INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Although exercise can provoke an acute asthmatic attack in asthmatic individuals, the possibility that healthy individuals may develop asthma because of sports activity has only recently been considered. In 1993, Larsson and coworkers reported high prevalences of self-reported asthma, asthmalike symptoms with bronchial hyperresponsiveness (BHR), and use of asthma medication in competitive cross-country skiers, and proposed that strenuous training at low temperatures with repeated inhalation of large amounts of cold air might increase the risk of asthma in healthy subjects (1). Although these findings were confirmed in later studies of skiers and long distance runners (2), the evidence supporting of this hypothesis is currently circumstantial. In healthy subjects, bronchial responsiveness to histamine is transiently increased after a short exposure to air at 17° C under conditions of both rest and exercise (5). In skiers, with an asthma diagnosis, the onset of respiratory symptoms occurs later in life in those with a diagnosis of asthma than in matched asthmatic controls (2). Further, exercise-induced symptoms are accompanied by BHR to methacholine, which is also persistent (1) and threefold more prevalent in athletes exposed to a colder winter climate (3). However, BHR is not specific for asthma, since it can also exist in healthy subjects, during viral infections, and in patients with other respiratory diseases (6)

In asthma, bronchial inflammation is present even patients with newly diagnosed disease (9, 10). The characteristic changes in the bronchial mucosa are epithelial damage; infiltration of eosinophils, lymphocytes, mast cells, and macrophages; and remodeling of the airway, with thickening of the epithelial basement membrane (BM) and enlargement of the bronchial smooth-muscle mass. An increased deposition of tenascin, an extracellular matrix protein, is also observed in the BM of asthmatic individuals, and may be a potential marker of the remodelling process in asthma (11).

In adolescent cross-country skiers, a mild to moderate degree of macroscopic inflammation of the proximal airways has been reported, with an increase in the percentage counts of lymphocytes and mast cells in bronchoalveolar lavage fluid (BALF) (12). We recently reported a 2.5-fold greater prevalence of lymphoid aggregates with some of the features of bronchus-associated lymphoid tissue in endobronchial biopsy specimens from these skiers as compared with healthy controls (13). To further characterize the morphologic changes in the bronchial mucosa in these athletes, we quantitated inflammatory cells and tenascin expression in the subepithelial BM of their endobronchial biopsy specimens and compared the results with those for subjects with mild asthma and healthy controls. We also assessed the association of these parameters with BHR to methacholine and with atopy.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects

Forty competitive cross-country skiing athletes from Sweden and Norway, without a prior diagnosis of asthma; 12 mildly asthmatic subjects; and 12 healthy nonathletic control subjects from Estonia were recruited for the study (Table 1). All were nonsmokers. Data about respiratory symptoms, respiratory allergy, use of antiasthmatic medication, training hours within the previous year, and competitive skiing experience were collected from skiers with a self-completed questionnaire (3). No skier had used antiinflammatory asthma medication within the previous year. The mean duration of annual training was 434 h (range: 200 to 630 h) and the mean duration of competitive skiing experience was 7.5 yr (range: 2 to 12 yr). Mildly asthmatic subjects had a history of asthmatic symptoms controlled by ₂-agonists and or theophylline, with BHR to histamine, a positive bronchodilator test, and an FEV₁ > 80% predicted. The mean duration of asthma was 8.9 yr (range: 2 to 30 yr). Five (42%) subjects had atopic asthma. Lung function of skiers was assessed by flow-volume spirometry with the Microlab 3300 Mk2 5 spirometer (Micro Medical Ltd., Gillingham, Kent, UK), and in the other subjects was assessed with the Jaeger Flowscreen spirometer (Erich Jaeger Laboratories, Würzburg, Germany). A positive reversibility test was defined as an increase in FEV₁ of at least 15% after 200 µg salbutamol (Ventolin; GlaxoWellcome, Greenford, UK) in the asthmatic group and after three doses of 0.1 mg rimiterol (Pulmadil; 3 M Health Care Ltd., Loughborough, UK) in the control group. Bronchial provocation with methacholine was performed in skiers as previously described, and BHR was defined as a decline in FEV₁ of at least 20% after administration of a maximal cumulative dose of 1,800 µg methacholine (PD₂₀ FEV₁  1,800 µg) (3). Atopy was defined in skiers as the presence of specific serum immunoglobulin (Ig) E to at least one of eight aeroallergens (house dust mite, cat, dog, horse, timothy grass and birch pollens, mugwort, and cladosporium) via the Phadiotop CAP test (Pharmacia Diagnostics, Uppsala, Sweden), and in asthmatic subjects and controls was defined as a positive skin prick test (> 3 mm in diameter) to a panel of 12 common aeroallergens (Allergologisk Laboratorium A/S, Horsholm, Denmark).

Written informed consent was obtained from all subjects and parents of those subjects younger than 18 yr of age. The study was approved by the local ethics committees for the participating institutions.

Bronchoscopy

Fiberoptic bronchoscopy was performed in skiers in the autumn, at the peak of their preseasonal training program, at the University Hospital of Trondheim, Norway, and in asthmatic patients and control subjects at the Tartu University Lung Hospital in Tartu, Estonia, in accordance with published international guidelines (14) and under premedication and local anaesthesia as previously described (13, 15). Bronchial biopsy specimens were taken from the second- and third-generation carinae, immediately snap-frozen in liquid nitrogen, and stored at 70° C. Subjects were not investigated within 4 wk after having an upper respiratory tract infection.

Processing of Bronchial Biopsies

Biopsy specimens were processed for immunohistochemistry and immunofluorescence as previously described (11, 13). Briefly, 5-µm-thick cryosections were fixed in precooled acetone (20° C) for 10 min. Eosinophils, mast cells, macrophages, T lymphocytes, and neutrophils were identified with the murine monoclonal antibodies (mAbs) EG2 (dilution = 1:50) (Kabi Pharmacia Diagnostic, Uppsala, Sweden), mast cell tryptase clone AA1 (1:500), BerMAC3 (1:25), anti-CD3 (1:1,000), and neutrophil elastase clone NP57 (1:2,000) (all from Dako A/S, Glostrup, Denmark), respectively, and were visualized with the alkaline phosphatase-anti-alkaline phosphatase (APAAP) method with rabbit antimouse antibody. Photographic slides of the stained cryosections were projected onto a calibrated digitizing tablet (Kurta IS/THREE; Kurta Corp., Phoenix, AZ). The density of the inflammatory cells in the whole biopsy specimen, disregarding damaged areas and areas with aggregates of > 50 cells, was computed with the AutoCad program, version 10.1 (Autodesk Inc., Sausalito, CA). For tenascin, cryosections were incubated with mouse mAb 100EB2 at room temperature for 30 min, washed in phosphate-buffered saline, and reincubated with fluorescein isothiocyanate conjugated sheep antimouse IgG (1:150; Jackson Immunosearch Laboratories, West Grove, PA). Areas containing cross-sections of BM were photographed, and the thickness of the tenascin-immunoreactive area was semiautomatically measured with a computerized image-analysis program.

Some of the specimens were fixed and prepared for electron microscopy as described previously (10), to create a more detailed view of the inflammatory cells and their relationship with other structural components, such as the blood vessels in the mucosa.

Slides and photocopies were coded by the same observer prior to analysis. Only slide preparations that contained cryosections from biopsies with bronchial epithelium, BM, lamina propria, and submucosa were evaluated.

Statistical Analysis

Data on cell counts and tenascin are presented as median values with interquartile ranges (IQRs), and were analyzed with the Mann-Whitney U test or Kruskal-Wallis test with Dunn's correction for multiple comparisons, as appropriate. Subgroup analysis by BHR and absence of atopy was also performed. Correlation coefficients (r) were calculated with Spearman's rank method. A value of p < 0.05 was considered statistically significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subject characteristics are presented in Table 1. The bronchodilator test was negative in all control subjects (change in FEV₁: 5.4 ± 4.1% [mean ± SD]) and positive in all asthmatic subjects (29.3 ± 17.4%). Thirty (75%) skiers were hyperresponsive to methacholine (median PD₂₀ FEV₁: 1,246 µg [IQR: 866 to 1,523] µg). Asthmatic symptoms were reported by 21 hyperresponsive and five nonhyperresponsive skiers. Of these, six subjects had consulted their physicians and reported the use of inhaled ₂-agonists. Coughing in episodes or in relation to exercise was reported by 26 (65%) skiers. Of 15 (38%) skiers with atopy, 12 were hyperresponsive to methacholine.

Inflammatory Cell Counts

Assessable bronchial biopsy specimens were obtained from all subjects. Because of unavailability of cryosections, neutrophil counts were not performed in control subjects and in three skiers. Data with cell counts are presented in Table 2 and Figures 1 and 2

View larger version (26K): [in this window] [in a new window]   Figure 1.   Density (cells/mm2) of macrophages, T lymphocytes, mast cells, and eosinophils in bronchial biopsy specimens from controls, skiers with and without BHR and asthmatic subjects. Horizontal bar = median value.

View larger version (12K): [in this window] [in a new window]   Figure 2.   Density (cells/mm2) of neutrophils in bronchial biopsy specimens from skiers with and without BHR and from asthmatic subjects. Horizontal bar = median value.

Group analysis showed that skiers had 43-fold (p < 0.001), 26-fold (p < 0.001), and twofold (p < 0.001) greater T-lymphocyte, macrophage, and eosinophil counts, respectively, than did controls. The skiers' neutrophil count was significantly greater than that of the asthmatic subjects, whereas the lymphocyte count was not significantly different, and the macrophage, eosinophil, and mast cell counts were lower. On subgroup analysis by nonatopic status, the neutrophil count in skiers was not significantly different and the eosinophil count was significantly lower than in asthmatic subjects. The mast cell count was greater in skiers than in controls. There were no significant differences in cell counts in nonhyperresponsive and hyperresponsive skiers. Both skier groups had greater macrophage and lymphocyte counts than controls (Figure 3A), whereas eosinophil counts were twofold greater in hyperresponsive skiers. The neutrophil count was greater in both hyperresponsive (Figure 3B) (86 cells/mm², p < 0.05) and nonhyperresponsive skiers (74 cells/mm², p < 0.05) than in asthmatic subjects.

View larger version (123K): [in this window] [in a new window]   View larger version (132K): [in this window] [in a new window]   Figure 3.   (A and B) Light microscopic image of a specimen is from the large airways of a hyperresponsive subject. Serial sections of the same specimen are stained with mAbs revealing CD3 antigen in all lymphocytes (A) and elastase in neutrophils (B). Both elastase- and CD3-positive cells (stained deep red with in the APAAP method; see METHODS) are seen in large numbers in the airway epithelium (small arrows) and lamina propria. The CD3-positive lymphocytes (A) (some of which are shown with arrowheads) are more numerous and clustered (open arrows) in the lamina propria than are neutrophils (B) (thick arrows), which are located mainly along the basement membrane (long thin arrow). In contrast, very few EG2-positive eosinophils were seen in the same area of the sections (data not shown). G = bronchial glands, L = airway lumen (original magnification: ×300).

Electron microscopy revealed that most of the blood vessels in the sections of biopsy specimens were capillaries and postcapillary venules. The postcapillary venules contained many leukocytes, especially neutrophils (Figure 4). Some of the neutrophils were seen to penetrate through the postcapillary venule wall into the lamina propria. The majority of the inflammatory cell population in the lamina propria consisted of small lymphocytes and neutrophils (Figure 5). Eosinophilic leukocytes, with their morphologically typical crystalloid core granules, were seen only occasionally. Because not all specimens were prepared for electron microscopy, the morphometric analyses of cells were done under light microscopy, using special staining with mAbs.

View larger version (164K): [in this window] [in a new window]   Figure 4.   Low-power transmission electron microscopy image showing postcapillary venules (V) in the lamina propria in a large airway specimen from a hyperresponsive elite skier. Many neutrophils (open arrows) are seen in the venules (original magnification: ×2,400). Bar = 10 µm.

View larger version (180K): [in this window] [in a new window]   Figure 5.   Low-power transmission electron microscopic image of large airways of a hyperresponsive subject. Some neutrophils (N) are present in the abnormal airway epithelium (E). The BM is thickened (open arrows), and many fibroblasts and lymphocytes are seen in the lamina propria beneath the epithelium (original magnification: ×2,400). Bar = 10 µm.

BM Tenascin

The tenascin-specific immunoreactivity band in the BM was significantly thicker in skiers (mean: 6.7 µm; IQR: 5.3 to 8.5 µm, p < 0.001) and asthmatic subjects (mean: 8.8 µm; 7.2 to 10.8 µm, p < 0.001) than in controls (mean: 0.8 µm; IQR 0 to 3.1 µm). When analyzed according to BHR, tenascin expression was increased in the order: controls < nonhyperresponsive skiers < hyperresponsive skiers < mildly asthmatic subjects. Tenascin expression in nonhyperresponsive skiers was not significantly different than in hyperresponsive skiers (Figure 6). On subgroup analysis according to nonatopic status, tenascin expression was found to be significantly greater in skiers (mean 6.5 µm; IQR: 5.2 to 7.5 µm, p < 0.001) and asthmatic subjects (8.8 µm; IQR: 7.2 to 9.7 µm, p < 0.001) than in controls.

View larger version (18K): [in this window] [in a new window]   Figure 6.   Thickness of tenascin immunoreactive band in subepithelial BM zone in controls, skiers with and without BHR, and asthmatic subjects. Horizontal bar = median value.

Correlations

In skiers, the T-lymphocyte count correlated significantly with counts of mast cells (n = 40, r = 0.33, p = 0.04) and macrophages (n = 40, r = 0.5, p = 0.001). Skiing experience in years was significantly correlated only with the macrophage count (n = 40, r = 0.42, p = 0.007). There were no significant correlations of cell counts with tenascin immunoreactivity. Cell counts and tenascin were not significantly correlated either PD₂0 FEV₁ or with bronchial responsiveness.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In this study, we observed a mucosal inflammatory cellular infiltrate and increased subepithelial tenascin deposition in the proximal airways of young, competitive cross-country skiers. Although no skier had a known medical diagnosis of asthma, the possibility of undiagnosed asthma cannot be excluded, since asthmalike symptoms and use of ₂-agonists were reported by 65% and 15% of the skiers, respectively. However, the degree of cellular infiltration in the skiers was different in several respects from that observed in steroid-naïve subjects with mild asthma. With the exception of T lymphocytes, skiers had a lesser degree of infiltration with eosinophils, mast cells, and macrophages. Moreover, skiers had a greater degree of neutrophil infiltration, which is not a significant feature in either atopic or nonatopic asthma (16, 17), suggesting that the inflammatory process in these athletes is different from that in asthmatic individuals.

The changes observed in the skiers in our study contrast with the absence of inflammatory changes in bronchial biopsy specimens from clinically healthy, asymptomatic sportsmen as reported by Power and associates (18). Moreover, Power and associates did not observe hyperresponsive subjects in their study, whereas in the present study, inflammatory changes were observed in hyperresponsive as well as in nonhyperresponsive athletes. We do not know why nonhyperresponsive skiers have inflammatory changes in their airways, but it is possible that these changes are related to repeated exposure of the proximal airways to inadequately conditioned air. Leucokyte infiltration and epithelial injury have been demonstrated in canine airways within 2 h of a 5-min period of hyperventilation with dry air at room temperature, with no evidence of resolution after 24 h (19). In contrast, our athletes regularly exposed themselves to lower ambient temperatures for prolonged periods, and to greater levels of minute ventilation, which are more than adequate to overwhelm the air-conditioning capacity of the upper airways (20).

The increase in tenascin expression in skiers, albeit to a lesser degree than in asthmatic subjects, may reflect ongoing healing and repair processes and remodelling of the airways. In adults, increased expression of this extracellular matrix protein is associated with the process of healing and repair after tissue injury or with oncogenesis (21). Interestingly, treatment with inhaled steroids during the pollen season has been reported to attenuate the increase in tenascin expression in birch pollen-sensitive asthmatic individuals (12). Whether these drugs have a similar effect in skiers should be clarified.

The changes observed in the present study should be considered to be a "snapshot" of the bronchial mucosa, taken at the peak of the cross country ski training season in the autumn. Traffic of inflammatory cells through tissues is continuous, as exemplified by the considerable intraindividual variation in inflammatory changes observed over a period of 1 mo in clinically stable patients with mild to moderate asthma (22). Although there were no strong correlations between the inflammatory indices and duration of skiing experience or training in the skiers in our study it must still be clarified whether the inflammatory changes that we observed persist or progress during the competitive season in the winter, with regression during the off season and on cessation from competitive skiing activity. Meanwhile, it would appear to be prudent to limit thermal and osmotic stress to the airways of these athletes.

In summary, young, elite, competitive cross-country skiers without a previous diagnosis of asthma have an inflammatory cellular exudate and increased subepithial expression of tenascin in the mucosa of their proximal bronchi. Repeated prolonged and intense exposure to inadequately conditioned air may be the factor precipitating this.