Infections with viruses causing upper respiratory tract infection (URI) are associated with increased leukotriene levels in the upper airways. Montelukast, a selective leukotriene-receptor antagonist, is an effective treatment of asthma and allergic rhinitis.
To determine whether prophylactic treatment with montelukast reduces the incidence and severity of URI in children.
A randomized, double-blind, placebo-controlled study was performed in 3 primary care pediatric ambulatory clinics in Israel. Healthy children aged 1 to 5 years were randomly assigned in a 1:1 ratio to receive 12-week treatment with 4 mg oral montelukast or look-alike placebo. Patients were excluded if they had a previous history of reactive airway disease. A study coordinator contacted the parents by phone once a week to obtain information regarding the occurrence of acute respiratory episodes. The parents received a diary card to record any acute symptoms of URI. The primary outcome measure was the number of URI episodes.
Three hundred children were recruited and randomly assigned into montelukast (n = 153) or placebo (n = 147) groups. One hundred thirty-one (85.6%) of the children treated with montelukast and 129 (87.7%) of the children treated with placebo completed 12 weeks of treatment. The number of weeks in which URI was reported was 30.4% in children treated with montelukast and 30.7% in children treated with placebo. There was no significant difference in any of the secondary variables between the groups.
In preschool-aged children, 12-week treatment with montelukast, compared with placebo, did not reduce the incidence of URI.
Letter in response to: Kozer et al. RCT of montelukast as prophylaxis for upper respiratory tract infections in children
Daniel Golshevsky1, Amy Linklater1, Sarah Arnup2, Kim Jachno2, Tom G Connell1,2 1Department of General Medicine, Royal Children's Hospital, Melbourne, Victoria, Australia 2Murdoch Children's Research Institute and Department of Paediatrics, University of Melbourne
Dear Editor, Randomized controlled trials (RCTs) in children with asthma have shown that compared to placebo, the daily administration of montelukast in addition to regular inhaled glucocorticoid treatment improves FEV1 and decreases the requirement for rescue medications (1). Furthermore, the intermittent use of montelukast has been associated with modest reductions in health-care resource utilization of health care resources and asthma interval symptoms (2-3). Preceding upper respiratory tract infection (URTI) is the most common trigger for an acute exacerbation of asthma in children but there are currently no data on the use of montelukast in preventing URTIs in 'healthy' children. We therefore read with interest the recent manuscript by Kozer et al that investigated the use of montelukast as prophylaxis in preventing URTIs in healthy children (4). Although we commend Pediatrics for publishing the results of this industry -sponsored trial that reported negative results, we have several concerns regarding the study design and methodology, data collection and analysis and urge caution in interpreting the overall results of the study.
Whilst the authors do provide some rationale for using montelukast in preventing URTI in healthy children, the overall design of their study is interesting. In the study, healthy children aged between 1 to 5 years were randomly assigned to receive either montelukast or placebo for a 12-week period (in an unspecified season) with the primary outcome being the number and duration of URTI episodes in children. Patients with known asthma were excluded. Although randomisation procedures were sound and the two groups well balanced demographically, the collection of follow-up data was suboptimal with less than one-third of parents completing a diary of URTI episodes in their children; questioning the validity of the findings presented and increasing the likelihood of recall bias. The decision to preferentially weigh complete data, compared with incomplete reported data is questionable and the rationale for such an approach was not provided.
The authors aimed to recruit 500 children to detect a difference of between 1.5 to 2 days in the duration of URTI episode between montelukast and placebo but due to poor recruitment the trial was terminated early. This was surprising given the number of patients included in the overall randomization was relatively large but we are not provided with the details of the overall time period during which children were recruited.
The presentation of the data in table 2 of the manuscript is confusing. Without providing an indication of the amount of missing data in calculating the average for each parameter analysed, the data presented in table 2 of the manuscript cannot be assumed to be representative of the entire study population. One way of handling missing data are to assume there were no occurrences of the parameter when the data was missing, or to assume that the frequency of each parameter was the same during the weeks when no data was obtained. Presenting the data as the average number of weeks when the parameter of interest occurred would have been preferable. The authors do not provide information on how outcome variables that occurred during a crossover period between weeks were counted.
Although the study was underpowered, the data presented in table 2 and table 3 of the manuscript show that children in the montelukast group were more likely to be admitted to hospital, have more febrile episodes, use more antibiotics and anti-pyretics compared to children in the placebo group, facts only briefly discussed by the authors in the discussion.
In their discussion, the authors highlight several medical conditions in children where montelukast has been shown to potentially beneficial. However, in the current study the authors' recruited 'healthy' children with no medical conditions. Despite being underpowered, had the current study reported a potential benefit for montelukast for the primary outcome, it is conceivable the authors may have advocated daily montelukast treatment for the prevention of URTI in 'healthy' children with no respiratory disease. The potential overuse of montelukast in this situation could be considered an example of therapeutic creep (5), where a treatment with beneficial effect in one group (ie children with asthma) is used in children where no efficacy has been shown (ie healthy children). A recent example of this has been the use of proton pump inhibitors in children with poorly controlled asthma from studies that showed a potential causal association with reflux (6). Importantly, a recent well- designed RCT not only showed no benefit of lansoprazole in poorly controlled asthma but highlighted significant potential safety concerns (7).
Kozer et al recognise the limitations of their study and conclude that montelukast did not reduce the number of URTI episodes in children. We await the results of future prospective well powered and designed RCTs with appropriate longitudinal follow-up to determine whether there will be any clinical benefit in administering montelukast to healthy children.
References 1. Bisgaard H, et al Montelukast reduces asthma exacerbations in 2- to 5- year-old children with intermittent asthma. Am J Respir Crit Care Med. 2005;171(4):315. 2. Knorr B, et al Montelukast for chronic asthma in 6- to 14-year-old children: a randomized, double-blind trial. Pediatric Montelukast Study Group. JAMA. 1998;279(15):1181. 3. Robertson CF, et al Short-course montelukast for intermittent asthma in children: a randomized controlled trial. Am J Respir Crit Care Med. 2007;175(4):323. 4. Kozer E, et al RCT of montelukast as prophylaxis for upper respiratory tract infections in children Pediatrics. 2012;129(2):e285-90. Epub 2012 Jan 4. 5. Martinez FD Children, Asthma and Proton Pump Inhibitors: Costs and Perils of Therapeutic Creep. JAMA. 2012;307(4):406-407. 6. Debley JS Prevalence and impact of gastroesophageal reflux in adolescents with asthma: A population-based study. Ped Pulm. 2006;41(5):475-481. 7. Holbrook JT et al. Lansoprazole for Children With Poorly Controlled Asthma. A Randomized Controlled Trial. JAMA 2012;307(4):373-380.
Conflict of Interest:
None declared