Assessment of corticosteroid response in pediatric patients with severe asthma by using a multidomain approach.

Abstract

There is no agreed upon definition of systemic corticosteroid response in asthmatic children. Moreover, pediatric severe therapy-resistant asthma (STRA) is heterogeneous, and thus response to steroids is unlikely to be uniform in all patients. We sought to evaluate the utility of a multidomain approach incorporating symptoms, lung function, and inflammation to determine steroid responsiveness in pediatric patients with STRA. Eighty-two children (median age, 12 years) with STRA received a clinically indicated dose of intramuscular steroid. Changes in 4 separate domains were assessed 4 weeks after intramuscular triamcinolone acetonide: normalization of (1) symptoms (Asthma Control Test score, >19/25 or 50% increase), (2) spirometric results (FEV1 ≥80% of predicted value or ≥15% increase), (3) fraction of exhaled nitric oxide levels (<24 ppb), and (4) sputum eosinophil counts (<2.5%). Fifty-four of 82 children had complete data in all 4 domains. Twenty-three (43%) of 54 children had a symptom response, 29 (54%) of 54 had a lung function response, 28 (52%) of 54 had a fraction of exhaled nitric oxide response, and 29 (54%) of 54 had a sputum eosinophil response. Although a similar proportion of children responded to systemic corticosteroids in each domain, there were no reliable predictors of a response pattern. Seven (13%) of 54 were complete responders (response in all domains), 8 (15%) of 54 were nonresponders (no response in any domain), and 39 (72%) of 54 were partial responders (response in ≥1 domain). A multidomain evaluation of systemic steroid responsiveness using pragmatic clinical assessments confirms childhood STRA is heterogeneous and that a complete response in symptoms and inflammatory and physiologic parameters is rare. Individual response patterns to systemic steroids might be useful in guiding the choice of add-on therapies in each child as a step toward achieving personalized medicine.