117
USING MARSSIMS TO INTERPRET ELEVATED BERYLLIUM SURFACE CONTAMINATION.
R. Brounstein, Bechtel-SAICComp LLC, Las Vegas, NV.
During the late 1960s, a number of experiments were conducted at the Nevada Test Site’s Area 25 that resulted in a dispersion of beryllium alloy. Approximately 3.5 miles away is the Yucca Mountain Project’s Sample Management Facility (SMF) where various types of soil and core samples are analyzed and stored. Throughout 2002 and 2003, an assessment was conducted at the SMF to determine the extent of beryllium contamination. While airborne samples resulted in nondetect measurements, surface swipes indicated detectable beryllium at levels greater than the release criteria of 0.2 micrograms per 100 centimeters squared, established by the Department of Energy per 10 CFR 850, “The Chronic Beryllium Disease Protection Program.” It has been hypothesized that these elevated levels are due to the natural beryllium content of the regional soils (up to 3 parts per million) and that many of the swipes picked up large quantities of material due to large soil deposition via the operations conducted at the SMF (ex. wet saw cutting) as opposed to legacy issues.
Using the Multi-Agency Radioactive Survey and Site Investigation Manual (MARSSIM) techniques, further investigations were conducted, based on a 95/95% confidence limit, to determine whether the beryllium detected in surface samples is due to natural soil conditions or prior legacy issues. As a result, MARSSIM assessment methodologies proved to be a valuable tool to determine whether 10 CFR 850 requirements were applicable.
118
SIMULATION OF RETROSPECTIVE AND PROSPECTIVE CHEMICAL EXPOSURE SCENARIOS USING
SULFUR HEXAFLUORIDE (SF6).
R. Moore, C. Simmons, Boelter & Yates Inc., Park Ridge, IL.
There often is a need to estimate chemical exposures which may have occurred in the past. It is not always possible to recreate these scenarios using the actual chemicals of interest due to safety and liability concerns. There are other situations where it is desirable to predict employee exposures during new process development. This presentation contains case studies, where sulfur hexafluoride (SF6) was used as a surrogate to simulate both past and future chemical-related exposure scenarios involving gases and vapors. In the first case discussed, an elevator graffiti cleaning operation on the ground floor of a high-rise condominium was simulated using SF6 to determine worst-case chemical levels which may have occurred as a result of this operation in a unit located on the 13th floor. The owner of the unit had made allegations that chemicals from this operation had migrated into the unit and caused irreversible damage to health. In the second case discussed, a rubber membrane roof installation operation was simulated. Sulfur hexafluoride was used as a surrogate to estimate possible worst-case chemical levels in occupied areas of the building. Certain occupants of the building had made allegations that chemicals from an operation of this nature on one particular day had entered their workplace and caused irreversible damage to their health. Both of these simulations assumed all of the solvent chemicals involved in the operations had evaporated. The third case presents how SF6 simulations are used in the semiconductor industry to predict possible worst-case employee exposures during equipment failures involving piping and processes that utilize pressurized gases and vapor-laden gas streams.
119
DEVELOPMENT OF A TAPE-STRIPPING METHOD TO QUANTIFY DERMAL EXPOSURE TO
HEXAMETHYLENE DIISOCYANATE.
K. Fent, K. Jayaraj, A. Gold, L. Ball, L. Nylander-French, University of North Carolina, Chapel Hill, NC.
Significant worker skin contact with hexamethylene diisocyanate (HDI) occurs during the application and manufacture of surface-coatings, foams, resins, and plastics. Respiratory sensitization and occupational asthma have long been associated with airborne isocyanate exposure, but a similar association with dermal exposure to isocyanates has largely been left unexplored, mainly due to nonexistent quantitative sampling methods. We have developed a noninvasive tape-stripping technique for sampling layers of the epidermis for determination of chemical concentrations in the skin. Quantification of HDI on tape-strip samples following derivatization reaction with 1-(2-methoxyphenyl)-piperazine was performed using liquid chromatography and mass spectrometry with electrospray ionization in positive ion mode. The derivative was synthesized independently and shown to be pure by proton NMR and melting point analysis. The ion at m/z 553.4, corresponding to the protonated molecular ion, was used for quantitation. This method was tested by applying HDI-containing products to tape-strip samples, performing derivatization and quantitative analysis, and determining the yield, which ranged from 87 to 93%. The limit of detection for this method is 0.05 ρmol/µL, while the limit of quantification is 0.5 ρmol/µL. This method is highly sensitive and specific and suitable for quantification of dermal exposure to HDI in occupational settings. This method will be used to measure dermal exposure to HDI in conjunction with inhalation and biological monitoring during spray-painting operations in order to investigate the contribution of dermal exposure to total body dose.
Supported by CDC/NIOSH R01-OH007598.
120
EVALUATION OF EXPOSURE TO CARBON MONOXIDE DURING FOREST PRESCRIBED BURN
ACTIVITIES.
K. Dunn, I. Devaux, A. Stock, J. Mott, CDC/NCEH, Atlanta, GA; L. Naeher, University of Georgia, Athens, GA.
As part of a 2004 study of smoke exposure and respiratory function in a cohort of forest fire fighters in the southeastern United States, we collected personal monitoring exposure data for PM2.5 and carbon monoxide (CO). CO sampling was performed using real-time Draeger single-gas monitors. A post-shift questionnaire was also administered to collect information on exposure and daily job tasks. In this abstract, we present the results of the CO exposure part of this study.
Over the three-week study period, firefighters were sampled during a total of 10 burn days, providing 87 total person-day CO measurements. Real-time CO exposure was measured on an average of 8.7 firefighters per burn day for an average shift of 11.3 hours. The controlled burns covered an average of 692 acres per burn (range 28 to 2745). Average CO concentrations were low (median = 1.23 ppm, max = 14.05 ppm), as much of the shift was spent preparing for and traveling to and from the fires.
The highest CO exposures occurred among firefighters working on foot in the holding of fires within specified boundaries (mean CO = 5.8 ppm) while the lowest were from on-site burn supervisors (mean CO = 1.7 ppm). Average CO concentrations increased in agreement with firefighter reported smoke exposure levels. Firefighters reporting the highest smoke level were exposed to the highest mean CO concentration (mean CO concentration = 9.8 ppm) compared with those reporting lower smoke levels (mean CO range = 1.5 to 4.1 ppm). Other indirect measures of exposure—such as number of acres burned—were poorly correlated with mean CO exposure (r2 = 0.03).
Real-time exposure monitoring provides good feedback to the firefighter on transient peak exposures, potentially allowing work-practice modification. Both particulate and CO monitoring provide important data that might help assess the long-term health effects associated with smoke exposure.
121
A RETROSPECTIVE EVALUATION OF NICKEL EXPOSURES UNDER NONSTATIONARY SCENARIOS.
S. Maberti, E. Symanski, The University of Texas at Houston, Houston, TX.
Background. While numerous studies have characterized the between- and within-worker sources of variation among different groups of workers, investigations within a single industry are limited because of scarce monitoring data. Thus, a database of over 4000 personal exposure measurements collected in the nickel-producing industry from 1973 to 1994, which was compiled for a separate investigation, provided a unique opportunity to carry out a comprehensive evaluation of workers’ exposure profiles from a common industrial sector. Methods. Exposure profiles of 160 job groups were characterized using a mixed effects model that considered changes in exposures over time. For each job group, trends in exposure levels and the within- and between-worker variance components were estimated and compared across industrial sector and by characteristics of the sampling campaign. Results. There was evidence of significant changes in exposure levels over time in 36% of the groups (-65 to 160% year-1, median -10.4% year-1), with the majority indicating trends towards lower exposure values. In general, groups from refining experienced faster declines than groups from other sectors. Not surprisingly, groups sampled before 1980 experienced faster declines in exposures than groups that were first sampled in later decades. Although both sources of variation appeared lower in refining compared to other sectors, the differences were not statistically significant. In evaluating homogeneity, a greater proportion of groups in refining (55%) compared to smelting (44%) and all other sectors (40%) had less than a two-fold difference in 95% of individual mean exposures. Conclusions. This approach is useful for retrospective studies, since it allows for an accurate characterization of the mean exposure profile as well as of the between- and within-worker sources of variation. Our results underscore the need for monitoring programs that include repeated measurements collected on the same worker to properly assess exposures over both the short- and long-term.
122
AN EVALUATION OF PATTERNS OF HOMOGENEITY IN EXPOSURE AMONG JOB GROUPS ACROSS
DIFFERENT INDUSTRIES: IMPLICATIONS FOR EPIDEMIOLOGIC STUDIES.
S. Maberti, E. Symanski, The University of Texas at Houston, Houston, TX.
Background. To produce groups with similar exposure profiles in epidemiologic studies, workers are often classified on the basis of characteristics related to location and tasks. A study that quantified the within- and between-worker sources of variation in exposure of 165 groups of workers provided evidence that grouping strategies that rely upon job title and location are not effective, since proportions of the groups with either homogeneous (22%) or heterogeneous (30%) exposures were similar. Consequently, such a grouping strategy could introduce exposure misclassification in some cases. Methodology. A literature review was conducted to identify studies published in the last 10 years that quantified the within- and between-worker sources of variation in exposure. In expanding upon the database that was previously analyzed, data were gathered on nearly 500 groups of workers and used to conduct a comprehensive evaluation of patterns of homogeneity in exposure. Results. Only 20% of the groups exhibited less than a two-fold difference in 95% of the individual mean exposures (bR95 < 2), while 25% had greater than a 100-fold difference. In evaluating homogeneity in shift-long exposures, 2% of the groups had wR95 values less than 2 and 15% had wR95 values greater than 100. Whereas individual mean levels were more homogeneous, on average, for workers exposed to gases/vapors as compared to aerosols or dermal agents (median bR95: 4.8, 8.2, and 9.8, respectively), the opposite pattern emerged for shift-long exposures (median wR95: 26.2, 16.1, and 13.7, respectively, for gases/ vapors, aerosols, and dermal agents). Differences were also detected when type of agent was stratified by industrial classification (chemical versus nonchemical). Conclusions. Our results corroborate previous observations that job group and location are not the most appropriate variables for defining homogeneous groups. Other variables should be evaluated to identify optimal strategies for classifying groups of workers with similar exposures.
123
DISSOCIATION OF HEXAVALENT CHROMIUM INTO SIMULATED LUNG FLUID.
M. Moran, P. LaPuma, U.S. Air Force, Bethesda, MD.
Purpose. OSHA’s proposal to reduce the ceiling limit for hexavalent chromium (0.1 mg/m3 as CrO3) to a more stringent eight-hour time-weighted average of 0.0005 mg/m3 (as Cr6+). This would make the continued use of Cr6+-containing primer paints very difficult. Two of three large-scale epidemiological studies indicate that painters using chromate-based paints don’t show an increased risk of cancer. In assessing toxicological risk to humans from the inhalation of particles, there’s considerable debate over the bioavailability of chemical consituents to the human receptor. Generally, a chemical is assigned specific toxicological values without considering the aerosol type or formulation involved with the chemical. Dose to the lung tissue is driven by the chemical’s dissociation from the particle into the lung mucosal fluid. If the paint matrix hinders the bioavailability of the chromate during the time that the particles are resident in the lungs, then the proposed OSHA standard would be overly restrictive in the case of painting operations. Hypothesis. Smaller particles will proportionately release more Cr6+ than larger particles. Particles generated from different paints (e.g., water versus solvent) will have differences in bioavailability. Methodology. Paint particles are generated using aviation primer paint and particles are collected and separated based on size using a six-stage viable cascade impactor. Each stage collects particles of a discrete particle size into a petri dish filled with a simulated lung fluid. The fraction of chromate that dissociates from the particles are measured against the total amount of chromate collected in each dish. Application. To date, a method to measure the release of a chemical from particles into lung fluid is not available. A method to quantify the fraction of a chemical released from a particle into lung fluid would aid significantly in understanding the actual dose when a chemical is incorporated within different types of aerosols.
124
RESPIRATORY HEALTH EFFECTS IN WELDERS.
D. Vinson, L. Conroy, T. Schoonover, S. Dorevitch, S. Erdal, R. Cohen, D. Tessier, University of Illinois at Chicago, Chicago, IL.
A group of workers at an engine manufacturing facility were recruited into a pilot study examining cross-shift changes in a series of respiratory inflammation indicators with exposure to metal fumes and irritant gases. Sixteen welders and 22 non-welders were selected. The subjects were tested on Mondays before and after the work shift, collecting blood, exhaled breath condensate (EBC), and exhaled air samples each time. Exposure to metal fumes, ozone, and nitrogen dioxide were measured during the workshift. Gas tungsten arc welding, shielded metal arc welding, and gas metal arc welding were performed in the facility.
Cytokines (IL-6 and IL-8) were measured in the exhaled breath condensate and blood serum samples. Exhaled air samples were analyzed for carbon monoxide (eCO) and nitric oxide (eNO).
Regression analysis and scatter plots were used to compare the exposure metals in the air to the response cytokines in blood and exhaled breath condensate, as well as to response eNO and eCO utilizing evening (PM) and cross-shift (XS) change levels. Exhaled NO and CO levels were found to be statistically significant in more than 100 modeling scenarios involving the exposure variables (p < 0.05). Five other statistically significant relationships were identified: PM blood IL-6 versus airborne manganese exposure (p = 0.0142); PM blood IL-6 versus O3 exposure (p = 0.0060); PM EBC IL-6 versus airborne cadmium exposure (p = 0.0351); XS EBC IL-6 versus cadmium exposure (p = 0.0060); and PM EBC IL-8 versus molybdenum exposure (p = 0.0002).
This pilot study showed increases in exhaled CO and NO and in exhaled cytokines with increasing exposure to metal fumes and irritant gases. Exhaled gases (eCO and eNO) and exhaled breath condensate should be further evaluated for their use in measuring lung inflammation due to occupational exposures.
125
METAL BIOMARKERS IN WELDING.
D. Vinson, L. Conroy, T. Schoonover, S. Dorevitch, S. Erdal, R. Cohen, D. Tessier, University of Illinois at Chicago, Chicago, IL.
A pilot study investigating biomarkers for metals was conducted in an engine manufacturing facility. Sixteen welders and 22 non-welders were selected. The subjects were tested on Mondays before and after the work shift, collecting blood and exhaled breath condensate (EBC) samples each time.
Breathing zone concentrations of metal fumes were monitored by fitting each subject with low flow pumps; cellulose ester membrane filters connected to the pumps and were placed near the breathing zone. The subjects were observed during their entire work shift, with their specific activities being recorded. Gas tungsten arc welding, shielded metal arc welding, and gas metal arc welding were utilized by these welders.
The average welding fume exposure in the welders was 861 µg/m3, with the exposure ranging from 45.8 to 2572 µg/m3. The average welding fume exposure in the non-welders was 89.2 µg/m3, with the exposure ranging from 3.6 to 374 µg/m3.
Regression analysis and scatter plots were used to compare the exposure metals in the air to the response metals in blood and exhaled breath condensate, utilizing evening (PM) levels as well as cross-shift (XS) levels. Four statistically significant relationships were identified in the EBC analysis: PM EBC cadmium versus exposure cadmium (p = 0.0018); XS EBC tin, controlling for smoking, versus exposure tin (p = 0.0422); total XS metals versus exposure welding fume, (p = 0.0186); and PM EBC zirconium versus PM blood zirconium (p = 0.0196). Twenty-three nearly-statistically significant (p < 0.2) relationships were also identified.
A literature review did not identify any studies utilizing exhaled breath condensate to monitor metals exposure. This study indicates that EBC, a noninvasive method, shows promise for monitoring welding fume exposure.
126
COMPARISON OF SHORT-TERM SAMPLE DATA WITH CORRESPONDING LONG-TERM RESULTS: STEL
DATA OVER-PREDICTS AN 8-HR TWA.
M. Weeks, F. Boelter, C. Simmons, Boelter & Yates Inc., Park Ridge, IL.
A series of short-term (30-minute) and long-term personal air samples were collected simultaneously during various activities in order to determine exposure to asbestos fibers. Samples were analyzed using NIOSH Method 7400 for phase contrast microscopy and NIOSH Method 7402 transmission electron microscopy. Time-weighted averages were calculated for 195 short-term samples, and were then compared with their 55 corresponding long-term samples. The results show that time-weighted averages of the short-term samples averaged 4.1 times higher than the corresponding long-term data, ranging from no difference to 18 times higher during the same sampling period. This indicates that using multiple, consecutive or nonconsecutive short-term samples in order to calculate a time-weighted average may produce results different from a long-term sample. The conclusion is that while short-term sampling is appropriate in certain situations (for instance, in determining compliance with STEL or ceiling limit values), the projection of long-term exposures from short-term data introduces significant uncertainty and may be inappropriate.
127
COMPARISON OF GRAVIMETRIC SAMPLING TO A DIRECT-READING AEROSOL MONITOR FOR
MEASURING DRYWALL DUST.
R. White, San Diego State University, San Diego, CA.
Workers who perform sanding of drywall joint compounds are often exposed to high concentrations of dusts, which may include such hazardous components as talc, calcite, mica, gypsum and even silica. A 1994 NIOSH Health Hazard Evaluation of a drywall sanding operation found that drywallers were exposed to concentrations up to 10 times the OSHA PEL for total nuisance dust of 15 mg/m3. Drywall workers have reported symptoms of eye, nose, throat, and respiratory tract irritation as well as phlegm production, coughing, and difficulty breathing. The hazardous components of drywall dust, along with ACGIH’s request for research involving alternate methods for quantifying total dust, presents the need for development of a simple, rapid, and cost-effective sampling method for measuring drywall dust concentrations. The objective of this study was to evaluate the effectiveness of the direct-reading HAZDUST III™ particulate monitor in estimating concentrations of dust generated during drywall sanding operations. Sixty sets of paired samples were collected with the HAZDUST III and gravimetric method over three sampling conditions to test the null hypothesis that there is no significant difference between traditional gravimetric sampling and real time sampling with the HAZDUST III particulate monitor. Three sampling conditions were created to simulate a range of drywall dust concentrations. A paired t test α = 0.05 was conducted on all three data sets, resulting in a p value of < 0.001 for all three conditions, rejecting the null hypothesis and indicating that there is a significant difference between dust concentrations measured between the two methods. It appears that the HAZDUST III optic system signal may be too low to produce an accurate display reading, meaning it does not adjust to the mass concentration of airborne particles. As a result, the instrument is prone to underestimating the airborne particulate concentration.
Posted May 30, 2005