association between cytokines in induced sputum and severity of chronic obstructive pulmonary...

ARTICLE IN PRESS

Respiratory Medicine (2006) 100, 846–854

KEYWORDCOPD;ExacerbatInduced spIL-6;IL-8;TNF-a

0954-6111/$ - sdoi:10.1016/j.r

�CorrespondE-mail addr

Association between cytokines in inducedsputum and severity of chronic obstructivepulmonary disease

Suleyman S. Hacievliyagila,�, Hakan Gunena, Levent C. Mutlua,Aysun B. Karabulutb, Ismail Temelb

aDepartment of Pulmonary Medicine, Inonu University, Turgut Ozal Medical Centre, Malatya, TurkeybDepartment of Biochemistry, Inonu University, Turgut Ozal Medical Center, Malatya, Turkey

Received 23 March 2005; accepted 30 August 2005

S

ion;utum;

ee front matter & 2005med.2005.08.022

ing author. Tel.: +90 42ess: suleymansavash@y

Summary Cytokines are known to be increased in induced sputum in chronicobstructive pulmonary disease (COPD). In this study, the relationship betweenthe levels of interleukin-6 (IL-6), interleukin-8 (IL-8), and tumour necrosis factor-a(TNF-a) in induced sputum of patients with exacerbation of COPD, and the severityof the disease, pulmonary function tests (PFT), arterial blood gases (ABG) werestudied.

Twenty-four patients with exacerbation of COPD were included in the study. Thepatients were grouped according to their PFT into two as: Group 1 (FEV1 below 50%of the predicted value, severe–very severe COPD, n ¼ 12) and, Group 2 (FEV1 above50% of the predicted value, mild–moderate COPD, n ¼ 12). The levels of IL-6, IL-8and TNF-a in induced sputum of the subjects were measured.

The mean levels of IL-6, IL-8 and TNF-a in induced sputum were found to behigher in Group 1 (severe–very severe COPD) than in Group 2 (mild–moderate COPD).The differences in IL-6 and IL-8 levels between groups were statistically significant(Po0:05). A significant correlation was observed between the IL-6 value and FEV1(r ¼ �0:435, P ¼ 0:034), FEV1/FVC (r ¼ �0:446, P ¼ 0:029), PaO2 (r ¼ �0:711,P ¼ 0:000), SaO2 (r ¼ �0:444, P ¼ 0:030) and disease duration (r ¼ 0:427,P ¼ 0:037), respectively. Also, the level of IL-8 in induced sputum was inverselycorrelated with FEV1 (r ¼ �0:562, P ¼ 0:004), PaO2 (r ¼ �0:540, P ¼ 0:006) and SaO2

(r ¼ �0:435, P ¼ 0:034). However, all three cytokines were positively correlatedwith the smoking load (r ¼ 0:653, P ¼ 0:001; r ¼ 0:439, P ¼ 0:032; r ¼ 0:649,P ¼ 0:001).

We conclude, therefore, that in exacerbated COPD cases with greater degrees ofobstruction of the airways have higher levels of cytokines in induced sputum. This

Elsevier Ltd. All rights reserved.

2341 0660.ahoo.com (S.S. Hacievliyagil).

ARTICLE IN PRESS

Association between cytokines in induced and severity of COPD 847

can be interpreted to mean that these cytokines are related to the clinicalparameters like the ABG and PFT and seem to be the determinant of the severity ofthe disease.& 2005 Elsevier Ltd. All rights reserved.

Introduction

Chronic obstructive pulmonary disease (COPD) is animportant lung disease characterized by progres-sive airway obstruction due to chronic inflamma-tion.1 Exacerbations in COPD lead to increases inthe number of patients reporting to the hospitaland the frequency of admissions. It also affectsadversely the quality of life of the patients andrestricts their daily activities.2 COPD exacerbationsare responsible for the increase in morbidity andmortality rates attributable to the disease.3

Cytokines are extracelluler signal proteins lessthan 80 kDa formed by various cell types in thebody. Interleukin-6 (IL-6) is secreted by monocytes,macrophages, T cells, B cells, fibroblasts, epithelialcells of the airways and endothelial cells. Under theactivation of IL-1b and TGF-b, IL-6 is secreted fromthe smooth muscles of the airways. IL-6 arepleotrophic cytokines with a role in the activation,proliferation and differentiation of T cells. IL-6 alsoserves as a terminal differentiating factor for Bcells, and induces immunoglobins like IgG, IgA andIgM.4 Interleukin-8 (IL-8), also known as CXCL8, is aCXC chemokine that is a potent chemoattractantfor neutrophils. In general, monocytes, tissue andalveolar macrophages, pulmonary epithelium, cellsof the smooth muscles of the airways, eosinophils,fibroblasts, and endothelial cells are its importantsources.5 Tumour necrosis factor-a (TNF-a) isproduced from several cells including T lympho-cytes, mast cells and cells of the airway epithe-lium. TNF-a controls cellular migration andpermeability and also stimulates the secretion ofGM-CSF, IL-6 and IL-8. It also activates thetranscription factor-kB (NF-kB) which in turnactivates the IL-8 gene in epithelial cells andmacrophages.6

The role inflammation and the cytokines play inthe pathogenesis of COPD have attracted moreinterest in recent times.4,6,7 In COPD patientselevation of cytokine levels are known to be evenmore apparent during exacerbations. This situationhas been demonstrated by the presence of inflam-matory cells and elevated levels of the cytokines inthe serum, induced sputum, bronchoalveolar la-vage (BAL) fluid and bronchoscopic biopsy speci-mens.8–11 Aaron et al.12 studied the levels of IL-8and TNF-a in induced sputum among COPD patients

during their stable periods, at exacerbation andduring the stable period a month after theexacerbation episode. The levels of IL-8 and TNF-a in induced sputum were observed to be higher inthe exacerbation period than in the stable period,falling again during the stable period a monthafterwards.

Many cytokines are increased or activated inCOPD, but their contribution to disease severity isstill unknown. In this study, the relationshipbetween the levels of IL-6, IL-8 and TNF-a ininduced sputum and the severity of the disease,pulmonary function tests (PFT), arterial bloodgases (ABG) were investigated.

Materials and method

A total of 30 male COPD patients reporting to thepulmonary diseases clinic of the Turgut OzalMedical Center, Inonu University, (Malatya, Turkey)with exacerbation of disease were enrolled. Diag-nosis of COPD was made based on the AmericanThoracic Society and European Respiratory Societycriteria,13 with exacerbation identified accordingto the definition by Anthonisen et al. which is basedon an increase in symptoms of dyspnoea, sputumvolume and sputum purulence with or withoutsymptoms of upper respiratory infection and thensubdivided depending on the number of symptoms.All patients with COPD exacerbation in our studyhave all major exacerbation symptoms (type Iexacerbation).14

The patients had no other pulmonary diseaseincluding asthma, bronchiectasia, pneumonia, tu-berculosis or lung cancer. All subjects were hospi-talized because of type I exacerbation of COPD andthey were no medications except for COPD. Eachpatient was given a brief explanation about thestudy and their consent for voluntary participationobtained. The clinical history of each patient wasobtained after which a through physical examina-tion was conducted. Routine laboratory analysis,echocardiography and radiologic studies were or-dered. PFTwere performed in all cases by using theVmax 22 device (SensorMedics Ltd., Yorba Linda,CA, USA) and ABG conducted for each patientbreathing room air. A total of six patients wereexcluded from the study for various reasons: one

ARTICLE IN PRESS

S.S. Hacievliyagil et al.848

patient due to elevated liver enzymes, two patientsdue to development of bronchospasm during theinduction stage for sputum, and for withdrawal ofconsent by another three patients. Therefore 24patients therefore completed the study.

The subjects were classified into two groupsbased on their FEV1 results. Those with an FEV1below 50% of the predicted value were placed inGroup 1 (severe–very severe COPD, n ¼ 12), whilethose with FEV1 above 50% of the predicted valueformed Group 2 (mild–moderate COPD, n ¼ 12).

Before the commencement of the treatment forthe exacerbation induced sputum samples weretaken from patients by ultrasonic nebulizators(Hirtz, Hico-Ultrasonat 806 EH, Koln-Germany).For this purpose, patients were made to rinse theirmouths with water after inhalation of a 3% salinesolution for 5–7min and then asked to expectoratedeeply for the 2ml induced sputum samples to becollected. In patients in whom adequate amount ofsample could not be collected saline inhalation wasrepeated at the same dose and subsequentlyincreased (at a concentration of 4.5%).15 Thesamples that were collected into a wide-covered

Table 1 Basic properties and laboratory values of patien

Group 1 (FEV1o

Subjects, n 12Age (years) 59.3711.7Duration of disease (years) 13.877.7Smoking load (pack-years) 41.4721.1Smoking status (yes/ex/never) 7/5/0IL-6 (pg/ml) 697.87662.8IL-8 (pg/ml) 10054.372330.2TNF-a (pg/ml) 134.67140.7FVC (% predicted) 63.5718.8FEV1 (% predicted) 33.879.7FEV1/FVC (%) 43.779.1FEF25%–75% 13.374.9PH 7.3970.8PaO2 (mmHg) 45.877.1PaCO2 (mmHg) 44.077.4SaO2 (%) 79.679.2HCO3

� (mmol/l) 29.077.0Hemoglobin (g/dl) 15.172.4Hematocrit (%) 47.178.2BUN (mg/dl) 22.8713.6Leukocyte count (109/l) 12366.775261.4Creatinine (mg/dl) 1.070.3Albumin (mg/dl) 2.670.2PAPs (mmHg) 38.6725.9

Data presented as mean7SD (standard deviation) or n, unless othtumor necrosis factor-a, FVC: forced vital capacity, FEV1: forcedflow, pH: arterial blood pH, PaO2: arterial oxygen tension, PaCsaturation, HCO3

�: arterial bicarbonate, BUN: blood urea nitrojen,

sterile 50 cc polypropylene plastic cups wereimmediately transferred to the laboratory afterlabelling for processing. The induced sputumsamples were analysed within an hour. The sputumsamples were prevented from contamination bysaliva by filtering through a double-layered sterilegauze immediately after collection. With the aid ofa forceps and spatula approximately 1–1.5 g wastaken from each sample and transferred into 10mlsterile polypropylene tubes with predeterminedweights and reweighed. After the addition of aphosphate buffer solution (PBS) that weighed ninetimes their fresh weight 5–6 pieces of silicone glassbeads were introduced into each tube and stirredwith a vortex for 10min until a homogenizedsolution was obtained. The solution was thereaftersubjected to centrifugation at +4 1C for 10min at1000g. The resulting supernatant was transportedinto polypropylene tubes in aliquots of 1ml. Afterthe addition of 10 ml 100mM phenyl methyl sulfonylfluoride (PMSF) into each tube they were coveredand turned upside down several times to ensureadequate mixing. Until analysed they were storedat �70 1C.16

ts in Groups 1 and 2.

50%) Group 2 (FEV1450%) P value

1259.879.3 NS13.377.0 NS28.6714.0 NS8/4/0 NS216.17350.8 0.0377159.372973.2 0.01496.37158.1 NS90.9720.9 0.00361.9718.8 0.00050.079.7 NS27.8715.1 0.0057.4570.4 0.0452.676.2 0.02140.878.5 NS85.776.2 0.00230.175.4 NS15.471.3 NS47.975.6 NS18.4710.4 NS9375.073252.7 NS0.870.2 NS3.270.5 NS30.6719.6 NS

erwise stated. IL-6: interleukin-6, IL-8: interleukin-8, TNF-a:expiratory volume in 1 s, FEF25%–75%: forced mid-expiratoryO2: arterial carbon dioxide tension, SaO2: arterial oxygenPAPs: systolic pulmonary artery pressure), NS: not significant.

ARTICLE IN PRESS


The IL-6, IL-8, TNF-a levels were measured by thechemiluminiscence method using an Immulite-1000analyzer and special kits (DPC [Diagnostic ProductCo., CA, USA]). The lower limit of detection forthese assays was 5 pg/ml for IL-6, 2 pg/ml for IL-8and 5 pg/ml for TNF-a. The inter- and intra-assaycoefficient of variation of these all assays was lessthan 10%. The samples which were higher thandetection limits of cytokines were reanalysed witha dilution of 0.9% serum saline solution for onetimes weight.

The patients were compared for demographicand laboratory parameters of subjects. The dataobtained were analysed with the SPSS version 10.0(Chicago, IL, USA) statistical program. Data areexpressed as means and standart deviation. Differ-ences between groups were analysed using Stu-dent’s t-test. Correlations between differentparameters were evaluated using the Pearson’scorrelation test. The significance level was con-sidered as Po0:05.

Results

The demographic properties and laboratory valuesof the subjects in Groups 1 and 2 are shown inTable 1. The differences of values of FVC, FEV1,FEF25%–75%, pH, PaO2 and SaO2 between groupsrevealed statistically significant (Po0:05). Theaverage induced sputum IL-6 level for Groups 1and 2 was 697.87662.8 and 216.17350.8 pg/ml,respectively (P ¼ 0:037). The level of IL-8 inthe induced sputum was 10054.372330.2 pg/mlfor Group 1 and 7159.372973.2 pg/ml for Group 2(P ¼ 0:014). The levels of TNF-a in induced sputumwere found to be 134.67140.7 and 96.37158.1 pg/ml for Groups 1 and 2, respectively. No statisticallysignificant difference was observed between thegroups in the TNF-a levels (P40:05).

As shown in Fig. 1a and b, the levels of IL-6and IL-8 in induced sputum correlated negativelywith FEV1 (r ¼ �0:435, P ¼ 0:034; r ¼ �0:562, P ¼0:004). In addition, the levels of IL-6 in inducedsputum significantly correlated, respectively, withFEV1/FVC (r ¼ �0:446, P ¼ 0:029, Fig. 1c) anddisease duration (r ¼ 0:427, P ¼ 0:037, Fig. 1d).Also, the levels of IL-6 and IL-8 in induced sputumcorrelated negatively with PaO2 (r ¼ �0:711, P ¼0:000, Fig. 1e; r ¼ �0:540, P ¼ 0:006, Fig. 1f) andSaO2 (r ¼ �0:444, P ¼ 0:030, Fig. 1g; r ¼ �0:435,P ¼ 0:034, Fig. 1(h). However, there was a sig-nificant correlation between levels of all threecytokines and smoking load (pack-years) (r ¼ 0:653,P ¼ 0:001; r ¼ 0:439, P ¼ 0:032; r ¼ 0:649, P ¼0:001, Fig. 2).

Discussion

In COPD there is a chronic inflammatory processwithin the airways that increases during an acuteepisode.1,6 The induced sputum method, a non-invasive method adequate enough in demonstratinginflammation in the airways such as that occuring inCOPD has been shown in earlier studies to be areliable method.15–18 In our study, the cytokinelevels in patients with exacerbation of COPD havebeen found to relate to the PFT, ABG, as well as tothe severity of disease.

Keatings et al.19 found high IL-8 levels in inducedsputum of COPD patients. In another study,Yamamoto et al.20 found higher IL-8 levels in thesputum of COPD patients than that in asthmaticpatients and normal controls. In their study,Vernooy et al.21 found higher induced sputum IL-8levels in 18 COPD patients compared to their 17smoking control subjects. Keatings et al.19 andYamamoto et al.20 demonstrated a positive correla-tion between the IL-8 level and the neutrophilproportions. This could not be verified in our studybecause characterization of cellular components ofthe sputum was not performed. On the other hand,all patients in this study had all the majorsymptoms of exacerbations as an increase insymptoms of dyspnoea, sputum volume and sputumpurulence (type I exacerbation).14,22 Gompertz etal.23 suggest that exacerbations of COPD defined bysputum colour, differ in the degree of bronchial andsystemic inflammation. They found purulent ex-acerbations were associated with marked bronchialand systemic inflammation and bacterial infection.Another lacks of present study, the bacteriology orviral cultures of sputum was not processed, so thispapers is not given any information about levels ofairway inflammation and bacterial infections orcolonizations. We found the cytokines levels in theinduced sputum more higher in severe or verysevere COPD patients than mild or moderate COPDpatients. It has been suggested that these cytokinesplay a role in airway inflammation and severity ofdisease.

With each episode of exacerbation there isgradual deterioration in the respiratory functionsof the patient. In those patients with frequentexacerbations the fall in FEV1 is much greater thanit is in those with less frequent exacerbations.24

Frequent exacerbations increase the severity of therespiratory tract inflammation. Bhowmik et al.8

found that the levels of IL-6 and IL-8 in inducedsputum of patients with three or more episodes peryear were higher than those with less frequentattacks. In contrast, Gompertz et al.25 found nosignificant differences between those with three or

ARTICLE IN PRESS


more episodes of exacerbation per year than thosewith less attacks. Soler et al.26 demonstrated anegative correlation between the FEV1 and the IL-6and IL-8 levels in BAL fluid. In contrast to thesestudies, however, Bhowmik et al.8 failed to estab-lish any relationship between the induced sputumIL-6 and IL-8 levels and the PFT parameters and thetime for improvement in symptoms and PEF values

FEV1

120100806040200

2000

1750

1500

1250

1000

750

500

250

0

FEV1/FVC706050403020100

2000

1750

1500

1250

1000

750

500

250

0

IL-6

(pg

/ml)

IL-6

(pg

/ml)

FEV1>50

FEV1<50

FEV1>50

FEV1<50

(a) (

(c) (

Figure 1 Correlations between (a) the levels of IL-6 and FEV(r ¼ �0:562, P ¼ 0:004); (c) the levels of IL-6 and FEV1/FVC (duration (r ¼ 0:427, P ¼ 0:037); (e) the levels of IL-6 and PaO(r ¼ �0:540, P ¼ 0:006); (g) the levels of IL-6 and SaO2 (r(r ¼ �0:435, P ¼ 0:034).

in the exacerbation period. Yamamoto et al.20

reported a negative correlation between the IL-8levels in induced sputum and the FEV1/FVC ratio inCOPD. Similarly we found a negative correlationbetween the levels of IL-6 in induced sputum andFEV1/FVC ratios in this study. Also, FEV1 of thepredicted value was inversely correlated with thelevels of IL-6 and IL-8 in induced sputum, but not

FEV1

120100806040200

14000

12000

10000

8000

6000

4000

2000

0

Disease duration403020100

IL-6

(pg

/ml)

2000

1750

1500

1250

1000

750

500

250

0

IL-8

(pg

/ml)

FEV1>50

FEV1<50

FEV1>50

FEV1<50

b)

d)

1 (r ¼ �0:435, P ¼ 0:034); (b) the levels of IL-8 and FEV1r ¼ �0:446, P ¼ 0:029); (d) the levels of IL-6 and disease

2 (r ¼ �0:711, P ¼ 0:000); (f) the levels of IL-8 and PaO2

¼ �0:444, P ¼ 0:030); (h) the levels of IL-8 and SaO2

ARTICLE IN PRESS

PaO2

706050403020100

2000

1750

1500

1250

1000

750

500

250

0

PaO2

706050403020100

14000

12000

10000

8000

6000

4000

2000

0

SaO2 SaO2

100806040200

2000

1750

1500

1250

1000

750

500

250

0100

9080

7060

5040

3020

100

14000

12000

10000

8000

6000

4000

2000

0

FEV1>50

FEV1<50

FEV1>50

FEV1<50

FEV1>50

FEV1<50

FEV1>50

FEV1<50

IL-8

(pg

/ml)

IL-6

(pg

/ml)

IL-8

(pg

/ml)

IL-6

(pg

/ml)

(e) (f)

(g) (h)

Figure 1 (Continued)


TNF-a levels. This indicates that the levels of IL-6and IL-8 in induced sputum might be related to theseverity of the respiratory tract obstruction. Andwe think that presence of high levels of cytokines ininduced sputum increases inflammation in theairways that in turn increases the susceptibility toexacerbations.

With hypoxia IL-6 and IL-8 are secreted from theendothelial cells.27,28 In addition, hypoxia isthought to be related to activation of the TNF-asystem in COPD patients.29 Yasuda et al.30 foundthat although the difference was not statisticallysignificant, the TNF-a and IL-6 levels among severe

COPD patients with hypoxemia was higher thantheir non-hypoxic controls who had mild–moderateCOPD. We found severe and very severe patientswho are more hypoxemic patients had highercytokines levels. However, there was a relationbetween the levels of IL-6 in induced sputum andhypoxemia, might due to increased inflammationassociated with hypoxemia.

Cigarette smoking is the most important cause ofCOPD. However, some smokers develop COPD andthe reason is still unknown.1 Bhowmik et al.8

established a correlation between the levels of IL-6 and IL-8 in induced sputum and the pack-years of

ARTICLE IN PRESS

Smoking load (pack-years)100806040200

TN

F-α

(pg/

ml)

600

500

400

300

200

100

0

FEV1>50

FEV1<50

Smoking load (pack-years)

100806040200

IL-8

(pg

/ml)

14000

12000

10000

8000

6000

4000

2000

0

FEV1>50

FEV1<50

Smoking load (pack-years)

100806040200

IL-6

(pg

/ml)

2000

1750

1500

1250

1000

750

500

250

0

FEV1>50

FEV1<50

(a) (b)

(c)

Figure 2 Correlations between (a) the levels of IL-6 and smoking load (pack-years) (r ¼ 0:653, P ¼ 0:001); (b) betweenthe levels of IL-8 and smoking load (pack-years) (r ¼ 0:439, P ¼ 0:032); (c) between the levels of TNF-a and smokingload (pack-years) (r ¼ 0:649, P ¼ 0:001).


cigarettes smoked. Wang et al.31 measured thelevels of IL-6 and IL-8 in induced sputum in 20stable COPD patients and compared them withsmoking and non-smoking control groups. The IL-8and TNF-a levels in the COPD group were found tobe significantly higher than the smoking and non-smoking groups with the IL-8 level in the smokinggroup also significantly higher than the non-smok-

ing group. The IL-6 levels, however, did not showany significant differences between the groups. Inour study a positive correlation was establishedbetween the levels of all three cytokines and thesmoking load (pack-years). This indicates thathigher smoking load had higher cytokine levels intheir induced sputum. The initiation of cytokinesrelease in COPD is likely to be due to the direct

ARTICLE IN PRESS


effect of cigarette smoking. Also, our study high-lighted a close link between smoking and levelsof cytokines in induced sputum and thereforesmoking needs to be prevented. On the other hand,correlations examined in induced sputum suggestthat those patients with a longer history of diseaseduration had higher cytokines levels. For thedetermination of disease duration, we used thedate of first visit to a doctor for chronic complaintsof cough, dyspnea and sputum production.

We therefore conclude that, in our study, COPDpatients with more marked obstruction had higherIL-6, IL-8 and TNF-a levels in their induced sputumsamples. In addition to its determinant role in thepathogenesis of COPD the levels of IL-6 and IL-8 ininduced sputum were found to be directly relatedto the ABG and PFT, parameters that determine theseverity of disease. The airway inflammation inthose patients with higher cytokine levels increasedleading to an increase in the rate of exacerbations.Reducing the rate of exacerbation episodes in COPDwill no doubt result in a marked fall in themorbidity and mortality of the disease. Currenttreatment is centered on the use of bronchodila-tors. As the importance of inflammation in thepathogenesis of COPD emerges treatment of COPD,just as in asthma will definitely be geared towardthe control of inflammation. By contrast, theinflammatory process in COPD appears to beresistant to corticosteroids, indicating that newanti-inflammatory treatments will be needed. It isour opinion, therefore, that anti-cytokine agents tobe developed will be an important leap forward inthe prevention of COPD and its progression as wellas minimalizing severity of disease.

Acknowledgements

The authors are grateful to Dr. Ozkan Kizkin for hishelpful advice and criticism.

References

1. GOLD Scientific Committee. Global strategy for the diag-nosis, management, and prevention of chronic obstructivepulmonary disease: NHLBI/WHO Global Initiative for chronicobstructive pulmonary disease (GOLD) workshop report.www.goldcopd.org. Updated 2004.

2. Seemungal TAR, Donaldson GC, Paul EA, et al. Effect ofexacerbation on quality of life in patients with chronicobstructive pulmonary disease. Am J Respir Crit Care Med1998;157:1418–22.

3. Wedzicha JA. Acute exacerbations of COPD. In: Barnes PJ,Drazen JM, Rennard S, Thomson NC, editors. Asthma and

COPD: basic mechanisms and clinical management. London:Academic Press; 2002. p. 727–36.

4. Chung KF. Cytokines. In: Barnes PJ, Drazen JM, Rennard S,Thomson NC, editors. Asthma and COPD: basic mechanismsand clinical management. London: Academic Press; 2002. p.261–71.

5. Pease JE, Williams TJ. Chemokines. In: Barnes PJ, DrazenJM, Rennard S, Thomson NC, editors. Asthma and COPD:basic mechanisms and clinical Management. London: Aca-demic Press; 2002. p. 255–60.

6. Chung KF. Cytokines in chronic obstructive pulmonarydisease. Eur Respir J 2001;18(Suppl 34):50–9.

7. Petty TL. COPD in perspective. Chest 2002;121(Suppl):116–20.

8. Bhowmik A, Seemungal TAR, Sapsford RJ, Wedzicha JA.Relation of sputum inflammatory markers to symptoms andlung function changes in COPD exacerbations. Thorax2000;55(2):114–20.

9. Martin TR, Raghu G, Maunder RJ, Springmeyer SC. Theeffects of chronic bronchitis and chronic airflow obstructionon lung cell populations recovered by bronchoalveolarlavage. Am Rev Respir Dis 1985;132:254–60.

10. Ollerenshaw SL, Woolcock AJ. Characteristics of theinflammation in biopsies from large airways of subjects withchronic airflow limitation. Am Rev Respir Dis 1992;145:922–7.

11. Wedzicha JA, Seemungal TA, MacCallum PK, et al. Acuteexacerbations of chronic obstructive pulmonary diseaseare accompanied by elevations of plasma fibrinogenand serum IL-6 levels. Thromb Haemost 2000;84(2):210–5.

12. Aaron SD, Angel JB, Lunau M, et al. Granulocyte inflamma-tory markers and airway infection during acute exacerbationof chronic obstructive pulmonary disease. Am J Respir CritCare Med 2001;163(2):349–55.

13. Celli BR, MacNee W, committee members. Standards forthe diagnosis and treatment of patients with COPD: asummary of the ATS/ERS position paper. Eur Respir J2004;23:932–46.

14. Anthonisen NR, Manfreda J, Warren CPW, et al. Antibiotictherapy in exacerbations of chronic obstructive pulmonarydisease. Ann Intern Med 1987;106(2):196–204.

15. Bhowmik A, Seemungal TA, Sapsford RJ, et al. Comparison ofspontaneous and induced sputum for investigation of airwayinflammation in chronic obstructive pulmonary disease.Thorax 1998;53(11):953–6.

16. Roland M, Bhowmik A, Sapsford RJ, et al. Sputum and plasmaendotelin-1 levels in exacerbations of chronic obstructivepulmonary disease. Thorax 2001;56:30–5.

17. Rytila PH, Lindqvist AE, Laitinen LA. Safety of sputuminduction in chronic obstructive pulmonary disease. EurRespir J 2000;15:1116–9.

18. Vlachos-Mayer H, Leigh R, Sharon RF, et al. Success andsafety of sputum induction in the clinical setting. Eur RespirJ 2000;16(5):997–1000.

19. Keatings VM, Collins PD, Scott DM, Barnes PJ. Differences ininterleukin-8 and tumor necrosis factor-alpha in inducedsputum from patients with chronic obstructive pulmonarydisease or asthma. Am J Respir Crit Care Med 1996;153(2):530–4.

20. Yamamoto C, Yoneda T, Yoshikawa M, et al. Airwayinflammation in COPD assessed by sputum levels of inter-leukin-8. Chest 1997;112(2):505–10.

21. Vernooy JH, Kucukaycan M, Jacobs JA, et al. Local andsystemic inflammation in patients with chronic obstructivepulmonary disease: soluble tumor necrosis factor receptors

http://www.goldcopd.org

ARTICLE IN PRESS


are increased in sputum. Am J Respir Crit Care Med2002;166(9):1218–24.

22. Wedzicha JA, Donaldson GC. Exacerbations of chronicobstructive pulmonary disease. Respir Care 2003;48(12):1204–13.

23. Gompertz S, O’Brien C, Bayley DL, et al. Changesin bronchial inflammation during acute exacerba-tions of chronic bronchitis. Eur Respir J 2001;17(6):1112–9.

24. Donaldson GC, Seemungal TAR, Bhowmik A, Wedzicha JA.Relationship between exacerbation frequency and lungfunction decline in chronic obstructive pulmonary disease.Thorax 2002;57(10):847–52.

25. Gompertz S, Bayley DL, Hill SL, Stockley RA. Relationshipbetween airway inflammation and the frequency of exacer-bations in patients with smoking related COPD. Thorax2001;56:36–41.

26. Soler N, Ewig S, Torres A, et al. Airway inflammation andbronchial microbial patterns in patients with stable COPD.Eur Respir J 1999;14(5):1015–22.

27. Karakurum M, Shreeniwas R, Chen J, et al. Hypoxicinduction of interleukin-8 gene expression in humanendothelial cells. J Clin Invest 1994;93(4):1564–70.

28. Yan SF, Tritto I, Pinsky D, et al. Induction of interleukin 6 (IL-6) by hypoxia in vascular cells. Central role of the bindingsite for nuclear factor-IL-6. J Biol Chem 1995;270(19):11463–71.

29. Takabatake N, Nakamura H, Abe S, et al. The relationshipbetween chronic hypoxemia and activation of the tumornecrosis factor-alpha system in patients with chronicobstructive pulmonary disease. Am J Respir Crit Care Med2000;161:1179–84.

30. Yasuda N, Gotoh K, Minatoguchi S, et al. An increase ofsoluble Fas, an inhibitor of apoptosis, associated withprogression of COPD. Respir Med 1998;92(8):993–9.

31. Wang S, Xu F, Chen Y. Detection and significance ofinterleukin-8,6, tumor necrosis factor-alpha in sputa frompatients with chronic obstructive pulmonary disease. Zhon-ghua Jie He He Hu Xi Za Zhi 2000 (abstract- PMID:11778258).

association between cytokines in induced sputum and severity of chronic obstructive pulmonary...

Documents