The Association between Oral Antibiotic Use and Asthma Exacerbation in Children seen in an Ambulatory Care Setting

Background: It is unclear whether the use of oral antibiotics in ambulatory settings may trigger asthma exacerbation, via gut dysbiosis, in children with asthma.

Objectives: To determine if exposure to oral antibiotics is associated with time-to-first exacerbation in children with newly diagnosed asthma.

Methods: We conducted a retrospective cohort study using the Quebec Asthma Pregnancy and Infant Database (1990-2010). Included children were 0-18 years old and treated for a first acute infection in the ambulatory care setting within the 2 years following a first asthma diagnosis. Only acute infections satisfying the positivity assumption, i.e. a probability of receiving antibiotics not equal to 0 or 1 via the propensity score, were included. Cohort entry (CE) was defined as 14 days after the first acute infection visit (T0) to avoid reverse causation from an exacerbation present at T0. Children not covered by Quebec’s public drug insurance plan, with other lung diseases, or receiving antibiotics 6 months prior to T0, were excluded. Children were followed from CE until the outcome date, end of drug insurance, or the 6-month follow-up period, whichever occurred first. The primary outcome was time-to-first asthma exacerbation, defined as any dispensed oral corticosteroid, emergency department visit, or hospitalization for asthma. Children receiving and not receiving oral antibiotics between T0 and CE were considered exposed and not exposed, respectively, from CE until the end of follow-up. Marginal structural Cox proportional hazards models, adjusted for confounders, estimated marginal hazard ratios (HRs) of asthma exacerbation for children exposed vs unexposed to any and spectrum-specific antibiotics.

Results: Of the 7854 children included, 4103 (52.2%) were exposed to antibiotics and 903 (11.5%) had exacerbation events over 42032 person-months of follow-up (crude incidence rate of 2.15 per 100 person-months). In children exposed vs unexposed to any antibiotic, the marginal hazard of exacerbation increased by 14% (HR 1.14, 95% CI 0.95-1.35). The HRs for children exposed to narrow-spectrum antibiotics and broad-spectrum antibiotics were 0.93 (95% CI 0.77-1.12) and 1.16 (95% CI 0.98-1.37), respectively, compared to those unexposed to antibiotics.

Conclusions: In children recently diagnosed with asthma, those receiving an antibiotic, particularly broad-spectrum, had shorter average time-to-first exacerbation than those not receiving antibiotics for acute infection indications that do not necessarily require them. Since inappropriate antibiotic prescribing rates are highly prevalent in children with asthma, our results suggest this practice could contribute to worse asthma outcomes in children.