Corticosteroid Therapy Yields Mixed Results in Children With Pneumonia

Laurie Barclay, MD

Jan 21, 2011

 

• 
• 
• 
• 
• Print
• Email

Editors' Recommendations

• Pneumonia: Corticosteroids Linked to Longer Hospital Stay
• Impact of Prior Systemic Corticosteroid Use in Patients Admitted With Community-Acquired Pneumonia
• Pneumococcal Vaccination Among Adults With Chronic Respiratory Diseases

Drug & Reference Information

• Aspiration Pneumonia
• Nursing Home Acquired Pneumonia
• Mycoplasmal Pneumonia

January 21, 2011 — Corticosteroid treatment in children with community-acquired pneumonia (CAP) who also have received a beta-agonist benefits those with wheezing but may lead to worse outcomes in children without wheezing, according to the results of a multicenter, retrospective cohort study published in the January issue of Pediatrics.

"Corticosteroids inhibit the expression of many proinflammatory cytokines released during the course of [CAP] infection," write Anna K. Weiss, MD, from Children's Hospital of Philadelphia, Pennsylvania, and colleagues. "Corticosteroids have been found in some studies to be associated with improved clinical outcomes in adults with pneumonia. No studies have investigated corticosteroid use in children with pneumonia."

The study goal was to examine outcomes associated with use of systemic corticosteroid therapy for children hospitalized with CAP, using data from 36 children's hospitals for 20,703 children aged 1 to 18 years (median age, 4 years). The primary exposure was the use of adjunct systemic corticosteroids, and the primary study endpoints were length of stay (LOS), readmission, and total costs of hospitalization. The investigators adjusted for potential confounders using multivariable regression models and propensity scores.

Median LOS was 3 days. Of the 20,703 children, 7234 (35%) received adjunctive treatment with corticosteroids. Readmission was needed in 245 patients (1.2%). Overall, systemic corticosteroid therapy was associated with shorter LOS, with an adjusted hazard ratio (HR) of 1.24 (95% confidence interval [CI], 1.18 - 1.30). Among children who received treatment with beta-agonists, LOS was shorter in children treated with corticosteroids than in children not treated with corticosteroids (adjusted HR, 1.36; 95% CI, 1.28 - 1.45).

In contrast, however, among children not treated with beta-agonists, LOS was longer in those given corticosteroids than in those not receiving them (adjusted HR, 0.85; 95% CI, 0.75 - 0.96), and hospital readmission was more likely (adjusted odds ratio, 1.97; 95% CI, 1.09 - 3.57).

"Results showed that corticosteroid treatment in children with pneumonia is common and its use is highly variable across institutions," the study authors write. "Although corticosteroid therapy may benefit children with acute wheezing treated with beta-agonists, corticosteroid therapy may lead to worse outcomes for children without wheezing."

Limitations of this study include possible inclusion of patients with simple asthma exacerbation rather than CAP, possible unmeasured confounding or residual confounding by indication, and the ability to record only readmissions that occurred at the same hospital as the index admission. In addition, the effect of adjunct corticosteroid therapy on other important outcomes such as progression of illness and the development of pneumonia-associated complications such as empyema could not be evaluated in this study.

"Our results do not support the routine use of corticosteroid treatment of children with CAP," the study authors conclude. "Our findings also have important implications for the design of future clinical trials, particularly with regard to planning of sample size and study cohorts. Because the practice of prescribing adjunct corticosteroids to children with CAP is both common and highly variable, a randomized trial is warranted to allow further exploration of which pediatric populations might benefit from systemic corticosteroid therapy."

The National Institutes of Health supported this study. The study authors have disclosed no relevant financial relationships.

Pediatrics. 2011;127:e255-e263.