INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Asthma is a chronic airway disease, characterized by episodic exacerbations of symptoms, often accompanied by transient airways obstruction, and an increase in airway hyperresponsiveness to specific or nonspecific stimuli (1). Prospective epidemiological studies in children as well as adults have shown that episodic falls in peak flow are often associated with respiratory virus infections, and rhinovirus infections in particular (2, 3). This fits in with data obtained by experimental rhinovirus infections in subjects with asthma, demonstrating an increase in both asthma symptoms and nonspecific airway hyperresponsiveness, thereby suggesting that rhinovirus infections can indeed induce mild exacerbations of asthma (4, 5). However, the development of variable airway obstruction has not yet been investigated after laboratory inoculation with rhinovirus. So far, several experimental studies have failed to demonstrate a significant increase in airway obstruction in patients with asthma, as measured by spirometry under standardized laboratory conditions (4).

It can be argued that frequent (daily) measurements of peak flow or FEV₁ are more sensitive measures of variable airway obstruction than less frequent laboratory recordings of lung function (1). In asthma, such daily variability of airway obstruction is associated with the inflammatory changes within the airway wall (7), while it improves in response to antiinflammatory treatment (8, 9). There is evidence that experimental rhinovirus infections are associated with an increase in airway inflammation (10), characterized by cellular infiltration of the airway wall (11). Such cellular infiltration may be accompanied by release of proinflammatory mediators, potentially promoting vasodilation, vascular leakage, and airway wall swelling, which enhances airway obstruction (13, 14). Therefore, in the present study we hypothesized that experimental rhinovirus infection augments variable airway obstruction in subjects with mild atopic asthma.

To test this hypothesis, we examined the effect of experimental rhinovirus 16 (RV16) infection on home recordings of FEV₁, in conjunction with cold and asthma symptom scores in a placebo-controlled parallel study in subjects with atopic asthma. In addition, we also measured lung function and airway hyperresponsiveness to histamine under standard conditions in a lung function laboratory.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects

Twenty-seven adult, nonsmoking subjects with asthma participated in the study. The patient characteristics have been published previously (5, 12), and are summarized in Table 1. The subjects had not used inhaled or oral corticosteroids for at least 3 mo preceding the study, nor had they used any other medication other than inhaled short-acting ₂ agonists on demand. They were atopic, as determined by skin prick test, to 12 common aeroallergens (at least 1 positive wheal, > 3 mm). The Leiden University Medical Center (LUMC, Leiden, The Netherlands) Medical Ethics Committee gave its approval for this study, and informed consent was obtained from all the subjects.

Design

The study had a parallel placebo-controlled design. Nineteen patients were inoculated with RV16 on two consecutive days (Days 0 and 1), and 8 patients received placebo. Before the study the patients were screened for inclusion and exclusion criteria. Home recordings of FEV₁, symptoms of common cold, and symptoms of asthma were noted by the subjects from 4 d before inoculation until 10 d afterward. Airway sensitivity was measured by histamine challenge tests 3 d before inoculation (Day 3) and on Days 4 and 11 after inoculation. Nasal lavage was performed and a blood sample was taken immediately before inoculation. Nasal washings were repeated on Days 2 and 9, and a second blood sample was taken 4 wk after inoculation.

RV16 Inoculation

The RV16 strain and stock were the same as used in previous experiments by others (15) and ourselves (4). The RV16 was cultured according to standards of good laboratory practice and the inoculum was tested to be safe for human in vivo use (16). RV16 was inoculated according to a previously described protocol (4). Briefly, 0.25-1.45 × 10⁴ times the 50% tissue culture infective dose (TCID₅₀) was administered by nasal inhalation of nebulized virus suspension (DeVilbiss 646; DeVilbiss, Somerset, PA), then by spraying the suspension into both nostrils (DeVilbiss 286, powered by a compressor), and finally by instilling droplets of the suspension into both nostrils, using a pipette. This procedure was repeated on the next day. In eight patients the diluent was administered in the same fashion.

Infection was confirmed by RV16-positive viral culture of nasal lavages, and/or by a fourfold or greater increase in RV16-specific serum neutralizing antibodies (5). In addition, all lavages were cultured in rhesus monkey kidney cells (LLC-MK2), Hep2 cell cultures, and HEL cell cultures at 37° C in order to exclude other intercurrent virus infections.

Home Recording of Symptoms and FEV₁

Throughout the study period, from Day 4 to Day 10, symptoms of common cold or asthma were evaluated with a questionnaire that was completed by the participants three times a day. Cold symptoms included sore throat, headache, nasal discharge, sneezing, stuffy nose, malaise, cough, and chills or fever, which were graded from 0 (absent) to 3 (severe) and added up to a total symptom score fore each recording time (maximum, 24). The maximal score after infection ("cold score") was used for correlation analysis (4, 5, 12, 15, 16). Asthma symptoms included breathlessness, wheeze, chest tightness, cough, and nocturnal symptoms. The cumulative asthma score of the first 5 d after infection, corrected for the baseline symptom score, was used for correlation analysis ("asthma score"; maximum, 156). In addition, the daily consumption of albuterol (number of 200-µg doses by metered dose inhaler [MDI]) was noted (5).

Each time after evaluation of symptoms, the patients performed three maximal forced expirations on a portable spirometer (Micro Spirometer; Micro Medical, Rochester, UK) in order to record FEV₁. the highest FEV₁ of three attempts was noted on the diary card. For statistical analysis, occasional missing values on the diary cards were replaced by interpolation of the two adjacent data points.

Laboratory Assessment of Lung Function, Airway Hyperresponsiveness

Histamine inhalation challenge tests were performed using histamine diphosphate in serial doubling concentrations ranging from 0.03 to 8.0 mg/ml as described previously (5, 12). The response was measured as FEV₁, by dry rolling spirometer (Spiroflow; Morgan, Rainham, UK). The tests were discontinued when FEV₁ decreased by more than 20% from the baseline value.

Statistical Analysis

Home-recorded FEV₁ was expressed as a percentage of the predicted value. In addition, the maximal degree of worsening in each individual was expressed as the lowest FEV₁ (percentage of personal best) (17) during the period before RV16 inoculation (period 0), and during three periods after inoculation: Days 0-3 (Period 1), Days 4-7 (Period 2), and Days 8-10 (Period 3). The response to histamine was expressed as the provocative concentration causing a 20% fall in FEV₁ (PC₂₀). Log-transformed values of PC₂₀ were used in the analysis and changes in PC₂₀ were expressed as doubling doses (DDs).

The effects of RV16 or placebo on symptoms, FEV₁, and PC₂₀ were explored by repeated measures analysis of variance (ANOVA) for each group separately, and for all subjects with RV16 or placebo as between-group factors, and with time as a within-group factor. This analysis was repeated for morning, afternoon, and evening recordings separately, in order to exclude the effect of variability during the day. In the case of significant ANOVA effects, paired and unpaired Student t tests were applied to analyze within-group and between-group effects, respectively. The Pearson correlation test was used for evaluation of associations between different parameters. p Values < 0.05 were considered significant. The summary statistics were expressed as means ± SEM.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

One of the RV16-treated subjects (Subject 9) dropped out of the study on Day 7 owing to a moderate exacerbation of asthma that required oral prednisone treatment, to which the subject responded well. RV16 infection was confirmed in all RV16-treated subjects, whereas in the placebo group all of the nasal lavages remained negative for respiratory viruses, without a rise in RV16-neutralizing antibodies (Table 1).

Symptoms

Before RV16 infection common cold symptoms were not different between the groups (ANOVA, p = 0.92). In the RV16 group symptoms of common cold started to increase from Day 0 onward, peaking on Day 2 and returning to baseline within 1 wk in most infected subjects (ANOVA, p < 0.001) (Figure 1A). This increase was significantly different from the placebo group (ANOVA, p < 0.001). The asthma score was also not different between the groups at baseline (ANOVA, p = 0.43). Asthma symptoms gradually increased from Day 0 onward, peaking on Days 2 and 3, and returning to baseline in most RV16-treated subjects within 1 wk (ANOVA, p < 0.001) (Figure 1B). This increase was significantly different from the placebo group (ANOVA, p = 0.04). Before RV16/placebo inoculation, the albuterol use (mean ± SEM) was 0.35 ± 0.13 doses per day. The RV16 group showed a nonsignificant increase in the use of albuterol (ANOVA, p = 0.13), which peaked on Day 3 (Figure 1C). This change was not significantly different from placebo (ANOVA, p = 0.09). The highest albuterol use was recorded by the patient who dropped out owing to an exacerbation of asthma.

View larger version (16K): [in this window] [in a new window]   Figure 1.   (A) Cold symptoms (excluding cough score) in RV16 group, expressed as total of three daily recordings. The closed symbols indicate the mean value. (B) Asthma symptoms in the RV16 group (including cough score), expressed as total of three daily recordings. The closed symbols indicate the mean value. (C ) Albuterol use in the RV16 group, expressed as total daily use in number of 200-µg doses. The closed symbols indicate the mean value. The open symbols connected by a dotted line represent the albuterol use of subject 9, who dropped out on Day 7 owing to an exacerbation of asthma.

Lung Function and Airway Hyperresponsiveness: Laboratory Recordings

As described previously, there were no significant changes in FEV₁ either within the group (ANOVA, p  0.13) or between the groups (ANOVA, p = 0.40) (5). In the placebo group there were no significant changes in PC₂₀ during the study (ANOVA, p = 0.67). In the RV16 group PC₂₀ decreased significantly by Day 4 as compared with baseline (p = 0.02), but not by Day 11 (p = 0.19). These changes were not significantly different from the changes in the placebo group (p  0.10) (5).

Home Recordings of FEV₁

During the days before inoculation the home-recorded FEV₁ was higher in the RV16 group as compared with the placebo group (mean ± SEM: RV16, 94.0 ± 2.4% pred; placebo, 82.4 ± 2.8% pred; ANOVA, p = 0.009). During the study, FEV₁ showed variability during the day (ANOVA, p < 0.001) (Figure 2). When considering morning, afternoon, and evening FEV₁ separately, there was only a slight increase in afternoon FEV₁ in the placebo group (ANOVA: morning, p = 0.16, afternoon, p = 0.01; evening, p = 0.14) (Figure 2). In the RV16 group FEV₁ decreased significantly (ANOVA: morning, p = 0.005; afternoon, p = 0.003; evening, p  0.001), reaching a minimum on Day 2 after inoculation, returning to baseline value during the following 3 d (Figure 2). These effects were significantly different between the groups (ANOVA: morning, p = 0.004; afternoon, p = 0.001; evening, p = 0.001). The lowest FEV₁ in period 0 was not significantly different between the groups (RV16, 85.6 ± 1.2% best; placebo, 81.5 ± 1.8% best; p = 0.07). In the RV16 group the lowest FEV₁ decreased in Period 1 (78.7 ± 2.6% best, p = 0.008), still tended to be decreased in Period 2 (82.8 ± 1.6% best, p = 0.06), and returned to baseline in Period 3 (85.4 ± 1.5% best, p = 0.77), whereas in the placebo group the lowest FEV₁ did not change significantly (ANOVA, p = 0.14) (Figure 3). The change in Period 1 as compared with Period 0 was significantly different between the groups (p = 0.04). In the RV16 group the maximal fall in home-recorded FEV₁ (% best) in Period 1 correlated significantly with the cold score (r = 0.47, p = 0.04) (Figure 4A), asthma score (r = 0.47, p = 0.04), and the change in PC₂₀ on Day 4 as compared with baseline (r = 0.50, p = 0.03) (Figure 4B). In the placebo group, only the maximal fall in home-recorded FEV₁ (% best) in Period 1 and asthma score were significantly correlated (r = 0.74, p = 0.04).

View larger version (18K): [in this window] [in a new window]   Figure 2.   Home-recorded FEV1 in the RV16 group (open circles) and the placebo group (solid squares), expressed as a percentage of baseline (mean ± SEM). RV16 or placebo was inoculated on Days 0 and 1.

View larger version (9K): [in this window] [in a new window]   Figure 3.   Lowest home-recorded FEV1 in the RV16 group (open circles) and the placebo group (solid squares), expressed as a percentage of personal best, during Period 0 (Days 4 to 1), Period 1 (Days 0 to 3), Period 2 (Days 4 to 7), and Period 3 (Days 8 to 10). RV16 or placebo was inoculated on Days 0 and 1.

View larger version (10K): [in this window] [in a new window]   Figure 4.   Correlations of cold score and the maximal fall in home-recorded FEV1 (percentage of personal best) during the acute phase after infection (period 1), (A), and of the maximal fall in home-recorded FEV1 (percentage of personal best) and the change in PC20 to histamine between Days 3 and 4 (expressed as doubling doses, DD) (B) in the RV16 group.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

This study demonstrates that experimental RV16 infection leads to a transient fall in daily home recordings of FEV₁ in subjects with asthma. This virus-induced decrease in FEV₁ appears to be significantly related to the accompanying cold symptoms and asthma symptoms, and to the increase in airway hyperresponsiveness to histamine. These data indicate that experimental RV16 infection leads to variable airway obstruction, suggesting that a rhinovirus-induced exacerbation of asthma may be associated with airway inflammation and airway wall swelling.

This is the first study demonstrating worsening of airways obstruction by experimental rhinovirus infection in asthma. This finding is in keeping with epidemiological data, showing that the detection of rhinovirus in the nasal lavage is associated in time with a drop in peak flow in patients with asthma (2, 3). The experimental design of the present study allows assessment of the time course of this phenomenon, and of the relationship with symptoms and changes in airway hyperresponsiveness. Previous experimental studies of patients with asthma, including the present one, have failed to demonstrate an effect of rhinovirus infection on laboratory recordings of FEV₁ (4, 11). This suggests that the timing or the procedure of the spirometric measurements in the lung function laboratory may mask the occurrence of variable airway obstruction, which is associated with asthma symptoms in everyday life.

The data in the present study were obtained by applying a controlled design, and carefully validated methods for RV16 inoculation (4, 5) and standardized laboratory measurement of lung function (18) and airway hyperresponsiveness (19). We chose to study home recordings of FEV₁ rather than peak flow, because FEV₁ may be a slightly more sensitive and more specific measure of airways obstruction, being less effort dependent than peak flow (18). It can be argued that the home recordings of FEV₁ may have been influenced by on-demand use of albuterol. However, albuterol use was generally low in these subjects with mild asthma, and did not increase significantly during the study in either group. Moreover, the use of albuterol would have led to an underestimation of airways obstruction, and can therefore not explain the fall in FEV₁ together with a rise in asthma symptoms in the RV16 group.

We observed a decrease in home-recorded FEV₁, whereas such a change could not be detected using laboratory recordings of FEV₁. Studies on exacerbations of asthma that were either spontaneous, presumably virus-induced (20), or due to experimental tapering of steroid treatment (21), have shown that home recordings are more sensitive in picking up changes in airway obstruction than are laboratory recordings of subjects with asthma. We speculate that this is also the case in our study. In the preceding studies (20, 21) peak expiratory flow rate (PEFR) rather than FEV₁ was recorded in the home setting. However, frequent recordings of FEV₁ show a pattern of variable airway obstruction that is qualitatively comparable to PEFR recordings in patients with asthma (22). Therefore, we consider the home recordings to be valid, thus allowing assessment of effects of RV16 infection on the development of airway obstruction. The observed differences between home and laboratory recordings may be attributed to factors such as the frequency and timing of the measurements (22) and/or the volume history preceding the measurements (23).

How can a rhinovirus cold lead to variable airway obstruction? There is evidence to suggest that experimental rhinovirus 16 infections are accompanied by an increase in airway inflammation (10), characterized by cellular infiltration of the airway wall, particularly with lymphocytes and eosinophils (11). Such inflammation appears to be accompanied by release of proinflammatory mediators from these cells (12) and airway tissues such as the epithelium (24, 25), which subsequently may cause airway smooth muscle contraction, vasodilation, and vascular leakage, leading to edema and airway wall swelling (13). It is highly likely that increased airway wall thickness, both internal and external from the airway smooth muscle layer, augments airway narrowing during smooth muscle contraction (14, 26, 27). This has been difficult to confirm in humans, because it probably occurs predominantly within the small airways (27), where asthmatic airway inflammation has indeed been demonstrated (28). It has been postulated that respiratory virus infections may induce small airway inflammation (31), but this remains to be established for rhinovirus colds in humans.

Our data suggest that the use of home-recorded FEV₁ (even when FEV₁ is measured only once daily) appears to be a sensitive parameter for early detection of virus-induced exacerbations of asthma, thereby allowing timely adjustment of bronchodilator and/or antiinflammatory therapy. As part of an adequate self-management plan, this could lead to prevention of severe airway obstruction or perhaps even the development of severe virus-associated exacerbations in asthma.

In conclusion, this study shows that experimental rhinovirus 16 infection induces the clinical and functional features of a mild exacerbation of asthma. This resembles exacerbations observed after natural rhinovirus infection, thereby allowing further experimental studies on both immunological mechanisms and therapeutic interventions, in order to optimize treatment of virus-induced exacerbations of asthma.