Association of early viral lower respiratory infections and subsequent development of atopy, a systematic review and meta-analysis of cohort studies

INTRODUCTION: Existing evidence on the relationship between childhood lower respiratory tract infections (LRTI) and the subsequent atopy development is controversial. We aimed to investigate an association between viral LRTI at <5 years and the development of atopy at > 2 years. METHODS: We conducted a search at Embase, Pubmed, Web of Science, and Global Index Medicus. We collected data from the included articles. We estimated the odds ratio and the 95% confidence intervals with a random effect model. We determined factors associated with atopy development after childhood LRTI using univariate and multivariate meta-regression analyses. We recorded this systematic review at PROSPERO with the number CRD42018116955. RESULTS: We included 24 studies. There was no relationship between viral LRTI at <5 years and skin prick test-diagnosed-atopy (OR = 1.2, [95% CI = 0.7–2.0]), unknown diagnosed-atopy (OR = 0.7, [95% CI = 0.4–1.3]), atopic dermatitis (OR = 1.2, [95% CI = 0.9–1.6]), hyperreactivity to pollen (OR = 0.8, [95% CI = 0.3–2.7]), food (OR = 0.8, [95% CI = 0.3–2.5]), or house dust mite (OR = 1.1, [95% CI = 0.6–2.2]). Although not confirmed in all studies with a symmetric distribution of the 23 confounding factors investigated, the overall analyses showed that there was a relationship between childhood viral LRTI at < 5 years and serum test diagnosed-atopy (OR = 2.0, [95% CI = 1.0–4.1]), allergic rhinoconjunctivitis (OR = 1.7, [95% CI = 1.1–2.9]), hyperreactivity diagnosed by serum tests with food (OR = 5.3, [1.7–16.7]) or inhaled allergens (OR = 4.2, [95% CI = 2.1–8.5]), or furred animals (OR = 0.6, [95% CI = 0.5–0.9]). CONCLUSION: These results suggest that there is no association between viral LRTI at < 5 years and the majority of categories of atopy studied during this work. These results, however, are not confirmed for the remaining categories of atopy and more particularly those diagnosed by serum tests. There is a real need to develop more accurate atopy diagnostic tools.