Association of Oral Corticosteroid Bursts With Severe Adverse Events in Children
Children are at higher risk for pneumonia (and at risk for GI bleed and sepsis) after burst steroid therapy of 14 days or less
Design
- Association of short course (burst) corticosteroid therapy to 4 major adverse events
- After a short duration of corticosteroid therapy, each participants act as his/ her own control.
- Events rate at 5-90 days prior to initiation of a corticosteroid burst were compared to 2 post-treatment periods (5-30 days and 31-90 days after treatment)
- 4 days washout period (events that occurred during a 4-day window both before and after corticosteroid use were dismissed as maybe due to other factors
Primary outcome
Risk of developing
- GI bleed
- sepsis
- pneumonia
- glaucoma
Secondary outcome
Subgroup analysis (2 groups)
- Those who received corticosteroid bursts for <7 days
- Those who received corticosteroid burst for >7 to 14 days
Population
- 4542623 pt record looked at, 23% included = 1064587 (51% boys vs 49% girls)
- De-identified medical claims records and prescription data from entire Taiwanese National Insurance Research Database (NHIRD) 1st Jan 2013 to 31st December 2017.
- 99% of their population is covered by this Medicare like scheme
- 1897858 (42% received at least 1 corticosteroid bursst during 5 -years study period)
Inclusion
- Continous use of oral corticosteroids for 14 days or less, as recorded on all prescription records.
Exclusion
- Pt >18 yo
- Rx systemic or tropical corticosteroid prior to 2013
- Diagnosis of primary outcome (GI bleed, sepsis, pneumonia or glaucoma prior to 2013)
- Received >1 corticosteroid burst during the observation period
- Continous oral corticosteroid Rx for >14 days
- Congenital anomalies or catastrophic illnesses
Analytic methods
- Self-controlled case series (participant act as their own control)
- Incidence rate ratios (IRRs) is calculated by comparing incidence rates of the severe adverse events within each post-treatment period with the incidence rates of the severe adverse events within the reference period.
Results:
- IRRs per 1000 person-years between study and control group is most significant for pneumonia = 9.35 (95% CI 9.19-9.51)
Conclusion
- Children are at higher risk for pneumonia (GI bleed and sepsis) after burst steroid therapy, the magnitude of increased risk is still very small in this study.
Practice implication
- The result challenge the conventional wisdom that burst corticosteroid are totally benign. In these children where steroid are given for croup, the symptomatic benefits may still reasonably outweigh the risks, in the right group of patient.