J. Cocker, P. Brough, J. Morton, Health & Safety Laboratory, Sheffield, United Kingdom.
Regular biological monitoring at a magnet manufacturing company showed raised urinary cobalt levels and prompted an occupational hygiene investigation to assess exposure and recommend improved controls. Personal air sampling was undertaken following HS(G)173 and MDHS14/3. Analysis was undertaken following MDHS 14/3 (gravimetry) and MDHS 30/2 (X-ray fluorescence). For biological monitoring purposes, workers provided urine samples at the end of work each day for eight consecutive workdays and levels of cobalt and other relevant metals were determined. Air monitoring indicated poor control. Five of the 11 results were in excess of the Workplace Exposure Limit (WEL) for cobalt (0.10 mg m-3 8hr TWA). The remaining six were all below the WEL, but five of these were in excess of 50% of the WEL for cobalt (i.e. >0.05 mg m-3 8hr TWA). Only one sample provided a result less than 50% of the WEL (i.e., <0.05 mg m-3 8hr TWA). Levels of cobalt in urine samples from 17 workers ranged from 7.2 to 778.5 µmol/mol creatinine and were, in the most part, below the German EKA value of 100 µmol/mol of creatinine, with a mean of 66.3 µmol/mol creatinine and a 90th percentile of 114.5 µmol/mol creatinine. The principal route of exposure to cobalt in the main workshop and adjacent areas was via inhalation. Incidences of BM results in excess of the German EKA Value may be attributed primarily to exposure via inhalation with possible additional exposure via accidental ingestion caused by poor employee hygiene and failure to use, or correctly use, PPE/RPE.
K. Ibrahim, J. Sparer, M. Stowe, C. Fleming, I. Collin-Hansen, C. Redlich, Y. Liu, Yale University, New Haven, CT.
Polyurethane paints utilizing isocyanate curing agents are widely used in autobody repair. Autobody workers are frequently in contact with recently painted automobile parts, which have been assumed to be fully cured. Tasks that are often performed on recently painted automobiles may include sanding, buffing and unmasking surfaces, posing a risk for isocyanate dermal exposure if uncured isocyanate groups (NCO) are present. The purpose of this study was to investigate the curing time of aliphatic isocyanates used in autobody refinishing and to assess factors that may impact NCO group availability such as paint formulation, curing method, temperature, and humidity. Twenty-one metal or plastic automobile pieces in four different shops were coated using four different brands of primer or clear coat and cured using heat mechanisms (e.g., baking in a booth, heat lamp) or air. Surface isocyanates were then evaluated using CLI Surface SWYPESM (a 225 cm2 pad that changes color upon contact with aliphatic isocyanates, indicating a positive test). Automobile pieces were evaluated during a period of hours to weeks after curing, utilizing 10 firm wipe strokes. The resulting color intensity was rated on a scale of 0–5 with 5 representing the highest isocyanate content. Geometric mean curing time was 40.9 hours (range: 0.8–640.0) across all parts, and 43.5 (range: 0.8–640.0) vs. 36.2 (range: 5.3– 334.5) for baking vs. air-dried. Free NCO groups remained present on the majority of sampled automobile surfaces for a substantial length of time: up to 120.2 hours for typical formulations and nearly a month for other formulations. This study demonstrates a potential risk for dermal exposure to uncured isocyanates among autobody workers.
M. Henriks-Eckerman, Finnish Institute of Occupational Health, Turku, Finland; T. Tuomi, R. Jolanki, R. Riala, K. Suuronen, Finnish Institute of Occupational Health, Helsinki, Finland.
Water-based metalworking fluids (MWF) may cause irritant and allergic contact dermatitis. Alkanolamines, used as corrosion inhibitors and pH adjusters, are known skin and respiratory irritants and sensitizers. The aim of this study was to measure overall exposure, both dermal and airborne, to MWFs by using alkanolamines as markers. The exposure measurements were carried out in nine machine workshops. After a 2-hour working period, the operating hand of 34 machinists was rinsed for 1 minute using 200 ml of 20% isopropanol in a plastic bag. After that, the machinists washed their hands with soap and warm water, and then blank samples were collected by using the same rinse-off method. Personal air samples were also collected during the 2 working hours onto acid-treated glass fiber filters. The filter samples were desorbed with methanol and analyzed by liquid chromatography with mass spectrometric detection (LC-MS). The rinse-off samples were also analyzed for alkanolamines by LC-MS. Most of the machinists used MWFs containing both monoethanolamine (EA) and triethanolamine (TEA). The overall exposure to alkanolamines was estimated by calculating the amounts absorbed by the airways and by the skin of the working hand. In this way, we showed that the exposure to EA takes place both through the airways as well as through the skin. The exposure to TEA was shown to be mainly dermal. The dermal exposure measurement was as easy to perform as the personal air sampling. Measurement of dermal exposure to alkanolamines is important in order to estimate the overall exposure to MWFs and to assess which preventive measurements are most efficient in reducing the exposure. This hand-rinsing method can also be used to determine the efficiency of protective gloves.
Y. Huang, C. Tu, National Kaohsiung First University of Science and Technology, Kaohsiung, Taiwan Republic of China.
Biomonitoring of occupational exposures is often compared with time-weighted average (TWA) measurements of exposure levels to evaluate the exposure-response relations. The assessment approach assumes a consistent exposure level and omits potential variations in exposure levels that might affect the biological monitoring outocme. From the perspective of body burden and clearance, if occupational exposure levels vary substantially over time, the exposure biomarker levels may also fluctuate over time, causing substantial uncertainties in the biomitoring outcome if data is only collected at the end of the work shift. In this study, the variability of workplace exposures to N,N-dimethylformamide (DMF) at a synthetic leather plant was evaluated and compared with monitoring outcome. Personal exposure levels were measured with passive sampler for the TWA levels. In addition, selected workers also wore a real-time organic solvent detector connected to a data logger, which recorded the variation of ambient DMF levels. Urine samples were collected at the beginning of the workday and before the end of the work shift. In addition, saliva samples were collected every two hours for one workday. The urine and saliva samples were analyzed for DMF levels and two of its metabolites, N-methylformamide (NMF) and N-acetyl-S-(N-methyl-carbamoyl)cysteine (AMCC). Results of the DMF monitoring showed a substantially high level of variability in detected DMF level. Although all TWA concentrations were below the 10 ppm regulatory limit, substantially different exposure patterns were detected among workers and across work days. In addition, higher urine and salivary DMF levels were found with end-of-shift samples among workers with elevated exposure levels closer to the end of the work shift. The occurrence frequency and level of higher occupational exposure levels might better explain the relation between occupational exposure and biomonitoring outcome. Further study on the extent of influence from exposure variability on biomonitoring outcome is warranted.
K. Krishnamoorthy, University of Louisiana-Lafayette, Lafayette, LA; T. Mathew, University of Maryland-Baltimore County, Baltimore, MD.
For the application of biomonitoring to occupational health practice, it is crucial that appropriate statistical methodology be used for the modeling and analysis of the relevant exposure data. This talk aims to develop a statistical framework for the biomonitoring scenario for situations where simple linear regression models can adequately model the relationship between a biomarker and the exposure level. For the problem of biomonitoring based on a simple linear regression model, we identify two classes of problems: the problem of prediction, and the problem of calibration or inverse regression. For example, for predicting urinary benzene levels from airborne benzene, we can treat the airborne benzene as the independent variable x and the urinary benzene as the dependent variable y. We now have a prediction problem, where it will be necessary to quantify the uncertainty in the prediction using a prediction interval. Also of interest is the computation of a tolerance interval or a simultaneous tolerance interval that is expected to contain most of the urinary benzene levels. As an example of the calibration problem, suppose we want to estimate the blood lead level (say, x) among exposed workers. When x is difficult, expensive, or time-consuming to measure, we can use a related variable y that is quicker and cheaper to measure, for example, saliva lead levels. Knowing the linear regression relationship between the two variables, we can estimate the blood lead level after measuring the corresponding saliva lead level. The uncertainty in the estimate can be quantified using appropriate confidence intervals. This talk will briefly discuss the relevant statistical methodology for addressing the above problems. The results will be illustrated using the analysis of exposure data.
S. Gaffney, D. Paustenbach, ChemRisk Inc., San Francisco, CA.
The OSHA benzene standard states that if a worker is exposed to benzene in an emergency situation and his/her end-of-shift urine contains phenol levels greater than 75 mg/L adjusted for specific gravity, further medical evaluation is required. ACGIH has published BEIs for benzene exposure of 500 µg t,t-muconic acid (ttMA)/g creatinine adjusted and 25 µg s-phenylmercapturic acid/g creatinine. However, much of the published literature puts into question the usefulness of these tests for persons exposed to less than 5 ppm in an 8-hour workday. A voluntary study was conducted where a worker was exposed to a mixture of chemicals for four days. The 8-hour TWA benzene exposures ranged from 0.1 to 0.83 ppm (mean = 0.47 ppm). Twenty-nine urine samples were collected during the four days of the study. Approximately 28% of the urine samples contained levels of phenol greater than 75 mg/L and 31% of the samples had levels of ttMA greater than the 500 µg/g creatinine. Thirty background urine samples were also collected from the volunteer, and approximately 7% had levels of phenol above 75 mg/L and 30% had levels of ttMA above the BEI. Statistically, phenol corrected for specific gravity was significantly greater in the urine samples collected during the exposure period than in the urine collected during the background period (p = 0.003). However, the levels of uncorrected phenol, phenol corrected for creatinine, uncorrected ttMA, and ttMA corrected for creatinine were not statistically different from background concentrations for this volunteer (p = 0.31, 0.79, 0.21, 0.21, respectively). These results support other studies that have shown that spot urine samples are clearly not reliable indicators of benzene exposure when exposures are around 1 ppm. Multiple background samples are essential to determine whether or not high levels of excreted metabolites correspond to workplace exposure rather than simply inter-individual differences in metabolism of foods in the diet.
J. Grassman, Brooklyn College, Brooklyn, NY; Y. Chernyak, Insititute of Occupational Health and Human Ecology of Siberia and Institute of Biophysics of Angarsk State Technical Academy, Angarsk, Russian Federation; E. Brodsky, Severtsov Insitute of Ecology and Evolution, Moscow, Russian Federation; A. Shelepchikov, E. Mir-Kadyrova, D. Feshin, Severtsov Institute of Ecology and Evolution, Moscow, Russian Federation; A. Merinova, Institute of Occupational Health and Human Ecology of Siberia and Institute of Biophysics of Angarsk State Technical Academy, Angarsk, Russian Federation.
In December 1992, more than 700 firefighters from the region near Shelekhov, Russia, were exposed to combustion products during a catastrophic three day industrial fire. The fire suppression took place inside a warehouse that was sufficiently large to permit the entry of firefighters and their trucks. An estimated 22 to 57 g of dioxins (in I-TEQDF) were formed during the combustion of 1,000 tons of various raw materials, mainly polyvinylchloride and polyethylene. Respiratory protection was not available for the majority of firefighters. Moreover, many of the firefighters with access to closed-circuit respirators were hesitant to use them. To assess their body burdens of dioxins, we collected serum samples in January 2004. Assessment of dioxin body burdens this long after the exposure event is informative because several dioxin congeners have half lives of 10–20 years. 7 polychlorinated dibenzo-p-dioxin (PCDD), 10 polychlorinated dibenzofuran (PCDF), and 12 polychlorinated biphenyl (PCB) congeners were measured in 20 firefighters randomly selected from the cohort of 165 firefighters. Firefighters were divided into groups according to their symptoms and participation in the Shelekhov fire. Overall, the firefighters had a total mean arithmetic TEQ of 153 pg/g lipid, with a median of 123 pg/g. This compares to serum levels of 20–30 pg/g lipid reported in populations with only environmental exposures. Dioxins were not significantly related to disability status, respirator use, or days at fire. Officers tended to have lower serum concentrations of PCDDs than lower-ranking firefighters. Firefighters participating in the Shelekhov fire had higher serum concentrations of PCDDs than the controls. These preliminary results suggest that this is a population with a high degree of dioxin exposure, but it does not conclusively identify participation in the Shelekhov fire or firefighting in general as the source.
J. Vietas, G. Talaska, University of Cincinnati, Cincinnati, OH.
Humans are considered the most sensitive species to arsenic exposure with increased risk to skin, lung, and bladder cancer. Epidemiologic studies of workers simultaneously exposed to benzo[a]pyrene (BaP) and arsenite (As) report additive to multiplicative effects. These studies are supported by both in vitro and in animal studies demonstrating an increase in BaP DNA adduct levels when co-treated with BaP and arsenite than when treated with BaP alone. Glutathione, the major thiol compound responsible for maintaining redox homeostasis, may provide cellular protection against arsenite’s ability to increase the likelihood of DNA damage. We characterized the effect of modulating glutathione levels, through the use of buthionine sulfoximine (BSO) and glutathione ethyl ester (GSHEE) treatment as well as by using glutathione-deficient knockout mice, on the formation of DNA adduct levels after co-exposure to arsenite and benzo(a)pyrene. Lung and skin tissues were analyzed for DNA adducts using 32P-postlabeling. Arsenic co-treatment increased average BaP adduct levels in both lung and skin; the increase was statistically significant in the lung (p = 0.048). A reduction in intracellular glutathione level increased BaP adduct levels, although only significantly in the skin of mice treated with BSO (p = 0.028). Treatment with GSHEE reduced adduct levels, although not significantly (p > 0.05) in any tissue measured. These results are consistent with previous in vitro findings and suggest that glutathione plays a role in arsenic’s ability to potentiate BaP DNA adduct formation.
Posted May 30, 2006