relationships in review. Toxicol. Appl. Pharmacol. 198, 243–252.