Thorax 1988;43:684-692

Number and activity of inflammatory cells inbronchoalveolar lavage fluid in asthma and theirrelation to airway responsivenessC KELLY, C WARD, C S STENTON, G BIRD, D J HENDRICK, E H WALTERS

From the Departments ofMedicine and Immunology, Newcastle General Hospital, Newcastle Upon Tyne

ABSTRACT Bronchial responsiveness to inhaled methacholine was measured four to six days beforefibreoptic bronchoscopy in 22 asthmatic patients (10 smokers) and 20 control subjects (12 smokers).The asthmatic patients had a baseline FEV, greater than 60% predicted and a PD20FEV, (provocativedose of methacholine causing a 20% fall in FEV,) of 0 006-3-7 mg. The 20 control subjects hadnormal pulmonary function and a PD20FEV, above the maximum cumulative dose of methacholineof 6-4 mg. Bronchoalveolar lavage ofa middle lobe segment (lingula in four subjects) was performedwith three sequential 60 ml aliquots of sterile saline. Cellular metabolic activity was stimulated withlatex in aliquots of resuspended cells, and measured by means of luminol enhanced chemilumines-cence to assess neutrophil activity and lucigenin enhanced chemiluminescence to assess macrophageactivity. Mean absolute total cell counts were similar in the asthmatic and control groups but therewere differences in differential cell counts, with a significant increase in eosinophil (p < 0 05) andlymphocyte (p < 0 05) counts in asthma. PD20FEV, was negatively correlated with percentageneutrophil counts (p < 0005). Luminol enhanced chemiluminescence/1000 neutrophils wasincreased about twofold in asthmatic subjects (p < 0 001), but was not correlated with PD20FEV1.Lucigenin enhanced chemiluminescence/1000 macrophages was increased nearly fourfold inasthmatic patients (p < 0-001) and showed a negative correlation with PD20FEV, (p < 0-01). Themacrophage count was increased twofold in current smokers in both groups, but other cell numberswere not altered significantly. Smoking did not affect cellular metabolic activity in either group. Thisstudy supports the idea that an inflammatory process is present in the airways of those with asthma,and suggests a relation between bronchial responsiveness and both neutrophil numbers andmacrophage activity.

Introduction

Inflammatory change within the airways, recognisedas a histopathological feature of asthma for manyyears,' has more recently been proposed as animportant factor in the development of bronchialhyperresponsiveness in man.2 Studies of the develop-ment of cellular inflammation in animal airways afterexposure to ozone4 and specific allergen5 have shownan association between the induction of transientairway hyperresponsiveness and neutrophil influx.When the cellular inflammatory response wasprevented by prior neutrophil depletion the develop-ment of bronchial hyperresponsiveness was

Address for reprint requests: Dr E H Walters, Chest Clinic, NewcastleGeneral Hospital, Newcastle upon Tyne NE4 6BE.

Accepted 25 May 1988

inhibited.6 The importance of an inflammatory cellinflux in the development of increased airwayresponsiveness has, however, been disputed in otheranimal models.8Lavage has now been performed in patients with

asthma in several centres, and has been generally welltolerated.9 Increased numbers of eosinophils havebeen recovered in lavage fluid from patients withallergic asthma"'2 and during allergen induced lateasthmatic reactions.'3"' Increase in neutrophils hasalso been found in lavage fluid from patients withasthma before914 and after allergen challenge'24 andafter inhalation of ozone'" and toluene diisocyanate,'5and mast cell numbers are increased in lavage fluid inatopic patients'6'5 and may be related to bronchialresponsiveness.'7

Circulating neutrophils from asthmatic patientsmay be activated after challenge with allergen,'920 and

684

on May 13, 2020 by guest. P

rotected by copyright.http://thorax.bm

j.com/

Thorax: first published as 10.1136/thx.43.9.684 on 1 S

eptember 1988. D

ownloaded from

Number and activity ofinflammatory cells in bronchoalveolar lavagefluid in asthmathis increase in activation may be inhibited bysteroids.2' The degree of activation of pulmonaryneutrophils has not been investigated, but increasingin vivo and in vitro evidence suggests that alveolarmacrophage function may be enhanced in asthma.2223There is little published evidence, however, of arelation between the activity of inflammatory cells inthe airway and airway hyperresponsiveness, a featurerelated to the patient's symptoms and treatmentrequirements.24 Luminol25 and lucigenin26 enhancedchemiluminescence have been used as markers ofreactive oxygen species released during the metabolicburst from neutrophils and macrophages respectivelyand have been adopted in this study as measures of cellfunction.The aims of this study were (1) to compare the cell

profile of lavage fluid and the metabolic activity ofpulmonary neutrophils and macrophages in 22patients with stable asthma showing a wide range ofbronchial responsiveness and in 20 non-asthmaticsubjects;- (2) to relate the numbers and activity ofinflammatory cells obtained from lavage fluid in thepatients with asthma to airway responsivenessmeasured as the provocative dose of methacholinecausing a 20% fall in FEV, (PD20FEV,).

Methods

SUBJECTSWe studied 22 patients with stable asthma and a widerange of PD20FEV, values in response to inhaledmethacholine and 20 control subjects with no evidenceof airflow obstruction. The mean age of the asthmaticpatients was 47 (range 17-73) years and of the controls53 (range 21-71) years. Ten asthmatic and 12 controlsubjects were smokers. Ethical approval for the studywas obtained from the Newcastle Area HealthAuthority's ethical committee, with the proviso thatall subjects must have a clinical indication forbronchoscopy. The nature of the investigation anddetails of the procedure were fully explained to eachindividual, and written consent was obtained.Indications for bronchoscopy among the asthmaticpatients included streaking haemoptysis (13), unduecough (8), and possible stridor (1). No asthmaticpatient had evidence, before or after bronchoscopy, ofa neoplasm, infection, or infarction, and none hadrecognised bronchiectasis. The control subjects hadeither a peripheral radiographic abnormality (9),unexplained haemoptysis (6), or cough (5). Four ofthis group proved to have lung cancer, always on theside opposite to the one that was lavaged. The studyfollowed internationally agreed guidelines27 and allsubjects fulfilled the following criteria: (1) age 16-75years, with an FEV, over 60% of the predicted value;(2) no oral corticosteroids for at least the previous

three months; (3) no evidence of a chest or upperrespiratory tract infection within eight weeks.

Medication in the asthmatic patients consisted ofregular inhaled bronchodilators alone in nine; sevenothers had low dose ( < 400 pg/day) inhaled steroids inaddition. Six were receiving no regular treatment.Airway responsiveness to methacholine to a maximumcumulative dose of 6-4 mg was measured in all sub-jects four to six days before bronchoscopy by a stan-dardised inhalation challenge technique with amicroprocessor controlled dosimeter.' A logarithmicdose-response curve was constructed and thePD20FEV, obtained by linear interpolation. Medica-tion was withheld for 12 hours before this test.Atopy was assessed by performing skinprick tests

with house dust mite, grass pollens, and Aspergillusspecies, and any other antigen for which the subjectgave a history suggesting sensitisation. A positiveresult was recorded if any reaction was greater thanthat induced by the histamine control.

Spirometry was performed before and 30 minutesafter an inhaled dose of 400 pg of salbutamol, andsingle breath transfer factor for carbon monoxide(TLCO) was measured in all current or former tobaccosmokers. All the asthmatic patients had an increase inFEV, of over 15% after inhaling salbutamol, a normalor raised TLCO value, and a PD20 methacholine valueof less than 6-4 mg. Thirteen of the 22 asthmaticpatients were atopic. None of the control subjects hada measurable PD20FEV, with 6-4 mg inhaledmethacholine, and all had FEV, and TLCO values over90% of the values predicted for age and height. Foursubjects were atopic.

BRONCHOALVEOLAR LAVAGESubjects took their usual medication two to threehours before bronchoscopy, which was performedwith an Olympus OT1O bronchoscope. No additionalbronchodilator was given. Salbutamol was availablefor inhalation throughout the procedure, but wasrequired after bronchoscopy by only one subject.Asthmatic patients were admitted overnight afterlavage for observation. Premedication with intra-muscular atropine 0 6 mg and papaveretum 10mg wasgiven 30 minutes before the procedure and 4 ml 1-5%(isotonic) lignocaine was given via a Porta-neb(Medic-aid Ltd) nebuliser compressor unit for 10minutes before bronchoscopy. Further 2 ml aliquotsof 1-5% lignocaine were administered throughout thebronchoscopy as required. The mean total volume oflignocaine given was 12 (range 8-16) ml. After routineendobronchial examination, and before any speci-mens were obtained for cytological or histologicalexamination, the bronchoscope was usually wedged ina segment of the middle lobe, but in three controlsubjects with right sided neoplasia a segment of the

685

on May 13, 2020 by guest. P

rotected by copyright.http://thorax.bm

j.com/

Thorax: first published as 10.1136/thx.43.9.684 on 1 S

eptember 1988. D

ownloaded from

Table 1 Mean (SD) percentage and absolute cell counts in bronchoalveolar lavagefluidfrom 22 asthmatic and 20 controlsubjects

EpithelialSubjects Total cells Macrophages Lymphocytes Neutrophils Eosinophils cells

Percentage countsControl 83-4(10 9) 12-4(9 9) 3-0(2 4) 0-2 (0-5) 1-0 (2 0)Asthmatic 71-8* (15-4) 20-8* (15-1) 4-4(3-3) 1-0* (1-2) 2-0 (2-2)Absolute counts (x 10'/l)Controls 1-88 (1-2) 1-61 (10) 0-19 (0-14) 0-06 (0-05) 0-003 (0-01) 0-02 (0-02)Asthmatics 1-90 (1-6) 1-42 (1-2) 0.34* (0 34) 0-08 (0-06) 0-022** (0-03) 0-04 (0-04)

*p < 0-05, **p < 0-001 in the comparison between asthmatic and control subjects (analysis of variance).

lingula was used. Lavage was performed with three 60ml aliquots of sterile phosphate buffered saline at37°C. The fluid was aspirated immediately into silicon-ised glassware by vacuum suction (50-100 mm Hg)and kept at 4°C until analysis. Supplementary oxygen(4 litres/min) was given throughout the procedure andcontinued for at least 30 minutes after its completion.

MORPHOLOGICAL STUDIESAnalysis was completed within two hours of lavage.The specimens were filtered through 200 gm2 sterilesteel mesh to remove mucus. Total cell counts werethen performed on each aspirate by two experiencedobservers using a Neubauer counting chamber andcounting a total of300 cells. The results were expressedas mean values of cells x 108/l. As the data wereskewed values were logarithmically transformed. Thegeometric mean ratio between the cells counts for eachobserver was 099 (95% CL = 094-1-05). Theaspirates were centrifuged at 1200 rev/min for fiveminutes and the cell pellets resuspended in cell medium199 at a concentration of 5 x 108/I. Cell viability wasassessed by tryphan blue exclusion, and cytospinpreparations were made with a Shandon II Cytospin(Shandon Southern Instruments, Sewickley, PA)using 100 pl of cell suspension. The slides were airdried and stained with May-Grunwald-Giemsa and300 cells were counted by scanning several fields,moving diagonally across the cell preparation tominimise possible artefacts induced by centrifugation.The results were expressed as mean values in absolute( x I0'/1) and percentage terms. Monoclonalantibodies to cytokeratin were used to stain severalslides at random and confirmed the accuracy with

which epithelial cells were being counted in Giemsastained preparations.

MEASUREMENT OF CHEMILUMINESCENCEFive hundred microlitres of cell suspension, contain-ing a calculated 250 000 cells, were added to 900 p1 of10-4 M solutions of either luminol or lucigenin andwarmed to 37°C. After the addition of 100 p1 of 5%unopsonised latex particles (Sigma Chemicals),chemiluminescence was measured with an LKB 1250luminometer. Assays were performed in triplicate, andthe mean peak height was recorded in mv. This wasexpressed both as total chemiluminescence and aschemiluminescence per 1000 neutrophils for luminoland chemiluminescence per 1000 macrophages forlucigenin to exclude variations due to differing cellprofiles. Lucigenin enhanced chemiluminescence haspreviously been used to represent alveolar macro-phage function,29 and recent work has shown that, cellfor cell, alveolar macrophages produce almost twice asmuch lucigenin enhanced chemiluminescence aspolymorphs. In our mixed cell populations, whichconsisted largely of macrophages, lucigenin enhancedchemiluminescence will predominantly reflectmacrophage activity.' Luminol enhancedchemiluminescence has often been used to assessneutrophil function,25 and is not produced insubstantial amounts by alveolar macrophages.30 Thewithin subject variance for assay was 2% for lucigeninenhanced chemiluminescence and 3% for luminolenhanced chemiluminescence.

Analysis ofvariance was used to compare data fromasthmatic and control subjects, and for analysis of the

Table 2 Mean (SD) luminol and lucigenin chemiluminescence (CL) in cellsfrom bronchoalveolar lavagefluid in 22asthmatic and 20 control subjects

Total luminol CL (mv)t Total lucigenin CL (mv)t Luminol CLIIOOO neutrophils Lucigenin CL/I000 macrophagesAsthmatics Controls Asthmatics Controls Asthmatics Controls Asthmatics Controls

6-77 (6.2)* 2-04 (2 2) 14-93 (14-2)** 4.39 (4 7) 0-64 (0-35)** 0-28 (0 20) 0-079 (0.060)** 0-022 (0 023)

*p < 0-005, **p < 0-001 in the comparison between asthmatic and control subjects (analysis of variance).tMixed aliquots of 250 000 cells.

686 Kelly, Ward, Stenton, Bird, Hendrick, Walters

on May 13, 2020 by guest. P

rotected by copyright.http://thorax.bm

j.com/

Thorax: first published as 10.1136/thx.43.9.684 on 1 S

eptember 1988. D

ownloaded from

Number and activity ofinflammatory cells in bronchoalveolar lavagefluid in asthmaPercentageneutrophilcount

r = 0.60, p < 0.005

00

0

0

Log PD20Fig 1 Relation between bronchial responsiveness tomethacholine (log PD2J) andpercentage neutrophil counts inbronchoalveolar lavagefluid in 22 asthmatic patients, with theline ofregression. Patients not taking regular inhaledcorticosteroids: smokers 0; non-smokers 0; patients takingregular inhaled corticosteroids: smokers 4p; non-smokers 9.

effects of smoking. All data were logarithmicallytransformed to reduce skewness. Linear regressionswere computed by least squares analysis, afterlogarithmic transformation of PD20FEV, values.

Results

Bronchoalveolar lavage was well tolerated and none ofthe subjects developed wheezing. The median totalvolume aspirated at lavage was similar for theasthmatic (86 (range 11-128) ml) and control subjects(89 (range 32-130) ml). Cell viability was over 92% inall specimens.

Percentagelymphocytecount60-

50-

40-

30-

20-

10-

0-0

r = 0.43, p < 0.05

LucigeninCL(mv/1 000macrophages)

0.20-

0.10-

0

00 *

r = 0.52

0 p < 0.05

00

ot00

0 1 2Log PD20

3 4

Fig 3 Relation between bronchial responsiveness tomethacholine (log PD211) and lucigenin chemiluminescence(CL) (mv/1000 macrophages) from bronchoalveolar lavagefluid in 22 asthmatic patients, with the line of regression.Patients not taking regular inhaled corticosteroids: smokers0; non-smokers 0; patients taking regular inhaledcorticosteroids: smokers 4p; non-smokers 9 .

LAVAGE CELL PROFILESThe total cell counts and the absolute numbers ofmacrophages, neutrophils and epithelial cells did notdiffer significantly between the two groups (table 1).The absolute and percentage eosinophil counts,

LuminolCL(mv/1 000neutrophils)2-

0

1-* 0 *

00

0 0 0* *0 0

.

0

II

I 2 3 4Log PD20

Fig 2 Relation between bronchial responsiveness tomethacholine (log PD2,) andpercentage lymphocyte countsin bronchoalveolar lavagefluid in 22 asthmatic patients, withthe line ofregression. Patients not taking regular inhaledcorticosteroids: smokers 0; non-smokers 0; patients takingregular inhaled corticosteroids: smokers (p; non-smokers 9.

00

r = 0.17

lb

00 0

t * 4 0 t#00 00 4

0

1 2

NS

0

o 00

3 4

Log PD20Fig 4 Relation between bronchial responsiveness tomethacholine (log PD2S) and luminol chemiluminescence(CL) (mv/J000 neutrophils) from bronchoalveolar lavagefluid in 22 asthmatic patients. Patients not taking regularinhaled corticosteroids: smokers 0; non-smokers 0; patientstaking regular inhaled corticosteroids: smokers 4); non-smokers 9.

I I I I

687

on May 13, 2020 by guest. P

rotected by copyright.http://thorax.bm

j.com/

Thorax: first published as 10.1136/thx.43.9.684 on 1 S

eptember 1988. D

ownloaded from

Table 3 Effect ofsmoking habit on the mean (SD) absolute cell counts in bronchoalveolar lavagefluidfrom asthmatic andcontrol groups

Absolute counts (x HP1/i)

EpithelialTotal count Macrophages Lymphocytes Neutrophils Eosinophils cells

CONTROLSSmokers (n= 12) 2-32 (1.13) 2-02 (1.03) 0-19 (0-17) 0-08 (0-06) 0-005 (0 004) 0-03 (0.02)Non-smokers (n= 8) 1.49* (0 77) 0.99* (0 53) 0-19 (0-21) 0-03 (0.02) 0 0-02 (0 01)ASTHMATICSSmokers (n= 10) 2 58 (2.17) 1-98 (1-57) 0-41 (0-47) 0 09 (0.06) 0-03 (0-03) 0-07 (0-07)Non-smokers (n= 12) 1.33* (0.78) 0-94* (0-54) 0-28 (0.18) 0-06 (0-06) 0 01 (0-01) 0 04 (0 04)

*p < 0-001 in the comparison between smokers and non-smokers in the asthmatic and control populations (analysis of variance).

however, were significantly greater in the asthmaticpatients than in the control subjects (p < 0-05), aswere absolute and percentage lymphocyte counts(p < 0.05). Percentage macrophage counts were lowerin the asthmatic patients than in the control subjects(p < 0-05), reflecting the other changes.

METABOLIC CELLULAR ACTIVITYMean total luminol enhanced chemiluminescence wasgreater in the cells obtained from asthmatic patientsthan in those from control subjects. Mean lucigeninenhanced chemiluminescence was also greater in cellsfrom asthmatic patients than in cells from controlsubjects. These differences are also highly significantwhen expressed as chemiluminescence per 1000neutrophils and per 1000 macrophages respectively(table 2).

RELATION BETWEEN CELL NUMBERS, FEVI, ANDBRONCHIAL RESPONSIVENESSInitial FEV, was not correlated significantly withPD20FEV1, cell counts, or indices of cellular activity ineither group. In the asthmatic patients PD20FEV, wascorrelated negatively with percentage neutrophilcounts (fig 1) but not with absolute neutrophil counts(r = -0. 17). There was a significant positive correla-tion between PD20FEV, and percentage lymphocyte

Table 4 Effect ofsmoking habit on the mean (SD) luminol(mv/1000 neutrophils) and lucigenin (mv/1000macrophages) chemiluminescence (CL) from cells obtainedby bronchoalveolar lavage in 22 asthmatic (10 smokers) and20 control subjects (12 smokers)

Luminol CLIIOOO Lucigenin CL/IOOOneutrophils macrophages

Asthmatics Controls Asthmatics Controls

Smokers 0 64 (0 44) 0-25 (0-18) 0 066 (0.057) 0-020 (0-024)Non-

smokers 0-63 (0-27) 0 30 (025) 0-091 (0-065) 0-024 (0-022)

None of the differences between smokers and non-smokers wassignificant (p > 0 60).

counts (fig 2) but no significant correlations betweenPD20FEV, and macrophage (r = 0-24), eosinophil(r = 0-30), or epithelial cell numbers (r = 0-03).

RELATION BETWEEN CELL ACTIVITY ANDBRONCHIAL RESPONSIVENESSThere was a significant negative correlation betweenbronchial responsiveness and lucigenin chemilumines-cence per 1000 macrophages (fig 3), but not betweenbronchial responsiveness and luminol chemilumines-cence per 1000 neutrophils (fig 4).

EFFECT OF SMOKING ON CELL NUMBERSSmoking was associated with an increased total cellcount in each group (table 3; p < 0-001). This wasalmost entirely due to an increased number of macro-phages (p < 0-001). There were more neutrophils andeosinophils in smokers than in non-smokers, but thesedifferences did not reach statistical significance.

EFFECT OF SMOKING ON CELL ACTIVITYTable 4 shows that smoking had no effect on eitherluminol enhanced or lucigenin enhanced chemi-luminescence. The relations between PD20FEV,and chemiluminescence were also uninfluenced bysmoking.

EFFECT OF INHALED STEROIDS ON CELL NUMBERSAND ACTIVITYThe asthmatic patients who regularly inhaled steroidsdid not differ in PD20FEV, values, cell numbers, orchemiluminescence values from the asthmatic groupas a whole (fig 1-4).

EFFECT OF ATOPY ON CELL NUMBERS ANDACTIVITYAtopy had no significant influence on cell numbers oron luminol enhanced or lucigenin enhancedchemiluminescence.

Discussion

In this study we have shown a significant increase in

688 Kelly, Ward, Stenton, Bird, Hendrick, Walters

on May 13, 2020 by guest. P

rotected by copyright.http://thorax.bm

j.com/

Thorax: first published as 10.1136/thx.43.9.684 on 1 S

eptember 1988. D

ownloaded from

Number and activity ofinflammatory cells in bronchoalveolar lavagefluid in asthmathe numbers ofeosinophils and lymphocytes obtainedin lavage fluid from patients with stable asthma; theincrease in lavage fluid neutrophils was non-sig-nificant, though the absolute numbers of neutrophilswere much greater than the numbers of eosinophils.The metabolic activity ofboth neutrophils and macro-phages obtained at lavage, assessed by stimulatedchemiluminescence, was increased in our asthmaticpatients.An increase in eosinophils in lavage fluid has been

described in patients with stable asthma by Godard etal9 and Tomioka et al,'6 and by others in response to achemotactic or allergic stimulus.'314 A correlationbetween eosinophil numbers and airway responsive-ness has also been reported in a small group ofindividuals with atopic asthma.'8 Our subjects weregenerally older than those in the study by Kirby et al'8(47 v 31 years). In our study eosinophils were no morenumerous in atopic than in non-atopic asthmaticpatients and their numbers in lavage fluid did notcorrelate significantly with bronchial responsiveness.We did, however, find a significant correlation be-tween percentage neutrophil count and bronchialresponsiveness in the asthmatic patients, a relation-ship not previously reported,'" although both Fabbriet al' and Seltzer et al,'° using isocyanate and ozonechallenges, have described large influxes of neutro-phils into the airways in association with the inductionof bronchial hyperresponsiveness. Both eosinophilsand neutrophils are capable of secreting widelyvarying inflammatory mediators, including leuko-triene B4, prostaglandins, and platelet activatingfactor; but it is not yet clear which has thepredominant role in inducing hyperresponsiveness inthe inflammatory reaction found in the airways inman.

Previous studies9"6 have reported an increase inlymphocyte numbers in stable asthma, Godard et al9finding an almost twofold increase in the percentage oflymphocytes in lavage fluid in patients who weresimilar to ours in mean age and in the amount oflavage fluid recovered. We found a significant positivecorrelation between the numbers of lymphocytesrecovered in lavage fluid and PD20FEV,. Circulatinglymphocytes have been reported to be activated inasthma3' and pulmonary lymphocytes are nowthought to play some part in mast cell recruitment.32Recently the amount of histamine releasing factorproduced by lymphocytes has been shown to becorrelated with bronchial responsiveness.33 Diaz et alhave shown changes in lymphocyte subsets afterallergen challenge,34 suggesting a possible immunemodulating effect. Further studies of lymphocytemarkers and function in asthma may clarify their rolein this condition.

Like most other workers, we were unable to show

any differences in absolute macrophage numbersbetween asthmatic and control subjects. Godard et al,9like us, reported a decrease in percentage macrophagecounts in asthmatic patients, but his absolute numbersdid not differ significantly from control values.We did not count the numbers of mast cells present

in our cytospin preparations as estimates of mast cellnumbers from slides not specifically prepared for thispurpose might have been inaccurate. Several differenttechniques have been used to fix and stain mast cells,and the numbers recorded in lavage fluid from patientswith stable asthma has varied considerably (from0.25%16 to 3%'7). Although mast cells probablyplay an important part in the development of theearly asthmatic reaction, and their numbers in lavagefluid may be correlated with bronchial responsive-ness,'718 their role in chronic asthma remains to beestablished.32

Seven of our asthmatic patients were regularlyinhaling corticosteroids. We could detect nodifferences in airway responsiveness or lavage resultsbetween these patients and the asthmatic subjects as awhole. We are currently investigating their effects onthe numbers and activation of cells obtained from theairways of asthmatic patients in more detail.We have previously shown that filtration of lavage

fluid to remove mucus can also reduce cell counts,largely through modest loss of macrophages.35Accurate cell counts and functional studies aredifficult to perform in the presence of mucus, and thesmall resulting change in cell counts was considered anecessary expedient. A chemically inert steel mesh wasused rather than the more frequently used cottongauze to avoid the risk of cell activation. As lavagefluid from all subjects was treated in an identicalmanner, routine filtration is unlikely to haveintroduced major inaccuracies.The production of superoxide radicals by macro-

phages is an important microbiocidal mechanism36that has been shown to occur after sensitisation withserum from asthmatic patients.37 Release of super-oxide radicals contributes to inflammation38 and theyare thought to be predominantly responsible forlucigenin enhanced chemiluminescence,39 which hasnow been used to assess alveolar macrophage activityin several conditions." Lucigenin enhancedchemiluminescence is abolished by superoxidedismutase (unpublished observation). The presentresults indicate that alveolar macrophage activity isdirectly related to the degree ofairway responsiveness,and suggest a pivotal role for this cell in thepathophysiology of asthma. Evidence that alveolarmacrophages have a role in asthma has previouslybeen suggested by the finding of increased levels of fglucuronidase4' and platelet activating factor22 inlavage fluid after local antigen challenge. These

689

on May 13, 2020 by guest. P

rotected by copyright.http://thorax.bm

j.com/

Thorax: first published as 10.1136/thx.43.9.684 on 1 S

eptember 1988. D

ownloaded from

690

mediators are thought to be released after macrophageactivation via IgE receptors, which have been found ingreater numbers on the surface of the alveolar macro-phages ofasthmatic patients than ofcontrol subjects.23Luminol enhanced chemiluminescence is myelo-

peroxidase dependent, and greatly depressed inpatients with myeloperoxidase deficient leucocytes.42A recent study in which luminol enhanced chemi-luminescence was increased in lavage fluid cells fromasthmatic patients was interpreted as indicatingincreased macrophage activity.43 Granulocyte num-bers, however, were closely correlated with luminolenhanced chemiluminescence, suggesting that activityof these cells, and not macrophages, was beingmeasured. This would be in keeping with our findings'"and those of Williams and Cole.25We also found luminol enhanced chemilumines-

cence to be increased in patients with asthma, andinterpret this as indicating neutrophil activation.There is already some evidence suggesting activationof neutrophils in asthma. Neutrophil complementrosettes were increased during both early and lateasthmatic reactions in patients challenged withallergen.920 These changes were immediately precededby a rise in circulating neutrophil chemotactic factor inboth studies, and could be reversed by corticosteroidtreatment,2' which also caused a decrease inleukotriene release by stimulated neutrophils.4 Ourdata suggest that neutrophil numbers but not activityare directly related to the degree of bronchialresponsiveness. Neutrophils appear to be uniformlyactive over a wide range of disease activity.

It has been suggested that cellular activity may bedepressed in the presence of neoplasia, so thepossibility that chemiluminescence was reduced in ourfour control subjects who had lung cancer must beconsidered, though the contralateral lung was lavagedin each case. The chemiluminescence results for thesesubjects were similar to the mean chemiluminescencevalues for the control group, so neoplasia in these foursubjects is unlikly to have affected the groupcomparisons.

Cigarette smoking is known to cause a twofold tofourfold increase in absolute lavage fluid cell counts,mainly owing to an increase in macrophages.45 In thepresent study there was a twofold increase inmacrophage numbers in smokers in both the controland the asthmatic group. There were more neutrophilsin smokers than in non-smokers, and among non-smokers the neutrophil count was higher in theasthmatic than the control subjects, though in bothcases the changes failed to reach statisticalsignificance. A smoking related increase in neutrophilsin lavage fluid has been described previously and maybe relevant to the development of flow limitation insmokers.'

Kelly, Ward, Stenton, Bird, Hendrick, WaltersSmoking did not alter luminol or lucigenin

enhanced chemiluminescence per 1000 neutrophils,but contributed to total chemiluminescence inasthmatic subjects by increasing cell numbers. Othershave found in vivo exposure to cigarette smoke to haveprofound effects on the numbers and function of bothpulmonary macrophages and circulating neutro-phils47 8 and, although alveolar macrophages fromsmokers appeared to be generally activated in vivo,certain functional characteristics are depressed bycigarette smoke.49 Alveolar macrophages fromsmokers have been reported to have increased glucoseconsumption.' Smoking had no effect on glucoseoxidation in another study,5' however, and, althoughthe production ofsuperoxide appeared to be greater inmacrophages obtained from the lungs of smokers,there was no concomitant increase in oxygen con-sumption per cell. More recent work has shown asignificant increase in superoxide production fromsmokers' macrophages only after specific chemicalstimulation.52 Our data suggest that smoking has noeffect on superoxide generation from macrophagesstimulated with 5% latex.

This study has shown correlations between thedegree of airway responsiveness and the numbers andactivity of certain inflammatory cells in lavage fluidfrom a group of patients with asthma. Although thefunction of inflammatory cells in asthmatic airways islikely to be complex, our findings are consistent with aprimary effector role for alveolar macrophages inasthma. These have the potential to induce an influx ofneutrophils into the airways by the release ofchemotactic factors.53 Once the neutrophil reaches theairway it appears to be activated to a similar extentacross a wide range of airway responsiveness.

We wish to acknowledge the support of the AsthmaResearch Council and the District ResearchCommittee for Newcastle upon Tyne.

References

1 Dunnill MS. The pathology of asthma with specialreference to changes in the bronchial mucosa. J ClinPathol 1960;13:27-33.

2 Chung KF. Role of inflammation in the hyperreactivityof the airways. Thorax 1986;41:657-62.

3 Walters EH, Parrish RW, Bevan C, Smith AP. Inductionof bronchial hypersensitivity: evidence for a role forprostaglandins. Thorax 1981;36:571-4.

4 Holtzman MJ, Fabbri LM, O'Byrne PM, et al. Im-portance of airway inflammation for hyperresponsive-ness induced by ozone. Am Rev Respir Dis 1983;127:686-90.

5 Marsh WR, Irvin CG, Behrens BL, Larsen GL.Pulmonary inflammation and changes in airwayreactivity after late asthmatic responses in an animalmodel. Am Rev Respir Dis 1984;129:4-9.

on May 13, 2020 by guest. P

rotected by copyright.http://thorax.bm

j.com/

Thorax: first published as 10.1136/thx.43.9.684 on 1 S

eptember 1988. D

ownloaded from

Number and activity ofinflammatory cells in bronchoalveolar lavagefluid in asthma6 O'Byrne PM, Walters EH, Gold BD, et al. Neutrophil

depletion inhibits airway hyper-responsivenessinduced by ozone exposure. Am Rev Respir Dis1984;130:214-9.

7 Murphy KR, Wilson MC, Irvin CG, et al. The require-ment for polymorphonuclear leucocytes in the lateasthmatic responses and heightened airways reactivityin an animal model. Am Rev Respir Dis 1986;134:62-8.

8 Murlas CG, Roum JH. Sequence of pathologic changesin airway mucosa of guinea pigs during ozone inducedbronchial hyperreactivity. Am Rev Respir Dis1985;131:314-20.

9 Godard P, Aubas P, Calvayrac P, Taib J, Michel FB.Endoscopie et lavage bronchoalveolaire chezl'asthmatique allergique. Nouv Press Med 1981;10:3141-8.

10 Seltzer J, Bigby B, Stulbarg M, et al. Ozone-inducedchange in bronchial reactivity to methacholine andairway inflammation in humans. J Appl Physiol1986;60: 1321-6.

11 Metzger WJ, Moseley P, Nugent K, Richerson HB,Hunninghake GW. Local antigen challenge andbronchoalveolar lavage of allergic asthmatic lungs.Chest 1985;87:155S-7S.

12 Metzger WJ, Richerson HB, Worden K, Monick M,Hunninghake GW. Bronchoalveolar lavage of allergicasthmatic patients following allergen broncho-provocation. Chest 1986;89:477-83.

13 Bruijnzeel PLB, de Monchy DJR, Verhagen J, KaufmannHF. The eosinophilic granulocyte-an activeparticipant in the late phase asthmatic reaction? ClinRespir Physiol 1986;suppl 7:54-61.

14 de Monchy JG, Kauffman H, Venge P, et al.Bronchoalveolar eosinophilia during allergen-inducedlate asthmatic reactions. Am Rev Respir Dis1985;131:373-6.

15 Fabbri LM, Boschetto P, Zocca E, et al. Bronchoalveolarneutrophilia during late asthmatic reactions induced bytoluene diisocyanate. Am Rev Respir Dis 1987;136:36-42.

16 Tomioka M, Ida S, Shindoh Y, Ishihara T, Takishima T.Mast cells in bronchoalveolar lavage of patients withbronchial asthma. Am Rev Respir Dis 1984;129: 1000-5.

17 Flint KC, Leung KBP, Hudspith BN, Brostoff J, PearceFL, Johnson NM. Bronchoalveolar mast cells inextrinsic asthma: a mechanism for the initiation ofantigen specific bronchoconstriction. Br Med J1985;291:923-6.

18 Kirby JG, Hargreave FB, Gleich G, O'Byrne PM.Bronchoalveolar cell profiles of asthmatic and non-asthmatic subjects. Am Rev Respir Dis 1987;136:379-83.

19 Durham SR, Carroll M, Walsh GM, Kay AB. Leukocyteactivation in allergen induced asthmatic reactions.N Engl J Med 1984;311: 1398-402.

20 Carroll MP, Durham SR, Walsh G, Kay AB. Activationof neutrophils and monocytes after allergen andhistamine induced bronchoconstriction. J Allergy ClinImmunol 1985;75:290-6.

21 Klempner NS, Gallin JI. Inhibition of neutrophil Fcreceptor function by corticosteroids. Clin Exp Immunol1978;34:137-42.

22 Arnoux B, Simoes Caeiro MH, Landes A, Mathieu M,Duroux P, Benveniste J. Alveolar macrophages fromasthmatic patients release platelet-activating factor(PAF-acether) and lyso-PAF-acether when stimulatedwith specific allergen [abstract]. Am Rev Respir Dis1982;125(suppl):70.

23 Tonnel A-B, Gosset P, Joseph M, Lassalle P, Dessant JP,Capron A. Alveolar macrophage and its participationin the inflammatory processes of allergic asthma. ClinRespir Physiol 1986;suppl 7:70-6.

24 Juniper EF, Frith PA, Hargreave FE. Airway responsive-ness to histamine and methacholine: relationship tominimum treatment to control symptoms of asthma.Thorax 1981 ;36:575-9.

25 Williams AJ, Cole PJ. Human bronchoalveolar lavagecells and luminol dependent chemiluminescence. J ClinPathol 1981 ;34:167-71.

26 Williams AJ, Cole PJ. Investigation of alveolarmacrophage function using lucigenin-dependentchemiluminescence. Thorax 1981 ;36:866-9.

27 Workshop on the investigative use of fibreopticbronchoscopy and bronchoalveolar lavage inasthmatic patients. Summary and recommendations.Chest 1985;88: 136-8.

28 Connolly MJ, Avery AJ, Walters EH, Hendrick DJ. Therelationship between bronchial responsiveness tomethacholine and bronchial responsiveness to his-tamine in asthmatic subjects. Pulmonary Pharmacology1988;1:53-8.

29 Wallaert B, Aerts C, Bart F, et al. Alveolar macrophagedysfunction in systemic lupus erythematosus. Am RevRespir Dis 1987;136:293-7.

30 Ward C, Kelly CA, Stenton CS, Duddridge M, HendrickDJ, Walters EH. Macrophage and neutrophilchemiluminescence in bronchoalveolar lavage fluid[abstract]. Clin Sci 1987;73(suppl 17):33-4.

31 Podleski WK, Grimes JR. Circulating hyperreactivelymphocytes in bronchial asthma. Clin ImmunolImmunopathol 1976;9:236-9.

32 Holgate ST, Hardy C, Howarth PH, Robinson C, ChurchMK, Agius RM. Bronchial mucosal mast cells andtheir implications in the pathogenesis of asthma. ClinRespir Physiol 1986;suppl 7:39-47.

33 Alan R, Kuna P, Rozniecki J, Kusminska B. Themagnitude of the spontaneous production ofhistamine-releasing factor by lymphocytes in vitrocorrelates with the state of bronchial hyperreactivity inpatients with asthma. J Allergy Clin Immunol1987;79: 103-8.

34 Gonzalez MC, Diaz P, Galleguillos FR, Ancic P,Cromwell 0, Kay AB. Allergen-induced recruitment ofbronchoalveolar helper (OKT4) and suppressor(OKT8) T-cells in asthma. Am Rev Respir Dis1987;136:6004.

35 Kelly CA, Ward C, Bird G, Hendrick DJ, Walters EH.The effect of filtration on cell counts inbronchoalveolar lavage fluid [abstract]. Thorax1986;41:727-

36 Murray HW, Cohn ZA. Macrophage oxygen-dependentantimicrobial activity. J Exp Med 1980;152:1596-609.

37 Joseph M, Tonnel A-B, Capron A, Voisin C. Enzyme riseand superoxide anion production by human alveolar

691

on May 13, 2020 by guest. P

rotected by copyright.http://thorax.bm

j.com/

Thorax: first published as 10.1136/thx.43.9.684 on 1 S

eptember 1988. D

ownloaded from

692macrophages stimulated with immunoglobulin E. ClinExp Immunol 1980;40:416-22.

38 Oyanagui Y. Inflammation and superoxide productionby macrophages. Agents Actions 1980;7(suppl):174-9.

39 Allen RC. Biochemiexcitation: Chemiluminescence andthe study of biological oxygenation reactions. In:Adam W, Cilento G, eds. Chemical and biologicalgeneration of excited states. New York: AcademicPress, 309-44.

40 Wallaert B, Bonniere B, Prin L, Cortot A, Tonnel AB,Voisin C. Primary biliary cirrhosis: subclinicalinflammatory alveolitis in patients with normal chestroentgenograms. Chest 1986;90:842-8.

41 Tonnel A-B, Joseph M, Gosset P, Fournier E, Capron A.Stimulation of alveolar macrophages in asthmaticpatients after local provocation test. Lancet1983;ii: 1406-8.

42 Rosen H, Klebanoff H. Chemiluminescence and super-oxide production by myeloperoxidase deficient leu-kocytes. J Clin Invest 1976;58:56-60.

43 Cluzel M, Damon M, Chanez P, et al. Enhanced alveolarcell luminol dependent chemiluminescence in asthma.JAllergy Clin Immunol 1987;80:195-201.

44 Gin W, Shaw RJ, Kay BA. Airways reversibility afterprednisolone therapy in chronic asthma is associatedwith alterations in leucocyte function. Am Rev RespirDis 1985;132:1 199-203.

45 Reynolds HH, Newball HH. Analysis of protein andrespiratory cells obtained from human lungs by bron-chial lavage. J Lab Clin Med 1974;84:559-73.

46 Martin TR, Raghu G, Maunder RJ, Springmeyer MC.

Kelly, Ward, Stenton, Bird, Hendrick, WaltersThe effects of chronic bronchitis and chronic airflowobstruction on lung cell populations recovered bybronchoalveolar lavage. Am Rev Respir Dis 1985;132:254-60.

47 Martin RR. Altered morphology and increased acidhydrolase content of pulmonary macrophages fromcigarette smokers. Am Rev Respir Dis 1973;107:596-601.

48 Harris JO, Olsen GN, Castle JR, Maloney AS.Comparison of proteolytic enzyme activity inpulmonary alveolar macrophages and bloodleucocytes in smokers and non-smokers. Am RevRespir Dis 1975;111:579-86.

49 Holt PG. Immune and inflammatory function in cigarettesmokers. Thorax 1987;42:241-9.

50 Harris JO, Swenson E, Johnson JE. Human alveolarmacrophages: comparison of phagocytic activity,glucose utilization and ultrastructure in smokers andnon-smokers. J Clin Invest 1970;49:2086-96.

51 Hoidal JR, Fox RB, Lemarbe PA, Perri R, Repine JE.Altered oxidative metabolic responses in vitro ofalveolar macrophages from asymptomatic cigarettesmokers. Am Rev Respir Dis 198 l;123:85-9.

52 McLeod R, Mack DA, McLeod EG, Campbell EJ, EstesRG. Alveolar macrophage function and inflammatorystimuli in smokers with and without obstructive air-ways disease. Am Rev Respir Dis 1985;131:377-84.

53 Gosset P, Tonnel A-B, Joseph M, et al. Secretion of achemotactic factor for neutrophils and eosinophils byalveolar macrophages from asthmatic patients.J Allergy Clin Immunol 1984;74:827-34.

on May 13, 2020 by guest. P

rotected by copyright.http://thorax.bm

j.com/

Thorax: first published as 10.1136/thx.43.9.684 on 1 S

eptember 1988. D

ownloaded from