Limitations and strengths
Our study has several limitations. First, it is single center, which limits its external validity. Second, the retrospective design increases the risk of information bias. Third, the use of corticosteroids was not strictly protocolized so that the criteria for initiating corticosteroids may have varied from one child to another. Fourth, we only collected data during the hospital stay: therefore, we cannot exclude that corticosteroids are associated with worse outcomes after hospital discharge. However, even though children with parapneumonic effusions may have severe clinical, radiological and lung function abnormalities during the acute phase of the disease, favorable and complete recovery is the rule in the months following infection29,30. Therefore, the benefits of corticosteroids should be expected during the hospital stay or maybe in the immediate post-hospital phase, but not in the long term. Fifth, microbiological, clinical and management changes may have occurred during the 15-year study period, and data observed 15 years ago may not be representative of what occurs nowadays. The main strengths of our study are the size of our cohort, and most importantly the fact that our study is the first to describe the successful use of corticosteroids in the management of parapneumonic effusion when non-surgical management is considered a failure.