Abstract
A reliable predictor of benefit from corticosteroid treatment in patients with chronic airflow limitation is needed. In a single-blind, sequential crossover trial of placebo and prednisone (30 mg/day) treatment, with each given for 2 wk, we investigated whether an increased proportion of sputum eosinophils ( ⩾ 3%) predicts a beneficial effect of prednisone in smokers with severe obstructive bronchitis. Patients were seen before and after each treatment. Clinical measurements were made blind to the laboratory findings and vice-versa. Eighteen of 20 patients completed the study. Eight had sputum eosinophilia and similar clinical and physiologic characteristics to those of 10 patients without a finding of sputum eosinophilia. Only in patients with sputum eosinophilia did prednisone, as compared with placebo, produce a statistically significant and clinically important mean effect on effort dyspnea of 0.8 (95% confidence interval [CI]: 0.3 to 1.2), p = 0.008, and in quality of life of 1.96 (95% CI: 0.5 to 3.3), p = 0.01, associated with a small improvement in FEV1 of 0.11 L (95% CI: − 0.04 to 0.23 L), p = 0.05. In these patients, prednisone also produced a significant decline in the median sputum eosinophil percentage, from 9.7% to 0.5% (p = 0.002), eosinophil cationic protein (ECP), from 6,000 μ g/L to 1,140 μ g/L (p < 0.001), and fibrinogen, from 25.3 mg/L to 5.4 mg/L (p < 0.001). These findings indicate that in smokers with severe airflow limitation, sputum eosinophilia predicts a beneficial effect of prednisone treatment. Improvement in FEV1, after prednisone treatment in this population, is small, and may not be appreciated in clinical practice.
Asthma and chronic obstructive bronchitis caused by cigarette smoking are common conditions characterized by airflow limitation and airway inflammation. Although these disorders are ordinarily distinct from one another, some patients with asthma develop chronic airflow limitation (CAL), and some smokers with chronic obstructive bronchitis have features of asthma (i.e., variable airflow limitation and airway hyperresponsiveness) (1, 2). The airway inflammation in asthma typically includes infiltration by activated eosinophils and mast cells (3). In contrast, obstructive bronchitis in smokers is characterized by an increase in activated lymphocytes (4), neutrophils (5), and macrophages (4), although eosinophilia has also been reported (6, 7). Both disorders include increased microvascular permeability, with plasma exudation and mucus hypersecretion (8).
Treatment with corticosteroids alleviates asthma and reduces airway eosinophilia (9). In contrast, the effects of steroid treatment in smokers with chronic obstructive bronchitis is controversial (10), even though some patients can benefit from corticosteroid treatment (11). Two studies (12, 13) suggested that sputum eosinophilia is a predictor of clinical benefit from steroid treatment, but little attention was given to these observations because sputum examination was considered to be unreliable. More recently, Chanez and colleagues (7) used bronchial biopsies and bronchoalveolar lavage (BAL) to examine the effects of open treatment with prednisolone in patients with chronic obstructive pulmonary disease (COPD). They showed that those who improved had the usual airway inflammatory features of asthma.
Recently, the method of sputum examination has been refined and shown to be a reliable (14), valid (14), and responsive (15) means of assessing airway inflammation. In addition, the method of inducing sputum with a saline aerosol can be modified to make it safe and successful even in patients with severe airflow limitation (15). Therefore, we performed a placebo-controlled, crossover study to investigate whether the presence of sputum eosinophils predicts benefit from prednisone in smokers with severe chronic obstructive bronchitis. The study was designed to be single blind and sequential, to avoid a possible carryover effect of prednisone treatment. To avoid bias, clinical measurements were performed blind to the laboratory characteristics, and vice versa.
Current or previous cigarette smokers with stable CAL and persistent mucoid sputum were recruited from the Firestone Regional Chest and Allergy Clinic, from the Respiratory Medicine Department of the John Hunter Hospital, and through other referrals. All had an FEV1/ VC ratio or baseline FEV1 < 60% after salbutamol (two puffs of 100 μg/puff), a late onset of respiratory symptoms, dyspnea with exercise limitation, and no history of asthma or episodic variability in symptoms (Table 1). Although 14 of the subjects had occasional wheezing, there was little or no bronchodilator response. All subjects were stable with the same regular treatment, without a recent increase in dyspnea, cough, or sputum production that caused any subject to seek medical attention. For at least 2 mo before the first study visit, the treatment included inhaled steroids (12 subjects) and bronchodilators (inhaled β2-agonist [nine subjects; 200 to 1,600 μg/d], inhaled ipatropium [five subjects; 160 to 320 μg/d] and theophylline [one subject; 600 mg/d]). Nine subjects had a history of recurrent respiratory infections, and to prevent the confounding influence of possible infection during the study, they received treatment (1 wk before the study) with cotrimoxazole (two tablets twice daily for a week) or amoxicillin (500 mg three times a day for a week), followed by half of the dose during the study. None of the subjects had other chronic respiratory or systemic conditions that could have interfered with the results of the study. The study was approved by the hospital research committees of St. Joseph's and John Hunter Hospitals, and all patients gave written informed consent.
| Sputum Eosinophils | ||||
|---|---|---|---|---|
| ⩾ 3% | < 3% | |||
| Subjects, n | 8 | 10 | ||
| Age, yr | 72 (63–81) | 63 (39–72) | ||
| Sex, m | 8 | 9 | ||
| Smoking, mean (SD) pack/yr | 31.3 (7.8) | 43.6 (41.0) | ||
| Atopic, n | 4 | 5 | ||
| FEV1, L | 0.9 (0.4–1.3) | 1.1 (0.4–1.3) | ||
| FEV1, % pred | 28 (17–38) | 29 (13–48) | ||
| FEV1/VC, % | 31 (18–43) | 36 (16–51) | ||
| ΔFEV1 after bronchodilator, %* | 4.3 (0.0–11.9) | 6.6 (1.6–12.4) | ||
| On inhaled steroids, n | 7 | 8 | ||
| Inhaled steroid dose, μg/d | 1,000 (0–2,000) | 1,000 (0–2,000) | ||
| Antibiotic, n† | 4 | 5 | ||
| Blood eosinophils, × 106/ L‡ | 0.3 (0.2) | 0.2 (0.2) | ||
The study was a two-center, before-and-after treatment, single-blind, placebo-controlled crossover study to investigate whether sputum eosinophilia of ⩾ 3% would predict clinical improvement from treatment with a short course of prednisone. Subjects received placebo in three tablets given twice daily for 2 wk, followed by three tablets of prednisone (15 mg) twice daily for 2 wk. Individual subjects made three morning visits, all at the same time. They were asked to withhold inhaled bronchodilators for at least 6 h before these appointments. At Visit 1, subject characteristics were documented. Subjects were instructed to continue their prestudy treatment, to record symptoms and medication use in a daily diary, and to contact a study physician if a worsening of symptoms occurred. Procedures performed at each visit were an assessment of quality of life (QoL), spirometry, and sputum and blood examination for inflammatory indices. In addition, at Visits 2 and 3, daily diaries were reviewed. A research assistant blinded to the study outcomes delivered the medications and explained how to use them. Compliance was assessed by counting remaining tablets, and for 15 subjects was over 95% and for the remainder was over 92%. The primary outcomes were improvement in QoL scores and pre- and post-bronchodilator FEV1. Secondary outcomes were changes in sputum total cell count; eosinophil and neutrophil counts; eosinophil cationic protein (ECP); myeloperoxidase (MPO); elastase; fibrinogen; blood eosinophils count; and serum ECP.
Subject characteristics were documented with a structured questionnaire. QoL was assessed by using the chronic respiratory disease questionnaire (CRQ), a 20-item questionnaire measuring four domains (dyspnea, fatigue, mastering, and emotions), that has been previously validated and fully described (16). Briefly, the first domain, dyspnea, was examined by asking subjects to specify five important and frequent activities in which they experienced shortness of breath. The intensity experienced during these five activities was measured serially, using a seven-point Likert scale. The same five activities identified on Visit 1 were used for all subsequent visits. Each of the other domains included five questions, and each was also measured serially with a seven-point Likert scale. During the study period, symptoms (chest tightness, shortness of breath, wheezing, cough, sputum production, and sputum characteristics) and the use of medications were recorded twice daily in a diary. Spirometry was performed with a PK Morgan 131 Spiroflow dry rolling-seal spirometer (Roxon Medi-Tech, Rexdale, ON, Canada) according to American Thoracic Society (ATS) specifications (17), and reference values were taken from Crapo (18). The baseline FEV1 was calculated as the best of three reproducible values (within a maximum change of 5%). Two expiratory and one inspiratory slow Vc measurements were made, in this sequence. Salbutamol (two puffs of 100 μg/puff) was inhaled through a large volume spacer (Ventahaler, 750 ml; Glaxo-Canada), and spirometry was repeated after 20 min.
Sputum induction was done as previously described (15) if subjects were unable to produce sputum spontaneously. Briefly, sputum was induced by inhalation of an aerosol of normal saline (NS) followed by hypertonic saline (1.5, 3, 4, and 5%) generated by a Fisoneb ultrasonic nebulizer (Canadian Medical Products, Ltd., Markham, ON, Canada) with an output of 0.87 ml/min and particle size of 5.58 μm aerodynamic mass-median diameter. The aerosol was inhaled for 1 or 2 min, according to the severity of the subject's airflow limitation, and FEV1 was measured. Subjects were asked to blow their noses, rinse their mouths, and swallow the water in the aerosol to minimize contamination with postnasal drip and saliva. They were instructed to cough up sputum into a sterile container whenever they felt that sputum might be present. These procedures were repeated sequentially with each concentration of saline for a total of up to 6 or 7 min of inhalation, until a sputum sample was obtained or a decrease in FEV1 ⩾ 10% occurred.
The appearance of the sputum was recorded (mucoid, mucopurulent, or purulent). Sputum selected from saliva was processed within 2 h as described by Pizzichini and colleagues (14). Briefly, sputum was treated by adding four volumes of 0.1% dithiothreitol (DTT) (Sputalysin 10%; Calbiochem Corp., San Diego, CA) followed by four volumes of Dulbecco's phosphate-buffered saline (D-PBS). The suspension was filtered through a 48-μm nylon gauze (BBSH Thompson, Scarborough, ON, Canada), and total counts of leukocytes and evaluations of cell viability (trypan blue exclusion method) were made. The filtrate was centrifuged at 3,000 rpm for 4 min, and the supernatant was aspirated and stored in Eppendorf tubes at −70° C. The cell suspension was adjusted to 1.0 × 106/ml, and cytocentrifuge preparations were made with 60 μl of the cell suspension (Shandon III cytocentrifuge; Shandon Southern Instruments, Sewickley, PA). One cytospin preparation was air dried and stained with Wright's stain. Four hundred nonsquamous cells were counted in Wright's-stained cytospin preparations, and the results were expressed as a percentage of the total nonsquamous cell count.
The concentrations of ECP and MPO in μg/L in the sputum supernatant were determined with a radioimmunoassay (RIA) (Kabi Pharmacia Diagnostics AB, Uppsala, Sweden). Fibrinogen was measured with a sandwich-type enzyme-linked immunosorbent assay (ELISA), using a rabbit antihuman fibrinogen antibody (Dako A080; Dakopatts A/S, Denmark). Interleukin-5 (IL-5) was measured in pg/ml with a quantitative “sandwich” enzyme immunoassay (Quantikine; R&D Systems, Inc., Minneapolis, MN). Elastase was measured with an ELISA, using human neutrophil elastase as previously described (19). The limits of detection of the fluid-phase assays for ECP, MPO, fibrinogen, IL-5, and elastase were, respectively, 2.0 μg/L, 8 μg/L, 0.79 μg/L, 7.8 pg/ml, and 0.25 μg/L. We have shown that both cell and ECP levels in selected induced sputum are similar to those in selected spontaneous sputum (20).
Venous blood was collected into a 5.0 mL tube containing ethylenediamine tetraacetic acid (EDTA) (K3 Vacutainer; Becton-Dickinson, Rutherford, NJ), and a differential white cell count was made with a Coulter STKS counter (Coulter Corp., Hialeah, FL). Serum was collected, centrifuged at 20° C at 1,500 rpm for 10 min, and stored at −20° C. The assay used to measure serum ECP was the same as that described for ECP in sputum.
All data were analyzed with the SPSS statistical package for Windows, release 7.0 (SPSS, Inc., Chicago, IL). Results are reported as median and as interquartile range (IQR) calculated with weighted averages. Dependent variables with nonnormal distribution (total cell count, eosinophils, metachromatic cells, ECP, fibrinogen, elastase, and MPO) were log transformed before analysis. Differences between groups at baseline were calculated with an unpaired t test with Bonferroni's adjustment for multiple comparisons. On the basis of our previous data (21), eosinophilia was defined as ⩾ 3% sputum eosinophils. Repeated-measures analysis of variance (ANOVA) was used in a model to examine the effect of treatment, within eosinophilic and noneosinophilic groups, on dependent variables (CRQ scores, FEV1, and sputum and blood inflammatory indices). Significance was accepted at the level of 95%, and the source of significant variation was identified with the Student–Newman–Keuls procedure to adjust the significance for multiple comparisons (22). The agreement between measurements of FEV1 at Visits 1 and 2, before and after bronchodilator use, was determined by calculating the intraclass correlation coefficient (R).
Twenty subjects were enrolled in the study. Two withdrew before completion of the placebo period, one for gastric intolerance to the study medication and the other for a complication unrelated to the study, and were not analyzed. The subject characteristics were similar in the two study groups (Table 1). The reproducibility of the pre- and postbronchodilator FEV1 measured at Visits 1 and 2 was high both for subjects with eosinophilic (R = 0.89 for both pre- and postbronchodilator FEV1) and noneosinophilic sputum (R = 0.84 and 0.90 for pre- and postbronchodilator FEV1, respectively).
The baseline CRQ scores and the pre- and postbronchodilator FEV1 values did not differ for subjects with eosinophilic and those with noneosinophilic sputum (Table 2). Seven subjects produced sputum spontaneously at each visit, three subjects had sputum induced on each occasion, and the remaining 10 subjects had either spontaneous or induced sputum. Spontaneous (n = 32) and induced sputum (n = 21) samples had similar median and IQR viability of cells (69.0 ± 42.8% versus 75.0 ± 19.7%, p = 0.4), and similar and trivial salivary squamous cell contamination (2.0 ± 1.4% versus 1.0 ± 1.5%, p = 0.3). One of the induced samples from one subject (on Visit 2) had too many degenerated cells to permit accurate cell counts on cytospin preparations; however, the supernatant was kept and examined. The inflammatory indices in sputum did not differ in the two study groups, with the exception of a higher proportion and number of eosinophils (5.4 ± 4.2% versus ± 1.0 ± 1.8%, p = 0.01, and 58.0 ± 374.0 × 104/ml versus 8.0 ± 8.0 × 104/ml, p < 0.001, respectively). IL-5 was under the limit of detection in all patients.
| Eosinophilic† | Noneosinophilic | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Outcome | Baseline | Placebo | Prednisone | Baseline | Placebo | Prednisone | ||||||
| Quality of life* | ||||||||||||
| Dyspnea | 4.0 (0.8) | 4.2 (0.5) | 5.0 (0.6)§ | 3.1 (1.3) | 3.3 (1.3) | 3.6 (1.6) | ||||||
| Fatigue | 4.2 (0.8) | 4.4 (0.8) | 4.7 (0.9) | 4.4 (1.2) | 4.3 (1.2) | 4.7 (1.1) | ||||||
| Emotions | 4.7 (1.0) | 4.8 (0.9) | 5.2 (0.9)§ | 4.9 (1.4) | 4.7 (1.2) | 4.8 (1.2) | ||||||
| Mastering | 4.7 (0.9) | 4.9 (0.7) | 5.3 (0.7) | 4.7 (1.7) | 4.5 (1.8) | 4.7 (1.7) | ||||||
| CRQ score | 17.6 (2.6) | 18.3 (2.3) | 20.2 (2.4)§ | 17.1 (4.6) | 16.9 (4.5) | 17.8 (4.5) | ||||||
| FEV1, L* | ||||||||||||
| Prebronchodilator | 0.89 (0.2) | 0.89 (0.3) | 1.0 (0.3)¶ | 0.9 (0.3) | 0.89 (0.3) | 0.9 (0.3) | ||||||
| Postbronchodilator | 1.0 (0.3) | 1.0 (0.3) | 1.2 (0.3)¶ | 1.2 (0.4) | 1.2 (0.4) | 1.2 (0.4) | ||||||
| Sputum indices*,‡ | ||||||||||||
| Squamous cells, % | 1.2 (2.3) | 2.0 (5.0) | 8 (9.0) | 0.6 (2.0) | 1.0 (1.6) | 2.0 (1.5) | ||||||
| Cell viability % | 64 (54) | 68 (30) | 65 (31) | 74 (16) | 66 (30) | 72 (20) | ||||||
| TCC, × 106/ml | 8.3 (8.1) | 7.6 (6.5) | 3.3 (15.1)¶ | 11.1 (14.1) | 8.7 (6.8) | 10.0 (7.7) | ||||||
| Eosinophils, % | 5.4 (4.2) | 9.7 (19.5) | 0.5 (1.7)§ | 1.0 (1.8) | 0.6 (1.6) | 0.3 (0.5) | ||||||
| Neutrophils, % | 73.4 (18.5) | 73.0 (21.3) | 84.5 (16.5) | 83.4 (10.9) | 81.7 (17.2) | 86.2 (14.2) | ||||||
| ECP, μg/L | 5,240 (5,720) | 6,000 (5,520) | 1,140 (728)‖ | 2,560 (4,916) | 1,460 (2,435) | 1,720 (4,002) | ||||||
| Fibrinogen, mg/L | 24.6 (38.1) | 25.3 (40.8) | 5.9 (4.2)‖ | 18.1 (35.4) | 29.4 (77.3) | 13.3 (27.9) | ||||||
| MPO, μg/L | 1,976 (5,304) | 4,224 (2,794) | 3,612 (4,666) | 3,256 (4,698) | 3,912 (3,048) | 4,056 (7,648) | ||||||
| Elastase, μg/L | 2,996 (643) | 3,160 (6,592) | 3,020 (5,665) | 2,512 (3,046) | 2,984 (8,008) | 3,176 (28,596) | ||||||
We examined the effects of treatment on CRQ scores, FEV1 (L) before and after bronchodilator use, and sputum inflammatory indices (Table 2 and Figures 1 and 2). Treatment with prednisone significantly improved dyspnea, QoL, and FEV1 in patients with sputum eosinophilia. The mean paired difference between prednisone and placebo treatment in the CQR for dyspnea was 0.8 points (95% confidence interval [CI]: 0.3 to 1.2 points) (p = 0.008), while that for QoL was 1.96 points (95% CI: 0.3 to 3.3 points) (p = 0.01), and the differences for the pre- and post-bronchodilator FEV1 were 0.1 L (95% CI: −0.04 to 0.23 L) (p = 0.05) and 0.11 L (95% CI: −0.09 to 0.22 L) (p = 0.05), respectively. In contrast, in patients without sputum eosinophilia, no improvement was observed in dyspnea (0.3 points [95% CI: −0.3 to 0.8 points]), QoL (0.9 points [95% CI −0.4 to 2.2 points]), or pre- and postbronchodilator FEV1 (−0.05 L [95% CI: −0.14 to 0.12 L] and 0.01 L [95% CI: −0.04 to 0.09 L]), respectively (p = 0.5). Prednisone treatment produced a significant decrease in sputum eosinophils, ECP, and fibrinogen in patients with sputum eosinophilia. It also produced a partial decrease in sputum fibrinogen in subjects with noneosinophilic inflammation. Sputum neutrophils and fluid-phase levels of elastase and MPO did not change after treatment in either group.
Fig. 1. Individual values of postbronchodilator FEV1, dyspnea domain, QoL, and sputum eosinophils. Closed circles are values from subjects with sputum eosinophilia (eosinophils ⩾ 3%) and open circles represent values from subjects without sputum eosinophilia. Horizontal bars represent mean values for postbronchodilator FEV1, dyspnea domain, QoL, and median values for sputum eosinophils. Dashed line represents upper limit of normal range. Prednisone treatment improved clinical outcomes only in subjects with sputum eosinophilia; *p = 0.05, **p < 0.01.
[More] [Minimize]
Fig. 2. Median changes in sputum neutrophils (%), elastase (μg/ L), fibrinogen (mg/L), and ECP (μg/L) after treatment with placebo and prednisone. Symbols as in Figure 1. Dashed line represents upper limit of normal range. There is a reduction in fibrinogen and ECP only in subjects with sputum eosinophilia. Prednisone treatment reduced fibrinogen in both groups, but this was significant only in subjects with sputum eosinophilia.
[More] [Minimize]Sputum induction was required on 21 occasions (Visit 1: six subjects, Visit 2, eight subjects, and Visit 3, seven subjects for subjects both with (n = 7) and without sputum eosinophilia (n = 14), and the mean duration of the procedure was similar in both groups at 5.6 min (95% CI: 3.4 to 7.7 min) and 5.3 min (95% CI: 4.1 to 6.3 min), respectively. The final saline aerosol concentration reached in each induction was 0.9%, 1.5%, and 3.0% on 10, two, and nine occasions, respectively. The FEV1 before induction was not different for patients with eosinophilic and those with noneosinophilic sputum, 1.2 L (95% CI: 1.1 to 1.4 L) and 1.2 L (95% CI: 1.0 to 1.4 L), respectively. After sputum induction there was a decrease in the FEV1 of 12.4% (95% CI: 3.4 to 21.5%) in patients with eosinophilic sputum, and of 18.6% (95% CI: 14.5 to 22.7%) in patients with noneosinophilic sputum (p = 0.1).
The results of this study show that although some patients with severe CAL caused by cigarette smoking can show sputum eosinophilia, they may not be distinguishable from patients without eosinophilia on the basis of physiologic parameters, QoL scores, blood eosinophil counts, and serum ECP levels, or on the basis of sputum ECP, fibrinogen, MPO, or elastase levels. Treatment of patients with sputum eosinophilia with prednisone at 30 mg/d for 2 wk produced a clinically important and statistically significant effect on dyspnea during day-to-day activities, and in QoL, and a small but significant improvement in FEV1. In addition, prednisone treatment completely abolished eosinophilia and resulted in significant reductions in fluid-phase ECP and fibrinogen in patients with sputum eosinophilia, whereas it had no significant effect on eosinophils, ECP, and fibrinogen in the sputum of subjects without sputum eosinophilia. These findings suggest that sputum eosinophilia predicts benefit from prednisone treatment in patients with severe CAL caused by cigarette smoking.
The finding is important, since there is no recognized predictor of benefit from corticosteroid treatment in these patients. If asthma is defined as variable airflow limitation (23), its absence or presence does not predict benefit (24, 25). In clinical practice, knowledge of whether corticosteroid treatment is beneficial is also problematic, and patients are often treated with an inhaled corticosteroid to give them the benefit of doubt. For this reason, we did not exclude patients taking regular treatment with inhaled steroids from the present study. In terms of outcome, steroid treatment would have minimized improvement, yet despite its use, significant improvement was demonstrated in patients with sputum eosinophilia. The improvement in FEV1 was small, an observation that may be relevant to the problem in practice of knowing whether steroid treatment is effective or not.
The small improvement in FEV1 in association with a large CI is a reflection of both the small sample size and the lack of improvement in some subjects. This could suggest, at first glance, that the evidence for eosinophilia as a predictor of treatment benefit is imprecise. The other outcome measurements, however, do not support this interpretation. The clinically important and significant treatment effect on dyspnea and in the overall QoL score, and the marked effect on airway inflammation observed only in the group of subjects with sputum eosinophilia, indicate that sputum eosinophilia in these patients is a predictor of response to prednisone treatment. The discrepancy between the changes in FEV1 and in the inflammatory indices examined in the study, which usually improve in parallel, may be a reflection of persistent structural changes observed in these subjects with severe CAL.
The results of the study support the observations of others that asthmatic-type inflammation can occur in smokers with or without CAL, that 20 to 30% of patients with CAL may experience a significant improvement in FEV1 from short-term corticosteroid use (7, 10, 11), and that sputum eosinophilia may predict those patients who will benefit from such therapy (6, 7, 13). Past studies with sputum have been inconclusive. The sensitivity of sputum eosinophilia for detecting a response to short-term corticosteroid therapy ranged from 0% (26, 27) to 100% (13, 28). The methodology for assessing sputum eosinophilia in these studies varied widely and was not validated. In particular, sputum eosinophils were scored semiquantitatively, using a scale that permitted eosinophil percentages of up to 10% to be scored as negative for eosinophilia. In the present study we used a refined method of sputum examination that is reliable and in which a proportion of eosinophils of greater than 3% is abnormal (14, 21). The results suggest that patients who have CAL with sputum eosinophilia will benefit from prednisone treatment. The benefit was demonstrated with QoL assessment and careful monitoring of FEV1.
The sputum eosinophilia in these patients was not accompanied by measurable levels of interleukin-5 (IL-5). In contrast, IL-5 has been measured in eosinophilic sputum of patients with uncontrolled (14) and exacerbated asthma (15). The results of the present study are in keeping with the findings of Saetta and colleagues (29), who were unable to demonstrate increased expression of IL-5 protein in the bronchial mucosa of patients with chronic bronchitis from smoking who were examined during an eosinophilic exacerbation. The inability to measure IL-5 in the sputum of these patients may have been methodologic or may indicate that other cytokines (IL-3, granulocyte-macrophage colony-stimulating factor [GM-CSF]) or chemokines (released on activation, normal T-cell expressed and secreted [RANTES], eotaxin), or different mechanisms, are involved in the recruitment of eosinophils into the airways.
The QoL questionnaire used in our study is of proven relevance to assessing patients with CAL (30). The improvement in QoL was restricted to the group of subjects with sputum eosinophilia. The improvement in dyspnea was associated with a small improvement in FEV1 that may be easily missed in clinical practice. This finding is common when investigating the effect of steroid treatment in severe CAL, and is not surprising. There are several determinants of airway caliber in CAL, including airway inflammation with a mixed inflammatory cell infiltrate, mucous gland hypertrophy, and loss of elastic recoil. Apart from airway inflammation with eosinophilia, these determinants are unlikely to be influenced by short-term corticosteroid use. Although other physiologic outcomes than FEV1 should be investigated in future studies, sputum eosinophilia appears to be a predictor of benefit of short-term prednisone therapy. If sputum induction with an aerosol of hypertonic saline is required, this and other studies indicate that it can be performed safely (15, 31).
Of the other sputum indices measured, none was able to predict benefit from prednisone treatment. This was so even for ECP, which is considered to be a marker of eosinophil activation. For some yet unexplained reason, increased levels of sputum ECP are observed in smokers with CAL (31). Although these levels were reduced by treatment in the eosinophilic group of subjects in the present study, they remained elevated and were unaffected by treatment in the noneosinophilic group. Sputum neutrophilia with activation and production of MPO and elastase is a recognized feature of smokers' obstructive bronchitis (32, 33), and was also unaffected by prednisone treatment. Sputum fibrinogen may be a marker of microvascular leakage as well as of local fibrinogen production. It is observed in several airway inflammatory conditions (14, 15, 34). In the present study, fibrinogen was reduced by prednisone treatment in both subject groups, although a statistically significant reduction was observed only in subjects with eosinophilic sputum.
In conclusion, patients with smoker's bronchitis and CAL may present with sputum eosinophilia, and this is associated with a clinically important response to short-term prednisone therapy as measured by QoL score and FEV1. In the present study, the improvement in these parameters was parallelled by a significant reduction in eosinophilia and eosinophil activation as indicated by sputum ECP levels, but not by changes in neutrophils or neutrophil proteases. Patients without sputum eosinophilia did not show clinical benefit from short-term prednisone therapy. These findings suggest that sputum eosinophilia is a useful measurement in the clinical management of patients with CAL, enabling the prediction of benefit from prednisone treatment.
The authors thank the patients who participated in this study, Dr. G. Guyatt for his suggestions on study design, G. Cox and S. Pugsley for help with recruiting the patients, M. Morris for help with clinical procedures, A. Efthimiadis for supervising the examination of sputum, S. Weston for helping with sputum cell counts, S. Evans for performing most of the fluid-phase measurements, and Diane Squillace for measurements of fluid-phase elastase. We also thank Apotex Inc., Etobicoke, Ontario, for the placebo tablets, and Pharmacia Diagnostics AB, Uppsala, Sweden, for the ECP assay kits.
Supported by a grant from Ontario Thoracic Society.
| 1. | Ramsdale E. H., Morris M. M., Roberts R. S., Hargreave F. E.Bronchial responsiveness to methacholine in chronic bronchitis: relationship to airflow obstruction and cold air responsiveness. Thorax391984912918 |
| 2. | Pride N. B.Bronchial hyperrectivity in smokers. Eur. Respir. J.11988485487 |
| 3. | Jeffery P. K.Bronchial biopsies and airway inflammation. Eur. Respir. J.9199615831587 |
| 4. | O'Shaugnessy T. C., Ansari T. W., Barnes N. C., Jeffery P. K.Inflammation in bronchial biopsies of subjects with chronic bronchitis: inverse relationship of CD8+ T lymphocyte with FEV1. Am. J. Respir. Crit. Care Med.1551997852827 |
| 5. | Laitinen A., Laitinen L. A.Cellular infiltrates in asthma and chronic obstructive pulmonary disease. Am. Rev. Respir. Dis.143199111591160 |
| 6. | Saetta M., Di Stefano A., Maestrelli P., Turato G., Ruggieri M. P., Calcagni P., Mapp C. E., Ciacia A., Fabbri L. M.Airway eosinophilia in chronic bronchitis during exacerbations. Am. J. Respir. Crit. Care Med.150199416461652 |
| 7. | Chanez P., Vignola A. M., O'Shaugnessy T., Enander I., Li D., Jeffery P. K., Bousquet J.Corticosteroid reversibility in COPD is related to features of asthma. Am. J. Respir. Crit. Care Med.155199715291534 |
| 8. | Corrigan C. J., Kay A. B.The role of inflammatory cells in the pathogenesis of asthma and chronic obstructive pulmonary disease. Am. Rev. Respir. Dis.143199111651168 |
| 9. | Barnes P. J.Drug therapy: a new approach to treatment of asthma. N. Engl. J. Med.321198915171527 |
| 10. | Schayck C. P., Grunsven P. M., Dekhuijzen P. N. R.Do patients with COPD benefit from treatment with inhaled corticosteroids? Eur. Respir. J.9199619691972 |
| 11. | Callahan C. M., Dittus R. S., Katz B. P.Oral corticosteroid therapy for patients with stable chronic obstructive pulmonary disease: a meta-analysis. Ann. Intern. Med.1141991216223 |
| 12. | Brown H. M.Treatment of chronic asthma with prednisolone: significance of eosinophils in the sputum. Lancet1195812451247 |
| 13. | Shim C., Stover C. E., Williams M. P.Response to corticosteroids in chronic bronchitis. 1978. J. Allergy Clin. Immunol.621978363367 |
| 14. | Pizzichini E., Pizzichini M. M. M., Efthimiadis A., Evans S., Morris M. M., Squillace D., Gleich G. J., Dolovich J., Hargreave F. E.Indices of airway inflammation in induced sputum: reproducibility and validity of cell and fluid phase measurements. Am. J. Respir. Crit. Care Med.1541996808817 |
| 15. | Pizzichini M. M. M., Pizzichini E., Clelland L., Efthimiadis A., Mahony J., Dolovich J., Hargreave F. E.Sputum in severe exacerbations of asthma: kinetics of inflammatory indices after prednisone treatment. Am. J. Respir. Crit. Care Med.155199715011508 |
| 16. | Guyatt G. H., Berman L. B., Townsed M., Pugsley S., Chambers L.A measure of quality of life for clinical trials in chronic lung disease. Thorax421987773778 |
| 17. | American Thoracic SocietyStandardization of spirometry: 1987 update. Am. Rev. Respir. Dis.136198712851298 |
| 18. | Crapo R. O., Morris A. H., Gardner R. M.Reference spirometric values using techniques and equipment that meets ATS recommendation. Am. Rev. Respir. Dis.1231981659694 |
| 19. | Hunt L. V., Gleich G. J., Ohnishi T., Weiler D. A., Mansfield E. S., Kita H., Sur S.Endotoxin contamination causes neutrophilia following pulmonary antigen challenge. Am. J. Respir. Crit. Care Med.149199414711475 |
| 20. | Pizzichini M. M. M., Popov T., Pizzichini E., Efthimiadis A., Hussak P., Evans S., Dolovich J., Hargreave F. E.Spontaneous and induced sputum compared. Am. J. Respir. Crit. Care Med.1541996866869 |
| 21. | Pizzichini E., Pizzichini M. M. M., Efthimiadis A., Dolovich J., Hargreave F. E.Measuring airway inflammation in asthma: eosinophils and ECP in induced sputum compared with peripheral blood. J. Allergy Clin. Immunol.991997539544 |
| 22. | Glantz, S. A., and B. K. Slinker. 1990. Repeated Measures. In S. A. Glantz and B. K. Slinker, editors. Primer of Applied Regression and Analysis of Variance. McGraw-Hill, New York. 381–420. |
| 23. | Scadding, J. G. 1983. Definition and clinical categories of asthma. In T. J. H. Clarck, and S. Godfrey, editors. Asthma, 2nd ed. Chapman and Hall, Cambridge, MA. 1–11. |
| 24. | Gibson P. G., Hargreave F. E., Girgis-Gabardo A., Morris M., Denburg J. A., Dolovich J.Chronic cough with eosinophilic bronchitis and examination for variable airflow obstruction and response to corticosteroid. Clin. Exp. Allergy251995127132 |
| 25. | Kerstjens H. A., Brand P. L., Quanjer P. H., van der Bruggen-Bogaarts B. A., Koeter G. H., Postma D. S.Variability of bronchodilator response and effects of inhaled corticosteroid treatment in obstructive airway disease: Dutch CNSLD Study group. Thorax481993722729 |
| 26. | Elliasson O., Hofmann J., Trueb D., Frederick D., McCormick J. R.Corticosteroids in COPD: a clinical trial and reassessment of the literature. Chest891986484490 |
| 27. | Mendella L. A., Manfreda J., Warren C. P. W., Anthonisen N. R.Steroid response in stable chronic obstructive pulmonary disease. Ann. Intern. Med.9619821721 |
| 28. | James A. J., Finucane K. E., Ryan G., Musk A. W.Bronchial responsiveness, lung mechanics, gas transfer, and corticosteroid response in patients with chronic airflow obstruction. Thorax431988916922 |
| 29. | Saetta M., Di Stefano A., Maestrelli P., Turato G., Mapp C. E., Pieno M., Ruggieri M. P., Zanguochi G., Del Prete G., Fabbri L. M.Airway eosinophilia and expression of interleukin-5 in asthma and in exacerbations of chronic bronchitis. Clin. Exp. Allergy261996766774 |
| 30. | Guyatt G. H., Townsend M., Keller J., Singer J., Nogradi S.Measuring functional status in chronic lung disease: conclusions from a randomized control trial. Respir. Med.831989293297 |
| 31. | Keatings V. M., Jatakanon A., Worsdell Y. M., Barnes P. J.Effects of inhaled and oral glucocorticosteoids on inflammatory indices in asthma and COPD. Am. J. Respir. Crit. Care Med.1551997542548 |
| 32. | Thompson A. B., Rennard S. I.Assessment of airways inflammation in chronic bronchitis. Eur. Respir. J.61993461464 |
| 33. | Keatings V. M., Barnes P. J.Granulocyte activation markers in induced sputum: comparison between chronic obstructive pulmonary disease, asthma and normal subjects. Am. J. Respir. Crit. Care Med.1551997449453 |
| 34. | Pizzichini M. M. M., Pizzichini E., Efthimiadis A., Chauhan A. J., Johnston S. L., Hussack P., Mahony J., Dolovich J., Hargreave F. E.Asthma and natural colds: inflammatory indices in induced sputum: a feasibility study. Am. J. Respir. Crit. Care Med.158199811781184 |
| 35. | Pepys, J. 1975. Skin test in diagnosis. In P. G. H. Gell, R. R. A. Coombs, and P. J. Lachmann, editors. Clinical Aspects of Immunology, 3rd ed. Blackwell Scientific Publications, Oxford. 55–80. |
Drs. Emilio Pizzichini and Marcia Pizzichini were supported by fellowships from Boehringer Ingleheim (Canada) Ltd. and C.A.P.E.S., Ministry of Education, Brazil.
