Key spirometric determinants of future airflow obstruction in children with asthma.

Background: There is increasing concern about children with asthma developing progressive lung function impairment. The objective of the present study was to identify the best spirometric determinants of subsequent development of airflow obstruction (AO) in children with asthma in the clinic setting. Methods: We assembled two retrospective cohort studies of children aged 6-17 years, managed in tertiary-care asthma clinics, with medical and drug coverage, and repeated spirometry testing. The primary outcome was AO, defined as pre-bronchodilation (pre-BD) forced expiratory volume in 1 s (FEV1)/forced vital capacity (FVC) ratio below the lower limit of normal (LLN). Multiple lung function parameters, prior to index visit, were adjusted for potential covariates/confounders in multivariable logistic regression models, by cohort and clinical scenario (≥1 versus ≥2 prior visits with spirometry); cohort estimates were meta-analysed using inverse-variance-weighted average. Results: Of 509 eligible children (mean age: 10 years), 17% subsequently developed AO. In patients with ≥1 prior visit, the likelihood of future AO independently increased by almost 4-fold (adjusted OR 3.91 (95% CI 2.54-6.01)) for every 1 z-score lower FEV1/FVC ratio. In patients with ≥2 prior visits, the likelihood of future AO increased by 3.31 (1.98-5.54) for every 1 z-score lower FEV1/FVC ratio at the last visit and by 1.50 (1.10-2.12) for every 1 z-score maximum between-visit variation in FEV1. Interpretation: Two spirometric parameters independently increased the likelihood of subsequently developing AO, namely FEV1/FVC in the low range of normal and high between-visit FEV1 variation, appearing as practical determinants of future impairment, before reaching the LLN.