DOI 10.1378/chest.06-2982 2007;132;98-105; Prepublished online June 5, 2007;Chest

 Kharitonov, Ian M. Adcock and Peter J. BarnesKittipong Maneechotesuwan, Sarah Essilfie-Quaye, Sergei A. 

*NeutrophilsWith Increased Sputum Interleukin-8 andCorticosteroid Withdrawal Is Associated Loss of Control of Asthma Following Inhaled

  http://chestjournal.chestpubs.org/content/132/1/98.full.html

services can be found online on the World Wide Web at: The online version of this article, along with updated information and 

ISSN:0012-3692)http://chestjournal.chestpubs.org/site/misc/reprints.xhtml(

written permission of the copyright holder.this article or PDF may be reproduced or distributed without the priorDundee Road, Northbrook, IL 60062. All rights reserved. No part of Copyright2007by the American College of Chest Physicians, 3300Physicians. It has been published monthly since 1935.

is the official journal of the American College of ChestChest

 © 2007 American College of Chest Physicians by guest on March 8, 2012chestjournal.chestpubs.orgDownloaded from

Loss of Control of Asthma FollowingInhaled Corticosteroid Withdrawal IsAssociated With Increased SputumInterleukin-8 and Neutrophils*

Kittipong Maneechotesuwan, MD, PhD†; Sarah Essilfie-Quaye, BSc;Sergei A. Kharitonov, MD, PhD; Ian M. Adcock, PhD; andPeter J. Barnes, DM, DSc, FCCP

Background: The role of neutrophils in exacerbations of asthma is poorly understood. Weexamined the effect of withdrawal of inhaled corticosteroids on sputum inflammatory indexes ina double-blind study in patients with moderate, stable asthma.Methods: Following a 2-week run in period, 24 subjects were randomized to receive eitherbudesonide (400 �g bid) or placebo, and the study was continued for another 10 weeks.Results: Loss of asthma control developed in 8 of 12 patients over the 10-week period of steroidwithdrawal, whereas only 1 of 10 patients with budesonide treatment had exacerbations. Thosewith an exacerbation had increased sputum interleukin (IL)-8 (p < 0.0001) and increased sputumneutrophil numbers (p < 0.0001) compared to those without an exacerbation. The significantelevation in sputum IL-8 and neutrophil counts initially occurred 2 weeks prior to an exacerba-tion. Sputum neutrophilia correlated positively with changes in IL-8 levels (r2 � 0.76, p � 0.01).Conclusions: Rapid withdrawal of inhaled corticosteroids results in an exacerbation of asthmathat is preceded by an increase in sputum neutrophils and IL-8 concentrations, in contrast to anincrease in eosinophils reported in previous studies in which inhaled steroids are slowly tapered.

(CHEST 2007; 132:98–105)

Key words: asthma; exacerbation; interleukin-8; neutrophil

Abbreviations: IL � interleukin; PC20 � provocative concentration of methacholine causing a 20% fall in FEV1

A lthough eosinophilic airway inflammation is rec-ognized as an important feature of some patients

with chronic, stable asthma,1,2 evidence supports animportant role of neutrophils in acute exacerba-tions.3–5 In stable asthma, there is typically an infil-trate of eosinophils, which occurs in response tocytokines such as interleukin (IL)-5, secreted by

T-helper type 2 lymphocytes.6 In contrast, duringacute exacerbations of asthma, the airway inflamma-tory response is both more intense and heteroge-neous.3–5,7 The increased intensity of airway inflam-mation is reflected by an elevated total cell count insputum.3–5,8 The heterogeneity of the inflammatoryresponse is reflected by an increase in neutrophils aswell as eosinophils, and there are reports of exacer-bations without sputum eosinophilia.3–5 There is also*From Airways Disease Section, National Heart and Lung Insti-

tute, Imperial College, London, UK.†Currently at the Division of Respiratory Disease and Tubercu-losis, Department of Internal Medicine, Siriraj Hospital, MahidolUniversity, Thailand.Funding for research was provided by the Royal Government ofThailand for sponsoring Dr. Maneechotesuwan, and AstraZeneca(Lund, Sweden).Drs. Barnes, Kharitonov, and Adcock obtained research fundingfrom several pharmaceutical companies interested in the treat-ment of asthma, including AstraZeneca, GlaxoSmithKline, andNovartis.

Manuscript received December 12, 2006; revision acceptedMarch 23, 2007.Reproduction of this article is prohibited without written permissionfrom the American College of Chest Physicians (www.chestjournal.org/misc/reprints.shtml).Correspondence to: Peter J. Barnes, DM, DSc, FCCP, AirwaysDisease Section, Dovehouse St, London, SW3 6LY, UK; e-mail:p.j.barnes@imperial.ac.ukDOI: 10.1378/chest.06-2982

Original ResearchASTHMA

98 Original Research

 © 2007 American College of Chest Physicians by guest on March 8, 2012chestjournal.chestpubs.orgDownloaded from

evidence for increased expression of the potentneutrophil chemoattractant IL-8,3 in addition to thepresence of IL-5. There is heterogeneity in theconcentrations of IL-8 in patients with moderateasthma and the strong correlation between the con-centration of IL-8 release by peripheral blood mono-nuclear cells9 and the frequency of asthma exacer-bations.10

Leukocytes, particularly neutrophils and eosino-phils, are the first cells recruited to the site of tissuedamage after an inflammatory insult.11 These cellsare then removed by apoptosis during the resolutionof the inflammatory response. Neutrophils are thepredominant inflammatory leukocyte characterizingairway inflammation in acute severe asthma, and thismay influence the clinical presentation, predisposingto a sudden severe attack.3–5 IL-8 is a key chemokineinvolved in the recruitment of neutrophils to the areaof inflammation, but few studies have characterizedthe kinetics of IL-8 production in sputum over aperiod of time prior to the onset of a clinical asthmaexacerbation. Because it is not possible to obtain thisinformation prospectively with a naturally occurringexacerbation, we induced loss of control of asthma bywithdrawing inhaled corticosteroids so that we wereable to make prospective measurements of inflam-matory indexes in induced sputum. This was done ina double-blind controlled manner so that steroidwithdrawal could be compared with continued ther-apy with inhaled corticosteroids.

Materials and Methods

Subjects

Twenty-four nonsmoking patients (10 men) with moderate,persistent asthma participated in the study. The inclusion criteriawere age 18 to 60 years of age and a history of stable asthma asdefined by the American Thoracic Society. Patient characteristicsare summarized in Table 1. All patients had a baseline FEV1� 70% of predicted, FEV1 reversibility � 15%, and requiredregular treatment with moderate doses of inhaled corticosteroids(beclomethasone dipropionate, 800 �g/d or equivalent) for � 3months. All patients were hyperresponsive as measured by aprovocative concentration of methacholine causing a 20% fall in

FEV1 (PC20) � 8 mg/mL. None of the patients had a history ofrespiratory disease other than asthma, and none required oralglucocorticoid treatment within 3 months before study entry andused any other medication except study medication during thetrial. The patients were included during a clinically well-con-trolled period without symptoms of an upper respiratory tractinfection for 4 weeks prior to the study. The study was approvedby the Royal Brompton and Harefield NHS Hospital Trust EthicsCommittee, and informed consent was given by all patients.

Design

In order to determine the onset of asthma exacerbation aftercorticosteroid withdrawal, the study was conducted in a random-ized, double-blind, placebo-controlled manner by comparing theeffects of 10 weeks of treatment with budesonide (400 �g bid orplacebo via Turbuhaler; AstraZeneca; Lund, Sweden). The studyperiods consisted of a 2-week run in phase and a 10-weekexacerbation-induction phase. Before the start of the study,patients were screened on 2 separate days. Long-acting �2-agonists and theophylline preparations were discontinued at least72 h before the first screening visit. On the first day, FEV1 wasmeasured before and after inhalation of 400 �g of albuterol bymetered-dose inhaler with a large-volume spacer device, and onthe second visit a methacholine challenge test was performed.The patients stopped regular asthma medication for 2 weeks.Thereafter, patients were randomized to receive either matchedplacebo or budesonide, 400 �g bid, via Turbuhaler. Patients whohad exacerbations during the run-in period were excluded fromthe study. Patients visited the hospital at the end of the run-inperiod, at 2-week intervals during the exacerbation-inductionperiod, and at the time of an exacerbation to undertake lungfunction testing and sputum induction. During the study, thepatients kept a diary to record morning and evening peakexpiratory flow measurements.

Run-in Phase

During the 2-week run-in phase, all patients had to have stabledisease without lower respiratory tract infection. Throughout thestudy, only terbutaline via Turbuhaler was permitted on anon-demand basis as a rescue inhaler. A study physician wasaccessible by telephone 24 h/d.

Induction of Exacerbation

Steroid withdrawal was performed by inhalation of placebo,and the study was discontinued when an exacerbation developedor when no exacerbation of asthma occurred within the 10-weekstudy period. Patients undergoing exacerbation during treatmentperiod were withdrawn from the study. Thereafter, inhaledbudesonide was increased to 800 �g bid, and systemic cortico-steroids (prednisolone 30 mg/d) were administered.

An exacerbation was defined as at least one of three crite-ria12,13: (1) a drop in morning peak flow � 20% below baseline(mean of the last 7 days run-in period) on 2 consecutive days; (2)wakening due to asthma on 2 consecutive nights, and requiringrescue medication; (3) � 50% increase in 24-h rescue medicationuse on at least 2 consecutive days compared to mean use duringthe last 7 days of the run-in period, which also exceeded theequivalent of four puffs of terbutaline. FEV1 was then measuredwithin 24 h of the exacerbation.

Lung Function Measurements and MethacholineChallenge Tests

FEV1 and FVC were measured using a dry wedge spirometer(Vitalograph; Buckingham, UK). Values are expressed as percent-

Table 1—Patient Characteristics*

Characteristics Steroid Withdrawal Budesonide

Age, yr 36.9 � 2.5 38.3 � 3.1Male/female gender, No. 4/8 5/7Atopy, No. 10 9Postbronchodilator FEV1, L 3.87 � 0.17 3.75 � 0.1Reversibilty, % 20.4 � 1.0 20.7 � 1.2PC20, mg/mL 0.98 � 0.26 0.99 � 0.27

*Data are presented as mean � SEM unless otherwise indicated.

www.chestjournal.org CHEST / 132 / 1 / JULY, 2007 99

 © 2007 American College of Chest Physicians by guest on March 8, 2012chestjournal.chestpubs.orgDownloaded from

age of predicted. Baseline values were measured after 15 min ofrest and taken as the highest of three readings. Single readingsonly were taken at other times. Airway hyperresponsiveness tomethacholine was assessed using the method of Sterk andcolleagues.13 Methacholine was inhaled by tidal breathing indoubling concentrations (0.015 to 32 mg/mL) for 2 min at 5-minintervals. Measurements of FEV1 were made at baseline andafter each dose. The challenge test was discontinued if FEV1dropped � 20% from baseline. PC20 was calculated by linearinterpolation of the log-dose response curves.13

Sputum Induction and Processing

Sputum was induced by inhalation for 15 min of 3.5% salinesolution via an ultrasonic nebulizer (model 2000; DeVilbis;Heston, UK), as previously described.14 After each inhalationperiod, patients were asked to rinse their mouth and wereencouraged to expectorate sputum. Briefly, the whole sputumsample was processed with dithiothreitol (Sigma Chemicals;Poole, UK). The homogenized sputum was centrifuged at 1,500gfor 10 min. The supernatant was separated and frozen at � 70°Cuntil further analysis. Total cell counts were made on a hemocy-tometer slide, using Kimura stain, and slides were prepared(Cytospin; Shandon; Runcorn, UK) and stained with May-Grun-wald-Giemsa stain. Differential cell counts were made by ablinded observer. Three hundred nonsquamous cells werecounted on two slides for each sample in a blind way. Differentialcell counts are expressed as percentages of nonsquamous cells.Two of the sputum samples from the same patient in budesonidetreatment group at different visits were withdrawn from analysisbecause of containing � 80% squamous cells (Table 2).

IL-8 Assay

The concentration of IL-8 in sputum supernatant was deter-mined using commercially available enzyme-linked immunosor-bent assay (R&D Systems; Minneapolis, MN) according to themanufacturer instructions.

Statistical Analysis

Results are reported as mean � SEM. We compared patientswho lost control of asthma with those who did not after cortico-steroid withdrawal. All statistical tests were two sided, andsignificance was accepted at the level of 95% and p � 0.05 using

statistical software (GraphPad Prism; GraphPad Software; SanDiego, CA). To determine whether changes in sputum neutrophilcounts and sputum IL-8 levels accompanied asthma exacerba-tion, the differences of the last-visit measurements from baselinewere used for analysis of correlation with Pearson product-moment technique. The data from subjects with and withoutexacerbation at each time point were categorized as a groupbefore analysis. Sputum neutrophil counts, percentage of neu-trophils, and IL-8 levels 2 weeks prior to an exacerbation werecompared to those of the week-8 inflammatory markers innonexacerbation group using the unpaired t test. The last-visitmeasurements of the exacerbation and nonexacerbation groupswere also evaluated by the unpaired t test.

Results

Patient Characteristics

Patient characteristics are shown in Table 1. Twopatients in the budesonide treatment group wereexcluded from the analysis: one was unable to attendfor all of the study visits, and the other providedinduced-sputum specimens containing squamouscell counts � 80%.

Exacerbations

Eight of 12 patients had an exacerbation over the10-week period of steroid withdrawal (Table 3). Theremaining four subjects did not have an exacerbationduring the 10 weeks of follow-up. Only 1 of 10patients in the budesonide treatment group had anasthma exacerbation at week 2 (Table 2).

Sputum IL-8 and Neutrophils

Overall, the patients with an exacerbation aftersteroid cessation had significantly increasing levels ofsputum IL-8, percentage of neutrophils, and sputumneutrophils 2 weeks prior to an exacerbation and atthe onset of exacerbation (Fig 1, top left, A, top right,

Table 2—Total and Specific Sputum Cell Counts After Budesonide*

Variables

Weeks of Budesonide Treatment

0 2 4 6 8 10

Patients, No. 10 10 10 10 10 10FEV1 , % predicted 72.3 � 1.5 73.4 � 2.1 71.8 � 0.9 73.7 � 0.6 73.2 � 1.1 72.8 � 1.7Exacerbation, No. 1Total cells, 106/mL 3.4 � 0.3 3.2 � 0.3 3.6 � 0.2 3.2 � 0.2 3.5 � 0.4 3.6 � 0.6Neutrophils, % 33.9 � 3.7 41.3 � 4.1 38.2 � 2.1 36.4 � 5.2 37.7 � 3.8 38.4 � 4.5Neutrophils, 106/mL 1.2 � 0.6 1.3 � 0.5 1.4 � 0.4 1.2 � 0.5 1.3 � 0.6 1.3 � 0.7Eosinophils, % 4.8 � 1.0 2.6 � 0.8 1.8 � 0.6 1.2 � 0.5 0.45 � 0.2 0.3 � 0.1Eosinophils, 106/mL 0.2 � 0.03 0.09 � 0.02 0.06 � 0.03 0.04 � 0.01 0.01 � 0.004 0.02 � 0.01Macrophages, % 57.9 � 3.4 53.3 � 3.9 57.6 � 2.7 60 � 5.5 59.7 � 4.8 59.5 � 4.9Macrophages, 106/mL 1.9 � 0.2 1.7 � 0.2 2.1 � 0.2 2.0 � 0.2 2.1 � 0.9 2.2 � 0.4Lymphocytes, % 3.3 � 0.9 2.8 � 0.7 2.5 � 0.7 2.4 � 0.6 2.1 � 1 2.0 � 0.7Lymphocytes, 106/mL 0.1 � 0.01 0.1 � 0.01 0.1 � 0.02 0.07 � 0.02 0.08 � 0.03 0.1 � 0.03

*Data are presented as mean � SEM for the patients indicated, unless otherwise indicated.

100 Original Research

 © 2007 American College of Chest Physicians by guest on March 8, 2012chestjournal.chestpubs.orgDownloaded from

B, and bottom, C, respectively; Fig 2). The signifi-cant increase in IL-8 levels in induced sputumoccurred at 2-week prior to an exacerbation and at

the onset of exacerbation when compared with non-exacerbation (Table 4) [mean, 1,832.1 � 151 ng/mLvs 1074.3 � 111.0 ng/mL, p � 0.001; and

Figure 1. Changes in sputum IL-8 (top left, A), percentage of neutrophils (top right, B), and number ofneutrophils (bottom, C) in patients with or without an exacerbation (exac). Exacerbation data were derived fromthose who were treated with placebo (eight cases) and with budesonide (one case), and the nonexacerbationdata included asthmatic subjects who were treated with budesonide (nine cases) and with placebo (fourcases). Data are shown as mean � SEM. Comparisons of concentrations of IL-8 and neutrophil numbers insputum were made between the results 2 weeks prior to an exacerbation and those of nonexacerbation atweek 8 and between data at the exacerbation visit and those of the week-10 nonexacerbation.

Table 3—Changes in Clinical Parameters and Sputum Neutrophils in Asthmatic Subjects Following InhaledCorticosteroid Withdrawal*

Variables

Weeks

0 2 4 6 8 10

Patients, No. 12 12 8 7 6 4FEV1, % predicted 73.5 � 0.9 66.9 � 1.1 67.5 � 1.6 67.7 � 1.7 67.9 � 1.7 71.1 � 0.4Exacerbations, No. 0 4 1 1 2 0Total cells, 106/mL 3.7 � 0.3 6.0 � 0.8 5.7 � 0.8 6.9 � 0.9 7.7 � 0.7 6.4 � 0.4Neutrophils, % 31.0 � 2.6 39.7 � 2.3 40.8 � 3.6 35.6 � 5.9 28.2 � 8.4 14.3 � 1.3Neutrophils, 106/mL 1.1 � 0.1 2.5 � 0.4 2.5 � 0.5 2.7 � 0.7 2.4 � 0.8 0.9 � 0.1Eosinophils, % 3.8 � 0.5 4.2 � 0.7 5.3 � 0.9 6.4 � 1.3 6.4 � 1.3 8.1 � 1.1Eosinophils, 106/mL 0.1 � 0.02 0.2 � 0.04 0.3 � 0.04 0.4 � 0.06 0.4 � 0.07 0.5 � 0.06Macrophages, % 60.6 � 2.7 50.9 � 2.6 50.2 � 2.9 51.8 � 5.7 60.9 � 7.1 73.6 � 3.0Macrophages, 106/mL 2.3 � 0.3 2.9 � 0.4 2.7 � 0.3 3.3 � 0.3 4.5 � 0.6 4.7 � 0.5Lymphocytes, % 4.2 � 0.7 4.5 � 0.7 3.8 � 0.9 4.9 � 0.8 4.7 � 0.5 4.5 � 1.7Lymphocytes, 106/mL 0.1 � 0.02 0.2 � 0.04 0.2 � 0.05 0.3 � 0.05 0.4 � 0.05 0.3 � 0.1

*Data are presented as mean � SEM for the patients indicated, unless otherwise indicated.

www.chestjournal.org CHEST / 132 / 1 / JULY, 2007 101

 © 2007 American College of Chest Physicians by guest on March 8, 2012chestjournal.chestpubs.orgDownloaded from

2,690.5 � 154.7 ng/mL vs 1,420.7 � 149.2 ng/mL,p � 0.0001, respectively]. The increase in percent-age of neutrophils was also observed at the sameperiod of time (Table 4) [mean, 42.0 � 3.3% vs23.6 � 3.5%, p � 0.02; and 51.1 � 1.9% vs29.5 � 4.2%, p � 0.0009, respectively]. Sputumneutrophilia correlated positively with changes inIL-8 levels (Fig 3) [r2 � 0.76, p � 0.01].

No significant differences were observed inbaseline total sputum cell and differential cellcounts between the budesonide treatment group

and steroid withdrawal groups (Tables 2, 3). Incontrast, there were significant differences in spu-tum lymphocyte counts (p � 0.01) and percentageof macrophages (p � 0.0001) at the onset of exac-erbation (Table 4; Fig 2, center, right, D, andbottom, left, E).

Sputum Eosinophils

No significant differences in the percentages ofeosinophils and absolute eosinophil number in spu-

Figure 2. Changes in percentage of eosinophils (top, left, A) and number of eosinophils (top, right,B), percentage of lymphocytes (center, left, C), number of lymphocytes (center, right, D),percentage of macrophages (bottom, left, E), and number of macrophages (bottom, right, F) insputum collected from asthmatic subjects with and without an exacerbation. The exacerbation datawere derived from those who were treated with placebo (eight cases) and with budesonide (onecase), and the nonexacerbation data included asthmatic subjects who were treated with budesonide(nine cases) and with placebo (four cases). All data are shown as mean � SEM. Comparisons ofeosinophil numbers lymphocyte numbers and macrophage numbers in sputum were made betweenthe results 2 weeks prior to an exacerbation and those of nonexacerbation at week 8 and betweendata on the exacerbation visit and those of the week-10 nonexacerbation. See Figure 1 legend forexpansion of abbreviation not used in text.

102 Original Research

 © 2007 American College of Chest Physicians by guest on March 8, 2012chestjournal.chestpubs.orgDownloaded from

tum were observed either at the onset of exacerba-tion or 2 weeks before an exacerbation, comparedwith nonexacerbation data (Table 4, Fig 2, top, left,A, and top, right, B).

Discussion

We evaluated the use of sputum neutrophils andIL-8 as predictors for the development of asthmaexacerbations. Our study indicates that neutrophils,rather than eosinophils, are associated with asthma

exacerbations following steroid withdrawal. This mayresult from the increase in sputum IL-8 observedprior to the onset of an exacerbation. Taken together,our findings demonstrated that sequential monitor-ing of sputum IL-8 and neutrophils might be usefulfor predicting an episode of severe exacerbation inmoderate persistent asthma after corticosteroidwithdrawal. This study also confirms that the mech-anisms underlying asthma exacerbation are highlyvariable.2,12,15–17

There is growing evidence that reducing asthmaexacerbations is achieved by targeting therapy toeosinophilic inflammation. An eosinophilic exacerba-tion is triggered by allergen exposure. However,several studies3–5,18 have demonstrated increasedneutrophil numbers in airway secretions and airwaytissue from patients with acute severe asthma. Spu-tum neutrophilia may be mediated by IL-8, a che-mokine that selectively attracts neutrophils. IL-8 isincreased in sputum of asthmatic subjects duringexacerbations in association with sputum neutrophil-ia.3,15,16 However, there is no clear relationshipbetween the type of airway inflammation and asth-matic exacerbations,19 possibly due to differences inthe methodology, specifically the use of steroidtapering.12 Furthermore, the pathophysiologicmechanisms underlying asthma exacerbations fol-lowing abrupt steroid withdrawal in moderate per-sistent asthmatics may be distinct from those seenwith gradual steroid tapering. The neutrophilic air-way inflammation in the present study may reflectthe sudden withdrawal of inhaled corticosteroids and

Figure 3. Relationship between changes in IL-8 concentrationsand neutrophil numbers in sputum obtained from the asthmaticpatients with severe exacerbations. Symbols represent individualasthmatic patients. IL-8 concentrations correlated positively andsignificantly with sputum neutrophilia.

Table 4—Measurements of Sputum Indices of Exacerbation and Nonexacerbation Groups*

VariablesExacerbation

(n � 9)Nonexacerbation

(n � 13) p Value95% Confidence Interval

of Difference

IL-8, ng/mL2 wk before visit 1,832.1 � 151 1,074.3 � 111.0 0.001 � 1,159.4 to � 356.3Last visit 2,690.5 � 154.7 1,420.7 � 149.2 � 0.0001 � 1,753 to � 786.8

Total cell counts, 106

2 wk before visit 5.1 � 0.9 4.65 � 0.5 0.69 � 1.77 to 2.61Last visit 8.5 � 0.8 4.63 � 0.5 0.0007 1.84 to 5.89

Neutrophils, %2 wk before visit 42.0 � 3.3 23.6 � 3.5 0.02 � 29.2 to � 7.4Last visit 51.1 � 1.9 29.5 � 4.2 0.0009 � 33.2 to � 9.99

Eosinophils, %2 wk before visit 4.3 � 0.4 2.8 � 1.1 0.24 � 4.1 to 1.1Last visit 2.7 � 0.6 2.7 � 1.0 0.99 � 2.96 to 2.99

Lymphocytes, %2 wk before visit 5.1 � 0.7 3.1 � 0.8 0.09 � 4.5 to 0.36Last visit 3.6 � 0.7 1.5 � 0.8 0.08 � 4.48 to 0.29

Macrophages, %2 wk before visit 50.3 � 3.8 70.1 � 3.5 0.0003 11.2 to 32.0Last visit 41.6 � 2.2 68.2 � 3.2 � 0.0001 17.6 to 35.7

*Data are presented as mean � SEM unless otherwise indicated.

www.chestjournal.org CHEST / 132 / 1 / JULY, 2007 103

 © 2007 American College of Chest Physicians by guest on March 8, 2012chestjournal.chestpubs.orgDownloaded from

reflect the kinetics of inflammatory cells in an acuteexacerbation when neutrophils are the first cells toinfiltrate the airways.

There are several studies12,17,19,20 suggesting thateosinophilic inflammation is associated with sputumasthma exacerbations in response to controlled with-drawal of steroids. In addition, targeting manage-ment based on normalizing sputum eosinophils isable to reduce asthma exacerbations.21–23 Unexpect-edly, the present study showed that a small numberof patients with steroid withdrawal did not haveasthma exacerbations, although they had eosinophilicinflammation. The patients in whom the dose ofinhaled budesonide was maintained showed a reduc-tion in sputum eosinophils, which may indicategreater compliance with inhaled corticosteroids dur-ing the trial period. The difference in the relation-ship between sputum eosinophils and asthma exac-erbation with the majority of current evidence is noteasily understood. We suggest that this may be dueto the fact that a 10-week period in our study isinadequate to monitor the development of eosino-philic exacerbation. This is supported by the study ofPizzichini and colleagues,24 who showed that break-through eosinophilic inflammation occurred at 4weeks following steroid reduction and subsequentasthma exacerbations developed within 12 weeksthereafter. Clinical exacerbation depends on theperception of breathlessness that is impaired byeosinophilic inflammation.25 Patients with eosino-philic inflammation who had no clinical worseningwere moderately severe in terms of FEV1 and mighthave impaired perception of dyspnea and thereforewere not aware of worsening asthma. Although thestrategy of titrating the inhaled corticosteroid doseagainst sputum eosinophil counts reduced exacerba-tions, it by no means eliminated them. It is possiblethat the residual exacerbations were neutrophilic innature.

Our results differ from those of Jatakanon et al20

and in’t Veen and colleagues,12 who induced loss ofcontrol in asthma by gradual tapering of the steroiddose, and both studies showed eosinophil predomi-nance during exacerbations. As suggested above,these results suggest that the pathophysiologic mech-anism(s) of asthma exacerbation induced by com-plete withdrawal may be different from that inducedby steroid tapering. This may have important clinicalconsequences in that patients with a loss of asthmacontrol as a result of stepping down inhaled cortico-steroid dosage should be able to attenuate eosinophilactions by increasing the inhaled steroid dose.23,26 Incontrast, exacerbations induced after abrupt steroidcessation may need antineutrophilic agents such aslong-acting �2-agonists to attenuate the neutrophilicinflammation.27

Increases in sputum IL-8 concentrations havebeen reported to precede an acute asthma attack,28

but the time course of this increase is unknown. Inanother study15 there was a significant elevation ofIL-8 levels in sputum samples from asthmatics whohad exacerbated compared to IL-8 concentrationsafter the resolution of the exacerbation. However,the onset of the elevation of IL-8 levels prior to thedevelopment of asthma exacerbation has not beenpreviously investigated. The present study demon-strates that a marked increase in IL-8 level occurred2 weeks before the symptomatic exacerbation (Fig 1,top left, A). Clinically, this may have importantimplications because there appears to be a windowduring which it may be possible to adjust asthmatreatment, such as the addition of long-acting �2-agonist, in an attempt to suppress IL-8 productionand subsequent neutrophilia leading to exacerbation.Although corticosteroids promote neutrophil survivalas a result of delay in neutrophil apoptosis in vitro,29

we did not detect sputum neutrophilia in patientswith budesonide treatment despite a failure to sup-press sputum IL-8 levels below that seen in therun-in period.

In summary, the findings in this study in contrastto earlier reports suggest that asthma exacerbationsfollowing inhaled corticosteroid withdrawal are asso-ciated with increased sputum IL-8 and neutrophils.Significant increase in sputum IL-8 and neutrophilinflux occurs 2 week prior to an exacerbation. Ourresults may be used to encourage asthmatic patientswho have poor compliance with inhaled corticoste-roid therapy are at risk for neutrophilic exacerbationif they abruptly stop their corticosteroid therapy.

References1 Brightling CE, Ward R, Woltmann G, et al. Induced sputum

inflammatory mediator concentrations in eosinophilic bron-chitis and asthma. Am J Respir Crit Care Med 2000; 162:878–882

2 Gibson PG, Norzila MZ, Fakes K, et al. Pattern of airwayinflammation and its determinants in children with acutesevere asthma. Pediatr Pulmonol 1999; 28:261–270

3 Fahy JV, Kim KW, Liu J, et al. Prominent neutrophilicinflammation in sputum from subjects with asthma exacerba-tion. J Allergy Clin Immunol 1995; 95:843–852

4 Lamblin C, Gosset P, Tillie-Leblond I, et al. Bronchialneutrophilia in patients with noninfectious status asthmaticus.Am J Respir Crit Care Med 1998; 157:394–402

5 Sur S, Crotty TB, Kephart GM, et al. Sudden onset fatalasthma: a distinct entity with few eosinophils and relativelymore neutrophils in the airway submucosa. Am Rev RespirDis 1993; 148:713–719

6 Busse WW, Lemanske RF. Asthma. N Engl J Med 2001;344:350–362

7 Barnes PJ. Cytokine-directed therapies for the treatment ofchronic airway diseases. Cytokine Growth Factor Rev 2003;14:511–522

104 Original Research

 © 2007 American College of Chest Physicians by guest on March 8, 2012chestjournal.chestpubs.orgDownloaded from

8 Kim CK, Kim JT, Kang H, et al. Sputum eosinophilia incough-variant asthma as a predictor of the subsequent devel-opment of classic asthma. Clin Exp Allergy 2003; 33:1409–1414

9 Johnston SL, Papi A, Monick MM, et al. Rhinoviruses induceinterleukin-8 mRNA and protein production in human mono-cytes. J Infect Dis 1997; 175:323–329

10 Kanazawa H, Kurihara N, Otsuka T, et al. Clinical signifi-cance of serum concentration of interleukin 8 in patients withbronchial asthma or chronic pulmonary emphysema. Respi-ration 1996; 63:236–240

11 Gilroy DW, Lawrence T, Perretti M, et al. Inflammatoryresolution: new opportunities for drug discovery. Nat RevDrug Discov 2004; 3:401–416

12 in’t Veen JC, Smits HH, Hiemstra PS, et al. Lung functionand sputum characteristics of patients with severe asthmaduring an induced exacerbation by double-blind steroidwithdrawal. Am J Respir Crit Care Med 1999; 160:93–99

13 Sterk PJ, Fabbri LM, Quanjer PH, et al. Airway responsive-ness: standardized challenge testing with pharmacological,physical and sensitizing stimuli in adults. Report WorkingParty Standardization of Lung Function Tests, EuropeanCommunity for Steel and Coal; Official Statement of theEuropean Respiratory Society. Eur Respir J Suppl 1993;16:53–83

14 Pizzichini MM, Pizzichini E, Efthimiadis A, et al. Asthma andnatural colds: inflammatory indices in induced sputum; afeasibility study. Am J Respir Crit Care Med 1998; 158:1178–1184

15 Norzila MZ, Fakes K, Henry RL, et al. Interleukin-8 secre-tion and neutrophil recruitment accompanies induced spu-tum eosinophil activation in children with acute asthma. Am JRespir Crit Care Med 2000; 161:769–774

16 Ordonez CL, Shaughnessy TE, Matthay MA, et al. Increasedneutrophil numbers and IL-8 levels in airway secretions inacute severe asthma: clinical and biologic significance. Am JRespir Crit Care Med 2000; 161:1185–1190

17 Pizzichini MM, Pizzichini E, Clelland L, et al. Sputum insevere exacerbations of asthma: kinetics of inflammatory

indices after prednisone treatment. Am J Respir Crit CareMed 1997; 155:1501–1508

18 Basyigit I, Yildiz F, Ozkara SK, et al. Inhaled corticosteroideffects both eosinophilic and non-eosinophilic inflammationin asthmatic patients. Mediat Inflamm 2004; 13:285–291

19 Gibson PG, Wong BJ, Hepperle MJ, et al. A research methodto induce and examine a mild exacerbation of asthma bywithdrawal of inhaled corticosteroid. Clin Exp Allergy 1992;22:525–532

20 Jatakanon A, Lim S, Barnes PJ. Changes in sputum eosino-phils predict loss of asthma control. Am J Respir Crit CareMed 2000; 161:64–72

21 Green RH, Brightling CE, McKenna S, et al. Asthma exac-erbations and sputum eosinophil counts: a randomised con-trolled trial. Lancet 2002; 360:1715–1721

22 Jayaram L, Pizzichini MM, Cook RJ, et al. Determiningasthma treatment by monitoring sputum cell counts: effect onexacerbations. Eur Respir J 2006; 27:483–494

23 Chlumsky J, Striz I, Terl M, et al. Strategy aimed at reductionof sputum eosinophils decreases exacerbation rate in patientswith asthma. J Int Med Res 2006; 34:129–139

24 Pizzichini MM, Pizzichini E, Clelland L, et al. Prednisone-dependent asthma: inflammatory indices in induced sputum.Eur Respir J 1999; 13:15–21

25 Veen JC, Smits HH, Ravensberg AJ, et al. Impaired percep-tion of dyspnea in patients with severe asthma: relation tosputum eosinophils. Am J Respir Crit Care Med 1998;158:1134–1141

26 Barnes PJ, Adcock IM. How do corticosteroids work inasthma? Ann Intern Med 2003; 139:359–370

27 Maneechotesuwan K, Essilfie-Quaye S, Meah S, et al. For-moterol attenuates neutrophilic airway inflammation inasthma. Chest 2005; 128:1936–1942

28 Kurashima K, Mukaida N, Fujimura M, et al. Increase ofchemokine levels in sputum precedes exacerbation of acuteasthma attacks. J Leukoc Biol 1996; 59:313–316

29 Meagher LC, Cousin JM, Seckl JR, et al. Opposing effects ofglucocorticoids on the rate of apoptosis in neutrophilic andeosinophilic granulocytes. J Immunol 1996; 156:4422–4428

www.chestjournal.org CHEST / 132 / 1 / JULY, 2007 105

 © 2007 American College of Chest Physicians by guest on March 8, 2012chestjournal.chestpubs.orgDownloaded from

DOI 10.1378/chest.06-2982; Prepublished online June 5, 2007; 2007;132; 98-105Chest

Ian M. Adcock and Peter J. BarnesKittipong Maneechotesuwan, Sarah Essilfie-Quaye, Sergei A. Kharitonov,

*NeutrophilsWithdrawal Is Associated With Increased Sputum Interleukin-8 and

Loss of Control of Asthma Following Inhaled Corticosteroid

 March 8, 2012This information is current as of

 

http://chestjournal.chestpubs.org/content/132/1/98.full.htmlUpdated Information and services can be found at:

Updated Information & Services

http://chestjournal.chestpubs.org/content/132/1/98.full.html#ref-list-1This article cites 29 articles, 15 of which can be accessed free at:

References

http://chestjournal.chestpubs.org/content/132/1/98.full.html#related-urlsThis article has been cited by 4 HighWire-hosted articles:

Cited Bys

http://www.chestpubs.org/site/misc/reprints.xhtmlfound online at: Information about reproducing this article in parts (figures, tables) or in its entirety can bePermissions & Licensing

http://www.chestpubs.org/site/misc/reprints.xhtmlInformation about ordering reprints can be found online:

Reprints

"Services" link to the right of the online article.Receive free e-mail alerts when new articles cite this article. To sign up, select the

Citation Alerts

PowerPoint slide format. See any online figure for directions. articles can be downloaded for teaching purposes inCHESTFigures that appear in Images in PowerPoint format

 © 2007 American College of Chest Physicians by guest on March 8, 2012chestjournal.chestpubs.orgDownloaded from