 |
|
 |
|
 |
 |
|
 |
 |
 |
 |
 |
 |
 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Davor Plavec, MD, PhD
Department for Pulmonary Function,
Children's Hospital Srebrnjak,
HR-10000 Zagreb, Croatia
To the Editor.—
In their recent Pediatrics article, García García et al1 conclude that montelukast is not inferior to fluticasone in treating mild persistent asthma by using asthma rescue-free days (RFDs) as their primary objective in a 1-year study that involved 6- to 14-year-old patients despite the fact that patients on montelukast had significantly more asthma attacks and required more rescue systemic and/or inhaled corticosteroids. Also, other secondary end points were significantly more improved in the fluticasone group (such as percentage of predicted forced expiratory volume in 1 second [FEV1], days with ß-receptor agonist use, and quality of life). The conclusion based on the primary end point (RFDs) is also doubtful. Although the authors calculated the noninferiority interval on the basis of prior studies and before recruitment, the results presented in their Table 2 for asthma RFDs clearly show that there is a statistically significant difference in the favor of fluticasone; the 95% confidence interval of mean difference between groups does not cross 0 (see also their results for FEV1 as percent predicted for comparison).
Other problems concerning the study protocol and conclusions that the authors draw from the results are also apparent. There is ample of evidence from clinical studies that a different affect is seen with asthma therapy on different outcome measures including symptoms, lung function, exacerbations, need for short-acting bronchodilators, RFDs, and quality of life. Effectiveness of asthma therapy should be assessed by using appropriate outcome measures. The poor correlation between different outcome measures implies that >1 measure is required for the assessment of asthma control. Thus, if we find improved lung function, it does not mean that symptoms are resolved and exacerbations may still occur.2 Thus, there is no simple outcome measure for asthma, and there is a danger if we talk about effectiveness or equality of asthma therapy on the basis of only one end point (as in this case with asthma RFDs). On the other hand, there was robust evidence that fluticasone was more effective in controlling mild persistent asthma in children than montelukast (almost all outcome measures were in favor of fluticasone).
The second problem was the recruitment criteria for mild persistent asthma. A recent study by Becharier et al3 showed that the classification of severity of asthma in children based on Global Initiative for Asthma guidelines (used in the Montelukast Study of Asthma in Children [MOSAIC] study) could lead to a serious mistake and include children from lower or higher severity categories. Results of the study done by Zeiger et al4 show that in patients with the lower level of severity, fluticasone could not show its superiority because improvement of outcome measures reaches a plateau (the same for fluticasone and montelukast), showing noninferiority in that subgroup. Zeiger et al also found that RFDs improved similarly over the short-term, but with prolonged open-label treatment, asthma control improved more in the fluticasone-treated group. On the basis of their results, they also conclude that classification criteria for mild persistent asthma may need to be reevaluated.
A number of approaches have been used to classify the severity of asthma. Following the Global Initiative for Asthma guidelines, asthma severity is classified before treatment is given on the basis of frequency of symptoms, medication use, and lung-function measures.5 According to the National Asthma Education and Prevention Program's guidelines for the diagnosis and management of asthma, asthma can be divided into 4 levels of severity: mild intermittent, mild persistent, moderate persistent, and severe persistent.6 The guidelines provide criteria for classifying FEV1 (as percent predicted) into 3 levels.5,6 According to the results of the Childhood Asthma Management Program Research Group, children with asthma may have a higher level of lung function than suggested in the guidelines' classification scheme.7
Furthermore, pulmonary-function measures are not always consistent with levels of asthma severity. Isolated measures of lung function, especially FEV1, are of little value in determining asthma severity, especially among patients who receive asthma-controller therapy3; on the other hand, objective measures of lung function are essential in severity criteria of asthma. It is also well known that spirometric assessment of lung function is not always interpreted correctly, especially in younger children. The use of controller medications complicates the assignment of asthma severity.
Serious consequences may arise from the misclassification of asthma severity. Many patients who are considered to have mild asthma may be undertreated, resulting in 
2 emergency-department visits per year that were required for 20% of patients with mild asthma.8 Undertreatment has an adverse impact on quality-adjusted life-years, morbidity, mortality, and treatment costs. Also, it could have an important impact on the results of the MOSAIC study (recruitment bias) if more severe patients would be recruited in the montelukast group. This would allow montelukast to show greater improvement, and fluticasone would plateau. Because improvements in both groups were compared in the MOSAIC study, this could lead to noninferiority results for montelukast.
The third problem was the conclusion about the safety of fluticasone compared with montelukast in asthmatic children. The authors of the MOSAIC study conclude that the patients in the fluticasone group had significantly lower (–0.41 cm) final height compared with those in the montelukast group even though this was not a planned comparison according to the published protocol. This conclusion is severely biased with the fact that groups were not of the same age span, there were more boys (64.8% vs 58.5%) in the montelukast group, and the age span also included adolescents in pubertal growth acceleration. Nothing is known about the actual sexual development status of the included patients between groups. Inclusion of more boys that reach higher final height and more subjects in the pubertal accelerated-growth phase could significantly bias the final result. Because the factors were not controlled and the randomization procedure was not planned to control for these factors that are very important for height velocity, no relevant conclusions could be made on the basis of the results of the MOSAIC study.
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
||
|
||
Home | My Pediatrics | Journal Information | Current Issue | Past Issues | Subscriptions & Services | Contact Us Click here for faster international access © 2006 American Academy of Pediatrics. All rights reserved. |
||
|
||
