Published ahead of print on December 18, 2002, doi:10.1164/rccm.200207-716BC
© 2003 American Thoracic Society
Comparison of Nasal and Oral Inhalation during Exhaled Breath Condensate CollectionDepartments of Pathophysiology, Clinical Chemistry, and Anesthesiology and Intensive Therapy, National Korányi Institute for Pulmonology, Semmelweis University of Medicine; and Oto-Rhino-Laryngology Department, Peterffy S. Hospital, Budapest, Hungary
Correspondence and requests for reprints should be addressed to Ildikó Horváth, M.D., D.Sc., National Korányi Institute for Pulmonology, Department of Pathophysiology, Budapest, Pihen
Analysis of exhaled breath condensate is a method for noninvasive assessment of the lung. Condensate can be collected with a nose clip (subjects inhale and exhale via the mouth) or without it (subjects inhale via the nose and exhale via the mouth), but the mode of inhalation may influence condensate volume and mediator levels. We compared condensate volume and adenosine, ammonia, and thromboxane B2 levels in young healthy volunteers (n = 25) in samples collected for 10 minutes from subjects with or without a nose clip. Patients with allergic rhinitis (n = 8) were also studied to assess the effect of upper airway inflammation on mediator levels. Adenosine, ammonia, and thromboxane B2 levels were determined by HPLC, spectrophotometry, and radioimmunoassay, respectively. Volume of condensate was significantly higher without nose clip than that with nose clip (mean ± SD, 2321 ± 736 µl and 1746 ± 400 µl, respectively; p = 0.0001). We found no significant difference in any mediator levels between these two collection modes in healthy volunteers, but adenosine showed a tendency to differ between oral and nasal inhalation in patients with allergic rhinitis. Our data indicate that whereas a greater volume of condensate can be obtained when subjects inhale through their noses, the mode of inhalation does not influence mediator levels in young healthy volunteers, but may affect these levels in patients with allergic rhinitis.
Key Words: airway inflammation • noninvasive monitoring • adenosine • ammonia • thromboxane There is an increasing interest in noninvasive monitoring of airway diseases. Recently, the collection of exhaled breath condensate (EBC) has been put forward as an easy to perform, convenient method for investigating different mediators from the pulmonary system (1–3). Although many studies have been published recently on EBC, there is no internationally accepted guideline for the performance of sample collection. One open question is the use of a nose clip. Wearing a nose clip, subjects are forced to inhale through their mouths and the nose clip prevents accidental exhalation through the nose. Measurements with a nose clip are preferred so that only mouth-conditioned exhaled air gets into the collection system (4); however, the use of a nose clip opens the nasopharyngeal velum, thus nasal air may at least partly contaminate the exhaled air. In the majority of studies, sample collection has been performed by this method (5–11). Without wearing the nose clip, patients inhale through their noses and exhale through their mouths. Although the mode of inhalation may have an influence on the volume and the composition of EBC, it has not been studied systematically. Our aim was to determine whether the mode of inhalation influences the volume of EBC and/or the level of different mediators in it. We examined EBC volume and level of three different mediators (adenosine, ammonia, thromboxane B2 [TXB2]) in young, healthy volunteers, sampling either with nose clip (inhalation via the mouth) or without (inhalation via the nose). We chose these mediators because they reflect different aspects of the airway function, reflect airway inflammation (12–14), and can be evaluated by completely different techniques. The measurement of adenosine in EBC samples by HPLC has been validated in our laboratory with good reproducibility (15). Therefore, adenosine was considered as the primary variable to assess the potential effect of nose clip use on mediator levels. Levels of ammonia and TXB2 were determined as secondary variables in this study. Because the upper airways and the mouth may influence mediator levels detected in EBC, we also intended to obtain information about the potential source of mediators. We therefore measured the concentration of these three mediators in nasal lavage, saliva, and EBC obtained through tracheostomy. Changes in mediator production in the upper airways during inflammation may alter the concentration of each mediator in EBC because they can be transported ("autoinhaled") to the lower airways and/or may be added to EBC in the mouth. Therefore, we compared mediator concentrations of EBC samples collected via nasal inhalation with those of samples collected with oral inhalation from symptomatic patients with allergic rhinitis. Some of the results of these studies have been previously reported in the form of an abstract (16).
Subjects A total of 25 healthy volunteers without a history of any chronic diseases or airway infections in the previous 2 weeks and 8 patients with allergic rhinitis were studied with and without nose clips (Table 1) . All patients were allergic to common aeroallergens and had symptoms of rhinitis at the time of the measurement. Salivary samples were obtained from 10 healthy subjects, and nasal lavage from 5 patients with allergic rhinitis. EBC was also collected from 10 mechanically ventilated patients (four with acute respiratory distress syndrom [ARDS], three with sepsis, and three postoperatively ventilated subjects without signs of lung disease) through tracheostomy. All participants were nonsmokers. Smoking habit was tested by NicCheck I (DynaGen Inc., Cambridge, MA) (17). The local Ethical Committee approved the protocol, and informed consent from each subject was obtained.
Study Design EBC was collected from healthy subjects and patients with allergic rhinitis twice for 10 minutes during tidal breathing with 15 minute breaks between measurements. The first sample was obtained with a nose clip, the second one without it. In 15 healthy volunteers,  E (E) was also measured during the collection. Samples were centrifuged immediately (at 0°C, 400 x g, for 1 minute) after collection to determine their volume precisely. Amylase concentration was measured to determine salivary contamination.
Sample Collections Nasal lavage was performed by 10 ml normal saline instilled into the nasal cavity with a pipette (18). Subjects tilted their heads back 45° for 10 seconds, then tilted their heads forward and the liquid dripped into a plastic basin. Saliva samples were obtained by centrifugation (at 0°C, 1,000 x g, for 10 minutes) from dental tampons chewed for 1–2 minutes.
Lung Function Test
Mediator Measurements Additional details are provided in an online supplement.
Statistical Analysis
Comparison of Collection with and without Nose Clip in Healthy Volunteers Volume of EBC. The volume of condensate was significantly higher when EBC was obtained without nose clip than with it (2,321 ± 736 µl versus 1,746 ± 400 µl, respectively; p = 0.0001, n = 25) (Figure 1) . The mean E was significantly higher when EBC was collected without nose clip than with it (11.2 L/minute versus 9.1 L/minute, respectively; p = 0.016, n = 15), and E showed significant correlation with the volume of EBC (r2 = 0.56, p < 0.0001).
Mediator levels in EBC. There was no significant difference in adenosine, ammonia, and TXB2 levels between the samples collected with and without nose clip (adenosine 8.3 ± 4.3 nM and 7.9 ± 3.9 nM, ammonia 73.9 ± 54.7 µM and 74.0 ± 57.0 µM, TXB2 30.9 ± 19.6 pg/ml and 24.9 ± 18.9 pg/ml, respectively; n = 25) (Figures 2A, 3A, and 4A) .
A significant, linear relationship was found in adenosine (r2 = 0.614, p < 0.0001) and in ammonia (r2 = 0.500, p < 0.0001) concentrations between the two collection methods, but not in TXB2 (r2 = 0.133, p = 0.073) (Figures 2B, 3B, and 4B). We found a relatively good agreement in adenosine concentration between the two collection methods with a mean difference of 0.4 ± 5.4 nM between the measurements (mean ± 2 SD) (Figure 2C). The agreement in ammonia and TXB2 levels was not as good (mean difference 0.0 ± 85.6 µM and 6.0 ± 43.3 pg/ml, respectively) (Figures 3C and 4C).
Relationship between different variables.
Source of Mediators in EBC In nasal lavage, the level of adenosine was 36 ± 16.6 nM, that of ammonia was 10.4 ± 13.6 µM, and that of TXB2 was 24.3 ± 5.7 pg/ml. There were no significant differences in the levels of any between nasal lavage and EBC samples. The adenosine concentration in EBC samples obtained directly from the lower airways through tracheostomy was 29.4 ± 23.3 nM, that of ammonia was 11.4 ± 15.8 µM (5 out of 10 subjects had no measurable ammonia level in EBC), and that of TXB2 was 44.2 ± 10.6 pg/ml. Ammonia concentration in EBC obtained through tracheostomy from mechanically ventilated patients was significantly lower than that found in EBC samples from healthy volunteers collected through the mouth either with oral (p < 0.05) or nasal inhalation (p < 0.05). No significant difference was detected between lower airway and oral sampling in adenosine and TXB2 concentrations.
Effect of Upper Airway Disease on Mediator Levels Neither ammonia nor TXB2 were significantly different in patients with allergic rhinitis compared with healthy volunteers.
Amylase
In the present study, we investigated whether the mode of inhalation had an influence on the volume or the mediator levels of EBC. The EBC volume was larger when the sample collection was performed without a nose clip, but no difference was observed in concentrations of adenosine, ammonia, and TXB2 in young, healthy volunteers.
Generally, 1 to 3 ml EBC samples can be obtained every 10 to 15 minutes. The duration of the collection, The concentration of adenosine (our primary variable) did not differ, but rather, showed a strong correlation between the two modes of sample collection with good agreement in healthy volunteers. Because adenosine can be determined in EBC with good repeatability (15), we could detect even small differences. Despite this fact, we did not detect changes in adenosine concentration with or without the nose clip. This observation indicates that the mode of inhalation did not affect the adenosine concentration in EBC in young, healthy volunteers. Further study is needed to establish the effect of mode of inhalation on elderly subjects and on patients with lung disorders. We did not find significant differences in the concentrations of our secondary variables (ammonia, TXB2) when comparing the two modes of EBC collection. Although ammonia level showed a strong correlation between the two collection methods, correlations of TXB2 and good agreement in both variables assessed by Bland and Altman test were lacking. This absence of correlation in the secondary variables may be a consequence of the lower reproducibility of the analytic methods (radioimmunoassay and spectrophotometric method) when compared with HPLC, taking into consideration that there was good agreement in adenosine measurement in this study. This assumption of lower reproducibility is in line with preliminary reports, which showed no relationship in the repeatability of measurements of the 8-isoprostane level in healthy subjects and those with asthma (25), or that the exhaled hydrogen peroxide had significant within-day variability and more than 40% higher intra-individual long-term variability (26). This emphasizes the need for the development of more precise and sensitive techniques for determination of these mediators. Although underlying lung diseases may influence mediator levels in EBC in ventilated patients, adenosine and TXB2 in EBC are most likely derived from the pulmonary system because their concentration were in the same range in EBC samples whether they were obtained through tracheostomy or via the mouth. However, the ammonia levels were much lower in samples from mechanically ventilated patients than in healthy volunteers. It is unlikely that the direct contamination with saliva had a considerable influence on our results because salivary amylase activity in our EBC samples was 0.0001 that of the saliva; but taking into consideration that ammonia is a volatile molecule, we believe that much of the EBC ammonia is derived from the mouth and/or upper airways, and added to the samples from the mouth by evaporation. This theory is in line with previously published data showing a marked difference in the ammonia levels between oral and direct lower airway sampling (27). The mean ammonia level we found in conventionally obtained EBC samples was somewhat lower than that found by other research groups (12, 27). This difference may be related to the lower collection temperature of our device, because lower sampling temperature causes lower ammonia concentration in EBC (28). Other factors, including the material of the collection surface and the use of different analytic methods, may have also played a part. It can be supposed that upper airway inflammation like allergic rhinitis may influence mediator concentrations in EBC when subjects inhale through the nose during sample collection. The ammonia and TXB2 concentrations did not verify this hypothesis because they did not differ between samples collected with oral or nasal inhalation in patients with allergic rhinitis. Furthermore, their concentrations were in the same range in samples from patients with allergic rhinitis and healthy subjects. However, a significant difference in adenosine level was found between patients with allergic rhinitis and healthy subjects, and a tendency to differ was shown between oral and nasal inhalation in patients with allergic rhinitis. These observations of our primary variable indicate that allergic rhinitis causes an increase in adenosine level in EBC. The slight disparity between oral and nasal inhalation may have been caused by the high nasal adenosine level that we have found through nasal lavage of these patients, suggesting that adenosine formed in the nose is released into the inhaled air and then is exhaled. This observation confirms the occurrence of aspiration of nasal secretions, which may be one possible pathophysiologic factor responsible for the comorbidity of asthma and allergic rhinitis (29–31). Long-term postnasal drip may initiate inflammation in the lower airways with elevated inflammatory mediator levels. This fact might explain why we found elevated adenosine levels in patients with allergic rhinitis without asthma compared with that of healthy subjects. Thus, adenosine may be a marker of subclinical inflammation of the lower airways similar to nitric oxide (32–34), although this hypothesis needs additional testing. In certain cases, EBC collection with inhalation through the nose (without using a nose clip) may have some advantages over collection wearing a nose clip. Some resistance occurs when inhaling through the mouth due to the valve of the mouthpiece. This resistance is inconvenient for patients, especially for those with airway obstruction. For these patients, inhalation through the nose is easier and less tiring. Moreover, inhalation through the nose is more natural and more convenient for patients because it does not dry up the mouth. The disadvantage of the removal of the nose clip (inhalation through the nose) is that patients must concentrate on exhaling through the mouth instead of the nose, which would be more natural. Consequently, this method of collection needs more attention. Furthermore, some subjects with severe upper airway diseases may not be able to inhale through their noses. EBC could be obtained from these patients only by oral breathing. This study was conducted to investigate EBC collection performed with tidal breathing. Therefore, its results cannot necessarily be applied to EBC collection with forced expiration (35). Determining the effect of the mode of inhalation requires further study under those settings. In summary, we have demonstrated that more condensate volume can be obtained when subjects inhale nasally rather than orally during EBC collection, and that the mode of inhalation does not affect the level of adenosine, ammonia, or TXB2 in young, healthy subjects. Nasal inflammation in allergic rhinitis does not influence the ammonia or TXB2 levels in EBC, but adenosine level shows a tendency to be elevated when EBC is collected with nasal inhalation in these patients. Our data suggest that adenosine and TXB2 levels in EBC are likely rooted in the lower airways. However, ammonia concentrations are higher when EBC is collected through the mouth than through tracheostomy; therefore, it is probably the case that the mouth and/or upper airways have a substantial influence on ammonia concentration in conventionally obtained EBC samples.
The authors thank Mrs. M. Mikoss, Mrs. M. Hernádi and Mrs. M. Kenéz for the collection and processing of breath condensate and saliva samples.
Supported by the Hungarian National Scientific Foundation (OTKA) (T030340) and Ministry of Welfare (ETT) (T-06/160/2001). This article has an online supplement, which is accessible from this issue's table of contents online at www.atsjournals.org Received in original form July 19, 2002; accepted in final form December 16, 2002
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
út 1., P.O. Box 1 H-1529, Hungary. E-mail: 
TOP
ABSTRACT
METHODS
