Inhaled Corticosteroids and the Beneficial Effect of Deep Inspiration in Asthma
http://www.100md.com
美国呼吸和危急护理医学 2005年第9期
Johns Hopkins Asthma and Allergy Center, Johns Hopkins University, School of Medicine, Baltimore, Maryland
Istituto di Medicina Generale e Pneumologia, Cattedra di Malattie dell'Apparato Respiratorio, University of Palermo, Palermo, Italy
ABSTRACT
Deep inspirationeCinduced bronchoprotection and bronchodilation are impaired in asthma. We evaluated the effect of inhaled glucocorticosteroids on these phenomena. Two groups of subjects with asthma, 9 with moderate/severe hyperresponsiveness to methacholine, and 12 with mild/borderline hyperresponsiveness to methacholine, received inhaled fluticasone (880 e daily) for 12 weeks. Serial bronchoprovocations were performed at Weeks 0, 6, and 12. The impact of deep inspirations on the airway response to methacholine was evaluated on the basis of inspiratory vital capacity and FEV1. Fluticasone produced a wide spectrum of changes in the beneficial effects of deep inspiration, but the mean changes were not significant. The magnitude of the steroid-induced changes in bronchoprotection by deep inspiration correlated with baseline log PC20 (the provocative concentration of methacholine causing a 20% fall in FEV1; higher log PC20 predicted improvement of the deep inspiration effect). The steroid-induced changes led to the emergence of strong positive correlations between the effects of deep inspiration and the methacholine log PC20 that did not exist at baseline. We conclude that deep inspirationeCinduced bronchoprotection can be restored by inhaled glucocorticosteroids only in individuals with mild hyperresponsiveness. After steroid treatment, the beneficial effects of deep inspiration become significant determinants of the magnitude of airway hyperresponsiveness.
Key Words: airway hyperresponsiveness; asthma; deep inspiration; fluticasone; methacholine
We and others (1eC6) have shown that deep inspiration exerts two potent beneficial effects on the airways of healthy subjects: the bronchoprotective effect, that is, the ability of deep inspirations performed before the inhalation of a spasmogen to prevent bronchial constriction; and the bronchodilatory effect, where deep inspirations taken immediately after the induction of bronchoconstriction improve airflow. In our previous work (1, 3), we have also found that the bronchoprotective effect of deep inspiration is greatly reduced in subjects with airway hyperresponsiveness, even when it is not accompanied by the diagnosis of asthma, and have suggested that the lack of deep inspirationeCinduced bronchoprotection is one of the major underlying causes of the hyperresponsive state. On the other hand, the bronchodilatory effect of deep inspiration is less discriminatory between airway hyperresponsiveness and normal responsiveness (3, 4, 6, 7).
Glucocorticosteroids are capable of reducing airway hyperresponsiveness (8), suggesting that this condition is somehow associated with airway inflammation. If so, it is possible that the loss of deep inspirationeCinduced bronchoprotection is linked to airway inflammation. The current study was designed to indirectly test this hypothesis by assessing the effect of inhaled glucocorticosteroid treatment on deep inspirationeCinduced bronchoprotection in volunteers with asthma.
We have previously reported that, as a group, individuals with borderline to mild airway hyperresponsiveness to methacholine demonstrate loss of deep inspirationeCinduced bronchoprotection to the same extent as those with moderate to severe hyperresponsiveness (3). This has raised the possibility that bronchoprotection by deep inspiration is the first normal function to be lost in the development of hyperresponsiveness. If so, and if inhaled glucocorticosteroids were to improve this function, their effect may be more evident in individuals with mild to borderline hyperresponsiveness, as opposed to those with a more advanced condition. To test this hypothesis, the design of this study has included two groups of volunteers with asthma: a group with borderline to mild hyperresponsiveness and a group with moderate to severe hyperresponsiveness.
METHODS
Subjects
We studied 21 subjects with asthma. Inclusion/exclusion criteria were identical to those employed in a previous study (3). All but one subject were skin-test positive to at least one aeroallergen. All subjects had intermittent or mild persistent asthma (9) and none had received inhaled or oral corticosteroids. Short-acting -agonists were withheld for at least 12 hours before each visit. For subjects who developed respiratory infections, bronchoprovocation sessions were postponed for at least 3 weeks. The study was designed to involve two groups, based on hyperresponsiveness to methacholine: (1) moderate to severe hyperresponsiveness (PC20 [provocative concentration of methacholine causing a 20% fall in FEV1] < 1 mg/ml, n = 9) and (2) mild to borderline hyperresponsiveness (1 < PC20 < 25 mg/ml, n = 10). We also entered two subjects with PC20 values of 44.7 and 45.4 mg/ml, respectively, but with a clinical history of asthma. The study was approved by the Johns Hopkins Bayview Institutional Review Board, and all subjects gave written, informed consent.
Study Design
The entire study consisted of two phases (Figure 1).
Phase 1. Each subject underwent a routine methacholine bronchoprovocation, using approximately half-logeCincreasing doses from 0.025 to 75 mg/ml (10), to assess the PC20. Thereafter, and on separate visits, a series of modified, single-dose bronchoprovocations (starting from 0.075 mg/ml) was performed as previously described (2) to determine (1) the single dose of methacholine inducing an at least 20% fall in FEV1 in the absence of deep inspirations (Figure 2a), (2) the ability of deep inspiration to reverse bronchoconstriction, expressed as percentage of bronchodilation for inspiratory vital capacity (IVC) and FEV1 (Figure 2b), and (3) the ability of deep inspiration to prevent bronchoconstriction, expressed as percentage of bronchoprotection for IVC and FEV1 (Figure 2c). This methodology is described in the online supplement.
Phase 2. All subjects received inhaled fluticasone (220 e, two inhalations twice daily) through a metered-dose inhaler and a spacer (AeroChamber; Trudell Medical International, London, ON, Canada), for 12 to 14 weeks. Each subject visited the laboratory every 2 weeks. At 6 weeks, and at the end of 12 weeks, the series of methacholine bronchoprovocations performed in Phase 1 were repeated.
Data Analysis
The log PC20, the percentage of bronchoprotection and the percentage of bronchodilation by deep inspirations, and the log of the single-dose methacholine required to reduce FEV1 by more than 20% in the absence of deep inspirations represented the outcomes of this study. To calculate bronchoprotection and bronchodilation, we employed the IVC as primary measure. The advantage is that the maneuver to determine IVC (from end-tidal inspiration to residual volume and then to total lung capacity) is free of the potential bronchodilatory effect of the deep inspiration that is involved in the measurement of FEV1 (11eC13). Treatment-induced changes from baseline over time were analyzed by repeated measures one-way analysis of variance (ANOVA) or by Friedman ANOVA, followed by appropriate post hoc tests. Linear regressions or the Spearman rank correlation were used to examine the relationship between conventionally measured airway responsiveness (expressed as log PC20) and deep inspirationeCinduced bronchoprotection and bronchodilation. Nonparametric statistics were used for post-steroid methacholine responsiveness values because, in five subjects who did not reach a PC20 after treatment, this parameter was assigned an arbitrary value (225 mg/ml). Significance was assumed for two-tailed values of p 0.05.
RESULTS
Phase 1
The demographic and lung function characteristics of the study participants are shown in Table 1. The two subjects who had low responsiveness at baseline were analyzed together with the "mild to borderline hyperresponsiveness" group. The primary analyses were conducted with and without these two subjects, with no qualitative impact on the outcome. The two study groups did not differ in terms of age (unpaired t test, p = 0.70), FEV1 % predicted (unpaired t test, p = 0.44), FVC % predicted (unpaired t test, p = 0.17), or FEV1/FVC (unpaired t test, p = 0.74). Also, the two groups did not differ with respect to the duration of asthma (23 ± 3.9 years in the mild to borderline hyperresponsiveness group vs. 28 ± 2.6 years in the moderate to severe hyperresponsiveness group; p = 0.33). By design, the magnitude of airway hyperresponsiveness, expressed as the PC20, was 15-fold different between the two subject groups (0.35 vs. 5.25 mg/ml, geometric mean). Table 2 presents information on the effects of the single-dose methacholine challenges. Notably, we met the target of inducing similar reductions in lung function with single doses of methacholine in the two groups. The single dose of methacholine employed to achieve the targeted change in lung function from baseline in the protocol devoid of deep inspirations was twofold lower in those subjects with moderate to severe hyperresponsiveness (0.52 mg/ml, geometric mean) than in those with mild to borderline hyperresponsiveness (1.02 mg/ml), the difference not being statistically significant. When all subjects were pooled into a single group, there was a significant correlation between the log PC20 to the conventional methacholine bronchoprovocation and the log of the single-dose methacholine required to reduce lung function in the absence of deep inspirations (linear regression, r2 = 0.21, p = 0.04).
As shown in Table 2, the bronchoprotective effect of deep inspiration as measured by IVC or FEV1 did not differ between the two groups. Also, this effect was not significantly different from zero (for IVC: one-sample t test, p = 0.14 and p = 0.40, moderate/severe and mild/borderline hyperresponsiveness, respectively; for FEV1: p = 0.07 and p = 0.20), supportive of our previous reports of loss of the effectiveness of deep inspirations to prevent bronchoconstriction in individuals with even borderline airway hyperresponsiveness (1, 3). The bronchoprotective ability of deep inspiration did not correlate with the degree of airway hyperresponsiveness as measured by the log PC20 (simple regression, r2 = 0.00, p = 0.99).
As we and others have previously described (3, 4, 6, 7), statistically significant bronchodilation by deep inspiration was present in both groups, as shown in Table 2, whether IVC or FEV1 was considered. No significant difference in the bronchodilatory effect of deep inspiration was found between the two groups. As with deep inspirationeCinduced bronchoprotection, when we assessed the relationship between the bronchodilatory ability of deep inspiration and the baseline magnitude of airway hyperresponsiveness (methacholine log PC20), no significant correlation was found (simple regression, r2 = 0.02, p = 0.53).
Phase 2
Twenty of 21 subjects completed the study. One subject (Subject 19) withdrew at 6 weeks of therapy for reasons not related to the study. During the study, four subjects experienced mild asthma exacerbations, presumably the result of respiratory viral infections; consequently, their next assessment was postponed for 3 weeks while continuing their inhaled corticosteroid use. None of the volunteers reported any adverse events associated with inhalation of the corticosteroids. No adverse events were recorded after any methacholine bronchoprovocation.
Treatment with inhaled fluticasone resulted in a small, but not significant, improvement in FEV1, and no improvement in FVC, when the two groups were analyzed separately. The increase in FEV1/FVC reached statistical significance in the mild to borderline hyperresponsiveness group (repeated measures ANOVA, p = 0.04). When the groups were combined, post-treatment FEV1 was significantly higher than the pretreatment value (repeated measures ANOVA, p = 0.04).
Fluticasone reduced airway responsiveness in both groups: significant changes in methacholine log PC20 occurred at Week 6 and improvement continued between Weeks 6 and 12. Despite the fact that the two groups started treatment with a more than 1-log difference in methacholine PC20, no significant difference was observed at Week 12 (Mann-Whitney U test, p = 0.28; Figure 3). Notably, when the conventional bronchoprovocation challenge was performed at the end of the study, five subjects had a less than 20% fall in FEV1 with the highest concentration of methacholine (75 mg/ml). Two of these five subjects belonged to the group with moderate to severe hyperresponsiveness, which indicates that the improvement in their PC20 was on the order of 2 logs. The PC20 value for these five subjects was arbitrarily determined as 225 mg/ml, a value that is approximately 0.5 log higher than the highest dose of methacholine administered (75 mg/ml).
Significantly higher single doses of methacholine had to be administered at 12 weeks of treatment to induce the targeted bronchoconstriction in the protocol devoid of deep inspirations in both groups (Figure 4). The change in the log of the single-dose methacholine for the mild to borderline hyperresponsiveness group was 0.79 (Friedman ANOVA, p = 0.01); for the moderate to severe hyperresponsiveness group, it was 0.76 (Friedman ANOVA, p = 0.02). Remarkably, the steroid-induced change in log PC20 from the conventional challenge was in absolute correlation with the change in log of the single-dose methacholine challenge, which was conducted in the absence of deep inspirations (r2 = 1.0).
The reduction in lung function in the single-dose methacholine challenges that were performed in the absence of deep inspirations at Weeks 6 and 12 was similar to that obtained at baseline (for the two subject groups combined, reduction in IVC: 26.3 vs. 27.9 and 27.6%, baseline vs. 6 and 12 weeks, repeated measured ANOVA, p = 0.81; reduction in FEV1: 33.0 vs. 34.5 and 34.8%, repeated measured ANOVA, p = 0.82). This was also true for each group analyzed separately, allowing for a direct comparison of the bronchoprotective and the bronchodilatory effects of deep inspiration between baseline and 6 and/or 12 weeks. The effect of fluticasone on bronchoprotection by deep inspirations is depicted in Figure 5 (top). We observed some improvement, albeit not statistically significant, in the ability of deep inspirations to protect airways from constriction in the group with mild to borderline hyperresponsiveness. Similar results were obtained when FEV1 was used as the spirometric outcome to calculate the bronchoprotective effect of deep inspiration. No significant difference was found between the two groups when the fluticasone-induced changes in the percentage of bronchoprotection IVC and FEV1 were compared.
Five subjects showed impressive improvement in the bronchoprotective effect of deep inspirations, reaching almost complete bronchoprotection (> 85%). Some of these subjects had dramatically reduced their airway responsiveness, as measured by conventional PC20. Yet we found no overall relationship between the change in log PC20 (log PC20, baseline to the end of treatment) and the change in percentage of bronchoprotection IVC induced by inhaled fluticasone (Spearman rank correlation, rs = 0.09, p = 0.69). On the other hand, all of these five subjects belonged in the group with mild to borderline hyperresponsiveness. Therefore, we analyzed the relationship between the baseline log PC20 and the change in percentage of bronchoprotection. This analysis yielded significant results, whether bronchoprotection by deep inspiration was calculated with IVC or FEV1 (simple regression, r2 = 0.31, p = 0.01, for percentage of bronchoprotection IVC; and r2 = 0.36, p = 0.005, for percentage of bronchoprotection FEV1; Figure 6a).
Fluticasone treatment did not have any effect on the bronchodilatory function of deep inspiration, as measured by IVC or FEV1, in either study group, or in the single-group analysis (IVC data in Figure 5, bottom). In contrast to our findings with deep inspirationeCinduced bronchoprotection, no relationship was found between the degree of baseline responsiveness and the changes in deep inspirationeCinduced bronchodilation after steroid treatment (Figure 6b).
In Phase 1 of the study (presteroid treatment), we found no relationship between the bronchoprotective effect of deep inspiration and airway responsiveness, as measured by the methacholine PC20. We reassessed the relationship between these two outcomes at the 12-week evaluation time point. At the end of the study, the percentage of bronchoprotection IVC correlated significantly with the log PC20 when the subject population was treated as a single group (Spearman rank correlation, rs = 0.58, p = 0.01; Figure 7). Analysis within each subgroup yielded a significant relationship only in the mild to borderline hyperresponsiveness group (Spearman rank correlation, rs = 0.73, p = 0.01). Similar findings were obtained when correlation analyses were performed with percent bronchoprotection FEV1 as the independent variable. We also performed these analyses with the bronchodilatory effect of deep inspiration as the independent outcome. For deep inspirationeCinduced bronchodilation, the percentage of bronchodilation IVC after steroid treatment correlated significantly with the methacholine log PC20 in the single-population analysis after steroid treatment (Spearman rank correlation, rs = 0.57, p = 0.01). The significant baseline relationship between the log PC20 of the conventional methacholine challenge and the log of the single-dose methacholine administered in the absence of deep inspirations remained virtually unchanged after fluticasone treatment compared with before treatment (rs = 0.58, p = 0.01, and rs = 0.47, p = 0.04, respectively).
DISCUSSION
The purpose of this study was to examine whether impairment of the beneficial effects of deep inspiration in individuals with asthma can be reversed by inhaled glucocorticosteroid treatment. The study was more focused on the bronchoprotective effect of deep inspiration, because the loss of this effect is strongly linked to airway hyperresponsiveness (3). The study was also designed to assess whether recovery of deep inspirationeCinduced bronchoprotection by glucocorticosteroids can occur more effectively in individuals with mild hyperresponsiveness, compared with those with moderate or severe hyperresponsiveness.
We generated a number of important observations. Twelve weeks of treatment with a moderately high dose of fluticasone did not produce statistically significant improvement in deep inspirationeCinduced bronchoprotection in either subject group or when the two groups were pooled for analysis (Figure 5). However, a few subjects had a remarkable improvement in bronchoprotection. This was not a random event, because these subjects were among those with the mildest degree of baseline hyperresponsiveness (Figure 6a). Also, this was not a regression to the mean phenomenon, because the relationship between the bronchoprotective effect of deep inspiration at baseline and the fluticasone-induced change in that effect was not statistically significant (data not shown). Equally interesting is the observation that individuals with the highest degree of baseline hyperresponsiveness showed an overall worsening of the protective effect of deep inspiration (Figure 5). Another important finding is that treatment with fluticasone converted an absent effect of deep inspirationeCinduced bronchoprotection and bronchodilation on airway hyperresponsiveness (as measured by the conventional log PC20) to a statistically significant effect (Figure 7). Finally, we found that inhaled glucocorticosteroid treatment resulted in significant improvement in airway responsiveness, when assessed by the single-dose methacholine bronchoprovocation technique, in the absence of deep inspirations (Figure 4).
It appears that bronchoprotection by deep inspiration is lost at an early stage of the inflammatory airway syndrome to which asthma belongs. We previously reported that this phenomenon is absent in those subjects with asthma with borderline to mild hyperresponsiveness, but also in individuals with allergic rhinitis and airway hyperresponsiveness (3). We are now showing that, at that early stage, the protective effect of deep inspiration can improve to normal levels with antiinflammatory treatment. In contrast, when severe hyperresponsiveness is present, the bronchoprotective effect of deep inspirations is clearly not improved with inhaled glucocorticosteroid treatment. Perhaps the loss of bronchoprotection is a permanent phenomenon at this stage, or perhaps treatment was not given for a long enough period of time. Although we did not measure markers of airway inflammation in this study, the remarkable improvement in the conventional PC20, even in the group of subjects with moderate to severe hyperresponsiveness, offers strong evidence that fluticasone exerted the expected antiinflammatory effects. Overall, these data suggest that the reduced protective effect of deep inspiration is an acquired phenomenon, possibly related to airway inflammation, and does not reflect a preexisting condition, as we had inferred in a previous publication (3).
By improving deep inspirationeCinduced bronchoprotection in subjects with mild hyperresponsiveness while worsening it in subjects with severe hyperresponsiveness, inhaled glucocorticosteroids led to the emergence of a significant relationship between the log PC20 to methacholine and the protective effect of deep inspiration (Figure 6). The absence of such a relationship at baseline was predictable from our previous work, where we have found bronchoprotection by deep inspiration equally lacking in asthmatic subjects with severe and with mild airway hyperresponsiveness (3). In the current study, not only did deep inspirationeCinduced bronchoprotection at the end of treatment correlate with airway responsiveness at the same time point but it also correlated equally well with airway responsiveness at the beginning of the study. It is reasonable to hypothesize that, at baseline, other active factors, presumably of an inflammatory nature, masked any relationship between the effect of deep inspiration and the log PC20. The impact of these factors may have been reduced or eliminated by the fluticasone treatment. The nature of such factors remains unknown at this point. Of note, our data suggest that, after removing the corticosteroid-sensitive elements of airway inflammation, a better understanding of the role of deep inspiration on airway responsiveness may be attained.
The design of this study allowed for tabulation of the dose of methacholine, which, when administered in the absence of deep inspirations, reduces lung function to a targeted degree. That dose is an outcome that reflects the component of airway responsiveness that is free of the effect of deep inspiration. Perhaps this is a true measure of airway smooth muscle reactivity, although other influencing elements (e.g., thickness of airway wall) cannot be excluded. In the absence of deep breaths, compared with baseline, significantly higher doses of methacholine (3.3- and 4.2-fold for subjects with asthma with moderate to severe hyperresponsiveness and for those with mild to borderline hyperresponsiveness, respectively) were required at the end of glucocorticosteroid treatment to induce the targeted reduction in lung function. Given the fact that glucorticosteroid treatment did not produce major overall improvement in the beneficial effects of deep inspiration, the primary contribution to the steroid-induced improvement in the methacholine PC20 can be assigned to this effect. Further dissection of the response to methacholine in the absence of lung inflation may lead to better understanding of the exact mechanism through which inhaled glucocorticosteroids exert their beneficial effects in asthma.
In conclusion, we found that inhaled glucocorticosteroids had a profound effect on the relationship between airway hyperresponsiveness and the beneficial effects of deep inspiration. Our data also suggest that deep inspirationeCinduced bronchoprotection is lost early in the pathologic process of asthma, at which point it may be recoverable. In more advanced stages, the loss of the beneficial effects of lung inflation is clearly resistant to glucocorticosteroids. Overall, early intervention with inhaled corticosteroids is advisable. Future studies should aim at addressing whether such treatment applied to individuals with airway hyperresponsiveness who do not have clinical asthma would prevent the full development of the disease.
This article has an online supplement, which is accessible from this issue's table of contents at www.atsjournals.org
REFERENCES
Kapsali T, Permutt S, Laube B, Scichilone N, Togias A. The potent bronchoprotective effect of deep inspiration and its absence in asthma. J Appl Physiol 2000;89:711eC720.
Scichilone N, Kapsali T, Permutt S, Togias A. Deep inspirationeCinduced bronchoprotection is stronger than bronchodilation. Am J Respir Crit Care Med 2000;162:910eC916.
Scichilone N, Permutt S, Togias A. The lack of the bronchoprotective and not the bronchodilatory ability of deep inspiration is associated with airways hyperresponsiveness. Am J Respir Crit Care Med 2001;163:413eC419.
King G, Moore B, Seow C, Pare P. Airway narrowing associated with inhibition of deep inspiration during methacholine inhalation in asthmatic. Am J Respir Crit Care Med 2001;164:216eC218.
Crimi E, Pellegrino R, Milanese M, Brusasco V. Deep breaths, methacholine and airway narrowing in healthy and mild asthmatic subjects. J Appl Physiol 2002;93:1384eC1390.
Jensen A, Atileh H, Suki B, Ingenito E, Lutchen K. Airway caliber in healthy and asthmatic subjects: effects of bronchial challenge and deep inspirations: signal transduction in smooth muscle [selected contribution]. J Appl Physiol 2001;91:506eC515.
Sundblad BM, Larsson K. Effect of deep inhalations after a bronchial methacholine provocation in asthmatic and non-asthmatic subjects. Respir Med 2002;96:477eC481.
Currie G, Fowler S, Lipworth B. Dose response of inhaled corticosteroids on bronchial hyperresponsiveness: a meta-analysis. Ann Allergy Asthma Immunol 2003;90:194eC198.
National Asthma Education and Prevention Program. Guidelines for the diagnosis and management of asthma: update on selected topics. Bethesda, MD: U.S. Department of Health and Human Services; 2002.
Chai H, Farr R, Froehlich L, Mathison D, McLean J, Rosenthal R, Sheffer A, Spector S, Townley R. Standardization of bronchial inhalation challenge procedures. J Allergy Clin Immunol 1975;56:323eC327.
Anderlind C, Kuroda N, Permutt S, Togias A. Validation of inspiratory vital capacity (IVC) as an outcome of single dose bronchoprovocation when deep inspiration (DI) is prohibited. Am J Respir Crit Care Med 2001;163:A829.
Scichilone N, Marchese R, Catalano F, Vignola AM, Togias A, Bellia V. The bronchodilatory effect of deep inspiration is absent in subjects with COPD. Chest 2004;125:2029eC2035.
Scichilone N, Marchese R, Catalano F, Togias A, Vignola AM, Bellia V. The bronchodilatory effect of deep inspiration diminishes with aging. Respir Med 2004;98:838eC843.(Nicola Scichilone, Solber)
Istituto di Medicina Generale e Pneumologia, Cattedra di Malattie dell'Apparato Respiratorio, University of Palermo, Palermo, Italy
ABSTRACT
Deep inspirationeCinduced bronchoprotection and bronchodilation are impaired in asthma. We evaluated the effect of inhaled glucocorticosteroids on these phenomena. Two groups of subjects with asthma, 9 with moderate/severe hyperresponsiveness to methacholine, and 12 with mild/borderline hyperresponsiveness to methacholine, received inhaled fluticasone (880 e daily) for 12 weeks. Serial bronchoprovocations were performed at Weeks 0, 6, and 12. The impact of deep inspirations on the airway response to methacholine was evaluated on the basis of inspiratory vital capacity and FEV1. Fluticasone produced a wide spectrum of changes in the beneficial effects of deep inspiration, but the mean changes were not significant. The magnitude of the steroid-induced changes in bronchoprotection by deep inspiration correlated with baseline log PC20 (the provocative concentration of methacholine causing a 20% fall in FEV1; higher log PC20 predicted improvement of the deep inspiration effect). The steroid-induced changes led to the emergence of strong positive correlations between the effects of deep inspiration and the methacholine log PC20 that did not exist at baseline. We conclude that deep inspirationeCinduced bronchoprotection can be restored by inhaled glucocorticosteroids only in individuals with mild hyperresponsiveness. After steroid treatment, the beneficial effects of deep inspiration become significant determinants of the magnitude of airway hyperresponsiveness.
Key Words: airway hyperresponsiveness; asthma; deep inspiration; fluticasone; methacholine
We and others (1eC6) have shown that deep inspiration exerts two potent beneficial effects on the airways of healthy subjects: the bronchoprotective effect, that is, the ability of deep inspirations performed before the inhalation of a spasmogen to prevent bronchial constriction; and the bronchodilatory effect, where deep inspirations taken immediately after the induction of bronchoconstriction improve airflow. In our previous work (1, 3), we have also found that the bronchoprotective effect of deep inspiration is greatly reduced in subjects with airway hyperresponsiveness, even when it is not accompanied by the diagnosis of asthma, and have suggested that the lack of deep inspirationeCinduced bronchoprotection is one of the major underlying causes of the hyperresponsive state. On the other hand, the bronchodilatory effect of deep inspiration is less discriminatory between airway hyperresponsiveness and normal responsiveness (3, 4, 6, 7).
Glucocorticosteroids are capable of reducing airway hyperresponsiveness (8), suggesting that this condition is somehow associated with airway inflammation. If so, it is possible that the loss of deep inspirationeCinduced bronchoprotection is linked to airway inflammation. The current study was designed to indirectly test this hypothesis by assessing the effect of inhaled glucocorticosteroid treatment on deep inspirationeCinduced bronchoprotection in volunteers with asthma.
We have previously reported that, as a group, individuals with borderline to mild airway hyperresponsiveness to methacholine demonstrate loss of deep inspirationeCinduced bronchoprotection to the same extent as those with moderate to severe hyperresponsiveness (3). This has raised the possibility that bronchoprotection by deep inspiration is the first normal function to be lost in the development of hyperresponsiveness. If so, and if inhaled glucocorticosteroids were to improve this function, their effect may be more evident in individuals with mild to borderline hyperresponsiveness, as opposed to those with a more advanced condition. To test this hypothesis, the design of this study has included two groups of volunteers with asthma: a group with borderline to mild hyperresponsiveness and a group with moderate to severe hyperresponsiveness.
METHODS
Subjects
We studied 21 subjects with asthma. Inclusion/exclusion criteria were identical to those employed in a previous study (3). All but one subject were skin-test positive to at least one aeroallergen. All subjects had intermittent or mild persistent asthma (9) and none had received inhaled or oral corticosteroids. Short-acting -agonists were withheld for at least 12 hours before each visit. For subjects who developed respiratory infections, bronchoprovocation sessions were postponed for at least 3 weeks. The study was designed to involve two groups, based on hyperresponsiveness to methacholine: (1) moderate to severe hyperresponsiveness (PC20 [provocative concentration of methacholine causing a 20% fall in FEV1] < 1 mg/ml, n = 9) and (2) mild to borderline hyperresponsiveness (1 < PC20 < 25 mg/ml, n = 10). We also entered two subjects with PC20 values of 44.7 and 45.4 mg/ml, respectively, but with a clinical history of asthma. The study was approved by the Johns Hopkins Bayview Institutional Review Board, and all subjects gave written, informed consent.
Study Design
The entire study consisted of two phases (Figure 1).
Phase 1. Each subject underwent a routine methacholine bronchoprovocation, using approximately half-logeCincreasing doses from 0.025 to 75 mg/ml (10), to assess the PC20. Thereafter, and on separate visits, a series of modified, single-dose bronchoprovocations (starting from 0.075 mg/ml) was performed as previously described (2) to determine (1) the single dose of methacholine inducing an at least 20% fall in FEV1 in the absence of deep inspirations (Figure 2a), (2) the ability of deep inspiration to reverse bronchoconstriction, expressed as percentage of bronchodilation for inspiratory vital capacity (IVC) and FEV1 (Figure 2b), and (3) the ability of deep inspiration to prevent bronchoconstriction, expressed as percentage of bronchoprotection for IVC and FEV1 (Figure 2c). This methodology is described in the online supplement.
Phase 2. All subjects received inhaled fluticasone (220 e, two inhalations twice daily) through a metered-dose inhaler and a spacer (AeroChamber; Trudell Medical International, London, ON, Canada), for 12 to 14 weeks. Each subject visited the laboratory every 2 weeks. At 6 weeks, and at the end of 12 weeks, the series of methacholine bronchoprovocations performed in Phase 1 were repeated.
Data Analysis
The log PC20, the percentage of bronchoprotection and the percentage of bronchodilation by deep inspirations, and the log of the single-dose methacholine required to reduce FEV1 by more than 20% in the absence of deep inspirations represented the outcomes of this study. To calculate bronchoprotection and bronchodilation, we employed the IVC as primary measure. The advantage is that the maneuver to determine IVC (from end-tidal inspiration to residual volume and then to total lung capacity) is free of the potential bronchodilatory effect of the deep inspiration that is involved in the measurement of FEV1 (11eC13). Treatment-induced changes from baseline over time were analyzed by repeated measures one-way analysis of variance (ANOVA) or by Friedman ANOVA, followed by appropriate post hoc tests. Linear regressions or the Spearman rank correlation were used to examine the relationship between conventionally measured airway responsiveness (expressed as log PC20) and deep inspirationeCinduced bronchoprotection and bronchodilation. Nonparametric statistics were used for post-steroid methacholine responsiveness values because, in five subjects who did not reach a PC20 after treatment, this parameter was assigned an arbitrary value (225 mg/ml). Significance was assumed for two-tailed values of p 0.05.
RESULTS
Phase 1
The demographic and lung function characteristics of the study participants are shown in Table 1. The two subjects who had low responsiveness at baseline were analyzed together with the "mild to borderline hyperresponsiveness" group. The primary analyses were conducted with and without these two subjects, with no qualitative impact on the outcome. The two study groups did not differ in terms of age (unpaired t test, p = 0.70), FEV1 % predicted (unpaired t test, p = 0.44), FVC % predicted (unpaired t test, p = 0.17), or FEV1/FVC (unpaired t test, p = 0.74). Also, the two groups did not differ with respect to the duration of asthma (23 ± 3.9 years in the mild to borderline hyperresponsiveness group vs. 28 ± 2.6 years in the moderate to severe hyperresponsiveness group; p = 0.33). By design, the magnitude of airway hyperresponsiveness, expressed as the PC20, was 15-fold different between the two subject groups (0.35 vs. 5.25 mg/ml, geometric mean). Table 2 presents information on the effects of the single-dose methacholine challenges. Notably, we met the target of inducing similar reductions in lung function with single doses of methacholine in the two groups. The single dose of methacholine employed to achieve the targeted change in lung function from baseline in the protocol devoid of deep inspirations was twofold lower in those subjects with moderate to severe hyperresponsiveness (0.52 mg/ml, geometric mean) than in those with mild to borderline hyperresponsiveness (1.02 mg/ml), the difference not being statistically significant. When all subjects were pooled into a single group, there was a significant correlation between the log PC20 to the conventional methacholine bronchoprovocation and the log of the single-dose methacholine required to reduce lung function in the absence of deep inspirations (linear regression, r2 = 0.21, p = 0.04).
As shown in Table 2, the bronchoprotective effect of deep inspiration as measured by IVC or FEV1 did not differ between the two groups. Also, this effect was not significantly different from zero (for IVC: one-sample t test, p = 0.14 and p = 0.40, moderate/severe and mild/borderline hyperresponsiveness, respectively; for FEV1: p = 0.07 and p = 0.20), supportive of our previous reports of loss of the effectiveness of deep inspirations to prevent bronchoconstriction in individuals with even borderline airway hyperresponsiveness (1, 3). The bronchoprotective ability of deep inspiration did not correlate with the degree of airway hyperresponsiveness as measured by the log PC20 (simple regression, r2 = 0.00, p = 0.99).
As we and others have previously described (3, 4, 6, 7), statistically significant bronchodilation by deep inspiration was present in both groups, as shown in Table 2, whether IVC or FEV1 was considered. No significant difference in the bronchodilatory effect of deep inspiration was found between the two groups. As with deep inspirationeCinduced bronchoprotection, when we assessed the relationship between the bronchodilatory ability of deep inspiration and the baseline magnitude of airway hyperresponsiveness (methacholine log PC20), no significant correlation was found (simple regression, r2 = 0.02, p = 0.53).
Phase 2
Twenty of 21 subjects completed the study. One subject (Subject 19) withdrew at 6 weeks of therapy for reasons not related to the study. During the study, four subjects experienced mild asthma exacerbations, presumably the result of respiratory viral infections; consequently, their next assessment was postponed for 3 weeks while continuing their inhaled corticosteroid use. None of the volunteers reported any adverse events associated with inhalation of the corticosteroids. No adverse events were recorded after any methacholine bronchoprovocation.
Treatment with inhaled fluticasone resulted in a small, but not significant, improvement in FEV1, and no improvement in FVC, when the two groups were analyzed separately. The increase in FEV1/FVC reached statistical significance in the mild to borderline hyperresponsiveness group (repeated measures ANOVA, p = 0.04). When the groups were combined, post-treatment FEV1 was significantly higher than the pretreatment value (repeated measures ANOVA, p = 0.04).
Fluticasone reduced airway responsiveness in both groups: significant changes in methacholine log PC20 occurred at Week 6 and improvement continued between Weeks 6 and 12. Despite the fact that the two groups started treatment with a more than 1-log difference in methacholine PC20, no significant difference was observed at Week 12 (Mann-Whitney U test, p = 0.28; Figure 3). Notably, when the conventional bronchoprovocation challenge was performed at the end of the study, five subjects had a less than 20% fall in FEV1 with the highest concentration of methacholine (75 mg/ml). Two of these five subjects belonged to the group with moderate to severe hyperresponsiveness, which indicates that the improvement in their PC20 was on the order of 2 logs. The PC20 value for these five subjects was arbitrarily determined as 225 mg/ml, a value that is approximately 0.5 log higher than the highest dose of methacholine administered (75 mg/ml).
Significantly higher single doses of methacholine had to be administered at 12 weeks of treatment to induce the targeted bronchoconstriction in the protocol devoid of deep inspirations in both groups (Figure 4). The change in the log of the single-dose methacholine for the mild to borderline hyperresponsiveness group was 0.79 (Friedman ANOVA, p = 0.01); for the moderate to severe hyperresponsiveness group, it was 0.76 (Friedman ANOVA, p = 0.02). Remarkably, the steroid-induced change in log PC20 from the conventional challenge was in absolute correlation with the change in log of the single-dose methacholine challenge, which was conducted in the absence of deep inspirations (r2 = 1.0).
The reduction in lung function in the single-dose methacholine challenges that were performed in the absence of deep inspirations at Weeks 6 and 12 was similar to that obtained at baseline (for the two subject groups combined, reduction in IVC: 26.3 vs. 27.9 and 27.6%, baseline vs. 6 and 12 weeks, repeated measured ANOVA, p = 0.81; reduction in FEV1: 33.0 vs. 34.5 and 34.8%, repeated measured ANOVA, p = 0.82). This was also true for each group analyzed separately, allowing for a direct comparison of the bronchoprotective and the bronchodilatory effects of deep inspiration between baseline and 6 and/or 12 weeks. The effect of fluticasone on bronchoprotection by deep inspirations is depicted in Figure 5 (top). We observed some improvement, albeit not statistically significant, in the ability of deep inspirations to protect airways from constriction in the group with mild to borderline hyperresponsiveness. Similar results were obtained when FEV1 was used as the spirometric outcome to calculate the bronchoprotective effect of deep inspiration. No significant difference was found between the two groups when the fluticasone-induced changes in the percentage of bronchoprotection IVC and FEV1 were compared.
Five subjects showed impressive improvement in the bronchoprotective effect of deep inspirations, reaching almost complete bronchoprotection (> 85%). Some of these subjects had dramatically reduced their airway responsiveness, as measured by conventional PC20. Yet we found no overall relationship between the change in log PC20 (log PC20, baseline to the end of treatment) and the change in percentage of bronchoprotection IVC induced by inhaled fluticasone (Spearman rank correlation, rs = 0.09, p = 0.69). On the other hand, all of these five subjects belonged in the group with mild to borderline hyperresponsiveness. Therefore, we analyzed the relationship between the baseline log PC20 and the change in percentage of bronchoprotection. This analysis yielded significant results, whether bronchoprotection by deep inspiration was calculated with IVC or FEV1 (simple regression, r2 = 0.31, p = 0.01, for percentage of bronchoprotection IVC; and r2 = 0.36, p = 0.005, for percentage of bronchoprotection FEV1; Figure 6a).
Fluticasone treatment did not have any effect on the bronchodilatory function of deep inspiration, as measured by IVC or FEV1, in either study group, or in the single-group analysis (IVC data in Figure 5, bottom). In contrast to our findings with deep inspirationeCinduced bronchoprotection, no relationship was found between the degree of baseline responsiveness and the changes in deep inspirationeCinduced bronchodilation after steroid treatment (Figure 6b).
In Phase 1 of the study (presteroid treatment), we found no relationship between the bronchoprotective effect of deep inspiration and airway responsiveness, as measured by the methacholine PC20. We reassessed the relationship between these two outcomes at the 12-week evaluation time point. At the end of the study, the percentage of bronchoprotection IVC correlated significantly with the log PC20 when the subject population was treated as a single group (Spearman rank correlation, rs = 0.58, p = 0.01; Figure 7). Analysis within each subgroup yielded a significant relationship only in the mild to borderline hyperresponsiveness group (Spearman rank correlation, rs = 0.73, p = 0.01). Similar findings were obtained when correlation analyses were performed with percent bronchoprotection FEV1 as the independent variable. We also performed these analyses with the bronchodilatory effect of deep inspiration as the independent outcome. For deep inspirationeCinduced bronchodilation, the percentage of bronchodilation IVC after steroid treatment correlated significantly with the methacholine log PC20 in the single-population analysis after steroid treatment (Spearman rank correlation, rs = 0.57, p = 0.01). The significant baseline relationship between the log PC20 of the conventional methacholine challenge and the log of the single-dose methacholine administered in the absence of deep inspirations remained virtually unchanged after fluticasone treatment compared with before treatment (rs = 0.58, p = 0.01, and rs = 0.47, p = 0.04, respectively).
DISCUSSION
The purpose of this study was to examine whether impairment of the beneficial effects of deep inspiration in individuals with asthma can be reversed by inhaled glucocorticosteroid treatment. The study was more focused on the bronchoprotective effect of deep inspiration, because the loss of this effect is strongly linked to airway hyperresponsiveness (3). The study was also designed to assess whether recovery of deep inspirationeCinduced bronchoprotection by glucocorticosteroids can occur more effectively in individuals with mild hyperresponsiveness, compared with those with moderate or severe hyperresponsiveness.
We generated a number of important observations. Twelve weeks of treatment with a moderately high dose of fluticasone did not produce statistically significant improvement in deep inspirationeCinduced bronchoprotection in either subject group or when the two groups were pooled for analysis (Figure 5). However, a few subjects had a remarkable improvement in bronchoprotection. This was not a random event, because these subjects were among those with the mildest degree of baseline hyperresponsiveness (Figure 6a). Also, this was not a regression to the mean phenomenon, because the relationship between the bronchoprotective effect of deep inspiration at baseline and the fluticasone-induced change in that effect was not statistically significant (data not shown). Equally interesting is the observation that individuals with the highest degree of baseline hyperresponsiveness showed an overall worsening of the protective effect of deep inspiration (Figure 5). Another important finding is that treatment with fluticasone converted an absent effect of deep inspirationeCinduced bronchoprotection and bronchodilation on airway hyperresponsiveness (as measured by the conventional log PC20) to a statistically significant effect (Figure 7). Finally, we found that inhaled glucocorticosteroid treatment resulted in significant improvement in airway responsiveness, when assessed by the single-dose methacholine bronchoprovocation technique, in the absence of deep inspirations (Figure 4).
It appears that bronchoprotection by deep inspiration is lost at an early stage of the inflammatory airway syndrome to which asthma belongs. We previously reported that this phenomenon is absent in those subjects with asthma with borderline to mild hyperresponsiveness, but also in individuals with allergic rhinitis and airway hyperresponsiveness (3). We are now showing that, at that early stage, the protective effect of deep inspiration can improve to normal levels with antiinflammatory treatment. In contrast, when severe hyperresponsiveness is present, the bronchoprotective effect of deep inspirations is clearly not improved with inhaled glucocorticosteroid treatment. Perhaps the loss of bronchoprotection is a permanent phenomenon at this stage, or perhaps treatment was not given for a long enough period of time. Although we did not measure markers of airway inflammation in this study, the remarkable improvement in the conventional PC20, even in the group of subjects with moderate to severe hyperresponsiveness, offers strong evidence that fluticasone exerted the expected antiinflammatory effects. Overall, these data suggest that the reduced protective effect of deep inspiration is an acquired phenomenon, possibly related to airway inflammation, and does not reflect a preexisting condition, as we had inferred in a previous publication (3).
By improving deep inspirationeCinduced bronchoprotection in subjects with mild hyperresponsiveness while worsening it in subjects with severe hyperresponsiveness, inhaled glucocorticosteroids led to the emergence of a significant relationship between the log PC20 to methacholine and the protective effect of deep inspiration (Figure 6). The absence of such a relationship at baseline was predictable from our previous work, where we have found bronchoprotection by deep inspiration equally lacking in asthmatic subjects with severe and with mild airway hyperresponsiveness (3). In the current study, not only did deep inspirationeCinduced bronchoprotection at the end of treatment correlate with airway responsiveness at the same time point but it also correlated equally well with airway responsiveness at the beginning of the study. It is reasonable to hypothesize that, at baseline, other active factors, presumably of an inflammatory nature, masked any relationship between the effect of deep inspiration and the log PC20. The impact of these factors may have been reduced or eliminated by the fluticasone treatment. The nature of such factors remains unknown at this point. Of note, our data suggest that, after removing the corticosteroid-sensitive elements of airway inflammation, a better understanding of the role of deep inspiration on airway responsiveness may be attained.
The design of this study allowed for tabulation of the dose of methacholine, which, when administered in the absence of deep inspirations, reduces lung function to a targeted degree. That dose is an outcome that reflects the component of airway responsiveness that is free of the effect of deep inspiration. Perhaps this is a true measure of airway smooth muscle reactivity, although other influencing elements (e.g., thickness of airway wall) cannot be excluded. In the absence of deep breaths, compared with baseline, significantly higher doses of methacholine (3.3- and 4.2-fold for subjects with asthma with moderate to severe hyperresponsiveness and for those with mild to borderline hyperresponsiveness, respectively) were required at the end of glucocorticosteroid treatment to induce the targeted reduction in lung function. Given the fact that glucorticosteroid treatment did not produce major overall improvement in the beneficial effects of deep inspiration, the primary contribution to the steroid-induced improvement in the methacholine PC20 can be assigned to this effect. Further dissection of the response to methacholine in the absence of lung inflation may lead to better understanding of the exact mechanism through which inhaled glucocorticosteroids exert their beneficial effects in asthma.
In conclusion, we found that inhaled glucocorticosteroids had a profound effect on the relationship between airway hyperresponsiveness and the beneficial effects of deep inspiration. Our data also suggest that deep inspirationeCinduced bronchoprotection is lost early in the pathologic process of asthma, at which point it may be recoverable. In more advanced stages, the loss of the beneficial effects of lung inflation is clearly resistant to glucocorticosteroids. Overall, early intervention with inhaled corticosteroids is advisable. Future studies should aim at addressing whether such treatment applied to individuals with airway hyperresponsiveness who do not have clinical asthma would prevent the full development of the disease.
This article has an online supplement, which is accessible from this issue's table of contents at www.atsjournals.org
REFERENCES
Kapsali T, Permutt S, Laube B, Scichilone N, Togias A. The potent bronchoprotective effect of deep inspiration and its absence in asthma. J Appl Physiol 2000;89:711eC720.
Scichilone N, Kapsali T, Permutt S, Togias A. Deep inspirationeCinduced bronchoprotection is stronger than bronchodilation. Am J Respir Crit Care Med 2000;162:910eC916.
Scichilone N, Permutt S, Togias A. The lack of the bronchoprotective and not the bronchodilatory ability of deep inspiration is associated with airways hyperresponsiveness. Am J Respir Crit Care Med 2001;163:413eC419.
King G, Moore B, Seow C, Pare P. Airway narrowing associated with inhibition of deep inspiration during methacholine inhalation in asthmatic. Am J Respir Crit Care Med 2001;164:216eC218.
Crimi E, Pellegrino R, Milanese M, Brusasco V. Deep breaths, methacholine and airway narrowing in healthy and mild asthmatic subjects. J Appl Physiol 2002;93:1384eC1390.
Jensen A, Atileh H, Suki B, Ingenito E, Lutchen K. Airway caliber in healthy and asthmatic subjects: effects of bronchial challenge and deep inspirations: signal transduction in smooth muscle [selected contribution]. J Appl Physiol 2001;91:506eC515.
Sundblad BM, Larsson K. Effect of deep inhalations after a bronchial methacholine provocation in asthmatic and non-asthmatic subjects. Respir Med 2002;96:477eC481.
Currie G, Fowler S, Lipworth B. Dose response of inhaled corticosteroids on bronchial hyperresponsiveness: a meta-analysis. Ann Allergy Asthma Immunol 2003;90:194eC198.
National Asthma Education and Prevention Program. Guidelines for the diagnosis and management of asthma: update on selected topics. Bethesda, MD: U.S. Department of Health and Human Services; 2002.
Chai H, Farr R, Froehlich L, Mathison D, McLean J, Rosenthal R, Sheffer A, Spector S, Townley R. Standardization of bronchial inhalation challenge procedures. J Allergy Clin Immunol 1975;56:323eC327.
Anderlind C, Kuroda N, Permutt S, Togias A. Validation of inspiratory vital capacity (IVC) as an outcome of single dose bronchoprovocation when deep inspiration (DI) is prohibited. Am J Respir Crit Care Med 2001;163:A829.
Scichilone N, Marchese R, Catalano F, Vignola AM, Togias A, Bellia V. The bronchodilatory effect of deep inspiration is absent in subjects with COPD. Chest 2004;125:2029eC2035.
Scichilone N, Marchese R, Catalano F, Togias A, Vignola AM, Bellia V. The bronchodilatory effect of deep inspiration diminishes with aging. Respir Med 2004;98:838eC843.(Nicola Scichilone, Solber)