Dermal Exposure to Polycyclic Aromatic Hydrocarbons among Road Pavers

Transcription

1 Ann. occup. Hyg., Vol. 49, No. 2, pp , 2005 # 2004 British Occupational Hygiene Society Published by Oxford University Press doi: /annhyg/meh094 Dermal Exposure to Polycyclic Aromatic Hydrocarbons among Road Pavers VIRPI VÄÄNÄNEN 1 *, MERVI HÄMEILÄ 1, PENTTI KALLIOKOSKI 2, ELINA NYKYRI 1 and PIRJO HEIKKILÄ 1 1 Finnish Institute of Occupational Health, Topeliuksenkatu 41 aa, FIN Helsinki, Finland; 2 University of Kuopio, Department of Environmental Science, PL 1627, FIN Kuopio, Finland Received 5 January 2004; in final form 9 August 2004 Objectives: Dermal exposure to polycyclic aromatic hydrocarbons (PAHs) and the role of an industrial by-product, coal fly ash, on workers PAH exposure were investigated during stone mastic asphalt (SMA) paving and remixing. Methods: PAH exposure was measured at eight sites during the laying of SMA containing coal fly ash or limestone (conventional SMA) as the filler. Six of the surveys were carried out during SMA paving and two during remixing of SMA (hot recycling at the paving site). Dermal PAH exposure was measured by hand washing (using sunflower oil and wiping with Kleenex tissues) before and after the work shift, and by placing exposure pads on the workers wrists during the work shift. The analyses included 15 native PAHs from the hand-washing samples determined using high-performance liquid chromatography equipped with a two-channel fluorescence detector and 16 native PAHs and four methylated PAHs from the exposure pads using gas chromatography with mass-selective detection. Results: The PAH results obtained using the pad and hand-washing methods (concentrations after the work shift) were equivalent and showed a strong correlation (r = 0.757, P < 0.001, N = 23 for total PAHs). There was a statistically significant difference between pre- and post-shift samples as measured by hand washing. The skin contamination by PAHs was significantly higher (P < 0.01) during remixing than during SMA paving. The variation in PAH contamination on the skin explained more of the variation in the excretion of urinary 1-hydroxypyrene and phenanthrols than the variation in the respiratory PAH concentrations. Conclusions: The industrial by-product investigated in asphalt, coal fly ash, had no statistically significant effect on the workers dermal PAH exposure. The dermal exposure of paving workers to PAHs was higher during remixing than during SMA paving. Keywords: coal fly ash; dermal PAH exposure; paving; polycyclic aromatic hydrocarbons; remixing; skin contamination INTRODUCTION Bitumen, which is used mainly in road paving and roofing, is a by-product of oil refining. It is a complex mixture of high molecular weight organic compounds including polycyclic aromatic hydrocarbons (PAHs). Asphalt is a mixture of bitumen with stone, sand, or filler. Stone mastic asphalt (SMA) contains about 6 w-% bitumen, 10 w-% filler (which in the current study was either coal fly ash or limestone) and 0.4 w-% fibre; the rest is composed of stones of various sizes. Coal fly ash is an industrial by-product *Author to whom correspondence should be addressed at the Department of Industrial Hygiene and Toxicology. Tel: ; fax: ; virpi.vaananen@ttl.fi of the combustion of coal and it may contain PAH compounds. The concentration of PAHs in coal fly ash depends on the origin of the coal, the burning temperature of the coal in the power plant and the combustion technique utilized in the power plant (Li et al., 1983; Gohda et al., 1993; de Raat et al., 1994; Borm, 1997). In remixing, the old layer of asphalt is removed by hot stripping and mixed with new asphalt at the paving site; it is then repaved onto the road. The amount of new asphalt is 20 25% in a repaving mixture. During SMA paving, the temperature of asphalt is raised to C; whereas the temperature of heated old asphalt may be as high as 250 C. It is known that the higher the temperature of the asphalt, the higher the proportion of 4 6-ring PAHs present in the fumes (Concawe, 1992). 167

2 168 V. Väänänen et al. Our earlier studies showed that the addition of an industrial by-product, coal fly ash, as the filler into SMA did not affect airborne PAH exposure nor the level of PAH metabolites in the urine of paving workers (Väänänen et al., 2003). Repaving operations have been reported to increase respiratory PAH exposure in paving crews (Burstyn et al., 2000). According to our earlier studies, repaving workers had a higher level of PAH metabolites in their urine than paving workers (Väänänen et al., 2003). The mutagenic potency of asphalt fumes collected during paving was higher than that of the laboratorygenerated fumes in Salmonella typhimurium tests without S9, and the remixing fumes were more mutagenic than the SMA paving fumes and laboratorygenerated fumes with S9 (Heikkilä et al., 2003). Many 2 3-ring PAHs can cause irritation, and some of the 4 6-ring PAHs are carcinogenic (IARC, 1973). The skin can be a target organ for carcinogens, but dermal application of coal tar, creosote and bitumen may also lead to the formation of DNA adducts in the lungs (Schoket et al., 1988a,b). Genevois et al. (1996) detected DNA adducts in skin, lung and lymphocytes after rats were painted with undiluted bitumen fume condensates. Laboratory-generated bitumen roofing fumes (Sivak et al., 1997; Niemeyer et al., 1988), raw roofing bitumen (Sivak et al., 1997) and bitumen paints (Robinson et al., 1984) have been shown to be carcinogenic in animal studies when applied dermally to mice. In a study by Emmet et al. (1981), however, raw roofing bitumen was not carcinogenic after dermal treatments. In the recent IARC (International Agency for Research on Cancer) European epidemiological study of cancer mortality among asphalt workers, the standardized mortality ratio (SMR) of lung cancer was slightly higher in those workers employed in jobs entailing exposure to bitumen (1.17, 95% CI ) than in construction workers (1.01, 95% CI ) (Boffetta et al., 2003a). Analysis restricted to road pavers and based on quantitative estimates of bitumen fume exposure suggests an association between lung cancer mortality and the average level of exposure to bitumen fumes (SMR = 1.23, 95% CI ) but not the duration of exposure or cumulative exposure (Boffetta et al., 2003b). Occupational exposure assessment has been based mainly on the air concentrations of the chemicals. However, some chemicals also have a skin notation in the occupational exposure limit (OEL) values, indicating that these chemicals are able to enter the body through the skin. Nevertheless, there are no standardized techniques for assessing skin exposure. Fenske (1993) has classified the methods of assessing dermal exposure into three groups: surrogate skin techniques, removal techniques and fluorescent tracer techniques. Surrogate skin techniques include patches (pads) and clothing which act as passive samplers (Soutar et al., 2000). Hand washing and skin wiping, which are common techniques used for dermal exposure sampling, belong to the removal techniques (Brouwer et al., 2000). In addition to these techniques, different sampling materials and methods of analysis complicate the comparison of the results of published studies. Occupational dermal PAH exposure has been most commonly measured using skin wipes (Wolff et al., 1989; Hicks, 1995; Kuljukka et al., 1997), hand washing (Zhou, 1997; Jongeneelen et al., 1988) and exposure pads (Jongeneelen et al., 1988; Van Rooij et al., 1992; Van Rooij, 1994). There are not many studies on the skin contamination of road paving workers. In road paving with petroleum asphalt, the geometric mean (GM) concentration in post-shift wiping samples has been measured as 1.4 ng/cm 2 for pyrene, which has often been used as a marker compound (Zhou, 1997). The skin contamination of workers appeared to increase 10-fold when the asphalt contained coal tar (Jongeneelen et al., 1988). In some work environments where PAH exposure has been high, skin contamination by PAHs has demonstrated a stronger correlation with urinary 1-hydroxypyrene (1-OHP) excretion than with the PAH concentration in the breathing zone air (Van Rooij et al., 1992, 1993a,c). The aim of this study was to investigate whether the use of recycled industrial by-products such as coal fly ash instead of limestone in SMA increases the exposure of paving workers to PAHs and modifies the genotoxicity of fumes. The exposure was studied by measuring the concentrations of airborne impurities, urinary PAH metabolites and dermal PAH exposure. The results for air and urinary PAH concentrations and the mutagenicity of fumes have been published earlier (Heikkilä et al., 2003; Väänänen et al., 2003). In this article, we present the dermal PAH exposure data. MATERIALS AND METHODS Subjects and study design The dermal PAH exposure of road pavers was studied at eight paving sites in Finland during the paving seasons in 1999 and In the Nordic countries, the paving season lasts from April to October. At six paving sites, SMA containing coal fly ash (SMA coal fly ash) or limestone (SMA lime) as the filler was used, and remixing (REM) of SMA containing coal fly ash or limestone was carried out at two sites. The SMA mixtures were laid at temperatures of C, but during remixing, the old asphalt layer was first heated to C, then scraped and mixed in situ with virgin asphalt (190 C). The heaters were warmed with liquefied gas.

3 Dermal PAH exposure among pavers 169 Paving teams with four to nine members (i.e. paver operator, screedman, shovellers/rakermen, adhesive sprayer, heater operators and traffic controllers) participated in the study. The paver operator sits on top of and controls the paving machine. The screedman operates the paving screed to give the desired dimensions, and stands in the back of the paving machine. The shovellers and rakermen correct manually spread asphalt with hand tools such as rakes and shovels. The adhesive sprayer applies bitumen emulsion onto the road before the spreading of new asphalt. Heater operators control the heaters, which are used at the remixing paving sites. Traffic controllers guide the passing traffic and are thus farther away from the paving site. The number of individual workers was 21; some of the pavers worked at several paving sites and thus gave more than one sample. The numbers of participating paving workers and samples collected are presented in Table 1. The paving machines were not equipped with a cabin or a ventilation system. The asphalt workers did not wear respiratory masks or use any barrier creams, but most of them wore gloves. The workers could not wash their hands or take a shower at the paving sites. SMA coal fly ash was rarely laid, and the paving companies had already planned their paving schedules before the paving season; we were therefore not able to randomize the selection of the sampling days or paving teams. Dermal exposure sampling and analysis Exposure pad method. At four SMA paving sites and two remixing paving sites, the workers wore exposure pads made of polypropylene (Millipore, AN1H4700, pore size 10 mm) on both wrists. The polypropylene prefilters were cut to fit into a plastic case (disk) (outer diameter 38 mm, inner diameter 29 mm) with a wristband. The exposure pad was worn like a watch. The effective sampling area of one pad was 6.6 cm 2 (13.2 cm 2 for two pads). Sampling time covered the whole work shift. Both pads from a worker were combined, and extracted ultrasonically with 10 ml cyclohexane (pro analysi, Merck)/dichloromethane (SupraSolv, Merck) mixture (4/1 w/w) for 60 min. Before the extraction, 20 ml of a mixture containing six deuterated PAH compounds with a concentration of 100 pg/ml [naphthalene-d8, biphenyl-d10, phenanthrene-d10, pyrene-d10, benzo(a)anthracene-d12, benzo(a) pyrene-d12, benzo(ghi)perylene-d12 (Chiron, Trondheim, Norway)] was added as an internal standard. The extract was evaporated to a volume of 100 ml under a gentle flow of nitrogen in a TurboVap Ò LV Evaporator (Zymark, PLD Finland Oy, Hopkinto, MA, USA). After evaporation, 4 ml of anthracened10, with a concentration of 1.2 ng/ml, was added as an instrument peak. The samples were analysed using high-resolution gas chromatography with massselective detection [HRGC MS system, AutoSpec-Q (Micromass, Altringham, UK)] using the high resolution selective ion monitoring (HR-SIM) technique. The molecular ions monitored are listed in Table 2. The ionization technique used was electron ionization (EI, 70 ev). The temperature of the ion source was set to 230 C and to 300 C in the transfer line. A 30 m 0.32 mm 0.25 mm capillary GC column (cross-linked 5% phenyl methylsilicone, HP-5) was Table 1. Number (N) of participating paving workers and traffic controllers, and the sampling days for each paving site Asphalt mixture, paving site, paving group SMACFA Air sampling Hand washing Exposure pads Urine sampling Sampling day Paving site a, Group 1 a 3 (1) b 4 (1) b 3 (1) b 4 (1) b Monday Paving site b, Group c 2 Wednesday Paving site c, Group 1 a Thursday SMAL Paving site a, Group 1 a 3 (1) b 4 (1) b 3 (1) b 4 (1) b Tuesday Paving site d, Group 3 4 (1) b 4 (1) b c 4 (1) b Wednesday Paving site e, Group 1 a 4 (1) b 4 (1) b 4 (1) b 4 (1) b Tuesday REMSMACFA, Group 4 d Thursday REMSMAL, Group 4 d Tuesday Total number of: Pavers e Traffic controllers SMACFA = stone mastic asphalt containing coal fly ash; SMAL = stone mastic asphalt containing limestone; REMSMACFA = remixing of SMACFA; REMSMAL = remixing of SMAL. a The paving operator and the screedman were the same person in Group 1, but the shoveller and rakerman were different people. b Number of traffic controllers in parentheses. c Samples were not collected. d All the paving workers were the same in Group 4. e The number of individual pavers was 21.

4 170 V. Väänänen et al. Table 2. Used molecular ions of PAH compounds analysed by HRGC/MS and recovery of the deuterated PAH compounds PAH Ion No. aromatic Note rings Naphthalene Analyte Recovery in samples a % SD (N = 26) Naphthalene-d IS Acenaphthylene Analyte Acenaphthene Analyte Biphenyl-d IS Fluorene Analyte Phenanthrene Analyte Anthracene Analyte Phenanthrene-d IS Methylphenanthrene Analyte 2,10/4,10-Dimethyl-phenanthrene Analyte 1,4-Dimethyl-phenanthrene Analyte Fluoranthene Analyte Pyrene Analyte Pyrene-d IS Benzo(a)anthracene Analyte Chrysene Analyte Benzo(a)anthracene-d IS Methylchrysene Analyte Benzo(b+k)fluoranthene Analyte Benzo(a)pyrene Analyte Benzo(a)pyrene-d IS Dibenzo(ah)anthracene Analyte Indeno(123cd)pyrene Analyte Benzo(ghi)perylene Analyte Benzo(ghi)perylene-d IS IS = internal standard; SD = standard deviation. a The samples had gone through sample preparation (evaporation). b The standards had not gone through sample preparation. Recovery in standards b % SD (N = 10) used. The pressure of the carrier gas (helium) was set at 15 psi. The injector temperature was 260 C. The column temperature was programmed as follows: 40 C for 1 min, an increase of 5 C/min up to 270 C, and holding at 270 C for 13 min. The pad analyses were performed for 16 PAH compounds listed by the EPA (Environmental Protection Agency) standard and four methylated PAH compounds (Table 2). Methylated PAHs were 1-methylphenanthrene, 2,10/4,10-dimethylphenanthrene, 1,4-dimethylphenanthrene and 5-methylchrysene. These methylated compounds were included because 1-methylphenanthrene and 1,4- and 4,10- dimethylphenanthrene are mutagenic compounds, and 5-methylchrysene has been found in animal studies to be carcinogenic (IARC, 1983; LaVoie et al., 1983). Quantification was based on the external liquid calibration standards and the correction with the deuterated internal standards and the instrument peak. The recoveries of the internal standards are given in Table 2. The quantitation limits for the PAH compounds were ng/cm 2. Hand-washing method. At all paving sites, the contamination of the workers hands was sampled by wiping the hands with sunflower oil and paper tissue (Jongeneelen et al., 1988). The surface area of the hands was estimated to be 820 cm 2 (Ness, 1994). At the beginning and at the end of the work shift, the workers hands were washed with 3 ml of sunflower oil followed by rubbing the hands together for 1 min. The oil was wiped from the hands with one Kleenex cleaning tissue ( mm, Kimberly Clark), which was stored in a glass vial. The analytes were extracted with 30 ml of dichloromethane (SupraSolv, Merck) in a shaking apparatus for 30 min, followed by 30 min sonication. Then 10 ml of the solvent was evaporated to a volume of 1 2 ml, and acetonitrile (HPLC grade, LabScan) was added to give a volume of 3 ml. After centrifugation, the samples were analysed using HPLC apparatus equipped with a two-channel fluorescence detector (HPLC FLD) following an in-house method (Kuusimäki et al., 2003). The limits of quantification (LOQ) were ng/cm 2. Recoveries for the

5 Dermal PAH exposure among pavers 171 detected PAH compounds varied from 30 to 83%, and were 75% for naphthalene, 83% for phenanthrene and 60% for pyrene. The relative standard deviation (RSD) varied from 5 to 35%, and was 14% for naphthalene, 9% for phenanthrene and 9% for pyrene. Determination of the air samples and the metabolites in the urine samples. Air samples were collected in the breathing zones of the asphalt workers. The PAH compounds were collected on Teflon filters (SKC , 2 mm, SKC Inc., Eighty Four, PA, USA) connected to XAD-2 adsorbent tubes (Orbo TM -43, Supelpak TM, Supelco, Bellefonte, PA, USA). A flow rate of 1 l/min was used and the sampling time was the whole work shift, about 8 h. The air samples were analysed after solvent extraction using the same HPLC FLD method as for the hand-washing samples (Kuusimäki et al., 2003; Väänänen et al., 2003). Urine samples were collected from the paving workers before and after the work shift. Metabolites of naphthalene, phenanthrene and pyrene in urine were analysed. Concentrations of PAH metabolites were adjusted to the creatinine to compensate for variations in the urine flow (Clark and Thomson, 1949). The urinary 1- and 2-naphthols were acidhydrolysed, cleaned on an Oasis HLB column and determined as their pentafluorobenzylbromide derivatives using gas chromatography with a massselective detector (GC MSD) using negative ion chemical ionization. 1-, 2-, 3-, 4- and 9-phenanthrols and 1-OHP were deconjugated from the glucuronide and sulphate conjugates using enzymatic hydrolysis, cleaned on solid phase Bond Elut C18 cartridges and analysed using HPLC FLD (Keimig and Morgan, 1986; Elovaara et al., 2003; Väänänen et al., 2003). Statistical methods The effect of coal fly ash and paving techniques on dermal contamination were studied using a Student s t-test for two independent samples. The differences in the PAH concentrations on the workers hands between pre-shift and post-shift samples were evaluated using a t-test for two paired samples. Pearson s correlation coefficient (r) and linear models were used to examine the strength of the relationship between the exposure pad and hand-washing methods. We examined the association between the urinary PAH metabolites and skin contamination or airborne PAHs using one-variable fixed-effect linear models (Väänänen et al., 2003). When we calculated the Pearson s correlation coefficients and used linear models, logarithmic transformation was performed on all values. In the analyses, results below the detection limit were replaced by values of half of the LOQ. RESULTS When the filler used with the SMA was coal fly ash, the pavers skin contamination with PAHs was slightly lower than when the filler was limestone. The GM of total PAHs measured from the pads was 4.6 ng/cm 2 during laying of SMA coal fly ash and 7.5 ng/cm 2 during laying of SMA limestone. During remixing the GM of total PAHs was 7.8 ng/cm 2 when the SMA contained coal fly ash and 10 ng/cm 2 when the SMA contained limestone. However, we could not find statistically significant differences (P > 0.05) in the dermal exposure of the paving workers between the use of coal fly ash and limestone as the filler in asphalt. The concentrations of PAH compounds on the pads on the workers wrist are presented in Table 3. Within the paving team, the screedman and the paver operator had the most contaminated hands. The GMs of total PAHs on their exposure pads were 13 and 8.9 ng/cm 2, respectively. There was no statistically significant difference in skin contamination between the paving jobs, but when the paver operators, screedmen, shovellers/rakermen and heater operators were compared with the traffic controllers, the difference in PAHs was statistically significant. The amount of pyrene on the exposure pads of screedmen and paver operators was about 40 times higher than the corresponding value on the exposure pads of the traffic controllers. The GMs of pyrene on their exposure pads were 1.45, 1.48 and 0.04 ng/cm 2, respectively. The hand-washing method revealed a statistically significant (P < 0.05) difference between the amounts of the PAHs in the pre- and post-shift samples (Table 4). The skin exposure results for 4 6-ring PAHs classified by job categories are presented in Fig. 1. The results showed that there was a strong linear relationship between the pad and hand-washing methods (concentration after the work shift). Pearson s correlation coefficient (r) was for pyrene (P < 0.001, N = 23), for 4 6-ring PAHs (P < 0.001, N = 23) and for total PAHs (P < 0.001, N = 23). The arithmetic means (AMs) for total PAHs and pyrene for all road pavers were the same using both methods, when the amount was calculated per cm 2. The results are shown in Fig. 2. The dermal exposure of the workers laying or remixing SMA differed significantly. The total amount of native PAHs (arithmetic mean) was 3-fold higher and the amount of 4 6-ring PAHs (arithmetic mean) was 6-fold higher during remixing work than during SMA paving. The highest difference between paving and remixing was observed in the concentrations of pyrene and fluoranthene (Table 4). The concentrations of pyrene and fluoranthene were more than seven times higher on the skin of the remixing workers than on the skin of the SMA workers.

6 172 V. Väänänen et al. Table 3. Concentrations of PAH compounds (ng/cm 2 ) on the pads on the workers wrists, by asphalt mixture and laying technique SMACFA (N = 7) SMAL (N = 7) REMSMACFA (N = 4) REMSMAL (N = 4) AM GM GSD Range AM GM GSD Range AM GM GSD Range AM GM GSD Range Naphthalene < Acenaphthylene a a 1.2 < < Acenaphthene < < Fluorene Phenanthrene Methylphenanthrene ,10/4,10-Dimethyl-phenanthrene ,4-Dimethyl-phenanthrene a a 1.0 <0.08 a a 1.0 <0.08 a a 1.0 <0.08 a a 1.0 <0.08 Anthracene < Fluoranthene Pyrene Benzo(a)anthracene 0.02 a 1.3 < < Chrysene Methylchrysene a a 1.0 <0.01 a a 1.0 <0.01 a a 1.0 <0.01 a a 1.0 <0.01 Benzo(b+k)fluoranthene a a 1.0 < < Benzo(a)pyrene a a 1.0 <0.02 a a 1.0 < < Dibenzo(ah)anthracene a a 1.0 <0.03 a a 1.0 <0.03 a a 1.0 <0.03 a a 1.2 <0.03 Indeno(123cd)pyrene a a 1.0 <0.02 a a 1.1 < < Benzo(ghi)perylene a a 1.1 < a a 1.2 < ring native PAHs ring native PAHs Total native PAHs Methylated PAHs AM = arithmetic mean; GM = geometric mean; GSD = standard deviation of geometric mean; SMACFA = stone mastic asphalt containing coal fly ash; SMAL = stone mastic asphalt containing limestone; REMSMACFA = remixing of SMACFA; REMSMAL = remixing of SMAL; N = number of observations. a All observations were below the detection limit.

7 Dermal PAH exposure among pavers 173 Table 4. Concentrations of PAH compounds on workers hands before and after work shifts (hand-washing method), by asphalt mixture and laying technique [concentration (ng/cm 2 )] N Naphthalene Phenanthrene Pyrene 2 3 aromatic ring PAH 4 6 aromatic ring PAH Total PAH AM GM GSD Range AM GM GSD Range AM GM GSD Range AM GM GSD Range AM GM GSD Range AM GM GSD Range SMACFA Pre-shift < < < Post-shift < SMAL Pre-shift < < < Post-shift < < REMSMACFA Pre-shift 4 a a 1.0 < Post-shift 4 a a 1.0 < REMSMAL Pre-shift 4 a a 1.0 < Post-shift 4 a a 1.0 < All pavers Pre-shift < < < Post-shift < < Traffic controller Pre-shift < < Post-shift AM = arithmetic mean; GM = geometric mean; GSD = standard deviation of geometric mean; SMACFA = stone mastic asphalt containing coal fly ash; SMAL = stone mastic asphalt containing limestone; REMSMACFA = remixing of SMACFA; REMSMAL = remixing of SMAL. The area of the hands has been estimated as 820 cm 2 (Ness, 1994, p. 330). N = number of observations. a All observations were below the detection limit.

8 174 V. Väänänen et al paver operator screedman Max pre-shift sample post-shift sample 4-6 ring PAHs, ng/cm shoveller/rakerman Mean heater operator adhessive sprayer traffic controller 0 Min Fig. 1. Skin exposure results for PAH compounds containing 4 6 aromatic rings, classified by job title. (A) Total PAHs Exposure pads Hand washing (post-shift) REMSMAL 50 REMSMACFA ng/cm SMACFA SMAL Asphalt workers (B) Pyrene Exposure pads Hand washing (post-shift) REMSMAL ng/cm SMACFA SMAL REMSMACFA Asphalt workers Fig. 2. Results (ng/cm 2 ) for (A) total PAH compounds and (B) pyrene on workers skin measured using exposure pads (HRGC/MS) and after work shifts by the hand-washing method (HPLC/FLD).

9 Dermal PAH exposure among pavers 175 The main PAH compounds were 2,10/4,10- dimethylphenanthrene, phenanthrene, 1-methylphenanthrene and naphthalene on the workers skin in SMA paving, whereas in remixing, 2,10/4,10- dimethylphenanthrene, pyrene, fluoranthene and phenanthrene were the main PAHs. In SMA paving, the proportion of quantified alkylated phenanthrenes was 52 w-% of the measured compounds including native PAHs and methylated PAHs. The proportion was 7 w-% for PAHs containing 4 6 aromatic rings and 41 w-% for PAHs containing 2 3 aromatic rings. In the remixing of SMA, the proportions of quantified alkylated phenanthrenes, PAHs containing 4 6 aromatic rings and PAHs containing 2 3 aromatic rings were 34, 22 and 44 w-%, respectively. The concentrations of the PAH compounds on the workers skin found using both sampling methods are presented in Tables 3 and 4, according to the laying technique and the asphalt mixture. Dermal naphthalene exposure was lower than phenanthrene exposure (Tables 3 and 4), and it did not correlate statistically significantly with the concentrations of urinary naphthols or with airborne naphthalene. The skin contamination by phenanthrene showed a strong linear relationship with the urinary phenanthrols. Pearson s correlation coefficients between the sum of the urinary phenanthrols and phenanthrene on the skin were (P = , N = 23 for the pad method) and (P = , N = 22 for the handwashing method). The airborne phenanthrene showed a strong linear relationship with phenanthrene on the skin (r = 0.733, P < , N = 25 for the pad method). The dermal pyrene results for both sampling methods correlated statistically significantly with the post-shift 1-OHP urine concentrations (r = 0.689, P = , N = 23 for the pads and r = 0.618, P = , N = 22 for hand washing). Also, the measured amount of pyrene on the skin of the road pavers showed a strong linear relationship with the airborne pyrene concentrations (r = 0.721, P < , N = 25 for the pad method). DISCUSSION The use of industrial by-products in asphalt is increasing. However, there are only a few studies on the occupational exposure of workers who are involved in recycled asphalt or in asphalt modified with industrial waste materials (Watts et al., 1998; Burstyn et al., 2000; Heikkilä et al., 2002, 2003). In our study, the addition of coal fly ash as the filler in asphalt did not increase the level of dermal PAH contamination. Our earlier studies also showed that the addition of an industrial by-product, coal fly ash, as the filler into the SMA did not affect the airborne PAH exposure nor the level of PAH metabolites in the urine of the paving workers (Väänänen et al., 2003). The dermal exposure of road pavers has been studied in only a few studies (Hicks, 1995; Zhou, 1997; Jongeneelen et al., 1988). In two studies, the paving asphalt was petroleum based and in one study the asphalt contained coal tar. In our study and in Zhou s study, where the asphalt was petroleum based, the measured skin contamination with pyrene was at the same level (Zhou, 1997). In our study, the GM of total PAHs was higher (6.8 ng/cm 2 determined by exposure pads, and 7.8 ng/cm 2 determined by hand washing) than that measured in Zhou s study (2.2 ng/cm 2 in the post-shift hand-washing samples), probably because Zhou measured only nine PAH compounds. In another study on petroleum asphalt, the skin contamination was measured by wiping the back of the hand or the forehead with a premoistened Whatman smear tab (Hicks, 1995). The limits of detection for separate PAH compounds were high in this study, and the amount of PAHs exceeded the limits in only a few samples. Our results confirm earlier published data that the dermal PAH contamination of road pavers using petroleum-based asphalt is about one-tenth that reported in workers exposed to bitumen containing coal tar or coal tar pitch (Jongeneelen et al., 1988; Wolff et al., 1989). The results of dermal exposure studies are not readily comparable because of different sampling techniques. We used two sampling and two analytical methods to measure the PAHs on the skin; one was polypropylene pads together with HRGC MS, and the other hand washing together with HPLC FLD. We used the methods to compare two generally used sampling techniques, to get information of the level of alkylated PAHs on skin and to confirm the concentration of native PAHs by two methods. The PAH results analysed using HPLC FLD from bitumen matrix have been criticized because of the poor resolution of 4 6-ring PAHs (Watts et al., 1998; Butler et al., 2000). The baseline in HPLC chromatograms is elevated owing to the high concentration of alkylated PAH compounds in the matrix. According to our unpublished results, both methods, HRGC MS and HPLC FLD, gave equal native PAH results. Exposure pads have several advantages, including their ease of application, small size, low cost and passive sampling, but they also have some drawbacks; for example, the pad material differs from natural skin and thus PAH adsorption into the pad may be different from PAH adsorption into the skin. The area of the exposure pad is small, which makes it difficult to assess the total dermal exposure. Also, the location of the pads is problematic: should they be worn outside or under the clothing (Ness, 1994)? On the other hand, hand washing and wiping methods may easily underestimate the level of skin absorption

10 176 V. Väänänen et al. because they measure the residue on the skin and not all the contamination is recovered during washing or wiping. In our study, the exposure pads measured the deposition of bitumen fumes, and to a lesser extent direct contact with contaminated surfaces or bitumen. The washing method measured the deposition of bitumen fumes as well as direct contact with bitumen, for example, via the work equipment. Sample preparation was more convenient with the exposure pads than with the hand-washing samples due to the sunflower oil. Nevertheless, the results of these two methods were equal (Fig. 2). In our study, dermal exposure to PAHs was higher in workers who were remixing SMA than in SMA paving workers, even though the breathing zone PAH concentrations did not differ statistically significantly between the remixing and SMA paving groups (Väänänen et al., 2003). One obvious reason for the higher skin contamination of the remixing workers was that only two out of eight workers wore gloves, whereas more than half of the SMA pavers used gloves and long-sleeved coats or shirts during their work shifts. Personal factors, such as the use of protective clothing, individual working methods, the frequency of changing work clothes and laundering, and personal hygiene can influence the skin contamination. With simple hygienic operations, skin contamination can be reduced considerably (Van Rooij et al., 1994; Lafontaine et al., 2002). However, some of these hygienic operations, such as hand washing, are not possible for workers at paving sites. A second reason for the difference may be the fact that remixing workers used light fuel oil for cleaning their equipment more carelessly than did SMA workers, who also used only vegetable oil for cleaning in two paving sites. Fuel oil contains PAHs (IARC, 1989), and Moen et al. (1996) have reported that workers who had oil contamination on their skin during their work in an engine room had increased concentrations of 1-OHP in their urine. In our study, the use of light fuel oil increased dermal pyrene and urinary 1-OHP concentrations, but not the concentration of pyrene in the breathing zone of the workers (P < 0.05). A third reason may be that according to our data, the airborne pyrene appeared more in the particle phase than in the vapour phase during remixing than during SMA paving, although the total amounts of airborne pyrene (both vapour and particle phase) were equal. Bitumen fumes contain high levels of phenanthrene and naphthalene and many of their alkylated derivatives, and these compounds are the main PAH groups in the fumes (Binet et al., 2002; Heikkilä et al., 2003). In our study, the amounts of 1-methylphenanthrene and 2,10/4,10-dimethylphenanthrene on the skin were high compared with the amount with native PAHs. In addition to these analysed alkylated PAHs, pavers skin is certainly contaminated with other alkylated PAHs and thiophenes. 1-Methylphenanthrene, 1,4- and 4,10-dimethylphenanthrenes are mutagenic to Salmonella typhimurium in the presence of metabolic activation, and 1,4- dimethylphenanthrene is also active as a tumour initiator (IARC, 1983; LaVoie et al., 1983). Binet et al. (2002) have suggested that some thiophenes are responsible for the genotoxicity of bitumen fumes. The amounts of methylated and sulphurcontaining PAHs in bitumen fumes are so high that the concentration of native PAHs alone may not be an appropriate indicator of exposure to carcinogenic PAHs. There are still no accepted standards or limit values for evaluating dermal exposure. However, many chemicals carry a skin notation in the OEL lists indicating that skin absorption is a possible route of exposure. It is known that bitumen fume condensates penetrate the skin rapidly owing to a reduction in viscosity attributable to the mixing of the vapour and particle phases (Genevois et al., 1996; Binet et al., 2002). The German committee on MAK (maximum workplace concentration) has assigned a skin notation to bitumen fumes because it has been shown in animal studies that carcinogenic compounds, present in bitumen fumes, are able to permeate the skin (Deutsche Forschungsgemeinschaft, 2001). Paradoxically, PAH compounds have no skin notation, although many studies have concluded that skin absorption of PAHs can be considerable (Van Rooij et al., 1992, 1993b,c, 1994; Elovaara et al., 1995). Estimates for the absorption of pyrene through the skin vary from 23 to 75% of the total dose (Van Rooij et al., 1993a; Brzeznicki et al., 1997; Lafontaine et al., 2002). In our study, skin contamination with PAHs showed a strong correlation with urinary PAH metabolites. The airborne pyrene correlated moderately with the urinary 1-OHP (r = 0.403, P = 0.027, N = 30) and the airborne phenanthrene showed a moderate correlation with the sum of urinary phenanthrols (r = 0.441, P = 0.015, N = 30). The differences in Pearson s correlation coefficient between our earlier published article (Väänänen et al., 2003) and this article resulted from the logarithmic transformation and the different number of samples: two samples have not been included in the statistical analysis because they were incomplete. According to the one-variable fixed-effect linear model, the variation in dermal pyrene contamination explained more of the variation in 1-OHP excretion (R 2 = 47% for the exposure pad method, R 2 = 38% for the hand-washing method) than did the airborne pyrene concentration (R 2 = 16%). Also, phenanthrene on the skin (R 2 = 27% for the exposure pad method and R 2 = 53% for the hand-washing method) had a greater effect on the variation in urinary phenanthrols than airborne phenanthrene (R 2 = 19%). Our conclusion is that skin contamination with PAHs increases

11 Dermal PAH exposure among pavers 177 significantly the total body burden of PAHs among paving workers. Acknowledgements Antti Hesso and Jarkko Tornaeus are thanked for performing the HRGC MS analysis. The authors gratefully acknowledge the assistance of Erkki Nykyri and Ritva Luukkonen in the statistical calculations, and the cooperation of Kimmo Peltonen, Asko Saarela and Petri Peltonen. This study was conducted in association with the project of the Finnish Research Programme on Environmental Health (SYTTY). The Academy of Finland, the Finnish Work Environment Fund, the Emil Aaltonen Foundation, the Finnish National Road Administration and the participating companies are thanked for financial support. REFERENCES Binet S, Pfohl-Leszkowicz A, Brandt H et al. (2002) Bitumen fumes: review of work on the potential risk to workers and the present knowledge on its origin. Sci Total Environ; 300: Boffetta P, Burstyn I, Partanen T et al. (2003a) Cancer mortality among European asphalt workers: an international epidemiological study. I. Results of the analysis based on job titles. Am J Ind Med; 43: Boffetta P, Burstyn I, Partanen T et al. (2003b) Cancer mortality among European asphalt workers: an international epidemiological study. II. Exposure to bitumen fume and other agents. Am J Ind Med; 43: Borm PJA. (1997) Toxicity and occupational health hazards of coal fly ash (CFA). A review of data and comparison to coal mine dust. Ann Occup Hyg; 41: Brouwer DH, Boeniger MF, van Hemmen J. (2000) Hand wash and manual skin wipes. Ann Occup Hyg; 44: Brzeznicki S, Jakubowski M, Czerski B. (1997) Elimination of 1-hydroxypyrene after human volunteer exposure to polycyclic aromatic hydrocarbons. Int Arch Occup Environ Health; 70: Burstyn I, Kromhout H, Kauppinen T et al. (2000) Statistical modelling of the determinants of historical exposure to bitumen and polycyclic aromatic hydrocarbons among paving workers. Ann Occup Hyg; 44: Butler MA, Burr G, Dankovic D et al. (2000) Hazard review: health effects of occupational exposure to asphalt. Cincinnati, OH, USA: National Institute for Occupational Safety and Health (NIOSH). Clark LCJ, Thomson HL. (1949) Determination of creatine and creatinine in urea. Anal Chem; 21: Concawe. (1992) Bitumens and bitumen derivates. Brussels: Concawe. de Raat WK, Boers JP, Bakker GL et al. (1994) Contribution of PAH and some of their nitrated derivatives to the mutagenicity of ambient airborne particles and coal fly ash. Sci Total Environ; 153: Deutsche Forschungsgemeinschaft. (2001) List of MAK and BAT values Commission for the Investigation of Health Hazards of Chemical Compounds in the Work Area. Report 37. Weinheim, Germany: Wiley-VCH. Elovaara E, Heikkilä P, Pyy L et al. (1995) Significance of dermal and respiratory uptake in creosote workers: exposure to polycyclic aromatic hydrocarbons and urinary excretion of 1-hydroxypyrene. Occup Environ Med; 52: Elovaara E, Väänänen V, Mikkola J. (2003) Simultaneous analysis of naphthols, phenanthrols and 1-hydroxypyrene as biomarkers of PAH exposure: intraindividual variance in the urinary metabolite excretion profiles caused by intervention with beta-naphtoflavone induction in the rat. Arch Toxicol; 77: Emmett EA, Bingham EM, Barkley W. (1981) A carcinogenic bioassay of certain roofing materials. Am J Ind Med; 2: Fenske RA. (1993) Dermal exposure assessment techniques. Ann Occup Hyg; 37: Genevois C, Brandt H, Bartsch H et al. (1996) Formation of DNA adducts in skin, lung and lymphocytes after skin painting of rats with undiluted bitumen or coal-tar fume condensates. Polycyclic Aromatic Compounds; 8: Gohda H, Hatano H, Hanai T et al. (1993) GC and GC MS analysis of polychlorinated dioxins, dibenzofurans and aromatic hydrocarbons in fly ash from coal-burning works. Chemosphere; 27: Heikkilä P, Riala R, Hämeilä M et al. (2002) Occupational exposure to bitumen during road paving. AIHA J; 63: Heikkilä P, Väänänen V, Hämeilä M et al. (2003) Mutagenicity of bitumen and asphalt fumes. Toxicol In Vitro; 17: Hicks JB. (1995) Asphalt industry cross-sectional exposure assessment study. Appl Occup Environ Hyg; 10: IARC. (1973) Certain polycyclic aromatic hydrocarbons and heterocyclic compounds. IARC monographs on the evaluation of carcinogenic risks to humans and their supplements. Vol. 3; Lyon: International Agency for Research on Cancer. IARC. (1983) Polynuclear aromatic compounds, part 1. Chemical, environmental and experimental data. IARC monographs on the evaluation of the carcinogenic risk of chemicals to humans. Vol. 32. Lyon: International Agency for Research on Cancer. IARC. (1989) Occupational exposures in petroleum refining; crude oil and major petroleum fuels. IARC monographs on the evaluation of carcinogenic risks to humans. Vol. 45. Lyon: International Agency for Research on Cancer. Jongeneelen FJ, Scheepers PT, Groenendijk A et al. (1988) Airborne concentrations, skin contamination, and urinary metabolite excretion of polycyclic aromatic hydrocarbons among paving workers exposed to coal tar derived road tars. Am Ind Hyg Assoc J; 49: Keimig S, Morgan D. (1986) Urinary 1-naphthol as a biological indicator of naphthalene exposure. Appl Ind Hyg; 1: Kuljukka T, Vaaranrinta R, Mutanen P et al. (1997) Assessment of occupational exposure to PAHs in an Estonian coke oven plant correlation of total external exposure to internal dose measured as 1-hydroxypyrene concentration. Biomarkers; 2: Kuusimäki L, Peltonen K, Mutanen P et al. (2003) Analysis of particle and vapour phase PAHS from the personal air samples of bus garage workers exposed to diesel exhaust. Ann Occup Hyg; 47: Lafontaine M, Gendre C, Morele Y et al. (2002) Excretion of urinary 1-hydroxypyrene in relation to the penetration routes of polycyclic aromatic hydrocarbons. Polycyclic Aromatic Compounds; 22: LaVoie EJ, Tulley-Freiler L, Bedenko V et al. (1983) Mutagenicity of substituted phenanthrenes in Salmonella typhimurium. Mutat Res; 116: Li AP, Clark RL, Hanson RL et al. (1983) Comparative mutagenicity of a coal combustion fly ash extract in Salmonella typhimurium and Chinese hamster ovary cells. Environ Mutagen; 5: Moen BE, Nilsson R, Nordlinder R et al. (1996) Assessment of exposure to polycyclic aromatic hydrocarbons in engine rooms by measurement of urinary 1-hydroxypyrene. Occup Environ Med; 53: Ness S. (1994) Surface and dermal monitoring for toxic exposures. New York: Van Nostrand Reinhold. Niemeyer RW, Thayer PS, Menziees KT et al. (1988) A comparison of the skin carcinogenicity of condensed roofing asphalt and coal tar pitch fumes. In: Tenth International Symposium on Polynuclear Aromatic Hydrocarbons, Columbus, OH, USA: Batelle Press. pp

12 178 V. Väänänen et al. Robinson M, Bull RJ, Munch J et al. (1984) Comparative carcinogenic and mutagenic activity of coal tar and petroleum asphalt paints used in potable water supply systems. J Appl Toxicol; 4: Schoket B, Hewer A, Grover PL et al. (1988a) Covalent binding of components of coal-tar, creosote and bitumen to the DNA of the skin and lungs of mice following topical application. Carcinogenesis; 9: Schoket B, Hewer A, Grover PL et al. (1988b) Formation of DNA adducts in human skin maintained in short-term organ culture and treated with coal-tar, creosote or bitumen. Int J Cancer; 42: Sivak A, Niemeier R, Lynch D et al. (1997) Skin carcinogenicity of condensed asphalt roofing fumes and their fractions following dermal application to mice. Cancer Lett; 117: Soutar A, Semple S, Aitken RJ et al. (2000) Use of patches and whole body sampling for the assessment of dermal exposure. Ann Occup Hyg; 44: Väänänen V, Hämeilä M, Kontsas H et al. (2003) Air concentrations and urinary metabolites of polycyclic aromatic hydrocarbons among paving and remixing workers. J Environ Monit; 5: Van Rooij JGM. (1994) Dermal exposure to polycyclic aromatic hydrocarbons among workers. Doctoral thesis. Nijmegen: Katholieke Universiteit Nijmegen. Van Rooij JG, Bodelier-Bade MM, De Looff AJ et al. (1992) Dermal exposure to polycyclic aromatic hydrocarbons among primary aluminium workers. Med Lav; 83: Van Rooij JG, Bodelier-Bade MM, Jongeneelen FJ. (1993a) Estimation of individual dermal and respiratory uptake of polycyclic aromatic hydrocarbons in 12 coke oven workers. Br J Ind Med; 50: Van Rooij JG, De Roos JH, Bodelier-Bade MM et al. (1993b) Absorption of polycyclic aromatic hydrocarbons through human skin: differences between anatomical sites and individuals. J Toxicol Environ Health; 38: Van Rooij JG, Van Lieshout EM, Bodelier-Bade MM et al. (1993c) Effect of the reduction of skin contamination on the internal dose of creosote workers exposed to polycyclic aromatic hydrocarbons. Scand J Work Environ Health; 19: Van Rooij JG, Bodelier-Bade MM, Hopmans PM et al. (1994) Reduction of urinary 1-hydroxypyrene excretion in cokeoven workers exposed to polycyclic aromatic hydrocarbons due to improved hygienic skin protective measures. Ann Occup Hyg; 38: Watts RR, Wallingford KM, Williams RW et al. (1998) Airborne exposures to PAH and PM2.5 particles for road paving workers applying conventional asphalt and crumb rubber modified asphalt. J Expo Anal Environ Epidemiol; 8: Wolff MS, Herbert R, Marcus M et al. (1989) Polycyclic aromatic hydrocarbon (PAH) residues on skin in relation to air levels among roofers. Arch Environ Health; 44: Zhou Q. (1997) Biomonitoring workers exposed to polycyclic aromatic hydrocarbons in asphalt during road paving. Doctoral thesis. Cincinnati, OH, USA: University of Cincinnati.