Prominent neutrophilic inflammation in sputum from subjects with asthma exacerbation

Respiratory pathophysiologic responses Prominent neutrophilic inflammation in sputum from subjects with asthma exacerbation John V. Fahy, ",b Kwang W o o Kim, b Jane Liu, b and Homer A. Boushey", b San Francisco, Calif. To infer possible mechanisms of acute airway inflammation and mucus hypersecretion in acute severe asthma, we performed cellular and biochemical analysis on sputum from 18 adults with acute severe asthma and compared the results with results of analysis of sputum from 12 adults with cystic fibrosis (CF). We found that in subjects with asthma neutrophils made up more than 75% of sputum cells in 10 samples whereas eosinophils made up more than 75% of cells in only three samples. Fifty percent of the subjects with asthma reported that their asthma exacerbation was precipitated by a respiratory tract infection, and these subjects had a significantly higher percentage of neutrophils in their sputum (85% +- 6% vs 57% +- 12%, p = 0.05). In the CF samples neutrophils made up more than 95% and eosinophils less than 1% of cells in all samples analyzed. Analysis of fluid phase chemicals in asthmatic and CF sputum samples showed that despite overall lower mean values of neutrophil elastase (27 +- 1I ixg/ml vs 466 + 121 txg/ml, p = 0.0001) and interleukin-8 (IL8) (55 +- 15 ng/ml vs 186 + 24 ng/ml, p = 0.0001), some of the asthmatic samples had values for these variables thai overlapped those in the CF samples. In addition, the asthmatic samples were distinguished by the presence of higher tryptase (10 + 7 U/L vs 0.9 + 0.9 U/L, p = 0.0001) and interleukin-6 (1166 +- 447 ng/ml vs 186 + 24 ng/m# p = 0.0001) levels and by a higher ratio of albumin to mucin-like glycoprotein (0.8 +- 0.5 vs 0.1 + 0.002, p = 0.02). DNA levels were lower in the asthmatic samples (0.5 +- 0.3 mg/ml vs 3.5 +- 1.2 mg/ml, p = 0.05). We conclude that neutrophils predominate more frequently than eosinophils as the major inflammatory cell in sputum from patients with asthma in acute exacerbation. We speculate that this may be because respiratory tract infections are a frequent precipitant of acute asthma. In addition, the high IL-8 levels and free neutrophil elastase activity observed in asthmatic sputum suggests that IL-8 may mediate airway neutrophilia in acute asthma and that neutrophil elastase may mediate mucin glycoprotein hypersecretion in acute asthma, as has been proposed for the mucin hypersecretion in CF. (J ALLERGY CLIN IMMUNOL 1995;95: 843-52.)

A l t h o u g h sputum production is a c o m m o n and troublesome s y m p t o m in acute severe asthma, 1 and inspissation of mucus plugs in the airways is a characteristic feature of p o s t m o r t e m studies o f asthma deaths, 2-s little is k n o w n about the nature From the aDepartment of Medicine and the bCardiovascular Research Institute, University of California, San Francisco. Supported by Program Project grant HL 24136 from the National Institutes of Health and by NRSA grant HL07185 (J.V.F.). Received for publication May 5, 1994; revised Sept. ]5, 1994; accepted for publication Oct. 3, 1994. Reprint requests: John V. Fahy, MD, Box 0130, University of California, San Francisco, 505 Parnassus Ave., San Francisco, CA 94143. Copyright © 1995 by Mosby-Year Book, Inc. 0091-6749/95 $3.00 + 0 1/1/62153

Abbreviations used CF: Cystic fibrosis GM-CSF: Granulocyte-macrophage colonystimulating factor IL: Interleukin MLG: Mucin-like glycoprotein PBS: Phosphate-buffered saline UCSF: University of California, San Francisco

or m e c h a n i s m s o f d i s t u r b a n c e s in m u c u s p r o d u c t i o n and clearance. In an initial a t t e m p t to e x a m i n e these i m p o r t a n t disturbances, we have examined the sputum spontaneously produced 843

844

Fahy et al.

J ALLERGYCLIN IMMUNOL APRIL 1995

by 18 adults with asthma who were seen for treatment of an acute exacerbation at the emergency departments of either of two urban university hospitals. Surprisingly, although many studies have analyzed airway secretions obtained from patients with exacerbations provoked by antigen challenge, relatively few have analyzed sputum from persons with asthma with spontaneous acute exacerbations 611 and not all these have analyzed the fluid phase of the sputum. Although analysis of the fluid phase of sputum from subjects with acute asthma yields only descriptive information, it may be important for at least three reasons. First, the analysis of secretagogues (chemicals that stimulate secretion from airway goblet and submucosal gland cells) in the sputum fluid phase may provide information on possible mechanisms of mucus hypersecretion in acute asthma. Neutrophil elastase, 12 chymase, 13 and eosinophil cationic protein, ~4 and a variety of other inflammatory mediators, is are potent secretagogues, and mucushypersecretion may be an important consequence of their actions. For example, 75% of the secretagogue activity of sputum from subjects with bronchiectasis is attributable to neutrophil elastase activity. ~6 Second, mucin glycoproteins, albumin, and D N A have important effects on viscoelastic properties of solutions, ~7-19 and their quantification in sputum can help determine how likely they are to contribute to the physical properties of the mucus plugs that obstruct the airways in acute asthma. Third, analysis of cytokine levels in asthmatic sputum may help identify cytokines involved in the airway inflammation of acute asthma. Thus in this study we not only analyzed the cellular constituents of sputum from subjects with acute asthma but also measured the concentrations of sputum constituents thought to be important for mucus hypersecretion (neutrophil elastase activity) and for the viscoelastic properties of sputum (mucin-like glycoprotein [MLG], DNA, a n d a l b u m i n ) . We also measured the concentrations of granulocyte-macrophage colony-stimulating factor (GM-CSF), interleukin(IL)-4, IL-8, and IL-6, cytokines thought to be important in the regulation of asthmatic inflammation. 2°-26 For comparative purposes we analyzed sputum expectorated by 12 patients with cystic fibrosis (CF) (cellular analysis of six samples and biochemical analysis of 12 samples).

METHODS Subjects with asthma Patients who had acute severe asthma diagnosed by emergency room physicians at San Francisco General Hospital and at Motfitt-Long Hospital at the University of California, San Francisco (UCSF) were invited to provide a sample of sputum and to complete a medical questionnaire. The medical questionnaire inquired about asthma history, asthma medications, concomitant illnesses (including allergies), history of cigarette smoking, events that triggered the current asthma attack, and the length of time of asthma exacerbation before care was sought in the emergency room. Subjects who had smoked cigarettes or any other recreational inhalant in the past 10 years were excluded. Also excluded were subjects who had evidence of congestive heart failure or who had a history of lung disease other than asthma. Eighteen subjects with asthma (10 men; mean age, 43.5 years) satisfied inclusion and exclusion criteria and provided a sputum sample. All had wheezing audible on chest examination, and 11 had peak flow rates that were indicative of severe airflow obstruction but that reversed at least partially with nebulized albuterol treatment. The remaining subjects had pulmonary function data in their medical records showing reversible airflow obstruction. All the subjects had long-standing asthma, many had had frequent visits to the emergency room for asthma, nine had had hospitalizations for management of severe asthma, two had previously been intubated and had received mechanical ventilatation for management of severe asthma, and three were steroid dependent because of their asthma. Six of the 18 subjects were formerly cigarette smokers but had not smoked cigarettes for at least 10 years. These and other clinical characteristics are presented in Table I.

Subjects with CF Sputum samples were collected from patients with CF. Eight of these samples were collected during the first 48 hours of l~ospitalization for treatment of acute exacerbations of CF-related lung disease at Mottitt-Long Hospital at UCSF, and four had sputum collected during outpatient visits to the adult CF clinic at UCSF. The diagnosis of CF was made on the basis of a typical clinical history, airway obstruction, and markedly elevated levels of sodium and chloride in repeated sweat tests. Other clinical data are presented in Table II. All subjects signed consent forms approved by the committee on human research at UCSF.

Sputum processing Sputum from patients with asthma and from those with CF was processed within 4 hours of collection. Sputum volume was determined and an equal volume of phosphate-buffered saline solution (PBS) was added and briefly mixed gently by vortex mixer before addition of a volume of 0.1% dithiothreitol in saline solution (10% Sputalysin, Behring Diagnostics Inc., Somerville, N.J.)

J ALLERGYCLIN IMMUNOL VOLUME 95, NUMBER4

Fahy et at.

845

TABLE I. C l i n i c a l c h a r a c t e r i s t i c s o f p a t i e n t s w i t h a s t h m a PEF (L/min)

Exacerbant

Subject No.

Sex/ age (yr)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

F/32 F/41 M/20 M/16 M/33 M/66 M/56 M/38 F/78 M/36 M/35 M/64 F/33 M/48 F/54

No Yes No No No Yes Yes Yes No No No Yes No Yes No

Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes No Yes Yes Yes

ND Yes ND Yes Yes Yes No Yes No Yes Yes No Yes No No

16 17 18

F/29 F/53 F/52

No No No

No Yes Yes

No No No

Exsmoker Atopy* RTIt

Other

ND

Length of PreMedications Antibiotic¢ exacerbation BD

ND Exercise

[3,T,IS [3,T,IS [3,T,IS [3,C [3, 3,T,IS 3,T,P ~,T, IS 3,IS,PT 3,T ~,IP,IS,P ~,IS ~,IS ~,T,IS,PT 3,PT

ND No ND No No Yes No No No No Yes No No No Yes

ND Days Days Days Hours Days Weeks Weeks Days Minutes Hours Minutes Days Weeks Minutes

ND ND None

[3,T,IS,PT [3 [3,T,IS

No No No

Weeks Hours Hours

ND

ND Cold air

ND

PostBD

330 240

37011 450

250 230 134 100

34011 35011 309 145

320

FEV1 (L)§ Pre- PostBD BD

2.2

3.0

2.9 0.9

3.2 1.2

1.3

1.6

420

150 360 110 320 235 380 140 Not done 150 315 125 160 200 250

[3, [3-Agonistmetered dose inhaler; BD, bronchodilation; C, inhaled sodium cromoglycate; IS, inhaled steroid; ND, not documented; P, prednisone maintenance therapy; PEF, peak expiratory flow; PT, prednisone taper started within 6 days of asthma exacerbation; RTI, respiratory tract infection; T, oral theophylline preparation. *Defined as a history of hay fever and/or a history of cough, wheeze, chest tightness, rhinorrhea, sneezing, or itchy or watery eyes on exposure to animals or house dust. ?Defined as a history of a head cold or a chest cold as the precipitant of the current asthma attack. :~Refers to whether subjects were taking an antibiotic on presentation to the emergency room. §Emergency room PEF not available on these subjects FEV1 data are from hospital records (within 5 years of emergency room visit) and are not the FEV1 of the subjects during the exacerbation. For many subjects the pre-BD value represents a PEF recording made after an initial albuterol treatment. IIAdmitted for management of asthma attack.

equal to the combined volume of sputum and PBS (four times the total dilution of origimil sample). The sample was then incubated in a shaking water bath at 37 ° C for 15 minutes with removal for brief gentle mixing at 5-minute intervals. An aliquot of diluted homogenized sputum was used for cell count and cell differential, as previously describedY Cell differentials inclusive and exclusive of squamous cells were calculated. The remaining hom0genized samples were centrifuged at 1037g for 5 minutes, and the supernatant was aspirated and stored at - 7 0 ° C for later analysis. Sputum samples from six of the patients with CF were not analyzed for cell counts and differentials. For these samples the initial mixing with PBS was omitted. In the subjects with asthma the median volume of sputum processed was 0.8 ml (range, 0.5 to 10.0 ml), and in the patients with CF it was 2.6 ml (range, 0.9 to 24.4 ml). The asthmatic sputum was mucoid or mucopurulent in appearance; the sputum from CF patients was grossly purulent in appearance.

Biochemical assays

The activity of free neutrophil elastase was determined with a chromogenic substrate specific for human neutrophil elastase (methoxysuccinyl-L-alanyl-L-alanylL-prolyl-L-valyl-p-nitroanilide)(Sigma Chemical Co., St. Louis, Mo.) as previously described? 6 IL-8, IL-4, IL-6, and GM-CSF levels were measured by specific and sensitive human immunoassays (ELISA, Quantikine, R & D Systems, Minneapolis, Minn.). MEG concentrations were also measured by ELISA, 28 and tryptase, eosinophil cationic protein, and albumin were measured with specific and sensitive radioimmunoassays. 27 D N A was measured by a microfluorometric assay with the Hoechst reagent. 29 The lower limits of detection for these chemical assays were as follows: neutrophil elastase, 10 - l a mol/L, IL-6, 4 pg/ml; IL-8, 35 pg/ml; IL-4, 35 pg/ml; GM-CSF, 8 pg/ml; tryptase, 2 U/L; eosinophil, cationic protein, 2 ng/ml; mucin, 2 ~xg/ml; albumin, 1 ~xg/ml; and DNA, 0.1 ng/ml.

846

F a h y et al.

J ALLERGY CLIN IMMUNOL APRIL 1995

TABLE II. Clinical characteristics of patients with CF Subject No,

Sex/age (yr)

FEV 1 (% predicted}

FVC (% predicted)

Sputum microbiology*

Admittedt

1 2 3 4 5 6 7 8 9 10 11 12

M/30 M/20 M/24 M/24 M/44 M/39 M/28 F/20 F/36 M/33 M/15 M/26

27 23 25 25 32 22 24 32 35 20 45 17

50 67 54 37 49 37 52 67 57 43 61 29

PA SA PA SA PA Aspergillus PS PA PC PA SA PA SA PA SA PA PA PA

Y Y N N Y Y Y Y N N Y Y

FVC, Forced vital capacity;PA, Pseudomonas aeruginosa; PC, Pseudomonas cepacia; SA, Staphylococcus aureus. *Sputum culture results were close to the time of sputum collection. tAdmission to the hospital for management of the exacerbationof CF-related lung disease.

Statistical analysis

Data were entered onto a computer spreadsheet (Excel, Microsoft Corp., Redmond, Wash.) and exported to a statistics program (Statview, Abacus Concepts Inc., Berkeley, Calif.) for descriptive and comparative statistics. Data are presented as mean _+ SEM, and the median values are also presented for most data. Because most of the data were not normally distributed, the Mann-Whitney U test was used in most instances to compare sputum data from the asthmatic and CF groups. An exception was the comparison of tryptase levels in sputum from the asthma and CF groups, in which the chi-square test was used. Spearman's rank order test was used to determine correlations between data. A probability value of less than 0.05, with twotailed tests, was considered significant. RESULTS

We found that neutrophils rather than eosinophils were more frequently the predominant cell in sputum from patients with acute severe asthma (Table III, Fig. 1). Neutrophils made up more than 75% of the nonsquamous cells in 10 samples and constituted more than 90% of cells in five samples. Eosinophils made up more than 75% of the nonsquamous cells in only three samples but were identifiable in all asthmatic sputum samples at lower percentages. Eight of the 16 subjects with asthma (50%) who provided data on whether a head or chest cold precipitated their asthma exacerbation said that it did (Table I). The mean percentage of neutrophils was higher in these subjects than in the others (85% vs 57%, p = 0.05), and the mean percentage of eosinophils was lower but not significantly so (5% vs 31%, p = 0.11).

The majority of samples had detectable free neutrophil elastase activity (11/17 sputum samples in which it was measured) (Table IV). IL-6 and IL-8 levels were detectable in the nanogram range in all samples, but assays for other cytokines (IL-4 and GM-CSF [n = 5]), sensitive to picogram quantities, yielded undetectable levels. Tryptase was detectable in the majority of asthmatic sputum samples but in only one of the CF sputum samples. MLG, albumin, and D N A levels in asthmatic sputum were each measurable in all samples in the milligram range (Table IV). The six sputum samples from patients with CF analyzed for cell differential contained significantly more neutrophils and fewer eosinophils than the 18 asthmatic sputum samples (Table III). Indeed, neutrophils constituted 96.5% to 100% of nonsquamous cells in these samples; eosinophils were identifiable in only two samples. The percentage of squamous cells in CF sputum was lower than in asthmatic sputum (0.9% _+ 0.3% vs 18% _+ 5.9%, p = 0.005). Free neutrophil elastase activity was significantly higher in the 12 CF sputum samples than in 17 asthmatic sputum samples. IL-6 levels in CF sputum were significantly lower than in asthmatic sputum, but IL-8 levels were significantly higher. M L G levels were significantly higher in CF sputum than in asthmatic sputum, but we found no significant difference in albumin levels. The ratio of albumin to M L G was significantly lower for CF sputum. Tryptase was detectable in significantly fewer CF sputum samples than asthmatic sputum samples (Table IV). We correlated neutrophil cell counts and neu-

Fahy et

J ALLERGY CLIN IMMUNOL VOLUME 95, NUMBER 4

al.

847

TABLE III. Total cell counts and cell percentages of nonsquamous cells in sputum from patients with asthma and those with CF

Subject No. Asthma group 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Mean Median SEM CF group 1 2 3 4 5 6 Mean Median SEM

Cell count (106/ml)

Epithelialcells (%)

Macrophages (%)

Neutrophils (%)

Lymphocytes (%)

Eosinophils (%)

7.6 20.2 2.7 47.7 24.5 49.3 10.7 2.9 2.1 13.2 0,5 20.2 30,0 12,9 36.0 18.5 3.7 7.6 17.2" 13.0 3.6

0.6 0.0 0.0 0.0 0.0 0.0 0.0 0.0 4.2 0.0 10.2 0.0 0.0 0.0 0.0 0.5 0.5 0.4 0.9 0.0 0.6

18.2 4.8 16.0 2.9 2.6 4.8 2.3 11.2 7.7 6.2 34.9 12.4 1.6 4.7 35.2 5.3 19.9 3.1 10.8"~ 5.8 2.5

4.5 94.3 69.1 96.9 95.6 94.2 50.8 83.1 87.4 68.2 54.0 84.2 96.8 82.9 63.1 0.0 77.1 10.3 67.3~ 80.0 7.6

0.3 0.2 0.0 0.0 0.0 0.0 0,0 0.0 0.0 0.0 0.0 0.0 0.2 0.3 0.5 0.0 1.5 0.0 0.2 0.0 0.1

76.4 0.7 14.9 0.2 1.8 1.0 47.0 5.6 0.7 25.4 1.0 3.3 1.3 12.2 1.3 94.2 1.0 86.0 20.8§ 2.5 7.6

0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

0.9 0.0 0.4 0.0 0.0 0.0 0.2 0.0 0.1

1~8 2.0 0.4 0.0 0.0 0.0 0.7 0.2 0.4

0.4 0.0 0.8 0.0 0.0 0.0 0.2 0.0 0.1

106.8 83.6 104.9 45.0 26.2 145.9 85.4 94.3 17.9

96.5 99.0 98.4 100.0 100.0 100.0 99.0 99.5 0.6

Symbols denote significant difference from CF sputum as follows: *, p = 0.001; t, P = 0.0003; :~,p = 0.0005; §, p = 0.0009. trophil elastase activity levels with IL-8 levels, neutrophil cell counts with neutrophil elastase activity levels, and neutrophil elastase activity levels with M L G levels for asthmatic and CF sputum samples analyzed together and then for each separate group. The neutrophil cell counts and the levels of free neutrophil elastase activity in sputum from both groups were significantly positively correlated (rs = 0.9, p = 0.0001), and this correlation held up for the 17 data pairs in the asthmatic group (r~. = 0.8, p = 0.003) but not for the six data pairs available for the CF group. The neutrophil cell counts and the levels of IL-8 in sputum from both groups were significantly positively correlated (rs = 0.8, p = 0.0003), and this correlation held up for the 17 data pairs available in the asthmatic group (r, = 0.6, p = 0.02) and had a trend toward

significance for the six data pairs available for the CF group (rs = 0.8, p = 0.08). No such correlation existed for eosinophil numbers and IL-8. There was no significant correlation between the levels of free nentrophil elastase activity and M L G levels when the data from both groups were observed together or separately. DISCUSSION In this study we analyzed the cellular and biochemical constituents of sputum from acutely ill subjects with asthma and compared the findings with those from analysis of sputum from patients with CF. Our main findings are that the predominant inflammatory cell in sputum from acutely ill subjects with asthma is more frequently the neutrophil than the eosinophil and that asthmatic

848

Fahy et al.

J ALLERGYCLINIMMUNOL APRIL 1995

TABLE IV. Free neutrophil elastase activity, tryptase, IL-6, IL-8, DNA, albumin, and MLG levels in sputum from patients with acute asthma and from patients with CF

Subject No.

Asthma group 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Mean Median SEM CF group 1 2 3 4 5 6 7 8 9 10 11 12 Mean Median SEM

Free neutrophil elastase activity (l~g/ml)

Tryptase (U/L)

ND 112.00 ND 92.00 150.00 IS ND 7.90 ND ND ND 49.60 11.00 6.40 32.00 0.08 0.13 0.14 27.1" 0.1 11.3 1400 400.0 480.0 470.0 180.0 140.0 39.7 60.8 780.0 945.0 49.8 645.0 465.9 435 121.4

IL-6 (ng/ml)

IL-8 (ng/ml)

DNA (mg/ ml)

Albumin (mg/ml)

MLG (mg/ml)

Albumin/ MLG ratio

8.00 ND 21.00 9.80 ND IS ND ND IS 55.00 IS ND 2.90 ND ND 12.70 2.10 IS 10.0" 0.0 6.7

IS 1360 940.0 500.0 300.0 800.0 440.0 120.0 1280 600.0 260.0 4320 612.0 680.0 7200 52.00 120.0 240.0 1166.1" 600.0 446.7

108.0 52.0 7.0 168.0 168.0 46.0 88.0 3.0 IS 16.0 4.0 32.0 32.0 5.0 176.0 14.0 5.0 3.0 54.5* 32.0 15.3

0.16 0.13 0.30 0.25 0.39 0.26 0.14 0.02 0.05 0.09 0.05 0.11 5.80 0.20 0.40 0.30 0.06 0.20 0.5"~ 0.2 0.3

1.70 0.35 1.07 0.47 1.20 1.20 0.80 2.00 IS 0.90 IS 0.60 IS 2.00 6.20 2.70 0.50 2.40 1.6 1.2 0.4

191.0 1.92 0.15 0.38 31.90 35.07 5.29 2.30 24.20 27.90 26.40 17.40 IS 4.60 15.30 5.60 1.70 9.60 23.65 9.6 10.9

0.01 0.18 7.13 1.24 0.04 0.03 0.15 0.90 IS 0.03 IS 0.04 IS 0.43 0.40 0.48 0.29 0.25 0.8§ 0.2 0.5

ND ND ND ND ND ND ND 10.3 ND ND ND ND 0.9 0 0.9

11.0 12.0 13.0 12.0 12.0 32.0 45.0 21.0 39.0 27.0 19.0 ND 20.2 15.4 3.8

236.0 248.0 204.0 128.0 62.0 336.0 90.0 192.0 186.0 180.0 84.0 288.0 186.2 189 24.3

3.20 2.50 2.20 1.80 0.50 1.70 1.40 IS 0.50 0.80 IS 0.90 1.5 1.5 0.3

20.7 27.0 15.4 35.3 17.3 18.7 36.4 23.6 85.5 86.0 130.4 174.0 55.9 31.2 15

0.15 0.09 0.14 0.05 0.03 0.09 0.04 IS 0.01 0.01 IS 0.01 0.1 0.005 0.002

13.2 9.5 4.3 4.9 1.7 1.6 1.2 0.8 1.7 0.7 1.3 0.6 3.5 1.6 1.2

IS, Insufficientsample; ND, not detectable. Symbolsdenote significant difference from CF sputum as follows: *,p = 0.0001; t, p = 0.05; ~:,p = 0.008; §,p = 0.02. sputum is also characterized by relatively high levels of DNA, albumin, and MLG. The pathophysiologic mechanisms responsible for these implied abnormalities in cellular infiltration, bronchovascular permeability, and mucus secretion are unknown but may be at least partially explained by the increased levels of IL-8 and IL-6, free neutrophil elastase activity, and tryptase measured in the fluid phase of the sputum. Although o u r findings for the cellular analysis of

asthmatic sputum confirm those of previous studies in showing that eosinophils are often present in asthmatic sputum samples, 6-s the surprising discovery from our analysis is the high proportion of sputum samples in which the neutrophil predominates. Neutrophils comprised more than 75% of the cells in 10 of 18 samples, and in five of these 10 comprised more than 90% of the cells. In contrast, eosinophils comprised more than 75% of the cells in only three of the 18 samples. Our findings are

Fahy et al.

J ALLERGY CLIN IMMUNOL VOLUME 95, NUMBER 4

50-

o o

849

100

80

40.

ego

°

o

o

O

xlO 6 /mL

3o2 2o2

o

o

60 8

%

8

40 o

130

o

O

lllll

20 o

0

°~°o° Total Cell Count

0

oo Neutrophila

8

ooo~oa~,

Eosinophils

FIG. 1. Total cell count and percentages of neutrophils and eosinophils in sputum from 18 subjects with asthma in acute exacerbation. Bars represent mean of data.

not directly comparable to other published studies because these studies do not always report on the neutrophil percentage in the sputum, the severity of the asthma attack in the subjects with asthma studied is variable, and in some instances only asthmatic subgroups are studied (e.g., patients with allergic asthma or asthma with noninfectious exacerbation). For example, in a study of sputum from 51 patients with acute asthma admitted to the hospital, Alfaro et al. 6 reported that eosinophils comprised a mean of 44% of the cells but did not report the percentages of other cell types in the sputum. Baigelman et al. 7 analyzed sputum from 11 patients with asthma with noninfectious exacerbations and found that eosinophils comprised 26% of the cells, but again they did not report the percentages of the other cell types. Gibson et al. s performed a full cell analysis on sputum from 10 patients with asthma with exacerbation seen in the outpatient clinic and found that eosinophils comprised 69% of the cells and neutrophils comprised 2%. In these studies the method of differentiating cells in sputum varies. Some investigators 7, s determined the cell differential on smears of mucus plugs extracted from the entire sputum samples. Others 6 collected sputum directly into 50% ethanol. Our approach was to determine the cell differential of cytocentrifuged preparations of nonsquamous cells in the entire sputum sample? 7 The effect these different methods might have on the cell differentials reported in sputum is uncertain. Our belief is that methodologic differences in sputum processing are less important than clinical differences in the asthmatic populations studied, specifically differences in the precipitant of the asthma exacerbation and differences in the severity of the asthma exacerbation. Neutrophils are not customarily regarded as important in the pathogenesis of acute asthma, but

recently reported histopathologic postmortem examinations of airways of persons with fatal asthma showed that neutrophils predominate over eosinophils in some patients? e, 31 One explanation for neutrophil predominance in airways during an acute asthmatic attack might be its precipitation by viral infection. Neutrophilic inflammation in the lower airways has been reported after influenza A infection 32 and in the upper airways after rhinovirus infection. 33,34 Notably, 50% of our subjects with asthma reported that a head or chest cold precipitated their asthma attack (Table I), and these subjects had a higher percentage of neutrophils in their sputum than the other subjects (85% vs 57%, p = 0.05). Of the subjects with greater than 90% neutrophils in their sputum, all five of those questioned reported symptoms of a preceding viral infection. Obviously, larger studies are needed to investigate more fully the relationship between the precipitants of acute, naturally occurring attacks of acute asthma and the inflammatory mechanisms they activate. It does appear possible, however, that the information provided by experimental aerosolized allergen challenge, the most frequently used model for acute asthma, may not be necessarily pertinent to the mechanisms of most attacks of asthma requiring emergency medical treatment, because allergen exposure does not appear to be the most common precipitant of such attacksY For our biochemical analysis of the asthmatic sputum samples collected in this study we focused on measurements that might shed light on the mechanism of mucus hypersecretion and cellular infiltration in acute asthma (cytoldnes and neutrophil elastase activity), and better describe the chemical constituents of the expectorated sputum (DNA, albumin, MLG). Of relevance to possible mechanisms of mucus hypersecretion is our finding

850

F a h y et al.

of detectable neutrophil elastase activity in many of the samples (Table IV). These elastase levels are in a range that has been shown to cause significant macromolecule secretion from bovine serous gland cells in vitro. 12 In this context our detection of tryptase in the asthmatic sputum samples (Table IV) is relevant, because it suggests that mast cell chymase, another preformed mast cell mediator and a potent secretagogu@ 3 may also be present. Of relevance to possible mechanisms of cellular infiltration in acute asthma is our finding of high levels of IL-8 in the sputum from subjects with asthma (Table IV). Sputum from patients with chronic bronchitis with similar IL-8 levels have been shown to cause chemoattraction of neutrophils in vitro. 36 A relationship between IL-8 and mucus hypersecretion has been proposed by Richman-Eisenstat et al. 36 These authors have suggested that mucin hypersecretion may result from an inflammatory cascade that begins with airway injury (infection, allergen exposure) and IL-8 secretion from epithelial cells or macrophages, and proceeds to IL-8-mediated chemoattraction of neutrophils to the airway and, after activation of neutrophils (by unknown mechanisms), to release of neutrophil proteases and neutrophil proteaseinduced mucin hypersecretion from airway secretory cells. Our findings provide some support for this hypothesis in that we found IL-8 levels to correlate with neutrophil counts, which in turn correlated with neutrophil elastase levels. Although free neutrophil elastase activity did not correlate with MLG levels, this does not negate the possibility of a link between these factors. The pertinent concentrations of free neutrophil elastase for its secretagogue effect are those in tissues immediately adjacent to submucosal gland cells. The distance between these sites and the airway lumen, and the additional influence of luminal protease inhibitors, could result in significant differences between the free elastase detectable in the airway luminal secretions and its concentration at its sites of action. The significance of our finding of very high IL-6 levels in asthmatic sputum is uncertain, but IL-6 is known to promote immunoglobulin production by B cells and to induce hepatic synthesis of acute phase proteins. 24 Increased expression of IL-6 in bronchial epithelial cells from subjects with asthma has been described, 21 and increased levels of IL-6 have been detected in bronchoalveolar lavage fluid from patients with symptomatic asthma. 2°,25 In addition, IL-6 is released in the cutaneous re-

J ALLERGY CLIN IMMUNOL APRIL 1995

sponse to allergen challenge in persons with atopy26 and, interestingly, antiinflammatory actions of IL-6 in the airway have also been described. 37 To provide a reference for our findings in sputum samples from patients with acute asthma, we compared the findings with a similar analysis we conducted on sputum from patients with CF. The mean values for total cell counts, neutrophil percentages, neutrophil elastase, and IL-8 were higher in the CF samples, but some of the asthmatic samples had overlapping values for these variables; that is, the values in some asthmatic samples were in the same range as the samples from patients with CF (Tables III and IV). Thus these two distinct airway diseases may share similar mechanisms for neutrophil recruitment and mucus hypersecretion. Other differences were evident in the cellular and biochemical constituents of sputum from patients with asthma and CF. The asthmatic samples were distinguished by the presence of eosinophils, by higher tryptase and IL-6 levels, and by a higher ratio of albumin to MLG (Table IV). This latter finding may indicate that in acute asthma, compared with CF-related lung disease, plasmaderived proteins contribute relatively more than mucin glycoproteins to luminal secretions. The albumin in asthmatic sputum may have important consequences for the physical properties of the sputum, because in vitro experiments have shown that albumin and mucin combine to form a highly asymmetric and thus highly viscous complex2 7 DNA also contributes to the abnormal viscoelastic properties of sputum from patients with chronic bronchitis and CF, 18,19 and aerosolized recombinant human DNase improves pulmonary function in patients, with CF 38,39 presumably by facilitating mucus clearance. Further studies are required, to determine whether the DNA levels we found in asthmatic sputum are high enough to contribute to secretion retention and whether recombinant DNase treatment of acute asthma might improve pulmonary function. In summary, our observations show that sputum from patients with asthma in acute exacerbation contains a mix of inflammatory cells, mucin glycoproteins, DNA, and albumin. Neutrophils rather than eosinophils were more frequently the predominant inflammatory cell in sputum samples from patients with asthma. The predominance of neutrophils and the high concentrations of IL-8, neutrophil elastase, and MLG in sputum from both asthmatic and CF groups suggests that despite the differences in the two conditions, the

J ALLERGY CLIN IMMUNOL VOLUME 95, NUMBER 4

m u c u s h y p e r s e c r e t i o n a s s o c i a t e d w i t h b o t h m a y in part involve the chemoattraction of neutrophils by IL-8 a n d t h e s t i m u l a t i o n o f s e c r e t i o n f r o m s u b m u cosal g l a n d s a n d g o b l e t cells b y n e u t r o p h i l elastase.

F a h y et al.

17.

18. We thank Carol Basbaum, PhD, and Walter Finkbeiner, M D for providing us with the A10G5 monoclonal antibody, and Alan Gelb, MD; Susan K. Ulrich, MS, RN; and Virginia Hill, RN, for assistance in the collection of sputum samples and characterization of the subjects. REFERENCES

1. Oppenshaw PJM, Turner-Warwick M. Observations on sputum production in patients with variable airflow obstruction: implications for the diagnosis of asthma and chronic bronchitis. Resp Med 1989;83:25-31. 2. Dunnill MS. The pathology of asthma, with special reference to changes in the bronchial mucosa, J Clin Invest 1960;13:27-33. 3. Dunnill MS, Massarella GR, Anderson JA. A comparison of the quantitative anatomy of the bronchi in normal subjects, in status asthmaticus, in chronic bronchitis, and in emphysema. Thorax 1969;24:176-9. 4. Earle BV. Fatal bronchial asthma. Thorax 1953;8:195-206. 5. Cardell BS, Pearson RSB. Death in asthmatics. Thorax 1959;14:341-52. 6. Alfaro C, Sharma OP, Navarro L, Glovsky MM. Inverse correlation of expiratory lung flows and sputum eosinophils in status asthmaticus. Ann Allergy 1989;69:251-4. 7. Baigelman W, Chodosh S, Pizzuto D, Cupples A. Sputum and blood eosinophils during corticosteroid treatment of acute exacerbations of asthma. Am J Med 1983;75:929-36. 8. Gibson PG, Girgis-Gabardo A, Morris MM, et al. Cellular characteristics of sputum from patients with asthma and chronic bronchitis. Thorax 1989;44:689-92. 9. Frigas E, Loegering DA, Solley GO, Farrow GM, Gleich GJ, Elevated levels of eosinophil granule major basic protein in the sputum of patients with bronchial asthma. Mayo Clin Proc 1981;56:345-53. 10. Shimura S, Sasaki T, Sasaki H, Takishima T. Chemical properties of bronchorrhea sputum in bronchial asthma. Chest 1988;94:1211-5. 11. Brogan TD, Ryley HC, Neale L, Yassa J. Soluble proteins of bronchopulmonary secretions from patients with cystic fibrosis, asthma, and bronchitis. Thorax 1975;30:72-9. 12. Sommerhoff CP, Nadel JA, Basbaum CB, Caughey GH. Neutrophil elastase and cathepsin G stimulate secretion from bovine airway gland serous cells. J Clin Invest 1990; 85:682-9. 13. Sommerhoff CP, Caughey GH, Finkbeiner WE, Lazarus SC, Basbaum CB, Nadel JA. Mast cell chymase: a potent secretagogue for airway gland serous cells. J Immunol 1989;142:2450-6. 14. Lundgren JD, Davey RT, Lundgren B, et al. Eosinophil cationic protein stimulates and major basic protein inhibits airway mucus secretion. J ALLERGYCLIN IMMUNOL1991; 87:689-98. 15. Lundgren JD, Shelhamer JH. Pathogenesis of airway mucus hypersecretion. J ALLERGYCLINIMMUNOL1990;85:399417. 16. Fahy JV, Schuster A, Ueki I, Boushey HA, Nadel JA.

19. 20.

21.

22.

23.

24. 25.

26.

27.

28.

29.

30.

31.

32. 33.

34.

851

Mucus hypersecretion in bronchiectasis: the role of neutrophil proteases. Am Rev Respir Dis 1992;146:1430-3. List SJ, Findlay BP, Forstner GG, Forstner JF. Enhancement of the viscosity of mucin by serum albumin. Biochem J 1978;175:565-71. Shak S, Capon D J, HeIImiss R, Marsters SA, Baker CL. Recombinant human DNase reduces the viscosity of cystic fibrosis sputum. Proc Natl Acid Sci U S A 1990;87:9188-92. Picot R, Das I, Reid L. Pus, deoxyribonucleic acid, and sputum viscosity. Thorax 1978;33:235-42. Mattoli S, Mattoso VL, Soloperto M, Allegra L, Fasoli A. Cellular and biochemical characteristics of bronchoalveolar lavage fluid in symptomatic non-allergic asthma. J ALLERGYCLIN IMMUNOL1988;81:41-7. Marini M, Vittori E, Hollemborg J, Mattoli S. Expression of the potent inflammatory cytokines, granulocyte-macrophage-colony-stimulating factor and interleukin-6 and interleukin-8, in bronchial epithelial cells of patients with asthma. J ALLEROYCLIN IMMUNOL1992;89:1001-9. Gosset P, Tsicopoulos A, Wallaert B, et al. Increased secretion of tumor necrosis factor c~ and interleukin-6 by alveolar macrophages consecutive to the development of the late asthmatic reaction. J ALLERGY CLIN IMMUNOL 1991;88:562-71. Gosset P, Tsicopoulos A, Wallaert B, Joseph M, Capron A, Tonnel AB. Tumor necrosis factor alpha and interleukin-6 production by human mononuclear phagocytes from allergic asthmatics after IgE-dependent stimulation. Am Rev Respir Dis 1992;146:768-74. Kelley J. Cytokines of the lung. Am Rev Respir Dis 1990;141: 765-88. Broide DH, Lotz M, Cuomo DA, Federman EC, Wasserman SI. Cytokines in symptomatic asthma airways. J ALLERGY CLIN IMMUNOL1992;89:958-67. Lee CE, Neuland HG, Teaford BF, et al. Interleukin-6 is released in the cutaneous response to allergen challenge in atopic individuals. J ALLERGY CLIN IMMUNOL 1992;89: 1010-20. Fahy JV, Liu J, Wong H, Boushey HA. Cellular and biochemical analysis of induced sputum from asthmatic and healthy subjects. Am Rev Respir Dis 1993;147:1126-31. Fahy JV, Steiger D J, Liu J, Basbanm CB, Finkbeiner WE, Boushey HA. Markers of mucus secretion and DNA levels in induced sputum from asthmatic and healthy subjects. Am Rev Respir Dis 1993;147:1132-37. Cesarone CF, Bolognesi C, Santi L. Improved microfluorometric DNA determination in biologic material using 33258 Hoechst. Anal Biocbem 1979;100:188-97. Sur S, Croty TB, Kephart GM, et al. Sudden-onset fatal asthma: a distinct entity with few eosinophils and relatively more neutrophils in the airway mucosa. Am Rev Respir Dis 1993;148:713-9. Fujisawa T, Kephart GM, Gray BH, Gleich GJ. The neutrophil and chronic airway inflammation. Am Rev Respir Dis 1990;141:689-97. Hers JF. Disturbances of ciliated epithelium due to influenza virus. Am Rev Respir Dis 1966;93:162-71. Nacleiro RM, Proud D, Lichtenstein LM, et al. Kinins are generated during experimental rhinovirus colds. J Infect Dis 1988;157:133-42. Levandowsky RA, Weaver CW, Jackson GG. Nasal secretion leukocyte populations determined by flow cytometry during acute rhinovirus infection. J Med Virol 1988;25:42332.

852

F a h y et al.

35. Braman SS, Kaemmerlen JT. Intensive care of status asthmaticus. JAMA 1990;264:366-8. 36. Richman-Eisenstat JBY, Jorens PG, Hebert C, Ueki I, Nadel JA. Interleukin-8: an important chemoattractant in sputum of patients with chronic inflammatory airway diseases. Am J Physiol 1993;264:IA13-8. 37. Ulich TR, Yin S, Guo K, Yi ES, Remick D, del Castillo J. Intratracheal injection of endotoxin and cytokines. II. Interleukin-6 and transforming growth factor beta inhibit acute inflammation. Am J Pathol 1991;138:1097-101.

J ALLERGY CLIN IMMUNOL APRIL 1995

38. Aitkin ML, Burke W, McDonald G, Shak S, Montgomery B, Smith A. Recombinant human DNase inhalation in normal subjects and patients with cystic fibrosis. JAMA 1992;262:1947-51. 39. Hubbard RC, McElvaney NG, Birrer P, et al. A preliminary study of aerosolized recombinant human deoxyribonuclease 1 in the treatment of cystic fibrosis. N Engl J Med 1992;326:812-5.