Published ahead of print on July 31, 2003, doi:10.1164/rccm.200212-1520OC
© 2003 American Thoracic Society Are Rhinovirus-induced Airway Responses in Asthma Aggravated by Chronic Allergen Exposure?Departments of Pulmonology, Medical Decision Making, and Virology, Leiden University Medical Center, Leiden, The Netherlands; and Respiratory Virus Research Laboratory, University of Wisconsin, Madison, Wisconsin Correspondence and requests for reprints should be addressed to Peter J. Sterk, M.D., Ph.D., Lung Function Laboratory, C2-P, Leiden University Medical Center, P.O. Box 9600, NL-2300 RC Leiden, The Netherlands. E-mail: p.j.sterk{at}lumc.nl
Airway inflammation in asthma may represent a favorable environment for respiratory viral infections, augmenting virus-induced exacerbations in asthma. We postulated that repeated low-dose allergen exposure preceding experimental rhinovirus 16 (RV16) infection increases the severity of RV-induced airway obstruction and inflammation. Thirty-six house dust miteallergic patients with mild to moderate asthma participated in a three-arm, parallel, placebo-controlled, double-blind study. Patients inhaled a low dose of house dust mite allergen for 10 subsequent working days (Days 15 and 812) and/or were subsequently infected with RV16 (Days 15 and 16). Allergen exposure resulted in a significant fall in FEV1 (p < 0.001) and provocative concentration of histamine causing a 20% fall in FEV1 (p < 0.001) and an increase in exhaled nitric oxide (p < 0.001) and percentage of sputum eosinophils (p < 0.001). RV16 infection led to a fall in FEV1 (p = 0.02) and increases in the percentage of sputum neutrophils (p = 0.01), sputum interleukin-8 (p = 0.04), and neutrophil elastase (p = 0.04). Successive allergen exposure and RV16 infection had no synergistic or additive effect on any of the clinical or inflammatory outcomes. In conclusion, repeated low-dose allergen exposure and RV16 infection induce distinct inflammatory profiles within the airways in asthma without apparent interaction between these two environmental triggers. This suggests that preceding allergen exposure, at the used dose and duration, is not a determinant of the severity of RV-induced exacerbations in patients with mild to moderate asthma.
Key Words: asthma exacerbation airway hyperresponsiveness inflammation sputum nasal lavage fluid Despite current guidelines on asthma management, exacerbations of asthma are still prevalent, often requiring additional therapy such as inhaled and/or oral steroids, sometimes including hospital admission. This occurs particularly in patients with relatively severe disease during the fall season (1). It is known that the presence of a common cold and allergen exposure are both associated with such exacerbations (24). Epidemiologic evidence has shown that the combination of these two triggers is particularly harmful to patients with asthma (5). However, there is lack of understanding regarding the pathogenic mechanisms involved. Respiratory viruses are major triggers of asthma exacerbations (2), in which rhinoviruses (RVs) appear to predominate (3, 6). In epidemiologic studies, it has been shown that common colds are associated with an increase in asthma symptoms, variable airways obstruction, airway responsiveness, and features of lower airway inflammation (3, 6, 7). Whereas experimental rhinovirus 16 (RV16) infections mimic these changes in patients with asthma (810), the clinical responses to such experimental infections are relatively mild. It has been suggested that host factors such as atopy or airway hyperresponsiveness influence the host susceptibility to virus-induced changes (11). Indeed, after upper respiratory tract infections, patients with asthma suffer from more frequent, severe, and longer lasting lower respiratory tract symptoms as compared with healthy subjects (12). Hence, the asthmatic "airway milieu" may represent a favorable environment for viral infections, which might be associated with allergic airway inflammation. The patient's natural exposure to environmental allergens can be mimicked using repeated low-dose allergen inhalations (13). Such exposures have been shown to worsen most of the disease features in patients with pre-existing atopic asthma. This has been demonstrated with respect to increased airway hyperresponsiveness (4, 14, 15) as well as for an increase in eosinophilic inflammation (4, 15, 16) and Th2/Th1 cytokine balance (1517). It can be postulated that such inflammation could facilitate asthma exacerbations caused by subsequent respiratory virus infection.
The interaction between repeated low-dose allergen exposure and subsequent RV infection is not only plausible from an epidemiologic standpoint (1, 5), but is also likely to occur at the cellular level within the airways. For instance, the presence of specific cytokines of ongoing allergic inflammation could lead to an increased expression of intercellular adhesion molecule-1, the receptor for the major group of RVs (18), and increased susceptibility to subsequent RV infection (19). Other possible mechanisms of interaction could be (1) switching of virus-specific CD8+ T cells from IFN- Therefore, in this study, we postulated that repeated low-dose allergen exposure preceding experimental RV16 infection increases the severity of an RV-induced asthma exacerbation and thus leads to a flare-up of symptoms, medication use, variability of airways obstruction, airway hyperresponsiveness, and airway inflammation. To that end, we exposed patients with mild to moderate atopic asthma to low doses of house dust mite allergen for 2 weeks, RV16 infection, or the combination of the two.
The extended version of the methods is available in the online supplement. Thirty-six steroid-naive atopic patients with mild to moderate asthma participated (Table 1) (see Table E1 in the online supplement for details of the patients' characteristics). The Medical Ethics Committee of the Leiden University Medical Center approved the study, and the subjects gave written informed consent before entering. The study had a placebo-controlled, double-blind, parallel, three-armed design (Figure 1) . The thirty-six patients were randomly assigned to either low-dose allergen exposure with placebo infection (allergen + placebo; n = 12), placebo exposure with RV16 infection (placebo + RV16; n = 12), or low-dose allergen exposure with subsequent RV16 infection (allergen + RV16; n = 12).
Screening high-dose allergen challenges were performed according to a standardized protocol (27, 28), with purified aqueous allergen extract of Dermatophagoides pteronyssinus (SQ 503; ALK Abelló, Nieuwegein, the Netherlands). Subjects had a documented early and late response. The noncumulative dose of allergen causing a fall in FEV1 of 5% from postdiluent baseline during this allergen challenge was selected as the low dose to be used for repeated allergen exposure. The patients inhaled this daily on working days as a single dose for 2 consecutive weeks (Figure 1). Spirometry was done before and 10, 20, and 30 minutes after inhalation on each occasion (16). RV16 inoculation was done by nasal administration of a total dose of 0.48.6 x 104 50% tissue culture infective dose (TCID50) diluted in 3 ml of Hanks' balanced salt solution to each subject according to a previously described procedure (8, 29). Confirmation of RV16 infection was established by a fourfold or greater increase in virus-specific neutralizing antibody titer in serum and/or by recovery of the virus from the nasal lavages (8, 29). FEV1, provocative concentration of histamine causing 20% fall in FEV1 (PC20FEV1) to histamine (27), exhaled nitric oxide (NO) measurements (30), sputum inductions, and nasal lavages were performed according to the study schedule represented in Figure 1. Asthma symptom scores, cold scores, salbutamol usage, and PEF measurements were recorded three times daily in diary cards during the 7 weeks of the study (16, 22, 31, 32). Sputum was induced and processed according a protocol that has been validated in our laboratory (16, 33). In the sputum supernatant, concentrations of eosinophil cationic protein and levels of neutrophil elastase and IL-8 were determined by enzyme immunoassay (16, 22, 3335). Nasal lavages were obtained using a modified "nasal pool device" (9, 22). The lavages recovered from the first (Hanks' balanced salt solution) and second nostril (phosphate-buffered saline) were used for confirmation of RV16 infection and to determine the cell numbers and mediator level, respectively. Levels of IL-8 were determined by ELISA. For sample size estimation and statistical power, see the online supplement. The cold score was analyzed using the Mann-Whitney U test to test between-group differences (SPSS 10.0; SPSS Inc., Chicago, IL). The longitudinal data were analyzed using a random-effects model to explore allergen-induced, RV16-induced, and allergen + RV16induced changes in asthma symptoms, ß2-agonist usage, PEF values, FEV1, (log-transformed) PC20FEV1, exhaled NO, and sputum and nasal lavage data (36). The STATA 6.0 xtreg procedure (StataCorp, College Station, TX) was used to fit the applied random-effects model. A p value of 0.05 or less was considered statistically significant.
In the majority of patients (21 out of 24) who were inoculated with RV16, RV infection was confirmed, whereas in three (patients 6, 13, and 20), it was not. Patients 28 and 29 in the allergen group had an intercurrent spontaneous "common cold" with RV. Two patients dropped out of the study because of an asthma exacerbation. Patient 24 had a spontaneous exacerbation during the beginning of the study, during the placebo-allergen period. Patient 3 developed an RV-induced exacerbation on the 2nd day of RV16 inoculation (Day 16), after 2 weeks of allergen exposure. The sums of the cold scores and asthma scores that day were 42 and 30, respectively. PEF decreased to 28% of personal best. The patient was withdrawn from the study and admitted to our hospital for 1 night. The patient was treated with oxygen, inhaled ipratropium/salbutamol, and prednisone orally and recovered quickly during the night and following days. The results from these seven patients were excluded from the analysis (see Table E1 for details). There were no differences in baseline values for any of the measurements between the groups (p > 0.06). The acute response to a low dose of allergen exposure as assessed by the mean maximal fall in FEV1 ± SD within 30 min after low-dose allergen exposure was 5.97 ± 1.68% in the combined allergen + placebo and allergen + RV16 groups as compared with 1.40 ± 0.63% in the placebo + RV16 group (p < 0.001). Between-intervention differences with corresponding 95% confidence intervals are outlined in Table 2 . There were significant changes during the 2 weeks of low-dose allergen exposure in most of the parameters as compared with placebo inhalations (discussed later here). However, the differences between the interventions were not maintained beyond the 2 weeks of allergen exposure for any of the parameters.
Asthma Symptom Score, Cold Score, PEF Measurements, and ß2-Agonist Usage During allergen exposure, the asthma symptom score (p = 0.001), lowest morning/highest PEF (p = 0.009), and ß2-agonist usage (p < 0.001) changed significantly (Figure 2) . RV16 infection either after allergen exposure or not after allergen exposure did not result in a significant further change of these parameters. After RV16 infection, the cold score increased significantly in both the allergen + RV16 group (median cold score, 7.5; range, 014, p = 0.048) and the placebo + RV16 group (cold score, 5.0; range, 314, p = 0.006) as compared with the allergen + placebo group (cold score, 2.0; range, 05). There were no significant differences in cold score between the allergen + RV16 group and the placebo + RV16 group (Figure 3) .
FEV1 and Airway Hyperresponsiveness Allergen exposure resulted in a significant decrease in FEV1 (p < 0.001) and PC20FEV1 (p < 0.001) (Figures 4 and 5) . RV16 infection also led to a drop in FEV1 (p = 0.02) but did not result in a significant decrease in PC20FEV1 (p = 0.2). Successive allergen exposure and RV16 infection showed a borderline significant increase in FEV1 (p = 0.052) and no additive changes in PC20FEV1 (p = 0.3).
Exhaled NO Exhaled NO increased during allergen exposure (p < 0.001), whereas it did not change significantly after RV16 infection (p = 0.8). Successive allergen exposure and RV16 infection did not change exhaled NO significantly (p = 0.9) as compared with allergen alone (Figure 6) .
Sputum Eosinophils, Neutrophils, Eosinophil Cationic Protein, Neutrophil Elastase, and IL-8 Repeated allergen exposure resulted in a significant increase in sputum eosinophils (p < 0.001), whereas RV16 infection significantly increased sputum neutrophils (p = 0.01) (Figure 7) . Moreover, RV16 infection also resulted in an increase in sputum eosinophil cationic protein, which was on the border of significance (p = 0.054) and significant increases in sputum neutrophil elastase (p = 0.04) and IL-8 (p = 0.04). RV16 infection after allergen exposure did not result in an additive effect in any of the parameters.
Nasal Lavage Eosinophils, Neutrophils, and IL-8 The percentage of nasal neutrophils (p = 0.005), as well as nasal IL-8 (p = 0.04), increased during RV16 infection, which was accompanied by a drop in the percentage of nasal eosinophils (p = 0.003). There were no significant additive changes in the allergen + RV16 group.
In this study, we found that "priming" the airways with repeated low-dose allergen exposure does not result in an aggravated response with regard to airways obstruction and airway inflammation after RV infection in patients with mild asthma. When given alone, allergen exposure and RV16 infection each resulted in a significant decrease in FEV1. Interestingly, PC20, exhaled NO, and the percentage of sputum eosinophils changed during allergen exposure, whereas sputum neutrophils, neutrophil elastase, and IL-8, and nasal neutrophils and IL-8 increased after RV16 infection. Hence, repeated low-dose allergen exposure and RV16 infection induce distinct inflammatory profiles within the airways in patients with asthma. However, we did not observe any interaction between these two environmental triggers. This indicates that preceding allergen exposure, in the used study design, is not a major determinant of the severity of RV-induced asthma exacerbations in patients with mild asthma. This is the first study in which the effects of repeated low-dose allergen exposure on the response to a subsequent RV infection in patients with asthma have been investigated. First, our findings during low-dose allergen exposure alone is in line with those of previous studies showing worsening of airway hyperresponsiveness (4, 14, 15), increased exhaled NO (16), and an increased eosinophilic inflammation (4, 15, 16). They also confirm inflammatory changes seen in sputum and nasal lavage after RV infection (9, 22). However, whereas there is good evidence of the reverse interaction, namely of RV infection on subsequent allergen-induced airways obstruction and inflammation (37, 38), we found no evidence that allergen exposure worsens subsequent RV-induced responses. Noticeably, our data are not suggestive of any protective effects of preceding allergen challenge on cold symptoms or nasal inflammation, as reported by Avila and colleagues (39). Therefore, it seems as if the epidemiologic evidence of allergen-virus interactions during asthma exacerbations (5) cannot simply be explained by allergen-induced "priming" of the responses to RV infection by the lower airways. The lack of interaction between repeated low-dose allergen exposure and subsequent RV16 infection could theoretically be due to method choices. Our design was mainly driven by safety considerations. We selected steroid-naive patients to avoid any confounding effect of antiinflammatory therapy. In addition, the low-dose allergen exposure was meant to be a mild inducer of inflammation, which by itself would not lead to an exacerbation of the disease (16). Nevertheless, this study and our previous one (16) have shown that this model results in definite physiologic and inflammatory changes. However, it cannot be excluded that there could be an interaction between the two interventions using a higher dose of allergen. It also cannot be excluded that the timing of subsequent RV infection may not have been optimal, as there was a 3-day interval between measuring the effects of repeated low-dose allergen and the RV16 inoculation. This may not effectively mimic the real-life situation and could be a possible reason for a lack of interaction between the two interventions. We used an RV16 inoculum that was known to be safe and for that very reason may exhibit limited virulence in vivo when compared with natural occurring common colds (810). Although we could not confirm RV-induced worsening of airway hyperresponsiveness (9), we did observe increases in cold scores next to an increase in neutrophil counts in sputum and nasal lavages. Finally, we cannot exclude that our failure to detect interaction between allergen exposure and RV16 infection was due to lack of statistical power. The absence of worsening in hyperresponsiveness after RV alone, in this and other studies (40), could be suggestive of this. However, it appeared that each of the triggers by themselves induced adequate (but distinct) inflammatory profiles within the airways, as required for potential interaction.
How can these results be explained? When developing our hypothesis, we postulated that ongoing allergic inflammation would facilitate or augment virus-induced proinflammatory responses in the airways. There are multiple potential pathways for this, which we did not specifically explore in this study. First, allergen-induced IL-4 release could impair viral clearance by a switch in CD8+ T-cell cytokine profile (IFN-
Nevertheless, clinical and epidemiologic data (5, 24) favor additive or synergistic responses to allergens and virus infections in asthma. It can be envisaged that such an interaction during asthma exacerbations can also be dependent on specific host characteristics or more severe degrees of the disease. First, patients with asthma who are susceptible for RV-induced asthma exacerbations could have relatively high IgE levels. Duff and colleagues (45) have shown that high IgE levels together with RV infections are synergistic risk factors for wheezing with colds. Furthermore, very recent data indicate that high levels of total serum IgE are also predictive of lower respiratory tract symptoms after experimental rhinovirus infection in patients with asthma (46). Second, host factors such as an altered balance between proinflammatory IL-1ß and antiinflammatory IL-1ra could also influence virus-induced inflammation and symptoms (4749). Finally, diminished antiviral activity could also occur in patients with a decreased production of IFN-
The authors sincerely thank all of the volunteers in the study and K. Grünberg (Department of Pathology, VU Medical Center, Amsterdam, The Netherlands) and J.E. Gern (Department of Pediatrics, University of Wisconsin Hospital, Madison, WI) for their expertise and advice. The authors acknowledge the expertise and technical assistance of H. van der Veen, M.C. Timmers, S.P.G. Manesse-Lazeroms, and A.C. van der Linden of the Department of Pulmonology and the technicians and E.P.A. de Klerk of the Department of Virology of the Leiden University Medical Center.
Supported by the Netherlands Asthma Foundation (Grant 98.06). This article has an online supplement, which is accessible from this issue's table of contents online at www.atsjournals.org Conflict of Interest Statement: J.d.K. has no declared conflict of interest; C.E.E. has no declared conflict of interest; J.K.S. has no declared conflict of interest; J.A.S. has no declared conflict of interest; C.J.G.vZ-v.D.H. has no declared conflict of interest; C.R.D. has no declared conflict of interest; K.F.R. has participated as a speaker in scientific meetings and has been a member of advisory boards for AstraZeneca, AltanaPharma, Boehringer, Pfizer, GSK, MSD, Novartis and Schering Plough. The Department of Pulmonology received research grants from AltanaPharma ($202,616), Novartis ($90,640), Bayer ($61,762), AstraZeneca ($103,155) and GSK ($299,495) in the years 2000 until 2002; P.S.H. and P.J.S. are staff members in the Department of Pulmonology that received research grants from AltanaPharma ($202,616), Novartis ($90,640), Bayer ($61,762), AstraZeneca ($103,155) and GSK ($299,495) in the years 2000 until 2002. Received in original form December 23, 2002; accepted in final form July 24, 2003
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||