Risk Assessment Study of Fluoride Salts: Probability-Impact Matrix of Renal and Hepatic Toxicity Markers.

The present risk assessment study of fluoride salts was conducted by oral administration of three different doses of sodium and potassium fluorides (NaF, KF) and zinc fluoride tetrahydrate (ZnF2 •4H2O) to male Wistar rats. The rats were divided into control and nine experimental groups, to which oral injections of 0.5 mL distilled water and 0.5 mL of fluoride solutions, respectively, were given. The dosage of fluoride compounds was adjusted to contain 2.1 mg (low-dose group, LG), 4.3 mg (mid-dose group, MG), and 5.4 mg fluoride per 200 g rat body weight (high-dose group, HG) corresponding to 5, 10, and 12.5 % of LD50 values for NaF. The 24-h urine volume, N-acetyl-ß-D-glucosaminidase (NAG) and creatinine clearance (Ccr) were measured as markers of possible acute renal impact. The levels of alanine aminotransferase (ALT) and aspartate aminotransferase (AST) were determined in serum samples as markers of acute hepatic impact. The levels of serum and urinary fluoride were determined to evaluate fluoride bioavailability. The results reveal that higher doses of NaF, KF, and ZnF2 induced renal damage as indicated by higher urinary NAG (p < 0.05 with ≥90th percentile of control). High doses of ZnF2 also induced a significant Ccr decrease (p < 0.05 with ≤10th percentile of control). Low doses of NaF and mid-doses of ZnF2 induced polyuria (p < 0.05 with ≥90th percentile of control) while medium doses of NaF and low doses of KF also induced liver damage, as indicated by a high level of AST (p < 0.05 with ≥90th percentile of control). These findings suggest that oral administration of fluoride is a potential, dose-dependent risk factor of renal tubular damage.

Comparison of the Biological Impacts of the Fluoride Compounds by Graphical Risk Visualization Map Technique.

Various fluoride compounds are widely used in industry. The present risk assessment study was conducted using a series of inorganic binary fluorides of the type XFn, where X(n) = Na(+), K(+), Li(+), Mg(2+), Ca(2+), Sr(2+), Ba(2+), Al(3+), Nd(3+), La(3+), Ce(3+), Sm(3+), Gd(3+), Y(3+), Yb(2+), and Zn(2+). The aqueous solutions of these salts were orally administrated to 16 experimental groups (one for each of the salts tested). The levels of fluoride, N-acetyl-ß-D-glucosaminidase in cumulative 24-h urine samples and creatinine clearance were measured to assess possible acute renal damages. The levels of fluoride, alanine aminotransferase, and aspartate aminotransferase were also determined in serum samples to assess possible acute hepatic damages. The results reveal that sodium fluoride (NaF), potassium fluoride (KF), and zinc fluoride tetrahydrate (ZnF2 (.)4H2O) can carry the fluoride ion into the bloodstream and that it is excreted via urine more readily than the other compounds tested. These fluorides were assigned the highest risk impact factor. Most of the rare earth fluorides are insoluble in water while those groups 2 and 13 of the periodic table are slightly soluble, so that they do not have a significant negative risk. These findings suggest that the biological impact of fluoride depends on the accompanying counter ion and its solubility. The risk map obtained in the present study shows that the graphical visualization map technique employed is a valuable new tool to assess the toxicological risk of chemical compounds.

Risk assessment visualization of rubidium compounds: comparison of renal and hepatic toxicities, in vivo.

Rubidium has been considered to be nontoxic. Its use includes thin film on glass deposition and as medical contrast medium. Recent technology innovations also involve the use of rubidium, but there is limited information about the biological effects of its various compounds. In the present risk assessment study, a series of rubidium compounds with different counter anions-acetate, bromide, carbonate, chloride, and fluoride-were orally administrated in a single dose to several groups of rats. Cumulative 24-h urine samples were obtained, and the levels of rubidium, fluoride, N-acetyl-ß-D-glucosaminidase and creatinine were measured to evaluate possible acute renal effects. Daily samples of serum were also obtained to determine the levels of aspartate and alanine aminotransferases to assess possible acute hepatic effects. Urinary rubidium excretion recovery of 8.0-10.5% shows that urine can be a useful diagnostic tool for rubidium exposure. The present results reveal that rubidium shows different biological effects depending on the counter anion. A pattern of large significant NAG leakage and elevation of ALT observed in rats treated with anhydrous rubidium fluoride indicates renal and hepatic toxicities that can be attributed to fluoride. The techniques reported in this study will be of help to assess the potential risks of toxicity of rubidium compounds with a variety of anions.

[Comparison of diagnostic criteria for the metabolic syndrome among Japanese university faculty].

OBJECTIVES: Diagnostic criteria for the metabolic syndrome (Mets) in Japan have been set by the Medical Committee of the Japanese Association of Medical Sciences (Med), the National Health and Nutrition Examination Survey (Nat), specific health checkups (Ckup), and second medical examination by Worker's Accident Compensation Insurance System (Wor). The purpose of this study was to compare classification of the metabolic syndrome by different organizational criteria and to investigate underlying differences. METHODS: All faculty members of a university in Osaka, Japan, underwent mandatory health checkups in September 2008. The demographic distribution included 769 males (mean age, 49 +/- 12 years) and 415 females (mean age, 43 +/- 10 years). Using the Med, Nat, Ckup and Wor criteria, individuals were assessed for the MetS and pre-metabolic syndrome (pre-Mets), strongly suspected metabolic syndrome (S-Mets) and assumed pre-metabolic syndrome (A-pre-Mets), as well as a positive support level (PSL) and a motivational support level (MSL). All faculty members were categorized into a morbid group (Mets, S-Mets, PSL, and FB) or a pre-morbid group (pre-Mets, A-pre-Mets, and MSL) based on medical data and smoking habits. The incidence of morbid and pre-morbid individuals was compared across the four criteria and analyzed based on gender and age (under 40 and 40 or over). RESULTS: Male incidences for the morbid and pre-morbid classifications were 17% and 20% with Med, 9% and 23% with Nat, 27% and 14% with Ckup, and 1.4% and 0% with Wor. There were significant differences across criteria sets in both the morbid and pre-morbid groups, with significantly greater numbers of males than females, and higher prevalences in those aged 40 or over than in their younger counterparts. Males aged under 40 classified into the pre-morbid group comprised 18% in Med, 16% in Nat, and 13% in Ckup. CONCLUSION: The different disease incidences found between Med and Ckup data in males aged 40 or over might be attributed to varying criteria for blood glucose levels, while Wor data may be influenced by the higher level of blood pressure set as a criterion with this approach. It will be important to continuously validate currently established criteria to identify the actual prevalence of MetS in Japan. Furthermore, incorporation of waist circumference and BMI for females, and a positive approach for young males, may be critical for future developments.

Toxicokinetics and metabolism deteriorated by acute nephrotoxicity after a single intravenous injection of hydrofluoric acid in rats.

OBJECTIVES: This study was designed to investigate the early dynamic state of hydrofluoric acid (HFA) in blood and urine as a model of accidental occupational exposure to a subtoxic dose of HFA. It was also aimed at determining the relationship between the kinetics and harmful effects of HFA on the kidney. METHODS: Rats received a single intravenous injection of HFA (3.2, 6.4, or 9.6 (LD(5)) mg/kg) or saline. The volume of each injection was 1 ml and the concentrations of HFA were 0.1, 0.2, and 0.3%, respectively. Ionized fluoride (F) was measured for the biological monitoring of HFA. Serum F concentrations were determined at 0, 5, 10, 30, 60, 120, and 300 min. Pharmacokinetic parameters were calculated with two-compartment modeling. Urine was directly collected from bladder for 300 min to determine the extent of the renal damage. RESULTS: AUC(0→300) values were significantly higher in the 9.6 mg/kg group than in the 3.2 and 6.4 groups. The total body clearance, V(1), V(2) and V(ss) were significantly lower in the 6.4 and 9.6 mg/kg groups than in the 3.2 mg/kg group. These results indicate that HFA was retained in blood. This could be a result of renal dysfunction. NAG/Cr and glucose excretion amount in urine were increased, and the clearance rate of F, urine volume and excretion amounts of electrolytes were decreased in the 9.6 mg/kg group compared with the saline group. These findings indicate renal tubular damage and a decrease in the amount of excretion of HFA from the kidney. CONCLUSIONS: We consider that acute nephrotoxicity of HFA caused renal injury, and the harmful effects of HFA were subsequently aggravated by its delayed metabolism.

Effects of sodium monofluoroacetate on glucose, amino-acid, and fatty-acid metabolism and risk assessment of glucose supplementation.

Sodium monofluoroacetate (SMFA; also known as compound 1080) is a highly toxic chemical; therefore, accidental exposure and intentional misuse are of great concern. SMFA intoxication is reportedly caused by the inhibition of aconitase. However, the pathogenesis underlying SMFA intoxication is not clear. This study was conducted to elucidate the acute effects of SMFA on glucose, amino-acid, and fatty-acid metabolism and to assess glucose supplementation as a possible alleviator or aggravator in SMFA intoxication. Rats were assigned to three groups: SMFA+saline, SMFA+glucose, and control (i.e., no SMFA), and blood samples were analyzed at 3 hours after SMFA or saline (control) administration. Additional rats were used for the monitoring of blood-glucose and lactate concentrations for 10 hour- and 14-day survival rates. SMFA increased the serum-citrate, serum-pyruvate, and blood-lactate concentrations. However, despite significant increases in these parameters when SMFA was administered with glucose, the effects on pH values were small and the survival rate was not changed. SMFA also increased the serum concentrations of free fatty acids, branched-chain amino acids, ammonia, urea, and calcium. The presence of glucose enhanced or suppressed these metabolic changes. Amphibolic intermediates in the tricarboxylic acid cycle might be supplied through the catabolism of proteins in SMFA intoxication. We conclude that other factors, in addition to the accumulation of lactate, citrate, and pyruvate, may affect survival rates, and that SMFA induces imbalances in glucose, amino-acid, and, fatty-acid metabolism. All these changes are inter-related and may contribute to SMFA intoxication.

Time-dependent changes of blood parameters and fluoride kinetics in rats after acute exposure to subtoxic hydrofluoric acid.

OBJECTIVES: In our previous study, we reported that even a sublethal dose of hydrofluoric acid (HFA) could cause acute toxic effects 60 min after intravenous injection. This study was designed to investigate the time- and dose-dependent changes associated with these disorders. The serum fluoride (F) kinetics are also considered in the discussion of the relationship between the concentrations of serum F and the disorders. METHODS: Rats were injected with HFA (1.6 or 9.6 mg/kg body weight) for the dose-response relationship study. For each dose, the rats were assigned to one of seven groups. Blood samples of the 0-min group were obtained from the carotid artery prior to injection as a control. The other six groups were labeled according to sampling times (5, 10, 30, 60, 120 and 300-min) in the time-dependent study. RESULTS: The 1.6 mg/kg dose decreased the ionized calcium (Ca2+) level significantly after 30 min, and it also decreased the total calcium (Ca) level after 300 min. The 9.6 mg/kg dose rapidly worsened renal dysfunction after 60 min. It increased the serum potassium level after 60 and 120 min and it decreased Ca and Ca2+ levels until 300 min. Although there was respiratory compensation, the base excess and HCO3(-) level and had not completely recovered by 300 min. CONCLUSIONS: Even low exposure to HFA caused renal dysfunction, and electrolyte abnormalities and metabolic acidosis lasted for several hours in rats. Therefore, persons involved in HFA accidental exposure should be closely monitored over time, even if the exposure is less than the sublethal dose.

Abnormalities in cadmium fluoride kinetics in serum, bile, and urine after single intravenous administration of toxic doses to rats.

Cadmium fluoride (CdF2, CdF for short) is the most lethal and hepatotoxic of all Cd-containing compounds. The toxic effects of CdF appear to depend on its detoxification and elimination. This study was designed to determine the early dynamics of the absorption, systemic distribution, and metabolism of CdF. The kinetics of cadmium and fluoride were investigated in the blood, bile, and urine of rats as a model of accidental occupational exposure to CdF. The serum concentration-time profiles measured after intravenous CdF (1.34, 2.67 or 4.01 mg/ per kg body weight) administration were analyzed by compartmental modeling using the WinNonlin program. Bile and urine were collected for 300 min after the administration. The kinetic profiles indicate that the clearance of Cd was diminished in the 2.67 and 4.01 mg/kg groups, leading to a persistently high serum Cd level. The mean total biliary excretions of Cd in the 2.67 and 4.01 mg/kg groups were significantly higher than that in the 1.34 mg/kg group. The abnormal kinetics of Cd was attributable to severe hepatic injury that diminished the capacity for Cd accumulation. The elimination of serum F was delayed in the 4.01 mg/kg group. The mean urinary F excretion amount was not significantly higher in the 4.01 mg/kg group than in the 2.67 mg/kg group. The abnormal kinetics of F was attributable to nephrotoxicity that diminished its elimination from the kidney.

Harmful effects and acute lethal toxicity of intravenous administration of low concentrations of hydrofluoric acid in rats.

The acute toxicity of hydrofluoric acid (HFA) was investigated in a 24-h lethal dose study of intravenous infusion in rats. The lethal dose lowest (LDLo) and LD50 were 13.1 and 17.4 mg/kg, respectively. Harmful systemic effects were also studied 1 h after acute sublethal exposure to HFA. The maximum dose was set at 9.6 mg/kg (LD5). Rats were injected with HFA (1.6, 3.2, 6.4 or 9.6 mg/kg), saline, sodium fluoride (NaF) or HCl solution. NaF and HCl solution concentrations corresponded to the F- and H+ concentrations of 9.6 mg/kg HFA. Blood urea nitrogen (BUN) and Cr were significantly increased in response to HFA concentrations greater than 3.2mg/kg. Acute glomerular dysfunction also occurred at HFA concentrations greater than 3.2 mg/kg. HCO3- and base excess (BE) were significantly decreased in the 6.4 and 9.6 mg/kg groups. Ca2+ was significantly decreased, and K+ was increased in the 9.6 mg/kg group. BUN was significantly increased in the NaF and HFA groups and was increased in the HFA group compared with that in the NaF group. Cr was significantly increased in the HFA group only. HCO3- and BE were significantly decreased in the NaF and HFA groups and were increased in the HFA group compared with values in the NaF group. Ca2+ was significantly decreased in the NaF and HFA groups, and K+ was significantly increased in the NaF and HFA groups. F- exposure directly affected serum electrolytes. Mortality was thought to be due to cardiac arrhythmia resulting from hypocalcemia and hyperkalemia. Metabolic acidosis and renal failure were more severe in response to HFA exposure than in response to NaF exposure because of more free F-, which has strong cytotoxicity, in the HFA group than in the NaF group. Lethal effects of HFA are promoted by exposure routes such as inhalation that cause rapid absorption into the body. Even low exposure to HFA can cause acute renal dysfunction, electrolyte abnormalities and metabolic acidosis. These complications result in a poor prognosis.

Strong acute toxicity, severe hepatic damage, renal injury and abnormal serum electrolytes after intravenous administration of cadmium fluoride in rats.

Cadmium fluoride (CdF) is commonly used as an insulator for ulta high speed mass telecommunications equipment, and there is a considerable risk that industrial workers will inhale CdF particles. Despite the possibility that acute exposure can cause harmful systemic effects, there are no studies to date that address the health consequences of acute CdF exposure. This study therefore aimed to determine the acute lethal dose of CdF and its effects on various target organs, including the liver and kidney. We also determined the effect of CdF on serum electrolytes and acid-base balance. The effective lethal dose was determined and dose-response study was conducted after intravenous administration of CdF in rats. The 24 h LD(50) of CdF was determined to be 3.29 mg/kg. The dose-response study used doses of 1.34, 2.67, 4.01 mg/kg CdF. Saline or sodium fluoride solution were used for controles. Severe hepatocellular injury was induced at doses greater than 2.67 mg/kg, as demonstrated by AST and ALT activities greater than 1,500 IU/l in rats injected with a dose of 4.01 mg/kg. Acute renal failure was induced at doses greater than 2.67 mg/kg. Decreased serum Ca, increased serum K and metabolic acidosis were induced at a dose of 4.01 mg/kg. Decreased serum Ca was caused by exposure to ionized F. CdF has the strongest lethal and hepatic toxicity among all Cd containing compounds.

Acute lethal toxicity, hyperkalemia associated with renal injury and hepatic damage after intravenous administration of cadmium nitrate in rats.

Cadmium nitrate Cd(NO(3))(2) (CdN) is commonly used in Ni-Cd battery factories. The possibility of accidental exposure to CdN is great. CdN is very soluble in water compared to other Cd compounds. Therefore, acute toxicity would be expected to be quick due to rapid absorption after exposure. However, the mechanisms of CdN toxicity have not been fully elucidated. We investigated the acute lethal toxicity and harmful systemic effects of acute exposure to large doses of CdN. The lethal dose and dose-response study of the liver and kidney were determined after intravenous administration of CdN in rats. The LD(50) of CdN was determined to be 5.5 mg/kg. Doses of 2.1, 4.2, 6.3 mg/kg were selected for the dose-response study. Liver injury was induced at doses greater than 4.2 mg/kg. Severe hepatic injury occurred in the 6.3 mg/kg group, which would have been caused by acute exposure to the high concentration of Cd that exceeded the critical concentration in hepatic tissue. A remarkable decrease in urine volume in the 6.3 mg/kg group indicated acute renal failure. A decrease in creatinine clearance suggested acute glomerular dysfunction at doses greater than 4.2 mg/kg. Increases in urinary N-acetyl-beta-D-glucosaminidase/creatinine, beta(2)-microglobulin and glucose in the 6.3 mg/kg group indicated proximal tubular injury. Secretion of K ion was also severely affected by proximal tubular injury and severe decreases in urine volume, and an increase in serum K ion was identified at doses greater than 4.2 mg/kg. Thus severe hyperkalemia might be associated with the cardiac-derived lethal toxicity of CdN.

An overview of boron, lithium, and strontium in human health and profiles of these elements in urine of Japanese.

The biological, medical and environmental roles of trace elements have attracted considerable attention over the years. In spite of their relevance in nutritional, occupational and toxicological aspects, there is still a lack of consistent and reliable measurement techniques and reliable information on reference values. In this review our understandings of the urinary profilings of boron, lithium and strontium are summarized and fundamental results obtained in our laboratory are discussed.Over the past decade we have successfully used inductively coupled plasma emission spectrometry for the determination of reference values for urinary concentrations of boron, lithium and strontium. Taking into account the short biological half-life of these elements and the fact that their major excretion route is via the kidney, urine was considered to be a suitable material for monitoring of exposure to these elements. We confirmed that urinary concentrations of boron, lithium and strontium follow a lognormal distribution. The geometric mean reference values and 95% confidence intervals were 798 µg/l (398-1599 µg/l) for boron, 23.5 µg/l (11.0-50.5 µg/l) for lithium and 143.9 µg/l (40.9-505.8 µg/l) for strontium. There were no discrepancies between our values and those previously reported. Our reference values and confidential intervals can be used as guidelines for the health screening of Japanese individuals to evaluate environmental or occupational exposure to these elements.

Survey of strontium in mineral waters sold in Japan: relations of strontium to other minerals and evaluation of mineral water as a possible dietary source of strontium.

The concentrations of strontium, calcium, and magnesium in 33 brands of natural mineral waters commercially available in Japan were determined by inductively coupled plasma-atomic emission spectrometry. The geometric mean values were 94.4 microg/L for strontium, 19.1 mg/L for calcium, and 2.82 mg/L for magnesium. Wide confidence intervals of 1.96-4539 microg/L for strontium, 0.865-421 mg/L for calcium, and 0.064-123 mg/L for magnesium were observed. The significant linear relationships among the three elements over a wide distribution range suggest that the synchronized variations of these elements are regulated by the natural ecosystem and not from accidental contamination from human activities or exceptionally high natural sources. Using the results of multiple linear regression analysis, the strontium concentration can be predicted by that of calcium with the appropriate power function. The results of this study suggest that mineral water can be an important nutritional source of strontium. As trace elements imbalance is often found in older patients with chronic renal failure, we propose that close attention of trace elements intake from trendy foods or beverages is necessary to prevent this hidden problem of a rapidly aging society.

Lethal acute lung injury and hypoglycemia after subcutaneous administration of monochloroacetic acid.

Hypoglycemia is suspected in the acute lethal toxicity induced by cutaneous exposure to monochloroacetic acid (MCA). Although it has been shown that hepato-renal dysfunction is involved, the mechanism and the target organs that directly affect mortality remain to be determined. We suspected respiratory failure as a main cause of death in some reported cases. We investigated dose-response effects, hypoglycemia, and lung injury in rats exposed to MCA. Serum glucose, blood gases, and parameters of alveolar injury in bronchoalveolar lavage fluid (BALF) were analysed 2 and 4 h after subcutaneous administration of MCA (108, 135 or 163 mg/kg). Apparent pulmonary injury and hypoglycemia were not identified 2 h after administration, but lactate dehydrogenase (LDH) and total cells in BALF were dose-dependently increased; and severe hypoglycemia was identified 4 h after administration. Blood gas analysis showed remarkable alveolar gas dysfunction as exchange in the 163 mg/kg group. Thus, hypoglycemia and lung injury appear to cause death in response to MCA exposure.

Fatality due to acute fluoride poisoning in the workplace.

We report a case of a 65-year-old worker who suffered a third-degree skin burn to 5% of his total body surface area as a result of being splashed in the face with hydrofluoric acid (HF). He died shortly thereafter without having received adequate first aid. His serum fluoride concentration was markedly increased at 6.38 mg/dl with hypocalcemia and hyperkalemia. We concluded that he died of HF poisoning. In this case he might have managed to avoid death if he had not been working alone and if he had received adequate first aid on the scene as soon as possible. We re-emphasize the need for the immediate initiation of first aid on the scene and the distribution of information on the risks of HF to workers.

Hepatic injury and gluconeogenesis after subcutaneous injection of monochloroacetic acid in rats.

OBJECTIVE: Monochloroacetic acid (MCA) is corrosive to skin, and causes not only chemical injury but also fatal systemic poisoning. Little is known about the cause of death. We studied the acute toxicity of MCA before clinical symptoms appeared in fasting rats. METHODS: Blood samples were analyzed 2 h after subcutaneous MCA injection (Ld(90): 162 mg/ml kg body weight). Control rats were injected with saline. RESULTS: Aspartate aminotransferase (AST) and alanine aminotransferase (ALT) were about 1.5-fold higher than in the controls, and mitochondrial AST (mAST) was 2-fold higher. Blood urea nitrogen and creatinine were significantly increased, while serum glucose was significantly decreased in the treated group. Lactate was 6-fold higher and pyruvate was 13-fold higher than in the controls. CONCLUSIONS: MCA caused injury to the liver and kidneys but these injuries were slight. However, the larger increase in mAST indicated that hepatocellular mitochondria were selectively targeted. Hepatocellular mitochondrial injury decreased gluconeogenesis and caused hypoglycemia and extremely high levels of lactate and pyruvate. Hypoglycemia and lactic acidosis were insidious before the critical symptoms appeared and this combination accelerated to death, affecting other organs such as the heart and brain. Nosotropic therapy of these abnormalities up to the appearance of symptoms may help to establish an early therapy for skin exposure to MCA.

X-ray fluorescence measurement of bone calcium, phosphorous and fluoride in a rat osteoporosis model.

The fluoride (F), calcium (Ca) and phosphorous (P) content in two parts of the bone mandible, lumbar spine and femur of rats fed a low mineral diet and corresponding controls were determined by X-ray fluorescence. The anterior part of the mandible, including teeth was labeled as part I while the posterior part, including the mandibular head but excluding the teeth was designed as part II. The bone Ca, P and F content of part I was the same for the experimental and control groups. The Ca, P and F content of part II was decreased in the low-mineral diet group (p < 0.01). The values for F were considerably lower than in the controls. The lumbar spine and femur were used as whole bones and showed a significantly lower values for the three elements measured in the experimental group (p < 0.01). Stress or other stimuli from mandibular movement and occlusal force may be involved in preventing bone density decline, while density loss could be a result of lower physical stimulation, which may exert a greater effect than the level of fluoride in bones. It is possible that the absence of significant difference in fluoride levels in part I of the mandible may have been caused by its accumulation in teeth.

Markers of cadmium exposure in workers in a cadmium pigment factory after changes in the exposure conditions.

The objective of this study was to assess changes in concentrations of cadmium in the blood (Cd-B), cadmium in the urine (Cd-U), beta2-microglobulin in the serum (beta2-mG-S) and beta2-microglobulin in the urine (beta2-mG-U) of workers at a cadmium (Cd) pigment factory in Japan in which exposure conditions improved. We evaluated reversibility of these markers in continuously employed workers in relation to changes in exposure levels resulting from improvements in the workplace and the reduced production of Cd. Our study involved both environmental and biological monitoring. Data were collected for four years. We measured the Cd concentration in the air of each work area, using the time-weighted average (TWA). Cd-B and Cd-U were measured in workers as direct indices of Cd exposure. beta2-mG-S and beta2-mG-U were measured as markers of renal tubular function. Exposure levels were high in all work areas, according to the criteria set by the American Conference of Governmental Industrial Hygienists (ACGIH). Workers' Cd-B and Cd-U concentrations reflected high levels of exposure. Correlation was found between these direct indices and beta2-mG-S concentrations. Since the second year, ambient Cd concentrations decreased and reacted markers have been improved. Our results suggest that Cd-B, Cd-U, beta2-mG-S and beta2-mG-U are appropriate markers for monitoring both the level of Cd exposure and the tubular function of workers. Reversibility of urinary low molecular weight protein was observed in the workers over the four years.