Metallic elements in exhaled breath condensate and serum of patients with exacerbation of chronic obstructive pulmonary disease

Abstract

Biomarkers in exacerbated chronic obstructive pulmonary disease may be useful in aiding diagnosis, defining specific phenotypes of disease, monitoring the disease and evaluating the effects of drugs. The aim of this study was the characterization of metallic elements in exhaled breath condensate and serum as novel biomarkers of exposure and susceptibility in exacerbated chronic obstructive pulmonary disease using reference analytical techniques. C-Reactive protein and procalcitonin were assessed as previously validated diagnostic and prognostic biomarkers which have been associated with disease exacerbation, thus useful as a basis of comparison with metal levels. Exhaled breath condensate and serum were obtained in 28 patients at the beginning of an episode of disease exacerbation and when they recovered. Trace elements and toxic metals were measured by inductively coupled plasma-mass spectrometry. Serum biomarkers were measured by immunoassay . Exhaled manganese and magnesium levels were influenced by exacerbation of chronic obstructive pulmonary disease, an increase in their concentrations—respectively by 20 and 50%—being observed at exacerbation in comparison with values obtained at recovery; serum elemental composition was not modified by exacerbation; serum levels of C-reactive protein and procalcitonin at exacerbation were higher than values at recovery. In outpatients who experienced a mild–moderate chronic obstructive pulmonary disease exacerbation, manganese and magnesium levels in exhaled breath condensate are elevated at admission in comparison with values at recovery, whereas no other changes were observed in metallic elements at both the pulmonary and systemic level.

---

Introduction

Chronic obstructive pulmonary disease (COPD) is a preventable and partially treatable disease with some significant extra pulmonary effects that may contribute to severity in individual patients. Its pulmonary component is characterized by airflow limitation that is not fully reversible. The airflow limitation is usually progressive and associated with an abnormal inflammatory response of the lung to noxious particles or gases.¹

The clinical course of COPD is frequently aggravated by episodes of acute worsening of respiratory symptoms with an increase in airflow obstruction and air trapping, which usually requires additional therapies.^1,2 COPD exacerbations accelerate the progressive decline in lung function and are important causes of morbidity and mortality associated with the disease.^3–5 COPD exacerbations are caused mainly by respiratory tract infections, but triggering factors also include non-infective causes, such as exposure to environmental pollutants. However, the cause of exacerbations cannot be identified in approximately one-third of all cases.^1,2 The heterogeneity of COPD exacerbations, due to their range of symptoms and ambiguous aetiology makes them difficult to define, classify and manage.⁶

Biomarkers with potential utility in the diagnosis and prognosis of COPD exacerbation are needed.^7–12

Exhaled breath condensate (EBC) obtained by cooling exhaled air, is a biological matrix representative of the composition of airway lining fluid, suitable to assess airway inflammation^13–15 and exposure to metallic elements polluting the general and working environment.^16–18 EBC collection is simple to perform and can be repeated several times without affecting airway function or inflammation.

Recently, we applied the elemental analysis of EBC to assess the target tissue dose of pneumotoxic metals and transition elements involved in redox systems implicated in the control of oxidative stress. We also proposed such an approach to develop new biomarkers of exposure and susceptibility in COPD patients.¹⁹ In EBC of patients with stable COPD, we found higher levels of toxic elements (lead, cadmium and aluminium) and lower levels of essential transition elements (copper, iron) compared to nonsmoking subjects.¹⁹ Recently, lower levels of iron have also been found by other authors in children with asthma compared to healthy controls.²⁰

The evaluation of metallic elements during COPD exacerbation may improve the knowledge of pathophysiological pulmonary mechanisms associated with this clinical condition. The working hypothesis is that trace elements have either activating or inhibiting roles in the defence systems of the respiratory tract (thus possibly a promoting role in COPD exacerbation), whereas toxic metals contained in tobacco smoke and in polluted ambient air may trigger mechanisms leading to COPD exacerbation. We also looked at blood biomarkers , such as C-reactive protein (CRP) and procalcitonin (PCT), as systemic biomarkers used for exacerbated COPD management.^7,10,11,21

Experimental

This study is part of a collaborative project entitled “Metals in exhaled breath condensate as COPD biomarkers ” supported by the National Heart, Blood and Lung Institute (grant R01 HL72323). The study was approved by both the University of Parma Ethics Committee and the Fallon Clinic Institutional Review Board; each subject was informed about the study protocol and signed a written consent agreement.

Study populations

Twenty-eight COPD outpatients were recruited at the Fallon Clinic Research Department (Worcester, MA) to take part in this study. Patients’ characteristics are shown in Table 1.

Table 1 Characteristics of COPD patients. Data are shown as no., mean ±SD, median (25–75%)

Male/female		14/14
a Lung function test performed after COPD exacerbation, in stable phase. b Type of COPD exacerbation according to Anthonisen criteria: type I = increased dyspnea, sputum volume and sputum purulence; type II = two of the above; type III = one of the above in addition to at least one of the following: upper respiratory infection within the past 5 days, fever without other causes, increased wheezing or cough, increase in respiratory rate or heart rate by 20% as compared with baseline. c p < 0.05 A vs. B.
Age/years		70.4 ± 8.4
Smokers/ex-smokers/nonsmokers		4/23/1
Pack/years (smokers and ex-smokers)		63.6 ± 43.5
GOLD stage^a	I (FEV1 ≥80% of pred.)	0
	II (80%≤FEV1 < 50% of pred.)	4
	III (50%≤FEV1 < 30% of pred.)	13
	IV (FEV1≤30% of pred.)	11
Anthonisen criteria^b	Type I	12
	Type II	5
	Type III	9

---

	Exacerbation (A)	Recovery (B)
FVC/l	1.84 ± 0.64	1.91 ± 0.43
FVC (%) predicted	53.81 ± 17.24	55.82 ± 4.88
FEV1/l	0.79 ± 0.24	0.89 ± 0.23^c
FEV1 (%) predicted	32.11 ± 11.97	35.25 ± 12.73
FEV1/FVC	0.44 ± 0.10	0.47 ± 0.12
FEF25–75% /l s^−1	0.29 ± 0.12	0.38 ± 0.23
FEF25–75% (%) predicted	15.42 ± 7.79	19.11 ± 13.76

---

The diagnosis of COPD was established according the global initiative for chronic obstructive lung disease (GOLD) criteria.¹

Tobacco smoke exposure was evaluated in terms of both self-reported current smoking status and serum cotinine levels. Patients who had stopped smoking for at least 1 year before recruitment were defined as ex-smokers.

Study design

COPD patients were assessed twice: at the beginning of an episode of COPD exacerbation and when they recovered. COPD exacerbation was defined as an acute worsening of respiratory symptoms (dyspnoea, cough, change in amount and purulence of sputum), that was beyond normal day-to-day variations of symptoms.¹ Patients were considered recovered from COPD exacerbation when their symptoms/signs significantly improved and additional therapies were stopped. In both cases, the diagnosis of COPD exacerbation and its recovery was confirmed by pulmonary physicians at the clinic. At the both times, patients came in for a clinic visit which included answering a medical questionnaire, spirometry, blood and EBC collection. COPD exacerbation was classified following the Anthonisen criteria based on the presence and different combination of three specific symptoms (dyspnea, sputum volume and purulence).²²

EBC collection and analysis

EBC was collected using a portable condenser (TURBO-DECCS; All Service; Parma, Italy) at collecting temperature of −5 °C as previously described.^19,23 Each subject was asked to breathe tidally for 15 min, through a two-way non re-breathing valve by which inspiratory and expiratory air is separated, and saliva was trapped. EBC samples were stored at −80 °C until transport on dry ice to the laboratory of Industrial Hygiene and Toxicology, University Hospital of Brescia (Italy), a laboratory certified for analysis of metallic elements, whose concentration in EBC was determined by inductively coupled plasma-mass spectrometry (ICP-MS ELAN 5000, Perkin Elmer, Wellesley, MA) as previously described.¹⁹ All disposable collection and analytical plastics were tested and accurately chosen to neither release nor adsorb any metallic element. The limits of detection (LOD, calculated as 3 SDs of blank) for all the tested metallic elements in EBC have been previously reported.¹⁹ In the case of Mn, NIST-certified standards containing 0.38 μg L⁻¹ give a signal intensity of about 5800 ions. Considering that an adequate level of accuracy may be maintained when the number of ions is greater than 150–200, the limit of quantification (LOQ) calculated as 3 LODs for Mn corresponds to 0.015 μg L⁻¹, well below the concentration we measured in EBC samples (0.07—0.4 μg L⁻¹). The LOQ value reported in literature for the same element in urine samples is 0.03 μg L⁻¹.²⁴ LODs and LOQs found in EBC were lower than that reported for urine by at least a factor of two,²⁴ probably because EBC is practically pure water without any matrix effect. The LOD for Mg was 0.5 μg L⁻¹.

Blood analysis

Serum obtained after centrifugation of venous blood samples were stored at −20 °C until transport in dry ice to the laboratory of Brescia (Italy) for metal analysis and to the laboratory of Parma (Italy) for protein and cotinine analysis. Whole blood was collected in BD Vacutainer® blood collection tubes, that were tested for trace element analyses (BD, Franklin Lakes, NJ). Moreover, when the tubes were treated as the samples, the release of metallic elements was always negligible. Metallic elements were determined using ICP-MS (ELAN 5000; Perkin Elmer; Wellesley, MA) with the method of Goullè et al. for plasma with almost the same LODs and LOQs,²⁴ except Mg, determined by flame photometry in a certified laboratory of Clinical Chemistry (Dept. of Pathology and Medicine of Laboratory, University Hospital of Parma, Italy). CRP was measured by nephelometry (BN 100, Dade Behring, Marburg), with a limit of detection of 0.175 mg l⁻¹. PCT was measured using 20 to 50 μl of serum by a time-resolved amplified cryptate emission technology assay (PCT sensitive Kryptor; B·R·A·H·M·S Hennigsdorf, Germany). The assay has a functional sensitivity of 0.06 μg l⁻¹ and LOD 0.02 μg l⁻¹.

Free cotinine was determined by liquid chromatography tandem mass spectrometry as previously described.²⁵

Spirometry

Spirometry was performed with a spirometer EasyOne™ (ndd. Medizintechnik, Zurich, Switzerland). The results of forced vital capacity (FVC), forced expiratory volume in the first second (FEV₁), ratio of FEV₁ to FVC and forced expiratory flow 25–75% (FEF _25–75%) are summarized in Table 1. Predicted values were assigned following the NHANES III criteria.²⁶

Statistics

Statistical analyses were performed using two software packages: Prism 4 (GraphPad; San Diego, CA) and SPSS 14.0 (SPSS; Chicago IL).

The study was originally planned assuming a two-sided SD exceeding the mean difference by 50% and to achieve a power of 95 and 80% for type I and type II errors, respectively. However, most data failed to meet the criteria for normality of the distribution using the Kolmogorov–Smirnov test. Thus, statistical analysis was based on non-parametric tests.

Comparisons between exacerbation and recovery were assessed using paired-sample Wilcoxon test for each of the tested variables. Comparison among groups was performed using Kruskal–Wallis test followed by the Dunn’s test for multiple comparisons. Correlations between variables were assessed using Spearman rank correlation. A significance level of 0.05 was chosen for all of statistical tests.

Results

Patients

All patients completed visits both at time of exacerbation and at a later time after recovery. There were no adverse effects to the study procedures. The time period between the first and second evaluation was 28 days (median value).

COPD exacerbations needed additional medical therapy for all patients, mainly inhaled bronchodilators, oral corticosteroids and antibiotics; the patients were treated at home. Following Anthonisen’s criteria of COPD exacerbation, 12 patients had type I, 5 subjects type II and 9 type III exacerbation. Two subjects were not classifiable according to Anthonisen criteria: in one case for absence of symptomatology information and in the other because only dispnea was mentioned.

Four subjects had serum cotinine values above 15 μg l⁻¹ (considered as cut-off point to differentiate non smokers from current smokers)^27,28 thus were considered current smokers, in agreement with data from the questionnaire. There were twenty-three ex-smokers and the reported number of years since stopping was 15.8 years (SD 11.8).

Lung function

Spirometric values are summarized in Table 1. Among the spirometric data, only FEV₁ values were lower to a statistically significant degree at admission in comparison with values at recovery (p = 0.02). FVC and FEF₂₅₋₇₅ levels were lower at admission, but without statistical significance.

EBC

Metallic elements in EBC are summarized in Table 2. Manganese (Mn) and magnesium (Mg) levels were elevated at admission in comparison with values at recovery, whereas no other significant variation in metallic element concentration was evident for any other metals. EBC Mn and Mg levels were positively correlated at admission (Fig. 1) and Mn negatively correlated with FEV₁ (Fig. 2) (the correlation between Mg and FEV₁ was close to statistical significance, data not shown). EBC metal concentrations seemed not affected by Anthonisen criteria of COPD exacerbation or by severity stage of disease classified in according to GOLD guidelines, although the number of patients were very small.

Fig. 1 Correlation between Mn and Mg levels in EBC samples collected at admission for exacerbation (A).

Fig. 2 Correlations between FEV₁ leves and EBC-Mn values assessed at the beginning of exacerbation (A).

Table 2 Metal levels in EBC and serum at exacerbation and recovery. Data are expressed as median (25–75th percentile). ND: not detectable; NM: not measured. *p < 0.05 exacerbation vs. recovery

Metallic elements/μg L^−1	Exacerbation	Recovery
a Levels measured in 18 out of 28 COPD patients. b Levels measured in 22 out of 28 COPD patients.
EBC
Al	1.45 (0.68–2.58)	1.00 (0.63–2.28)
Pb	0.10 (0.07–0.20)	0.17 (0.04–0.20)
Cd	ND (ND-0.048)	ND (ND-0.014)
Ni	0.20 (0.10–0.40)	0.20 (0.05–0.47)
Mn	0.25 (0.09–0.40)	0.20 (0.07–0.28)*
Se	0.85 (0.40–1.95)	1.15 (0.55–1.95)
Cu	0.37 (ND-0.78)	0.30 (ND-0.79)
Fe	0.85 (0.50–1.40)	0.70 (0.40–1.92)
Mg	9.44 (3.08–13.81)	5.06 (2.59–8.90)*
Serum
Al	NM	NM
Pb	0.41 (0.11–1.03)	0.40 (0.19–0.82)
Cd	0.37 (0.05–0.98)	0.28 (0.09–0.65)
Ni	0.70 (0.32–1.19)	0.70 (0.30–0.94)
Mn	1.39 (0.90–1.80)	1.50 (0.99–2.08)
Se	129 (122–139)	137 (115.5–149)
Cu	NM	NM
Fe	NM	NM
Mg/mg L^−1	22 (19–24)^a	20 (18–25)^b

---

Blood

No metal showed changes during the course of COPD exacerbation or was influenced by Anthonisen criteria of COPD exacerbation or the severity stage of COPD, classified according to GOLD guidelines.

At admission, PCT and CRP values were significantly higher than values measured at recovery [0.09(0.07–0.13) μg L⁻¹vs. 0.08(0.06–0.11) μg L⁻¹ for PCT, 11.6(3.5–29.5) mg L⁻¹vs. 3.7(1.6–8.8) mg L⁻¹ for PCR, p < 0.05 in both cases]. Neither protein concentrations were associated with Anthonisen criteria, whereas subdividing the patients according to the severity stage of COPD, the highest values were observed in those COPD patients with lower FEV₁ ad admission, reaching statistical significance for PCT only (Fig. 3).

Fig. 3 Serum levels of CRP (a) and PCT (b) at the beginning of exacerbation (A). The COPD patients were classified on the basis of severity stage of disease in stage II (n = 4) stage III (n = 13) and stage IV (n = 11). Symbols and errors bars represent the median and the interquartile range.

A positive correlation was found between serum levels of CRP and PCT at both times of sampling (Fig. 4).

Fig. 4 Correlation between CRP and PCT, at the beginning (a) and after COPD exacerbation (b). Serum levels of CRP and PCT were measured at admission for exacerbation (A) and at after recovery (B).

Discussion

The study had three important findings: (1) Mn and Mg levels in EBC are influenced by COPD exacerbation, as an increase in their concentrations was observed at exacerbation in comparison with values obtained at recovery; (2) serum elemental composition was not modified by COPD exacerbation; (3) serum levels of CRP and PCT at exacerbation were higher than values at recovery, though only mild changes were observed in these well established markers of exacerbation, thus excluding bacterial infection as a cause of exacerbations, which were all treated at home.

EBC elements

In this study, the focus was on possible changes in patterns of exhaled metallic elements associated with COPD exacerbation. Most elements didn’t change at all, despite their deviation from values observed in healthy subjects. For example, Fe and Cu in EBC of COPD patients were similar to those already observed in another series of COPD patients and in smokers, who showed an important decrease (by an order of magnitude in the case of Fe) as compared to healthy controls.¹⁹

Mn and Mg—the latter not examined in previous studies—were significantly higher during acute illness than after recovery. In particular, Mn level showed a gradual reduction in its concentration in 16 patients, whereas no changes were seen in 9 patients, and a small increase in 3 patients. A progressive reduction in Mg levels was observed in 16 patients whereas an opposite trend was noted in the remaining 12 patients. Interestingly, all 16 patients showing a reduction of Mn also showed a reduction in Mg concentrations. In addition, these elements were positively correlated with each other and both were negatively correlated with FEV₁ at admission.

The concordant change in the levels of Mn and Mg could suggest either similar biological actions or a similar body burden due to common polluting sources.

In this regard, both elements have been reported to be involved in respiratory homeostasis: Mn is a co-factor of mitochondrial superoxide dismutase (SOD), thereby playing a detoxifying role against oxidative lung injury.²⁹ In exacerbated COPD patients, elevated EBC Mn levels might mirror an enhanced activity of mitochondrial SOD. This is in line with a study of Sadowska et al., showing an increase in lung antioxidant defences during acute exacerbation of COPD.³⁰

Mg plays an important role in airway smooth-muscle relaxation and bronchodilation, stabilization of mast cells, neurohumoral mediator release and mucociliary clearance.^31,32 We speculate that elevated Mg levels observed during exacerbation could reflect an attempt to increase lung concentration of this element, useful to counterbalance obstruction mechanisms associated with exacerbation. Relying on its effects on excitable membranes of smooth muscles, Mg supplementation has been proposed in the management of exacerbated COPD patients, with some positive results.^33,34

We noted a discrepancy between EBC and serum levels, which however is not inconsistent with the literature: in fact, systemic administration of Mg i.v. was found to be effective both in acute asthma and COPD, whereas its local administration as aerosol was not, thus suggesting that lining-fluid and serum may not be in equilibrium for certain components that are metabolically active at a given site.³⁵ Furthermore, the elevation of Mg lung burden at COPD exacerbation may be associated with other meaningful physiological effects, such as the reduction of lung hyperinflation. Amaral et al.³⁶ noted that treatment with Mg was associated with a significant decrease in residual volume.

Mg and Mn concentrations may reflect a physiological response to the acute phase of the diseases. Interestingly, electron paramagnetic resonance (EPR) studies showed that Mn²⁺ can often replace Mg²⁺ in biological functions such as nucleic-acid processing enzymes.³⁷

We are aware that information obtained by means of elemental analysis may be ambiguous, not least because certain transition elements can be essential or toxic depending on their valence, physical state and burden. We can not exclude that Mg and Mn levels could be altered by ambient Mg and Mn sources,³⁸ although this explanation is inconsistent with the observed data trend. In fact, the concentration of other toxic metallic elements—contained in tobacco smoke and in polluted environments—did not change during the course of exacerbation. On the contrary, we noted a good short term reproducibility of other exposure biomarker levels at the same sampling times when Mg and Mn levels did change. Such changes in both elements during exacerbation probably indicate an endogenous origin.

The levels of all other elements in EBC samples were not affected by the severity stage of COPD classified in according to GOLD guidelines or on the Anthonisen criteria of COPD exacerbation. The observed increase in Mn levels in EBC of exacerbated COPD patients is also in line with our previous manuscript on clinically stable COPD patients:¹⁹ in fact, among different groups of patients, we showed that the highest range in Mn levels was observed in COPD patients. We speculate that those COPD patients with highest Mn levels in EBC are more prone to develop an exacerbation. Further longitudinal studies are warranted to deal with this issue.

Metallic elements in blood

We did not detect any significant change of elemental metals during course of COPD exacerbation, their levels being similar to reference levels.^39–41 A hypothesis accounting for the observed normal serum values is the mild–moderate degree of exacerbation that our patients experienced.

Whereas Mn show an opposite trend in serum as compared to EBC, compatible with a different meaning in the two compartments (e.g., supply vs.excretion), changes in serum Mg concentration follow the same direction of EBC, but are an order of magnitude higher than corresponding EBC levels. In this case, it is conceivable that short term changes in a peripheral organ are not reflected by systemic circulation.

There are few published data on the serum elemental metal composition during COPD exacerbation. In a retrospective study,⁴² lower Mg serum concentrations were found in patients undergoing an exacerbation in comparison with COPD stable patients; however, in agreement with our data, serum Mg concentrations were found to be within the reference interval.

Serum proteins

PCT and CRP were assessed as previously validated diagnostic and prognostic biomarkers which have been associated with COPD exacerbation, thus useful as a basis of comparison with EBC and serum metal levels.

The PCT values we observed, although slightly higher at admission in comparison with values at recovery,^11,21 were within the reference interval from our laboratory (0–0.5 μg l⁻¹); moreover, most PCT values (26 out of 28) were below 0.25 μg l⁻¹, that is a suggested possible cut-off for the presence of bacterial infection in COPD exacerbated patients for which antibiotic treatment is encouraged.²¹ In agreement with previous studies,^10,11,43 CRP levels were significantly elevated at the time of exacerbation and decreased substantially thereafter. In nearly 50% of the study patients at the time of exacerbation, CRP values were within the reference interval (<10 mg l⁻¹) as reported in several studies;^10,42 only 4 patients had CRP levels higher than 50 mg l⁻¹, which has been reported as a cut-off value predictive of significant bacterial infection in lower respiratory tract infection.⁴⁴

Limitations

A limitation of the present study could be represented by poor control on potential confounders, such as time spent outdoors, and diet, which are difficult to record over a relative long period if time. Furthermore, we acknowledge the relatively low number of exacerbated COPD patients as a potential limitation of our study, though repeated measures within the same subjects were compared, thereby increasing the statistical power. However, the assessment of actual statistical power of the present study was impossible, due to uneven and mostly non-normal distribution of elemental concentrations in biological media used in this study, and subsequent use of non parametric tests. Unfortunately, to the best of our knowledge, non-parametric tests are not available to assess statistical power.

Conclusion

Our results show that in outpatients experiencing a mild–moderate COPD exacerbation, Mn and Mg levels in EBC are elevated compared with values at recovery. No other metallic elements showed significant changes during the period of COPD exacerbation, thereby indirectly confirming the validity and reproducibility of EBC elemental characterization, both in terms of burden of toxic elements and depletion of oxidative stress and inflammation-associated elements. Serum levels, including those of Mn and Mg, were unchanged during COPD exacerbation. Interestingly, these elements share properties and biological functions, in which they are often interchangeable. The validity of EBC Mn and Mg levels as biomarkers of susceptibility in patients with COPD exacerbation is still unknown and requires further investigation.

Competing interests

The authors declare that they have no financial or non-financial competing interests. The academic salary of Susan R. Sama is funded by various grants from NIH. Donald K. Milton receives NIH and CDC grants.

Acknowledgements

This study was supported by National Heart, Blood and Lung Institute, grant R01 HL72323. We would like to acknowledge the Research Nurses Anne McDonald and Kathleen Allain, who conducted the fieldwork for this study. We are grateful for their dedication and hard work. We would also like to thank Margalit Lai and Christine Andersson for their assistance in this work.

References

1. K. F. Rabe, S. Hurd, A. Anzueto, P. J. Barnes, S. A. Buist, P. Calverley, Y. Fukuchi, C. Jenkins, R. Rodriguez-Roisin, C. van Weel and J. Zielinski, Am. J. Respir. Crit. Care Med., 2007, 176, 532–555.
2. B. R. Celli and P. J. Barnes, Eur. Respir. J., 2007, 29, 1224–1238.
3. J. A. Wedzicha, Chest, 2002, 121, 136S–141S.
4. G. C. Donaldson, T. A. Seemungal, A. Bhowmik and J. A. Wedzicha, Thorax, 2002, 57, 847–852.
5. T. A. Seemungal, G. C. Donaldson, A. Bhowmik, D. J. Jeffries and J. A. Wedzicha, Am. J. Respir. Crit. Care Med., 2000, 161, 1608–1613.
6. L. G. Franciosi, C. P. Page, B. R. Celli, M. Cazzola, M. J. Walzer, M. Danhof, K. F. Rabe and O. E. Della Pasqua, Respir. Res., 2006, 7, 74.
7. J. R. Hurst, G. C. Donaldson, W. R. Perera, T. M. Wilkinson, J. A. Bilello, G. W. Hagan, R. S. Vessey and J. A. Wedzicha, Am. J. Respir. Crit. Care Med., 2006, 174, 867–874.
8. S. Bozinovski, A. Hutchinson, M. Thompson, L. Macgregor, J. Black, E. Giannakis, A. S. Karlsson, R. Silvestrini, D. Smallwood, R. Vlahos, L. B. Irving and G. P. Anderson, Am. J. Respir. Crit. Care Med., 2008, 177, 269–278.
9. W. B. Gerritsen, J. Asin, P. Zanen, J. M. van den Bosch and F. J. Haas, Respir. Med., 2005, 99, 84–90.
10. N. Weis and T. Almdal, Eur. J. Intern. Med., 2006, 17, 88–91.
11. D. Stolz, M. Christ-Crain, N. G. Morgenthaler, J. Leuppi, D. Miedinger, R. Bingisser, C. Müller, J. Struck, B. Müller, M. Tamm and M. Copeptin, Chest, 2007, 131, 1058–1067.
12. M. Cazzola, W. MacNee, F. J. Martinez, K. F. Rabe, L. G. Franciosi, P. J. Barnes, V. Brusasco, P. S. Burge, P. M. Calverley, B. R. Celli, P. W. Jones, D. A. Mahler, B. Make, M. Miravitlles, C. P. Page, P. Palange, D. Parr, M. Pistolesi, S. I. Rennard, M. P. Rutten-van Mölken, R. Stockley, S. D. Sullivan, J. A. Wedzicha and E. F. Wouters, on behalf of the American Thoracic Society/European Respiratory Society Task Force on outcomes of COPD, Eur. Respir. J., 2008, 31, 416–469.
13. S. A. Kharitonov and P. J. Barnes, Chest, 2006, 130, 1541–1546.
14. A. Antczak and P. Górski, Int. J. Occup. Med. Environ. Health, 2002, 15, 317–323.
15. P. Montuschi, Clin. Chim. Acta, 2005, 356, 22–34.
16. M. Goldoni, S. Catalani, G. De Palma, P. Manini, O. Acampa, M. Corradi, R. Bergonzi, P. Apostoli and A. Mutti, Environ. Health Perspect., 2004, 112, 1293–1298.
17. M. Goldoni, A. Caglieri, D. Poli, M. V. Vettori, M. Corradi, P. Apostoli and A. Mutti, Anal. Chim. Acta, 2006, 562, 229–235.
18. A. Caglieri, M. Goldoni, O. Acampa, R. Andreoli, M. V. Vettori, M. Corradi, P. Apostoli and A. Mutti, Environ. Health Perspect., 2006, 114, 542–546.
19. A. Mutti, M. Corradi, M. Goldoni, M. V. Vettori, A. Bernard and P. Apostoli, Chest, 2006, 129, 1288–1297.
20. Z. Vlasic, S. Dodig, I. Cepelak, R. Z. Topic, J. Zivcic, B. Nogalo and M. Turkalj, J. Asthma, 2009, 46, 81–85.
21. D. Stolz, M. Christ-Crain, R. Bingisser, J. Leuppi, D. Miedinger, C. Muller, P. Huber, B. Muller and M. Tamm, Chest, 2007, 131, 9–19.
22. N. R. Anthonisen, J. Manfreda, C. P. Warren, E. S. Hershfield, G. K. Harding and N. A. Nelson, Ann. Intern. Med., 1987, 106, 196–204.
23. M. Goldoni, A. Caglieri, R. Andreoli, D. Poli, P. Manini, M. V. Vettori, M. Corradi and A. Mutti, BMC Pulm. Med., 2005, 5, 10.
24. J. P. Goullè, L. Mahieu, J. Castermant, N. Neveu, L. Bonneau, G. Lainè, D. Bouige and C. Lacroix, For. Sci. Int., 2005, 153, 39–44.
25. P. Manini, G. De Palma, R. Andreoli, D. Poli, M. Petyx, M. Corradi, A. Mutti and P. Apostoli, Toxicol. Lett., 2008, 181, 25–30.
26. J. L. Hankinson, J. R. Odencrantz and K. B. Fedan, Am. J. Respir. Crit. Care Med., 1999, 159, 179–87.
27. F. Seccareccia, P. Zuccaro, R. Pacifici, P. Meli, F. Pannozzo, K. M. Freeman, A. Santaquilani and S. Giampaoli, Eur. J. Epidemiol., 2003, 18, 487–492.
28. V. Binnie, S. McHug, L. Macpherson, B. Borland, K. Moir and K. Malik, Oral Dis., 2004, 10, 287–293.
29. I. Rahman, S. K. Biswas and A. Kode, Eur. J. Pharmacol., 2006, 533, 222–239.
30. A. M. Sadowska, C. Luyten, A. M. Vints, J. Verbraecken, D. V. Ranst and W. A. De Backer, Respirology, 2006, 11, 741–747.
31. K. I. Gourgoulianis, G. Chatziparasidis, A. Chatziefthimiou and P. A. Molyvdas, J. Aerosol Med., 2001, 14, 301–307.
32. J. Britton, I. Pavord, K. Richards, A. Wisniewski, A. Knox, S. Lewis, A. Tattersfield and S. Weiss, Lancet, 1994, 344, 357–62.
33. M. S. Skorodin, M. F. Tenholder, B. Yetter, K. A. Owen, R. F. Waller, S. Khandelwahl, K. Maki, T. Rohail and N. D'Alfonso, Arch. Intern. Med., 1995, 155, 496–500.
34. J. Abreu González, C. Hernández García, P. Abreu González, C. Martín García and A. Jiménez, Arch. Bronconeumol., 2006, 42, 384–387.
35. B. H. Rowe and C. A. Camargo, Curr. Opin. Pulm. Med., 2008, 14(1), 70–76.
36. A. F. do Amaral, A. L. Rodrigues, J. T. Filho and V. Vannucchi, JAB Martines, Med. Sci. Monit., 2008, 14(10), CR524–529.
37. C. Sissi and M. Palumbo, Nucleic Acids Res., 2009, 37, 702–711.
38. S. Tubek and S. Tubek, Biol. Trace Elem. Res., 2008, 123, 270–276.
39. Centers for Disease Control and Prevention. Third National Report on Human Exposure to Environmental Chemicals.2005, CDC, Atlanta, GA. NCEH Pub. No. 05-0570.
40. http://www.cdc.gov/exposurereport/pdf/thirdreport.pdf .
41. V. Lyengar and J. Woittiez, Clin. Chem., 1988, 34, 474–81.
42. C. Minoia, E. Sabbioni, P. Apostoli, R. Pietra, L. Pozzoli, M. Gallorini, G. Nicolaou, L. Alessio and E. Capodoglio, Sci. Total Environ., 1990, 95, 89–105.
43. H. S. Aziz, A. I. Blamoun, M. K. Shubair, M. M. Ismail, V. A. Debari and M. A. Khan, Ann. Clin. Lab. Sci., 2005, 35, 423–427.
44. D. Dev, E. Wallace, R. Sankaran, J. Cunniffe, J. R. W. Govan, C. G. Wathen and X. S. Emmanuel, Respir. Med., 1998, 92, 664–667.
45. W. R. Perera, J. R. Hurst, T. M. A. Wilkinson, R. J. Sapsford, H. Mullerova, G. C. Donaldson and J. A. Wedzicha, Eur. Respir. J., 2007, 29, 527–534.

---