Upper Airway Sampling

THOMAS FRISCHER and EUGENIO BARALDI

University Childrens Hospital, Vienna, Austria; and Pulmonary Function Laboratory, School of Medicine, University of Padua, Padua, Italy

	INTRODUCTION

TOP INTRODUCTION WHAT DO WE KNOW? WHAT DO WE NEED... HOW CAN WE ACHIEVE... REFERENCES

The role of airway inflammation in asthmatic disease has mainly been determined from studies in the adult population. Data from these patients have demonstrated an association between the concentration of inflammatory cells and markers in samples from the lower airways and airway responsiveness and the clinical severity of asthma. Airway inflammation is therefore a useful marker of disease activity. There is less information concerning airway inflammation in young children, in whom the asthma phenotype is thought to be more heterogeneous and in whom access to the lower airways is a major obstacle.

The upper airway, specifically the nasal epithelium, shares many properties with the lower airway. Moreover, it is easily accessible and nasal lavage (NAL) procedures are well tolerated even by infants as young as 4 wk old. In comparison with serum analysis, measurements of inflammatory markers can generally be assumed to reflect airway pathology more directly. Therefore, the determination of inflammatory markers in the nasal epithelium fluid would be attractive for both clinical and epidemiological purposes.

The response of the upper airway to allergens in individuals with hay fever (1), response to viral infection (2, 3), and response to different exogenous harmful substances, such as ozone (4), passive smoking, or particles (5) have been studied extensively. This article focuses on methodological considerations for the use of upper airway sampling in the relatively neglected field of pediatric asthma. In particular, we address the extent to which nasal samples can provide information on lower airway inflammation.

	WHAT DO WE KNOW?

TOP INTRODUCTION WHAT DO WE KNOW? WHAT DO WE NEED... HOW CAN WE ACHIEVE... REFERENCES

Sampling Methods

There is no standardized method for the collection of upper airway epithelial lining fluid. It may be collected either by simple nasopharyngeal aspiration or by instilling a predefined volume of either Ringer lactate or saline into the nasal cavity, either with a needleless syringe or a pump spray (6). Fluid is then recollected by a vacuum pump or by microsuction, using standard capillary glass tubes. The suction catheters may be rinsed with a specific volume of diluent after the collection of the specimen. Furthermore, secretions can be collected via absorption by cotton strips or rubber foam samplers. Simply blowing the nose has also been studied. Similar to the provoked sputum technique, the volume of an NAL specimen can be increased by prior inhalation of hypertonic saline (7). All methods are based on the need to obtain a sufficient sample volume without disturbing the mucosa.

Nasopharyngeal aspiration is not always possible because of lack of secretion, particularly in healthy children, which is an issue for epidemiological studies where data are required for the whole population. However, the advantage of this method is a more precise determination of "true" concentrations of mediators or inflammatory cells. On the basis of these considerations, nasopharyngeal aspirates are more often performed in a clinical setting with selected children suffering from upper respiratory illness (URI) or lower respiratory illness (LRI). Blowing the nose and suction techniques are also restricted by the lack of data that can be collected, because in healthy subjects there is only a sparse layer of secretions on the nasal mucosa (6).

In contrast, absorption techniques almost always generate data if samples are processed quickly, but the concentration of markers may be influenced by irritation of the nasal mucosa. For children, the issue of irritation and thus noncompliance might pose a problem when field studies are performed.

In a study by Klimek and Rasp (6) comparing seven different methods of collection in healthy adults, the highest concentrations of eosinophil cationic protein (ECP) were measured with the suction techniques and the lowest were measured with the lavage techniques; the absorption techniques delivered intermediate results. For the latter two methods, sufficient samples were obtained in approximately two-thirds of subjects, whereas with the former method sufficient samples were obtained in less than 50% of individuals. The authors therefore suggested that the absorption methods and the lavage techniques were superior to suction techniques. Unfortunately, in this methodological paper, no data were generated on reproducibility of the methods.

Nasal Lavage Techniques

With a cooperative child of 7 yr or older, approximately 50% of a 5-ml volume of saline instilled into each nasal cavity can be recollected (4, 8). The method is performed by having the child sit on a chair, with head bent to his/her back. The child is instructed to hold his/her breath and to make a hiccup sound to close the soft palate and thus prevent the fluid being swallowed. The diluent should be warmed to body temperature and should be isotonic to prevent irritation of the nasal mucosa. The child is asked to hold his/her breath for 10 s and then expel the fluid into a specimen cup. If cytokines are to be studied, proteinase inhibitors such as phosphoramidon or leupeptin should be added to the specimen cup before collection of the expelled fluid. This method can be applied safely to almost every child; it is well tolerated and because of its simplicity can be used in field studies involving children. Noah and coworkers (11) performed nasal lavage in 13-yr-old children using a metered-dose nasal inhaler that was actuated five times on each side of the nose, delivering approximately 7 ml with a median return of 37%. For smaller children, lavages have been performed with individuals in the supine position (12), a technique initially devised by Balfour-Lynn (3, 13). Prewarmed saline (2 ml) was instilled into each nasal cavity and immediately aspirated into a specimen trap by inserting a flexible suction catheter. After washing each nostril, 1 ml of saline was aspirated through the catheter to rinse secretions into the trap. For analysis of cellular components, samples must be processed quickly. First, mucus should be removed by filtering the lavage through a layer of gauze, although this method does selectively remove cells. Alternatively, if volumes of secretion are low, the mucolytic agent dithioerithritol (Sputalysin) can be added to the sample. After centrifugation, total cell count can be obtained by a hemocytometer and differential cell counts by preparation of cytocentrifuge slides. Viability of cells can be assessed by staining with trypan blue. Aliquots of the supernatant can then be stored at 70° C for later analysis of soluble markers.

Issue of Dilution

The main issue in terms of interpreting lavage data is the unknown dilution when analyzing the samples. In a methodological study by Heikkinen and coworkers (14) involving children with a URI, the dilution factor ranged between 1.8 and 432 (median, 11.2). In this elegant study, it was suggested that adjustment of the marker of interest by total protein content of the sample provides the best correlation with the "true" concentration as measured in direct nasopharyngeal aspirates in the same children at the same time points. However, these conclusions can be applied only to children without URI, as vascular leakage due to inflammation may alter protein concentrations. In our experience, total protein cannot always be detected in NAL of healthy children and adjustment for albumin does not increase the reliability of ECP concentrations. To adjust for dilution, inulin or lithium chloride can also be added to the lavage fluid (3, 13, 15). Without correction for dilution, the intraindividual variability of any marker in the NAL will increase and thus decrease the statistical power to detect differences between groups. If significant differences between groups are observed, it is likely that a biological effect is present. Dilution is also an issue for measurement of low concentrations of proteins in the NAL, which often will be below the detection limit of the specific assay. A comparison of the NAL technique with other methods of sampling the upper airways has shown that NAL is more often associated with missing data because of values falling below the detection limit (6).

Reproducibility

Few authors have looked at short-term reproducibility (within days) of the NAL. Good reproducibility has been shown for total cell count, differential cell count, and ECP in a limited number of subjects (n = 4) (7). We have studied 24 children less than 1 yr old who were free of infection at the time of NAL. NAL was collected by a protocol similar to that reported by Ingram and coworkers (12). For ECP values obtained at 0, 24, and 72 h after NAL, reproducibility was poor (Figure 1). This may have been due to the repeated occurrence of URIs in this age group and to the adverse effects of environmental agents (9).

View larger version (18K): [in this window] [in a new window]   Figure 1.   Reproducibility of nasal lavage concentrations of ECP (µg/L) at 0, 24, and 72 h (tests 1, 2, and 3, respectively) after nasal lavage in 65 children less than 1 yr of age.

Association with the Pediatric Asthma Phenotype

Given these shortcomings, it is surprising that markers in a single NAL specimen have been shown to differentiate between children with asthma and nonasthmatic children with atopy and to demonstrate some association with the severity of disease (11, 12). Infants with bronchial obstruction during respiratory infection have significantly higher levels of ECP in nasal washes than patients without obstruction (16). Furthermore, measurement of nasopharyngeal ECP concentrations in children with acute bronchiolitis showed that high nasal ECP was a predictor of subsequent hospitalization and physician-diagnosed bronchial obstruction (17). To investigate whether nasal ECP can be used to monitor asthma, repeated NALs have been performed in a limited number of patients with mild to moderate persistent asthma (10). For the group of patients as a whole, ECP tended to increase during exacerbations. However, as this did not occur for every single patient on all occasions, prediction of exacerbations by nasal ECP was not possible.

Does Nasal Lavage Reflect Inflammation in the Lower Airways?

Whereas NAL is an appropriate tool with which to study upper airway disease mechanisms, particularly in allergic rhinitis (1), there are few data on the association between upper and lower airway inflammation in children. In exposure studies with ozone (4) or swine dust in adults (5), changes in NAL reflect, to some extent, inflammation in the lower airway. In experimentally induced infection with rhinovirus, the virus RNA can be found both in the upper and lower airway (2). In wheezy infants, an increase in tumor necrosis factor  concentrations in NAL has been observed, particularly in children infected with the respiratory syncytial virus (13). These data suggest that local inflammation, which promotes asthma exacerbations in the susceptible host, can be studied by the NAL technique. However, more research and adequately controlled trails are needed.

	WHAT DO WE NEED TO KNOW?

TOP INTRODUCTION WHAT DO WE KNOW? WHAT DO WE NEED... HOW CAN WE ACHIEVE... REFERENCES

1. If the objective of analyzing nasal secretions is to obtain information on the state of inflammation in the lower airway in the asthmatic population, further studies are necessary to investigate the relationship between bronchoalveolar lavage and NAL data in pediatric (or adolescent) patients with asthma. Because there is always the possibility that this relationship is confounded by URI or atopy, studies should be performed in patients stratified for these disorders.

2. What is the relationship between NAL and other indices of airway inflammation such as bronchial responsiveness, peak flow variability, or provoked sputum parameters?

3. Can NAL be used to characterize the wheezing infant and identify those at risk for persistent disease?

4. Is adjustment for dilution always necessary, and if so, what is the ideal method with the NAL technique?

5. Can NAL differentiate children with recurrent cough due to URI from children with asthma?

	HOW CAN WE ACHIEVE THIS?

TOP INTRODUCTION WHAT DO WE KNOW? WHAT DO WE NEED... HOW CAN WE ACHIEVE... REFERENCES

1. The most critical point is whether NAL can be used as a surrogate for lower airway sampling. To rapidly resolve this issue, young children undergoing bronchoscopy or bronchoalveolar lavage (BAL) should be evaluated for NAL.

2. When other markers of inflammation are measured in children (NO in exhaled air, etc.), an NAL should be performed simultaneously.

3. Future NAL studies should account for dilution by adding exogenous substances as proposed by Balfour-Lynn and coworkers (3).

4. Studies that prospectively follow cohorts of at-risk children and perform repeat NAL should determine whether nasal inflammation is a risk factor for the subsequent development of lower airway disease.

	Footnotes

Correspondence and requests for reprints should be addressed to Thomas Frischer, M.D., University Childrens Hospital, Währingergürtel 18-20, A-1090 Vienna, Austria. E-mail: thomas.frischer{at}akh-wien.ac.at

	References

TOP INTRODUCTION WHAT DO WE KNOW? WHAT DO WE NEED... HOW CAN WE ACHIEVE... REFERENCES

1. Sim, T. C., L. M. Reece, K. A. Hilsmeier, J. A. Grant, and R. Alam. 1995. Secretion of chemokines and other cytokines in allergen-induced nasal responses: inhibition by topical steroid treatment. Am. J. Respir. Crit. Care Med. 152: 927-933 [Abstract].

2. Gern, J. E., D. M. Galagan, N. N. Jarjour, E. C. Kick, and W. W. Busse. 1997. Detection of rhinovirus RNA in lower airway cells during experimentally induced infection. Am. J. Respir. Crit. Care Med. 155: 1159-1161 [Abstract].

3. Balfour-Lynn, I. M., B. Valman, M. Silverman, and A. D. Webster. 1993. Nasal IgA response in wheezy infants. Arch. Dis. Child. 68: 472-476 [Abstract].

4. Frischer, T., A. Pullwit, J. Kühr, N. Haschke, M. Studnicka, and G. Lubec. 1997. Aromatic hydroxylation in nasal lavage fluid following ambient ozone exposure. Free Radic. Biol. Med. 22: 201-207 [Medline].

5. Cormier, Y., C. Duchaine, E. Israel-Assayag, G. Bedard, M. Laviolette, and J. Dosman. 1997. Effects of repeated swine building exposures on normal naive subjects. Eur. Respir. J. 10: 1516-1522 [Abstract].

6. Klimek, L., and G. Rasp. 1999. Norm values for eosinophil cationic protein in nasal secretions: influence of specimen collection. Clin. Exp. Allergy 29: 367-374 [Medline].

7. Melillo, G., G. Balzano, F. Stefanelli, C. Lorio, E. De-Angelis, and E. Melillo. 1998. Ultrasonic nebulization of hypertonic solution: a new method for obtaining specimens from nasal mucosa for morphologic and biochemical analysis in allergic rhinitis. Allergy 53: 794-797 [Medline].

8. Koller, D. Y., G. Halmerbauer, T. Frischer, and B. Roithner. 1999. Assessment of eosinophil granule proteins in various body fluids: is there a relation to clinical variables in childhood asthma? Clin. Exp. Allergy 29: 786-793 [Medline].

9. Wojnarowski, C., M. Studnicka, J. Kuhr, D. Y. Koller, N. Haschke, C. Gartner, S. Renz, and T. Frischer. 1998. Determinants of eosinophil cationic protein in nasal lavages in children. Clin. Exp. Allergy 28: 300-305 [Medline].

10. Wojnarowski, C., B. Roithner, D. Y. Killer, G. Halmrebauer, C. Gartner, E. Tauber, and T. Frischer. 1999. Lack of relationship between eosinophil cationic protein and protein X in nasal lavage and urine and the severity of childhood asthma in a 6-month follow-up study. Clin. Exp. Allergy 29: 926-932 [Medline].

11. Noah, T. L., F. W. Henderson, M. M. Henry, D. B. Peden, and R. B. Devlin. 1995. Nasal lavage cytokines in normal, allergic, and asthmatic school-age children. Am. J. Respir. Crit. Care Med. 152: 1290-1296 [Abstract].

12. Ingram, J. M., G. P. Rakes, G. E. Hoover, T. A. Platts-Mills, and P. W. Heymann. 1995. Eosinophil cationic protein in serum and nasal washes from wheezing infants and children. J. Pediatr. 127: 558-564 [Medline].

13. Balfour-Lynn, I. M., H. B. Valman, R. Wellings, A. D. Webster, G. W. Taylor, and M. Silverman. 1994. Tumor necrosis factor- and leukotriene E4 production in wheezy infants. Clin. Exp. Allergy 24: 121-126 [Medline].

14. Heikkinen, T., M. Shenoy, R. M. Goldblum, and T. Chonmaitree. 1999. Quantification of cytokines and inflammatory mediators in samples of nasopharyngeal secretions with unknown dilution. Pediatr. Res. 45: 230-234 [Medline].

15. Virolainen, A., M. J. Makela, E. Esko, J. Jero, G. Alfthan, J. Sundvall, and M. Leinonen. 1993. New method to assess dilution of secretions for immunological and microbiological assays. J. Clin. Microbiol. 31: 1382-1384 [Abstract/Free Full Text].

16. Garofalo, R., L. Kimpen, R. Welliver, and P. Ogra. 1992. Eosinophil degranulation in the respiratory tract during naturally acquired respiratory syncytial virus infection. J. Pediatr. 120: 28-32 [Medline].

17. Reijonen, T. M., M. Korppi, M. Kleemola, K. Savolainen, L. Kuikka, I. Mononen, and K. Remes. 1997. Nasopharyngeal eosinophil cationic protein in bronchiolitis: relation to viral findings and subsequent wheezing. Pediatr. Pulmonol. 24: 35-41 [Medline].