© 2005 American Thoracic Society doi: 10.1164/rccm.2502003
Inhaled Corticosteroid Dose Reduction in Childhood AsthmaIs Nitrosopnea Informative?University of Virginia Pediatrics Charlottesville, Virginia Inhaled corticosteroid therapy represents the cornerstone of antiinflammatory asthma management (1). However, airway inflammation is not monitored during inhaled corticosteroid therapy. Instead, dosing is determined by symptom history and by spirometry, variables that are not consistently related to the degree of airway inflammation (1, 2). This situation is a bit like following symptoms and serum creatinine while treating nephritis, without examining urine sediments or obtaining a renal biopsy. The inability to monitor the primary therapeutic outcome in asthma is particularly concerning in the case of children, in whom unnecessarily high doses of inhaled corticosteroids have the potential to adversely affect linear growth (1, 3). In this issue of the AJRCCM (pp. 1077–1082), Zacharasiewicz and coworkers present data suggesting that successful lowering of inhaled corticosteroid dose in children can be predicted by two markers demonstrating quiescent inflammation: (1) an expired nitric oxide (eNO) level below 22 ppb, and (2) an absence of eosinophils in induced sputum (4). There is agreement that the two measures of high eNO levels (hypernitrosopnea [5]) and increased numbers of sputum eosinophils are associated with one another in asthma, and that they both decrease with corticosteroid therapy (6–10). Though the relationship of these markers to asthmatic airway inflammation is complex (7, 9–11), limited data suggest that one or both measures may be useful in predicting the success of inhaled corticosteroid dose reduction in adults (12, 13). Zacharasiewicz and coworkers now extend these observations to a population of 40 children, 6 to 17 years of age, with moderate asthma (4). Weaning decisions were made based on clinical symptoms and spirometry at 2-month intervals. Weaning failure was defined as a change in weekly albuterol dosing requirements from less than three times a week to more than five times per week, and/or a requirement for systemic corticosteroids. Using receiver operator characteristic curves, the authors determined that both an eNO value less than 22 ppb and an absence of sputum eosinophils had strong predictive values for identifying children who would succeed at corticosteroid weaning. However, the positive predictive values for identifying children (above these thresholds) who would fail weaning were only in the range of 40 to 50%. Put another way, if a child had values below these thresholds, it appeared to be reasonably safe to decrease the dose of corticosteroids; however, if the values were above these thresholds, it was less clear whether or not the dose could be reduced. Of note, neither atopy nor exhaled breath concentrations of leukotrienes or nitrogen oxides were helpful in predicting exacerbations. Bronchial hyperreactivity was potentially helpful, but was not as strong a marker as eNO or sputum eosinophil percent. Sputum induction is not consistently successful in children (4, 14, 15). Further, the process of obtaining and analyzing induced sputum is time consuming, labor intensive, and unpleasant. Nitrosopnea measurement, on the other hand, is faster and generally easier (7, 8). Therefore, the principal practical application of the Zacharasiewicz data may be to suggest that an eNO level less than 22 ppb, in the context of good clinical control and spirometry, may provide modest reassurance that inhaled corticosteroid dosing can be successfully reduced—at least for 2 months—in children with moderate asthma. There are several important caveats that must be raised concerning the Zacharasiewicz data. First, the population was inhomogeneous, and included both adolescent and pre-adolescent children. The subjects were predominantly (75%) male. The initial dose of inhaled corticosteroid at enrollment was variable. The population was small, and only 15 subjects experienced an exacerbation. These factors limit the power of analyses (for example, of the confounding effect of age/puberty on eNO). Moreover, there was not clear accounting for seasonality, viral infections, or patient adherence, and some parents refused to follow the recommendation to decrease dosing. It is important to remember that there are many determinants of eNO concentration. Hypernitrosopnea does not simply reflect increased activity of inducible nitric oxide synthase (an effect primarily regulated by Th1, as opposed to Th2, cytokines). The activity of airway neuronal nitric oxide synthase, pH regulatory enzymes, S-nitrosothiol metabolic enzymes, prokaryotic colonizing species, and superoxide production can each influence eNO (7, 9). Moreover, there is a substantial population of children with severe asthma who have quite low eNO (16). These hyponitrosopneic children have both low breath condensate pH and extremely high breath condensate levels of nitrogen oxides. Such children might be characterized as having a "fire in the airways," with NO being oxidized ("burned off") at low pH to form cytotoxic species such as peroxynitrous and nitrous acids (9). These data also suggest that asthma—particularly severe asthma—is not one disease, but is a collection of complex biochemical abnormalities superimposed on allergic and other types of airways inflammation (2, 9). In this sense, noninvasive inflammometry may serve to characterize asthmatic subphenotypes amenable to specific types of therapy and/or inhaled corticosteroid dose reduction. Analyses of data from large asthma network populations will be required to determine which markers of inflammation will predict successful steroid weaning in well defined subsets of children. In conclusion, the data of Zacharasiewicz and colleagues suggest that children on inhaled corticosteroids whose asthma is clinically stable, and who have an eNO concentration less than 22 ppb, may tolerate a reduction in inhaled corticosteroid dose for at least a period of 2 months. The results of this study also suggest that the absence of sputum eosinophils is a useful predictor of successful inhaled corticosteroid weaning, and that the degree of bronchial hyperresponsiveness adds information. Because determination of both of these latter measures is cumbersome and somewhat unpleasant—particularly for repeated assessment at serial clinic visits—the eNO threshold may be the most useful inflammation gauge in the pediatric lung function laboratory. However, the data fall short of being sufficient to support formal recommendation. At best, the eNO threshold represents a nonspecific marker for inflammation; it is only the first step in the long journey toward scientific management of anti-inflammatory therapy based on measurement of airway inflammation. FOOTNOTES Conflict of Interest Statement: B.G. is a minority stockholder in Respiratory Research, Inc. REFERENCES
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
