© 2002 American Thoracic Society
Improvement in Bronchial Hyperresponsiveness with Inhaled Corticosteroids in Children with AsthmaImportance of Family History of Bronchial HyperresponsivenessDepartment of Pediatrics and Clinical Research Institute, Seoul National University Hospital; and Department of Pediatrics, Inje University Sanggye Paik Hospital, Seoul, Republic of Korea Correspondence and requests for reprints should be addressed to Dr. Young Yull Koh, Department of Pediatrics, Seoul National University Hospital, 28 Yongon-dong, Chongno-gu, Seoul 110-744, Republic of Korea. E-mail: kohyy{at}plaza.snu.ac.kr
The extent of improvement in bronchial hyperresponsiveness (BHR) with corticosteroids varies considerably among patients with asthma, although predictive factors for improvement are largely unknown. We tested the hypothesis that the improvement may vary according to family history of BHR. Children with atopic asthma (n = 121) received inhaled budesonide (800 µg per day) regularly for 6 months. Methacholine provocative concentration causing a 20% fall in FEV1 was measured before treatment and again after 3 and 6 months of treatment. A methacholine challenge test was also performed in each patient's parents, and the results were analyzed with regard to their children's response to corticosteroid therapy. When the children were classified into large (n = 40) and small (n = 40) improvement groups after 6 months of treatment, the prevalence of BHR and the bronchial responsiveness index were higher in parents of the small improvement group (28.8%, 1.145 ± 0.104) than in parents of the large improvement group (6.3%, 1.095 ± 0.064; both, p < 0.01). The magnitude of improvement in BHR at 6 months was lower in children with at least one parent with BHR (n = 45; 1.666 ± 1.244 doubling doses) than in children with non-BHR parents (n = 76; 2.531 ± 1.726, p < 0.01). Our results suggest that a family history of BHR may be an important factor in the sensitivity of BHR of individuals with asthma to inhaled corticosteroids.
Key Words: bronchial hyperreactivity • corticosteroids • familial predisposition
Bronchial hyperresponsiveness (BHR) is a characteristic feature of asthma (1), and its degree correlates with asthma severity and the need for therapy (2). Despite extensive investigation, the pathogenesis of BHR remains unclear, although it is now becoming evident that inflammation of the airway may play an important role (3). Corticosteroids are currently the most effective antiinflammatory drugs available for the treatment of asthma (4). Several clinical studies have shown that BHR is reduced by prolonged treatment with inhaled corticosteroids (5, 6), and the reduction in BHR has been proposed as a potential consequence of corticosteroid action, contributing to its beneficial effect in the treatment of asthma. The effects of inhaled corticosteroids on bronchial reactivity become apparent after several weeks of therapy, and those effects are both dose- and time-dependent (7, 8). However, the extent of improvement in BHR with corticosteroids varies considerably among patients (6, 7, 9). Some patients show a dramatic improvement, even a return to normal levels of bronchial responsiveness, whereas others remain hyperresponsive with little change. Several studies have examined various asthma features that might be important in predicting the magnitude of response (5, 10). However, the factors predictive of improvement in BHR with inhaled corticosteroids are largely unknown. The fact that BHR is not always responsive to corticosteroids may reflect the contribution of two components: corticosteroid-responsive hyperreactivity is related to airway inflammation and corticosteroid-unresponsive hyperreactivity pertains to the underlying BHR (11). The mechanism(s) of this component of BHR remains unknown. There are, however, a number of arguments suggesting that the underlying BHR may differ from individual to individual and that genetic factors may control this underlying hyperresponsiveness (12, 13). We hypothesized that the extent to which corticosteroids reduce BHR may be controlled by familial factors. Specifically, we desired to test the hypothesis that the improvement in BHR with corticosteroid therapy varies according to the family history of BHR. To test the hypothesis, we performed a methacholine challenge test in the parents of children with asthma whose BHR had been assessed in response to corticosteroid therapy, and we analyzed the results with regard to their children's response to corticosteroids.
Children with atopic asthma and their parents were targeted for this study. Asthma was diagnosed according to the criteria of the ATS (14), and atopy was defined as at least one positive skin prick test among a panel of 12 common aeroallergens. All patients had mild to moderate asthmatic symptoms, which were controlled by an as-needed inhaled bronchodilator and the regular use of prophylactic medications (inhaled nedocromil sodium or inhaled corticosteroid). The study commenced with a run-in period of 6 weeks, during which prophylactic medications were tapered off and discontinued 4 weeks before the baseline measurements (spirometry and methacholine provocation test). To be eligible for the study, FEV1 had to be 70% or more of the predicted value (15) and the patients were required to have a PC20 methacholine below 8 mg/ml. After the run-in period, the patients received inhaled budesonide in a dose of 400 µg twice a day via a multidose dry powder inhaler (Turbuhaler; Astra Draco, Lund, Sweden). After 3 and 6 months of treatment, the patients underwent spirometry and a methacholine inhalation provocation test. The study was performed during the months of December–July to avoid the autumn season, in which symptoms of house dust mite atopy are most prominent (16). Both natural parents of the children with asthma underwent spirometry and a bronchial challenge test with methacholine. Parents who had been diagnosed as having asthma were included in the study, but parents with FEV1 levels lower than 70% of the predicted value (17) were excluded. Methacholine bronchial challenges were performed using a modification of the method described by Chai and colleagues (18). Methacholine (Sigma Chemical, St. Louis, MO) at concentrations of 0.075, 0.15, 0.3, 0.625, 1.25, 2.5, 5, 10, and 25 mg/ml was prepared by dilution in buffered saline (pH 7.4). For the follow-up measurements in the children, two more concentrations (50 and 100 mg/ml) were added. Methacholine PC20 was calculated by interpolation between two adjacent data points if FEV1 fell by more than 20%. As most parents did not experience a 20% drop in FEV1 after inhalation up to the highest concentration of methacholine (25 mg/ml), PC20 calculation was made impossible. Therefore, the log of the slope of the percentage decline from baseline FEV1 after the last dose of methacholine, per unit concentration of methacholine (bronchial responsiveness index: BR index), was calculated for all the parents, as described by Burrows and colleagues (19).
Statistical Analysis
One hundred thirty-six children entered the study. Nine children were unable to complete the study protocol: four due to failure to perform the tests within 4 weeks of a scheduled visit because of viral respiratory infection or oral steroid use, three due to poor compliance with inhaled budesonide, and two due to loss of follow-up. Among the parents of the 127 children who completed the budesonide study, three pairs of parents themselves declined to participate in the study, and either member of another three pairs could not perform the tests due to poor technique of forced expiratory maneuver or lower FEV1 level. Complete data were therefore available for 121 children and their parents (Table 1) .
Table 2 summarizes the levels of FEV1 and methacholine PC20 before and after budesonide treatment in the children. The mean FEV1 values showed a small but significant improvement during the first 3 months of treatment, but no further improvement was observed during the following 3 months. Progressive increases in methacholine PC20 were noted after 3 and 6 months of treatment (comparisons with pretreatment value, both p < 0.01; comparison between 3- and 6-month values, p < 0.01). Changes in methacholine PC20 with respect to the pretreatment value, expressed as doubling doses, were 1.510 ± 1.237 (mean ± SD) and 2.210 ± 1.614 after 3 and 6 months of treatment, respectively. The magnitude of improvement showed wide variation. At 3 months of treatment, the doubling dose ranged from -1.227 to 4.560, with 10 subjects displaying greater than a 10-fold improvement in PC20 and 15 showing a deterioration. At 6 months of treatment, the doubling dose ranged from -2.322 to 6.001, with 29 subjects displaying greater than a 10-fold improvement and six still showing a deterioration.
Among the patients, the large and small improvement groups were defined as those whose response fell within the upper one-third and the lower one-third, respectively, in the rank of doubling doses. Table 3 summarizes the clinical characteristics of patients classified as large and small improvement groups at two time-points. No significant differences with respect to sex, age, duration of illness, serum total IgE, skin test sensitivity, class, or duration of prophylactic medication, environmental factors, and pretreatment FEV1 or methacholine PC20 were observed between these two groups, after either 3 or 6 months of treatment.
Figure 1 shows the BR index and the prevalence of BHR in the parents segregated according to the classification of their children into either the small improvement or large improvement group after 3 months (left panel) and 6 months (right panel) of treatment, respectively. The parents of children with small improvement at 3 months of treatment had a BR index of 1.136 ± 0.099, which is not statistically different (p = 0.107) from the value (1.113 ± 0.083) of parents of children with large improvement. The prevalence of BHR was 25% (20/80) in the former group of parents and 15% (12/80) in the latter group of parents but this difference was not significant (p = 0.114). When the parents were segregated according to the children's response at 6 months, the BR index was significantly higher among the parents of children with small improvement (1.145 ± 0.104) than the parents of children with large improvement (1.095 ± 0.064; p = 0.001). The index was not significantly different between the two father groups (1.120 ± 0.095 versus 1.088 ± 0.056, p = 0.070), whereas it was significantly higher in mothers of children with small improvement (1.170 ± 0.107) than in mothers of children with large improvement (1.102 ± 0.070; p = 0.001). Thus, the maternal degree of bronchial responsiveness seemed to be more important in predicting a child's response to inhaled corticosteroids than the paternal degree of bronchial responsiveness. The prevalence of BHR was also higher among the parents of children with small improvement after 6 months of treatment (28.8%, 23/80) than among the parents of children with large improvement (6.3%, 5/80; p = 0.001). The difference was significant both in fathers (10/40 versus 2/40, p = 0.012) and in mothers (13/40 versus 3/40, p = 0.005).
Parental data were also analyzed after categorizing the children on the basis of two doubling doses, which is considered a clinically important change (22), after 3 and 6 months of treatment (data not shown). Neither the BR index nor the prevalence of BHR was significantly different between parents (n = 154) of children with doubling doses less than two after 3 months of treatment and parents (n = 88) of children with doubling doses of two or more. When the parents were divided according to their children's response at 6 months, the BR index was significantly higher in parents (n = 108) of children with doubling doses less than two, than parents (n = 134) of children with doubling doses of two or more (1.141 ± 0.10 versus 1.107 ± 0.08, p = 0.006). The prevalence of BHR was also higher in the former group of parents (26.9%, 29/108) than in the latter group (13.4%, 18/134; p = 0.009). Figure 2 shows the baseline methacholine PC20 levels (left panel) and changes in methacholine PC20, expressed as doubling doses (right panel), in the two groups of children divided according to whether or not their parents have BHR. The baseline methacholine PC20 was not significantly different between the children (n = 45) having one (n = 43) or both parents (n = 2) with BHR and those (n = 76) having non-BHR parents (1.40 [0.47–4.20] versus 1.63 [0.68–3.93], p = 0.433). After 3 months of treatment, the changes in PC20 were not significantly different between the children of BHR-positive parents and those of BHR-negative parents (1.303 ± 1.136 versus 1.633 ± 1.284 doubling doses, p = 0.144). At 6 months, however, the children of BHR-positive parents showed a change of 1.666 ± 1.244 doubling doses, compared with 2.531 ± 1.726 doubling doses in the children of BHR-negative parents (p = 0.002). Similar figures were observed in the frequency of children classified as being in a small improvement group at the two time-points. At 3 months of treatment, 18 (40%) of the children of BHR-positive parents were in the small improvement group, compared with 22 (28.9%) of the children of BHR-negative parents (p = 0.212). At 6 months, however, 21 (46.7%) of the children of BHR-positive parents were in the small improvement group, compared with 19 (25%) of the children of BHR-negative parents (p = 0.034).
We have shown that treatment with inhaled budesonide for 6 months reduced bronchial responsiveness to methacholine, with large variations in the degree of reduction experienced by children with asthma. The parents of children with a small reduction were found to have a higher prevalence of BHR and a greater degree of bronchial responsiveness than the parents of children demonstrating a large reduction. Furthermore, comparisons between children with asthma of BHR-positive parents and those of BHR-negative parents showed that methacholine PC20 levels before the treatment were similar, but the changes in PC20 after 6 months of treatment were significantly lower in the former than in the latter group. In the present study, the average increase in PC20 after 3 months of treatment with budesonide was 1.51 doubling doses of methacholine, and after 6 months, it was 2.21 doubling doses. These degrees of improvement are in line with those observed in previous studies (5–8), i.e., of the order of one or two doubling doses, and are both dose and time dependent. To ensure the maximum therapeutic effect, we chose to administer a dose of inhaled budesonide (800 µg per day) that is slightly higher than that usually prescribed. It has been reported that marked improvement in BHR was already apparent after 3 months (5) or 6 months (23) of treatment, and the continuation of therapy resulted in only a slight, but not significant, further improvement. In light of these results, we opted to assess bronchial responsiveness up to 6 months of treatment, although it is possible that the responsiveness could have been reduced further by prolonged treatment. It would be of interest to continue the corticosteroid therapy until BHR reaches a plateau in each individual and to analyze parental data with regard to changes in the children's BHR. The mechanism underlying the efficacy of corticosteroids in controlling bronchial responsiveness presumably stems from their ability to control multiple components of airway inflammation (24). Some studies have shown that inhaled corticosteroids cause a decrease of BHR, which is paralleled by a decrease of eosinophils and other inflammatory cells in bronchial mucosa (25). However, the majority of studies have demonstrated a relative dissociation between the decrease of inflammatory cells and the decrease of BHR (26). For example, BHR can still be present after years of treatment with corticosteroids despite an absence of histologic evidence of inflammation in bronchial mucosal biopsies (27). Furthermore, the results of studies on the relationships between BHR and airway inflammation at the baseline are largely inconsistent (28, 29). It has been proposed that BHR in asthma is caused by the interaction between underlying BHR and airway changes due to the inflammatory process and that genetic factors might control the underlying BHR (12, 13). If these inferences are correct, the extent to which corticosteroids reduce BHR may be influenced by familial factors because the corticosteroid-unresponsive component of BHR could reflect an asthmatic's underlying BHR. To our knowledge, this is the first study in which bronchial responsiveness has been assessed in families of children with asthma in connection with the children's response to corticosteroids. Our results have shown that the BR index and the prevalence of BHR were significantly higher in parents of children with small improvement during treatment with corticosteroids than in parents of children with large improvement, when the children were grouped at 6 months of treatment. These observations suggest that the extent of improvement in BHR with corticosteroids in children with asthma depends on the degree of bronchial responsiveness in their parents. Methacholine PC20 measurements before the treatment were not significantly different between the children with asthma of BHR-positive parents and those of BHR-negative parents, but changes in methacholine PC20 after treatment were lower in the former than in the latter group, with the difference reaching statistical significance after 6 months of treatment. Furthermore, children of the BHR-positive parents were more frequently classified in the small improvement group after 6 months of treatment than were the children of the BHR-negative parents. These observations suggest that children with asthma, with a familial history of BHR have a larger component of BHR that is not ameliorated by corticosteroids than those without such a history. Studies of BHR in parents and nuclear families of subjects with asthma have unanimously agreed that familial, possibly genetic, factors are involved in its genesis (30, 31). Our results suggest that the corticosteroid-unresponsive component of BHR rather than BHR per se in each individual may more closely reflect the contribution by familial factors. Apart from familial factors, we examined various asthma features that we believed might be important in predicting the magnitude of response. From clinical practice, we expected patients with relatively strong BHR to show a greater response to inhaled corticosteroids than patients who were less hyperresponsive. In our analysis, however, PC20 before the treatment did not seem predictive of improvement in PC20 after treatment with inhaled corticosteroids. This is consistent with other studies (6, 7). Kerstjens and colleagues (10), investigating a group of patients with asthma and chronic obstructive pulmonary disease, demonstrated that a higher initial IgE level predicts a more favorable course of BHR in patients who are treated with an inhaled corticosteroid. However, our study does not confirm their results. This may be explained in part by differences in patient characteristics and study design. There is evidence to suggest that asthma acts through a chronic inflammatory process, with an increased thickness of the airway wall and hypertrophy of airway smooth muscle (32). BHR may be the result of such airway remodeling and a component of BHR caused by this may not be influenced by corticosteroids (33). It would have been ideal to include only newly referred patients with asthma who were not already receiving inhaled corticosteroids, like Haahtela and coworkers (34) did, but we were unable to enroll enough new patients within the predetermined entry period. Although we cannot estimate to what degree airway remodeling contributes to BHR in our subjects, this factor is unlikely to have significantly affected our results because the duration of asthma or of maintenance prophylactic therapy was similar between the two groups (large improvement and small improvement) in our study. It may be argued that the small improvement with corticosteroids shown by children with high residual BHR and a high degree of bronchial responsiveness in their parents may be due to their sharing a common "highly allergic" environment. Presumably, children with a high residual BHR may have sustained airway inflammation, provoked by an ongoing environmental trigger, which may also have had an impact on parental BHR. A recent study has shown that rigorous allergen avoidance has additional benefit on clinical or inflammatory markers and BHR in patients with persistent asthma already being treated with inhaled corticosteroids (35). In the present study, although the environmental factors with respect to exposure to tobacco smoke, pets in the house, or residence location were not different between those with large and small improvement with corticosteroids, we cannot rule out the possibility that other unexplored environmental factors may have contributed to our finding. In conclusion, instituting inhaled corticosteroid therapy in children with asthma can provide significant improvement in bronchial responsiveness, with the magnitude of response showing wide variation. The extent of improvement in children was found to be dependent on the degree of bronchial responsiveness in their parents. Our results suggest that family history of BHR may be an important factor in the sensitivity of BHR of individuals with asthma to inhaled corticosteroids.
This study was supported by BK21 Human Life Sciences, Seoul National University.
This article has an online data supplement, which is accessible from this issue's table of contents online at www.atsjournals.org Received in original form January 31, 2001; accepted in final form April 24, 2002
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
INTRODUCTION
log10PC20 [log10(PC20 after the treatment) - log10(PC20 before the treatment)]/log102 (
