September - December 2002: 
Volume 15, Issue 3

Click on the image to download the Issue in PDF format.


Hydrogen peroxide in expired breath condensate: relationship with airway inflammation and asthma severity
Objective: To investigate which cells are the main source of hydrogen peroxide (H2O2) production in stable asthma and the possible relationship between H2O2 levels on the one hand, and airway inflammation and disease severity on the other. Material and methods: Fifty stable asthmatics with disease severity ranging from mild to moderate were studied. H2O2 was measured in expired breath condensate and its relationship with variables indicating asthma severity [e.g. FEV1% pred, peak expiratory flow rate (PEFR) variability, symptoms score and airway responsiveness to histamine] or airway inflammation [differential cell counts in induced sputum, levels of eosinophil cationic protein (ECP) in sputum supernatant] was assessed. Results: Our results showed that mean (95% CI) concentrations of H2O2 were significantly elevated in patients with asthma compared to the values obtained from control subjects (.67, .56 .77 vs .2, .16 .24 mM, respectively, P<.0001). The difference was primarily due to the significantly increased H2O2 levels observed in moderate asthma patients as compared to those observed in patients with mild persistent or mild intermittent asthma (.95, .76 1.12 μM vs .59, .47 .7, vs .27, .23 .32 μM, respectively, P<.0001). H2O2 levels were positively related to sputum eosinophilia, as well as to ECP concentrations. A positive relationship was also found between H2O2 levels and neutrophil counts in patients with moderate asthma. H2O2 levels were also associated with symptoms score and PEFR variability. There was an inverse relationship between H2O2 levels and FEV1% pred. Further analysis showed that the relationship between H2O2 levels and the examined variables was significant only in moderate asthma patients not receiving inhaled steroids. Conclusions: Eosinophils are the main H2O2 producing cells in asthma of varying severity, whereas neutrophils might also contribute to the much higher H2O2 levels observed in the more severe cases. The predictive value of H2O2 levels as regards disease severity and inflammatory activity is limited and depends on the use of inhaled steroids and the severity of the disease itself. Pneumon 2002, 15(3):263-274.
Full text


Asthma is a chronic inflammatory disease of the airways characterized by reversible bronchial obstruction and increased bronchial responsiveness.1 Disease severity assessment is based on various variables, including the presence and frequency of symptoms, medical treatment, results of pulmonary function tests. In the recent years the method of sputum induction has been widely used for the assessment of the inflammatory process.2 This methodology facilitates the analysis of both the inflammatory cells and mediators implicated in the pathogenesis of asthma.

Oxidative stress has a key role in the pathogenesis of a variety of respiratory diseases, including asthma.3,4 It is also known that another characteristic asthma feature is the activation of a variety of inflammatory cells, including eosinophils, macrophages and neutrophils, leading to the production of oxidants such as hydrogen peroxide (H2O2).5 H2O2 levels reflect the severity of oxidative stress and are determined in expired breath condensate. H2O2 measurements in expired breath condensate provide a non-invasive and highly reproducible6 method for assessing the extent and severity of airway inflammatory damage; such assessments have been confirmed by examination of bronchoscopic material and sputum analysis.8 

Still, many questions regarding the cellular origin of H2O2 and, mainly, its value in the assessment of the activity and the severity of the disease, which would be supported by its correlation with established indices of inflammation as well as with clinical symptoms and functional deficits, remain unanswered. The purposes of the present study are to demonstrate the cellular origin of H2O2 in asthma patients and examine the possible correlation of H2O2 levels in expired breath condensate with established indicators of inflammation, clinical severity and functional deficits in asthma. Furthermore, we examined whether these correlations were affected by either the use of inhaled steroids or the clinical stage of the disease. Total and percent (%) cell counts in induced sputum along with eosinophil cationic protein (ECP) in sputum supernatant were used as indicators of the inflammatory activity. Clinical severity was assessed by a symptom scoring scale, whereas pulmonary function tests included standard spirometry, determinations of peak expiratory flow rate (PEFR) variability and bronchial provocation testing with histamine to assess bronchial hyperresponsiveness.

Material and methods


Fifty asthma patients (39 males, mean age 29, SD 6, range 19‑43 years, mean FEV1 82, SD 16, range 49‑110% pred) were studied (Table 1). Thirty two of these patients were atopic as determined by positive skin prick test results in at least one of six common allergens. Asthma diagnosis was established using the criteria of the American Heart and Lung Institute.1 Patients were included in the study only if they were clinically stable and had not suffered an exacerbation of the disease due to any possible cause for at least 4 weeks prior to their enrollment. Twenty patients were on inhaled steroids (propionic fluticasone 500‑1000 μg daily or budesonide 400‑800 μg daily). None of the patients received additional anti-inflammatory drugs, including leukotriene antagonists and theophyllin. Furthermore, none of the patients received mucolutic drugs or supplemental oxygen therapy. Twenty-two patients were using long-acting β2-agonists twice daily (some of these patients also used short-acting β2-agonists occasionally for symptom relief); the remaining twenty-eight patients used long- or short-acting β2-agonists only for symptom relief.

Asthma patients were classified into three severity categories following the criteria of the American Heart and Lung Institute.1 We identified 10 cases of mild intermittent, 20 cases of mild persistent and 20 cases of moderate asthma.

Fifteen normal subjects (11 males, mean age 33, SD 7, range 19‑37 years, Table 1) were also included in the study as control group. They all were non-smokers and had no upper or lower airway infection for at least six weeks prior to enrollment. None of the control subjects had a history of chronic disease or received long-term medical treatment; none of the control subjects had a history of atopy and skin prick tests for six common allergens were negative. All the control subjects had normal spirometry values (FEV1 95, SD 3, range 87‑121% pred) and nonreactive airways in bronchial provocation test with histamine (PD20 >.800, mean value 1.45, range .95‑1.9 mg).


All study subjects underwent expired breath condensate collection for H2O2 measurements, sputum induction and collection, as well as pulmonary function testing by standard spirometry. The collection of both expired breath condensate and sputum, and spirometry testing were performed on the same day for each individual patient. H2O2 measurements, as well as induced sputum cell content analysis were performed blind to the clinical and functional status of the patients.

H2O2 levels were determined and compared with the values obtained from the control group. Comprehensive comparisons between subgroups of patients, as defined by the relevant international classification guidelines regarding clinical severity, and controls followed. The effect of inhaled steroids on H2O2 levels in each subgroup of patients was recorded. H2O2 levels correlated with variables reflecting clinical and functional severity as well as with indices of inflammatory activity, including cell content of induced sputum and ECP concentrations in sputum supernatant.

Expired breath condensate collection and hydrogen peroxide measurements

Expired breath condensate was collected during morning hours using an alternative method for cooling the collection tubes. The device that is normally intended for cold air provocation tests was set to produce air at a temperature of -15 to -18 oC with an output of 80 liters per minute. The one end of the 30 cm long double-lumen glass tube was connected to the cold air production port, whereas the study subjects exhaled through the other end of the tube. On the top of the tube there was a two-way valve (Heins Rudolf) that permitted the distinction between the inspiratory and expiratory phase of a respiration cycle. Before starting the expired breath condensate collection procedure, all subjects rinsed their mouth and pharynx with water. During the procedure, the subjects were in sitting position and breathed steady and calmly for 10 minutes. Expired breath condensate was collected at the end of the tube connected to the cold air machine ("outlet"), and was stored at -70 oC for further processing. The above-described method provides at least 1 ml of condensate in a 3 ml plastic tube. Determinations of H2O2 in all condensates were performed on the collection day and then again within 20 days from collection. Repeat measurements were performed at four time points (minimum 2 days and maximum 20 days from collection). All samples were examined for the presence of saliva using the amylase detection method. In this method, amylase is detected by spectrophotometry using a reagent manufactured by KONE Instr. Finland; if present in the sample, α-amylase together with α-glycosidase contained in the reagent hydrolyze the substrate p-nitrophenyl-α-δ-maltoheptaside to glucose and p-nitrophenyl. The absorbance of p-nitrophenyl is then measured at 405 nm (37 oC) for 2 minutes. Saliva was added in two samples in order to examine the sensitivity of the method. Amylase could not be detected in any of the study samples using the above-described method. However, when saliva was added, amylase was detected at a concentration of >5000 IU. The repeatability of H2O2 measurements was examined with repeat measurements on two separate days in 5 normal subjects and 10 patients, both randomly selected. The stability of the samples was examined in 10 subjects (5 patients); 4 ml samples were obtained for this purpose. This amount was divided in four equal portions, which were used for measuring H2O2 on days 2, 7, 14 and 21, the latter being the maximum time elapsed between condensate collection and H2O2 measurement for all samples. The repeatability of H2O2 measurements and the stability of the samples were assessed as previously described in the literature.6 

Hydrogen peroxide levels were measured using an enzyme assay, as previously reported.9,10 In short, 250 μl of a 420 μM 3΄,3,5,5΄-tetramethylbenzidine solution (in 0.42 M citrate buffer, pH 3.8) and 10 μl of a 52.5 U/ml horseradish peroxidase solution (HRP, Sigma Chemicals, St. Louis, USA) were mixed with 250 μl of expired breath condensate for 20 minutes at room temperature. The mixture underwent oxidation at a pH equal to 1 with 10 μl of a 18 N sulphate solution. The product of this reaction was quantitated at 450 nm with a double beam spectrophotometer (Unicon 940, Kontron Instr.). An automatic analyzer (model EL 312 Biotec) was also used for this measurement. The lower detection limit of the method was 0.1 μM H2O2.

Sputum induction

Sputum was induced as previously reported,11 taking all necessary preventive measures for assuring patient safety. All patients underwent bronchodilatory pretreatment with inhalations of a β2-agonist solution (200‑400 μg salbutamol) vaporized in an air-mask. FEV1 was measured fifteen minutes later. That FEV1 value was used as reference for the measurements that followed during the procedure of sputum induction. If this reference FEV1 value was <60% of predicted or < 1 liter, the procedure started with normal saline inhalations, while gradually reducing the time during which the mixture was inhaled. If FEV1 was further reduced by 10‑19%, an appropriate dose of a bronchodilator was given and the procedure continued once FEV1 returned to baseline. If the additional reduction in FEV1 was greater than 20%, the procedure was withheld and, once again, bronchodilators were given at an appropriate dose. In the majority of the patients, 3.5% hypertonic saline inhalations generated by a DeVilbiss ultrasonic nebulizer (2696 Somerset PA, USA) were used for sputum induction. Both patients and controls were asked to blow their nose and rinse their mouth and pharynx to minimize the possibility of sputum contamination with saliva and post-nasal drip. All patients wore a nose clip during the procedure. They were instructed to cough deeply every 2‑3 minutes, taking of course into account the functional severity of the disease. The first sputum sample was rejected since it is believed to contain a high percentage of epithelial cells.13 At least 2 ml of induced sputum were collected in a plastic tube. Reliable samples were identified as those with an epithelial cell percentage <30% of total inflammatory cell count. Final samples were stained by May-Grόnwald-Giemsa stain and the number of cells was calculated. Two slides were used and at least 400 inflammatory cells were counted. The inflammatory cell counts were expressed as both absolute numbers of cells per gram of sputum and percentages (%) of total non-epithelial cell count. The total cell counts were expressed as multiples of 106 ml-1. H2O2 values were not known at the time of determining the cellular content of induced sputum. ECP levels in the supernatant of centrifuged sputum were measured in bronchial asthma patients using a fluorometric enzyme immunoassay (Pharmacia, Uppsala, Sweden), as reported in the literature.2 All sputum measurements were performed on the same day for all patients in a group.

Pulmonary function testing

FEV1 measurements were performed using a Vica-test, Mignhard, NL spirometer. The best value of three attempts expressed as a percentage of the predicted value was used in the study. Both controls and mild asthmatics underwent bronchial provocation testing with histamine in order to exclude the control subjects who showed bronchial hyperresponsiveness in the former group, and to determine the severity of asthma based on the grade of bronchial hyperresponsiveness on the latter. Provocation testing with histamine was performed using the APS Jaeger system, Wόrzburg, Germany, with a Sandoz Jaeger nebulizer. PD20 represents the provocative concentration of histamine causing a 20% fall in FEV1; a PD20 >.800 mg indicates absence of bronchial hyperresponsiveness. PD20 was calculated by drawing the semilogarithmic dose-response curve.14 PEFR was measured in all asthma patients twice daily (morning and evening) and intraday variability was calculated as follows: [evening value - morning value/mean value of the two measurements] x 100.

Assessment variables

Drug treatment

The effect of inhaled steroids on H2O2 levels as well as on other assessment variables was further investigated in all patients; in this analysis, patients were subdivided in subgroups across the disease severity spectrum.

Clinical severity score was calculated using a well-established scoring system.15 All asthma patients were instructed to record their daily symptoms (cough, daytime wheezing, shortness of breath, and the presence of bedtime symptoms) for a two-week period; symptoms were graded on a 0‑3 scale (0=absence of symptoms, 1=mild symptoms, 2=moderate symptoms, 3=severe symptoms). Scores for all symptoms were added and a mean score for the whole assessment period was calculated. Pulmonary function (FEV1, PEFR) remained stable in the period from day 1 to day 14 (symptom recording period).

Statistical analysis

Subjects' characteristics were expressed as mean values plus or minus standard deviation (SD), with the value range in parentheses. Comparisons between assessment parameters were reported as mean differences with 95% confidence intervals (CI). Data were assessed for normal distribution using the Shapiro Wilk's test; if normally distributed, paired t-test was used for comparisons. The non-parametric Mann Whitney test was used for comparisons between not normally distributed data. The differences in the study variables between patient subgroups were assessed by one-way analysis of variance (ANOVA) with the appropriate corrective post-hoc test (Bonferroni) for multiple comparisons. Significant associations between normally distributed variables were sought by Pearson's correlation coefficient, whereas Spearman's correlation coefficient was calculated for not normally distributed data. Rank correlation was assessed by a multiple correlation statistical model (SPSS 10 for Windows). Significance was defined as a P value of <.05.

Figure 1. A) Hydrogen peroxide (H2O2) levels in expired breath condensate from patients with asthma (total number of patients n=50l, patients with mild intermittent n=10Ñ, mild persistent n=20t, and moderate asthma n=20s) and controls (n=15m). Depicted values correspond to individual study subjects. H2O2 levels were significantly higher in asthma patients compared to controls (p<.0001). Moderate asthmatics had significantly higher H2O2 levels than both mild intermittent and mild persistent asthmatics (p<.0001). Horizontal lines indicate mean values.

B) Percentages (%) of non-squamous cells in induced sputum from patients with asthma (total of patients, patients with mild intermittent, mild persistent and moderate asthma) and controls. A single asterisk indicates significantly higher neutrophil percentage in patients with moderate asthma compared with mild asthmatics and controls (p<.05). A double asterisk indicates significantly higher eosinophil percentages in patients with moderate asthma compared with mild asthmatics and controls (p<.001). A triple asterisk indicates significantly lower macrophage percentages in patients with moderate asthma compared with mild asthmatics and controls (p<.0001). Data are expressed as mean values.

C) Hydrogen peroxide (H2O2) levels in expired breath condensate from asthma patients treated with inhaled steroids (IS) (total number of patients n=20π, patients with mild persistent n=10Ñ, and moderate asthma n=10m) and asthma patients not receiving inhaled steroids (total number of patients n=20n, patients with mild persistent n=10t, and moderate asthma n=10l). Depicted values correspond to individual study subjects. H2O2 levels were significantly lower in asthma patients treated with inhaled steroids compared to those that were not receiving such treatment (p<.05 for the total of patients, p<.001 for patients with mild persistent and p<.0001 for patients with moderate asthma). Horizontal lines indicate mean values.


As regards the precision of the method and the agreement between measurements performed on two different days, the mean difference of the obtained values was .09 (.03) μΜ in the patient group and .07 (.04) μM in the control group. The comparison between H2O2 measurements performed on portions of the same original sample on four different days showed non-significant differences [.54 (.21) μM on day 2, .58 (.3) μM on day 7, .6 (.3) on day 14 and .57 (.4) μM on day 21, P=.43] denoting the stability of H2O2 in frozen samples.

The mean (95% CI) H2O2 concentrations in expired breath condensate were significantly higher in asthmatics than in controls (.67, .56‑.77 μM versus .2, .16‑.24 μM, P<.0001, Figure 1A). This significant difference was primarily due to the remarkably increased H2O2 levels in patients with moderate asthma as compared to those with either mild intermittent or mild persistent asthma (.94, .76‑1.12 μM versus .27, .23‑.32, versus .59, .47‑.7, respectively P<.0001, Figure 1A). However, patients from across the spectrum of asthma severity (mild intermittent, mild persistent and moderate asthma) had significantly higher H2O2 levels compared with controls (P=.04, P<.001 and P<.0001, respectively). There were no significant differences between atopic and non atopic patients (.7, 95% CI .6‑.8 versus .6, 95% CI .4‑.8, P=.08). Eosinophil percentages were higher in all patient groups compared with controls, with an upward trend in relation to increasing disease severity (9.3, 95% CI .1‑.6, 7x103/g, 95% CI 2‑8 in controls, 7%, 95% CI 2‑9 , 98x103/g, 95% CI 71‑114 in all asthmatics as a single group, 2%, 95% CI .8‑3, 21x103/g, 95% CI 14‑27 in mild intermittent asthma, 5%, 95% CI 2‑7, 74x103/g, 95% CI 63‑81 in mild persistent asthma, and 11.5%, 95% CI 4‑13, 164x103/g, 95% CI 137‑198 in moderate asthma patients, Figure 1B, Table 1, P<.001, percentage significantly higher in moderate asthma patients). Macrophage percentages in moderate asthmatics were consistently lower than those in patients with mild intermittent or mild persistent disease (57%, 95% CI 51‑60, 518x103/g, 95% CI 435‑562, 72%, 95% CI 67‑74, 737x103/g, 95% CI 637‑816, and 70%, 95% CI 69‑75, 621x103/g, 95% CI 577‑703, respectively, Figure 1B, P<.0001). Borderline significance was found in neutrophil absolute counts and percentages between patients with moderate asthma and patients with mild intermittent or mild persistent disease (31%, 95% CI 28‑35, 255x103/g, 95% CI 211‑273, 25%, 95% CI 20‑26, 224x103/g, 95% CI 195‑237, and 24%, 95% CI 21‑25, 201x103/g, 95% CI 189‑220, respectively, Figure 1B, P<.05). Significant differences in ECP concentrations in sputum supernatant were also noted, with higher levels recorded in moderate asthmatics (Table 1, P<.001).

Furthermore, there was a significant difference in mean (95% CI) H2O2 concentrations between patients on inhaled steroids and those not receiving such agents (.66, .52‑.8 μM versus .87, .68‑1.05 μΜ, P<.05, Figure 1C). This difference persisted when the subgroups of mild persistent and moderate asthma were separately examined as regards the effect of inhaled steroids on H2O2 concentrations (.65, .4‑.87 μΜ versus .52, .41‑.64 μM, P<.001, and 1.09, .083‑1.34 μΜ versus .74, .46‑1.02 μΜ, P.ooo1, respectively. Figure 1C). Significant lower eosinophil counts in induced sputum and ECP levels in sputum supernatant were found in patients on inhaled steroids (Table 1, P<.001 and P<.0001, respectively).

A significant positive relationship was detected between H2O2 levels, ECP concentrations and eosinophil counts in induced sputum in all asthmatics (Table 2). Similar associations were found between H2O2 levels, clinical severity score and PEFR variability (Table 2). H2O2 levels were inversely related to FEV1% pred (Table 2). A significant relationship between H2O2 levels and induced sputum macrophage and polymorphonuclear leukocyte counts, or between H2O2 levels and grade of bronchial hyperresponsiveness, as expressed by PD20 of histamine, could not be established (Table 2 for cellular content, rs=.02, P=077 for PD20). When patients were divided in subgroups based on disease severity and the use of inhaled steroids, the above-mentioned significant relationships were not maintained, with the exception of moderate asthmatics and, especially, the subgroup of patients who were not receiving inhaled steroids (Table 3). Further analysis of the correlations between cellular content and H2O2 levels revealed a positive relationship between H2O2 levels and polymorphonuclear leukocyte count in moderate asthmatics (r=.6, P=.005, Table 3); eosinophil counts also correlated with H2O2 (Table 3). A significant positive rank correlation between H2O2 levels as a constant variable and induced sputum eosinophil and neutropil counts was observed in moderate asthmatics (r=.64, P=.0002).


In the present study we demonstrated that H2O2 levels in expired breath condensate from patients with asthma are increased, primarily due to the eosinophilic inflammation involved in the disease process, although neutrophils may have an enhanced contribution to H2O2 production in more severe disease. H2O2 levels depend on clinical severity and the use of inhaled steroids. These findings indicate the presence of increased oxidative stress in the airways of asthma patients but, at the same time, limit the value of H2O2 levels in the assessment of the inflammatory process and clinical severity.

Oxidative stress is the result of increased oxidant production, which is coupled with the stimulation of antioxidant defenses. Normally, oxidants produced in the airways are inactivated by a variety of antioxidant sys systems present in the respiratory system; the major intracellular antioxidants include catalase, superoxide dismutase and glutathione. The latter is produced by the selenium-dependent enzyme glutathione peroxidase. Increased H2O2 levels in expired breath condensate from patients with asthma suggest increased oxidant production and/or decreased antioxidant capacity. This has been confirmed by studies documenting decreased antioxidant activity in asthma patients. Reduced glutathione peroxidase activity has been detected in platelets from asthma patients; this reduction appears to be related to serum selenium concentrations.16 Similar findings have been reported regarding superoxide dismutase activity in bronchoalveolar lavage and epithelial cells.17 It should be stressed, however, that the above-described findings were absent in patients with mild asthma, especially in those treated with inhaled steroids. Subsequently, considering the findings of the present study, oxidative stress is less prominent in mild intermittent asthma and is effectively limited by the use of inhaled steroids in mild persistent and moderate asthma. In conjunction with literature data regarding antioxidant activity, our results lead to the plausible conclusion that in mild forms of asthma, oxidants are effectively inactivated by antioxidant systems, whereas in severe asthma increased oxidant production outweighs antioxidant defense. Inhaled steroids enhance the effectiveness of antioxidant systems by suppressing the inflammatory process. 

A variety of structural and inflammatory cells are activated by mediators released in the airways of asthma patients, leading to increased oxidant production.5 Nevertheless, the exact role of oxidants in the pathogenesis of asthma is still unclear, since it has not been possible to determine which cells produce oxidants as well as whether oxidant production differs across the clinical spectrum of disease severity. In many studies, eosinophilic inflammation has been shown to characterize mild asthma, whereas the presence of neutrophils has been considered an indicator of severe asthma.15 In particular, the presence of neutrophils in induced sputum is a feature of symptomatic disease unresponsive to inhaled steroids. Sputum induction is a reliable, non-invasive method for the assessment of airway inflammation; its value in the determinations of inflammatory cell counts and mediators, such as ECP, has been pointed out in many studies.2,12,15 The presence of eosinophils in induced sputum is currently held as the most valuable marker of airway inflammation, since it reflects bronchial biopsy results, decreased pulmonary function parameters and clinical severity scores.18,19 Our findings are in agreement with the above-described literature data, but go one step further emphasizing that H2O2 production primarily occurs in eosinophils, with additional H2O2 amounts produced by neutrophils in more severe forms of the disease. In other words, eosinophils are the main H2O2 producing cells, whereas neutrophils have a significant supplementary contribution in severe asthma. The latter suggestion is strongly supported by the significant rank correlation between eosinophil and neutrophil H2O2 production in moderate asthma. In addition to the above cited views, the results of the present study are consistent with previously reported work on the primary role of eosinophils in H2O2 production in asthma,20,21 with neutrophils being considered the site of production of large amounts of oxidants in a more general sense.22,23 

Findings similar to ours have also been reported by another study with similar design but quite different methodology.24 Differences were mainly to do with severity criteria by which patients were divided in subgroups, since international guidelines for asthma classification were not applied. In addition, rather than assessing the total cell content, only eosinophils were included in cell analyses. Eventually, established indicators of asthma severity (PEFR variability, symptom score) were disregarded.

The above controlled study is incapable of establishing any association between inhaled steroids and H2O2 levels. Only double-blind placebo-controlled clinical trials can provide this kind of evidence. One such clinical trial has shown that H2O2 levels decreased significantly in subjects treated with inhaled steroids for 4 weeks, compared to subjects treated with placebo.25 Our study, as well as a similar previous one,24 also showed that the use of inhaled steroids resulted in a decrease in H2O2 levels. However, another study has concluded that inhaled steroids do not affect H2O2 production by eosinophils.26 Such a finding is in sharp contrast with the generally accepted view that asthma is characterized by an eosinophilic inflammatory process, and the established regulatory effect of inhaled steroids on the activity of eosinophils.27 

In essence, differences in H2O2 levels between patients on inhaled steroids and those not receiving such treatment reflect differences in the underlying eosinophilic inflammation rather than differences in oxidative stress. This is confirmed by the strong correlation between H2O2 levels on the one hand and eosinophil counts or ECP concentrations on the other in patients not receiving inhaled steroids, as well as by the significant association between H2O2 levels and neutrophil counts in induced sputum irrespective of the use of inhaled steroids. This is due to the recognized failure of steroids to inhibit H2O2 production by neutrophils.28 Therefore the evidence suggests that the effect of steroids on H2O2 production in asthmatics is the result of their suppressive effect on the underlying eosinophilic inflammation.

In conclusion, the present study demonstrates that H2O2 levels in expired breath condensate from asthma patients are increased, but are of limited value in the assessment of the inflammatory process and the clinical severity, due to their dependence on the use of inhaled steroids and disease severity itself. However, until further research provides clear evidence, we shall not know whether the therapeutic administration of antioxidants can, by reducing oxidative stress, cause any meaningful clinical and functional improvement in asthma patients.


  1.  National Institutes of Health. National Health, Lung and Blood Institute 1997. Guidelines for the diagnosis and management of asthma. NIH Publication No 97-4051A.

  2.  Pizzichini E, Pizzichini MMM, Efthimiadis A et al. Indices of airway inflammation in induced sputum: reproducibility and validity of cell and fluid phase measurements. Am J Respir Crit Care Med 1996, 154: 308-17.

  3.  Dohlman AW, Black HR, Royall JA. Expired breath hydrogen peroxide is a marker of acute airway inflammation in paediatric patients with asthma. Am Rev Respir Dis 1993, 148: 955-60.

  4.  Montuschi P, Corradi M, Giabattoni G et al. Increased 8-isoprostane a marker of oxidative stress in exhaled condensate of asthma patients. Am J Respir Crit Care Med 1999, 160: 216-20.

  5.  Barnes PJ. Reactive oxygen species and airway inflammation. Free Rad. Biol. &Med 1990, 9: 235-43.

  6.  Jobsis Q, Raatgeep HC, Schellekens SL, et al. Hydrogen peroxide in exhaled air of healthy children: reference values. Eur Respir J 1998, 12: 483-85.

  7.  Arias-Diaz J, Vara E, Torres-Mellero J et al. Nitrite/nitrate levels in bronchoalveolar lavage fluid of lung cancer patients. Cancer 1994, 74: 1546-51.

  8.  Kanazawa H, Shoji S, Yamada M et al. Increased levels of nitric oxide derivatives in induced sputum in patients with asthma. J Allergy Clin Immunol 1997, 99: 624-29.

  9.  Gallati H, Pracht I. Horseradish peroxidase: kinetic studies and optimization of peroxidase activity determination using the substrates H2O2 and 3, 3', 5, 5'-tetramethylbenzidine. J Clin Chem Clin Biochem 1985, 23: 453-60.

10.  Loukides S, Horvath I, Wodehouse T, et al. Elevated levels of expired breath hydrogen peroxide in bronchiectasis. Am J Respir Crit Care Med 1998, 158: 991-94.

11.  Keatings VM, Evans DJ, O'Connor BJ, et al. Cellular profiles in asthmatic airways: a comparison of induced sputum, bronchial washings and bronchoalveolar lavage. Thorax 1997, 52: 372-74.

12.  Pizzichini MMM, Pizzichini E, Clelland L et al. Sputum in severe exacerbations of asthma: kinetics of inflammatory indices after prednisone treatment. Am J Respir Crit Care Med 1997, 155: 1501-08.

13.  Keatings VM, O'Connor BJ, Wright LG, et al. Late response to allergen is associated with increased concentrations on TNF-a and interleukin-5 in induced sputum. J Allergy Clin Immunol 1997, 99: 693-98.

14.  Sterk PJ, Fabbri LM, Quanjer PH et al. Airway responsiveness: Standardized challenge tests with pharmacological, physical and sensitizing stimuli in adults. Eur Respir J (Suppl) 1993, 16: 53-83.

15.  Louis R, Lau KCL, Bron AO et al. The relationship between airways inflammation and asthma severity. Am J Respir Crit Care Med 2000, 161: 9-16.

16.  Misso NL, Powers KA, Gillon RL et al. Reduced platelet glutathione peroxidase and selenium concentration in atopic asthmatic patients. Clin Exp Allergy1996, 26: 838-47.

17.  Smith CAD, Harrison DJ. Association between polymorphism in gene for microsomal epoxide hydrolase and susceptibility to emphysema. Lancet 1997. 350, 630-33.

18.  Maestrelli P, Saetta M, di Stefano A et al. Comparison of leukocyte counts in sputum, bronchial biopsies and bronchoalveolar lavage. Am J Respir Crit Care Med 1997, 155: 542-48.

19.  Lim S, Jatakanon A, John M et al. Effect of inhaled budesonide on lung function and airway inflammation: assessment by various inflammatory markers in mild asthma. Am J Respir Crit Care Med 1999, 159: 22-30.

20.  Raible DG, Mohanty JG, Jaffe JS et al. Hydrogen peroxide release from human eosinophils on fibronectin: scopoletin enhances eosinophil activation. Free Radic Biol Med 2000, 28: 1652-60.

21.  Ayars G, Altman LC, McManus MM et al. Injurious effect of the eosinophil peroxide-Hydrogen Peroxide-Halide system and major basic protein on human nasal epithelium in vitro. Am Rev Respir Dis 1989, 140: 125-31.

22.  Vachier I, Channez P, Le Doucen C et al. Enhancement of reactive oxygen species formation in stable and unstable asthmatic patients. Eur Respir J 1994, 7: 1585-92.

23.  Repine JE, Bast AALT, Lankhorst IDA and the oxidative stress study group. Oxidative stress in chronic obstructive pulmonary disease. Am J Respir Crit Care Med 1997, 156: 341-57.

24.  Horvath I, Donnelly LE, Kiss A et al. Combined use of exhaled hydrogen peroxide and nitric oxide in monitoring asthma. Am J Respir Crit Care Med 1998, 158: 1042-46.

25.  Antczak A, Kurmanowska Z, Kasielski M, et al. Inhaled glucorticosteroids decrease hydrogen peroxide level in expired air condensate in asthmatic patients. Respir Med 2000, 94: 416-21.

26.  Lantero S, Oddera S, Silvestri M et al. Budesonide down-regulates eosinophil locomotion but has no effects on ECP release or on hydrogen peroxide production. Lung 1999, 177: 219-28.

27.  Keatings VM, Jatakanon A, Worsdell M et al. Effect of inhaled and oral glucocorticosteroids on inflammatory indices in asthma and COPD. Am J Respir Crit Care Med 1997, 155: 542-48.

28.  McLeish KR, Miller FN, Stelzer GT et al. Mechanism by which methylprednisolone inhibits acute immune complex induced changes in vascular permeability. Inflammation 1986, 10: 321-32.