Document RaL2VJbDjB7B0K43Mz0Q2BqZX

' , > ir a _ \ --\ \.. ' pp ms. :im; 2'u i: Hr: i! -I> < Hr upatn mu I H\ uonc S<w :c(\ t'lil'ii'I'A'J iKlnul l 'imorsti> Pros- 1)01 !D Id'i'/aiinh'. ii/mc`ji)37 ! '' - Analysis of Particle and Vapour Phase PAHs from the Personal Air Samples of Bus Garage Workers Exposed to Diesel Exhaust k PLAINTIFF'S 2 EXHIBIT LEEA KUUSIMAKI*, KIMMO PELTONEN, PERTTI MUTANEN and I g KIRSTI SAVELA 1 f%0?/ Finnish Institute of Occupational Health, Topeliuksenkatu 41 aA, FIN-00250 Helsinki, Finland Received 9 September 2002; in final form 3 February 2003 The levels of particle and vapour phase polycyclic aromatic hydrocarbons (PAHs) derived from the diesel exhaust compounds in bus garage work were measured in winter and in summer. Five personal air samples were collected from the breathing zones of 22 garage workers every other day of consecutive weeks. Control samples (n = 22) were collected from office workers in Helsinki. Fifteen PAHs in the air samples were analysed by HPLC using a fluorescence detector. Statistically significant differences were observed between total PAH levels of the exposed workers (2241 and 1245 ng/mJ) and the control group (254 and 275 ng/m-') in both winter (P < 0.001) and summer (P < 0.001). Phenanthrene, pyrene, benzo[^/']perylene and fluoranthene were the major compounds in the particle phase, and naphthalene, phenanthrene and fluorene in the vapour phase. About 98% of PAHs measured were related to the vapour phase compounds, whereas the high molecular weight PAH compounds were detected only in the particle phase. The PAH levels in the garages were twice as high (P < 0.001) in winter as in summer. Even though the exposure levels were low in the bus garages, the low level does not allow conclusions to be drawn about the possible adverse health effects due to exposure to diesel exhaust. Keywords: air sampling; diesel exhaust; occupational exposure; polycyclic aromatic hydrocarbons INTRODUCTION The diesel exhaust emitted from heavy and light duty engines is a mixture of gases, vapours, semivolatile compounds and particles. In the 1980s, animal experiments demonstrated that exposure to diesel emissions impaired respiratory function and was suspected of causing pulmonary carcinoma (Heinrich et al., 1986). Since then several studies have shown that exposure to diesel engine emissions can provoke asthma, cardiovascular diseases and an increased lung cancer risk in several occupations (Boffetta et al., 1988; Briiske-Hohlfeld et al., 1999; Hong et al., 2002). On the other hand, some contradictory results have also been reported (Muscat and Wynder, 1995; Valberg and Watson, 2000). It has recently been proposed that the fine and ultrafine particles may *Author to whom correspondence should be addressed. Tel: +358 9 47472614; fax: +358 9 47472114; e-mail: leea.kuusimaki @ ttl.fi cause the most harmful effects to the lungs (MacNee and Donaldson, 2000). About 90% by mass of diesel exhaust particles are <1 pm in diameter. They are formed in the combus tion chamber of the engine exhaust system and are released as respirable emissions into the atmosphere after agglomeration and condensation (Kleeman et al., 2000). The particles, which contain elemental carbon and organic compounds, represent a good adsorption surface for the hundreds of compounds produced after incomplete combustion in diesel engines. Well-known groups of compounds released within the emissions are polycyclic aromatic hydro carbons (PAHs) and nitrated PAHs (Pederson and Siak, 1981; Miguel et al.. 1998). Polyaromatics may also originate from unburned diesel fuel, lubricating oil and the pyrosynthesis of low molecular weight polyaromatics. - PAHs with two or three aromatic rings are mostly detected in the vapour phase, whereas those with between four and seven fused rings are found mostly in the particle phase. In addition to 389 390 L. Kuusimaki et al. PAHs, there is also a mixture of several compounds such as nitrogen, carbon and sulphur oxides, formaldehyde and butadiene in the gas phase of diesel exhaust (IPCS, 1998). . Anecent report by Gerde et al. (2001) focused on the release of benzo[a]pyrene (B[a]P) from the surface coating of diesel particle soot. They reported a rapidly released pulse of B[a]P, which was quickly adsorbed through the alveolar epithelium after inhal ation. B[a]P appeared mostly unmetabolized in the circulation, but after 1 h B[a]P was systemically metabolized. The data indicate that absorption through the alveolar epithelium is an important route of entry to the circulation of unmetabolized B[a]P. Several occupational studies have reported B[a]P levels in various job categories due to its major carcinogenic potential. Exposure to B[a]P levels of- 1.5-184 ng/m3 have been measured among truck drivers and bus garage workers (Guillemin et al.. 1992; Hemminki et al.. 1994; Schoket et al.. 1999). Estimates of the occupational exposure to diesel exhaust in Finland showed that during 1990-1993, -39000 workers were exposed via land transport. In 1997. exposure to PAHs alone was notified as having affected 1545 machine and engine repair workers and foundry workers (Kauppinen, 2001). Reformulated fuels with aromatic hydrocarbon contents <20% by vol and sulphur content <0.005% by mass are used in Finland by virtually all heavy and light duty diesel engines in trucks, buses and other vehicles. Two types of reformulated diesel fuels are used due to the seasonal changes in the climatic conditions. The main differences in the reformulated summer and winter grade diesel fuels are the density and cetane number (Mikkonen et al.. 1997). This study reports the analysis of PAHs derived from diesel engine emissions in the personal air samples of bus garage workers. The distributions of 15 vapour and particle phase PAHs in summer and winter were measured and the levels of PAHs in garages were estimated. SUBJECTS AND METHODS Study population A group of 20 male and two female non-smoking bus garage workers participated in this study. Personal data, job type and daily working hours were obtained from questionnaires completed by every participant. The average age and employment of garage workers were 48 yr (range 30-59) and 26 yr (range 9-41), respectively. The control group consisted of 48 non-smoking office workers and research scientists, who had no occupational exposure to PAHs. The average age was 49 yr (range 25-61) and duration of employment 16 yr (range 1-43). The working time in the garages was 7.5 h. The garage workers consisted of mechanics, electricians, w elders. painters, foremen and workers who cleaned, washed and refuelled the buses. The work was done completely inside in both the study and the control groups. The bus fleet was no older than 5-6 yr, and 200 buses were serviced daily in three garages. Typi cally, buses were driven from outdoors into a large hall, where engine repair, washing, cleaning and refuel ling were carried out. Inside the halls, the engines were started and stopped several times and the diesel exhaust occasionally became elevated. Especially in the winter, diesel exhaust drifted into the repair halls when the engines were cold-started outdoors close to the garages. Description of the bus garages Garage number 1 was built in 1962 and partially remodelled in 1988. The size of the hall is 2000 m2 and 70 buses are kept in this hall. About 400 buses are fixed or cleaned there every month! Detailed information on ventilation is not available, but the ventilation operates in summer and in winter. The exhaust pipes of the buses were connected to a pipe line to carry the exhaust of running engines out of the hall. The ventilation removes heat, but the air is not circulated. The incoming air is filtered. Garage number 2 was built in 1952 and remodelled in 1992. No buses are kept in this 1800 m2 hall. About 140 buses visit this repair and cleaning hall per month. No detailed information on the ventilation is available, but the ventilation system is very similar to garage number 1. Garage number 3 was built in 1966 and the total size of the hall is 3750 m2. The hall is divided into a repair shop and a washing line, which process about 200 buses per month. The air is circulated in the hall and some fresh outdoor air is conducted into the hall. This garage will be remodelled in 2004. Air sampling Outdoor temperature during the sampling period varied from -24 to 0C in winter and from 9 to 17C in summer. Five air samples were collected from the breathing zone of each garage worker by sampling every other day of consecutive weeks. Each person carried around the waist a pump connected via a silicon tube to the air sampling device, which was attached to the clothing on the right shoulder -10 cm away from the mouth. The sampling strategy was similar in summer and winter. Small battery powered pumps (SKC Models 222 and 224 and DuPont model S2500) were used to pump air through the sampling device containing a filter and an adsorbent tube. Particles were collected on a polytetrafluoroethylene (PTFE) filter (ZefluorTM. 37 mm diameter. 2.0 pm pore size; SKC Inc ), providing a very low back ground in the analysis and an efficient retention of agglomerated particles and PAHs bound to carbon due to the matri\ of randomlv oriented fibres, which L PAMs and bus aaraae worker*. 391 are wound or bonded together in several layers (depth filter) and retain particles on the surface and throughout the matrix (ASTM, 1999). The PTFE filter was placed in a closed face aerosol analysis monitor (field monitor MAWP037 AO; Millipore), which was connected to the XAD-2 adsorbent (226-30: SKC Inc.) collecting vapour phase compounds. Air samples were collected during a full working day with a flow rate of 1.4 1/min for 5-8 h (400-700 I/sample). Sampling devices were protected against light during and after sampling bv wrapping them in aluminium foil. The samples were stored at -20C before the analyses. Using the personal air sampling devices, the control samples were collected as stationary air samples from the offices in which the control persons worked 8 h/day. Analytical methods The air samples collected on filters and XAD-2 adsorbents were extracted with 5 ml of cyclohexane and 1 ml of acetonitrile, respectively. Extraction was carried out by sonication for 30 min at room tempera ture (Yrjanheikki et al., 1995). After cyclohexane extraction, the solvent was concentrated under nitrogen and changed to acetonitrile at 40C. Acetonitrile was added dropwise while evaporating cyclohexane in order to prevent the samples drying out. Cyclohexane as the upper layer easily evaporated and the final volume was adjusted to 1 ml with acetonitrile. Before the analyses all samples were filtered through a GHP filter (0.45 |4m Bulk Acrodisc Syringe Filter; Gelman). A standard mixture containing 16 PAHs in acetonitrile was obtained from Ehrenstorfer (Augsburg, Germany). Fifteen PAHs were quantified; only acenaphthylene was not detected, due to its weak fluorescence. Standards were prepared by spiking known amounts of PAH mixture onto the filter and XAD-2 absorbent. As background impuri ties, naphthalene (1.2-11 ng/sample), acenaphthene (7-20 ng/sample), fluorene (5-23 ng/sample) and phenanthrene (0.5-18 ng/sample) were analysed in the XAD-2 absorbent. These background values were subtracted from the corresponding compounds measured in the air samples. PAH analyses were carried out according to the methods of the National Institute of Safety and Health (NIOSH methods 5506 and 5800) (NIOSH, 1998). The HPLC appar atus (HP1100; Agilent Technologies, Waldbronn, Germany) was equipped with a fluorescence detector with two programmable emission channels (FLD1A and FLD1B). A guard column (10 x 3 mm; ChromPack) and an analytical PAH column (100 x 3 mm, 5 pm particle size; ChromPack) separated the samples at a flow rate of 0.3 ml/min. Typically, 15 |il in acetonitrile was injected into the HPLC. A gradient of water (solvent A) with an increasing acetonitrile content from 40 to 100% (solvent B) over 16 min was used, followed by 100% B for 13 min. Between the injections, the column was washed with 100% acetonitrile for 2-3 min at a flow rate of 0.6 ml/min (Makela and Pyy, 1995). Statistical evaluation The results are presented as arithmetic means of PAH concentrations; when calculating the means, observations below the limit of quantification (LOQ) were given a value of 0. A repeated measurement method was applied for statistical comparisons between the exposure and control groups with respect to PAH concentrations (Diggle et al., 1994; Littell et al., 1996). In this method, the correlation of five measurements from the same person was taken into account. A logarithmic transformation was performed to obtain a normal distribution of the data (values below the LOQs were replaced with values of half the LOQs). The Spearman rank correlation coefficient was used in correlation analysis. RESULTS Recovery and limit of quantification (LOQ) Recoveries for 15 PAHs in the particle and vapour phases with the relative standard deviations ( RSD) were determined using six parallel samples spiked with two concentrations (4.0 and 20 ng/ml) of the PAH standards (Table 1). The recovery of each PAH obtained from the filters spiked with 20 ng/ml ranged from 72 to 122% ( 2-12%), being clearly dependent on the individual PAH. The recovery varied from 82 to 108% ( 5-20%) at the low concentration of 4.0 ng/ml (data not shown). Seven PAHs with between two and four fused aromatic rings were analysed from the XAD-2 adsorbent. The recoveries for naphthalene, acenaphthene, fluorene, phen anthrene, anthracene, fluoranthene and pyrene at a concentration level of 200 ng/ml were between 57 and 84% (+5-26%). The LOQ values calculated for 700 1 of air varied from 0.9 to 3.7 ng/m3 for particu late PAH and from 3.3 to 50 ng/m3 for vapour phase PAHs (Table 1). The limit of detection (LOD) for PAHs varied from 0.02 to 1.1 ng/sample and the dayto-day variation at a concentration level of 5.0 ng/ml was 7-17%. Figure 1 shows an example of the HPLC chromatograms analysed for PAHs in the standard mixture (Fig. la), an air sample of the vapour phase (Fig. lb) and the particle phase (Fig. lc) obtained from an exposed worker. Garage workers were mainly exposed to naphtha lene, fluorene and phenanthrene in winter and summer, although other low molecular weight PAHs in the vapour phase were also detected (Fig. 2a). The concentrations of naphthalene, fluorene and phenan threne were about 2-fold higher in winter when compared with those in summer. In winter, mean levels of 1285, 167 and 730 ng/m3 were determined for naphthalene, fluorene and phenanthrene, respect- 392 L. Kuusimaki el al. Table 1. The recovery and the limit of quantification (LOQa) of the measured PAH compounds Compound 1 Naphthalene 2 Acenaphthene 3 Fluorene 4 Phenanthrene 5 Antracene 6 Fluoranthene 7 Pyrene 8 Benzo[n]anthracene 9 Chrysene 10 Benzo[i>jfluoranthene 11 Benzo[k]fluoranthene 12 Benzo[n]pyrene 13 Dibenzo[a/t]antracene 14 Benzo[g/t/]perylene 15 Indeno[1.2,3-ct/]pyrene Excitation Emission wavelength PFTE filter wavelength (nm) (nm) l/2b Recovery (RSD. n = 6) 230 350/360 92 + 7 245 350/450 78 12 245 350/450 88 7 245 350/450 101 12 245 413/450 72 9 245 450/413 95 + 6 245 413/450 87 + 7 270 413/450 77 + 3 270 413/450 85 + 3 270 450/413 103 + 2 270 413/450 122 + 2 270 413/450 75 + 5 305 413/487 94 + 9 305 413/487 76 + 7 305 487 87 + 3 LOQ (ng/m3) 3.7 2.7 1.6 3.5 0.9 1.8 1.3 1.1 0.9 0.6 1.3 0.9 2.3 2.0 1.0 'LOQ. 10 times standard deviation + the smallest calibration standard 0.2-0.4 ng/ml. bl. wavelength used for quantification; 2. wavelength used for qualification. ND. not determined. XAD-2 Recovery LOQ (RSD. n = 6) (ng/m3) 84+ 19 50 60 26 25 73 + 5 41 69+ 15 23 67 + 6 ii 70 + 7 5.1 57 10 5.3 ND 3.3 ND 3.3 ND 2.8 ND 3.9 ND 4.5 ND 3.3 ND 4.5 ND 6.2 ively, in the vapour phase (Table 2). In the summer samples, the corresponding values were 760, 64 and 405 ng/m3. The means of four other PAH compounds detected in the vapour phase were invariably below 6.6 ng/m3 in both winter and summer. Naphthalene was only detected in the vapour phase, accounting in winter for 58% and in summer for 62% of all vapour phase PAHs. The level of anthracene was low in both phases and fluoranthene and pyrene were slightly higher in the particle phase. Fluorene and phenanthrene were determined in winter and summer at concentrations from 50 to 200 times lower in the particle than in the vapour phase. Thirteen PAH compounds were determined in the particle air samples (Fig. 2b). The mean value of the exposure to particle phase PAHs was 25 ng/m3 (range ND--494 ng/m3, n = 106) in winter and 8.5 ng/m3 (range ND-46 ng/m3, n - 111) in summer. A statistic ally significant exposure difference between the exposed and controls in winter (P < 0.001) and in summer (P < 0.005) was observed as far as PAHs present in the particle phase were concerned. The most abundant PAHs in particle phase samples in winter were phenanthrene, fluoranthene, pyrene, B[a)P and benzo[g/']perylene. In summer, the PAH profile in particles was slightly different, with the most abundant compounds being fluorene, phenanthrene and fluoranthrene. The highest concentration in the particle phase, in both winter and summer, was analysed for the non-carcinogenic phenanthrene. The total PAH concentrations of weak (benzo[]anthracene. benzo[g/7/]perylene and chrysene) and strong carcino gens (benzo[/?]fluoranthene. benzo[ A)fluoranthene. B[a]P. dibenzo[n/i]anthracene and indenoj 1.2.3cc/]pyrene) in all air samples analysed were less than 2%. The average levels of B[a]P and pyrene were 2.9 and 3.6 ng/m3 in winter and 0.6 and 0.8 ng/m3 in summer. In winter, the exposure to B[a]P and pyrene was about four times higher than in summer. A significant correlation (r = 0.78. P < 0.001, n = 44) between B[a]P and pyrene concentration in the personal air samples was observed (Fig. 3). The levels of PAHs in the three bus garages are shown in Fig. 4a. The mean level of PAHs was about the same (1000 ng/m3) in garages numbers I and 2 in summer, but garage number 3 had a significantly higher (P < 0.001. df = 80) mean level (1700 ng/m3). Elevated levels were observed in all garages in winter time (1500-2400 ng/m3). However, garage number 2 had a slightly lower level of PAHs in winter than garages 1 (P < 0.00I. df = 70) and 3 (P < 0.001. df = 64). The mean total PAH concentrations of 2241 and 1245 ng/m3 in the exposed workers' air samples in winter and summer, respectively, were significantly higher (P < 0.001, df = 38.3 and df = 35.7) than the 254 and 275 ng/m3 of the control group (Fig. 4b). Large differences in levels of PAHs were observed between the air samples (890-5960 ng/m3 in winter and 247-3260 ng/m3 in summer). The concentrations of PAHs measured in the bus garages were signifi cantly higher in winter than in summer (P < 0.001, df = 43.7), however, no seasonal variation in the control group (P = 0.93. df = 2.12) was observed. PAHs ami hu S '-Ml ,ILV 'A orkcrs 393 fable 2. Exposure ol bus garage workers ami control persons to PAHs in winter ami summer (corrected tor recosery) Compound-1 Volatile PAH (ng/m-*) Napht Ac Flu Phe Ant Fit P Sum of 7 PAH Bus garages Winter (n = 104 or I061') Mean Range Controls Summer In = 106 or II1L) Winter Oi = 5) Mean Range Mean Range 1285 640-3362 760 204-2450 226 160-303 6.6 ND-232 ND ND 167 53-936 64 ND--160 ND 770 63-2664 405 43-1180 33 25-53 0.4 ND-29 ND ND 0.4 ND-20 1.3 ND-40 ND 2.6 ND--10 0.3 ND-10 ND 2198dd 875-5465 1230d 247-3230 254 190-355 Summer tn = 17) Mean Range 230 100-510 ND ND 40 ND-215 ND ND ND 275 100-720 PAH on particles (ng/m1 ) Flu 1.6 Phe 4.1 Ant 0.2 Fit 3.1 P 3.2 BaA 1.0 Chr 1.7 BbF 1.5 BkF 0.4 B1 a ] P 2.9 DBA ND BPe 3.1 IP 2.3 Sum of particle PAH 25dl' Z(particles+volatile) 22411,1 ND-24 ND-50 ND -10 ND-79 ND-105 ND-29 ND-24 ND-17 ND-11 ND-60 ND-58 ND-34 ND--+94 890-5960 1.4 1.9 0.01 1.4 0.8 0.3 0.2 0.5 0.06 0.6 ND 0.6 0.2 8.5" 1245d ND-28 ND-15 ND-1.1 ND-8.9 ND-12 ND-3.3 ND-2.5 ND-2.6 ND-1.4 ND-8.9 ND-6.3 ND-2.8 ND--+6 247-3260 ND ND ND ND ND ND ND ND ND ND ND ND ND ND 254 190-355 ND ND ND ND ND ND ND ND ND ND ND ND ND ND 275 100-720 '`Napht. naphthalene; Ac. acenaphthene; Flu. fluorene; Phe. phenanthrene: Ant. anthracene; Fit. fluoranthene; P. pyrene; BaA. benzo[</]anthracene; Chr. chrysene; BbF. benzo[i>]fluoranthene; BkF. benzo[k]fluoranthene; B[a]P, benzo[n]pyrene; DBA. dibenzo|u/?]amhracene; BPe. benzo[g/tt]perylene; IP, indeno[ 1,2.3-/]pyrene. hDifferent number of particle (106) and volatile (104) samples due to losses in sample preparation. `Different number of particle (111) and volatile (106) samples due to losses in sample preparation. dP < 0.001 when result is compared with controls. eP < 0.005 when result is compared with controls. 'P < 0.001 when winter and summer samples are compared. ND, not detected. < LOQ. The exposure as a function of the job title is shown in Fig. 4c. In general, exposure was lower in all groups in summer than in winter, but as far as the exposure was concerned there was only an insignifi cant difference between the groups in any season. The workers were divided into three groups according to high (>2500 ng/m3), medium (1000-- 2500 ng/m3) and low (<1000 ng/m3) exposure to PAHs. In the winter and summer, the high exposure group consisted of mechanics (winter 43%, summer 0%), welders (winter 29%, summer 0%), refuelling persons (winter 14%, summer 0%) and foremen and electricians (winter 14%, summer 0%). There were mechanics (winter 60%, summer 47%), welders (winter 7%, summer 20%), refuelling persons (winter 7%, summer 15%) and foremen and electricians (winter 27%, summer 20%) in the medium exposure group. None of the workers were in the low exposure group in winter and only mechanics (67%), foremen and electricians (33%) were in this particular group in summer. DISCUSSION There are a few occupational studies evaluating the distribution of vapour and particle phase PAHs origin ating from diesel exhausts. The complex mixture of diesel exhaust has been assessed in earlier studies using diesel particulate matter, carbon and gaseous components as surrogates (Heinrich et al., 1986; Muscat and Wynder, 1995; Briiske-Hohlfeld et al., 1999; MacNee and Donaldson, 2000; Valberg and 394 L. Kuusimaki el al. Fig. 1 . An example of HPLC chromatograms of the PAH standard mixture at a concentration level of 1 ng/ml (a) and of the vapour phase (b) and particle phase (c) air samples of an exposed worker. FLD1A and FLD1B refer to the two emission channels used. PAH compounds analysed were: 1, naphthalene; 2, acenaphthene; 3. fluorene; 4, phenanthrene; 5, anthracene; 6. fluoranthene; 7. pyrene; 8. benzo[]anthracene; 9. chrysene; 10. benzo[F]fluoranthene; 11. benzo[/t]fluoranthene; 12. benzo[]pyrene: 13. dibenzo[z)/t|anthracene: 14. benzo[g/i/]per\ lene: 15. indenof 1,2.3-r</]pyrene. 1400 PAHs tind bus 'jjijlil' workers W5 Fig. 2. The average concentration (ng/m1) of vapour (a) and particle (b) phase air samples of the exposed garage workers analysed for single PAH compounds in winter (filled circles) and in summer (open triangles). Fig. 3. The correlation (r = 0.784, P < 0.0001, n = 44) between the average pyrene and B[a]P air concentrations among exposed garage workers. The dashed lines display the 95% . confidence limits. Watson, 2000; Hong et a/., 2002). This study evalu ated the seasonal exposure of 22 bus garage workers to 15 PAH compounds in personal air samples. The measurements confirmed that -98% of PAHs meas ured were related to the vapour phase compounds. Our data are in line with other reports, which have indicated that if a back-up adsorbent is not used, three to five ring PAH compounds are lost during sampling (Davis et al., 1987; Hansen et al., 1994; Kleeman et al., 2000). High molecular weight PAHs (between four and seven rings) were only detected in the particle phase. Other factors that influence the distri bution of PAHs between the two phases are tempera ture, air sampler, filter type, particulate matrix and equilibrium concentrations (Davis et al., 1987). Most of the vapour phase PAH compounds are not carcin ogens, but their measurement is justified by the fact that some of them are important precursors of highly mutagenic nitro-PAHs (Arey et al., 1986; Atkinson and Arey, 1994). Naphthalene accounted for -85% of the total PAH load in an iron foundry study (Hansen et al., 1994), whereas in our study the naphthalene concentrations (0.6-3.3 |4g/m3) in the bus garages ranged from 60 to 80% of total PAHs. Naphthalene, fluorene and phenanthrene were significant constitu ents of total PAHs analysed in a new Finnish coking plant (Yrjanheikki et al., 1995). This observation is similar to our data, in which they were also the major 396 L. Kuusimaki et al. Y///X Winter Electricians Fig. 4. A box plot presentation of PAHs in garage work. The levels of PAHs in three different garages (a), the difference between the exposed group and the control group tb) and the difference in exposure between the job groups (c>. The arithmetic mean (filled circles), minimum (filled inverted triangles) and maximum (filled triangles) values are shown and three horizontal lines from the top of the box indicate the number of data points that are below 75. 50 and 25<;r of all measurements. tween t filled m the PAIL and bus garage workers 397 constituents, but present at much lower concentra REFERENCES tions. We noticed only minor variations in the mean total PAH concentrations between the three garages. In the ASTM. (1999) Standard practice for evaluation of air assay media by the monodisperse DOP (dioctyl phthalate) smoke test. D2986-95a. West Conshohocken. PA: American Soci largest garage, the average PAH level in both seasons ety for Testing and Materials. (1698 ng/m3) was at the same level as that of the smallest garage (1245 ng/m3). The highest PAH exposure (2140 ng/m3) was detected in the mid-size Arev .1, Ziclinska B. Atkinson R et al. (1986) The formation of nitro-PAHs from the gas-phase reactions of fluoranthene and pyrene with the OH radical in the presence of NOx. Atmos Environ: 20: 2339-45. garage, apparently being attributable to its old venti lation system. The exposure to PAHs among the control subjects was lower or at the same level as that reported in other studies (Guillemin et al., 1992; Atkinson R. Arey J. (1994) Atmospheric chemistry of gasphase polycyclic aromatic hydrocarbons: formation of atmospheric mutagens. Environ Health Perspecc. 102 (suppl. 4): 1 17-26. Boffetta P, Stellman SD. Garfinkel L. (1988) Diesel exhaust Schoket et al., 1999). We analysed 12 PAHs in the particles, five being non-carcinogenic and seven carcinogenic compounds; dibenzo[a/t]anthracene was not detected. The concentrations of these PAHs in the exposure and mortality among males in the American cancer society prospective study. Am J Ind Med: 14: 403-15. Briiske-Hohlfeld 1, Mohner M. Ahrens W et al. (1999) Lung cancer risk in male workers occupationally exposed to diesel motor emissions in Germany. Am i Ind Med; 36: 404-14. particle phase were about 140-fold lower in summer Chiang PC. Chiang YC, Chang EE. Chang SC. (1996) Charac and 90-fold lower in winter than the concentration of total PAHs in the vapour phase. Biomonitoring studies conducted in the garages terizations of hazardous air pollutants emitted from motor vehicles. Toxicol Environ Chem: 56: 85-104. Davis CS. Fellin P. Otson R. (1987) A review of sampling methods for polycyclic aromatic hydrocarbons in air. J Air have shown higher (15-184 ng/m3) levels of BfajP Pollut Control Assoc: 37: 1397-408. compared with our results (Hemminki et al., 1994; Schoket et al., 1999). The low B[a]P exposure detected in our study is in better agreement with data de Pereira P. Andrade JB, Miguel AH. (2002) Measurements of semivolatile and particulate polycyclic aromatic hydrocar bons in a bus station and an urban tunnel in Salvador, Brazil. J Environ Monit; 4: 558-61. obtained from truck drivers, a bus station (0.77-3.45 Diggle PJ, Liang KY, Zeger SL. (1994) Analysis of longi ng/m3) and ambient air measurements (mean 0.62 tudinal data. Oxford: Clarendon Press. 0.85 ng/m3) (Guillemin et al., 1992; Schoket et al., 1999; Zmirou et al.. 2000; de Pereira et al., 2002). Gerde P. Muggenburg BA. Lundborg ML, Dahl AR. (2001) The rapid alveolar absorption of diesel soot-adsorbed benzo(a)pyrene: bioavailability, metabolism and dosimetry Ratios between PAHs such as B[a]P and pyrene of an inhaled particle-carcinogen. Carcinogenesis; 22: 741 -- and B(a]P and benzo[g/t/']perylene measured in the 9. "" ! air samples have been used to characterize the origin Guillemin MP. Herrera H, Huynh CK. Droz PO, Vu Due T. (1992) Occupational exposure of truck drivers to dust and of PAHs (Masclet et al., 1986; Chiang et al., 1996; polynuclear aromatic hydrocarcbon: a pilot study in Geneva, Zmirou et al., 2000). In general, a ratio >1 between Switzerland. Int Arch Occup Environ Health; 63: 439-47. high and low molecular weight PAHs corresponds to gasoline and a ratio <1 to diesel vehicles. In our study, the ratio 0.01 indicates that the PAHs were attributable to diesel-driven vehicles. The Hansen AM. Ornland 0. Poulsen OM et al. (1994) Correlation between work process-related exposure to polycyclic aromatic hydrocarbons and urinary levels of a-naphthol, |3naphthylamine and 1-hydroxypyrene in iron foundry work ers. Int Arch Occup Environ Health; 65: 385-94. B[a]P:benzo[g/?(]perylene ratio has been used as indicative of exposure to vehicle or heating sources. Our data show ratios of 1.12 (in winter) and 1.02 (in summer), values that are in agreement with the study Heinrich U, Muhle H, Takenaka S et at. (1986) Chronic effects on the respiratory tract of hamsters, mice and rats after long term inhalation of high concentrations of filtered and unfil tered diesel engine emissions. J Appl Toxicol; 6: 383-95. Hemminki K, Soderling J, Ericson P, Norbeck HE. Segerback of Masclet et al. (1986), emphasizing the origin of exposure to vehicle emissions. The seasonal variation in total PAH exposure in the garages was 2-fold higher in winter than in summer. D. (1994) DNA adducts among personnel servicing and loading diesel vehicles. Carcinogenesis; 15: 767-9. Hong YC. Lee JT, Kim H, Ha EH. Schwartz J, Christiani DC. (2002) Effects of air pollution on acute stroke mortality. Environ Health Perspect; 110: 187-91. The concentrations of vapour phase PAHs were 2-10 times higher in winter than in summer. In the particles, a 1- to 20-fold difference was observed. Although not the case for all individual PAHs. the IPCS. (1998) Selected non-heterocyclic polycyclic aromatic hydrocarbons. Environmental Health Criteria 202. Geneva: WHO International Programme on Chemical Safety. Kauppinen T. (2001) Finnish occupational exposure databases. Appl Occup Environ Hyg; 16: 154-8. difference in total PAHs concentration between Kleeman MJ, Schauer JJ. Cass GR. (2000) Size and composi winter and summer was statistically significant. tion distribution of fine particulate matter emitted from motor vehicle. Environ Sci Technol; 34: 1132^12. Littell RC, Milliken GA. Stroup WW. Wolfinger RD. (1996) Acknowledgements--The bus garage workers of Helsinki City SAS system for mixed models. Cary, NC: SAS Institute Inc.. Transport and the control persons are thanked for their valuable p. 633. contribution to this investigation. The Finnish Work Environ MacNee W, Donaldson K. (2000) How can ultrafine particles ment Fund and Helsinki City Transport financially supported be responsible for increased mortality. Monaldi Arch Chest this study. Dis; 55: 135-9. 398 L. Kuusimaki et al. Makela M, Pyy L. (1995) Effect of temperature on retention time reproducibility and on the use of programmable fluores cence detection of fifteen polycyclic aromatic hydrocarbons. J Chromatogr A; 699: 49-57. Masclet P, Mouvier G, Nikolaou K. (1986) Relative decay index and sources of polycyclic aromatic hydrocarbons. Atmos Environ; 20: 439-46. Miguel AH, Kirchstetter TW, Harley RA. (1998) On-road emission of particulate polycyclic aromatic hydrocarbons and black carbon from gasoline and diesel vehicles. Environ Sci Technol; 32: 450-5. Mikkonen S, Kiiski U, Makela M et al. (1997) Reformulated diesel fuel--four years experience in Finland, SAE Techni cal Paper Series, paper no. 971634, pp. 1-15. Warrendale, PA: Society of Automotive Engineers. Muscat JE, Wynder EL. (1995) Diesel engine exhaust and lung cancer: an unproven association. Environ Health Perspect; 103: 812-8. NIOSH. (1998) Manual of analytical methods. Issue 3. Wash ington, DC: National Institute of Safety and Health. Avail able at: http://www.cdc.gov/niosh/nmam/nmammenu.html. Pederson T, Siak J. (1981) The role of nitroaromatic com pounds in the direct acting mutagenicity of diesel particle extracts. J Appl Toxicol; 1: 54--60. Schoket B, Poirier MC, Mayer G et al (1999) Biomonitoring of human genotoxicity induced by complex occupational exposures. Mutat Res; 445: 193-203. Valberg PA, Watson AY. (2000) Lack of concordance between reported lung-cancer risk levels and occupation-specific diesel-exhaust exposure. Inhalat Toxicol; 12(suppl. 1): 199 208. Yrjanheikki E, Pyy L, Hakala E, Lapinlampi T, Lisko A, Vahakangas K. (1995) Exposure to polycyclic aromatic hydrocarbons in a new coking plant. Am Ind Hyg Assoc J; 56: 782-7. Zmirou D, Masclet P, Boudet C, Dor F, Dechenaux J. (2000) Personal exposure to atmospheric polycyclic aromatic hydrocarbons in general adult population and lung cancer risk. J Occup Environ Med; 42: 121-6.