Pretreatment with Albuterol versus Montelukast for Exercise-Induced Bronchospasm in Children
Hengameh H. Raissy, PharmD; Michelle Harkins, MD; Franceska Kelly, BS; H. William Kelly, PharmD
Study Objectives: To compare pretreatment with albuterol versus montelukast added to the current asthma regimen for protection against exercise-induced bronchospasm in children with mild-to-moderate asthma, and to determine whether cysteinyl leukotriene (Cys-LT) concentrations measured in the exhaled breath condensate correlated with response to montelukast.
Design: Prospective, randomized, double-blind, double-dummy, crossover study.
Setting: Asthma clinic at a university-affiliated medical center.
Patients: Eleven children aged 7–17 years with physician-diagnosed mild-to-moderate asthma for at least 6 months and with self-reported exercise-induced bronchospasm (defined as ≥ 15% decrease in forced expiratory volume in 1 sec [FEV1] at screening and baseline visit).
Intervention: Patients were randomly assigned to receive 3–7 days of oral montelukast 5–10 mg/day or 2 puffs of an albuterol metered-dose inhaler just before an exercise challenge and then were crossed over to the alternate therapy for the last visit.
Measurements and Main Results: Serial spirometry was performed before and at 0, 5, 10, 15, 30, 45, and 60 minutes after the exercise challenge at each visit. Measurement of exhaled breath condensate was performed at the screening visit and study visits 1 and 2. The primary outcome was the maximum change in FEV1 after exercise. Secondary outcomes were the area under the curve for FEV1 (expressed as percentage decrease from baseline) during the first 60 minutes (AUC0–60) after exercise and the proportion of patients in whom exercise-induced bronchospasm was prevented (defined as < 15% decrease in FEV1 after exercise challenge). The mean ± SD maximum decrease in FEV1 was 27.5 ± 7.9% at baseline. Patients receiving montelukast had an 18.3 ± 13.7% decrease in FEV1 compared with 0.7 ± 1.6% in patients receiving albuterol (p=0.002, paired t test). Exercise-induced bronchospasm was prevented in 100% of the patients receiving albuterol compared with 55% receiving montelukast (p<0.05, McNemar’s test). The AUC0–60 was significantly smaller with albuterol compared with montelukast (p<0.001, Wilcoxon signed rank test). No correlations were found between Cys-LT concentration and the severity of exercise-induced bronchospasm or the response to montelukast.
Conclusion: Pretreatment with albuterol is more effective than montelukast for prevention of exercise-induced bronchospasm in children with asthma.
Exercise-induced bronchospasm is common in children with asthma,[1,2] and exercise limitation is a primary complaint in children with mild-tomoderate asthma. Although exercise-induced bronchospasm may be the only asthma symptom or some patients, it may result from an overall lack of asthma control.
Exercise-induced bronchospasm has a discrete pathophysiology that remains poorly understood. The cysteinyl leukotrienes (Cys-LTs) LTC4, LTD4, and LTE4 have been detected in exhaled breath condensate in children with asthma[3-5]and have been reported to be higher in those with exercise-induced bronchospasm.
Long-term controller drugs may have a role in the management of exercise-induced bronchospasm. Inhalation of corticosteroids has been shown to attenuate the exercise-induced bronchospasm response, although not completely abate it.[7,8] The long-acting β2-agonist salmeterol inhibits exercise-induced bronchospasm for 12 hours after a single dose; however, after long-term monotherapy the duration of protection is only about 4 hours.[9,10] Results of comparative studies of long-term montelukast and salmeterol monotherapy for exercise-induced bronchospasm, where the exercise-induced bronchospasm challenge is performed 12 hours after the last dose of montelukast or salmeterol, show that montelukast provides superior attenuation of the bronchospasm. Long-term treatment with montelukast reduces exercise-induced bronchospasm by 20-50%[11-13] but does not appear to prevent the bronchospasm to the same extent as pretreatment with an inhaled short-acting β2-agonist administered just before exercise, although no direct comparisons have been performed. In addition, up to 50% of patients may not respond to montelukast. Recently, it has been suggested that a reduction in the Cys-LTs correlates with initial Cys-LT concentrations before treatment with montelukast.[15,16]
In April 2007, montelukast gained United States Food and Drug Administration (FDA)-approved labeling as a single dose for prevention of exercise-induced bronchospasm in patients aged 15 years or older. Currently, to our knowledge, no data exist that compare standard of care (i.e., pretreatment with albuterol) with montelukast for prevention of exercise-induced bronchospasm in children. Therefore, our primary hypothesis in this study was that pretreatment with albuterol would provide superior protection against exercise-induced bronchospasm compared with montelukast in children with asthma. A secondary objective was to determine whether Cys-LT concentrations measured in the exhaled breath condensate correlated with response to montelukast.
This prospective, randomized, double-blind, double-dummy, crossover clinical trial was conducted from November 1, 2005-April 30, 2007. Patients aged 7-17 years with physiciandiagnosed asthma for at least 6 months in addition to self-reported exercise-induced bronchospasm were screened. Long-term controller drugs were allowed if patients were receiving a stable dosage for at least 4 weeks. Patients were excluded if they had a history of cardiac dysfunction, were unable to perform exercise challenge or spirometry, used montelukast for asthma management, had upper respiratory infection in the previous 4 weeks, or used oral corticosteroids in the previous 3 months. Informed consent and authorization for the use and disclosure of Protected Health Information were obtained according to institutional review board-approved processes of the University of New Mexico.
The study consisted of four visits: screening visit, baseline visit, study visit 1, and study visit 2 (Figure 1). Exercise-induced bronchospasm was assessed at the screening visit followed by a baseline visit in 1-14 days, when a second exercise challenge was performed. Patients were required to have a positive exercise challenge, defined as a 15% or greater decrease in forced expiratory volume in 1 second (FEV1), at both the screening and baseline visits to qualify. At the end of the baseline visit, eligible patients were randomly assigned to receive either montelukast capsules 5-10 mg (depending on age) or matching placebo capsules to be taken every night. Study visit 1 was scheduled 3-7 days later (Figure 1). At this visit, exhaled breath condensate was measured, and baseline spirometry was performed. Patients who had received montelukast were then instructed to use 2 puffs of a placebo metered-dose inhaler (MDI) without a spacer 15 minutes before the exercise challenge; patients who had received placebo used 2 puffs (90 µg/puff) of an albuterol MDI 15 minutes before the exercise challenge. At the end of study visit 1, patients were crossed over to the alternative therapy and scheduled for study visit 2.
Figure 1. (click image to zoom)Schematic of the study procedures.
To allow double blinding of the study drug, placebo and montelukast capsules were prepared from chewable tablets by the investigational pharmacist, based on experience from another study. The study drugs were dispensed for each subject according to a random code. Patients were instructed to take placebo or montelukast capsules at 9:00 P.M. every night and to bring all their unused study drugs to the clinic at study visits 1 and 2 in order to monitor adherence. All visits were scheduled for 7:30 A.M. (± 30 min), so the exercise challenge could be performed approximately 12 hours (± 30 min) after the last dose of montelukast.
Preexercise spirometry was performed 5 minutes before the challenge. The preexercise FEV1 was required to be at least 70% of the predicted value and within 20% of the baseline visit value at study visits 1 and 2. A standardized exercise challenge was performed on a treadmill. Workload was increased until 80-90% of the maximum heart rate (220 minus age) was achieved in the first 2 minutes, and exercise was sustained for 6 minutes. The individual workload established at enrollment for each patient was used as a starting point at each visit and was adjusted as needed to achieve 80-90% of the maximum heart rate.
Spirometry was performed immediately after exercise (time 0) and at 5, 10, 15, 20, 30, and 60 minutes. Patients performed two or three maneuvers at each testing time. A positive exercise challenge was defined as a decrease in FEV1 from the preexercise value by at least 15%. A standard spirometer (Sensormedics, Yorba Linda, CA) was used to perform spirometric maneuvers that achieved American Thoracic Society acceptability and reproducibility criteria, and the best FEV1 from each set of measurements was used for the analysis. Patients were discharged from the clinic when their FEV1 was at least within 5% of their baseline. After 60 minutes of exercise, if patients were still symptomatic or their FEV1 was not within 5% of baseline, 2-4 puffs of albuterol MDI were administered, and their FEV1 was measured in 15 minutes. Patients were instructed to withhold their short-acting β2-agonist and cromolyn for 6 hours and long-acting β2-agonist for 12 hours before the exercise challenge.
At the investigators’ discretion, the challenge could be stopped and albuterol could be administered. Patients were discontinued from the study if they were not compliant with their drug therapies, had an upper respiratory infection with worsening of asthma symptoms, required treatment with oral corticosteroids, or had a change in regular asthma drug therapy.
Collection of Exhaled Breath Condensate
Samples of exhaled breath condensate were collected at the beginning of study visits 1 and 2 with use of a disposable collection kit (RTube; Respiratory Research, Inc., Charlottesville, VA). Patients were asked to breathe tidally for 10 minutes with no nose clip. The samples were stored at -70°C until analyzed at the end of the study.
The Cys-LT concentrations were measured by Cayman Chemical (Ann Arbor, MI) using a specific enzyme immunoassay with a lower limit of detection of 7 pg/ml. Samples below the lower limit of detection were concentrated by a factor of 3-4 to determine their concentration. The exhaled breath condensate was collected and analyzed in all of the qualified patients and the last 11 patients with a negative exercise challenge.
The primary outcome was the maximum change in FEV1 after exercise. Secondary outcomes were the area under the curve for FEV1 (expressed as percentage decrease from baseline FEV1 before challenge on each day) in the first 60 minutes (AUC0-60) after exercise and the proportion of patients in whom exercise-induced bronchospasm was prevented. Only the area below the preexercise baseline FEV1 was used in calculating the AUC.
This study was designed to have a 90% power to detect at least 13.7% difference in the mean change for the maximum percentage decrease in FEV1 between albuterol and montelukast based on an estimated standard deviation of 25.0 and with a significance level (α) of 0.05 by using a two-sided one-sample t test and a correlation of 0.25.
Data are presented as mean ± SD. The clinical outcomes and Cys-LT levels were compared by using the paired t test and Wilcoxon signed rank test as appropriate. Correlations were evaluated by Spearman rank test between Cys-LT levels and severity of exercise-induced bronchospasm or the response to montelukast compared with baseline visit.
McNemar’s test was used to compare the percentage of patients in whom exercise-induced bronchospasm was prevented (defined as < 15% decrease in FEV1 after exercise) in each treatment group.
Ninety-one patients were recruited for the study. Thirteen patients were either lost to follow-up or decided to discontinue their participation for personal reasons (e.g., parents’ work, school work). Seventy-eight patients completed the screening visit, 13 of whom had exercise-induced bronchospasm and qualified for the study: 11 of these patients completed all the visits, and two were discontinued from the study (one patient had an episode of hyperventilation at study visit 1 after 3 minutes of running on the treadmill, and the other had upper respiratory infection requiring a course of prednisone after his screening visit). Baseline data and demographics for the 11 patients who completed the study are presented in Table 1 .
Adherence to the study drug was 100%. Throughout the study, no significant difference was seen in preexercise FEV1 between the baseline visit and study visits 1 and 2.
The mean ± SD maximum percentage decrease in FEV1 after exercise at baseline visit was 27.5 ± 7.9%. The maximum percentage decrease in FEV1 after exercise was reduced significantly with albuterol compared with montelukast (0.7 ± 1.6% vs 18.3 ± 13.7%, p=0.002; Table 2 ). The FEV1 AUC0-60 was significantly smaller with albuterol compared with montelukast (p<0.001; Figure 2). Exercise-induced bronchospasm was prevented in 100% of the patients receiving albuterol compared with 55% of patients receiving montelukast (p<0.05).
Figure 2. (click image to zoom)Changes in forced expiratory volume in 1 second (FEV1) after exercise challenge at baseline and after pretreatment with albuterol and montelukast.
Mean ± SD Cys-LT concentrations did not significantly differ between patients receiving montelukast and patients receiving placebo (3.16 ± 1.4 and 5.8 ± 4.2 pg/ml, respectively, p=0.09). Although patients with a negative exercise challenge had lower Cys-LT concentrations compared with those with exercise-induced bronchospasm, the difference was not statistically significant (2.72 ± 2.9 and 5.8 ± 4.2 pg/ml, respectively, p=0.08). No correlations were found between Cys-LT concentration and the severity of exercise-induced bronchospasm or the response to montelukast
We found that pretreatment with an albuterol MDI had greater efficacy than pretreatment with montelukast in the prevention of exercise-induced bronchospasm in children, as measured by maximum percentage decrease in FEV1. The FEV1 AUC0-60 after exercise challenge was also significantly smaller with albuterol compared with montelukast. Significantly more patients (100%) using pretreatment with albuterol had prevention of exercise-induced bronchospasm compared with only 55% of those taking montelukast. The results seen with montelukast are consistent with those of previous studies which found that montelukast attenuates exercise-induced bronchospasm in asthmatic children when compared with placebo or longacting β2-agonists.[11-13,19-22] Furthermore, our study provides evidence that the recommended standard pretreatment with albuterol is significantly more effective in blocking exercise-induced bronchospasm than is montelukast, supporting current guideline recommendations.
The magnitude of effect and the onset of action of montelukast on exercise-induced bronchospasm in children have been investigated. One group of authors suggested that treatment with montelukast in children aged 8-14 years can provide significant protection against exercise-induced bronchospasm.19 Mean ± SD maximum percentage decrease in FEV1 improved from 36.5 ± 10.2% at baseline to 27.6 ± 14.4% after 8 weeks of treatment with montelukast once/day (p<0.001). Other investigators reported that the exercise-induced bronchospasm protection was evident after 3 days of treatment with montelukast (maximum percentage decrease in FEV1 of 17.7% compared with 33.3% at baseline). These results were confirmed in a similar study in children aged 6-14 years. Montelukast was taken for 2 days, and an exercise challenge was performed 20-24 hours after the last dose of montelukast. Montelukast caused a significant reduction in maximum decrease in FEV1 (18 ± 13%) compared with placebo (26 ± 14%). The FEV1 AUC0-60 after exercise challenge was also significantly smaller with montelukast compared with placebo (265 ± 272%•min vs 590 ± 705%•min, respectively).
We designed our study to assess the protective effect of montelukast against exercise-induced bronchospasm in children when montelukast provided its maximum effect at 12 hours after administration. The onset and duration of effect on exercise-induced bronchospasm after one dose of montelukast were evaluated in children in a double-blind, randomized, placebo-controlled, crossover design. Children performed an exercise challenge 2, 12, and 24 hours after a single dose of montelukast. Only at 12 hours after dosing did montelukast produce a significant protection against exercise-induced bronchospasm compared with placebo, measured by mean ± SD maximum percentage decrease in FEV1 (9.78 ± 1.85% vs 18.69 ± 2.83%). A significant difference was seen in the percentage of patients receiving montelukast (50%) compared with placebo (31%), with exerciseinduced bronchospasm protection defined as less than 15% decrease in maximum FEV1 only at 12 hours after dosing (p=0.02).
In a similar study in adult patients with asthma, exercise-induced bronchospasm protection was measured at 2, 12, and 24 hours after a single dose of montelukast. Mean ± SD maximum percentage decrease in FEV1 was 10.8 ± 7.9%, 8.4 ± 7.5%, and 8.3 ± 7.3%, respectively, at 2, 12, and 24 hours after dosing, which was significantly smaller compare with placebo (22.3 ± 13.1%, 16.1 ± 10.2%, and 16.9 ± 11.7%, respectively). Maximum effect was evident at 12 hours after the dose and persisted up to 24 hours. The FEV1 AUC0-60 after exercise challenge was also significantly smaller with montelukast at 2, 12, and 24 hours, respectively, after the dose (182.46 ± 235.9%•min, 152.27 ± 179.5%•min, and 120.06 ± 119.8%•min) compared with placebo (702.13 ± 675%•min, 389.55 ± 317.1%•min, and 409.5 ± 426.6%•min). Percentages of patients who had a 10% or less decrease in FEV1 were 47%, 63%, and 67% in the montelukast group at each time point, respectively; these decreases in FEV1 were significantly lower than the decreases at each time point in 20%, 35%, and 33% of the patients receiving placebo.
In addition, we purposefully allowed patients to continue taking whatever β2-agonists they were prescribed, as previous studies demonstrated increased sensitivity to exercise and decreased protection against exercise-induced bronchospasm from β2-agonists after long-term administration. Thus, we minimized the potential efficacy of pretreatment with albuterol. Nonetheless, the guideline recommendation to use 2 puffs of albuterol before exercise as a standard of care for prevention of exercise-induced bronchospasm was more effective than montelukast. In our study, none of the patients experienced exerciseinduced bronchospasm while receiving albuterol compared with 55% of patients receiving montelukast who still had a minimum of 15% decrease in their FEV1 after exercise. A recent report by the American Academy of Allergy, Asthma and Immunology Work Group suggests a lower cutoff point of at least 10% decrease in FEV1 to be sufficient for the diagnosis of exercise-induced bronchospasm, provided the decrease is associated with symptoms. Using this cutoff, only 27% of our patients receiving montelukast had protection against exerciseinduced bronchospasm when maximum effect of montelukast is expected to be seen.
Previous studies have suggested that exhaled Cys-LT concentration may be a noninvasive marker of airway inflammation in pediatric asthma.[27,28] Furthermore, exhaled Cys-LT concentrations have been reported to be significantly higher in children with unstable asthma compared with steroid-naïve children or those with stable mild to- moderate asthma who are taking inhaled corticosteroids.[4,29] One group of authors did not find a significant difference between patients with mild intermittent asthma (steroid naïve) and healthy subjects. Another group evaluated 19 children with asthma (11 receiving inhaled corticosteroids and eight who were steroid naïve) for exercise-induced bronchospasm. Exhaled Cys-LT concentration was higher in children with exercise-induced bronchospasm compared with those without exercise-induced bronchospasm or the healthy control group (42.2 vs 11.7 and 5.8 pg/ml, respectively, p<0.05 for both comparisons). Three days of treatment with montelukast lowered the concentration of exhaled Cys-LT to where there was no significant difference from that in the control group. A significant correlation was seen between the exhaled Cys-LT concentration and maximum decrease in FEV1 after exercise (r=0.7, p<0.01). We did not see a significant difference in exhaled Cys-LT concentration between those with a negative exercise challenge and those with exercise-induced bronchospasm. This disparity may be explained by our patients having relatively mild asthma, as their exhaled Cys-LT concentration was close to that of healthy control subjects reported in other studies. Only two of our patients with exercise-induced bronchospasm and five of those with a negative exercise challenge were receiving inhaled corticosteroids. We also did not find a correlation between exhaled Cys-LT concentration and maximum decrease in FEV1 after exercise or the response to montelukast. The Cys-LT in exhaled breath condensate, although attractive due to its simple and safe method of collection, may be a more useful marker for the severity of asthma than for predicting exercise-induced bronchospasm. Although our study suggests a minimal role of Cys-LT in exercise-induced bronchospasm, it was not powered to detect a difference in exhaled breath condensate Cys-LT concentrations, so the data must be considered preliminary.
Recently, the FDA approved labeling of montelukast for prevention of exercise-induced bronchospasm in patients with asthma who are aged 15 years or older. The approval was based on the clinical trials comparing montelukast with placebo.[25,30] These studies have shown that montelukast can attenuate exercise-induced bronchospasm. To our knowledge, our study is the first controlled clinical trial comparing the protective effect of montelukast with albuterol, which is considered the standard-of-care treatment.
We appreciate that we had a small number of patients in the study; however, based on previous data, we would have expected to see a smaller difference between the treatment groups, and the study was powered to detect the anticipated difference. Only two of our patients were receiving long-term controller drugs for asthma, but they all used albuterol as needed for symptoms and prophylactically before exercise.