|
|
| |||||||||||||||||||||||||||||||||||||||||||
Background Inhaled
Methods In a multicenter, double-blind study, we randomly assigned 255 patients with mild asthma to inhale albuterol either on a regular schedule (126 patients) or only as needed (129 patients). The patients were followed for 16 weeks.
Results The primary outcome indicator, peak expiratory air flow measured in the morning, did not change significantly during the treatment period in the scheduled (416 liters per minute after the run-in period and 414 liters per minute after the treatment period) or the as-needed (424 liters per minute at both times) treatment groups (P = 0.71). There were no significant differences between the two groups in peak flow variability, forced expiratory volume in one second, the number of puffs of supplemental albuterol needed, asthma symptoms, asthma quality-of-life score, or airway responsiveness to methacholine. The statistically significant differences between the groups in evening peak flow and in the short-term bronchodilator response to inhaled albuterol were small and judged to be clinically unimportant.
Conclusions In patients with mild asthma, neither deleterious nor beneficial effects derived from the regular use of inhaled albuterol beyond those derived from use of the drug as needed. Inhaled albuterol should be prescribed for patients with mild asthma on an as-needed basis.
-agonists are the most commonly used treatment for asthma, but data suggest that regularly scheduled use of these agents may have a deleterious effect on the control of asthma. We compared the effects of regularly scheduled use of inhaled albuterol with those of albuterol used only as needed in patients with mild chronic, stable asthma.
-agonists.1,2,3,4 Before 1990, a number of standard reference works recommended the use of inhaled -agonist medications on a regular basis5,6,7,8 as a treatment for mild asthma. The recommendation, based on evidence that this approach resulted in better control of asthma than the use of -agonists on an as-needed basis,9,10,11,12 has been reinforced by subsequent studies.13,14 In 1990 and in a follow-up paper in 1993, Sears and coworkers15,16 reported that regular use of -agonists was associated with diminished asthma control and suggested that the use of -agonists could account for increasing worldwide asthma morbidity. Subsequent reports have also reinforced these observations.17,18,19 However, none of these studies, either in favor of or against the regularly scheduled use of -agonists, have involved sufficiently large cohorts of patients with clinically uniform disease for a long enough period to be used in drafting recommendations for treatment. Indeed, the debate about the safety and efficacy of -agonists as a class continues.20,21,22,23,24,25,26,27,28,29,30,31,32
Since mild asthma is the most prevalent form of asthma and the type for which inhaled -agonists are most likely to be the sole therapy,3,4 the Asthma Clinical Research Network of the National Heart, Lung, and Blood Institute undertook a study to test the following hypothesis: In patients with mild asthma whose only asthma treatment is inhaled -agonists, the addition of scheduled treatment with inhaled -agonists to treatment on an as-needed basis will result in no effect on the control of asthma. In this report we summarize the results of a 26-week randomized, multicenter, double-blind, placebo-controlled trial in patients with mild asthma. The trial compared the control of asthma in a group of patients treated with regularly scheduled albuterol with that in a group of patients treated with albuterol only as needed.
Methods
Patient Recruitment
Patients with mild asthma, as defined by the criteria shown in Table 1, were recruited from existing study populations and by advertising. Eligible patients entered a six-week single-blind run-in period, during which they used a placebo inhaler on a regular basis (two inhalations four times a day) and took supplemental puffs of open-label albuterol as needed. During the run-in period, patients were evaluated three times at two-week intervals, at which time asthma control was assessed by the review of a number of criteria.
|
Patients were randomly assigned to a treatment group if over the six-week period their asthma was clinically stable and they demonstrated their ability to comply with the study procedures, as indicated by the regular use of the placebo inhaler (monitored by a Chronolog recording device) and their ability to record their peak flow (twice daily, using a Mini-Wright peak-flow meter [Clement Clarke, Columbus, Ohio]) and asthma symptoms (once daily) in a diary. The treatments assigned consisted of either inhaled albuterol on a regular basis (two inhalations four times a day) plus albuterol as needed or inhaled placebo on a regular basis (two inhalations four times a day) plus albuterol as needed. Albuterol and placebo inhalers were generously supplied by Schering-Plough (Memphis, Tenn.). Patients were instructed to have their regularly scheduled inhalations in the morning after recording their morning peak flow, at midday, in the late afternoon, and on retiring to sleep after recording their evening peak flow. They were instructed to allow at least four hours between their regularly scheduled inhalation in the late afternoon and the recording of their evening peak flow.
Patient Treatment
Over the ensuing 16 weeks, while patients received blinded treatment, the control of asthma was monitored daily, through peak flow rates and symptoms recorded by patients, as well as during clinic visits, which were scheduled every two to three weeks. At the completion of the randomized-treatment period, all the patients were switched to single-blind treatment with inhaled placebo for a four-week withdrawal period; during this time patients continued to use open-label albuterol as needed.
Seven outcome indicators were monitored: peak flow, the symptom record, quality of life, the change in the forced expiratory volume in one second (FEV1) in response to an inhaled bronchodilator, the concentration of methacholine required to decrease the FEV1 by 20 percent (PC20), asthma exacerbations, and treatment failure. Peak flow, the primary outcome indicator, was measured twice daily by patients using a Mini-Wright peak-flow meter; the best of three efforts was recorded. Patients recorded their asthma symptoms and the number of puffs of supplemental albuterol used daily. Asthma symptoms were recorded on a 4-point scale, with 0 representing no symptoms and 3 representing severe symptoms. Asthma-specific quality-of-life scores were recorded during clinic visits, with an instrument validated by other investigators.33 To determine the spirometric response to an inhaled bronchodilator, the difference in the FEV1 before and 15 minutes after two inhalations of albuterol was measured (and reported as percent improvement) during clinic visits when responsiveness to methacholine was not tested. Patients refrained from taking their study medications for at least eight hours before all clinic visits. To measure PC20 for methacholine, methacholine aerosols were generated with a nebulizer (model 646, DeVilbiss Health Care, Somerset, Pa.) and a calibrated dosimeter (S&M Instruments, Doylestown, Pa.). The PC20 for methacholine was determined by standard procedures.34 Asthma exacerbations were monitored during each clinic visit; patients were asked about their asthma control, and all asthma exacerbations were recorded. An asthma exacerbation was defined as an increase in symptoms of cough, chest tightness, or wheezing in association with one or more of the following: an increase over the base-line use of supplemental -agonist treatments of 8 or more puffs per 24 hours for a period of 48 hours, the use of 16 or more puffs of a supplemental -agonist per 24 hours for a period of 48 hours, or a fall in peak flow of 35 percent or more from the best three-day average (morning and evening) during the run-in period. Treatment was considered to have failed if patients who had asthma exacerbations and were treated with increased doses of -agonists did not respond adequately — that is, if they continued to meet the criteria for exacerbation. Such patients were treated with a short course of prednisone, as determined by their physicians; their data continued to be collected, and they remained in the trial (in accordance with the intention-to-treat method).
Standardization and Quality-Assurance Techniques
All clinical laboratory tests — that is, measurements of lung function, skin testing for allergies, methacholine challenges, and quality-of-life assessments — were performed at each center with the use of equipment and procedures that were standardized for the entire network. Workers participating in the network were tested to ensure proficiency and uniformity in all network-related skills and had to pass certification examinations before the data they gathered could be used in the network. All results of spirometric testing (Collins Eagle 2 spirometer, Quincy, Mass.), including that for the methacholine challenge, were confirmed by a single network member. Peak-flow meters were tested against spirometers during each clinic visit and were replaced if they failed to meet previously established performance standards. A distributed data-entry system allowed each clinical center to submit its data over the Internet directly to the Data Coordinating Center. The Data Coordinating Center entered the data a second time to verify it.
Compliance
Each patient was given a digital wristwatch with multiple alarms to improve treatment compliance. In addition, Chronolog recording devices were used with the randomly assigned metered-dose inhalers to provide an electronic record of the date and time of inhaler use.
Statistical Analysis
Morning peak flow was chosen as the primary outcome variable for the calculation of sample size. A minimum of 200 patients made it possible to detect a difference of 25 liters per minute between groups with 80 percent statistical power. A goal of recruiting 250 randomized patients was established on the assumption that the dropout rate would be less than 20 percent. This sample size also provided 80 percent statistical power to detect a difference of 0.19 liter in FEV1 and 0.70 doubling dilution in the PC20 values for methacholine.
Response variables — that is, peak-flow values, medication use, and asthma symptoms — from the patients' diary cards were averaged each week. Because of the longitudinal nature of most of the response variables, a mixed-effects linear model was applied35,36; this approach allowed all data obtained to be used, not just the data obtained at a single visit. For each response variable, a segmented linear model was fitted with an intercept and with slopes for the last 4 weeks of the run-in period, the first 5 weeks of the treatment period, the remaining 11 weeks of the treatment period, and the withdrawal period. The "break point" after five weeks of randomized treatment was chosen on the basis of rates of asthma exacerbation reported by Sears et al.15 For each outcome measure, values were calculated from the models for the end of the run-in period, for the end of the double-blind–treatment period, and for the end of the withdrawal period. This statistical model was determined before the start of the study, and therefore other models were not considered during data analysis. The groups were compared with respect to rates of treatment failure with the use of Fisher's exact test. To ensure patient safety, an interim analysis was conducted after approximately 40 percent of the randomized patients had completed the trial or withdrawn consent; as a result of this analysis, the P value considered to indicate statistical significance was reduced from 0.05 to 0.03 for the final analyses.37,38
Results
Enrollment and Retention
Of the subjects recruited, 255 were eligible for enrollment at the end of the six-week run-in period and were randomly assigned to receive double-blind treatment (Table 2). There were no significant differences between the treatment groups with respect to any of the indexes monitored. During the period of randomized treatment and withdrawal, 25 subjects dropped out of the trial — 10 in the scheduled-treatment group and 15 in the treatment-as-needed group. Two hundred thirty patients completed the entire trial.
|
Compliance with the use of inhaled medication, either active or placebo, on a regular basis was greater than 80 percent, as indicated by Chronolog treatment records and an analysis of diary cards. Of the 3172 scheduled visits to patients' clinical centers, 26 were missed, for a rate of compliance of over 99 percent.
Asthma Exacerbations
Asthma was exacerbated 24 times (11 times in the scheduled-treatment group and 13 times in the treatment-as-needed group) during the active treatment period and 4 times during the withdrawal period (twice in each treatment group). The 28 exacerbations occurred in 12 patients in the scheduled-treatment group and 11 patients in the treatment-as-needed group.
Treatment Failures
Treatment was considered to have failed in 11 patients during the 16-week period of randomized treatment (5 in the scheduled-treatment group and 6 in the treatment-as-needed group) and in 2 during the withdrawal period (both in the scheduled-treatment group). There were three visits to the emergency room for asthma (two in the scheduled-treatment group and one in the treatment-as-needed group). No patients were hospitalized for asthma during the trial, and none died. There were no significant differences in any of the event rates between the two treatment groups.
Efficacy Outcomes
Lung function (indicated by morning peak flow, evening peak flow, peak-flow variability, FEV1, -agonist responsiveness, and PC20) and asthma symptoms (determined by the number of uses of the supplemental -agonist metered-dose inhalers, diary scores, and quality-of-life scores) as derived from the regression analysis performed for each patient group are shown in Figure 1 and Table 3. Graphic displays of values predicted by the model as compared with sample means showed excellent goodness of fit by the statistical model (data not shown). There were no significant differences in morning peak flow between the two treatment groups (Table 3). Even though the average use of albuterol was 9.3 puffs per day in the scheduled-use group and 1.6 puffs per day in the treatment-as-needed group, the extra use of medication did not lead to differences in peak-flow variability, FEV1, supplemental albuterol use, asthma symptoms, quality of life, or PC20.
|
|
Discussion
Over the past five years, health professionals have been concerned about potential deleterious effects of inhaled -agonists on asthma control.20,21,22,23,24,25,26,27,28,29,30,31,32 In part because of concern about potential adverse effects of regular -agonist use, most current guidelines for the management of asthma recommend their use on an as-needed basis only,3,4,40 even though there is no sound clinical evidence on which to base this recommendation. Our study provides reassurance about the use of inhaled albuterol by clearly demonstrating, in patients with mild asthma, that its regular use is not associated with a deleterious effect on asthma control. At the same time, however, in this group of patients with mild asthma we were unable to demonstrate any additional beneficial effect of regularly scheduled treatment with inhaled albuterol beyond that achieved with albuterol used on an as-needed basis only.
We observed no change attributable to treatment in our primary outcome variable, morning peak flow. The only effect during randomized treatment that was attributable to the different treatment regimens was the change in bronchodilator responsiveness. The FEV1 response to treatment with albuterol was less than 12 percent in both treatment groups before the initiation of randomized therapy, probably because the patients had mild asthma. The small increase in bronchodilator responses observed in the scheduled-treatment group was somewhat surprising, since many have speculated that the regularly scheduled use of inhaled -agonists may produce tolerance to their effects. Although we have no specific explanation for these findings, our data provide clinically meaningful reassurance that tolerance to the acute effect of inhaled albuterol does not occur with its regularly scheduled use.
The only other change attributable to the difference in treatment regimens occurred in evening peak flow between the end of treatment and the end of withdrawal. Evening peak flow rose slightly, but not significantly, during randomized treatment in the scheduled-use group and then fell significantly after the end of randomized treatment. We attribute this effect to the residual action of inhaled albuterol. Even though patients taking their scheduled albuterol were instructed to withhold it for four hours before measuring evening peak flow, albuterol is known to have a residual bronchodilator effect up to six hours after administration.13 This observation also provides a basis for our finding that variability in peak flow increased with treatment and then decreased during withdrawal in the scheduled-treatment group.
An important aspect of our trial design was the incorporation of a "withdrawal" period, when all patients were treated with inhaled albuterol only as needed. This aspect of the study design allowed us to ascertain whether the scheduled use of inhaled albuterol masked underlying changes in lung function or asthma symptoms. We compared asthma control during the run-in and withdrawal periods to determine whether lung function deteriorated during scheduled treatment with albuterol. We found no statistically or clinically significant changes in any of the outcome indicators monitored; thus, there were no masked deleterious effects of scheduled albuterol use.
Although there were no other effects attributable to differences in treatment between the groups, it is of interest to examine one of the statistically significant changes that occurred in one of the treatment groups. In the two weeks after the start of treatment with regularly scheduled inhaled albuterol, we observed a small increase (one half of a doubling dilution) in airway responsiveness. This increase in airway reactivity did not progress as the scheduled treatment continued, disappeared as soon as the scheduled treatment was stopped, and was not associated with a change in any other index of asthma control. Thus, we were able to confirm what has been noted by others — that the scheduled use of inhaled -agonists slightly increases airway responsiveness to methacholine.16 However, our data extend these findings in an important way by demonstrating that this change in airway responsiveness is not associated with any other alteration in asthma control in patients with mild asthma.
Our study differs in a number of ways from previously published studies addressing the same general question. Unlike the study by Sears and colleagues,15,16 in which over 75 percent of the patients were using inhaled corticosteroids, our study was limited to patients with mild asthma whose mean lung function was near normal — that is, whose FEV1 values were approximately 90 percent of the predicted value — and whose only asthma treatment was inhaled -agonists. As a consequence of this difference in asthma severity, our patients had far fewer asthma exacerbations: fewer than 1 in 10 of our patients had a significant asthma exacerbation, as compared with 3.1 exacerbations per patient in the study by Sears et al. Since our data are relevant only to patients with mild asthma, it may not be appropriate to extrapolate our findings to patients with more severe asthma.
Over the past five years, there has been substantial debate about inhaled -agonists as a class of asthma treatment.20-32 In this study, despite a more than fivefold difference in the amount of -agonist administered (9.3 puffs vs. 1.6 puffs per day), there was no clinically significant difference in asthma control between the treatment groups, indicating that patients with mild asthma should receive inhaled albuterol on an as-needed basis only; this approach also reduces the cost of medication. On the other hand, if, in an individual case, a patient with mild asthma and his or her physician perceive that scheduled treatment is beneficial, our results indicate that this practice will not be detrimental in this population.
Supported by grants (U10 HL 51831, U10 HL 51834, U10 HL 51843, U10 HL 51810, U10 HL 51823, and U10 HL 51845) from the National Institutes of Health. Copies of the complete study protocol can be obtained from the Asthma Clinical Research Network Data Coordinating Center, Center for Biostatistics, Milton S. Hershey Medical Center, 500 University Dr., Rm. C5863, P.O.B. 850, Hershey, PA 17033-0850.
We are indebted to the following collaborators and coordinators at each center: William McGarry, B.S., R.R.T., C.P.F.T., and Nitaina Cruz, B.S., Boston; Monica Kraft, M.D., Juno Pak, B.S., and Joseph D. Spahn, M.D., Denver; Kathleen Cox, M.S., and Ann Smith, Madison, Wis.; Janice Kubis, R.N., Philadelphia; Theresa Ward, R.N., B.S.N., Grace Hardie, R.N., M.S., P.C.N.S., and Anase Taharka, M.D., San Francisco; Donna K. Baker, B.S., Hershey, Pa.; and Jean G. Ford, M.D., and Diane McLean, Ph.D., New York.
Source Information
From Brigham and Women's Hospital and Harvard Medical School, Boston (J.M.D., E.I.); the University of California at San Francisco, San Francisco (H.A.B., J.V.F., S.C.L.); Milton S. Hershey Medical Center, Hershey, Pa. (V.M.C.); Thomas Jefferson University, Philadelphia (J.E.F., S.P.P.); the University of Wisconsin, Madison (R.F.L., C.S.); and the National Jewish Center for Immunology and Respiratory Medicine, Denver (R.J.M., S.J.S.). Additional contributing authors were Reuben Cherniack, M.D. (National Jewish Center for Immunology and Respiratory Medicine, Denver), and Suzanne Hurd, Ph.D. (National Heart, Lung, and Blood Institute, Bethesda, Md.).
Address reprint requests to Dr. Drazen at the Respiratory Disease Division, Brigham and Women's Hospital, 75 Francis St., Boston, MA 02115.
References
-agonists and the risk of death and near death from asthma. N Engl J Med 1992;326:501-506. [Abstract]
2 agonists in asthma: much ado about nothing? Adverse effects are not proved. BMJ 1994;309:795-796. 2 agonists in asthma: much ado about nothing? Still cause for concern. BMJ 1994;309:794-795. -Adrenergic bronchodilators. N Engl J Med 1995;333:499-506. Asthma Clinical Research Network Investigators: C. Fanta, A. Fischer, and A. Quizon, Harvard Medical School, Boston; C.V. Chambers, K. Epstein, S.M. McGeady, and J. Diamond, Thomas Jefferson University, Philadelphia; S.L. Janson, University of California at San Francisco, San Francisco; S.J. Kunselman, J.K. Martel, C.W. Zwillich, and E.A. Mauger, Data Coordinating Center, Milton S. Hershey Medical Center, Hershey, Pa.
Protocol Review Committee for this study: G.W. Hunninghake, Iowa City, Iowa; R.O. Crapo, Salt Lake City; R. Nicklas, Washington, D.C.; R.C. Strunk, St. Louis; J.E. Connett, Minneapolis; M.R. Sears, Hamilton, Ont., Canada; and D.P. Tashkin, Los Angeles.
Data Safety Monitoring Board for this study: N.R. Anthonisen, Winnipeg, Man., Canada; B. Layman, Waynesboro, Pa; S.A. Murphy, Albuquerque, N.M.; S. Redline and M.D. Schluchter, Cleveland; G. Shapiro, Seattle; and M.C. Wu, National Heart, Lung, and Blood Institute, Bethesda, Md.
| |||||||||||||||||||||||||||||||||||||||||||
Related Letters:
Albuterol in Mild Asthma
Suissa S., Ernst P., Zahger D., Drazen J. M., Israel E., The National Heart, Lung, and Blood Institute's Asthma Clinical Research Network
Extract |
Full Text
N Engl J Med 1997;
336:729-730, Mar 6, 1997.
Correspondence
Effect of Inhaled Formoterol and Budesonide on Exacerbations of Asthma
FitzGerald J. M., Salvi S., Pauwels R. A., Tattersfield A. E., Löfdahl C.-G.
Extract |
Full Text
N Engl J Med 1998;
338:1071-1073, Apr 9, 1998.
Correspondence
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2008 Massachusetts Medical Society. All rights reserved. |
Previous
