Outcome after early respiratory infection: Prospective, population‐based designs needed

Abstract

Santos et al. published their interesting results on the association between hospitalization for pneumonia without wheezing at <24 months of age and subsequent asthma up to preschool age. 1 The cases were retrospectively charted from the registers of two hospitals, and two controls were matched for each case. The number of cases, however, was only 55, and the authors did not describe whether the controls were collected from population registers (population-based data collection) or from hospital registers (hospital-based data collection). 1,2 The selection of controls is important and carries a similar risk for biased data to that of the cases. 3 The cases of pneumonia were confirmed on radiology but the radiological criteria or the principles of radiological evaluation were not given. The patients with wheezing on admission were excluded. Asthma diagnoses and treatments up to preschool age were confirmed by doctors but reported by parents. Only 36 (56%) of the children with early-life pneumonia attended follow up. The authors have also reported recently that hospitalization for pneumonia had no influence on the occurrence of respiratory infections up to preschool age. 2 The main result of the study was that the prevalence of asthma ever up to preschool age was higher in children hospitalized for pneumonia at <24 months of age than in controls (41.6% vs 22.6%), with an adjusted prevalence ratio of 2.06 (95% confidence interval: 1.10-3.62). In addition, there was a tendency towards more asthma and treatment for asthma at preschool age, but these findings were not statistically significant. 1 The authors included potential confounders in the model with no description of which adjustments they used and why.The authors described the main shortcomings of the study as being the retrospective design, the small number of cases leading to under-powering of the study and the lack of data concerning other respiratory infections in early childhood other than the index episodes leading to hospitalization. 1 In addition, the authors noted that the exclusion of children who wheezed on admission reduced later asthma prevalence figures. This is true, of course, but the purpose of the study was to evaluate if there was an association between early-life pneumonia and preschool asthma independently from early-life wheezing. The link from early-life wheezing to later asthma is well established. 4 Additional shortcomings, not discussed by the authors, were the poor attendance rate, only 56%, and the lack of radiological classification of chest radiographs. The etiology of pneumonia, especially pneumococcal etiology, is difficult to assess in clinical practice and requires advanced methods applied in a prospective study setting.The age limit used for early-childhood pneumonia or other respiratory infections has been 24 months or even 36 months in many studies. 4 In post-bronchiolitis studies, the outcome has been highly age dependent. The prevalence of asthma at preschool age has been significantly higher in children admitted at age >1...

Pneumonia during the first 2 years of life and asthma in preschool‐age children

“…We thank Korppi 1 for his comments on our study 2 . We would like to address the following issues raised by him.…”