Airways inflammation in subjects chronic bronchitis ... · brief, the substrate solution containedSAPNA 50mg/dl (final concentration) in Tris buffer 0-2 M + CaCl2 50mM, pH 8-0, +

Thorax 1994;49:1211-1216

Airways inflammation in subjects with chronicbronchitis who have never smoked

Mirco Lusuardi, Armando Capelli, Carlo G Cerutti, Elisa L Spada, Claudio F Donner

Division of PulmonaryDisease, Clinica delLavoro Foundation,IRCCS, MedicalCenter ofRehabilitation, I-28010Veruno (NO), ItalyM LusuardiA CapelliC G CeruttiE L SpadaC F Donner

Reprint requests to:Dr M Lusuardi.Received 4 February 1994Retumed to authors11 May 1994Revised version received11 July 1994Accepted for publication12 September 1994

AbstractBackground - Smoking is the single mostcommon cause of chronic bronchitis butthe disease can also occur in non-smokers.Alterations in the lung responsible for thedisease, such as oxidant/antioxidant andprotease/antiprotease imbalance, havebeen investigated in smokers. The aim ofour study was to evaluate local cellularand soluble factors (albumin, immuno-globulins, proteases, a,-antitrypsin, andtransferrin) that may be involved in thedevelopment of chronic bronchitis in sub-jects who have never smoked.Methods - Sixteen clinically stable patientswith chronic bronchitis who had neverbeen smokers were studied and 17 healthynon-smokers served as controls. All sub-jects underwent bronchoalveolar lavage(BAL). Total and differential cell countsand concentrations of the main proteins(albumin, immunoglobulins, complementfractions, a,-antitrypsin, and transferrin)were measured. Elastase-like activity wasassessed in cells and supernatants. To es-timate the oxidant burden the release ofsuperoxide anion (02-) from native cellpopulations was evaluated.Results - Recovery of BAL fluid was re-duced in older individuals in both thechronic bronchitis and control groups.There was no difference in total cell count,but neutrophil percentage count washigher in those with chronic bronchitis(median (range) 3.5 (16.14.2)) than incontrols (1.3 (0O5-3.7)). These differenceswere most pronounced in the first re-covery, representative ofthe bronchial lav-age. There was no difference in bronchialepithelial cells. Total proteins and albuminlevels were comparable and IgG, IgA, IgM,C3, C4, transferrin and a,-antitrypsin val-ues standardised to albumin did not showany significant differences. No differencesin elastase-like levels in supernatants weredetected. In cell lysates elastase-likeactivity x 107 cells (macrophages + neu-trophils) was increased in patients withchronic bronchitis (0.25 (0.06-4.3) com-pared with controls 0-08 (0.03-0.9) pgPPEeq). The release of 02- both at base-line and after opsonised zymosan pha-gocytosis did not show any differences.Correlation analysis between FEV, andBAL fluid data showed a negative cor-relation only with neutrophils/ml.Conclusions - Clinically stable non-smokers with chronic bronchitis show noalterations of local immune components,

oxidant burden, and free elastase-like ac-tivity in BAL fluids, while the content ofelastase-like activity in phagocytic cells isincreased. As in smokers, bronchial neu-trophilia is the most significant cellularmodification which correlates with the de-gree of airflow obstruction.

(Thorax 1994;49:1211-1216)

Cigarette smoking is considered the majorcause of chronic bronchitis in western coun-tries. Nevertheless, recent studies suggest thatin non-smokers chronic bronchitis may be as-sociated with other environmental risk factors.'

Local factors in the lung which may play apart in causing the disease have been in-vestigated in smokers. These include oxidant/antioxidant, protease/antiprotease imbalance,and alteration ofimmune defences,2-5 althougha precise index of individual susceptibility tothe deleterious effects of smoking has yet to beidentified.6 Only a few patients with chronicbronchitis claim never to have smoked in theirlife. In such cases risk factors other than tobaccosmoking have been suggested, including age,an inherited deficiency of antiprotease, defectsof immunity with recurrent respiratory in-fections, environmental factors such as chronicinhalation of irritants (fumes, dusts) in theworkplace, and general pollution.7

Bronchial and bronchoalveolar lavage (BAL)have been used to characterise the intraluminalinflammatory components in chronic bron-chitis, but these have rarely been compared withhistological findings in the bronchial mucosa.59No series has reported only on patients withchronic bronchitis who have never smoked.Smoking induces major modifications of BALfluid data irrespective of the presence of overtdisease, and thus potentially represents a con-founding factor.41'The aim of our study was to evaluate local

cellular and soluble factors potentially involvedin the development of chronic bronchitis insubjects who had never been smokers. In par-ticular, we sought to determine whether a localdefect of immune components or an increasein oxidant and protease burden may have someimportance irrespective of smoking habits.

MethodsSTUDY POPULATIONThe BAL data from 16 clinically stable, wellnourished patients affected by chronic bron-chitis were studied. Table 1 gives the demo-graphic data. None had any history ofsmoking.

1211

on April 30, 2020 by guest. P

rotected by copyright.http://thorax.bm

j.com/

Thorax: first published as 10.1136/thx.49.12.1211 on 1 D

ecember 1994. D

ownloaded from

Lusuardi, Capelli, Cerutti, Spada, Donner

Table 1 Median (range) demographic and functionaldata

Controls Chronic bronchitis(n= 17) (n=16) p

Sex (M:F) 12:5 4:12Age (years) 45 (28-69) 65 (49-78) 0-001FEV, (% pred) 90 (81-119) 64 (38-84) 0-0001FVC (% pred) 91 (76-120) 94 (69-125) NSFEF2 -75 (% pred) 87 (71-117) 54 (26-93) 0-001

Chronic bronchitis was diagnosed according tothe MRC standard criteria - that is, the pres-

ence of cough and sputum production on mostdays of the month for at least three months a

year in the last two years." Specific risk factorsfor chronic bronchitis, apart from age (fivepatients >65 years) were identified in only sixpatients who had been exposed to inorganic(two patients) or organic (four patients) dustsin the workplace, but not within three years ofthe start of the study. No clinical or functionalevidence of interstitial lung disease was found.A history of recurrent respiratory infectionsduring childhood and/or adolescence was re-

ported by four patients. Significant passivesmoking was denied by all the patients. Noneof the patients with chronic bronchitis hadoal-antitrypsin deficiency or systemic immunedefects.Pulmonary function tests included the meas-

urement of forced expiratory volume in one

second (FEV,), forced vital capacity (FVC),and forced expiratory flow between 25% and75% ofvital capacity (FEF25 15). In the patientswith chronic bronchitis FEV, measurement wasrepeated 15 minutes after inhalation of 200 ig

salbutamol. The degree of emphysema was notassessed by objective means such as computedtomographic (CT) scanning or measurementof the lung transfer factor (TLCO).

Patients with respiratory disorders other thanchronic bronchitis, including asthma, in-

fectious diseases, atopy, immunodeficiencyconditions, autoimmune disorders, malig-nancies or clinically significant cardio-vascular, neurological, endocrine and haemato-logical disorders were excluded. Fine cutCT scanning was not available, so a com-

ponent ofbronchiectasis could not be excluded.Only treatment with bronchodilators was per-mitted. All the patients with chronic bronchitisused inhaled bronchodilators either regularlyor on an as needed basis; 11 were on treatmentwith oral sustained-release theophylline.Patients were not studied within one month ofan exacerbation.

Seventeen healthy volunteers who had never

smoked served as controls, and their demo-graphic data are summarised in table 1. Sincethe controls and the patients with chronic bron-chitis were not age matched, the controls were

divided into two subgroups of five elderly ofmedian (range) age 63 (55-69) years and eightyounger (36 (28-42) years) individuals to studythe possible effect of age on BAL data.The study was carried out under the super-

vision of the institutional review board atthe Medical Center ofRehabilitation in Veruno,Italy, and the subjects gave written informedconsent.

BRONCHOALVEOLAR LAVAGE AND PROCESSINGOF SPECIMENSTo evaluate local airway inflammatory com-ponents BAL was performed using a previouslydescribed technique.'2 Three 50 ml boluses ofsterile saline warmed to 37°C were instilledthrough the working channel of a fibreopticbronchoscope. Each bolus was immediatelyremoved by gentle suction with a syringe andkept separately. Fluids were filtered through asingle layer of sterile gauze.

Before centrifugation cytocentrifuge (Cyto-spin II, Shandon, London, UK) slide pre-parations of native fluids were made in du-plicate for each recovery and stained with May-Grumwald-Giemsa. A minimum of 500 cellsper slide were examined at a magnification ofx 1000. Cell differential analysis was carriedout on the individual and pooled recoveries.Bronchoalveolar cells were separated by cent-rifugation (Beckman TJ 6, Beckman, Fullerton,California, USA) at 400g for 15 minutes at4°C. The samples were then resuspended incell culture medium RPMI 1640 (BoehringerMannheim GmbH, Mannheim, Germany).

BIOCHEMICAL ANALYSISProteins and protein fractionsSupernatants were immediately frozen at- 80°C and assays were performed within 15days. The levels of albumin, IgG, IgA, IgM,C3, C4, ocl-antitrypsin, and transferrin in therecovered BAL fluid samples were assessedusing a modified immunoturbidimetricmethod'3 (URIN-PAK immuno MICROALBfor albumin; SERA-PAK immuno for the othersubstances; Miles Italiana, Cavenago Brianza,Milan, Italy). All BAL fluid samples were un-concentrated. The immunoturbidimetric re-action required a 30 minute incubation forall the assays; the absorbance readings wereperformed at 340 nm using a COMPUR M2000 CS2 spectrophotometer (Bayer Diag-nostic-Electronic, Munchen, Germany). Allsamples were tested in duplicate.

Elastase-like activityThe elastase burden was evaluated on super-natants (free elastase-like activity) and lys-ates of the entire cell population (potentiallyreleasable elastase-like activity) using a solublesynthetic substrate, Succinyl-Ala-Ala-Ala-p-Nitro-Anilide (SAPNA), according to a mo-dification of the method of Bieth et al. Inbrief, the substrate solution contained SAPNA50 mg/dl (final concentration) in Tris buffer0-2 M + CaCl2 50 mM, pH 8-0, + di-methylsulphoxide 10% (v/v) + Triton X-1000-003% (v/v) (all reagents from Sigma Chem-icals, St Louis, Missouri, USA). The workingsolution, composed of 1-9 ml of the substratesolution and 100 gl of the sample, was in-cubated for 24 hours at 37°C. The absorbancereadings were performed at 405 nm using aCOMPUR M 2000 CS2 spectrophotometer(Bayer Diagnostic-Electronic, Munchen, Ger-many). All samples were tested in duplicate.

Calibration curves were produced with pan-

1212

on April 30, 2020 by guest. P

rotected by copyright.http://thorax.bm

j.com/

Thorax: first published as 10.1136/thx.49.12.1211 on 1 D

ecember 1994. D

ownloaded from

Airways inflammation in subjects with chronic bronchitis who have never smoked

Table 2 Median (range) cell counts and differentials in the three sequential and pooled BAL fluid recovery volumes

Ist recovery 2nd recovery 3rd recovery Pooled recoveries

Controls CB p Controls CB p Controls CB p Controls CB p

Cells/ml x 105 1-3 (0 5-4 1) 1 3 (0 3-48) NS 1-9 (1-14-6) 1-7 (1 0-3 6) NS 2-2 (1 0-5 5) 1-8 (0 7-4 8) NS 2-0 (1 1-4 0) 1 6 (1 1-3 7) NSMacrophages (%) 87 (78-97) 79 (46-90) 0-0002 90 (82-97) 88 (77-97) NS 90 (80-98) 90 (72-96) NS 89 (81-97) 87 (76-94) NSLymphocytes (%) 8-3 (1-2-17-4) 7-3 (24-21-4) NS 8-8 (1 3-17 0) 6-6 (1-8-21-3) NS 8-9 (10-18-4) 6-0 (1-5-26-6) NS 9-2 (1-6-17-6) 6-9 (2-3-18 4) NSNeutrophils (%) 2-1 (0-9-5-7) 10-3 (20-21-8) 0 0001 0-8 (022-23) 2-0 (0-5-11-9) 0 01 0-8 (022-24) 1-7 (0 5-8 4) 0-007 1-3 (0 5-3 7) 3-5 (1-6-14-2) 0 0001Eosinophils (%) 02 (0-2 9) 0-3 (0-12 5) NS 0-3 (01-4) 0 3 (0-45) NS 0-2 (0-07) 0 3 (0-22) NS 0-3 (01-4) 0-4 (0-5 1) NSBasophils (%) 0 (0-06) 0 (0-0-2) NS 0 (0-04) 0 (0-0) 0-005 0 (0-04) 0 (0-0 8) NS 0-1 (0-04) 0 (0-0 4) 0-02

CB = chronic bronchitis.

creatic porcine elastase (PPE). Data are re-ported in zig PPEeq (eq= equivalent) asstandardised to albumin levels (supematants),or per 107 cells considering only macrophagesand granulocytes together, the only cells con-taining a significant quantity of elastase.

PhagocytosisTo obtain data that would most closely reflectfindings in vivo the entire BAL cell populationwas used for the measurement of phagocytosis.The BAL cells were rinsed three times in phos-phate buffered saline (PBS) and then sus-pended in RPMI 1640 + 10% fetal calf serum(FCS, Boehringer Mannheim GmbH, Mann-heim, Germany). Viability was assessed usingthe trypan blue dye exclusion test. Superoxideanion release was assessed both at baselineand after opsonised zymosan phagocytosis assuperoxide dismutase (SOD) inhibitable re-duction of ferricytochrome C according to amodified Bellavite's assay.'5 Four sterile testtubes labelled IA, iB, 2A, 2B were preparedfor each cell suspension with 400 ,ul of a KrebsRinger phosphate (KRP) buffer solution con-taining 0-25 mM/l cytochrome C and 5 mMIlglucose. Ten ,ul SOD solution (2-5 mg/ml; 3000U/mg protein) were added to tubes 1B and 2B.Tubes 2A and 2B contained zymosan, 1 g/l,opsonised with group AB human serum (10%)in a thermostat bath to 37°C for 15 minutesto evaluate the phagocytic activity. Tubes 1Aand 1B, containing no zymosan, were used toevaluate the basal production of superoxideanion. Following the addition of 100,l of al-veolar cell suspension the tubes were incubatedin a water bath for 10 minutes at 37°C withcontinuous shaking (about 100 rpm). The re-action was stopped with 2 ml of ice cold KRPbuffer. The tubes were centrifuged at 15OOgfor 10 minutes. The absorbance of cell-freesupematants was measured at 550 nm and datawere multiplied by the dilution factor and di-vided by the extinction coefficient (ismol/l) ofcytochrome C determined at 550 nm (0-0189).The difference between the values of 02-

production with and without zymosan is re-ported as a phagocytic index. 02- productionwas expressed in nM 02-/ 106/ 10 minutes.

STATISTICAL ANALYSISAll data are reported as median (range).Differences in BAL fluid data were evaluatedaccording to the Mann-Whitney U test as thedistribution of data did not appear to be nor-mal. Probability values greater than 95% wereconsidered statistically significant. The Spear-

man's rank correlation coefficient was used tostudy correlations between FEV, and BAL fluiddata. Analyses were performed using amicrocomputer (Macintosh LC) and Stat-view + Graphics software (Abacus Concept).

ResultsThe demographic and functional data of pa-tient and control subjects are shown in table1. Pulmonary function tests in the patients withchronic bronchitis showed mild to moderateairways obstruction. Reversibility after inhaledsalbutamol was always < 15%. The groups werenot matched for age. The comparison of thetwo control subgroups showed a significantdifference in recovery of BAL fluid (oldest 69(63-98) ml v youngest subjects 93 (74-107)ml, p = 0-02) which was not related to differ-ences in respiratory function (in both groupsthe mean FEV, was 91% predicted). Con-sidering the whole case series, recovery ofBALfluid was reduced in patients with chronic bron-chitis (70 (42-108) ml) compared with thecontrol group (86 (63-114) ml, p=002).However, when the five elderly control subjectsand the patients with chronic bronchitis werecompared, there were no differences in BALfluid recovery. The highly significant differencein FEV, between controls and patients withchronic bronchitis (table 1) suggests that agewas the main factor responsible for the re-duction in recoveries.

Cellular components were not influenced byage. Cell counts were superimposable (table2). Viability was similar in the two groups:controls 98 (91-99)%, chronic bronchitis 95(92-98)%. The differential cell counts revealedsignificant differences in the percentage of neu-trophils (table 2) and absolute values (chronicbronchitis 5-2 (2*6-31*9)/ml x 103, controls 2-1(1 1-8 5)/ml x 103, p<0-001). Although thesedifferences were accounted for mainly by thefirst recovery, representative of the bronchialenvironment, a significantly higher percentageof neutrophils was also present in the secondand third recoveries from the patients withchronic bronchitis (table 2). The number ofmacrophages was significantly reduced in thefirst recovery sample (table 2). Basophil cellswere significantly reduced in the subjects withchronic bronchitis (table 2). Surprisingly, thepercentage of bronchial epithelial cells in thefirst recovery was similar in the patients withchronic bronchitis (1 8 (0 5-9 5)) and the con-trol subjects (2-3 (0 5-3 5)).

Absolute, but not standardised values ofpro-tein components tended to be influenced byage but not significantly. Total proteins and

1213

on April 30, 2020 by guest. P

rotected by copyright.http://thorax.bm

j.com/

Thorax: first published as 10.1136/thx.49.12.1211 on 1 D

ecember 1994. D

ownloaded from

Lusuardi, Capelli, Cerutti, Spada, Donner

Table 3 Median (range) biochemical data from BAL fluid supernatants

Controls Chronic bronchitis p

Total proteins (mg/dl) 8 5 (5 0-11-8) 9 2 (5 0-17 8) NSAlbumin (mg/dl) 3 0 (0 9-7 5) 3 2 (1 3-9-0) NSIgG/albumin 0.2 (0 01-0 5) 0 26 (0 05-1 26) NSIgA/albumin 0 09 (0 004-0 7) 0 17 (0 018 0-6) NSIgM/albumin 0 08 (0 01-0 14) 0 04 (0 02-0 16) NSC3/albumin 0.043 (0 001-0 22) 0 04 (0-0 13) NSC4/albumin 0 012 (0 007-0 1) 0-023 (0 004-0 06) NS,Y-antitrypsin/albumin 0-033 (0-002-0 52) 0(033 (0 008-0 09) NSTransferrin/albumin 0 125 (0 015-0 25) 0 071 (0 004-0 18) NS

(4.3)

E Controls

ED Chronic bronchitis

0

0

0

Supernatant ELA Intracellular ELA(pg PPEeq/mg albumin) (pg PPEeq/107 cells)

Figure I Box whisker plots of the elastase-like activity (ELA) in bronchoalveolar lavagefluids in non-smoker patients with chronic bronchitis compared with normal controls.Since the data are not normally distributed the plots show the median, the interquartilerange, and the 10th and 90th percentiles. Individual data points which lie outside theI 0th and 90th percentiles are shown individually. See text for levels of significance.

160

E Controls

E Chronic bronchitis140

120

100*

80*

60

0

T I

IVe e e e

% %7 7

tt eRLe {

Baseline Phagocytosis

and numbers of neutrophils and macrophages.There were no significant differences in therelease of 02 at baseline or after opsonisedzymosan phagocytosis (fig 2), nor did the com-parison of the phagocytic index reveal any sig-nificant differences.The correlation between FEV, and BAL fluid

data was significant only for neutrophils/ml (fig3). The same trend was shown for albumin butdid not reach statistical significance.

DiscussionThis study was conducted to evaluate and com-pare the characteristics ofBAL fluids in a groupof subjects with chronic bronchitis who hadnever smoked and a group of non-smokinghealthy controls.The two groups were not matched for age

and the mean age of the bronchitic group wassignificantly higher than that of the controls.In spite of this disparity, the only result clearlyinfluenced by age was the amount of fluidrecovered. Interestingly, bronchial obstructiondid not have a major effect on reduced recoveryvolumes, at least in patients with mild to mod-erate airflow limitation. A previous study alsofound that bronchial obstruction was notassociated with significantly lower recoveryvolumes. 1 6The finding of similar cell numbers in the

subjects with chronic bronchitis and the con-trols strongly suggests that the disease itself isnot responsible for an increase in cells. Thelavage fluid obtained from the subjects withchronic bronchitis differed in cellular pro-portions by an average 3-4 fold increase inneutrophils. The magnitude of increase wasalways greatest at the first recovery and lowestat the third, and the increase in percentage ofneutrophils in the first recovery accounted forthe reduced percentage of macrophages. Sim-ilar alterations in cell differentials have beendescribed in smokers.'6 Cigarette smoking isassociated with a steady increase of neutrophilsin the airways'7 but, according to our results,chronic bronchitis is characterised by intra-luminal neutrophilia irrespective of smoking.Smoking is one, but by no means the only,factor to cause airways inflammation associatedwith heightened recruitment of neutrophils. As

Figure 2 Box whisker plots of the O release fronm the native bronchoalveolar lavage cellpopulation at baseline and after phagocytosis with opsonised zymosan in non-smokerpatients with chronic bronchitis and normal controls. Since the data are not normallydistributed the plots show the median, the interquartile range, and the 10th and 90thpercentiles. Individual data points which lie outside the 10th and 90th percentiles areshown individually.

albumin levels were superimposable; there wereno significant differences in IgG, IgA, IgM,C3, C4, transferrin and oct-antitrypsin valuesstandardised to albumin (table 3).

Elastase-like levels in supematants were

comparable (fig 1). In cell lysates elastaseactivity x 1 O cells (macrophages + neu-

trophils) was significantly increased in patientswith chronic bronchitis (p<0 05) (fig 1). Nocorrelation was found between elastase levels

35

30

25x

E 20

aC 15

= 10a)

r=-0.54p = 0.036

.

0

5 0@ 0

* *

0 1,1,1,1,1,1,35 40 45 50 55 60 65 70 75 80 85

FEV1 (%)

Figure 3 Correlation between percentage predicted FEV,and neutrophil count in bronchoalveolar lavage fluid(absolute values x 1 3).

c

E

0C)

ZW

1214

I0 ""Ill,

kh -J.-

.e-- .

on April 30, 2020 by guest. P

rotected by copyright.http://thorax.bm

j.com/

Thorax: first published as 10.1136/thx.49.12.1211 on 1 D

ecember 1994. D

ownloaded from

Airways inflammation in subjects with chronic bronchitis who have never smoked

found in smokers with chronic bronchitis, therewas a significant correlation between the num-ber of neutrophils and the degree of bronchialobstruction. In clinically stable bronchiticpatients, the number of neutrophils in the BALfluid may serve as an index of the severity ofbronchial obstruction. Such information wouldbe useful in clinical studies - for example, forevaluating the local efficacy of anti-in-flammatory treatment."The part played by neutrophils in the patho-

genesis of chronic bronchitis remains elu-sive. Basic alterations in peripheral neutrophilfunctions have been described in chronic ob-structive lung disease - for example, an en-hanced response to chemotactic factors andextracellular proteolysis.'9 Recent data frompatients with chronic bronchitis showed thatincreased expression of adhesion molecules inthe bronchial mucosal vessels was associatedwith tissue neutrophils,20 but we do not knowwhether they are the cause or the effect of theanatomical and functional damage. Thompsonet al found that epithelial cells from the sub-mucosal glands in the large airways of bovinescan release increased quantities of neutrophilchemotactic factor when stimulated by endo-toxin.2' Functional studies in smokers haveshown that alveolar macrophages release morechemotactic factors responsible for neutrophilrecruitment.'0 It would also be interesting toevaluate the release of chemoattractant agentsfrom the mucosal mononuclear cells whichpredominate over the other types of in-flammatory cells in the bronchial walls,22 23 witha view to elucidating the features ofactivation.23Inactivation of chemotactic factor inhibitor hasalso been described, but in association withtobacco smoke.24What then is the status of alveolar macro-

phages in patients with chronic bronchitiswho have never smoked? Macrophages werenot increased, and were morphologically ident-ical to those in the controls when studied byoptical microscopy. Functional studies of theentire cell population (02- release and pha-gocytosis) were not consistent with a state ofactivation; however, chemotactic factors werenot directly assessed.The increase in proteases in cells could, the-

oretically, cause damage to the bronchial mu-cosa. However, the quantity of elastase-likeactivity released into fluids was similar in thepatients with chronic bronchitis and the healthycontrols. In the light of the characteristics ofextracellular proteolytic processes, proteasesstored in resting phagocytic cells may be moreimportant than the aliquot released into thealveolar lining fluid which can easily be neut-ralised by cxl-antitrypsin. 19 25

Superoxide anion, another important causeof damage to the mucosal cells, was increasedboth at baseline and after phagocytic sti-mulation. This increase was not significant des-pite the augmentation of viable neutrophils.Mucus in the airways can inhibit macrophagephagocytosis.26 Bronchodilator therapy shouldalso be taken into account, as methylxanthinesand P2 agonists have anti-inflammatory effectsnot only in vitro but also in vivo.'527 The

comparison ofBAL fluid data from the patientswho were taking theophylline and those whowere not revealed no differences. On the otherhand, other factors may have triggered ma-crophage and neutrophil activities. Secretionsin patients with chronic bronchitis are fre-quently contaminated by bacteria28 whoseproducts can stimulate phagocytic cells.The shedding of epithelial cells can be held

as a marker ofdamage. In keeping with others'6we observed no increase in ciliated epithelialcells in BAL fluid samples. The differencesbetween the percentage of ciliated cells iden-tified in this study and previously reportedpercentages2 may depend on the techniquesused to recover the fluids. Importantly, wefound a 4-5 fold increase in the number ofepithelial cells when BAL fluids were recoveredby mechanical aspiration rather than manualgentle aspiration using a syringe (unpublishedobservations).

Because specific staining was not used it ispossible that the reduced number of basophilsresulted from increased degranulation whichrendered accurate quantification difficult.The analysis of soluble factors in super-

natants revealed no difference in total pro-tein and albumin levels, denoting the absenceof significant exudation. The levels of a,-anti-trypsin and transferrin, important componentsof the protease/antiprotease and oxidant/anti-oxidant29 balances respectively, were also notsignificantly different. Functional activity wasnot, however, investigated. On the other hand,given that the burden of endogenous and exo-genous oxidants was not increased in ourpatients with chronic bronchitis, we considerit unlikely that their levels of oc-antitrypsininactivation would differ from those of thecontrols.The evaluation of soluble immune com-

ponents revealed insignificantly increased pro-portions of IgA. Although the interpretation ofbiochemical data in BAL fluid remains con-troversial,30 chronic bronchitis does not seemto be associated with a quantitative defect ofIgA secretion. Furthermore, the cellular im-mune defence mechanisms, and phagocytosisin particular, as demonstrated by the super-oxide anion release after challenge with opson-ised zymosan, were not depressed in ourpatients with chronic bronchitis. Even insmokers the data on lung immune defencesand their clinical relevance are controversial.'"Our data on immune defences in patients withchronic bronchitis who had never smoked agreewith previous observations that, with the ex-ception of a limited number of subjects whoare affected by inherited disorders of definiteimmune components, chronic bronchitis is notassociated with a defect but rather with anenhancement of local immune defences inresponse to a failure of other non-immunedefence mechanisms such as cough andmucociliary clearance.3' It is unlikely that anabsolute defect in local immune defences playsa crucial part in the pathogenesis of chronicbronchitis or in recurrent exacerbations. Func-tional defects are more plausible32 but need tobe substantiated by further studies.

1215

on April 30, 2020 by guest. P

rotected by copyright.http://thorax.bm

j.com/

Thorax: first published as 10.1136/thx.49.12.1211 on 1 D

ecember 1994. D

ownloaded from

Lusuardi, Capelli, Cerutti, Spada, Donner

In summary, although only a few patientswith chronic bronchitis are non-smokers, thecharacterisation of intraluminal cells and sol-uble factors compared with normal controlswithout the confounding effect of smoking canlead us to conclude that local immune defences(Ig, phagocytosis) are not significantly differ-ent, the global oxidant burden is not increased,and the elastase-like activity content of alveolarphagocytic cells is increased, but the com-ponent released in the alveolar milieu is withinthe normal range. The increase in proteaseburden may be an important cause of mucosaldamage even in clinically stable patients.Finally, bronchial neutrophilia was the mostsignificant characteristic in our patientsand correlated with airways obstruction. Thus,at the bronchial level luminal signs ofinflammation in patients with chronic bron-chitis correlate with functional impairment,irrespective of smoking habits.

The authors are indebted to Gillian Jarvis for her editorialassistance. This study was supported in part by the grant"Ricerca Corrente" from the Ministry of Health, Rome, Italy.

1 Neuberger M, Kundi M, Friedl HP. Environmental factorsin chronic obstructive pulmonary disease. Eur Respir Rev1992;2: 144-8.

2 Thompson AB, Daughton D, Robbins RA, Ghafouri MA,Oehlerking M, Rennard SI. Intraluminal airway in-flammation in chronic bronchitis. Am Rev RespirDis 1989;140:1527-37.

3 Fujita J, Nelson NL, Daughton DM, Dobry CA, SpurzemJR, Irino S, et al. Evaluation of elastase and antielastasebalance in patients with chronic bronchitis and pulmonaryemphysema. Am Rev Respir Dis 1990;142:57-62.

4 Luisetti M, Meloni F, Ballabio P, Leo G. Role of bronchialand bronchoalveolar lavage in chronic obstructive lungdisease. Monaldi Arch Chest Dis 1993;48:54-7.

5 Stockley RA, Burnett D. Bronchoalveolar lavage and thestudy of proteinases and antiproteinases in the pa-thogenesis of chronic obstructive lung disease. MonaldiArch Chest Dis 1993;48:245-53.

6 Harris CC. Tobacco smoke and lung disease: who is sus-ceptible? Ann Intern Med 1986;105:607-9.

7 Davies RJ, Sheinman BD. Smoking and atmospheric pol-lution. _Allergy Clin Immunol 1986;78(5, part 2):1031-5.

8 Rossi GA, Pesci A, Zanon P, Olivieri D. Inflammatorychanges in the airway lumen and in the bronchial mucosaof patients with chronic bronchitis. Monaldi Arch ChestDis 1993;48:399-402.

9 Hobson JE, Wright JL, Wiggs BR, Hogg JC. Comparisonof the cell content of lung lavage fluid with the presenceof emphysema and peripheral airways inflammation inresected lungs. Respiration 1986;50:1-8.

10 Costabel U, Guzman J. Effect of smoking on broncho-alveolar lavage constituents. Eur Respir Jf 1992;5:776-9.

11 American Thoracic Society. Standards for the diagnosis andcare ofpatients with chronic obstructive pulmonary diseaseand asthma. Am Rev Respir Dis 1987;136:225-44.

12 Velluti G, Capelli 0, Lusuardi M, Braghiroli A, Pellegrino

M, Milanti G, et al. Bronchoalveolar lavage in the normallung. First of three parts: protein, enzymatic and ionicfeatures. Respiration 1983;44:403-10.

13 Capelli A, Cerutti CG, Lusuardi M, Colombo M, DonnerCF. Methods for evaluation of protein components inbronchoalveolar lavage fluids. Clin Chem 1993;39:2201-2.

14 BiethJ, Spiess B, Wermuth CG. The synthesis and analyticaluse of a highly sensitive and convenient substrate ofelastase. Biochem Med 1974;11:350-7.

15 Capelli A, Lusuardi M, Carli S, Zaccaria S, Trombetta N,Donner CF. In vitro effect of P2-agonists on bacterialkilling and superoxide anion (O2-) release from alveolarmacrophages of patients with chronic bronchitis. Chest1993;104:481-6.

16 Martin TR, Raghu CR, Maunder RJ, Springmeyer SC.The effects of chronic bronchitis and chronic air-flowobstruction on lung cell populations recovered by bron-choalveolar lavage. Am Rev Respir Dis 1985;132:254-60.

17 Hunninghake GW, Crystal RG. Cigarette smoke and lungdestruction: accumulation of neutrophils in the lungs ofcigarette smokers. Am Rev Respir Dis 1983;128:833-8.

18 Thompson AB, Mueller MB, Heires AJ, Bohling TL,Daughton D, Yancey SW, et al. Aerosolized beclo-methasone in chronic bronchitis. Improved pulmonaryfunction and diminished airway inflammation. Am RevRespirDis 1992;146:389-95.

19 Burnett D, Chamba A, Hill SL, Stockley RA. Neutrophilsfrom subjects with chronic obstructive lung disease showenhanced chemotaxis and extracellular proteolysis. Lancet1987;ii:1043-6.

20 Di Stefano A, Roggeri A, Maestrelli P, Turato G, FerraressoA, Drigo R, et al. Vascular expression of adhesion mo-lecules in bronchial biopsies of subjects with chronicbronchitis. Eur RespirJ7 1993;6(Suppl 17):309s.

21 Thompson AB, Heires A, Bohling T. Stimulated bovinetracheal submucosal gland cells secrete neutrophilchemotactic activity. Am Rev Respir Dis 1991;143(4, part2):A137.

22 Jeffery PK. Morphology of the airway wall in asthma andin chronic obstructive pulmonary disease. Am Rev RespirDis 1991;143:1152-8.

23 Saetta M, Di Stefano A, Maestrelli P, Ferraresso A, DrigoR, Potena A, et al. Activated T-lymphocytes and ma-crophages in bronchial mucosa of subjects with chronicbronchitis. Am Rev Respir Dis 1993;147:301-6.

24 Robbins RA, Gossman GL, Nelson KJ, Koyama S, Thomp-son AB, Rennard SI. Inactivation of chemotactic factorinactivator by cigarette smoke. Am Rev Respir Dis 1990;142:763-8.

25 Weiss SJ. Tissue destruction by neutrophils. N Engl J Med1989;320:365-76.

26 Wanner A. The role of mucus in chronic obstructive pul-monary disease. Chest 1990;97(Suppl 2):1 s-15s.

27 Lusuardi M, Capelli A, Donner CF. Interaction betweendrugs and lung immune defences in COPD: local evalu-ation with bronchoalveolar lavage. Monaldi Arch Chest Dis1993;48:411-4.

28 Spada EL, Tinivella A, Carli S, Zaccaria S, Lusuardi M,Sbaffi A, et al. Proposal of an easy method to improveroutine sputum bacteriology. Respiration 1989;56: 137-46.

29 Pacht ER, Davis WB. Role of transferrin and ceruloplasminin antioxidant activity of lung epithelial lining fluid. JApplPhysiol 1988;64:2092-9.

30 European Society of Pneumology Task Group on BAL(Klech H, Pohl W, eds). Technical recommendations andguidelines for bronchoalveolar lavage (BAL). Eur RespirJ'1989;2:561-85.

31 Bilton D, Leung AYT, Hill SL, Stockley RA. Defective lungdefences. Eur Respir Rev 1992;2:159-65.

32 Vecchiarelli A, Dottorini M, Puliti M, Todisco T, CenciE, Bistoni F. Defective candidacidal activity of alveolarmacrophages and peripheral blood monocytes frompatients with chronic obstructive pulmonary disease. AmRev Respir Dis 1991;143:1049-54.

1216

on April 30, 2020 by guest. P

rotected by copyright.http://thorax.bm

j.com/

Thorax: first published as 10.1136/thx.49.12.1211 on 1 D

ecember 1994. D

ownloaded from