| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Guest Access | Sign In via User Name/Password
|
|
|
|||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Department of Pneumonology and Clinical Research Unit (Drs. Loukides, Papatheodorou, and Panagou), Athens Army General Hospital, Athens, Greece; and the Department of Pneumonology (Drs. Bouros and Siafakas), Medical School, University of Crete, Heraklion, Greece.
Correspondence to: Stelios Loukides, MD, Smolika 2, 16673, Athens, Greece; e-mail ssat@hol.gr
 |
Abstract |
|---|
 TOP Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Setting: Inpatient respiratory unit and outpatient clinic in tertiary-care hospital.
Patients: Fifty stable asthmatic patients with disease severity ranging from mild to moderate.
Methods: H2O2 was measured in expired breath condensate and was correlated with variables expressing both asthma severity (ie, FEV1 percent predicted, peak expiratory flow rate [PEFR] variability, symptom score, and histamine airways responsiveness) and airway inflammation (ie, differential cell counts from induced sputum and levels of eosinophil cationic protein [ECP]).
Results: The mean (95% confidence interval [CI]) concentration of H2O2 was significantly elevated in patients with asthma compared to that in control subjects (mean, 0.67 µM [95% CI, 0.56 to 0.77 µM] vs 0.2 µM [95% CI, 0.16 to 0.24 µM]; p < 0.0001). The difference was primarily due to the elevation of H2O2 in patients with moderate asthma whose expired breath H2O2 level of 0.95 µM (95% CI, 0.76 to 1.12 µM) was significantly higher from that of patients with mild-persistent and mild-intermittent asthma (mean, 0.59 µM [95% CI, 0.47 to 0.7 µM] and 0.27 [95% CI, 0.23 to 0.32 µM], respectively; p < 0.0001). H2O2 concentration was positively related to sputum eosinophilia as well as to ECP concentration. A similar correlation was found between H2O2 and neutrophils in patients with moderate asthma. A positive correlation was observed between H2O2 level, symptom score, and PEFR variability. H2O2 level was negatively related to FEV1 percent predicted. Further analysis showed that only patients with moderate asthma who were not receiving inhaled steroids were found to have a strong relationship with the variables tested.
Conclusions: Eosinophils are the predominate cells that generate H2O2 in all forms of the disease, while neutrophils might be responsible for the highest levels that are observed in the more severe forms of the disease. The role of H2O2 concentration in predicting the severity of the disease as well as in the inflammatory process is limited and depends on the use of inhaled steroid therapy and the classification of the severity of the disease.
Key Words: airway inflammation • hydrogen peroxide • severity • stable asthma
|
Introduction |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Oxidative stress plays an important pathogenetic role in many inflammatory diseases including asthma.3 4 The activation of inflammatory cells such as neutrophils, eosinophils, and macrophages is seen in patients with asthma as well as in those with other inflammatory airway diseases. Activated inflammatory cells respond with a respiratory burst, which results in the production of reactive oxygen species (ROS) such as hydrogen peroxide (H2O2).5 H2O2 reflects the underlying state of oxidative stress and has been measured in exhaled breath condensate, a simple noninvasive technique that is highly reproducible,6 which reflects abnormalities that are noted in specimens obtained bronchoscopically7 and in the sputum.8
However, a number of important issues regarding the value of ascertaining H2O2 levels in asthma patients remain unresolved, the main one being whether the oxidative stress seen in stable patients with asthma contributes to airway inflammation as well as to the classification of the severity. Our primary aim was to evaluate whether there is an association between H2O2 concentration in expired breath condensate and airway inflammation or disease severity. Furthermore, we investigated which cells are the main source of H2O2 production in stable patients with asthma and whether these cells differ according to the severity classification as well as according to the use of inhaled corticosteroids (ICS). As indexes of airway inflammation in sputum, we studied validated variables, including differential cell counts and eosinophil cationic protein (ECP) levels. Disease severity was assessed by a symptom score, the results of lung function tests, and airway responsiveness to histamine. To account for the possible confounding effects of ICS, further analysis was conducted after subdividing the patients on this basis.
|
Materials and Methods |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
The Ethics Committee of Athens Army General Hospital approved the study protocol, and all participants gave informed written consent.
Assessment of Disease Severity
On enrollment into the study, all patients with mild asthma
underwent histamine challenge (ie,
PD20, 0.273 mg; range, 0.05 to 0.61 mg). All the
patients with asthma were instructed to record accurately for 2 weeks
their daily asthma symptoms (ie, cough, daytime wheezing,
nighttime asthma, and daytime breathlessness), bronchodilator
requirements, and peak expiratory flow rate (PEFR) measurements twice
daily (morning and evening).9
Simultaneously, they had to
comply with the terms of receiving their regular medication. The
foresaid asthma symptoms were scored by the patients every day using a
subjective scoring system of 0 to 3 (0, none; 1, mild; 2, moderate; 3,
severe). The symptoms were added for each day, and a mean daily score
was extracted for the period of assessment. No attempt was made to
modify their treatment, which remained unchanged for 4 weeks before
sputum induction and H2O2
collection. There were no differences in lung function (ie,
FEV1 and PEFR) between days 1 and 14 of the
period during which disease severity was assessed.
Classification of Asthmatic Subjects According to Asthma Severity
We used the recommendations for the classification of asthma
severity of the NHLBI workshop on the global strategy for
asthma1
to classify patients into the following three
asthma categories: mild intermittent; mild persistent; and moderate.
Ten patients were classified as having mild-intermittent asthma, 20
patients were classified as having mild-persistent asthma, and the
remaining 20 patients were classified as having moderate asthma (Table 1) .
Analysis for the Effects of ICS
In a subanalysis performed to account for the use of ICS, we
excluded patients with mild-intermittent asthma. The remaining 40
patients were divided into the following two subgroups: those who
received ICS (n = 20); and those who were steroid-naive (n = 20).
Patients with mild-persistent asthma and moderate asthma also were
subdivided according to the use of ICS (10 patients in each subgroup
received ICS).
Lung Function
FEV1 was measured using a dry spirometer
(Vica-test; Mijnhardt; Rotterdam, Netherlands). The best value
of three maneuvers was expressed as a percentage of the predicted
value. Airway responsiveness was measured by histamine provocation
challenge10
in healthy subjects and in patients with mild
asthma. The PD20 was calculated by the
interpolation of the semi-logarithmic dose-response curve. The diurnal
PEFR variation was calculated by the formula (evening and morning
PEFR/the mean of the evening and morning PEFR) x 100, and the
results are shown as the mean daily variation.
Collection of Expired Breath Condensate and
H2O2 Measurement
Expired breath condensate was collected in the morning using an
alternative way of cooling the tubes in order to enhance the formation
of condensate. A heat-exchanger unit (RHES; Jaeger; Wuerzburg,
Germany) was used to produce cold air of -15 to -18°C at an airflow
of 80 L/min. A double-jacketed glass tube of 30 cm length was
specifically adapted to the cold air system and a two-way
unidirectional valve (Hans Rudolf; Kansas City, MO) was connected to
the tube in order to separate inspiration from expiration. After
rinsing their mouths, study participants were seated comfortably in a
chair wearing nose clips and breathed in a relaxed manner
(ie, tidal breathing) for 10 min. The breath condensate was
collected at the other end of the tube and was immediately stored at
-70°C for later analysis. According to this design, salivary
contamination was highly unlikely and was easily observed, as the
proximal cold air connection was 20 cm from the mouthpiece.
Approximately 1 mL breath condensate was collected in a 2-mL sterile
plastic tube. H2O2
measurements were performed on the same day. To examine the
repeatability of H2O2
measurements within subjects, the condensate was collected from 5
healthy subjects and from 10 patients on 2 consecutive days. To assess
the stability of H2O2 in
the frozen condensate, 4 mL condensate was collected from eight
subjects (four patients). The above concentration was divided into 1-mL
aliquots, in which H2O2
concentrations were determined after 2 days, 1 week, 2 weeks, and 3
weeks of storage (which was the maximum time between collection and
measurement in the whole samples). The repeatability of exhaled
H2O2 measurements and the
stability of the H2O2
frozen samples was estimated as previously described.6
All condensate samples were tested for salivary contamination by the
determination of amylase activity. Amylase activity was carried out
spectophotometrically (kinetic method) using a commercial reagent kit
(KONE Instruments; Espoo; Finland). In this procedure, the
-amylase of the sample and the enzyme -glucosidase
hydrolyzes the substrate
p-nitrophenyl--D-maltoheptaoside to
glucose and p-nitrophenol. The liberation of
p-nitrophenol is followed at 405 nm (37°C) for 2 min. Two
samples were spiked with saliva to ensure that this could be detectable
by our method of detection. In all the samples, no amylase was detected
using the method described, suggesting that no contamination of breath
condensate with saliva had occurred. The samples, which were spiked
with saliva, showed levels of > 5000 IU salivary amylase.
H2O2 concentration was determined by an enzymatic assay, as previously described.11 12 Briefly, 250 µL 420 µM 3',3,5,5'-tetramethylbenzidine (dissolved in 0.42 M citrate buffer; pH 3.8) and 10 µL 52.5 U/mL horseradish peroxidase (Sigma Chemical Co; St Louis, MO) were reacted with 250 µL condensate for 20 min at room temperature. Subsequently, the mixture was acidified to a pH of 1 with 10 µL 18 N sulfuric acid. The reaction product was quantitated at an absorbency of 450 nm using a double-beam spectrophotometer (Uvicon 940; Kontron Instruments; Zurich, Switzerland). A standard curve was performed for the assay with a limit of determination of 0.1 µM H2O2.
Sputum Induction and Processing
Sputum was induced by the inhalation of an aerosol of 3.5%
hypertonic saline solution that was generated by an ultrasonic
nebulizer (model 2696; DeVilbiss; Somerset PA)13
that was
used with modifications to improve its safety.14
At least
2 mL sputum was collected into a sterile container. The sample from the
first cough was discarded as it was heavily contaminated with squamous
epithelial cells. An adequate sample was defined if the number of
squamous epithelial cells was < 30% of the total number of
inflammatory cells.15
Cytospin slides were prepared
and stained with May-Grunwald-Giemsa stain. The person who performed
the differential cell counts was not aware of the clinical and
functional status of the patients or of the expired breath condensate
measurements. Two slides were used for counting, and at least 400
inflammatory cells were counted for each slide. The ECP was analyzed as
previously described14
using a fluorometric enzyme
immunoassay (Pharmacia; Uppsala, Sweden). The inflammatory cells in
sputum are shown as both absolute counts (ie, number of
cells per gram of sputum) and as percentages of the total number of
nonsquamous cells. Condensate and sputa were obtained after the period
during which the disease severity was assessed. Sputum measurements
were performed on the same day for all the patients.
Statistical Analysis
Data concerning the characteristics of participants are shown as
the mean (SD; range). Data concerning the comparisons among the various
parameters in the study groups are given as the mean with 95%
confidence intervals (CIs). The data were examined for normal
distributions, and when the data were not normally distributed the
nonparametric Mann-Whitney test was used for statistical comparisons.
For normally distributed data, paired t tests were used for
statistical comparisons. The normality of distribution was tested with
the Shapiro Wilks test. Data from the whole study group of asthma
patients were not normally distributed, while the data from subgroups
of asthma patients were normally distributed. Parameters from patients
with mild-intermittent, mild-persistent, and moderate asthma were
compared using the one-way analysis of variance with an appropriate
post hoc test (ie, Bonferroni correction) for
multiple comparisons. Pearson’s correlation coefficient was used to
investigate the relation between the normally distributed parameters.
For non-normally distributed parameters, Spearman’s correlation
coefficient was used. The codependence of
H2O2 production both from
neutrophils and eosinophils in patients with moderate asthma was
estimated by multilinear regression analysis (SPSS, version 10.0
Windows; SPSS; Chicago, IL). As a dependent variable, we used the
H2O2 concentration in
expired breath condensate, and as independent variable we used the
percentage of eosinophils and neutrophils in induced sputum samples. A
p value < 0.05 was considered to be significant.
|
Results |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The mean H2O2 concentration in expired breath condensate of all asthmatic subjects was significantly higher than that from healthy control subjects (0.67 µM [95% CI, 0.56 to 0.77 µM] vs 0.2 µM [95% CI, 0.16 to 0.24 µM], respectively; p < 0.0001) [Fig 1 , top]. The difference was primarily due to the elevation of H2O2 concentrations in patients with moderate asthma whose expired breath H2O2 level was significantly higher than that of those with mild-intermittent and mild-persistent asthma (0.94 µM [95% CI, 0.76 to 1.12 µM], 0.27 µM [95% CI, 0.23 to 0.32 µM], and 0.59 µM [95% CI, 0.47 to 0.7 µM], respectively; p < 0.0001) [Fig 1 , top]. Patients with mild-intermittent, mild-persistent, and moderate asthma had significantly higher levels of H2O2 compared with those of healthy subjects (p = 0.04, p < 0.001, and p < 0.0001, respectively). There was no significant difference between patients with atopy and those with no atopy (0.7 [95% CI, 0.6 to 0.8] vs 0.6 [95% CI, 0.4 to 0.8], respectively; p = 0.08). All four groups of asthma patients had higher levels of sputum eosinophilia than did control subjects, with a progressive increase that was related to asthma severity (control subjects, 0.3% [95% CI, 0.1 to 0.6%] and 7 x 103/g [95% CI, 2 to 8 x 103/g]; all asthma patients, 7% [95% CI, 2 to 9%] and 98 x 103/g [95% CI, 71 to 114 x 103/g]; patients with mild-intermittent asthma, 2% [95% CI, 0.8 to 3%] and 21 x 103/g [95% CI, 14 to 27 x 103/g]; patients with mild-persistent asthma, 5% [95% CI, 2 to 7%] and 74 x 103/g [95% CI, 63 to 81 x 103/g]; and patients with moderate asthma, 11.5% [95% CI, 4 to 13%] and 164 x 103/g [95% CI, 137 to 198 x 103/g]) [Fig 1 , middle; Table 1 ; p < 0.001 and significantly higher in patients with moderate asthma]. The percentage and number of macrophages were significantly lower in patients with moderate asthma compared both to those with mild-intermittent asthma and mild-persistent asthma (moderate asthma, 57% [95% CI, 51 to 60%] and 518 x 103g [95% CI, 435 to 562 x 103g]; mild-intermittent asthma, 72% [95% CI, 67 to 74%] and 737 x 103g [95% CI, 637 to 816 x 103g]; and mild-persistent asthma, 70% [95% CI, 69 to 75%] and 621 x 103g [95% CI, 577 to 703 x 103g]) [Fig 1 , middle; p < 0.0001]. Furthermore, there were significant differences in the relative neutrophil counts between patients with moderate asthma and patients both with mild-intermittent and mild-persistent asthma (moderate asthma, 31% [95% CI, 28 to 35%] and 255 x 103g [95% CI, 211 to 273 x 103g]; mild-intermittent asthma, 25% [95% CI, 20 to 26%] and 224 x 103g [95% CI, 195 to 237 x 103g]; and mild-persistent asthma, 24% [95% CI, 21 to 25%] and 201 x 103g [95% CI, 189 to 220 x 103g]) [Fig 1 , middle; p < 0.05]. Significant differences among the groups were detected for the ECP concentration with a progressive increase, which was related to asthma severity (Table 1 ; p < 0.001 significantly higher concentration in patients with moderate asthma).
|
Associations Among H2O2 Concentration,
Inflammatory Indexes in Induced Sputum, Lung Function, and Severity
Scores in Subjects With Asthma
In all asthmatic subjects, there was a positive correlation
between H2O2 levels and
both ECP concentrations and eosinophil counts. (Table 2
). Similar results were observed between
H2O2 levels and both
symptom score and PEFR variability (Table 2)
.
H2O2 was inversely
correlated with FEV1 percent predicted (Table 2) .
H2O2 levels did not
correlate with macrophage and neutrophil counts as well as with the
PD20 (see Table 2
for the sputum results;
PD20, r = 0.02 and p = 077). The
above correlations did not remain significant after dividing the
patients on the basis of disease severity and ICS use. They were
present only in patients with moderate asthma (Fig 2
, top) and in patients not treated with ICS (Table 2)
.
Furthermore, in patients with moderate asthma a positive correlation
between neutrophils counts and
H2O2 levels
(r = 0.6; p = 0.005) was observed (Table 2
; Fig 2
,
bottom). When we further analyzed the patients with
mild-persistent asthma and moderate asthma on the basis of ICS use, the
above significant correlations existed only in those with moderate
asthma who had not been treated with ICS (Table 3
). However, the patients with moderate asthma who had been treated with
ICS showed positive correlations both with sputum neutrophilia and
eosinophilia (Table 3)
. A statistically significant correlation also
was found when we analyzed the codependence of
H2O2 production from both
eosinophils and neutrophils in patients with moderate asthma
(eosinophils, adjusted R2 = 0.72;
p = 0.04; neutrophils, p = 0.014).
|
|
|
|
Discussion |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Oxidative stress describes an imbalance between ROS and antioxidants. The normal production of oxidants is counteracted by several antioxidative mechanisms in the human respiratory tract. The major intracellular antioxidants in the airways are catalase, superoxide dismutase, and glutathione, which are formed by the selenium-dependent enzyme glutathione peroxidase. An increased concentration of exhaled H2O2 may represent an increased production of oxidants and/or a reduced free radical-scavenging capacity in the airways of asthmatic subjects. In addition to the increased production of ROS in patients with asthma, there may be a deficiency of antioxidant defenses. Glutathione peroxidase activity is reduced in the platelets of asthmatic patients and is correlated with the reduction in serum selenium concentrations.16 There is a reduced superoxide dismutase activity in both cells obtained by BAL and in the epithelial cells of asthmatic patients.17 Interestingly, there is no abnormality in antioxidant levels in patients with mild asthma, and specifically in those whose disease is being controlled with ICS. Our results showed that oxidative stress was not increased in patients with mild-intermittent asthma, irrespective of their use of ICS, and is significantly affected by the use of ICS in patients with all forms of the disease. Taking all these points together, we interpret our findings as following: in patients with mild asthma, antioxidant defenses are adequate to overcome the increased production of ROS. However, as the severity of the disease progresses the production of ROS overcomes the antioxidant defenses. In patients with more severe forms of the disease, in which the increased production of ROS overcomes the antioxidant defenses, therapy with ICS significantly affect the imbalance between oxidants and antioxidants by decreasing the ROS production through the control of the inflammatory process.
Many inflammatory and structural cells that are activated in the airways of asthma patients result in the marked production of ROS, including H2O2.5 However, the precise role of ROS in the pathophysiology of the inflammatory process remains unclear due to the difficulty in the identification of which cells produce ROS in the lungs and bronchi of asthmatic subjects as well as to the difficulty in specifying the differences in these cells among the various states of the disease. One conclusion of the studies conducted so far is that eosinophilic inflammation is a characteristic feature of intermittent and mild asthma, whereas the association of eosinophilia and neutrophilia characterizes more severe forms of the disease.9 Moreover, sputum neutrophil counts were particularly evident in asthma patients who remained symptomatic despite treatment with ICS. Sputum induction is considered to be a noninvasive means to monitor inflammation. The value of the measurement of inflammatory cells and mediators such as ECP in induced sputum from asthma patients has been highlighted in several studies.2 9 14 The present evidence suggests that the number of sputum eosinophils may possibly be the most useful parameter that directly reflects the intensity of the inflammation measured in bronchial biopsy specimens, lung function impairment, and, finally, disease severity.18 19 In the present study, we found a significant correlation among eosinophil counts in induced sputum, disease severity, lung function impairment, and H2O2 concentration in expired breath condensate. Further analysis showed that the above correlations existed only in patients with moderate asthma in whom a strong correlation between the number of neutrophils and H2O2 levels also was observed. Taking all these points together, and based on the above data from the literature, we believe that eosinophils are the predominant cells that generate ROS in patients with all forms of the disease, while neutrophils might be responsible for the highest levels that are observed in the more severe forms of the disease. These findings are supported mainly by the observations of our study, which showed a statistically significant correlation of the codependence of H2O2 production from both eosinophils and neutrophils in patients with moderate asthma. Additionally, they are partially supported by previous observations which demonstrated that eosinophils are the predominant cells that produce H2O220 21 in asthma patients, while neutrophils from asthma patients released higher quantities of ROS in comparison to healthy control subjects.22 Moreover, the accumulation of neutrophils in the airways is associated with the extensive release of several molecules, particularly proteases and oxidants.23 However, in mild forms of the disease active treatment with ICS leads to low levels of H2O2 by suppressing the eosinophilic inflammation and controlling the symptoms. Similar results regarding the correlation data between eosinophils and H2O2 concentrations also were observed in another study with a design similar to that of our study.24 However, the above study classified patients as stable and unstable without using the NHLBI guidelines, which are more acceptable for the classification of disease severity. Moreover, it correlated H2O2 only with eosinophils and did not provide data for the other inflammatory cells (ie, macrophages and neutrophils) as well as mediators such as ECP concentration. Finally, our study was more detailed regarding the parameters used for asthma evaluation (ie, symptom score and PEFR variability) offering new information about the correlation of H2O2 with both clinical severity and the inflammatory process.
The present cross-sectional study cannot demonstrate a causal relationship between ICS use and H2O2 concentration. In a prospective controlled study,25 therapy with ICS reduced the levels of H2O2 in the expired breath condensate over a 4-week period, indicating an anti-inflammatory effect on oxidative stress. Data from a previous study24 also have suggested that the differences in exhaled H2O2 levels between steroid-treated and steroid-naïve patients imply differences in oxidative stress. However, other studies26 support the idea that steroids do not decrease H2O2 release from eosinophils. The current evidence suggests that there are significant quantitative differences in the number of eosinophils in sputum in relation to disease severity, directly reflecting the magnitude of airway inflammation.9 Moreover, sputum eosinophils may be the most useful parameter directly reflecting airway inflammation that can be directed, modulated, and controlled by ICS.27 Since the available evidence suggests that therapy with corticosteroids is considerably more effective when there is eosinophilic inflammation, this should logically lead to the plausible explanation that the differences in exhaled H2O2 levels between steroid-treated and steroid-naive patients did not imply differences in oxidative stress but reflected differences in the eosinophilic inflammation of the disease. The above theory is partially supported by the finding that the correlation between H2O2 levels and both sputum eosinophilia and ECP concentration seems to exist only in steroid-naive patients, as well as by the strong correlation between neutrophil and H2O2 levels, irrespective of the use of ICS. These observations as well as those of previous reports that demonstrated that steroids failed to alter H2O2 production from polymorphonuclear leukocytes,28 confirm the initial theory that therapy with ICS influences H2O2 production through their inhibitory role in eosinophilic recruitment to the airways.
The results of this study indicate that the role of exhaled H2O2, an index of oxidative stress, in predicting the severity of the disease as well as the course of the inflammatory process is limited. It depends on the use of inhaled steroids and the classification of the severity of the disease. Prospective controlled studies are needed to confirm the effects of anti-inflammatory drugs and oxygen radical scavengers on both the airway inflammatory indexes and ROS, in order to apply the technique of expired breath condensate in the evaluation of inflammatory pulmonary disorders.
|
Footnotes |
|---|
Received for publication December 20, 2000. Accepted for publication September 5, 2001.
|
References |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  R. P. Bowler and J. D. Crapo Oxidative Stress in Airways: Is There a Role for Extracellular Superoxide Dismutase? Am. J. Respir. Crit. Care Med., December 15, 2002; 166(12): S38 - 43. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
, patients with mild-intermittent
asthma [n = 10]; 
, patients with mild-persistent asthma
[n = 20]; 
, patients with moderate asthma [n = 20]; 
,
healthy subjects [normals] [n = 15]). Each symbol represents one
individual. The concentration of H2O2 was
significantly higher in patients with asthma compared to healthy
subjects (p < 0.0001). Patients with moderate asthma had significant
higher values compared to those with mild-intermittent and
mild-persistent asthma (p < 0.0001). The horizontal bar
represents mean values. Middle: the percentage and
number of differential cell counts in the induced sputum of patients
with asthma (ie, all patients, patients with
mild-intermittent asthma, patients with mild-persistent asthma, and
patients with moderate asthma) and in healthy subjects. * = a
statistically significant higher percentage of neutrophils in the
induced sputum of patients with moderate asthma compared to healthy
subjects and patients in the other asthma subgroups (p < 0.05).
** = a statistically significant higher percentage of
eosinophils in the induced sputum of patients with moderate asthma
compared to healthy subjects and patients in the other asthma subgroups
(p < 0.001). *** = a significantly lower percentage of
macrophages in the induced sputum of patients with
moderate asthma compared to healthy subjects and patients in the other
asthma subgroups (p < 0.0001). Bottom:
H2O2 concentration in the expired breath
condensate of patients with asthma who were treated with ICS (+; 
,
all patients [n = 20]; 
, all patients [n = 20];
