Ann. occup. Hyg., Vol. 48, No. 3, pp. 245-255, 2004
© 2004 British Occupational Hygiene Society
Published by Oxford University Press
Determination of Dermal Exposures During Mixing, Spraying and Wiping Activities
Institute of Occupational Medicine, Research Park North, Riccarton, Edinburgh EH14 4AP, UK
Received 2 June 2003; in final form 8 January 2004
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONREFERENCES |
|---|
Dermal exposure measurements were collected as part of RISKOFDERM, a European dermal exposure study which aims to improve the understanding of the nature and range of dermal exposures to hazardous substances throughout the European Union. Exposure measurements were collected to enable a predictive model to be developed for regulatory risk assessment purposes. In this paper dermal exposure data are presented for three generic job tasks: spray painting, wiping surfaces and mixing/dilution of formulations. The particular workplace settings included a dockyard and three medical laboratories. In the dockyard the tasks involved spray application and mixing of anti-foulant paint. For laboratory workers the observed tasks were preparation of biocide solution and wiping of surfaces with the disinfectant. Each dermal exposure measurement was derived from the mass of trace analyte on cotton gloves and 11 fabric patches, which were cut from whole-body dosimeters, representing the main anatomical areas of the body. The percentage mass of trace analyte in the formulation was determined by analysis to enable the total mass of the product on the anatomical areas to be calculated. The sampling periods were recorded to enable calculation of the dermal exposure rate, which is expressed as µg total formulation/cm2/h. The geometric mean dermal exposure rate for the hands during spray painting was 2760 µg/cm2/h (n = 24). The exposure rate for the rest of the body was 175 µg/cm2/h (n = 35). Mixing of the paint involved higher exposure rates for both the hands and body, with a geometric mean of 31 200 µg/cm2/h (n = 9) for the hands and 327 µg/cm2/h (n = 14) for the rest of the body. For small-scale routine disinfection of surfaces using small quantities of biocide the principal anatomical area affected was the hands, with a geometric mean dermal exposure rate of 1840 µg/cm2/h (n = 6). During systematic disinfection of laboratory surfaces with larger quantities of the biocide solution, the geometric mean dermal exposure rate for the hands was increased to 139 000 µg/cm2/h (n = 24). In this case there was increased exposure of the body: principally the arms, legs, chest and head. The measured dermal exposure rate during preparation of the biocides (mixing) was very low, with a geometric mean value for the hands of 13 µg/cm2/h (n = 16). There was a high level of variability observed in the results within each task. It is suggested that dermal exposures are partly dependent on human behaviour and on the occurrence of accidental contact with contaminated surfaces. This makes interpretation of the results difficult for predictive risk assessment purposes.
Keywords: anti-foulant; biocide; dermal exposure; disinfection; paint spraying; patch sampling; wiping
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONREFERENCES |
|---|
RISKOFDERM is a multi-centre pan-European project which aims to improve the understanding of the nature and range of dermal exposure to hazardous substances throughout the European Union. The project comprises four separate elements, which are: (i) identification of the main determinants of exposure; (ii) measurement of exposures in various workplace scenarios; (iii) development of a predictive exposure model; (iv) development of a practical risk assessment tool. The work described in this paper relates to the second work part, i.e. the measurement of dermal exposure in various workplace scenarios. The full details of this project are described in RISKOFDERM (2001, 2002).
The objectives of the project and the overall approach taken are described in some detail in the companion paper by Rajan-Sithamparanadarajah et al. (2004). In brief the work is concerned with the identification and assessment of exposure for certain generic tasks, defined as dermal exposure operation (DEO) units. A total of six DEO units have been defined and these represent general categories of work for which dermal exposure may occur. Each of the DEO units has one or more (up to 14) scenarios that are considered to be examples of typical industrial processes for the relevant DEO unit. The basis behind the definition of the DEO units is described by Rajan-Sithamparanadarajah et al. (2004) and in RISKOFDERM (2001).
A model of dermal exposure has been described by Schneider et al. (1999) in which the 3-dimensional layer of contamination on the skin (the skin contaminant layer) would be the ‘actual exposure’ as described in HSE exposure assessment document EH74/3 (Health & Safety Executive, 1999a). The outer clothing contaminant layer is sometimes referred to as the ‘potential exposure’, since this is the mass of contamination which could potentially come into contact with the skin (if protective gloves or clothing are not worn). For regulatory risk assessment purposes it is common practice to use measurements of potential exposure to determine the level of risk associated with the substance of interest.
Assessment of the risks from dermal contact with harmful substances is more complex than for exposure by inhalation. It is accepted that there is considerable scope for temporal and spatial variations of dermal contamination due to worker and process related factors (e.g. Vermeulen et al., 2000). Conventional techniques for assessing dermal exposure are unable to take account of temporal variations (within-day) in either the skin contaminant layer or the outer clothing contaminant layer. At best, we can evaluate the level of contamination on the skin surface or on the outer clothing by skin wiping, washing or by analysing patches applied to the skin or clothing. However, the contamination may be lost over time due to overloading, evaporation, penetration or a combination of all three.
One common method of estimating potential exposure involves collecting and analysing patch samples, which are deemed to be representative of the anatomical areas of interest. It is then possible to extrapolate whole body exposures by summing the assumed contamination levels for all the different anatomical areas. Where a patch sample is splashed or comes into direct contact with contaminated surfaces it may not be truly representative of the anatomical area under consideration and this may cause over-estimation of exposure when whole body exposures are calculated. Conversely, if splashes occur but miss the patch sample, this would lead to an under-estimate. It is possible to avoid these errors by analysing contaminant deposition on the outer clothing by cutting up and analysing the whole garment or by systematically quantifying the contaminants by some other means, e.g. using a portable X-ray fluorescence spectrometer (Wheeler and Warren, 2002). However, these methods tend to be highly labour intensive and/or involve expensive specialist analytical equipment. The various methods for assessing dermal exposure assessment methods and their limitations have been described in the literature (e.g. Fenske, 2000; Soutar et al., 2000; Vermeulen et al., 2000).
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONREFERENCES |
|---|
The principal objective of the work described here was to collect dermal exposure information for three scenarios: spraying for coating, mixing/diluting and wiping.
For the ‘spraying for coating’ and the ‘mixing/ diluting’ scenarios the chosen industry sector was shipbuilding/ship repair, in which exposure to marine anti-fouling paint was assessed. Marine anti-fouling paint is used to prevent growth of marine organisms on ships’ hulls and is usually applied by spray coating in dry dock at the time of building and regularly thereafter. This painting scenario was investigated during an extended field study at a single dry dock in which a large naval ship was undergoing a refit. For the wiping scenario, exposure to public health biocides in hospitals and pathology laboratories was assessed. It was also possible to observe a mixing/ dilution task in the laboratories, when the biocide solutions were being prepared.
It was agreed within the RISKOFDERM study that 30 whole-body and hand exposure measurements should be obtained for each DEO unit. This was a pragmatic approach intended to ensure that a broad range of tasks and workplaces could be covered for each DEO.
Measurement methods
In this work we used a sampling method based on the OECD protocol for sampling dermal exposure to pesticides (OECD, 1997). This included sampling the outer clothing contaminant layer or ‘potential exposure’ by analysing gloves and whole-body dosimeters. The same measurement methods were used in each of the scenarios.
For the dockyard workers white polycotton hooded coveralls were used as whole-body dosimeters. For the remaining workplaces SontaraTM hooded overalls were used. In each case, 11 sample patches were marked out onto the overalls before exposure, according to the OECD convention. Each patch measured 10 x 10 cm and the locations were as per the OECD sampling protocol. The whole body dosimeters were worn over the normal working clothes of the subject, except where these were already contaminated. In such cases the contaminated clothing was removed or a clean disposable overall was placed between the personal clothing and the whole-body dosimeter.
After exposure the overalls were collected and taken to a clean area where they were hung up on wire coat hangers and the patches cut out and immediately placed in labelled sample jars.
White cotton fourchette gloves were used to assess exposure to the hands. They were worn on the outside of the protective gloves usually provided for the task, thereby providing a measure of potential exposure. A new pair of protective gloves were put on before the cotton sampling gloves were placed over them, to prevent any initial contamination of the samplers. After sampling, each glove was removed from the wearer and placed in individual glass jars.
The time of sampling was measured, which allowed the deposition rate for the contaminants of interest to be calculated for the given DEO unit. Other qualitative information about worker posture, the presence of ventilation or protective equipment was gathered and recorded using a standard questionnaire form developed for use by all of the RISKOFDERM partners. This information was to be used by others in combination with the quantitative exposure data for development of the predictive exposure model.
For every subject a blank glove and blank patch sample was collected. These samples were analysed to check for contamination introduced on site. Production of field spikes proved difficult for anti-foulant paint as it was not possible to dispense accurately small quantities of the substance due to the high viscosity of the formulation.
As part of the method validation, predetermined quantities of a copper standard solution (Aldrich catalogue no. 35,618-2) were applied to patches and gloves using a pipette. This procedure was repeated on different patches and gloves using samples of working biocide collected at the time of survey. These spike samples were prepared and analysed to determine the recovery efficiency for each analyte.
Chemical analysis
Samples of the anti-foulant paint and the biocide solution were analysed to determine the percentage content of the relevant analyte in the formulations. This was done so that dermal exposures for the total formulation could be extrapolated from the mass of analyte detected on the dermal samples. For the anti-foulant spray painting and paint mixing tasks in the dockyard, the analyte was copper. For the mixing/diluting and wiping tasks in the laboratories, the analyte was potassium.
Samples of the anti-foulant paint were collected from the actual containers being used at the time of survey and these were placed in sealed glass jars for transport to the laboratory. The percentage mass of copper in the paint was determined according to the following procedure. A sample of paint was applied to a preweighed glass slide, which was immediately reweighed in order to determine the mass of wet paint available for analysis. The sample was dried at room temperature for a period of 48 h and then reweighed. This enabled the solid content to be determined, although this was not used in any of the exposure calculations. The dry paint sample was then acid digested and analysed to determine the mass of elemental copper. It was therefore possible to obtain a measure of the percentage mass of analyte in the total mass of paint formulation. It is debatable whether the mass of wet paint should be used for exposure assessment, since much of the solvent is likely to be vaporized during spraying. However, this convention was adopted to enable a worst case exposure estimate to be developed.
Samples of the working biocide solution were also collected from the working containers and placed in sealed glass jars. These solutions were analysed directly for elemental potassium. Again, the percentage mass of analyte in the biocide formulation was calculated.
The individual patch and glove samples exposed to anti-foulant were digested in 10% nitric acid. The acid digests were heated to near boiling point for 3 h and were then left to cool. The resultant solution was filtered through a 5 µm membrane filter, ensuring the container was thoroughly rinsed with distilled water. The resultant filtrate was made up to 100 ml.
All samples were analysed by inductively coupled plasma atomic emission spectroscopy (ICP/AES). The samples were analysed at the IOM analytical laboratory, which holds accreditation for the analysis of copper, potassium and other metals, by ICP/AES. The documented in-house method, based on Occupational Safety and Health Administration (OSHA) method 121 (OSHA, 1991) is accredited by the UK Accreditation Service (UKAS) under UKAS accreditation no. 0374.
Samples exposed to the biocide solution were also analysed by ICP/AES. However, the samples were extracted using caesium chloride and analysed in accordance with the procedure described in HSE method MDHS 95 (Health & Safety Executive, 1999b).
Calibration standards were prepared using known weights of analar grade reagents and the sample masses were determined with reference to these standards. Each result was then converted to a blank-corrected, recovery efficiency-corrected mass, taking into account the quantities of analyte contained in the blank and spike samples.
Data analysis
As indicated, chemical analyses yielded estimates of the mass of analyte on each of the patches. From the measured concentration of the analyte in the formulation, the mass of analyte on each patch sample was scaled up to provide an estimate of the mass of formulation on the patch. This was then scaled up to estimate the mass of formulation on the relevant anatomical area defined in the OECD method under the assumption of a uniform surface distribution.
As part of the RISKOFDERM project, common approaches were developed to deal with results that were below the limit of quantification (LOQ) and limit of detection (LOD). Values <LOD were calculated as 0, values >LOD but <LOQ were calculated as LOQ.
This approach enabled the results to be calculated as a contamination mass (in mg) on the represented area. While this metric is important in evaluating the exposure risks in specific situations, it is necessary for predictive purposes to normalize the data to deposition rates. A second metric is therefore used to represent the rate at which contamination builds up on the anatomical areas studied. This is defined as the dermal exposure rate and is evaluated for the mass of formulation (paint or biocide solution) per unit area divided by the exposure duration. Dermal exposure rates are expressed in terms of µg/cm2/h, which represents the time and surface area corrected dermal exposure rate for the given workplace scenario. This approach could enable exposure modellers to extrapolate dermal exposures for other substances or formulations used in other similar situations.
Discussion of the exposure data is therefore restricted to results for the formulation, rather than the analyte. Values for dermal exposure rates were calculated separately for the hands and for the rest of the body. In all cases values for left and right hands were summed to produce an overall value for both hands. Also, the individual values for the corresponding left and right limbs were summed to produce estimates for these different anatomical areas.
For the calculations of the concentration on the hands and body (excluding hands) the surface areas used were 820 and 18 720 cm2, respectively.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONREFERENCES |
|---|
The potential dermal exposure levels for each of the five observed workplace scenarios are summarized in box plot diagrams Figs 1 and 2. The data for the hands (Fig. 1) and the whole body excluding the hands (Fig. 2) are considered separately.
|
|
In these plots the boxes include data from the lower 5th and the upper 95th percentiles. Outlying data points are indicated by excursions outside the box. The notch in each of the boxes indicates the position of the median.
It is possible to illustrate the variability of dermal exposure by examining the individual results. Figures 3, 4, 5, 6 and 7 provide an illustration of the spatial distribution for the main anatomical areas in terms of the dermal exposure rate for each of the different scenarios observed.
|
|
|
|
|
A total of 79 pairs of glove samples were collected to establish exposure of the hands and 1 of these was <LOQ. For the whole body exposures, each measurement comprised 11 patch samples as previously described. A total of 95 whole body measurements were obtained, 10 of which were <LOQ (6 obtained during mixing/diluting biocides and 4 during biocide wiping).
Spraying for coating
The anti-fouling paint used during the survey exercise contained copper(I) oxide and various volatile organic solvents. Two different types of paint formulations were used, which contained 
36.9 and 20% copper by mass of formulation. Individual exposure measurements for the paint formulations were determined according to the relevant percentage copper mass ratio. The average recovery efficiency for the sampling media used for copper-based anti-foulant (cotton patches and gloves combined) was 1.03 (range 0.95–1.09, n = 19).
A total of 35 whole-body dermal exposure measurements and 24 hand exposure measurements were collected for this scenario, each measurement relating to the spraying task over a measured period of time (fewer hand exposure samples were collected due to initial problems fitting the samplers on top of the available protective gloves). In order to obtain this number of samples it was necessary to make a number of repeat measurements from each of the eight workers involved, the samples being collected over a 4 day period.
Up to eight workers were involved in the spray coating work on any given day. The workers were organized into four teams in which one operator carried out the spraying work (spray operator) and the second worker (line operator) assisted with the paint lines. Since the work was being done in small compartments and the pairs of workers were in close proximity to each other, the two jobs were considered to be similar enough to justify including the line men in the monitoring programme.
The ship was a very large naval vessel and the spray operators worked from scaffolding erected alongside. The scaffolding was covered by polythene screens in order to create a number of relatively enclosed spaces or compartments. It is usual practice to enclose a naval vessel in this way for security considerations, and this may mean that exposures are higher than would be measured in a commercial dry dock where marine vessels tend to be spray painted without enclosures.
The pattern of work dictated the duration of our sampling activities, although this matched the relevant DEO unit as closely as possible. In each case the spray operators were fitted with sampling suits before they ascended the scaffold and wore them until they returned at their break or lunch or the end of the shift. Hence, each sampling period was of the order of 1–2 h. The exposure duration was considered to be from the time the sampling dosimeter was put on to the time of removal.
The results for the sprayers and the linemen taken together show that the geometric mean dermal exposure rate for the hands during spray painting was 2760 µg/cm2/h (n = 24). The geometric mean dermal exposure rate for the rest of the body was 175 µg/cm2/h (n = 35). The geometric mean dermal exposure rates for the sprayers’ hands and body areas were 2600 (n = 11) and 166 µg/cm2/h (n = 18), respectively. The corresponding exposure rates for the linemen are 2900 (n = 13) and 184 µg/cm2/h (n = 17). These two data sets were analysed using SPSS version 11.0. Since the data are log normally distributed, they were transformed using natural logs and a two-sample t-test was carried out to compare hand and body exposures for the two jobs. There were no significant differences between the two sets of data (P = 0.801 for the hands and P = 0.794 for body exposures).
Mixing/diluting
This mixing/diluting task observed in the dockyard was carried out in support of the anti-foulant spray painting. In this case only one worker was involved. This job is generally known in this industry as the ‘pot man’. The task involves opening and mixing 25 l containers of the paint and inserting the pump system into the containers. Large quantities of paints were being used and depending on the situation up to 200 l could be processed during a 2 h period. The observed task was unaffected by spray painting work as the mixing station was remote from the area being painted.
Again, samples were collected for periods of 1–2 h, as this corresponded to the spray painting work described above.
The worker wore standard polycotton overalls and nitrile gloves when handling the paint. No respiratory protection was required for the pot man, as he was located in the open air, well away from the spray paint area. However, the worker’s hands were often contaminated from splashes and from contact with containers or the painting equipment. In certain cases the worker’s bare skin became contaminated and the paint was cleaned from the skin by wiping the hands with rags soaked in thinners, usually xylene.
A total of 14 measurements for mixing marine anti-fouling paints in a dockyard were collected. Again, it was necessary to make repeat measurements so that the required total of 30 samples could be obtained for this DEO unit.
This task involved higher exposures to both the hands and body than for spray painting, with exposure rates of 31 200 µg/cm2/h (n = 9) for the hands and 327 µg/cm2/h (n = 14) for the rest of the body.
The second mixing/diluting scenario involved the preparation of a disinfectant solution in hospitals and laboratories, which were visited principally to investigate the wiping scenario.
The biocide being used was VirkonTM, manufactured by Antec International, supplied in powder form in 50 g foil sachets. The product contains potassium peroxomonosulphate as an active ingredient. The product was dispensed in measured quantities with water to a standard 1% aqueous solution. This solution contained 0.17% potassium (derived by analysis of bulk samples). The work was carried out in a ‘clean’ environment in comparison to the dockyard and was of much shorter duration, typically 5–10 min for each event.
The biocide solution was prepared in one of two ways depending on the mode of use. First, gram quantities of biocide powder were measured out in a glass beaker and mixed with water. This was then poured into polypropylene wash bottles, which were used to apply the biocide to small areas of workbench during routine work. Secondly, the entire 50 g sachet was poured into a plastic bucket and filled with 5 l of water. This was then used for a systematic clean up of potentially contaminated surfaces, as part of a maintenance operation.
A total of 16 whole-body exposure measurements were collected for these mixing/diluting tasks. Typically 3–4 people were using the biocide at each of the locations, so a number of repeat measurements were obtained using the same subjects over different days.
The measured dermal exposure rate during mixing of the biocide was very low, with a geometric mean value for the hands of 13 µg/cm2/h (n = 16) and 0.35 µg/cm2/h (n = 16) for the rest of the body. The average recovery efficiency for potassium for gloves and patches combined was 0.99 (range 0.82–1.12, n = 31).
Wiping
This task is related to the biocide mixing/diluting tasks described in the previous section and the product being used was the same.
Two main modes of use were observed. First, a systematic disinfection of fume cupboards and work benches was carried out. Typically this scenario lasted 10–15 min per person each time it was done. For this work the subjects immersed a sponge in a plastic bucket containing 5 l of the solution and used this to wipe down the worktops and other surfaces. The surfaces were rubbed with the scouring pad surface of the sponge in order to remove stains and dried-on residues, so there was significant pressure applied to the surfaces.
The second mode of use was for small-scale disinfection of surfaces used for analysis of faecal samples. These had been collected as part of a local colorectal cancer screening programme and the laboratory workers applied the disinfectant before taking breaks away from their workstations. Typically, the wiping scenario lasted 5 min per person each time it was done. The biocide solution was applied from a 250 ml capacity wash bottle with squirting nozzle, rather than an aerosol spray. The surfaces were then wiped down with paper towels using low physical effort.
A total of 30 dermal exposure measurements were collected for this DEO unit. Again it was necessary to obtain repeat measurements from the same subjects at each location. While repeat measurements were obtained from the same individuals, this was done for a range of different wiping tasks, involving many changes in posture, so that the measurements could be considered to cover a diverse range of conditions as far as possible. For the large-scale disinfecting work it is estimated that each of the subjects carried out 1 h of wiping per day. In the faecal sample preparation unit, wiping was carried out for 15 min in total per person per day.
For the small-scale disinfection task the principal anatomical area affected was the hands, with a geometric mean dermal exposure rate of 1840 µg/cm2/h (n = 6). During systematic disinfection of laboratory surfaces with larger quantities of the biocide solution, the geometric mean dermal exposure rate for the hands was increased to 139 000 µg/cm2/h (n = 24). In this case there was increased exposure to the body: principally the arms, legs, chest and head. It was noted that the glove samples were heavily contaminated with the biocide solution due to partial immersion into the container followed by contact with a wet sponge.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONREFERENCES |
|---|
The results range over five orders of magnitude, which makes it difficult to compare results across the different tasks. However, it is clear that in all five scenarios the hands have the greatest potential for exposure. In the case of spraying one might expect that the body would receive a high level of exposure and this was found to be the case. However, it is surprising that the paint mixing/diluting task had very similar levels of dermal exposure to the body compared with the spray painting task. Furthermore, dermal exposure of the hands during mixing were much greater than for spraying. This is broadly in line with the observation that the workers hands became grossly contaminated from handling the paint containers. Also, the body areas became smeared with paint from the hands or from accidental contact with the paint containers and other contaminated surfaces, which explains why such high levels were recorded for the whole body exposures.
There were inevitably interruptions to the spray painting work during the sampling period. These were due to movement of personnel and equipment across scaffold levels and from one level to another. This may explain some of the variability in the dermal exposure data. It was not possible to observe the workers at all times due to limitations on access to the work location so it is not possible to correct for these interruptions. However, there were no rest breaks taken in the work areas and there were no major hold-ups due to equipment failures, so the measured levels may be considered typical of spray painting operations in a dockyard environment. It is accepted that the measured exposures may not be typical of other types of spray painting work.
The small-scale mixing and diluting tasks in the laboratories involved handling of gram quantities of biocide powder over short periods of time and, consequently, the measured exposures are relatively low.
For ease of interpretation, the wiping scenarios are considered separately. The disinfection of pathology laboratory surfaces involved significant contact with the biocide solutions in comparison with the routine disinfection of surfaces in the health screening laboratory. Consequently, in the former case (large-scale wiping) the sampling gloves became saturated due to immersion in the container holding the biocide solution. This caused splashing, and biocide was also transferred to the legs and arms as the workers leant over or knelt down on contaminated surfaces.
Figures 3 and 4 show the two mixing scenarios. These tasks are very different both in terms of the work environment and the quantities of formulation and methods of work employed. The magnitudes of exposure levels and patterns of exposure are therefore quite different. In the dockyard paint mixing scenario, the principle region of exposure of the body is clearly the legs. This is broadly in line with the findings of Garrod et al. (1999), where it was reported that splashing and uneven distribution of anti-foulant on the clothing could cause scaling errors. This was indeed noted on this occasion and must be taken into account when interpreting the results.
In the case of the biocide mixing, the exposure of the body is very low. This is simply because the task duration is very short for the laboratory work (5 min) and only very small quantities (<25 g) of the biocide powder are handled at any one time. It is interesting to note that the results are dominated by one very large result for the head in one sample. This could be explained by the subject touching the overall’s hood with contaminated hands, although this person did not receive a particularly high exposure to the hands in this case. Whatever the source of contamination to the head, this illustrates that the results are greatly skewed where the patches are splashed or smeared with contaminant. For optimum accuracy the patch sampler should be representative of the anatomical areas under observation, but this seldom happens in practice. Sources of errors for this method are discussed in more detail later.
The pattern in the spraying of anti-fouling paint is illustrated in Fig. 5 and the data is arranged so that the exposures for the linemen and sprayers can be compared. The pattern of exposure appears quite random, although it is clear from this that exposure of the hands is most important. In some cases the upper regions of the body receive maximal exposure and in others the maximal exposures are of the lower body regions. This is probably explained by individual differences in posture and orientation to the paint spray over time and between different subjects. There is an indication that perhaps exposure of the torso and legs occurs more frequently for the linemen than for the sprayers, but overall there were no significant differences in the hand and whole-body exposures for these two jobs.
Figure 6 illustrates the hand and whole-body exposures for the large-scale wiping task. The exposure results for the hands represent an upper level of potential exposure due to saturation of the sampling gloves. Apart from the hands, exposure of the forearms was highest and most frequent. The forearms were certainly exposed when workers were stretching out across the benches, but the measurements may have been exaggerated by a wicking effect on the sleeves, i.e. where the solution migrates along the sleeve of the sampling dosimeter due to saturation at one end. This was observed when the hands were heavily exposed and the sleeves of the sampling overall were in contact with the sampling gloves. There is also some exposure of the upper and lower legs, which is almost certainly due to the fact that the subjects were required to kneel down on the floor when it was being wiped, thus contaminating the knees, shins and feet. In a few cases there were measurable exposures of the head, chest, back and upper arm areas. These are all probably due to accidental contact with the surfaces being cleaned.
The pattern of exposure for the small-scale wiping task shows no distinct features (see Fig. 7) and since the measured values are so low, it is likely that the measured results are due to background ‘noise’ from environmental contamination.
It should be noted that within all of these results there were a number of repeat measurements taken from the same individuals. The data have not been analysed to determine if there are any inter-worker relationships.
|
CONCLUSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONREFERENCES |
|---|
Measurements of dermal exposure have been carried out for three dermal DEO units covering five different workplace scenarios. In all, 95 individual dermal exposure measurements have been obtained. All of the measurements relate to specific tasks, rather than being a measure of daily average exposure. All the dermal exposure measurements were obtained by patch and glove sampling and the results have been extrapolated from a trace analyte to mass of formulation on the body and hands. The measurements are also expressed, first, in terms of the mass of formulation on the body and hands and, secondly, standardized by time as dermal exposure rates. The purpose of obtaining the measurements and summarizing them in this way was to enable a generic dermal exposure model to be developed, for regulatory risk assessment purposes.
There were practical problems in gaining access to a wide range of workplaces and consequently fewer workplaces were surveyed than was originally intended. This meant that repeat measurements were collected from the same subjects, e.g. over different days, so that a complement of 30 exposure measurements could be obtained. However, there was a high level of variability in the workplace conditions and in the way the tasks were performed by individuals. It is our opinion that the effect of including these repeat measurements does not seriously compromise the validity of the data. However, the fact that the measurements were collected from only a few different workplaces means that the pattern and level of exposures, particularly for the anti-foulant spray painting and paint mixing tasks, may not be typical of other dockyards where these tasks are carried out.
Even within this relatively small selection of data a number of interesting observations have been made. These include the following.
For anti-foulant painting whole body exposures for those involved in mixing/dilution, in terms of both mass and rate, was similar to (actually larger than) that for the sprayers. This is despite longer sample duration for the spray operators than for the mixing/dilution operator.
The dermal exposure rate in the large-scale paint mixing/dilution scenario in the dockyard is of the order of 2000 times greater for the hands than in the small-scale biocide mixing/dilution scenario. While the comparison between the two tasks is valid on the basis of the DEO definition, it is not surprising that there are significant differences in the level of exposure since the work environments, quantities and physical characteristics of the substances being used are all very different.
The dermal exposure rates in the large-scale use biocide wiping scenario are of the order of 100 times greater than the small-scale biocide wiping scenario, where much larger differences might have been expected. However, it was observed that the sampling gloves became saturated in the large-scale wiping scenario.
As the above examples illustrate, the limitations of current dermal exposure measurement methods need to be considered. In comparison with inhalation exposure assessment, measurements of dermal exposures are more difficult to interpret. In the case of inhalation exposure we take exposures to be the biologically relevant fraction of the airborne concentration that is inhaled (or potentially inhaled) multiplied by the duration of exposure. No such equivalent metric exists for dermal exposure and, in the meantime, we consider dermal exposure to be the mass of contaminant on the skin contaminant layer (actual exposure) or the mass of contaminant on the outer clothing (potential exposure).
However, liquids or dusts seldom form a single uniform layer across the exposed skin or outer clothing. This means that patch sampling is not always representative and whole body exposure estimates based on this method may be subject to errors. For example, if a large splash falls onto a patch located on the subject’s arm, scaling this up may lead to a large overestimate. (If the splash misses the patch, an underestimate may occur.) Furthermore, it is accepted that there are areas of the skin which do not often become exposed, such as the inner surfaces of the arms and legs and natural creases such as the inside of the elbows and backs of the knees. The HSE publication EH74/3 (Health & Safety Executive, 1999a) provides details of various correction factors which are intended to account for non-uniform distribution of the contaminant. In this paper the data are presented in an uncorrected format.
While it is possible to use cotton patches and cotton gloves as dermal samplers to evaluate potential or actual exposures, these materials may act as reservoirs for the substances and quickly become saturated (e.g. Soutar et al., 2000). It is therefore not possible to determine whether the measured level of surface contamination results from a gradual accumulation of surface deposits across the measurement period or the result of a single massive exposure event at any time. Heavily contaminated samplers are also likely to cause cross-contamination to other anatomical areas while they are being worn by the subject. This is often seen as wicking, e.g. from the gloves to the sleeves, as observed for the wiping scenario in this study.
The notion of studying particular DEO units so that exposure levels can be attributed to generic task descriptions (for modelling purposes) is plausible, but in practice is more problematical. This is because real tasks can rarely be classified uniquely as one DEO unit for very long. In the course of this study tasks appeared to comprise an amalgamation of different DEO units or they would merge into each other very quickly, depending on the behaviour of the particular subject. This is illustrated by the biocide wiping task in which the subjects placed a sponge into a bucket of the biocide, squeezed out excess liquid and then wiped the sponge over the surfaces to be cleaned. There is clearly an element of this task that would be considered to be immersion. The task is wiping, but the immersion is closely related and may dominate the measured potential exposure. It is debatable whether the exposure measurements are more consistent with an immersion task or a wiping task. The other tasks included in this study also displayed DEO merging, but to a lesser extent.
Although current dermal exposure assessment methods have inherent limitations, this study has provided some useful standardized data where none previously existed and has given further insight into the theory and practice of dermal exposure methodology.
Acknowledgements—The authors wish to acknowledge Joop van Hemmen, who provided leadership of the RISKOFDERM project team, and Paul Evans with Bob Rajan for their leadership of work part 2. Particular thanks are due to Martin Roff of HSL, who provided invaluable technical advice and guidance for the sampling surveys. The authors wish to thank the members of the IOM project team, which included Dr Anne Soutar, Karen Creely, Andrew Apsley (from IOM) and Dr Sean Semple (of Aberdeen University) whose hard work and dedication made it possible to collect so many samples. Thanks are also due to Carolyn McGonagle and colleagues at IOM who carried out the sample analyses. The RISKOFDERM project was funded by the European Commission 5th Framework Quality of Life programme as research contract QLKA4-CT-1999-01107. Participation of the IOM in RISKOFDERM was co-funded by the HSE, through RSU contract 4101/R51.2000.
|
FOOTNOTES |
|---|
* Author to whom correspondence should be addressed. Tel: +44-870-850-5131; fax: +44-870-850-5132; e-mail: graeme.hughson@iomhq.org.uk
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONREFERENCES |
|---|
Fenske, RA. (2000) Dermal exposure: a decade of real progress. Ann Occup Hyg; 44: 489–91.
Garrod ANI, Guiver R, Rimmer DA. (1999) Potential exposure of amateurs (consumers) through painting wood preservative and antifoulant preparations. Ann Occup Hyg; 44: 421–6.
Health & Safety Executive. (1999a) Dermal exposure to non-agricultural pesticides (EH74/3). Exposure assessment document. Sudbury: HSE Books.
Health & Safety Executive. (1999b) Measurement of personal exposure of metalworking machine operators to airborne water-mix metalworking fluid (MDHS 95). Sudbury: HSE Books.
OHSA. (1991) Method ID121: metal and metalloid particulates in workplace atmospheres. Washington, DC: Occupational Safety and Health Administration, US Department of Labor.
OECD. (1997) Guidance document for the conduct of studies of occupational exposure to pesticides during agricultural application. Series on testing and assessment. Environmental Health and Safety Publication no. 9. Paris: OECD.
Rajan-Sithamparanadarajah R, Roff M, Delgado P et al. (2004) Patterns of dermal exposures to hazardous substances in European Union workplaces. Ann Occup Hyg; 48: 285–297.
RISKOFDERM. (1999) Contract number QLK4-CT-1999-01107, 5th Framework Programme of the European Union. Luxemburg: European Commission.
RISKOFDERM. (2001) Risk assessment for occupational dermal exposure to chemicals, First year report. Zeist, The Netherlands: TNO Nutrition and Food Research.
RISKOFDERM. (2002) Risk assessment for occupational dermal exposure to chemicals, Second year report. Zeist, The Netherlands: TNO Nutrition and Food Research.
Schneider T, Vermeulen R, Brouwer DH, Cherrie JW, Kromhout H, Fogh CL. (1999) Conceptual model for assessment of dermal exposure. Occup Environ Med; 56: 765–73.[Abstract]
Soutar A, Semple S, Aitken RJ, Robertson A. (2000) Use of patches and whole body sampling for the assessment of dermal exposure. Ann Occup Hyg; 44: 511–8.
Vermeulen R, Heideman J, Bos RP, Kromhout H. (2000) Identification of dermal exposure pathways in the rubber manufactuirng industry. Ann Occup Hyg; 44: 533–41.
Wheeler JP, Warren ND. (2002) A dirichlet tessellation-based sampling scheme for measuring whole body exposure. Ann Occup Hyg; 46: 209–17.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  P. Boogaard Biomonitoring as a tool in the human health risk characterization of dermal exposure Human and Experimental Toxicology, April 1, 2008; 27(4): 297 - 305. [Abstract] [PDF]
|
|
|
|
 |
|
 A Pronk, F Yu, J Vlaanderen, E Tielemans, L Preller, I Bobeldijk, J A Deddens, U Latza, X Baur, and D Heederik Dermal, inhalation, and internal exposure to 1,6-HDI and its oligomers in car body repair shop workers and industrial spray painters Occup. Environ. Med., September 1, 2006; 63(9): 624 - 631. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. MARQUART, N. D. WARREN, J. LAITINEN, and J. J. VAN HEMMEN Default Values for Assessment of Potential Dermal Exposure of the Hands to Industrial Chemicals in the Scope of Regulatory Risk Assessments Ann. Hyg., July 1, 2006; 50(5): 469 - 489. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. D. WARREN, H. MARQUART, Y. CHRISTOPHER, J. LAITINEN, and J. J. VAN HEMMEN Task-based Dermal Exposure Models for Regulatory Risk Assessment Ann. Hyg., July 1, 2006; 50(5): 491 - 503. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. RAJAN-SITHAMPARANADARAJAH, M. ROFF, P. DELGADO, K. ERIKSSON, W. FRANSMAN, J. H. J. GIJSBERS, G. HUGHSON, M. MAKINEN, and J. J. VAN HEMMEN Patterns of Dermal Exposure to Hazardous Substances in European Union Workplaces Ann. Hyg., April 1, 2004; 48(3): 285 - 297. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||