Int Arch Occup Environ Health (2010) 83:803–811 DOI 10.1007/s00420-010-0516-4

O R I G I N A L A R T I CL E

Biological eVect markers in exhaled breath condensate and biomonitoring in welders: impact of smoking and protection equipment Monika Gube · Joachim Ebel · Peter Brand · Thomas Göen · Karl Holzinger · Uwe Reisgen · Thomas Kraus

Received: 3 September 2009 / Accepted: 19 January 2010 / Published online: 4 February 2010 © Springer-Verlag 2010

Abstract Purpose The objective of this study was to investigate the eVect of welding as well as the impact of smoking and protection measures on biological eVect markers in exhaled breath condensate. Additionally, biomonitoring of chromium, aluminium and nickel in urine was performed to quantify internal exposure. Methods Exhaled breath condensate (EBC) and urine samples of 45 male welders and 24 male non-exposed control subjects were collected on Friday pre-shift and after 8 h of work post-shift. In EBC, biological eVect markers such as malondialdehyde, nitrite, nitrate, 3-nitrotyrosine, tyrosine, hydroxyproline, proline, H2O2 and pH-value were measured while aluminium, nickel, and chromium were measured in the urine samples. Results Although internal exposure to aluminium, nickel and chromium in this study was low, welders showed signiWcantly increased concentrations of all these parameters at baseline compared to non-exposed controls. Moreover, welders had higher nitrate concentrations in EBC at base-

M. Gube · J. Ebel · P. Brand (&) · T. Göen · T. Kraus Institute for Occupational and Social Medicine, Medical Faculty, RWTH Aachen University, Pauwelsstraße 30, 52074 Aachen, Germany e-mail: [email protected] T. Göen Institute for Occupational, Social and Environmental Medicine, University Erlangen-Nuremberg, Schillerstraße 29, 91054 Erlangen, Germany K. Holzinger · U. Reisgen ISF, Welding and Joining Institute, RWTH Aachen University, Pontstraße 49, 52062 Aachen, Germany

line and after shift. Nitrate concentration was considerably lower after shift if personal protection equipment was used. H2O2 was increased only when subjects smoked during shift. Conclusion It has been shown that welding-associated long-term and short-term health eVects could be detected in a population of welders. The results also showed that using personal protection equipment is of high importance and H2O2 may be an eVect marker associated with smoking rather than with welding fumes, while nitrate in EBC seems to be sensitive to welding fume exposure. Keywords Welding · Biological eVect markers · Exhaled breath condensate · Biomonitoring

Introduction The ongoing debate about protection from Wne and ultraWne aerosol particles which is a current hot focus of public dispute is not only a question of public health in general but also has a great inXuence on the development of workplace threshold values—for example, for welding fumes and gases. UltraWne particles (diameter <0.1 m), which are the main constituent of welding fumes, are of substantial interest because of their particular toxicological characteristics (André et al. 2006; Borm 2004; Brown et al. 2001; Kreyling et al. 2004; Oberdörster et al. 1990, 1995; Seaton et al. 1995; Wiebert et al. 2006). Since the toxicity of welding fumes is dependent on multiple factors such as welding technique and used materials, exposure to welding fumes and gases pose a challenge with regard to medical and toxicological assessment. Dependent on the used materials and welding technique, welding fumes contain potentially harmful agents like hexavalent chromium, nickel oxide, Xuoride and

123

804

many other organic or inorganic substances in diVerent concentrations. These agents are able to induce health eVects in humans such as inXammatory reactions and oxidative or nitrosative stress (Antonini et al. 1998; Han et al. 2005; Hurley et al. 2003; Kim et al. 2005; Palmer et al. 2006; Scharrer et al. 2007; Sjögren et al. 2002, 2006). Various studies indicate that the analysis of markers of inXammation, nitrosative and oxidative stress in exhaled breath condensate samples (EBC) (Balint et al. 2001; Barregard et al. 2008; Boyce et al. 2006; Broding et al. 2009; Caglieri et al. 2006; Horváth et al. 2005; Kharitonov and Barnes 2001; Romieu et al. 2008) may be a promising tool for the non-invasive assessment of inXammatory reactions in the lungs due to exposure to environmental and occupational pollutants. In this study, exhaled breath condensate (EBC) was used to investigate the eVect of welding on biological eVect markers. Additionally, biomonitoring of chromium, aluminium and nickel in urine was performed. All measurements were taken on Friday pre- and post-shift.

Methods Subjects Forty-Wve male welders and 24 male non-exposed control subjects were included into the study. Inclusion criteria for the welders were as follows: (1) age of less than 50, (2) welding occupational history of more than 10 years, (3) one of the following welding processes was used at the time of the study, shielded manual metal arc welding (MMA), gas metal arc welding with CO2 (MAGC) or with gas mixture (i. e. Argon and CO2) (MAG-M), tungsten inert gas welding (TIG) and metal inert gas welding (MIG), (4) informed written consent. The 45 recruited welders were divided in subgroups according to the predominant welding techniques used. Eleven welders mainly used MIG welding, 13 welders TIG, 10 welders MAG-M, 4 welders MAG-C and only 2 welders MMA. The predominant welding technique was coded by a numerical predominant welding technique index. Similarly, the company aYliation was coded by a company aYliation index. Five welders had welding fume exposure in the past but not at the time of the study. These subjects were classiWed as welders but without current exposure (weekly exposure = 0, Friday exposure = 0, predominant welding technique = control). Exposure to cigarette smoke and to welding fumes was quantiWed by anamnesis. Using a questionnaire, all subjects were asked how many years they had been smoking, their average cigarette consumption and if they had smoked during the work shift studied. Additionally, cotinine level was measured in urine. Based on the total years of prevailing

123

Int Arch Occup Environ Health (2010) 83:803–811

welding activity, the lifetime welding fume exposure in the group of welders was quantiWed by “welding years”. Moreover, exposure during the study week was quantiWed by the number of welding hours within the week before Wrst examination (weekly exposure) and by the number of welding hours on the study day (Friday exposure). The control group comprised 24 healthy adult males without any history of lung disease and no welding fume exposure and was either employees of the university hospital or non-exposed employees of the participating companies. Demographic data of the study population are shown in Table 1. The protocol of the study was approved by the ethics committee of the medical school of RWTH Aachen University. Informed written consent was obtained from each subject prior to inclusion. Study design The current study was performed at the workplaces of seven companies Friday pre-shift (T1) (=baseline) and after 8 h of work at the end of the shift (T2). The study protocol included a questionnaire and the exhaled breath condensate, blood and urine samples. Additional lung function measurements were taken. The results from this study part were published elsewhere (Gube et al. 2009). Ambient monitoring Aerosol mass concentrations and gas concentrations at the diVerent workplaces were spot-checked during the work shift on Friday. Measurements were undertaken according to the European norm ISO 15011-1, 2 and ISO 10882-1, 2. Gravimetrical measurements of the aerosol mass concentration were taken for 2 h using Wbre glass Wlters and sampling heads for inspirable dust and dust inhalable into the alveolar region (according to EN ISO 10882-1). Concentrations of gaseous NO2, CO 2, CO and ozone were assessed using Dräger tubes (Drägerwerk AG & Co. KGaA, Lübeck, Germany). Both, aerosol and gas samples were taken close to the heads selected workers. Biological monitoring of metals The determination of chromium, nickel and aluminium in urine using graphite furnace atom absorption spectrophotometry was conducted according to analytical methods recommended by the Deutsche Forschungsgemeinschaft (DFG) (Deutsche Forschungsgemeinschaft 2008). All data were generated under good laboratory practice conditions, applying internal as well as external quality control measures.

Int Arch Occup Environ Health (2010) 83:803–811 Table 1 Anthropometric data of the study population

Parameter

805

Unit

Number Age Height

non-w

24

w

45

Median

SD

IQR

Min

Max

non-w

40.04

41

8.63

13

24

54

w

41.76

43

8.52

14

20

57

non-w

178

176.78

6.15

9.75

170

190

179.04

179

6.59

9

162

197

non-w

89

85

16.21

26

68

126

w

87.49

87

12.75

15

61

140

Smoking

non-w

10n/6s/8ex

w

16n/21s/8ex

PY

non-w

7.74

1.75

10.36

15

0

32.5 60

Cotinine Welding years

cm

Mean

w Weight

non-w non-welders, w welders, SD standard deviation, IQR interquartile range, PY pack-years, n never smoker, s smoker, ex ex-smoker

Years

Group

kg

g/l

w

10.81

5.1

14.37

19.5

0

non-w

277.13

9

600

23.5

2

1,830

w

664.79

29.5

914

1,630

1

2,786

non-w

0

0

0

0

0

0

w

20.23

20

9.97

15

4

44

Assessment of breath condensate biomarkers Collection of breath condensate Exhaled breath condensate (EBC) was collected using the ECOSCREEN condenser from Jaeger (Höchberg, Germany) according to published recommendations (Horváth et al. 2005). The subjects wore a nose clip to ensure oral breathing. They were instructed to breath tidally through a mouthpiece connected to a condenser until a volume of 200 litres of exhaled air was achieved (as measured by the EcoVent Xow meter). The exhaled air entered and left the condensing chamber (tempered at ¡20°C) through one-way valves, keeping the chamber closed throughout the process. The EBC samples were aliquoted and stored at ¡80°C until analysis. Chemical analysis of biomarkers in EBC Malondialdehyde (MDA) in EBC was analysed using the method of Andreoli et al.(Andreoli et al. 2003) with slight modiWcations. BrieXy, MDA was derivatized with dinitrophenylhydrazine (DNPH) at 37°C for 8 h, and the resulting DNPH derivatives were separated by HPLC and speciWcally determined using tandem mass spectrometry in the ESI-negative mode. Calibration was based on aqueous standards in the range of 5–200 nM/l. Methyl Malondialdehyde (MeMDA) was synthesized according to Claesen et al. (2001) and was used as internal standard for the quantitative determination of MDA. The limit of detection for MDA in EBC was determined to be 2 nM/l, based on a signal-to-noise ratio of 3. Between-day precision for the determination of MDA in EBC has been determined to be

14.9% based on a aqueous sample that was spiked with 25 nM MDA/l and processed in every analytical series (n = 34). Nitrite and nitrate in EBC were determined photometrically according to the Griess reaction using a quantiWcation kit purchased by Roche Diagnostics (Dziedzic et al. 2003). QuantiWcation of both nitrite and nitrate in EBC was achieved by two measurements: at Wrst, nitrite was measured by the Griess reaction and then nitrate was reduced enzymatically to nitrite by nitrate reductase. Nitrate content of EBC was calculated based on the diVerence between both measurements. The limit of detection was determined to be 1 M for nitrite and nitrate, respectively. Betweenday precision for nitrite and nitrate (n = 26) was determined to be 8.7 and 2.1% at concentrations of 3 and 9 M/l, respectively. Simultaneous measurement of 3-nitrotyrosine, tyrosine, hydroxyproline and proline in EBC was undertaken according to our previous publication (Conventz et al. 2007). In brief, 1 ml of EBC was freeze-dried and reconstituted in 50 l of HILIC eluent for concentration purposes. Separation and quantiWcation of the analytes was achieved by hydrophilic interaction liquid chromatography (HILIC) coupled to tandem mass spectrometry using isotopically labelled analogues of the analytes as internal standards. The limit of detection was 5 ng/l for 3-nitrotyrosine and hydroxyproline and 500 ng/l for tyrosine and proline. Between-day precision (n = 8) for proline, hydroxyproline, tyrosine and 3-nitrotyrosine in EBC was determined to be 2.6, 13.2, 6.0 and 8.2%, respectively. Values below detection limit were replaced by one half of the detection limit for the statistical analysis.

123

806

Measurement of H2O2 and pH-value in EBC was taken immediately after sample collection using the ECoCheck of VIASYS-HealthCare (Höchberg, Germany). Measurement of H2O2 is based on the chemical reaction of H2O2 by peroxidase. Via a mediator, electrons are carried to an electrode and they are detected amperometrically. The measurement of pH-values was taken using an unbreakable and stable pH-sensor which allowed measurements in highly diluted solutions. This was more eVective than the conventional practice of using a glass electrode. It is worth mentioning that EBC represents a highly variable matrix. The dilution of droplets generated in the respiratory tract by water vapour varies considerably, as already observed by other authors (EVros et al. 2002). Consequently, it is necessary to adjust the results of EBC analyses to accommodate for this eVect. According to our observations, non-volatile substances like amino acids or peptides that are certainly present within the airway-lining Xuid of the lung would be ideally suited to correct for the degree of dilution of the lining Xuid in the collected EBC. Therefore, simultaneous measurements of 3-nitrotyrosine, hydroxyproline as well as their amino acid precursors tyrosine and proline were determined in the EBC samples in order to correct for this dilution eVect. Consequently, the quotient of the concentrations of the eVect parameters to the amino acids was calculated and used in the statistical analysis. As tyrosine and proline in EBC are highly correlated, tyrosine served as reference parameter for MDA, nitrite, nitrate and 3-nitrotyrosine, while proline served as reference parameter for hydroxyproline. Statistics All statistical analysis was performed using the SPSS software (Version 15.0.1). DiVerences between study groups were assessed using the non-parametric Mann–Whitney test (NPAR TESTS). In order to investigate the relation between the various endpoint variables and parameters quantifying exposure to cigarette and welding smoke, an analysis of variance was calculated. Since the biomonitoring and EBC parameters showed no normal distribution the data were logarithmized. These transformed data were in good agreement with the normal distribution (Shapiro–Wilk Test, p > 0.25). The analysis of variance was then calculated with the logarithmized endpoint parameters as dependent variable, and the company aYliation index and the predominant welding technique index as nominal factors and the cigarette packyears, the cotinine level, the welding years, the weekly welding exposure (hours) and the Friday welding exposure (hours) as metric covariates (UNIANOVA). In order to assess the acute eVect of a welding working shift, the change in endpoint variables during the shift was

123

Int Arch Occup Environ Health (2010) 83:803–811

quantiWed by the diVerences between the after shift and the before shift value of each parameter on Friday for the welders sub-population. The null hypothesis that this diVerence is zero was tested using Student’s t-test. Data are graphically presented as Box and Whisker Plots with the lowest dot representing the 5% percentile, the lower whisker the 10% percentile, the lower side of the box the 25% percentile, the line within the box the median, the upper side of the box the 75% percentile, the upper whisker the 90% percentile and highest dot the 95% percentile.

Results Ambient monitoring At the various workplaces under investigation, the median inhalable dust concentration was 4.9 mg/m3 (0.1–9.6 mg/ m³) and the median concentration of dust respirable into the alveolar region was 0.7 mg/m3 (0–2.2 mg/m³). The ozone concentration was between 0.11 mg/m3 and 5.4 mg/m3 (median 0.32 mg/m3), the NO2 concentration between 0.21 mg/m3 and 4.1 mg/m3 (median 1.0 mg/m3), the CO2 concentration between 0.02 and 0.04% and the median CO concentration below 2.5 mg/m3 (range: <2.5 and 50 mg/m3). Biological monitoring Medians and interquartile ranges of the biomonitoring parameters before and after shift are given in Table 2. The average concentrations of metals in the pre-shift urine of the subjects were 1.4 § 0.22 g/l for chromium, 2.8 § 0.46 g/l for nickel and 22.18 § 5.92 g/l for aluminium. The average creatinine concentration was 1.54 § 0.12 g/l, resulting in normalized metal values of 0.91 § 0.22 g/g creatinine for chromium, 1.82 § 0.32 g/g creatinine for nickel and 11.7 § 1.8 g/g creatinine for aluminium. Exhaled breath condensate The results of the analysis of markers in the exhaled breath condensate (before and after shift) are summarized in Table 2. DiVerences between welders and control subjects For the baseline data obtained before shift, the following signiWcant diVerences between welders and the control subjects were found (Mann–Whitney test): Welders showed a higher H2O2/tyrosine ratio concentration in the EBC (56 § 19 pM/g vs. 6.6 § 2.2 pM/g, p = 0.032), a higher nitrate/tyrosine ratio in EBC (0.71 § 0.13 nM/g vs. 0.22 § 0.04 nM/g, p = 0.002) (Fig. 1) and a higher

Int Arch Occup Environ Health (2010) 83:803–811 Table 2 Medians and interquartile ranges (IQR) of the biomonitoring parameters and markers in the exhaled breath condensate (EBC) measured in the welder population before and after work shift

Parameter

807

Before shift Median

After shift IQR

Median

IQR

Chromium/creatinine (g/g)

0.32

0.97

0.54

1.39

Nickel/creatinine (g/g)

1.11

1.6

1.47

2.27

11.2

9.05

8.38

Aluminium/creatinine (g/g)

7.14

Nitrite/tyrosine (nM/g)

0.68

0.92

0.66

0.76

Nitrate/tyrosine (nM/g)

0.39

0.87

0.36

0.97 3.16

MDA/tyrosine (pM/g)

2.06

2.67

2.06

Nitrotyrosine/tyrosine (ng/g)

0.009

0.02

0.007

0.01

Hydroxyproline/proline (ng/g)

0.0063

0.0034

0.006

0.003

H2O2/tyrosine (pM/g) pH

36.53 6

61.6 1.25

35.52 6

55.74 0.7

Fig. 1 Nitrate/tyrosine ratio in EBC at baseline for the welder and non-welder groups

Fig. 2 Aluminium relative to creatinine in urine for the welder and non-welder groups

cotinine concentration (751 § 157 g/l) than that of control subjects (343 § 170 g/l, p = 0.041). All other parameters (nitrite, MDA, nitrotyrosine, hydroxyproline, pH) showed no signiWcant diVerences. In addition, the welding-related metals were increased in the urine of the welders group: chromium/creatinine: 0.91 § 0.22 g/g versus 0.07 § 0.03 g/g, p = 0.001; nickel/creatinine: 1.82 § 0.32 g/g versus 0.73 § 0.10 g/g, p = 0.008; and aluminium/creatinine: 11.7 § 1.8 g/g versus 4.4 § 0.59 g/g, p < 0.001 (Fig. 2).

and cigarette pack-years, cotinine level, welding years, weekly welding exposure (hours) and welding exposure (hours) on Friday as metric covariates. It turned out that the nitrite/tyrosine ratio was dependent on the predominant welding technique (p = 0.043). The highest values for nitrite were found in the sub-population with MMA welding as predominant welding technique. The nitrate/tyrosine ratio and the nitrotyrosine/tyrosine ratio were dependent on the company aYliation index (p = 0.24, p = 0.008, respectively).

Factors inXuencing EBC and biomonitoring parameters Assessment of the intra shift eVect In order to investigate which parameters determine EBC and biomonitoring parameters, an analysis of variance was calculated. Therefore, a linear model was calculated with the logarithms of EBC and biomonitoring parameters as dependent variables, and the company aYliation index and predominant welding technique index as nominal factors,

In order to assess changes of the EBC and biomonitoring parameters during the work shift, the diVerence between the post-shift and pre-shift value of each parameter was calculated for each subject. The null hypothesis that this diVerence is zero was tested using Student t-Test. It turned out

123

808

Fig. 3 H2O2/tyrosine ratio in EBC in the welder sub-populations who smoked during shift and those who did not smoke

Int Arch Occup Environ Health (2010) 83:803–811

Fig. 4 Nitrate/tyrosine ratio in EBC in the welder sub-populations who used personal protection equipment and those who did not

that no signiWcant intra shift eVect could be detected for any of the endpoint parameters. InXuence of smoking After the working shift, subjects were asked if they had smoked during the work day. Subjects who smoked during work shift showed signiWcantly increased H2O2 concentrations after shift when compared to subjects who did not smoke (p = 0.019) (Fig. 3). H2O2 values before shift were similar in both groups. Although a similar trend was found between non-welders who smoked and non-welders who did not smoke, this trend failed statistical signiWcance. InXuence of personal protection equipment After the working shift, subjects were asked if they had used personal protection equipment during welding. Subjects who did not use such equipment showed signiWcantly increased nitrite/tyrosine ratios (p = 0.004), increased nitrate/tyrosine ratios (p < 0.001) (Fig. 4) and increased MDA/tyrosine ratios (p = 0.002) (Fig. 5). Welders who did not use any protection equipment during working shift showed increased nitrite/tyrosine ratios in EBC.

Discussion In this study, EBC and urine samples were collected in a population of male welders and non-welders to assess biological eVect markers in EBC and to monitor the inner

123

Fig. 5 MDA/tyrosine ratio in EBC in the welder sub-populations who used personal protection equipment and those who did not

exposure of welders to commonly used welding compounds. Measurements were taken at two diVerent timepoints: pre-work shift and post-work shift on Friday. The aim of this study was to explore whether diVerences in the various endpoint parameters can be detected in order to obtain information about inXammation, nitrosative and oxidative stress in welders, especially with respect to the inXuence of smoking habits and the use of protection equipment. The measurement of welding-related metals like aluminium, chromium, and nickel in the urine of welders in this study had signiWcantly increased concentrations for all of these parameters compared to controls at baseline.

Int Arch Occup Environ Health (2010) 83:803–811

However, the mean value of chromium was 1.40 g/l in the welders whereas the BAR threshold value [Biologischer ArbeitsstoV-Referenzwert recommended by the Deutsche Forschungsgemeinschaft (Deutsche Forschungsgemeinschaft 2009)] is 0.6 g/l for chromium that reXects exposure to chromium of the adult, non-exposed population. The EKA value (Expositionsäquivalente für krebserzeugende ArbeitsstoVe (DFG), exposure equivalents for carcinogenic substances) for chromium, however, is 15 g/l indicating that the internal chromium exposure of the welders in this study was only slightly higher than for an unexposed population. For nickel, a concentration of 2.8 g/l was found in welders which is slightly lower than the BAR threshold for nickel in urine: 3 g/l. The aluminium/creatinine ratio was 11.7 g/g in this study. BAT (Biologischer ArbeitsstoVToleranzwert recommended by DFG) for aluminium is 60 g/g creatinine in urine. Overall, it may be concluded that the internal baseline exposure of the subjects was rather low. Since the concentration of chromium, nickel and aluminium did not increase during the work shift under consideration, it may be concluded that welding fume exposure during this work shift did not contribute to the internal exposure to chromium, nickel and aluminium. It has also been shown that the ambient aerosol concentration was relatively low (inhalable dust 4.9 mg/m3, respirable dust 0.7 mg/m3), and consequently, the low internal exposure of the subjects was plausible. Nevertheless, welders had, even before shift, higher nitrate concentrations in the EBC when compared to non-welders. Nitrate concentration was furthermore signiWcantly dependent on the company aYliation and the nitrite concentration on the predominant welding technique. It turned out that welders who predominantly used the MMA welding technique showed the highest nitrite values in the EBC. These data suggest that welding inXuences nitrogen metabolism in the lungs even on a long-term scale, and it may be speculated that these changes may be associated with nitrosative stress. An acute intra shift eVect probably was not seen in this study because the exposure during that shift was relatively low. Both nitrate and MDA showed a signiWcant dependency on the company aYliation of the welders—adjusted for welding techniques. Although this Wnding could not be explained, it may be speculated that these diVerences are due to the local conditions at the workplaces such as ventilation and exhaustion. Welders showed a considerably higher cotinine concentration in urine than control subjects which could be explained by the higher number of smokers within the group of welders. Furthermore, a higher H2O2 concentration was found in the EBC of this group. Further analysis of the results showed that H2O2 was signiWcantly increased after shift in subjects who had smoked during shift in comparison with those who had not smoked during shift. Before

809

shift, both groups showed similar values of H2O2. This trend was also observed in control subjects, although in this group the diVerence did not reach statistical signiWcance. These results indicate that H2O2 may be inXuenced by cigarette smoking rather than by welding. Furthermore, the inXuence of using personal protection equipment was investigated. In subjects who did not use equipment such as a dust mask, we found signiWcantly increased concentrations of nitrite , nitrate and MDA in EBC after work shift, whereas subjects who used protection equipment showed signiWcantly lower values of these endpoints. These results indicate that, although it was not possible to detect an acute eVect of welding in the total welder population, an acute eVect can be identiWed in the sub-population who did not use personal protection equipment. Additionally, these results again emphasize the importance of personal protection equipment for welders. In general, analysis of biomarkers in EBC is complicated by a highly variable matrix. This phenomenon which is due to a variable dilution of the droplets generated in the respiratory tract by water vapour has already been described by other authors (EVros et al. 2002). Consequently, there is a strong need to correct the results of EBC analyses for the dilution. Gessner et al. (Gessner et al. 2001) pointed out that proteins may be a suitable reference substance to correct for dilution eVects. However, according to our previous investigations (Conventz et al. 2007), non-volatile substances like amino acids are well suited to correct for the degree of dilution of the lining Xuid in the collected EBCs. Consequently, we chose to determine the amino acids tyrosine and proline (which are the biochemical precursors of 3-nitrotyrosine and trans-hydroxyproline) as “internal standards” to normalize for all measured biomarker concentrations. The EBC biomarkers investigated in the present study part were nitrate, nitrite, malondialdehyde (MDA), hydroxyproline and nitrotyrosine. A very recent study reported the elevated levels of nitrate in EBC of patients with asthma compared to a control group and found good correlations between the concentrations of H2O2 and nitrite/ nitrate (Ueno et al. 2008). The authors concluded that these biomarkers are very suitable for the evaluation of airway inXammation. Several studies indicated that nitrate and nitrite concentrations are aVected by the exposure to exogeneous noxious substances: Murgia et al. (2006) have shown that nitrite was signiWcantly higher in subjects occupationally exposed to chromium when compared to control subjects. Balint et al. (2001) demonstrated that nitrate is increased by short-term smoking. Both nitrite and nitrotyrosine were not changed by smoking. Malondialdehyde (MDA) is generated within autoxidation of polyunsaturated fatty acids through reactive oxygen species. As these fatty acids are also part of cell

123

810

Int Arch Occup Environ Health (2010) 83:803–811

membranes, malondialdehyde is considered to be a suitable eVect biomarker of oxidative stress. Previous studies showed elevated MDA levels among chrome-plating workers that were highly correlated with H2O2 levels and chromium levels in the corresponding EBC (Caglieri et al. 2006). Another study showed the eVect of lung oxidative stress at high altitude on malondialdehyde levels in EBC (Araneda et al. 2005). Barregard et al. (2008) showed that air pollution (wood smoke) increases MDA levels in breath condensate and Romieu et al. (2008) found a relation of MDA levels with, both, ambient air pollution and lung function. So our results were in line with conclusions in the literature cited elsewhere.

Conclusion Although exposure conditions were quite heterogeneous and intensity of exposure was low, in this study, it has been shown that welding-associated long-term and short-term eVects could be detected in a population of welders using biomarker analyses in EBC. The results also showed that using personal protection equipment is of high importance in the welding profession and H2O2 may be an eVect marker associated with smoking rather than with welding fumes, while nitrate in EBC seems to be sensitive to welding fume exposure. Thus, EBC should be considered as a non-invasive and reliable method to assess welding fumeassociated health eVects for occupational medicine research. Acknowledgments The authors would like to thank Dr. Alice Mueller-Lux, MD, Dr. Angela Conventz, PhD, Anita Musiol, Karina Müller and Dr. Thomas Schettgen, PhD, for their assistance during the measurements in the companies and in the laboratory. The authors also thank Kathy Bischof for her review of and comments on the manuscript. This study was supported by the ‘Arbeitsgemeinschaft industrieller Forschungsvereinigungen ‘‘Otto von Guericke’’ e.V.’, AIF-Grant 14437. ConXict of interest statement conXict of interest.

The authors declare that they have no

References André E, Stoeger T, Takenaka S, Bahnweg M, Ritter B, Karg E, Lentner B, Reinhard C, Schulz H, Wjst M (2006) Inhalation of ultraWne carbon particles triggers biphasic pro-inXammatory response in the mouse lung. Eur Respir J 28:275–285 Andreoli R, Manini P, Corradi M, Mutti A, Niessen WM (2003) Determination of patterns of biologically relevant aldehydes in exhaled breath condensate of healthy subjects by liquid chromatography/ atmospheric chemical ionization tandem mass spectrometry. Rapid Commun Mass Spectrom 17:637–645 Antonini JM, Clarke RW, Krishna Murthy GG, Sreekanthan P, Jenkins N, Eagar TW, Brain JD (1998) Freshly generated stainless steel

123

welding fume induces greater lung inXammation in rats as compared to aged fume. Toxicol Lett 98:77–86 Araneda OF, Garcia C, Lagos N, Quiroga G, Cajigal J, Salazar MP, Behn C (2005) Lung oxidative stress as related to exercise and altitude. Lipid peroxidation evidence in exhaled breath condensate: a possible predictor of acute mountain sickness. Eur J Appl Physiol 95:383–390 Balint B, Donnelly LE, Hanazawa T, Kharitonov SA, Barnes PJ (2001) Increased nitric oxide metabolites in exhaled breath condensate after exposure to tobacco smoke. Thorax 56:456–461 Barregard L, Sallsten G, Andersson L, Almstrand AC, Gustafson P, Andersson M, Olin AC (2008) Experimental exposure to wood smoke: eVects on airway inXammation and oxidative stress. Occup Environ Med 65:319–324 Borm PJA (2004) Partikeltoxikologie: Vom Steinkohlenbergbau zur Nanotechnologie. Zbl Arbeitsmed 54:188–197 Boyce PD, Kim JY, Weissman DN, Hunt J, Christiani DC (2006) pH increase observed in exhaled breath condensate from welding fume exposure. J Occup Environ Med 48:353–356 Broding HC, Michalke B, Goen T, Drexler H (2009) Comparison between exhaled breath condensate analysis as a marker for cobalt and tungsten exposure and biomonitoring in workers of a hard metal alloy processing plant. Int Arch Occup Environ Health 82:565–573 Brown DM, Wilson MR, MacNee W, Stone V, Donaldson K (2001) Size-dependent proinXammatory eVects of ultraWne polystyrene particles: a role for surface area. Toxicol Appl Pharmacol 175:191–199 Caglieri A, Goldoni M, Acampa O, Andreoli R, Vettori M, Corradi M, Apostoli P, Mutti A (2006) The eVect of inhaled chromium on diVerent exhaled breath condensate biomarkers among chromeplating workers. Environ Health Perspect 114:542–546 Claeson K, Thorsen G, Karlberg B (2001) Methyl malondialdehyde as an internal standard for the determination of malondialdehyde. J Chromatogr B Biomed Sci Appl 751:315–323 Conventz A, Musiol A, Brodowsky C, Müller-Lux A, Dewes P, Kraus T, Schettgen T (2007) Simultaneous determination of 3-nitrotyrosine, tyrosine, hydroxyproline and proline in exhaled breath condensate by hydrophilic interaction liquid chromatography/ electrospray ionization tandem mass spectrometry. J Chromatogr B 860:78–85 Deutsche Forschungsgemeinschaft (2008) Analytische Methoden zur Prüfung gesundheitsschädlicher ArbeitsstoVe—Analysen in biologischem Material. Wiley-VCH, Weinheim Deutsche Forschungsgemeinschaft (2009) MAK- und BAT-WerteListe 2009. Wiley-VCH, Weinheim Dziedzic B, Mazanowska-Gajdowicz J, Walczewska A, Sarniak A, Nowak D (2003) Comparison of cadmium and enzyme-catalyzed nitrate reduction for determination of NO2¡/NO3¡ in breath condensate. Clin Chim Acta 335:65–74 EVros RM, Hoagland KW, Bosbous M, Castillo D, Foss B, Dunning M, Gare M, Lin W, Sun F (2002) Dilution of respiratory solutes in exhaled condensates. Am J Respir Crit Care Med 165:663–669 Gessner C, Kuhn H, Seyfarth HJ, Pankau H, Winkler J, Schauer J, Wirtz H (2001) Factors inXuencing breath condensate volume. Pneumologie 55:414–419 Gube M, Brand P, Conventz A, Ebel J, Goeen T, Holzinger K, Muller-Lux A, Reisgen U, Schettgen T, Kraus T (2009) Spirometry, impulse oscillometry and capnovolumetry in welders and healthy male subjects. Respir Med 103:1350–1357 Han SG, Kim Y, Kashon ML, Pack DL, Castranova V, Vallyathan V (2005) Correlates of oxidative stress and free-radical activity in serum from asymptomatic shipyard welders. Am J Respir Crit Care Med 172:1541–1548 Horváth I, Hunt J, Barnes P, Alving K, Antczak A, Baraldi E, Becher G, Beurden Mv, Corradi M, Dekhuijzen R, Dweik R, Dwyer T,

Int Arch Occup Environ Health (2010) 83:803–811 EVros R, Erzurum S, Gaston B, Gessner C, Greening A, Ho L, Hohlfeld J, Jöbsis Q, Laskowski D, Loukides S, Marlin D, Montuschi P, Olin A, Redington A, Reinhold P, Rensen Ev, Rubinstein I, SilkoV P, Toren K, Vass G, Vogelberg C, Wirtz H, Condensate AETFoEB (2005) Exhaled breath condensate: methodological recommendations and unresolved questions. Eur Respir J 26:523–548 Hurley J, Cherrie J, Donaldson K, Seaton A, Tran C (2003) Assessment of health eVects of long-term occupational exposure to tunnel dust in the London underground. Research Report TM/03/02 Kharitonov S, Barnes P (2001) Exhaled markers of pulmonary disease. Am J Respir Crit Care Med 163:1693–1722 Kim J, Chen J, Boyce P, Christiani D (2005) Exposure to welding fumes is associated with acute systemic inXammatory responses. Occup Environ Med 62:157–163 Kreyling WG, Semmler M, Müller W (2004) Health eVects of ultraWne particles. J Aerosol Sci 35:1155–1156 Murgia N, Muzi G, Dell’ Omo M, Montuschi P, Melchiorri D, Ciabattoni G, Abbritti EP, Orazi N, Sapia IE, Abbritti G (2006) Induced sputum, exhaled breath condensate and nasal lavage Xuid in electroplating workers exposed to chromium. Int J Immunopathol Pharmacol 19:67–71 Oberdörster G, Ferin J, Finkelstein G, Wade P, Corson N (1990) Increased pulmonary toxicity of ultraWne particles? II. Lung lavage studies. J Aerosol Sci 21:384–387 Oberdörster G, Gelein RM, Ferin J, Weiss B (1995) Association of particulate air pollution and acute mortality: involvement of ultraWne particles? Inhal Toxicol 7:111–124

811 Palmer KT, McNeill Love RM, Poole JR, Coggon D, Frew AJ, Linaker CH, Shute JK (2006) No inXammatory responses to the occupational inhalation of metal fume. Eur Respir J 27:366–373 Romieu I, Barraza-Villarreal A, Escamilla-Nunez C, Almstrand AC, Diaz-Sanchez D, Sly PD, Olin AC (2008) Exhaled breath malondialdehyde as a marker of eVect of exposure to air pollution in children with asthma. J Allergy Clin Immunol 121: 903–9 e6 Scharrer E, Hessel H, Kronseder A, Guth W, Rolinski B, Jorres RA, Radon K, Schierl R, Angerer P, Nowak D (2007) Heart rate variability, hemostatic and acute inXammatory blood parameters in healthy adults after short-term exposure to welding fume. Int Arch Occup Environ Health 80:265–272 Seaton A, Godden D, MacNee W, Donaldson K (1995) Particulate air pollution and acute health eVects. Lancet 345:176–178 Sjogren B, Gyntelberg F, Hilt B (2006) Ischemic heart disease and welding in Scandinavian studies. Scand J Work Environ Health Suppl 2:50–53 Sjögren B, Fossum T, Lindh T, Weiner J (2002) Welding and ischemic heart disease. Int J Occup Environ Health 8:309–311 Ueno T, Kataoka M, Hirano A, Iio K, Tanimoto Y, Kanehiro A, Okada C, Soda R, Takahashi K, Tanimoto M (2008) InXammatory markers in exhaled breath condensate from patients with asthma. Respirology 13:654–663 Wiebert P, Sanchez-Crespo A, Seitz J, Falk R, Philipson K, Kreyling W, Möller W, Sommerer K, Larsson S, Svartengren M (2006) Negligible clearance of ultraWne particles retained in healthy and aVected human lungs. Eur Respir J 28:286–290

123