Introduction The IARC (1987)1 has classified arsenic as a group 1 human carcinogen. Chronic exposure to inorganic arsenic can cause cancerous1-4 and non-cancerous health hazards5,6 in humans. Arsenic can get entry into the human body via drinking water, eating food, inhaling dust, and/or ingesting soil. With chronic and continuing exposure, steady-state concentrations of arsenic in blood and urine are achieved; these have been the potential to serve as biomarkers of arsenic exposure7. However, urinary arsenic is a reflection of As excretion and not actual tissue burden8, and significant complexities are introduced when urinary arsenic concentrations are normalized for urinary creatinine in order to adjust for dilution factor9,10. Hall et al. (2006)7 have been suggested that with chronic exposure, blood arsenic which receives inputs not only from recent exogenous exposure but also from tissue compartments – may better reflect an individual's total internal As burden. The main organ for arsenic metabolism is the liver, but the metabolic pathway of inorganic arsenic is not yet fully clarified11,12. Trivalent arsenic species are more ready to cross cell membrane and inorganic pentavalent arsenate in mostly reduced to trivalent arsenite in the blood stream before entering the cells for further metabolism13,14. Inorganic arsenic is metabolized in the body by alternating reduction of pentavalent arsenic to trivalent and addition of a methyl group from S-adenosylmethionine as methyl donor11,15 (Fig. 1). Figure 1. Metabolism of Inorg-As. Determination of the concentrations and the proportional distribution of the various arsenic species including the inorganic arsenic and the methylated metabolites in urine can give a reflection of the capacity to methylate inorganic arsenic in human body. The ratio between MMAV and inorganic arsenic (PMI, primary methylation index) and the ratio between DMAV to MMAV (SMI, secondary methylation index) are also used to assess the arsenic methylation capacity of the first and second methylation step, respectively. Several studies have been shown an increasing prevalence of arsenic-related toxic effects with increasing % MMA in urine5,16,17 and probably an increased concentrations of the highly toxic MMAIII) at cellular level18,19. A number of studies have shown associations between the severity of arsenic related health effects and nutritional status20-22. Lower Se intake is associated with enhance As toxicity20,23-25 and lower urinary Se levels were associated with increased % inorg As and decreased % DMA in urine25. Another study has been reported that subjects with higher intakes of Zn had lower % MMA and higher % DMA in urine21. Zinc (Zn) has been linked to decrease arsenic toxicity in some studies2,26. Trace elements are well known to play an important role in the maintenance of health27. Thus, monitoring the status of trace elements is of critical importance in human health. Today, biomonitoring of trace elements in human blood and urine have become an important tool for measuring trace elements status28-31. It is well known that the concentrations of pollutants in spot urine sample are highly dependent on the dilution of the sample caused by variation in the intake of fluids, physical activity, temperature, etc32. Commonly applied method to control for this variation is adjustment by the creatinine concentration in urine9,32,33. However, creatinine is a waste product formed by the spontaneous, essentially irreversible dehydration of body creatine and creatine phosphate from muscle metabolism and meat intake9,34,35. Thus, urinary creatinine (U-cre) varies by gender, age, body size, racelethnicity, diet, renal function, etc9,36,37. Recent studies have been reported that urinary arsenic levels (µg/L) were found significantly correlated with urinary creatinine levels10,38,39. Gamble et al. (2005)10,38 has found that higher urinary creatinine is associated with reduced risk for premalignant skin lesions among the arsenic exposed population in Bangladesh and folic acid supplementation significantly increased urinary creatinine. But Hindwood et al. (2002)33 have been suggested that creatinine adjustment of urinary inorganic arsenic (Inorg-As) concentrations may not be required in population studies investigating environmental exposure. To date, other metal or metalloids that may influence arsenic methylation are largely unknown. The aims of this study were to assess the influence of trace elements for biotransformation process of inorganic arsenic and the correlations between the concentrations of trace element in blood as well as urine and urinary arsenic metabolites among the population in Lagunera area of Mexico, who drunk water containing arsenic in range 38 to 116 µg/L. Our results suggest that trace elements had influenced arsenic methylation process in humans, but it was concentrations dependent. The results also suggested that urinary creatinine adjustment might be over-estimated of urinary trace element concentrations due to low concentrations of urinary creatinine for the people with low arsenic in urine. Materials And Methods Reagents. The chemicals used and there sources are as follows: Sodium arsenate (ACS reagent grade) from MCB Reagents (Cincinnati, OH); dimethylarsinic acid (sodium salt), ammonium phosphate (dibasic), and glutathione (GSH) from Sigma Chemical Co. (St. Louis, MO); sodium m-arsenite and ammonium nitrate from Sigma-Aldrich Co. (St. Louis, MO); disodium methylarsenate from ChemService, Inc. (West Chester, PA). The arsenic and other elements standard solution was from SPEX Certiprep (Metuchen, NJ). Freeze-dried urine reference material for toxic elements (SRM 2670a) and frozen bovine blood reference material for toxic metals (SRM 966) from National Institute of Standards & Technology (NIST, Gaithersburg, MD 20899). Triton X-100 was from Pharmacia Biotech (Uppsala, Sweden). All other chemicals were analytical reagent grade or the highest quality obtainable. Water was doubly deionized and distilled. Subjects. Urine and blood samples were collected from 191 subjects (98 females and 93 males), aged 18-77 years in the Lagunera area of Mexico. There were five groups, based on total arsenic concentration (38-116 µg/L) in their drinking water. Urine and Blood Collection. All collecting containers were soaked overnight in 2% nitric acid (Baker analyzed for trace metal analysis) (J. T. Baker, Inc. Phillipsburg, NJ) and rinsed with double distilled and deionized water. All plastic measuring and collecting equipment were similarly washed, sealed in bags, placed in locked footlockers, and transported by air to the site of the study at the same time as the investigators. After collection, urine sample was immediately frozen in a portable icebox containing dry ice. Blood was collected by venous puncture, into vacutainers containing EDTA, transferred to the vial, and immediately frozen. The samples were kept frozen while being transported to the University of Arizona, Tucson where they were stored at -70° C before analysis. Arsenic Species Analysis. Frozen urine samples were thawed at room temperature, filtered with a 0.45 µm filter (Nanosep MF Centrifugal Devices, Pall Life Sciences, Ann Arbor, MI), and diluted 5-fold using Milli-Q water before injection. An HPLC-ICP-MS (High Performance Liquid Chromatography- Inductively Coupled Plasma-Mass Spectrometry) speciation method40 was modified for the measurement of arsenic species including AsB by author. The HPLC system consisted of a PerkinElmer Series 200 HPLC with an anion exchange column (PRP-X100, 10 µm, 250 x 4.6 mm, Hamilton Company, Nevada). The mobile phase (pH 8.5) contained 10 mM ammonium nitrate and 10 mM ammonium phosphate (dibasic) at a flow rate of 1 ml/min. The column temperature was maintained at 30° C. An ELAN DRCe ICP-MS (PerkinElmer) with a cyclonic quartz spray chamber and Meinhard nebulizer was used as a detector for the analysis of arsenic species [AsB, AsV, AsIII, MMAV, and DMAV] in urine at 4° C. The operating parameters were as follows: Rf power, 1400 W; plasma gas flow, 15 L/min; nebulizer gas flow, 0.82 L/min; auxiliary gas flow, 1.2 L/min; oxygen flow for DRC, 0.87 mL/min; and arsenic was measured at m/z 91. The working detection limits and accuracy of this analytical method were as follows: The working detection limits were 0.80 - 1.75 µg/L for arsenic metabolites. Accuracy values were calculated by spiking standard compounds of all five species (5 µg/L) in urine samples. The recoveries of the added compounds were 98-103%. Standard samples (5 µg/L) containing all five arsenic species were also analyzed after analysis the urine samples each day. The values of mean ± SE for AsB, AsV, AsIII, MMAV, and DMAV were found 4.86 ± 0.08, 5.09 ± 0.11, 5.16 ± 0.11, 5.02 ± 0.10, and 4.90 ± 0.05, respectively. Trace elements analysis in urine. Urine samples in acid washed polypropylene tubes were digested with nitric acid (5: 1) while a water bath for 40 min at 70° C. Freeze-dried urine reference material for toxic elements containing arsenic at a level of 220 ± 10 µg As/L was used for quality control and to validate the assay. After acid digestion, analysis of this standard by ICP-MS yielded a range of 216.0 - 236.0 µg As/L with a range of recoveries of 98.18 - 107.27 %. We also analyzed the spiking standard compounds of all the arsenic species [for AsB, AsV, AsIII, MMAV, and DMAV] at levels of 10 µg total As /L and 20 µg total As /L. The recoveries of the spiking samples were 104.20 % (10.42 ± 0.13 µg As/L) and 97.70 % (19.54 ± 0.24 µg As/L), respectively. After acid digestion, analyzed trace elements in urine samples collected from the subjects and NIST reference urine samples. The recoveries of Se, Zn, Co, Cu, Mn, Ni, Cd, Pb, and Hg in NIST reference urine were 92.16 %, 93.01 %, 101.00 %, 94.77 %, 106.06 %, 100.84 %, 109.70 %, 100.72 %, and 94.28 %, respectively. The multi-element standard solutions were digested and diluted using the same procedure and dilution factors (as the samples) for preparation of the calibration curve. The calibration correlation coefficients (r2) of the elements were greater than 0.999. Trace element analysis in whole blood. Whole blood samples were analyzed for total As, Se, Zn, Co, Cu, Mn, Ni, Cd, Pb, and Hg concentrations using Perkin Elmer Elan DRCe ICP-MS. Inductively coupled plasma mass spectrometry method for elements in whole blood was developed (with modifications) based on published method41. Whole blood samples were thawed, thoroughly mixed, diluted 50 times with diluents containing 0.65% HNO3 + 0.1% Triton X-100, and centrifuged for 10 min (3500 rpm at 4° C) with the supernatant reserved for analysis. The multi-element standard solutions were prepared from stock standard solution with 0.65% HNO3 + 0.1% Triton X-100. Three working mercury standard solutions, viz., 1, 2.5, and 5 µg/L were prepared from stock standard solution with 0.65% HNO3 + 0.1% Triton X-100, added gold (200 ppb), and mixed well. Five working other elements (As, Se, Zn, Co, Cu, Pb, Cd, Ni, and Mn) standard solutions, viz., 5, 10, 20, 50, and 100 µg/L were prepared from stock standard solution with same diluents, added internal standards (Ga, In, & Re; 100 µg/L of each), and mixed well. The calibration correlation coefficients (r2) of the elements were greater than 0.999. Frozen bovine blood reference material for toxic metals was used for quality control and to validate the assay. The reference sample was thawed in ice, mixed thoroughly, and diluted 50 times with diluents containing 0.65% HNO3 + 0.1% Triton X-100, and centrifuged for 10 min (3500 rpm at 4° C) with the supernatant reserved for analysis. The recoveries of the elements in the reference bovine blood samples were very close to the certified values (Table 1). The recoveries of Pb, Cd, and Hg in the reference bovine Table 1. Recovery data for NIST SRM ‘Bovine Blood’ and Spike Recoveries for human blood. Values are the mean ± SE (n=3 for bovine blood and n = 8 for human blood) blood samples were 92 %, 107 %, and 97 %, respectively. We also analyzed the spiking standard elements in the human blood samples and also the quality control (QC) standard samples. The spiking and QC samples were prepared and analyzed using the same procedures as the human blood samples. The recoveries of the elements in the spiking and QC samples are shown in Table 1 and 2, respectively. The rinse solution contained 2% HNO3 + 1% Triton X 100. Table 2. Recovery data for trace elements in the quality control standard (QCS). Values are the mean ± SE (n=8) Creatinine Measurement. Creatinine concentration in urine was determined using the Randox Creatinine Colorimetric kit (San Diego, CA), which is based on the reaction of creatinine with picric acid in alkaline solution, forming a colored complex, measured at 492 nm42. Statistical Analysis: The means and standard error (SE) were calculated. The unpaired t test (GraphPad Software, Inc., 2005) was used to analyze the significance difference. The correlation coefficients for different variables were tested using the Spearman rank order correlation test (Richard Lowry, 1998-2009). P values less than 0.05 were considered significant. Results In this study, there were five groups (Gps) of participants based on total arsenic concentration in their drinking water. The general characteristics of the study population have been previously described in detail (Manuscript submitted). Human blood samples were collected from 191 subjects. After collection, blood samples were transferred to the Nalgene vials (Nalge Nunc International, NY) and immediately frozen. We did not acid wash the Nalgene vials. But, we analyzed arsenic and other elements in rinse solution (2% HNO3) of Nalgene vials. The concentrations of the trace elements in rinse solution of the vials were below the MDL except Zn (2.42±0.79 µg/L). Study population. In this study, out of 191 participants in Lagunera area of Mexico, 98 were females (F) and 93 were males (M). The average age of females versus males was not statistically significant. Trace element concentrations in urine and whole blood. Figure 2 shows the distribution of the concentrations of trace elements in urine of different arsenic exposure groups in Lagunera area of Mexico. The concentrations of trace elements in urine and blood are reported in Table 3. The element concentrations in urine expressed as ug/g cre were higher for females (F) compared to males (M). But the concentrations of most elements in blood were opposite for females compared to males. The mean concentrations of Cd, Hg, Se, Co, Cu, and Mn were significantly higher in urine for females than males (p<0.01, p<0.05, p<0.001, p<0.001, p<0.01, & p<0.001, respectively). The concentrations of As, Pb, Zn, and Ni in urine were not statistically different between females and males. A significant difference of the concentration between females and males was found with respect to As, Pb, Zn, Cu, and Mn in blood (Table 3 and Fig. 3). The mean concentrations of As, Pb, and Zn in blood were significantly lower for females compared to males (p<0.0001, p<0.0001, and p<0.01, respectively). But, the mean concentrations of Cu and Mn were significantly higher for females than males (p<0.0001 and p<0.001, respectively). The concentrations of Cd, Se, Co, and Ni in blood were not statistically significant between females and males. More than 92% and 88% of the blood samples contained below working MDL concentrations of Hg (<1.10 µg/L) for females and males, respectively. Figure 2. The distribution of the concentrations of trace elements in urine of different arsenic exposure groups. Table 3. Trace elements concentrations in urine and whole blood for females (F) and males (M). Values are the mean ± SE (F, n=98 and M, n=93) Figure 3. Concentrations of trace elements in whole blood of females and males. Values are the mean ± SE. (* statistically significant) The mean concentrations of trace elements in urine followed the order: As> Pb> Hg> Cd (toxic elements) and Zn> Se> Cu> Ni> Mn> Co (essential elements) for both females and males. But in blood it was little different, for toxic elements: Pb> As> Cd, and essential elements: Zn> Cu> Se> Mn> Ni> Co for both females and males. Correlations of the concentrations of elements in blood versus ages of the arsenic exposed people in Mexico. Zinc (Zn) and copper (Cu) concentrations in blood were positively and significantly correlated with age of females and males, respectively (Table 4). But manganese (Mn) concentrations were negatively and significantly correlated with age of females but not of males. The correlations between other elements and ages were not statistically significant for both females and males. Table 4. Spearman correlation coefficients for bloodelements concentrations (µg/L) versus ages (years) of females and males of our study groups. Correlations of the concentrations of As in blood versus As concentrations in drinking water or urine for females and males. Total arsenic concentrations in bloods expressed as µg/L were strongly and positively correlated with arsenic concentrations in drinking water expressed as µg/L and urinary arsenic concentrations expressed as µg/L or µg/g creatinine for both females and males (Table 5). Table 5. Spearman correlation coefficients for As concentrations in bloods versus As concentrations in drinking water as well as urinesof females and males. Correlations of the concentrations of As and other elements in blood. Cobalt (Co) and nickle (Ni) concentrations in blood were positively and significantly correlated with As concentrations in blood of females (rs= +0.32, p<0.01 and rs= +0.57, p<0.000001, respectively) (Table 6). For males, blood Table 6. Spearman correlation coefficients (rs) for As versus other elements concentrations in bloods of females and males. As concentrations were positively and significantly correlated with blood Se, Co, Cu, Ni, and Mn concentrations (rs= +0.23, p<0.05; rs= +0.41, p<0.0001; rs= +0.39, p<0.0001; rs= +0.41, p<0.0001; and rs= +0.39, p<0.001, respectively). Influence of the relative concentrations of other trace elements to arsenic in urine on the percentage of urinary arsenic metabolites. There were better correlations between (a) the ratio of other element (Se, Zn, Mn, Ni, or Hg) to arsenic (As) in urine (µg/g cre) and the percentage of urinary arsenic metabolites than (b) the correlations found between the corresponding element concentrations expressed as µg/g cre in urine and percentage of urinary arsenic metabolites (Table 7). Table 7. Comparison of spearman correlation coefficients between (a) the ratio of the concentrations of other element to As or (b) the concentration of element (µg/g creatinine) in urines and the percentage (%) of urinary As metabolitesof females and males. Statistically significant correlations were not found between the concentrations (µg/g cre) of Se, Zn, Mn as well as Ni and % inorg As, % MMA, % DMA, as well as the ratios of % DMA to % MMA in urine for females. But, the ratios of the concentrations of Se, Mn, as well as Ni to As expressed as µg/g cre were positively and significantly correlated with % inorg As (rs= +0.29, p<0.01; rs= +0.23, p<0.05; and rs= +0.22, p<0.05, respectively) as well as % MMA (rs= +0.25, p<0.05; rs= +0.26, p<0.01; and rs= +0.21, p<0.05, respectively), and negatively correlated with % DMA (rs= -0.34, p<0.001; rs= -0.23, p<0.05; and rs= - 0.27, p<0.01, respectively) as well as the ratios of % DMA to % MMA (rs= -0.31, p<0.01; rs= -0.28, p<0.01; and rs= -0.25, p<0.05, respectively) in urine for females (Table 7). For males, the ratios of the concentrations (µg/g cre)) of Se or Zn to As than the concentrations of Se or Zn were more strongly and positively correlated with % inorg As levels (rs= +0.26, p<0.05 vs. rs= +0.17 not significant and rs= +0.22, p<0.05 vs. rs= +0.086 not significant, respectively), but more strongly and negatively correlated with % DMA levels (rs= -0.25, p<0.05 vs. rs= -0.12 not significant and rs= -0.19 not significant vs. rs= -0.008 not significant, respectively) in urines (Table 7). The correlations between the ratio of the concentrations of Mn to As and the percentage of urinary arsenic metabolites were more significant than the correlation found between Mn concentrations and the percentage of arsenic metabolites (with % inorg As: rs= +0.42, p<0.0001 vs. rs= +0.32, p<0.01, respectively; with % MMA: rs= +0.22, p<0.05 vs. rs= +0.20, p<0.05, respectively; with % DMA: rs= -0.42, p<0.0001 vs. rs= -0.29, p<0.01, respectively; with the ratios of %MMA to % inorg As: rs= -0.25, p<0.05 vs rs= -0.16 not significant, respectively; as well as with the ratios of % DMA to % MMA: rs= -0.31, p<0.01 and rs= -0.27, p<0.01, respectively) in urine for males. These relations were more significantly correlated in urine for males than females. It was also interesting that the correlation between (a) the ratio of Ni to As in urine (µg/g cre) and the percentage of urinary arsenic metabolites as well as (b) the correlation found between the Ni concentrations expressed as µg/g cre in urine and percentage of urinary arsenic metabolites were not statically significant. The results show (Table 7) that the correlations were more significant between the percentage of arsenic metabolites and the ratios of the concentrations of Hg to As than the concentrations of Hg in urine for both females and males. The ratio of the concentrations of Hg to As expressed as µg/g cre were positively and significantly correlated with % inorg As (rs= +0.41, p<0.0001 vs. rs= +0.34, p<0.001, respectively) as well as % MMA (rs= +0.20, p<0.05 and rs= +0.09 not significant, respectively), and negatively correlated with % DMA (rs= -0.39, p<0.0001 and rs= -0.33, p<0.001), the ratios of %MMA to % inorgAs (rs= -0.25, p<0.05 and rs= -0.28, p<0.01, respectively) as well as the ratios of % DMA to % MMA (rs= -0.27, p<0.01 and rs= -0.16 not significant, respectively ) in urine for females. For males, the ratio of the concentrations of Hg to As expressed as µg/g cre were also positively and significantly correlated with % inorg As (rs= +0.47, p<0.00001 vs. rs= +0.33, p<0.01, respectively) as well as % MMA (rs= +0.17 not significant vs. rs= +0.15 not significant, respectively), and negatively correlated with % DMA (rs= -0.44, p<0.0001 vs. rs= -0.27, p<0.01, respectively), the ratios of %MMA to % inorgAs (rs= -0.30, p<0.01, vs. rs= -0.19 not significant, respectively) as well as the ratios of % DMA to % MMA (rs= -0.29, p<0.01 vs. rs= -0.23, p<0.05, respectively ) in urine for males. Influence of the relative concentrations of trace elements to arsenic in blood on the percentage of urinary arsenic metabolites. The ratios of the concentrations of Se, Zn, Mn, as well as Cu to As than the corresponding element concentrations in blood were more significantly and positively correlated with % inorg As (rs= +0.36, p<0.001 vs. rs= +0.11 not significant; rs= +0.24, P<0.05 vs. rs= -0.10 not significant; rs= +0.41, p<0.0001 vs. rs= +0.29, p<0.01; and rs= +0.36, p<0.01 vs. rs= +0.06 not significant, respectively), and negatively correlated with % DMA (rs= -0.34, p<0.001 vs. rs= -0.12 not significant; rs= -0.24, p<0.05 vs. rs=-0.005 not significant; rs= - 0.28, p<0.01 vs. rs=-0.15 not Table 8. Comparison of spearman correlation coefficients between (a) the ratio of the concentrations of other element to As or (b) the concentration of element (µg/g creatinine) in bloods and the percentage (%) of urinary As metabolitesof females and males. significant; and rs= -0.26, p<0.05 vs. rs= +0.05 not significant, respectively), and with the ratios of % DMA to % MMA (rs= -0.32, p<0.01 vs. rs=-0.063 not significant; rs= -0.29, p<0.01 vs. rs=-0.049 not significant; rs= -0.24, p<0.05 vs. rs=-0.059 not significant; and rs= -0.22, p<0.05 vs. rs=+0.15 not significant, respectively) in urine for females (Table 8). The ratios of the concentrations of Se or Zn to As than the concentrations of Se or Zn in blood were also more positively and significantly correlated with % MMA (rs= +0.27, p<0.01 vs. rs= +0.033 not significant and rs= +0.26, p<0.05 vs. rs= +0.05 not significant, respectively) in urine for females. Strong correlations also found between the ratios of % MMA to % inorg As in urine and the ratios of the concentrations of Mn or Cu to As in blood (rs= -0.264, p<0.01 and rs= - 0.243, p<0.05, respectively) than the concentrations of Mn or Cu (rs= -0.24, p<0.05 and rs= -0.17 not significant, respectively) in blood for females. For males, better and significant correlations found between the ratios of the concentrations of Se, Zn, Mn, or Cu to As in blood and the percentage of urinary arsenic metabolites than the correlations found between the corresponding element concentrations in blood and the percentage of urinary arsenic metabolites (Table 8). The ratios of the concentrations of Se, Zn, or Mn to As in blood were more positively correlated with % inorg As in urine (rs= +0.25, p<0.05 vs rs= -0.043 not significant; rs= +0.26, p<0.05 vs. rs= +0.036 not significant; and rs= +0.31, p<0.01 vs rs= +0.015 not significant, respectively). The ratios of % MMA to % inorg As in urine were negatively correlated with the ratios of the concentrations of Se, Zn or Mn to As (rs= -0.20 not significant; rs= -0.23, p<0.05; and rs= -0.27, p<0.05, respectively) in blood. But these correlations were not statistically significant with the concentrations of Se, Zn or Mn. The ratios of the concentrations of Se, Zn, Mn, or Cu to As in blood were negatively correlated with the ratios of % DMA to %MMA in blood for males, but these correlations were not statistically significant. Concentrations of trace elements in urine expressed as µg/L versus µg/g cre. The mean concentrations of Cd, Pb, Hg, Se, Zn, Co, Cu, Ni, and Mn expressed as µg/L were not significantly difference between females and males with the exception of Zn concentration, which was significantly higher in urine for males compared to females (p<0.05) (Table 9A). Table 9. Concentrations of trace elements in urine expressed as ‘µg/L’ and ‘µg/g cre’ for females (F) and males (M). However, the mean concentrations of elements (Cd, Hg, Se, Co, Cu, and Mn) after creatinine adjustment expressed as ug/g cre were significantly higher in urine for females compared to males (p<0.01, p<0.05, p<0.001, p<0.001, p<0.01, and p<0.001, respectively) with the exception of Pb, Zn, and Ni concentrations (Table 9B). The mean concentrations (µg/g cre) of Pb, Zn, and Ni were also higher in urines for females compared to males but were not statistically significant. After adjustment, the above urinary elements concentrations expressed as µg/g cre were significantly higher than the concentrations of these elements without adjustment expressed as µg/L in urines for females (Table 9C). This was not so for males with the exception for the mean concentrations of Hg (p<0.01), Se (p<0.01), and Mn (<0.0001) after creatinine adjustment. Concentrations of trace elements expressed as ‘µg/L’ versus ‘µg/g cre’ in urine for low concentrations (<= 50 µg As/L) and high concentrations (> 50 µg As/L) As groups. We found that adjusted urinary As and other trace element concentrations were significantly higher than the unadjusted concentrations in the case of the urine groups with arsenic concentrations less or equal to 50 µg As/L (Table 10A). These differences were not statistically significant for the urine groups with arsenic concentrations >50 µg As/L (Table 10B). The mean urinary creatinine concentrations were significantly low for the lower concentrations As (<= 50 µg/L) in urine groups compared to higher concentrations As (> 50 µg/L) in urine groups (0.32±0.02 g cre/L urine vs. 1.12±0.07 g cre/L urine (p<0.01), respectively). Ages were not significantly difference between these two groups (p=0.12). Table 10. Difference of concentrations of trace elements expressed as ‘µg/L’ versus ‘µg/g cre’ in urine for low concentrations (<= 50 µg As/L urine) and high concentrations (> 50 µg As/L urine) As groups. Influence of creatinine adjustment on urinary trace elements concentrations. The urinary As, Cd, Pb, Hg, Zn, Co, Cu, Ni, and Mn concentrations expressed as µg/L were positively and significantly correlated with the urinary concentrations of these elements expressed as ug/g cre for both females and males with the exception of Se concentrations (Table 11). The correlation between urinary Se concentrations expressed, as µg/L and µg/g cre was not statistically significant for both females and males. Table 11. Spearman correlation coefficients (rs) for urinaryelements concentrations, ‘µg/L’ versus ‘µg/g cre’ forfemales and males. Influence of arsenic and creatinine concentrations on the other trace elements concentrations in urine. Urinary total arsenic concentrations expressed as µg/L were positively and strongly correlated with U-Cre concentrations [rs= +0.86, p<0.01 (n=191 including females and males both), rs= +0.85,