Biological monitoring of exposure to solvents using the chemical itself in urine: application to toluene.

OBJECTIVE: Biomonitoring of solvents using the unchanged substance in urine as exposure indicator is still relatively scarce due to some discrepancies between the results reported in the literature. Based on the assessment of toluene exposure, the aim of this work was to evaluate the effects of some steps likely to bias the results and to measure urinary toluene both in volunteers experimentally exposed and in workers of rotogravure factories. METHODS: Static headspace was used for toluene analysis. o-Cresol was also measured for comparison. Urine collection, storage and conservation conditions were studied to evaluate possible loss or contamination of toluene in controlled situations applied to six volunteers in an exposure chamber according to four scenarios with exposure at stable levels from 10 to 50 ppm. Kinetics of elimination of toluene were determined over 24 h. A field study was then carried out in a total of 29 workers from two rotogravure printing facilities. RESULTS: Potential contamination during urine collection in the field is confirmed to be a real problem but technical precautions for sampling, storage and analysis can be easily followed to control the situation. In the volunteers at rest, urinary toluene showed a rapid increase after 2 h with a steady level after about 3 h. At 47.1 ppm the mean cumulated excretion was about 0.005% of the amount of the toluene ventilated. Correlation between the toluene levels in air and in end of exposure urinary sample was excellent (r = 0.965). In the field study, the median personal exposure to toluene was 32 ppm (range 3.6-148). According to the correlations between environmental and biological monitoring data, the post-shift urinary toluene (r = 0.921) and o-cresol (r = 0.873) concentrations were, respectively, 75.6 microg/l and 0.76 mg/g creatinine for 50 ppm toluene personal exposure. The corresponding urinary toluene concentration before the next shift was 11 microg/l (r = 0.883). CONCLUSION: Urinary toluene was shown once more time a very interesting surrogate to o-cresol and could be recommended as a biomarker of choice for solvent exposure.

Biological monitoring of workers exposed to 4,4'-methylenediphenyl diisocyanate (MDI) in 19 French polyurethane industries.

OBJECTIVES: To study the range of urinary levels of 4,4'-methylenedianiline (MDA), a metabolite of methylenediphenyl diisocyanate (MDI), across factories in the polyurethane industries and to evaluate the validity of this biomarker to assess MDI occupational exposure. METHODS: Workers exposed to MDI, as well as non-occupationally exposed subjects, were studied and pre- and post-shift urine samples were collected from 169 workers of 19 French factories and 120 controls. Details on work activities and practices were collected by a questionnaire and workers were classified into three job categories. The identification and quantification of the total urinary MDA were performed by high-performance liquid chromatography with electrochemical detection (HPLC/EC). RESULTS: For all the factories, MDA was detectable in 73% of the post-shift urine samples. These post-shift values, in the range of <0.10 (detection limit)-23.60 microg/l, were significantly higher than those of the pre-shift samples. Urinary MDA levels in the control group were in the range of < 0.10-0.80 microg/l. The degree of automation of the mixing operation (polyols and MDI) appears as a determinant in the extent of exposure levels. The highest amounts of MDA in urine were found in the spraying or hot processes. The excretion levels of the workers directly exposed to the hardener containing the MDI monomer were significantly higher than those of the other workers. In addition, skin exposure to MDI monomer or to polyurethane resin during the curing step were always associated with significant MDA levels in urine. CONCLUSIONS: Total MDA in post-shift urine samples is a reliable biomarker to assess occupational exposure to MDI in various industrial applications and to help factories to improve their manufacturing processes and working practices. A biological guiding value not exceeding 7 microg/l (5 microg/g creatinine) could be proposed in France.

Determination of N-nitrosodiethanolamine in urine by gas chromatography thermal energy analysis: application in workers exposed to aqueous metalworking fluids.

OBJECTIVE: The aim of this study was to describe a detailed and validated methodology designed for the analysis of carcinogenic N-nitrosodiethanolamine (NDELA) down to sub-microgram/l levels in urine and its application to a number of workers exposed to NDELA-contaminated aqueous metalworking fluids (MWF). METHODS: Following a work-up procedure based on solid-phase extraction of NDELA, the urinary extracts were analysed without derivatization by gas chromatography on a polar wide-bore column with chemiluminescent detection using a thermal energy analyser (TEA). N-Nitroso-(2-hydroxypropyl)amine was used as an internal standard. The method was applied to 12 workers using "nitrite-free" or "nitrite-formulated" MWF and to 15 unexposed subjects. The NDELA content of the MWF was also determined using a similar, but simpler method able to easily quantify NDELA down to at least 0.1 mg/l. RESULTS: Contamination by NDELA traces of some chemicals used for the sample preparation, particularly ethyl formate, must be carefully checked since it can give rise to false-positive results of up to 1 or 2 micrograms/l. The response was linear in the range of 0-500 micrograms/l. Between 0.5 and 10 micrograms/l, the recovery rate was close to 95%, while repeatability ranged from 12.5 to 6.4% (n = 5). The detection limit was 0.3 microgram/l (Signal/noise = 3). No detectable NDELA could be observed in the control workers. There was no significant increase in NDELA levels at the end of shift spot samples from an exposed worker over 1 week. Higher NDELA concentrations were found in two workers (4.3 and 10.7 micrograms/l) exposed to "nitrite-formulated" fluids (contaminated with 65 and 18 mg NDELA per 1, respectively) than in nine workers (range, 0.4-1.3 micrograms/l exposed to "nitrite-free" fluids with lower levels of NDELA (range, 0.5-6.6 mg/l). CONCLUSION: The detailed methodology described in this work and applied to a limited industrial situation was found to be suitable for monitoring NDELA in the urine of workers exposed to aqueous MWF. A much larger screening has been undertaken with the aim of obtaining better information on the real exposure of workers sometimes exposed to "nitrite-formulated" fluids that are still used.

Biological monitoring of workers exposed to 4,4'-methylene-bis-(2-orthochloroaniline) (MOCA). I. A new and easy determination of "free" and "total" MOCA in urine.

OBJECTIVE: The objective of the study was to validate a new and simple method to determine MOCA in the urine of exposed workers. METHODS: The separation, identification and quantification of urinary MOCA were performed in spiked urines by a sensitive and practical high-performance liquid chromatography (HPLC) method and applied to urine samples of 11 workers occupationally exposed to MOCA; the postshift urinary levels of MOCA in their urine samples with and without hydrolysis, "total" and "free" MOCA respectively, were determined. In addition, we investigated the use of citric or sulfamic acid as preservatives of urine samples. RESULTS: The "total" and "free" MOCA were extracted with isooctane from hydrolysed and nonhydrolysed 20-ml urine samples respectively. After evaporation, the residue was dissolved in 4 ml of 2.10(-2) M aqueous hydrochloric acid and analysed by an isocratic HPLC system using both ultraviolet (UV) detection at 244 nm and electrochemical detection working in oxidation mode (0.9 V) with an Ag/AgCl reference electrode. Mobile phase (50% acetonitrile in water containing 0.4% acetate buffer solution pH = 4.6) was used to complete the 20-min analysis. "Free" and "total" MOCA were chromatographed on a reversed-phase C8 column (5 microns, 250 mm x 4 mm). The standard curve of MOCA was linear over the range 5-500 micrograms/l in human urine. The detection limit was 1 microgram/l for a 20-microliter injection volume; the repeatability ranged from 5.6 to 1.3% (n = 6) for spiked urines at 5 and 500 micrograms/l, with a percentage recovery of 94 +/- 3%. The reproducibility of the method was 7.3% (n = 4) for spiked urine at 10 micrograms/l. The use of sulfamic acid as a preservative of urine samples is important to improve the precision and accuracy of the analysis. CONCLUSION: The results indicate that these analytical procedures using conventional apparatus may be used routinely and reliably with large numbers of urine samples for biological monitoring of the exposure to MOCA. The occupational exposure to MOCA in some factories in France is studied in the second part of this work.

Biological monitoring of workers exposed to 4,4'-methylene-bis-(2-orthochloroaniline) (MOCA). II. Comparative interest of "free" and "total" MOCA in the urine of exposed workers.

OBJECTIVES: The objectives of the study were to investigate the best urinary marker of exposure to MOCA in urine samples of exposed workers and to study its applicability in exposure evaluation in polyurethane resin production plants. In addition, our intention was to contribute to the question of biological target value in order to reduce exposure levels. METHODS: Urinary MOCA markers were measured in urine samples collected at the end of the workshifts. 40 workers from four factories were observed for three consecutive days in the same week. "Free" MOCA in non-acid-stabilized urines, "total" MOCA in urines after alkaline hydrolysis, "acid-labile" MOCA in sulfamic or citric acid-protected urines were measured in all urine samples. MOCA liberated by heating the non-acid-stabilized urines ("heat-labile" MOCA) was also measured in 17 urine samples of exposed workers. RESULTS: The median value of "free" MOCA was lower than that of "sulfamic acid-labile" MOCA: 6.0 and 12.5 micrograms/l respectively. The difference between "free" and "total" MOCA was statistically significant (Mann-Whitney test, P = 0.023) for the urine samples collected without acid. Correlations between urinary MOCA concentrations in hydrolysed and non-hydrolysed urine samples were high (r = 0.984, 0.966 and 0.950; P < 0.001) for urine samples with sulfamic acid, with citric acid and without acid respectively. For all the factories, the postshift urinary MOCA concentrations ranged between 1 microgram/l (detection limit) and 570 micrograms/l; 25% of "sulfamic acid-labile" MOCA concentrations exceeded 50 micrograms/l. Workers handling crystallized MOCA excreted the highest amounts of MOCA in urine. The urinary MOCA concentrations (median) were: 84.0 micrograms/l (mixer), 15.5 micrograms/l (moulder). 59.0 micrograms/l (maintenance) and 3.0 micrograms/l (others). CONCLUSIONS: MOCA measured in sulfamic acid-protected urine samples without hydrolysis provides a more practical and reliable biomarker than "total" MOCA (after hydrolysis) or "free" MOCA. A biological guiding value of 20 micrograms/l expressed as "sulfamic acid-labile" MOCA is proposed.

Determination of urinary 4,4'-methylenedianiline and its acetylated metabolites by solid-phase extraction and HPLC analysis with UV and electrochemical detection.

An analytical procedure based on solid-phase extraction and high-performance liquid chromatography using both ultraviolet and electrochemical detection was developed to determine, without derivatization, stable urinary forms of 4,4'-methylenedianiline (MDA) and of its acetylated metabolites at the micrograms/l level, in post-shift urine from 63 exposed workers. The determination of MDA, N-acetyl MDA (MAMDA) and N,N'-diacetyl MDA (DAMDA) was achieved on the non-hydrolysed urine samples, and that of total MDA on urine samples after alkaline hydrolysis. It was necessary to protect urine samples with sulfamic acid in order to stabilize amines and to improve the precision and accuracy of the analyses. Under these conditions, unstable labile conjugates were determined as their parent stable amine. The distribution of total MDA, MDA, MAMDA and DAMDA was assessed in 116 urine samples. Their relative concentrations (arithmetic means) were found to be in the following order: total MDA > MAMDA > MDA > DAMDA. While MAMDA represented more than 50% of total MDA, MDA and DAMDA were lower than 15% and 3% respectively. Acetylation of MDA, described as a possible pathway of detoxication, is confirmed to be an important metabolization route in humans, essentially through the monoacetylated metabolite. However, the individual ratio MAMDA/total MDA was found to vary widely (roughly from 0% to 100%).

trans,trans-Muconic acid, a reliable biological indicator for the detection of individual benzene exposure down to the ppm level.

trans,trans-Muconic acid (2,4-hexadienedioic acid) (t,t-MA) is a minor benzene metabolite which can be used as a biological indicator for benzene exposure. The purpose of the study was to evaluate the limits of use of t,t-MA for detection and quantification of occupational exposures to benzene, particularly on an individual scale, phenol being used as the metabolite of reference. A simple and sensitive method previously described by the authors was carried out to analyse t,t-MA in 105 end-of-shift urinary samples from 23 workers exposed to benzene used as an extraction solvent for "concretes" recovery in the perfume industry. Good correlations were found between atmospheric benzene and both metabolites (uncorrected or corrected for creatinine) or between the metabolites themselves, with correlation coefficients from 0.81 to 0.91 (P < 0.0001). Correlation- coefficients were not improved after correction for creatinine. The overall individual benzene exposure range, median, and arithmetic mean were respectively 0.1-75, 4.5, and 9.0 ppm with corresponding t,t-MA excretion of 0.1-47.9, 5.2 and 8.9 mg/l (uncorrected) and phenol excretion of 1.4-298, 30.9, and 42.2 mg/l (uncorrected). In the control group (145 determinations for t,t-MA and 76 for phenol from 79 individuals) the range, median, and arithmetic mean were respectively < 0.04-0.66, 0.08, and 0.13 mg/l (uncorrected t,t-MA) and 1.5-42.0, 9.85 and 11.3 mg/l (uncorrected phenol). t,t-MA was far more specific than phenol and could be easily and practically used to estimate with a given probability the upper or lower corresponding benzene concentrations down to around the ppm level.(ABSTRACT TRUNCATED AT 250 WORDS)

Improvement in HPLC analysis of urinary trans,trans-muconic acid, a promising substitute for phenol in the assessment of benzene exposure.

Urinary trans,trans-muconic acid (t,t-MA), a minor metabolite of benzene, is a potential candidate for biological monitoring of benzene. A clean-up procedure using SPE extraction cartridges was applied to urinary samples in order to improve the reliability of t,t-MA determinations by HPLC-UV greatly and to carry out convenient analyses on a routine scale, particularly at low levels of t,t-MA concentrations. The detection limit of the method is low enough to measure urinary t,t-MA at a concentration of 0.05-0.1 mg/l. The recovery rates and relative standard deviations from spiked urines (1 mg/l to 20 mg/l) were about 90% and 5%, respectively. t,t-MA was found to be rapidly excreted by rats and humans. In rats the background range never exceeded 0.5 mg/l with a mean concentration around 0.3 mg/l. In 49 human blank urines, t,t-MA average and median-value were respectively around 0.2 and less than 0.1 mg/l with a range of less than 0.1 to 0.5 mg/l. Experimental exposure of rats for 1 h to 10.2 ppm of benzene induced urinary excretion of 13 mg/l of t,t-MA during a 6-h post-exposure period while occupational exposures to 2.6 ppm (mean exposure level during 5 d-8 h) and 7 ppm (4 h) of benzene resulted in urinary excretion of 2.1 (mean excretion level) and 6.5 mg/l respectively at the end of the exposure. In humans, t,t-MA has a similar half-time as phenol. Analysis of urinary t,t-MA seems to be a better indicator than phenol for the assessment of exposure to low levels of benzene. Ingestion of 200 mg of sorbic acid, the only other known precursor of t,t-MA, interfered minimally with the background excretion of t,t-MA.