Published ahead of print on June 26, 2008, doi:10.1164/rccm.200706-910OC
© 2008 American Thoracic Society doi: 10.1164/rccm.200706-910OC
Study of Montelukast for the Treatment of Respiratory Symptoms of Post–Respiratory Syncytial Virus Bronchiolitis in Children1 Danish Pediatric Asthma Center, Copenhagen University Hospital, Copenhagen, Denmark; 2 Department of Pediatric Pulmonology, Hospital para el Niño Poblano, Puebla, Mexico; 3 Department of Pediatric Medicine, KK Women's and Children's Hospital, Singapore; 4 Children's Hospital of Oakland, Oakland, California; 5 Pediatric Pulmonology, Netcare Krugersdorp Hospital, Johannesburg, South Africa; and 6 Merck Research Laboratories, Rahway, New Jersey Correspondence and requests for reprints should be addressed to Hans Bisgaard, M.D., Danish Pediatric Asthma Center, Department of Pediatrics, Copenhagen University Hospital, Gentofte, DK-2900 Copenhagen, Denmark. E-mail: bisgaard{at}copsac.dk
Rationale: A pilot study (Bisgaard H; Study Group on Montelukast and Respiratory Syncytial Virus. A randomized trial of montelukast in respiratory syncytial virus postbronchiolitis. Am J Respir Crit Care Med 2003;167:379–383) reported the efficacy of montelukast in post–respiratory syncytial virus (RSV) bronchiolitic respiratory symptoms. Objectives: To evaluate the efficacy and safety of montelukast, 4 and 8 mg, in treating recurrent respiratory symptoms of post-RSV bronchiolitis in children in a large, multicenter study. Methods: This was a double-blind study of 3- to 24-month-old children who had been hospitalized for a first or second episode of physician-diagnosed RSV bronchiolitis and who tested positive for RSV. Patients (n = 979) were randomized to placebo or to montelukast at 4 or 8 mg/day for 4 weeks (period I) and 20 weeks (period II). The primary end point was percentage symptom-free days (%SFD; day with no daytime cough, wheeze, and shortness of breath, and no nighttime cough).
Measurements and Main Results: No significant differences were seen between montelukast and placebo in %SFD over period I: mean ± SD for placebo and for montelukast at 4 and 8 mg were 37.0 ± 30.7, 38.6 ± 30.4, and 38.5 ± 29.9, respectively. Least-squares mean differences (95% confidence interval) between montelukast (4 mg) and placebo and between montelukast (8 mg) and placebo were 1.9% (–2.9, 6.7) and 1.6% (–3.2, 6.5), respectively. Secondary end points were similar across treatments. Both doses were generally well tolerated. During the first two treatment weeks, average %SFD was approximately 29%. In post hoc analyses of patients (n = 523) with persistent symptoms (%SFD Conclusions: In this study, montelukast did not improve respiratory symptoms of post-RSV bronchiolitis in children. Clinical trial registered with www.clinicaltrials.gov (NCT00076973).
Key Words: post–respiratory syncytial virus bronchiolitic asthma symptoms • symptom-free days • wheeze • leukotriene receptor antagonist
Viral infection of the lower airways in infancy often results in bronchiolitis, an inflammatory disorder that is associated with rhinitis, tachypnea, wheezing, cough, crackles, use of accessory muscles, and/or nasal flaring (1). The most common etiology in bronchiolitis is respiratory syncytial virus (RSV) (2, 3). The acute phase of infection in RSV-induced bronchiolitis may lead to hospitalization in 1–2% of infants (4). Many patients continue to have postbronchiolitic recurrent episodes of lower airway obstruction, such as cough and/or wheezing, which may continue for months and sometimes years after the acute infection has resolved (3, 5–7). Several studies have suggested that RSV-induced bronchiolitis in infancy may be a risk factor for such subsequent wheezing and asthma symptoms (8–11). Cytokines from neutrophils and neurotrophic factors and receptors may contribute to airway inflammation during RSV infection (12, 13). It remains unclear whether the immune response to RSV alters airway function and increases the risk of asthma (11, 14) or whether preexisting immune responses or airway mechanics that underlie asthma susceptibility also predispose the patient to a severe response to RSV infection. Few effective options are available for the treatment of post-RSV bronchiolitic recurrent episodes of wheeze, cough, and other respiratory symptoms (1, 3, 4). A pilot study (15) reported a potential role for the cysteinyl leukotriene receptor antagonist montelukast, an established asthma controller, in relieving respiratory symptoms and reducing exacerbations in post-RSV bronchiolitis. This result is supportive of several observations implicating cysteinyl leukotrienes in this disorder. Cysteinyl leukotriene levels in the bronchoalveolar lavage of children with acute RSV-induced bronchiolitis were significantly higher than in control subjects (16). Similarly, significantly elevated levels of leukotriene C4 were seen in the nasopharyngeal secretion of infants with RSV-induced bronchiolitis compared with those with RSV illness limited to upper respiratory symptoms; leukotriene levels were detectable at progressively lower concentrations in samples obtained up to 28 days after the onset of illness (17). Urinary leukotriene E4 levels were eightfold higher in infants with RSV bronchiolitis than in control subjects without respiratory infection (18). The present study was designed to evaluate the efficacy and safety of two doses of montelukast in treating the recurrent episodes of respiratory symptoms post-RSV bronchiolitis in children. Doses selected for this study included the 4-mg dose approved for asthma in children as young as 6 months of age in some countries and a twofold higher dose (8 mg). On the basis of safety data from previous montelukast studies (19, 20), there were no safety issues precluding the evaluation of an 8-mg dose in these young children. Some of the results of this study have been previously reported in the form of abstracts (21–23).
Study Design This study was conducted between July 2003 and October 2006 in 118 centers across 6 continents. Patients were randomly allocated to receive once-daily montelukast 4-mg granules, montelukast 8-mg granules, or placebo. The study comprised two double-blind periods: a 4-week period I and a 20-week period II of extended treatment. Patients were allocated to treatment assignment according to a computer-generated randomized allocation schedule. Allocation numbers were centrally assigned in sequential order via an interactive voice response system. All study personnel and patients remained blinded to treatment allocation throughout the study; the code was revealed to the researchers once recruitment, data collection, and laboratory analyses were completed.
Patients As-needed ipratropium and/or epinephrine was permitted before or during hospitalization. About half the randomized patients were allowed treatment of bronchiolitis with corticosteroids during hospitalization. One oral corticosteroid rescue per month was allowed in both treatment periods. Throughout the study, patients were permitted as-needed short-acting β-agonist therapy for treatment of respiratory symptoms and the use of oxygen, nutrition, and intravenous fluids according to the investigator's usual clinical practice. The study (protocol 272) was approved by ethics review committees or institutional review boards for each study site. Informed consent was obtained from the parent/legal guardian of each patient.
Evaluations Other end points included the percentage of cough-free days, percentage of days without β-agonist use, percentage of patients with systemic corticosteroid rescues, and the average of the four individual daily symptom scores. Safety and tolerability were assessed by clinical evaluations (physical examinations), adverse experience monitoring, and laboratory safety tests. As prespecified in the study protocol, an adverse experience (AE) included any unfavorable and unintended change or worsening in the structure, function, or chemistry of the body temporally associated with treatment. A serious AE included any AE that resulted in death or a persistent or significant disability/incapacity; was life threatening; resulted in or prolonged an existing inpatient hospitalization; was a congenital anomaly/birth defect, a cancer, or an overdose; or any medical event that, in the investigator's appropriate medical judgment, could jeopardize the patient.
Statistical Analysis
For a sample size of 300 patients per treatment arm, the study had 90% power to detect an 8% point difference between each dose of montelukast and placebo for the primary end point (with
Patients and Baseline Characteristics Of 1,182 patients screened for inclusion in the study, 979 were randomized: 328 to placebo, 327 to montelukast (4 mg), and 324 to montelukast (8 mg); 203 patients were not randomized (Figure 1). A total of 745 patients completed the study.
Baseline characteristics of patients in each treatment group are described in Table 1. Most (58.6%) were boys; the mean age at randomization of boys was 7.7 ± 4.6 months and of girls was 7.9 ± 4.5 months. Most patients (94.4%) were hospitalized for at least 24 hours. To allow the inclusion of the small number of patients hospitalized for less than 24 hours, a respiratory severity score was used. This score was developed to assess the severity of bronchiolitis based on clinical attributes of bronchiolitis (cough, wheeze, audible rales/crackles/rhonchi, and paradoxical chest movements [retractions]), the respiratory rate, and the level of oxygen saturation (based on pulse oximetry) at baseline. A separate end-point validation study conducted concurrently with the present study showed that patients with a documented baseline score of at least 5 (on a 0–8 point scale) and treated in a hospital or urgent/emergent-care setting for less than 24 hours had a post-discharge symptom burden similar to that of patients hospitalized for at least 24 hours (24).
All baseline characteristics were similar across the treatment groups. The mean patient-reported compliance (ratio of the number of days the study drug was taken to the number of study days) was greater than 97% in each treatment group for periods I + II.
Efficacy Evaluations
The secondary end point of percentage of bronchiolitis-free days during period I was similar between montelukast (4 mg) and placebo and between montelukast (8 mg) and placebo (see Table 2). No significant trends for treatment group differences were seen during period I (P = 0.537). Furthermore, the secondary end point of percentage of patients with exacerbations in periods I + II was similar across treatment groups: 33% of the patients had at least one exacerbation in the placebo group, 31% in the montelukast 4-mg group, and 32% in the montelukast 8-mg group. No significant trends for treatment group differences were seen (P = 0.565). Results for other end points are also shown in Table 2. These include percentage of cough-free days, percentage of days without β-agonist use, and the average of individual daily symptom scores, all measured over period I; and percentage of symptom-free days, percentage of bronchiolitis-free days, average of individual daily symptom scores, percentage of patients with systemic corticosteroid use, and percentage of patients with health care resource use, all assessed over periods I + II. No significant differences were seen between the treatment groups for any of these end points. Treatment effects during period I were consistent among the prespecified subgroups (geographic region, sex, race, age group, use of corticosteroids for the episode of bronchiolitis at randomization, hospitalization for 24 hours or more because of bronchiolitis, time in hospital for initial current bronchiolitis episode, oxygen saturation percentage, and baseline respiratory severity score) (data not shown). Treatment effects were also consistent among the prespecified subgroups on the basis of first or second episode of bronchiolitis. Similarly, treatment effects did not appear to be affected by the number of days of bronchiolitic symptoms before randomization or presence of a positive or abnormal RAST test.
Post Hoc Analyses Among those patients with persistent symptoms, 30.5% of the patients receiving montelukast (4 mg and 8 mg combined) and 24.4% of the patients receiving placebo were symptom free over the second half of period I; the odds ratio comparing the treatments was 1.36 (95% CI, 1.01–1.81) (Figure 2A). Changes in the number of patients with symptom-free days were small over period I among those patients with less persistent symptoms (Figure 2B).
Safety Montelukast at 4 and 8 mg was generally well tolerated; the safety profiles of both doses assessed over period I and over periods I + II were generally comparable to that of placebo. No significant trend from placebo through montelukast at 4 and 8 mg was detected with respect to the percentage of patients with one or more clinical adverse experiences (AEs; 81.4, 84.1, and 82.1%, respectively), drug-related AEs (4.1, 4.4, and 2.8%, respectively), serious AEs (16.0, 13.0, and 16.6%, respectively), or AEs leading to discontinuation from the study (2.8, 1.9, and 2.8%, respectively). Table 3 lists the most frequent (clinical and serious clinical) AEs in either montelukast group regardless of causality over periods I + II.
Laboratory AEs were infrequent, and the percentage of patients experiencing one or more laboratory AEs was similar among the treatment groups (8.4, 9.0, and 8.2%, respectively). No significant trend from placebo through montelukast (8 mg) was detected over period I and over periods I + II with respect to the proportion of patients with one or more laboratory AEs, drug-related laboratory AEs, serious laboratory AEs, laboratory AEs leading to discontinuation (Table 4), or drug-related laboratory AEs leading to discontinuation. One patient (placebo) had a serious laboratory AE in the study. In addition, the proportion of patients outside predefined limits of change over period I and periods I + II for selected laboratory parameters (hematocrit, white blood cell count, platelet count, aspartate aminotransferase, alanine aminotransferase, serum alkaline phosphatase, and serum total bilirubin) showed no significant trend from placebo through 8 mg for any of the comparisons, except for increase in platelet count during period I and periods I + II.
Two patients died during the study; neither death was determined by the investigator to be drug related. One patient (4 mg of montelukast) died of acute respiratory distress syndrome and sepsis while hospitalized for RSV-induced bronchiolitis. The second patient (8 mg of montelukast) died of thermal burns in a house fire.
The results of this study show that montelukast at two separate doses did not alleviate post-RSV bronchiolitic respiratory symptoms in 3- to 24-month-old infants over a period of 4 weeks (primary analysis) and over a subsequent 20-week extended treatment period. The key outcome measures employed in this study were chosen to provide information of clinical relevance about patients with RSV-induced bronchiolitic respiratory symptoms. The primary end point was a composite measure that evaluated, over a 24-hour period, the presence or absence of cough, wheeze, and dyspnea, all symptoms commonly reported by caregivers (24). This end point was similar to that employed in a pilot study evaluating montelukast in post-RSV bronchiolitis (15), except for the assessment of activity limitation in the earlier study, a measure that we prospectively deemed difficult to estimate uniformly across the age range of this large, multicenter and multicultural study. The secondary end points of percentage of bronchiolitis-free days and the proportion of patients with exacerbation assessed both symptoms and rescue interventions that provide a clinical measure of the control of respiratory symptoms. These measures of symptoms and health care use associated with post-bronchiolitis have been validated (25). Because of the long-term duration of respiratory symptoms after acute RSV-induced bronchiolitis (3), the extended 20-week, double-blind period permitted an evaluation of whether longer term treatment provides continued benefit. Focusing on hospitalized patients allowed the selection of a patient population with similar levels of severity of RSV-induced bronchiolitis at baseline. We also wanted to use a uniform approach to determine the severity of illness of infants treated for RSV-induced bronchiolitis in the urgent/emergent-care setting (e.g., emergency department) without hospital admission. For this purpose, a respiratory severity score was developed. The approximately 6% of patients included in the study who had a stay of less than 24 hours in a hospital or urgent/emergent-care setting and a respiratory severity score of 5 or more are expected to have a baseline severity of disease similar to that of patients hospitalized for 24 hours or more. The treatment effect of montelukast was similar in the subgroups of patients hospitalized for less than 24 hours or for 24 hours or more. The efficacy results of our study differed from those previously reported (15). That was a double-blind pilot study of postbronchiolitic respiratory symptoms in 3- to 36-month-old infants after hospitalization for acute RSV-induced bronchiolitis. Patients were randomized to montelukast at 5 mg/day or placebo for 28 days. The primary end point was a 24-hour day free of nighttime and daytime cough, wheeze, dyspnea and activity limitations. The results showed that over the 4-week treatment period, the median percentage of symptom-free days was 22% in the montelukast group, compared with a median of 4% in the placebo group (P = 0.015). The study by Bisgaard and coworkers (15) did not allow steroid use. In the present study, however, almost one in four patients had received corticosteroids during hospitalization, and oral corticosteroid rescue was allowed during treatment. Treatment effects were similar between patients with and without baseline steroid use. In general, patients in the previous study (15) had more persistent postbronchiolitic respiratory symptoms. The mean percentage of symptom-free days during the first 2 weeks of treatment (which was similar across the two treatment groups) was 17% in that study, compared with about 29% in our present study. Because the results in our study showed that some children have persistent respiratory symptom after hospitalization whereas others do not, we asked whether a treatment effect of montelukast could be detected in patients with persistent symptoms. In post hoc analyses of patients with a mean percentage of symptom-free days not exceeding 30% during the first 2 weeks of treatment, differences between the two doses of montelukast and placebo were seen in several end points. The effect size was comparable to that seen with controller treatment of persistent asthma in young children, which is smaller than later in life (26, 27). These preliminary findings need to be replicated in further studies. In this context, it must be pointed out that although the number of symptom-free days may provide a measure of the persistence of post-RSV bronchiolitic respiratory symptoms, it remains uncertain whether this end point assessment bears any correlation with the severity of such symptoms. Understanding correlates and predictors of increased respiratory burden in children as a result of lower respiratory tract infections such as are due to RSV remains an important task. Daily administration of montelukast was well tolerated at both the 4- and 8-mg dose levels during the duration of the study. The 4-mg dose of montelukast was previously shown to be well tolerated in children 6 months to 5 years of age (26–28). In conclusion, montelukast at 4 and 8 mg did not alleviate RSV-induced bronchiolitic respiratory symptoms in this study of 3- to 24-month-old children over a 24-week period. In spite of this overall negative result, post hoc analyses seemed to suggest that patients with more persistent symptoms may show more improvement in the primary end point when treated with montelukast. Further studies of children with persistent respiratory symptoms post-RSV bronchiolitis are required to prospectively determine effective therapy for this population.
Supported by a grant from Merck Research Laboratories.
* A list of investigators may be found at the end of this article. Originally Published in Press as DOI: 10.1164/rccm.200706-910OC on June 26, 2008 Conflict of Interest Statement: H.B. has been a consultant to, paid lecturer for, and holds sponsored grants from Aerocrine, AstraZeneca, Altana, GlaxoSmithKline, Merck, MedImmune, NeoLab, and Pfizer. The COPSAC clinical research unit has in the last 3 years received research grants from the following industry partners in increasing order: Merck, GlaxoSmithKline, and AstraZeneca. A.F.-N. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. A.G. has participated as a speaker in scientific meetings or courses organized and financed by various pharmaceutical companies (GlaxoSmithKline, AstraZeneca, Merck Sharpe & Dohme [MSD], Wyeth, and Abbott). A.G. is also on the advisory board for AstraZeneca for 2007. P.A. has participated in speaker programs for Merck and Medimmune. P.A. has owned about 3,000 shares of Merck stocks for more than 20 years, and P.A. had a Merck grant for the current manuscript. A.H. has delivered lectures for GlaxoSmithKline for which he received $3,000 in total. A.H. has lectured for Schering Plough for which he received $600 in total and has chaired meetings and lectured for MSD and received $500 in total. M.-P.M. is a paid employee of Merck & Co., Inc., owns company stock, and has stock options granted by Merck & Co., Inc. J.-L.M. is a paid employee of Merck & Co., Inc., owns company stock, and has stock options granted by Merck & Co., Inc. S.B.D. is a paid employee of Merck & Co., Inc., owns company stock, and has stock options granted by Merck & Co., Inc. T.F.R. is a paid employee of Merck & Co., Inc., owns company stock, and has stock options granted by Merck & Co., Inc. B.A.K. is a paid employee of Merck & Co., Inc., owns company stock, and has stock options granted by Merck & Co., Inc. List of Investigators: D. A. Fitzgerald, R. Loh, M. Nissen (Australia); K. Desager, A. Malfroot (Belgium); J. P. Lotufo, J. Rodrigues, N. Rosario Filho (Brazil); M. Boza, M. Calvo, R.A. Pinto (Chile); R. Aristizabal, G. Merchan, A. Perez (Colombia); M. N. Hermansen, E. Mosfeldt-Jeppesen, M. Pedersen, V. Petersen (Denmark); O. Ruuskanen (Finland); A. Deschildre, B. Fauroux, R. Klink (France); H.G. Bresser, M. Henschen, F. Riedel, S. Schmitt-Grohé, L. Vogler, J. Freihorst, A. Von Berg (Germany); Y. Lau (Hong Kong); A. Ghosh, M. Gupta, H. Jain, S. Kabra, D. Shivpuri (India); A. Fiocchi, M. Pifferi, G. A. Rossi (Italy); U. Schierloh (Luxembourg); J. De Bruyne (Malaysia); A. Flores, J. Sienra Monge (Mexico); F. T. Guerra Lillo, L. Monsante (Peru); A. Jiao, M. Sumpaico (Philippines); A. Emeryk, A. Lagun, T. Malaczynska, I. Stelmach (Poland); J.M. Lopes Dos Santos, M.F. Praca (Portugal); J. Santana (Puerto Rico); A. E. Goh, D. Goh Yam Thiam (Singapore); A. Halkas, O.F. Jooma, H. Lewis, A.I. Manjra, A. Puterman, A. Van Niekerk, J. Vermeulen, H. C. Weber (South Africa); R. Busquets Monge, N. Cobos-Barroso, J. Elorz Lambarri, M. Garcia de Miguel, M. L. Garcia-Garcia, A. Mirada Vives, S. Nevot Falco, A. Nieto-Garcia, J. Perez-Frias, J.R. Villa-Asensi (Spain); L. Nilsson, J. Svedmyr, P. Thunqvist (Sweden); J. L. Huang, K.H. Lue (Taiwan); A. H. Assa'Ad, P. Azimi, J. Bardelas, J. Carl, A. Chidekel, B. Congeni, K. Coopersmith, J. DeVincenzo, M. Dyson, R. J. Fink, C. Harris, K. Haver, P. Hiatt, C. Johnson, R. Kravitz, G. Kurland, W. La Via, H. Lee, B. Lesnick, S. Mc Colley, S. Medford, I. Melamed, J. Moy, K. Murphy, B. Nickerson, H. Panitch, I. Paul, G. Piedimonte, M.H. Rathore, R. Reddy, L. Roberto, H. Sadeghi, V. Sanchez-Bal, S. Sheikh, D. Skoner, R. Steele, J. Sullivan, D. Thomas, W. Warren, N. Wilson (United States). Received in original form June 22, 2007; accepted in final form June 26, 2008
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
AT A GLANCE COMMENTARY
30% over Weeks 1–2), differences in %SFD were seen between montelukast and placebo over Weeks 3–24: difference were 5.7 (0.0, 11.3) for montelukast (4 mg) minus placebo and 5.9 (0.1, 11.7) for montelukast (8 mg) minus placebo. 
= 0.05, a two-sided test, and an assumed within-patient standard deviation of 30%). Power calculations were based on data from Bisgaard and coworkers (
