Annals of Occupational Hygiene Advance Access originally published online on May 26, 2006
Annals of Occupational Hygiene 2006 50(6):609-621; doi:10.1093/annhyg/mel024
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Assessing Isocyanate Exposures in Polyurethane Industry Sectors Using Biological and Air Monitoring Methods
1 Institute of Occupational Medicine Research Avenue North, Riccarton, Edinburgh, UK
2 Health and Safety Laboratory Harpur Hill, Buxton, UK
*Author to whom correspondence should be addressed. E-mail: karen.creely{at}iom-world.org
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODOLOGYRESULTSDISCUSSIONCONCLUSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Isocyanates, as a chemical group, are considered to be the biggest cause of occupational asthma in the UK. Monitoring of airborne exposures to total isocyanate is costly, requiring considerable expertise, both in terms of sample collection and chemical analysis and cannot be used to assess the effectiveness of protection from wearing respiratory protective equipment (RPE). Biological monitoring by analysis of metabolites in urine can be a relatively simple and inexpensive way to assess exposure to isocyanates. It may also be a useful way to evaluate the effectiveness of control measures in place. In this study biological and inhalation monitoring were undertaken to assess exposure in a variety of workplaces in the non-motor vehicle repair sector. Companies selected to participate in the survey included only those judged to be using good working practices when using isocyanate formulations. This included companies that used isocyanates to produce moulded polyurethane products, insulation material and those involved in industrial painting. Air samples were collected by personal monitoring and were analysed for total isocyanate content. Urine samples were collected soon after exposure and analysed for the metabolites of different isocyanate species, allowing calculation of the total metabolite concentration. Details of the control measures used and observed contamination of exposed skin were also recorded. A total of 21 companies agreed to participate in the study, with exposure measurements being collected from 22 sites. The airborne isocyanate concentrations were generally very low (range 0.0005–0.066 mg m–3). A total of 50 of the 70 samples were <0.001 mg m–3, the limit of quantification (LOQ), therefore samples below the LOQ were assigned a value of 1/2 LOQ (0.0005 mg m–3). Of the 70 samples, 67 were below the current workplace exposure limit of 0.02 mg m–3. The highest inhalation exposures occurred during spray painting activities in a truck manufacturing company (0.066 mg m–3) and also during spray application of polyurethane foam insulation (0.023 mg m–3). The most commonly detected isocyanate in the urine was hexamethylene diisocyanate, which was detected in 21 instances. The geometric mean total isocyanate metabolite concentration for the dataset was 0.29 µmol mol–1 creatinine (range 0.05–12.64 µmol mol–1 creatinine). A total of 23 samples collected were above the agreed biological monitoring guidance value of 1.0 µmol mol–1 creatinine. Activities that resulted in the highest biological monitoring results of the dataset included mixing and casting of polyurethane products (12.64 µmol mol–1 creatinine), semi-automatic moulding (4.80 µmol mol–1 creatinine) and resin application (3.91 µmol mol–1 creatinine). The biological monitoring results show that despite low airborne isocyanate concentrations, it was possible to demonstrate biological uptake. This tends to suggest high sensitivity of the biological monitoring method and/or that in some instances the RPE being used by operators was not effective or that absorption may have occurred via dermal or other routes of exposure. This study demonstrates that biological monitoring is a useful tool when assessing worker exposure to isocyanates, providing a more complete picture on the efficacy of control measures in place than is possible by air monitoring alone. The results also demonstrated that where control measures were judged to be adequate, most biological samples were close to or <1 µmol mol–1 creatinine, the agreed biological monitoring benchmark.
Keywords: biological monitoring • inhalation exposure • isocyanates
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODOLOGYRESULTSDISCUSSIONCONCLUSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Isocyanates are a family of highly reactive, low molecular weight chemicals and are considered to be one of the biggest causes of occupational asthma in the UK (Huggins et al., 2001). The most commonly used isocyanates include diphenyl methane diisocyanate (MDI), toluene diisocyanate (TDI), hexamethylene diisocyanate (HDI) and isophorone diisocyanate (IPDI). Isocyanates are the raw materials from which all polyurethane products are made and are widely used in the manufacture of flexible and rigid foams, coatings such as paints and varnishes and elastomers (Klees and Ott, 1999). The greatest number of cases of isocyanate-induced asthma is observed in spray painters within the motor vehicle repair (MVR) sector (Huggins et al., 2001), and the determinants of exposure have been investigated in detail (Woskie et al., 2004). However, isocyanates are increasingly being used in general manufacturing, construction and for insulation purposes, and relatively little is known about the numbers of workers potentially exposed, their methods of work and the types of control measures used. This paper originates from a comprehensive study of isocyanate users in the UK covering the MVR and other industries (Cowie et al., 2005). Based on the extrapolation of figures obtained from telephone and postal questionnaires of MVR and non-MVR sectors, it was estimated that within the general UK population there are
6200 MVR companies using isocyanates, with 15 000 employees being directly exposed. For non-MVR sectors it was estimated that 500 UK companies use isocyanates, with 7000 employees being directly exposed, although it was considered that these numbers may in fact be higher owing to lack of a comprehensive listing of all relevant companies. Inhalation exposure data from non-MVR sectors was also collected in the study, and in this paper we also consider biological monitoring data obtained from the non-MVR industry sector, carried out simultaneously, but not reported by Cowie et al. (2005). Estimation of personal exposure to isocyanates is usually done by measuring airborne concentrations of isocyanates by personal sampling methods, and this method is described in the methods of determination of hazardous substances (MDHS) 25/3 (HSE, 1999). However, the air monitoring method is not particularly well suited to workplace conditions and the chemical analysis is complex. It is widely held that the collection and analysis of air samples requires considerable expertise (Streicher et al., 2000; White, 2006), which tends to make the procedure relatively costly. While air monitoring can provide information about the effectiveness of engineering controls such as enclosures and local exhaust ventilation (LEV), it is difficult to use it to assess the effectiveness of respiratory protective equipment (RPE). Alternatively, biological monitoring by analysis of metabolites in urine is a relatively simple and inexpensive way to determine the effectiveness of all the controls applied to a particular workplace and can provide direct information on personal exposure. Several volunteer and occupational studies show a good correlation between airborne concentrations of isocyanates and their hydrolysable adducts in urine (TDI: Rosenberg and Saviolainen, 1986, Brorson et al., 1991, Maitre et al., 1993, Persson et al., 1993, Kaaria et al., 2001a; HDI: Maitre et al., 1996, Tinnerberg et al., 1995; IPDI: Tinnerberg et al., 1995). The UK Health and Safety Executive (HSE) has been using biological monitoring to help assess occupational exposure to isocyanates, particularly in the MVR sector (Williams et al., 1999) and also in a wide range of other industries.
In October 2005, HSE's Working group on Action To control CHemicals (WATCH) committee using data from over 2700 urine samples agreed on the establishment of a biological monitoring guidance value (BMGV) for isocyanate metabolites in urine, set at 1.0 µmol urinary diamines per mol creatinine released by hydrolysis of protein conjugates of HDI, TDI, MDI or IPDI. This value was set on the basis that a concentration of urinary diamines at or below this level is associated with good control of exposure (HSE, 2005a).
In this study, we assessed workers inhalation and biological exposure to isocyanates in companies using these products for non-MVR purposes in order to determine the exposure levels that are achievable using recognized good working practices and exposure control methods for a range of different tasks and categories of use. This study did not assess exposures during the thermal degradation of polyurethane products where isocyanate subunits may be re-created resulting in exposure to isocyanates (Rosenberg et al., 2002).
|
METHODOLOGY |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODOLOGYRESULTSDISCUSSIONCONCLUSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Company selection
In Part 1 of the study by Cowie et al. (2005) a comprehensive telephone and postal survey of isocyanate users was undertaken. A subset of companies included in this survey was asked to participate in a workplace monitoring survey. Some additional companies were also recruited by contacting members of industry associations involved with isocyanate uses. This included members of the British Rubber Manufacturers Association (BRMA) and the British Urethane Foam Manufacturers Association (BrUFMA).
The sample selection aimed to include only those companies using ‘good practice’ in the handling and use of isocyanates. This was intended to ensure that exposure levels were consistent with what could be achieved by using good working practices and exposure control methods. In order to be considered, the company's processes either had to be contained or be used with LEV and RPE.
A total of 122 companies were contacted by letter and then by telephone to explain the remit of the workplace surveys and establish willingness to participate in the study. Those companies that agreed to participate were then categorized according to the way they used isocyanates, and a total of eight task categories were identified. These were chemical processing, coating/spreading; gluing/sealing; mixing; moulding/injecting; painting (brush or roller); painting (spray) and polyurethane spraying. Most companies were uniquely classified according to a single category of use; however, for several companies more than one type of operation was carried out.
Sampling and analysis of airborne isocyanates
Inhalation exposure sampling was carried out for specific tasks, e.g. spray painting, moulding, gluing, etc. Personal air sampling was carried out according to the HSE method MDHS 25/3 (HSE, 1999), using IOM inhalable dust sampling heads with chemically treated glass fibre filters loaded into stainless steel filter cassettes. The filters were pre-treated with 1-(2-methyoxyphenyl) piperazine (1-2MP) solution, as the derivatising agent. The method used is recommended for sampling vapour phase isocyanates, whereas mixtures of airborne particles and vapours should be sampled using an impinger with an in-line chemically treated filter between the impinger and sampling pump. The scientific justification of this method has been discussed by White (2006). However, this sampling train was not always used in our study owing to the risk of spillage of the solvent onto the wearer. Although, it is possible for the filters to become clogged (MDHS 25/3), the performance of the sample was assessed by measuring the sampling flow regularly over the monitoring period. In instances where the flow rate was not maintained to within ±5% of the nominal value the sample was discarded.
Immediately after sampling the filters were placed in vials containing sufficient 1–2 MP to completely cover the filter. This was done in order to ensure complete derivatisation of any residual free isocyanates. The samples were analysed for total isocyanate content (expressed as –NCO group) at the laboratory of the Institute of Occupational Medicine (IOM), Edinburgh. The procedure used high performance liquid chromatography, in accordance with the method described in MDHS 25/3 (HSE, 1999). Peaks within the isocyanate range that were not isocyanate monomers were included as pre-polymer, unless proven to be interferences. The pre-polymer values were calculated using the ratios stipulated within MDHS 25/3 (HSE, 1999). The detection limit for this method was established at 0.02 µg total –NCO in each filter sample. The IOM laboratory is accredited for this analysis under the United Kingdom Accreditation Service (UKAS) accreditation no. 0374.
The results from this analysis were used to calculate the total airborne isocyanate concentration for each sample in terms of the total NCO content. The limit of quantification (LOQ) for routine occupational hygiene samples of this type was 0.001 mg m–3. Samples below the LOQ were assigned a value of 
LOQ (0.0005 mg m–3).
Sampling and analysis of biological samples
All workers included in the air sampling surveys were asked to give their informed consent and submit a urine sample.
Since the biological half-life of the urinary metabolites of HDI, TDI and IPDI are roughly 1–2 h (Brorson et al., 1990, 1991; Tinnerberg et al., 1995), samples were collected as soon as possible after the exposure period had ended. Samples were either collected immediately after completion of the task (where exposure was over a specific period of time) or at the end of the shift (where exposure was evenly spread over the day).
All urine samples were collected in polystyrene universal containers (30 ml) bottles pretreated with citric acid (0.5 g) and submitted to the Health and Safety Laboratory (HSL), Buxton, for analysis based on the method by Williams et al. (1999) & Rosenberg et al. (2002) but using analogue internal standards. Briefly; aliquots of the samples (2 ml) had internal standards (100 µl heptane diamine 1 µM and ethylenediamine 5 µM) added and were acidified with concentrated sulphuric acid (200 µl). These were then incubated in sealed tubes at 100°C for 90 min to release free isocyanate-derived diamines from protein and/or other conjugates in urine. After cooling the urine was made alkaline with 2 ml of 10 M sodium hydroxide and extracted with diethylether (4 ml). A portion (3 ml) of each ether extract was removed to clean tubes and the solvent evaporated under nitrogen. The residue was derivatised with heptafluorobutyric anhydride (50 µl) in toluene (500 µl) in closed tubes at 55°C for 1 h. After cooling the excess derivatising agent was removed under nitrogen and the residue reconstituted in toluene (100 µl). Injections (1 µl) were made splitless (350°C, 30 s) into a capillary column (30 m x 0.3 mm BP5 1 µm) at 150°C increasing at 10°C min–1 to 240°C then at 20°C min–1 to 300°C. Detection was by negative ion CI (methane) mass spectrometry using a Hewlett Packard 5973 mass selective detector. 2,4-Toluene diamine (TDA) and 2,6-TDA were used as markers of TDI exposure; 1,6-hexane diamine (HDA) as a marker of HDI exposure; 4, 4 methylenedianiline (MDA) as an indicator of MDI exposure and isophorone diamine (IPDA) as an indicator of isophorone diisocynate exposure. The detection limit (3x background) of the method was 1 nmol l–1, the coefficient of variation for within day analysis was 5% and for day to day analysis 12% at 200 nmol l–1. The results from this analysis were then used to calculate the total urinary isocyanate concentration for each sample in terms of the total metabolite content to allow comparison with the ‘total’ isocyanate in air. Samples below the LOQ were assigned a value of LOQ (0.05 µmol mol–1 creatinine) to allow data analysis.
Collection of contextual information and evaluation of control measures
Details about the working practices and production rates for each company were recorded to help ascertain whether working conditions were typical. These were recorded on a proforma, which was also used to collect information about the type of personal protective equipment (PPE) and RPE that was used, what condition it was in, whether ventilation equipment was used and whether it performed adequately. The suitability of the protective clothing and respiratory protection for the specific application was evaluated with reference to the manufacturer's guidance and from published guidance. In addition, details about observed contamination of clothing and exposed skin were recorded in order to help assess the pattern and extent of dermal contact with the chemical substances used.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODOLOGYRESULTSDISCUSSIONCONCLUSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Of the 122 companies invited to participate in the study 21 companies agreed (17% of the invited study population), resulting in 22 participating sites. Table 1 shows the type of business carried out by each site and the categories of use of isocyanates. The commercial vehicle manufacturer with site codes 11 and 18 was a single company, but with a number of different depots operating across the country. Since the two sites identified for sampling were separated geographically and also subject to different management regimes it was considered appropriate to code these as separate operating units. Based on the estimation that there were 500 companies using isocyanates throughout the UK (Cowie et al., 2005), and not considering that some of these may have multiple sites, the population included in the study represented 4.2% of the UK population of non-MVR isocyanate users.
|
After the site visits were completed it was evident that there was some overlap between the original allocated task categories. More appropriate categories were, therefore, developed to help summarize the exposure data, taking into account the nature of the work observed. These were as follows:
- Painting (roller) (one company)—The mixing and subsequent application of paints onto the hull of a ship using long handled roller implements. This work was undertaken outside, with no physical protection from the elements. The painters wore cotton/leather rigger type gloves. The isocyanate present in the curing agent was HDI.
- Quality assurance/control (three companies)—This involves a range of tasks, for example, collecting and dispensing small quantities of mixtures (<100 ml) for transfer to the laboratory for testing. Tasks were either naturally ventilated, carried out in fume cupboards or with LEV. Disposable gloves were worn in all cases except one. MDI and TDI were the most common isocyanates present in these tasks.
- Mixing and casting (two companies)—The manufacture of polyurethane items involving the mixing and pouring of isocyanate components into moulds, which were left to cure for a required time before removal. LEV was present and either heat-resistant or disposable latex gloves were worn. MDI was present in most formulations, with IPDI being present in only one of those used.
- TDI plant operations (one company)—TDI pre-polymers were manufactured using two reaction vessels through which TDI monomer was reacted. LEV was present at the workstations and PVC gauntlets were worn during the task. TDI was obviously the isocyanate of concern in this operation.
- Moulding (automatic) (one company)—Block rigid foam panels were manufactured using a fully automated enclosed production line with LEV, with operators being stationed at various intervals to ensure the process was running smoothly and to specification. Rigger type gloves were worn by the operators. The isocyanate formulation used in the process contained MDI.
- Saw operations (two companies)—The cutting of block rigid foam/composite panels by an automated process, being supervised by operators to ensure the required specifications were being met. These processes were either enclosed or had LEV present to capture contaminants. The isocyanate formulations used in these processes contained MDI.
- Unloading/storage finished product (three companies)—This involved the removal and storage of finished items from the process and includes processes such as palletising and fork lift removal. Natural or dilution ventilation was in place and rigger type gloves were worn in two instances. The isocyanate formulations used in the primary processes carried out at these sites contained MDI.
- Moulding (semi-automatic) (two companies)—In these tasks polyurethane foam was used in an injection moulding process to produce dashboard elements for cars or seat mouldings and other similar products. The polyurethane components were mixed and injected into the mould, being left to cure for the required period of time before removal. Lint-free or disposable latex gloves were normally worn, with the processes being either enclosed or naturally ventilated. The polyurethane products used in these companies contained MDI.
- Coating/spreading (supervising) (four companies)—The coating and spreading task included the manufacture of laminated flexible packaging materials and the bonding of plywood panels to block rigid foam slabs. The laminating process essentially involved the dispensing of adhesive onto a roller, coating the packaging material as it passed through the laminating machine. A second film was then bonded to the pre-coated film and fed onto a spool that was then removed for storage. Block rigid foam panels were placed onto a conveyor belt and sprayed with adhesive and stacked on a trolley, placing plywood on top for pressing and curing. In two instances half mask respirators with A2 organic filters were worn during mixing and disposable latex or vinyl gloves were worn in all except two instances. LEV and/or natural ventilation were used during these tasks. The adhesives used in these companies all contained MDI.
- Resin operator (one company)—Isocyanate resins were mixed and dispensed onto sheets of polyester/vinyl labels using a fully automated application system, with the coated labels passing through a series of heaters before being removed and placed on a cooling rack. LEV and dilution ventilation was present. The polyurethane resin hardener contained IPDI.
- Painting (spray) (three companies)—The application of isocyanate-based paints using spray equipment, for example, in the vehicle manufacturing industry. All spray-painting activities were undertaken in a spray booth, with air-fed visors also being worn. Anti-static, disposable vinyl, or latex or lint-free gloves under chemical protective gloves were also worn. The top coat hardener of the paint formulations used contained HDI.
- Glazing operator (one company)—During this task glazing panels were fixed to vehicle window frames using an isocyanate adhesive/sealant. The operators applied activator onto the panels and adhesive was then applied by robot. The operators then lifted and pressed the panels into position. The area was naturally ventilated and stringknit gloves were worn, which provided no chemical protection. The adhesive contained MDI.
- Polyurethane spraying (three companies)—This includes the spray application of polyurethane to the interior of an agricultural building, onto copper cylinders and a range of metal substrates. Air-fed visors or half-mask respirators were worn during spray application in all except one instance, with the tasks being undertaken in a spray booth or a naturally ventilated area. Chemical-resistant gloves were also worn during spray activities in all except one instance. MDI was the isocyanate of interest in these three companies.
Inhalation exposure
A total of 160 personal exposure measurements were obtained during this study. This included both long-term average exposures and task-based exposure measurements. However, only 70 workers provided urine samples so only their associated air samples (n = 70) are discussed here. Information on the dataset of all 160 inhalation exposure measurements can be found in Cowie et al. (2005).
Exposures were generally low, with only 20 samples showing air concentrations 
0.001mg m–3, the LOQ. The inhalation results are summarized into the task categories in Table 2. Samples below the LOQ were assigned a value of LOQ (0.0005 mg m–3).
|
Overall, the airborne concentrations were low, (Geometric mean (GM) = 0.0009 mg m–3; range 0.0005–0.0658 mg m–3). The greatest spreads in the range of measurements obtained were for spray painting, coating and spreading, semi-automatic moulding and polyurethane spraying. The highest exposure measurements were obtained for semi-automatic moulding (GM = 0.0021 mg m–3) and spraying polyurethane foam insulation (GM = 0.0020 mg m–3). In three instances the average exposure exceeded the 8 h time weighted average workplace exposure limit (WEL) of 0.02 mg m–3 (HSE, 2005b). These measurements related to the spraying of polyurethane foam insulation (0.0226 mg m–3) and spray painting of trucks (0.0360 and 0.0658 mg m–3). Spray booths or other ventilation equipment that was not working properly was available in the spray painting situations, although the workers were protected by the use of supplied air breathing apparatus. All other inhalation exposure measurements were below the WEL, with the lowest exposures recorded being obtained for the operators involved in glazing, resin application tasks, TDI operations and automatic moulding activities where all the measurements collected were below the LOQ. These had a combination of LEV and either dilution ventilation or enclosed processes.
The inhalation exposure measurements were summarized by type of ventilation control in place and are detailed in Table 3. Operators working in spray booths or on naturally ventilated processes had the greatest range of exposures (0.0005–0.0658 and 0.0005–0.0226 mg m–3, respectively), with operators working on processes with just dilution or dilution and LEV all having exposures less than the LOQ. Two of the instances where the average exposure exceeded the 8 h time weighted average WEL occurred when spray booths were used (0.036 and 0.066mg m–3), with the remaining exposure exceeding the WEL occurring when natural ventilation was used.
|
Biological monitoring
The majority of urine samples (67 out of 70) were analysed for all isocyanate metabolites (MDA, 2,4-TDA; 2,6-TDA; 1,6-HDA and IPDA); however, three samples at the start of the project were only analysed for the metabolites of the specific isocyanate thought to be present (MDA), giving a total of 70 samples. The consent obtained from the workers was modified to allow analysis of ‘isocyanate metabolites’ rather than ‘metabolites of x’ (single isocyanate). The total urinary isocyanate metabolite content was calculated and the biological monitoring results are also summarized by task categories (Table 4).
|
Overall, the geometric mean total isocyanate metabolite level for the dataset was 0.29 µmol mol–1 creatinine, (range 0.05–12.64 µmol mol–1 creatinine). HDA was the most commonly detected metabolite in the urine samples, being detected in 21 instances. The least commonly detected isocyanate was 2,4-TDA, being detected on only two occasions, both following mixing and casting operations. In the majority of instances, exposure to only one isocyanate metabolite was detected; however, there were several tasks that resulted in multiple isocyanate metabolite exposure. These included mixing and casting (all five isocyanate metabolites), polyurethane spraying and quality assurance/control (2,6-TDA and MDA for MDI), semi-automatic moulding and coating and spreading (1,6-HDA and MDA for MDI).
Scrutiny of the material safety data sheets for the products in use did not explain the origin of all the isocyanate metabolites and the air samples analysis did not reveal the presence of unexpected airborne isocyanate groups. This may indicate that the analytical method for the air samples was unable to explicitly detect all the isocyanate groups present or that the route of exposure for these compounds was by routes other than inhalation.
Twenty-three operators were found to have total isocyanate metabolite exposures >1.0 µmol mol–1 creatinine, the agreed BMGV for isocyanates (HSE, 2005a). Activities resulting in high biological levels included resin application (3.9 µmol mol–1 creatinine, based on one result), mixing and casting (GM = 2.33 µmol mol–1 creatinine) and glazing (GM = 0.92 µmol mol–1 creatinine). The highest total metabolite exposure occurred in a casting operator and was found to be 12.64 µmol mol–1 creatinine, and of the eight mixing and casting operatives, five had metabolite levels in excess of 5 µmol mol–1 creatinine. The lowest biological monitoring results were observed in storage tasks and sawing operations during the manufacture of composite panels, with the GM of these tasks being <0.1 µmol mol–1 creatinine). A total of 33 operators were found to have all their metabolite exposures less than the limit of detection. This included all except one of the unloading/storage of finished products operatives, 6 of the 8 coaters/spreaders, 2 of the 3 saw operators, 2 of the 4 automatic moulding operatives and 5 of the 13 spray painters.
Total isocyanate metabolite exposure and types of control measures used in the tasks were also explored (Table 5). The suitability and appropriateness of the gloves and RPE being used is not assessed in Table 5, nor is the effectiveness of the ventilation controls in place. Although the data suggest that operators wearing gloves had higher exposures, it is interesting to note that the urinary metabolite concentrations for operators with obvious dermal contamination was over two times that of operators thought not to have received any. The geometric mean exposure level for both RPE users and non-users was the same. Operators working on enclosed processes had the greatest exposure; however, this result was based on only two measurements (GM = 4.18 µmol mol–1 creatinine). All operators working on processes where dilution ventilation was used had exposures below the LOQ, though this was only based on two measurements.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODOLOGYRESULTSDISCUSSIONCONCLUSIONACKNOWLEDGEMENTSREFERENCES |
|---|
The 21 companies, resulting in 22 separate sites included in the workplace surveys, represented a wide range of non-MVR isocyanate users in the UK.
It was difficult to recruit companies into the monitoring programme because of concerns regarding time and resource pressure, and company confidentiality. As with any study of this nature there is always concern regarding how representative the recruited companies are. Based on the information collected by telephone or postal surveys of isocyanates users in non-MVR sectors (Cowie et al., 2005), in our judgement, the companies that did participate were a representative cross-section of isocyanate users in the UK, biased towards those having good working practices. The sampling surveys identified a number of important issues. In particular, they showed that small and medium-sized enterprises are dynamic workplaces, and often the production processes are only partly developed. Chemical formulations often need to be manufactured or blended in-house to meet certain specifications, rather than being bought in. This resulted in extensive manual mixing and pouring tasks, and although they were carried out relatively infrequently, did represent a significant potential for both inhalation and dermal exposure. While it was our original intention to include companies that only applied good practice controls it was clear that in some instances procedures were deficient and improvements to ventilation solutions, PPE/RPE and managerial controls could easily be made.
Reliable analysis of inhalation exposure is difficult and requires a high level of expertise (Streicher et al., 2000; White, 2006). The existing exposure limits for isocyanates are based on the total isocyanate group, i.e. the sum of the monomeric and polymeric forms of these chemicals, which should be included in the exposure assessment. While analysis of monomeric diisocyanate compounds should be reasonably straightforward, it is a difficult technical procedure to quantify levels of polyisocyanates in exposure samples. Since polyurethane products may contain significantly more polyisocyanate compounds than monomeric isocyanates, it is possible that workplace exposures may be underestimated owing to the problems associated with the methodologies used (Streicher et al., 2000). It should also be noted that this study did not include assessment of exposures during the thermal degradation of polyurethane products where isocyanate subunits may be re-created resulting in exposure to isocyanates.
This not withstanding, the airborne concentrations for the majority of workplace tasks appeared to be low, and mostly below the limit of detection for the measurement method. There were only four personal exposure measurements in total, which might be regarded as significant, using a nominal value of of the 8 h WEL, i.e. 0.01 mg m–3 as an indicator. These samples were collected from four different workplaces and predominately related to spray application of coatings. These results are supported by the findings of other researchers who have noted similar isocyanate concentrations during industrial spray painting (England et al., 2000). In two of these workers control relied on air-fed masks and no metabolites were detected in urine. One worker used a half-mask and metabolites were detected in urine underlining the difficulties in selection and use of respiratory protection. The fourth worker was exposed to mostly polymeric HDI (no free HDI monomer was detected) and did not use any respiratory protection. Five other workers had lower personal airborne exposures (range 0.0005–0.0072 mg m–3) to essentially polymeric HDI, and HDA was also found in their urine. This suggests either a remarkable sensitivity of the biological monitoring method if it is only detecting isocyanate monomer or the method may also detect exposures to polymeric isocyantes, possibly by hydrolysis of polymeric conjugates. An alternative possibility is that the low levels of diamines found were due to non-occupational exposure. However, although isocyanate-related diamines have been reported in urine of people not occupationally exposed to isocyanates (Rosenberg et al., 2002; Sennbro et al., 2005) the levels found were below the detection limit of the method used in this study.
Based on the maximum exposures observed in the inhalation dataset, the highest exposure measurements were obtained from spray painting and polyurethane spraying operations in which the level of control was considered to be either rudimentary or working incorrectly (based on the occupational hygienists subjective assessment). These exposures were above the 8 h time weighted average WEL. Nevertheless, in all these cases the workers involved were or appeared to be adequately protected by supplied air breathing apparatus. These companies are probably not consistent with the inclusion criteria for the survey design, i.e. that the companies included should apply ‘good practice’. However, these particular work applications probably represent the most hazardous types of use of isocyanates and help illustrate typical exposures in less well-controlled environments. The exposure levels for polyurethane processes, not involving spraying, such as resin application and moulding, were very low and often below the limit of detection for the analytical method. This seems to be broadly in line with the findings of the limited available research in this particular sector (Sennbro et al., 2004).
In the majority of cases where there was potential for inhalation exposure, the correct type of RPE was being used. Furthermore, the RPE used by companies carrying out more hazardous operations was generally well maintained. However, under normal situations where there was no significant risk of inhalation exposure, a wide range of different types of RPE had been provided, mainly as a precautionary measure for dealing with spillages. Most of these were inappropriate for use with isocyanates and may help explain why the total urinary metabolite concentrations were not notably different between RPE users and non-users. While the RPE devices may have been intended to provide protection against organic solvents for example, the workers did not always seem to appreciate this difference and it is conceivable that they may have relied on these devices, perhaps when carrying out some non-standard work where isocyanate concentrations were higher than normal. It is also possible that in instances where the correct RPE had in fact been supplied, this was not in fact being correctly used by the operator, which may also help explain the higher urinary metabolite concentrations observed. Further analysis of the biological sampling data revealed that in 10 of the 23 samples above the BMGV, HDA the marker of HDI exposure was the most common metabolite, thus, suggesting that in those instances the inhalation route of exposure predominated owing to the higher volatility of this isocyanate. However, the inhalation results suggest that it is reasonable to conclude that exposures can easily be controlled to the current 8 h average WEL levels using basic exposure control methods.
While the airborne concentrations across the majority of tasks were controlled to acceptable levels, there were numerous examples of poor chemical handling methods, resulting in potential for inhalation exposure or skin contact, particularly during spray painting and manual mixing and pouring tasks. Many companies had not implemented robust dermal exposure control measures and the types of protective gloves selected by companies were a compromise between chemical protection and dexterity requirements. Consequently workers may have been poorly protected because of the gloves being inappropriate for use with the isocyanate products, which may explain why operators wearing gloves had higher urinary metabolite concentrations than those who did not. In addition, there were cases, e.g. during spray painting, where there was evidence of exposure to other parts of the body, including the arms, face and neck. This is likely to mean that there is significant potential for dermal exposure across the wide range of industrial isocyanate users. This is important because operators judged to have experienced dermal contamination to isocyanates during the sampling period were found to have urinary metabolite concentrations over two times that of operators who had not. This suggests that the dermal route may in some cases be a major contributor to total body exposure to isocyanates. It is also recognized that isocyanates are corrosive to the skin and eyes, and may result in skin sensitization (Goossens et al., 2002), so it is important for companies to provide and maintain appropriate personal protection regimes. Furthermore, there are suggestions that dermal exposure to isocyanates may result in respiratory sensitization (Johnson et al., 2004; Vanoirbeek et al., 2004) and while this link has not definitely been established, it provides additional incentive to control dermal contact as far as possible.
There is scope also for applying biological monitoring methods in the exposure assessment process. Isocyanate compounds have different volatilities with the relative proportion that can be contributed to inhalation and dermal exposure varying, in part, with this characteristic. Biological monitoring allows assessment of all the isocyanate compounds rather than sampling those with higher risk of becoming airborne. Urinary metabolites may be a more useful indicator of exposure than airborne concentrations, since they can be reliably quantified with only limited potential for interference from other chemicals. However, it is important that this method is supported by observations of the process so that results can be put in context and the key routes of exposure determined.
In most cases urine samples were collected immediately after a specific task, while in other cases samples were taken at the end of the shift where tasks may have been carried out on multiple occasions. Given the relatively short half-lives for all isocyanates (except possibly MDI) the concentration of metabolites found in urine mostly reflects the exposure during the period between the previous urine voiding and sample collection (typically 2–4 h). If tasks were repeated during the day there might be a small additional contribution or bias in the end of exposure samples from earlier exposures; however, it was felt that this was a preferable alternative to collecting samples only at the end of shift, which would have risked not detecting brief exposures some hours earlier. The biological samples showed in several instances that operators had absorbed isocyanates despite using control measures such as RPE, ventilated work areas and protective gloves. Indeed, 23 of the 70 samples were above the recently agreed BMGV of 1.0 µmol mol–1 creatinine. This finding is similar to that reported by Kaaria et al. (2001b) who report detecting MDA in urine from 97% of workers (n = 57) exposed to MDI moulding rigid polyurethane foam even though the MDI in air was below the detection limit for the air method for 67% of the workers. Schutze et al. (1995) also found that MDI exposure could be detected in more workers by measurements of urine metabolites than by air monitoring.
Two biological samples from different companies indicated significant exposures to isocyanates. For the first of these, collected in a conveyor belt manufacturing company, an operator's exposure to isocyanates was assessed during the manufacture of a polyurethane splicing paste that involved mixing and decanting tasks. LEV was present and the airborne concentrations of isocyanates were all very low and certainly much less than the WEL. Disposable vinyl gloves were worn, along with a half mask respirator during mixing. The end-of-shift urine sample indicated 16.3 µmol MDA per mol creatinine, suggesting a high exposure to MDI. However, this result did not correspond to the working practices and level of contact with the isocyanate formulations observed. The safety data sheet for the product also stated that it contained only very low residual levels of isocyanate. A possible explanation for the elevated MDA levels was that exposure to MDA itself occurred, rather than it being a direct metabolite of MDI exposure. Further investigation revealed that 3 days prior to sampling, a batch of paste containing MDA was mixed by the operator. It is possible that this previous task involving MDA contributed to the urine result, possibly due to inadequate handling procedures and the result was therefore removed from the dataset. This illustrates the usefulness of biological sampling as a means of identifying cryptic exposures where they might not normally be identified.
The second biological sample was obtained from a company involved in casting, moulding and spraying polyurethane foam onto a range of substrates. Again, the airborne concentrations of isocyanates were well below the WEL and LEV was also present. Despite this, one of the operator's end-of-shift urine sample indicated a markedly higher exposure of 12.64 µmol mol–1 creatinine. The gloves provided were noted to provide limited chemical protection and skin contamination was also evident. It is possible that skin absorption of the products occurred, which contributed to the total body exposure and this result was, therefore, retained in the dataset.
Both these findings highlight the importance of the role of biological monitoring, demonstrating that although operators airborne exposure may be minimal, significant exposure can occur via other routes such as dermal and ingestion, which would not normally be quantified. The results also demonstrate the role of biological monitoring in assisting with the assessment of the effectiveness of the control measures in place and can also be used to highlight whether RPE are being used correctly. The conveyor belt operator sample also illustrates the importance of collecting information on possible exposure to chemicals that are known interferences to the sampling and analytical procedures and how a result may highlight a problem, which may have otherwise gone unnoticed.
Forty seven operators were found to have total metabolite exposures 
1.0 µmol mol–1 creatinine and in the majority of instances the controls in place were judged to be adequate and included the use of skin protection as well as ventilation controls. Where exposure controls were judged to be deficient in some way, urinary diamine was normally detected. It is, therefore, suggested that total exposures <1.0 µmol mol–1 creatinine, the agreed BMGV, can be achieved where good practice and control of exposure is achieved. The observation of 23 urine samples with diamine levels above the BMGV rather than the expected 7 (this value being obtained on the basis that the BMGV is based on the 90th percentile of places with good control) could underline the basic point that an assumption about the level of control in a workplace is often not supported by observation of the actual work practices. In this study several of the 22 workplaces could have improved their control of isocyanate with relatively simple and inexpensive techniques. Perhaps this should be a stimulus to occupational hygienists and regulators to visit more workplaces.
|
CONCLUSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODOLOGYRESULTSDISCUSSIONCONCLUSIONACKNOWLEDGEMENTSREFERENCES |
|---|
It was our original intention to include only companies that applied good practice exposure controls, but it was clear that in some cases their procedures were deficient. In many instances, there was scope for improved managerial controls or technical solutions such as enclosed isocyanate delivery and dispensing systems, or more efficient containment or ventilated booths. The airborne concentrations across the majority of tasks were found to be controlled to acceptable levels, with only three airborne samples being in excess of the current WEL, although these workers were protected through the use of appropriate RPE. However, air sampling cannot be used to assess the effectiveness of protection for workers wearing RPE. Biological monitoring methods, targeted at urinary metabolites are an acceptable method for assessing control (both engineering and behavioural aspects) and give direct information on personal exposure. The urine sample results revealed that measurable biological uptake occurred, with 23 of the 70 urine results being in excess of the agreed BMGV. This suggests absorption may have occurred via dermal, inhalation and possibly even ingestion routes of exposure. Indeed in several of these instances the effectiveness of the protective gloves in providing adequate chemical protection was questionable and dermal contamination was clearly evident. The results also demonstrated that where control measures were judged to be adequate, most samples were close to or <1.0 µmol mol–1 creatinine. This demonstrates that biological monitoring is a useful tool when assessing worker exposure to isocyanates, giving a more complete picture on the efficacy of the control measures in place than is possible by air monitoring alone, providing that the methodology is supported by observations of the process in question so that the results can be put into context and the key routes of exposure determined.
|
ACKNOWLEDGEMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODOLOGYRESULTSDISCUSSIONCONCLUSIONACKNOWLEDGEMENTSREFERENCES |
|---|
The authors would like to thank the management and workers of all the companies who participated in the air and biological monitoring surveys. We would also like to thank the staff at the IOM and HSL laboratories for their analysis of the inhalation and biological samples collected. Our thanks also go to Dr Anne Soutar (IOM) and Dr Sean Semple (University of Aberdeen) for their comments on the paper. This work was funded by the UK Health and Safety Executive under contract number 4305/R51.229.
Received November 9, 2005; in final form February 21, 2006
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODOLOGYRESULTSDISCUSSIONCONCLUSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Brorson T, Skarping G, Nielsen J. (1990) Biological monitoring of isocyanates and related amines II. Test chamber exposure of humans to 1,6 hexamethylenediisocyanate (HDI). Int Arch Occup Environ Health 62:385–89.[CrossRef][ISI][Medline]
Brorson T, Skarping G, Sango C. (1991) Biological monitoring of isocyanates and related amines IV. 2,4 and 2,6 toluenediamine in hydrolysed plasma and urine after test chamber exposure of humans to 2,4 and 2,6 toluene diisocyanate. Int Arch Occup Environ Health 63:253–59.[CrossRef][ISI][Medline]
Cowie HA, Hughson GW, Creely KS, et al. (2005) An occupational hygiene assessment of the use and control of isocyanates in the UK. HSE Research Report 311. (Health and Safety Executive, HSE Books, Sudbury, England).
England E, Key-Schwartz R, Lesage J, et al. (2000) Comparison of sampling methods for monomer and polyisocyanates of 1,6–hexamethylene diisocyanate during spray finishing operations. Appl Occup Environ Hyg 15:472–78.[CrossRef][Medline]
Goossens A, Detienne T, Bruze M. (2002) Occupational allergic contact dermatitis caused by isocyanates. Contact Dermatitis 47:304–8.[CrossRef][ISI][Medline]
HSE. (1999) Methods for the determination of hazardous substances: organic isocyanates in air. MDHS 25/3. (Health and Safety Executive, HSE Books, Sudbury, England).
HSE. (2005a) Biological monitoring for isocyanates. WATCH 2005/13. Available from: http://www.hse.gov.uk/aboutus/hsc/iacs/acts/watch/051005/13.pdf, accessed 17 October 2005.
HSE. (2005b) Workplace exposure limits, EH40/2005. (Health and Safety Executive, HSE Books, Sudbury, England).
Huggins V, Anees W, Robertson AS, et al. (2001) SHIELD 2001: A surveillance scheme of occupational asthma in the Midlands. Available from: http://www.occupational-asthma.com/occupationalasthma/shield/shield2001.shtml.
Johnson VJ, Matheson JM, Luster MI. (2004) Animal models for diisocyanate asthma: answers for lingering questions. Curr Opin Allergy Clin Immunol 4:105–10.[CrossRef][Medline]
Kaaria K, Hirvonen A, Norppa H, et al. (2001a) Exposure to 2,4- and 2,6- toluene diisocyanate (TDI) during production of flexible foam: determination of airborne TDI and urinary 2,4- and 2,6–toluenediamine (TDA). Analyst 126:1025–31.[CrossRef][Medline]
Kaaria K, Hirvonen A, Norppa H, et al. (2001b) Exposure to 4, 4'-methylenediphenyl diisocyanate (MDI) during moulding of rigid polyurethane foam: determination of airborne MDI and urinary 4, 4'-methylendianiline (MDA). Analyst 126:476–9.
Klees JE and Ott MG. (1999) Diisocyanates in polyurethane plastics applications. Occup Med 14:759–76.[Medline]
Maitre A, Berode M, Perdix A, et al. (1993) Biological monitoring of occupational exposure to toluene diisocyanate. Int Arch Occup Environ Health 65:97–100.[CrossRef][ISI][Medline]
Maitre A, Berode M, Perdix A, et al. (1996) Urinary hexane diamine as an indicator of occupational exposure to hexamethylenediisocyanate. Int Arch Occup Environ Health 69:65–8.[CrossRef][ISI][Medline]
Persson P, Dalene M, Skarping G, et al. (1993) Biological monitoring of occupational exposure to toluene diisocyantate: measurement of toluenediamine in hydrolysed urine and plasma by gas chromatography-mass spectrometry. Br J Ind Med 50:1111–8.
Rosenberg C and Saviolainen H. (1986) Determination of occupational exposure to toluenediisocyanate by biological monitoring. J Chromatogr 367:385–92.[CrossRef][ISI][Medline]
Rosenberg C, Nikkila K, Henriks-Ekerman ML, et al. (2002) Biological Monitoring of aromatic diisocyanates in workers exposed to thermal degradation products of polyurethanes. J Environ Monit 4:711–6.[CrossRef][ISI][Medline]
Schutze D, Sepai O, Lewalter J, et al. (1995) Biomonitoring of workers exposure to 4,4'-methylenedianiline or 4,4'-methylenediphenyl diisocyanate. Carcinogenesis 16:573–82.
Sennbro CJ, Lindh CH, Ostin A, et al. (2004) A survey of airborne isocyanate exposure in 13 Swedish polyurethane industries. Ann Occup Hyg 48:405–14.
Sennbro CJ, Littorin M, Tinnerberg H, et al. (2005) Upper reference limits for biomarkers of exposure to aromatic diisocyanates. Int Arch Occup Environ Health 78:541–6.[CrossRef][ISI][Medline]
Streicher RP, Reh CM, Key-Schwartz RJ, et al. (2000) Determination of airborne isocyanate exposure: considerations in method selection. Am Ind Hyg Assoc J 61:544–56.
Tinnerberg H, Skarping G, Dalene M, et al. (1995) Test chamber exposure of humans to 16 hexamethylene diisocyanate and isophorone diisocyanate. Int Arch Occup Environ Health 67:367–74.[CrossRef][ISI][Medline]
Vanoirbeek JA, Tarkowski M, Ceuppens JL, et al. (2004) Respiratory response to toluene diisocyanate depends on prior frequency and concentration of dermal sensitization in mice. Toxicol Sci 80:310–21.
White J. (2006) MDHS 25 revisited; development of MDHS 25/3, the determination of organic isocyanates in air. Ann Occup Hyg 50:15–27.
Williams NR, Jones K, Cocker J. (1999) Biological monitoring to assess exposure from use of isocyanates in motor vehicle repair. Occup Environ Med 56:598–601.[Abstract]
Woskie SR, Sparer J, Gore RJ, et al. (2004) Determinants of isocyanate exposures in auto body repair and refinishing shops. Ann Occup Hyg 48:393–403.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  H. Tinnerberg and C. Mattsson Usage of Air Monitoring and Biomarkers of Isocyanate Exposure to Assess the Effect of a Control Intervention Ann. Hyg., April 1, 2008; 52(3): 187 - 194. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Bello, C. A. Redlich, M. H. Stowe, J. Sparer, S. R. Woskie, R. P. Streicher, H. D. Hosgood, and Y. Liu Skin Exposure to Aliphatic Polyisocyanates in the Auto Body Repair and Refinishing Industry: II. A Quantitative Assessment Ann. Hyg., March 1, 2008; 52(2): 117 - 124. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. White Reply Ann. Hyg., October 1, 2007; 51(7): 648 - 649. [Full Text] [PDF]
|
|
|
|
|
|
G. W. Hughson, K. S. Creely, J. Cocker, and K. Jones Reply Ann. Hyg., October 1, 2007; 51(7): 646 - 647. [Full Text] [PDF]
|
|
|
|
|
|
M. J. Walker Isocyanate Sampling and Analysis Ann. Hyg., October 1, 2007; 51(7): 645 - 646. [Full Text] [PDF]
|
|
|
|
 |
|
 J. Cocker Biological monitoring for isocyanates Occup. Med., September 1, 2007; 57(6): 391 - 393. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||