Comparison of three methods assessing spirometry bronchodilator responsiveness in children

Abstract

Introduction/Aim: There is limited consensus in the literature about what constitutes significant bronchodilator responsiveness (BDR) or how BDR should be expressed. According to the BTS asthma management guidelines (2019) a clinically significant BDR is a change in post-bronchodilator FEV 1 of ≥12% relative to baseline values (Method 1). The 2005 ATS/ERS interpretative strategies (Pellegrino, 2005) recommend a ≥12% and ≥200 mL increase in FEV 1 and/or FVC from baseline values (Method 2). The new ERS/ATS (Stanojevic, 2022) interpretative strategies recommend BDR be classified as a change of ≥10% relative to the predicted value for FEV 1 or FVC (Method 3). Nearly all the published BDR recommendations were derived from studies performed on adults. Our aim was to evaluate BDR in children using these three published methods. Method: A retrospective cross-sectional analysis was performed on spirometry data collected at the Queensland Children's Hospital from 2019 to 2022. Descriptive statistics were used to summarise cohort demographics, and multivariable logistic regression to assess the influence of age, sex, height and baseline FEV 1 (% predicted) on BDR. Results: 1376 acceptable pairs of pre- and post-bronchodilator spirometry tests were analysed (53% male, 11 ± 3.4 years). Of the 428 responders (positive BDR by any method), 393 (92%) children had a positive BDR using Method 1, 297 (69%) using Method 2, and 343 (80%) using Method 3 (p = 0.03 vs. Method 1). When mutually adjusting for age, sex, height and baseline FEV1, BDR determined using Methods 1, 2, and 3 was influenced by age and sex, while Method 2 was further influenced by height (Table 1). For each method, a BDR was ≥30% more likely in males. Conclusion: Age and sex continued to influence BDR even after adjustment for predicted values generated using the GLI-2012 reference equations for all three methods, including for Method 3, previously reported to minimise biases in BDR assessment. Further investigation is needed to validate these approaches in children.