Comparison of bronchodilator responsiveness in asthmatic children using 2021 or 2005 ATS/ERS guidelines

Detection of bronchodilator responsiveness, along with spirometric airflow obstruction, are cornerstones of the diagnosis of asthma in children.1 The 2005 ATS/ERS interpretative strategies for lung function tests defined a positive bronchodilator response (BDR) as an increase in the forced expiratory volume in 1 s (FEV1) or forced vital capacity (FVC) of ≥12% of the baseline value and ≥200 mL (12%Base).2 The 2021 ERS/ATS technical standard on lung function interpretative strategies recommended changing the definition of a positive BDR to an increase in FEV1 or FVC of >10% of the predicted value (10%Pred). The evidence upon which this recommendation was based did not include children. The upper limit of the range of BDR in FVC and FEV1 in healthy adults has been reported as 9.6%–11.6% of the predicted value in various studies; this range is not known in children. In addition, a threshold for BDR relative to the predicted value has been associated with survival in adults but not evaluated in children. Our objective was to evaluate the concordance of BDR according to the two definitions in a cohort of children and to explore the characteristics of FEV1 and FVC associated with discordance in BDR classification according to the two definitions. We conducted a retrospective, single-center study using anonymized 2021 data from the pulmonary function laboratory of APHP-Armand Trousseau Hospital, Paris, France. At our institution, BDR is routinely evaluated only in those with a diagnosis of asthma or chronic cough, so the cohort was limited to children with these diagnoses. Spirometry (BodyBox; Medisoft) was performed according to the ATS/ERS3 standard by experienced technicians. Results were expressed as z-scores based on the Global Lung Function Initiative reference ranges.4 The Kruskal–Wallis test was used for between-group comparisons and the χ2 test for comparisons of proportions (GraphPad Prism® version 6.01). The lung function database complies with the French authority for data protection (CNIL) and with the General Data Protection Regulation (GDPR) for personal data protection. Children and their parents were informed of the possible use of the childrens' anonymized lung function results and did not opt out. The study was approved by the Institutional Review Board of the French Société de Pneumologie de Langue Française (CEPRO 2023-007). In 2021, 3177 spirometry tests were conducted at our institution. Including only children referred for asthma or chronic cough, we retrieved 1365 tests with baseline and post-BD spirometry measurements among 1224 children. The mean (SD) age at first test in 1224 was 11.6 (3.3) years, range: 5.4–18 years; 1188 (97%) children had a diagnosis of asthma; 429 (35%) were female; 822 (67%) were of European ancestry, 260 (21%) African or Caribbean, 9 South-East Asian (0.7%), and 133 (11%) other. The distribution of BDR according to the 2005 (12%Base) versus 2021 (10%Pred) definitions is shown in Table 1. Among the 1365 tests, 259 (19%) were BDR+ by one or both definitions. More tests were classified as BDR+ using the 10%Pred definition compared to the 12%Base definition (n = 245/1365, 17.9% vs. 201/1365, 14.7%; p = .023). The distribution of baseline spirometric indices according to BDR definition is shown in Figure 1. Tests classified as BDR+ only by the 2021 definition (12%Base−/10%Pred+; n = 58) had higher baseline FEV1, FVC, and FEV1/FVC z-scores than those classified as BDR+ only by the 2005 definition (12%Base+/10%Pred−; n = 14) (p < .0001). Tests classified as BDR+ only by the 2021 definition (12%Base−/10%Pred+) were more likely to have a normal FEV1 or FEV1/FVC than those fulfilling both BDR+ definitions (10%Base+/12%Pred+; n = 187) (p < .00001). Lastly, all of the 72 tests with discordant classification of BDR based on change in FEV1 were similarly discordant based on FVC. Among children with a diagnosis of asthma or chronic cough, we found that implementing the 2021 recommendation for BDR interpretation leads to a small but significant increase in the number of tests classified as BDR+ (among 1365 tests, n = 201 classified as BDR+ by the 2005 guidelines vs. n = 245 by the 2021 guidelines). Tests classified as BDR+ by the 2021 but not the 2005 guidelines were generally those with the highest baseline FEV1 and without airflow obstruction. Interestingly, Suh et al.5 showed, in a cohort of 138 children with mild to moderate asthma (baseline FEV1 90.2 ± 13.1%Pred) that almost all children with an FEV1 that increased by 9% or more of the predicted value after albuterol had bronchial hyperresponsiveness to methacholine (46/47) or to AMP (45/47). Therefore, by identifying a positive BDR in children with normal baseline spirometry, the new definition could be useful in diagnosing asthma in those with a compatible clinical phenotype. As expected, we found that the very small number of children (n = 14) reclassified from positive BDR by the 2005 definition to negative by the 2021 definition were those with the lowest baseline FEV1 z-score, in whom a small increase in FEV1 postbronchodilator would meet the 12%Base criterion but not the 10%Pred criterion. These children all had an abnormally low baseline FEV1 (z-score < −1.64) and in 13/14 cases had baseline airflow obstruction. Further studies will be needed to evaluate the clinical implications of reclassifying these individuals as having fixed rather than reversible airflow obstruction. This is the first study on this subject conducted in a large pediatric cohort. In agreement with our results, in a retrospective study of spirometry in adults (n = 3983) and children (n = 475) with asthma, Li et al.6 recently showed that children were slightly more likely to be classified as BDR+ with the 2021 compared to the 2005 definition (64.6% vs. 63.1%) while the opposite trend was observed in adults (51.4% vs. 63.3%), reinforcing that BDR results from adults cannot be extrapolated to children. Our study has some limitations. It is a single center, retrospective study, though the technicians and physicians working in our pediatric lung function test laboratory follow international standards for spirometry performance.3 In addition, it focuses on spirometry, without detailed clinical or treatment data. Finally, the children included in this cohort did not have restrictive lung disease due to chest wall deformities or neuromuscular disease, so our results cannot be generalized to these populations. In conclusion, adoption of the 2021 10%Pred definition for a positive BDR results in a small increase in the proportion of tests classified as positive. The likelihood of reclassification from BDR − to + is highest in children with normal FEV1 and no or minimal airflow obstruction. In addition, with the new definition, a small number of children will be reclassified from BDR + to −. These children typically have a low baseline FEV1 so a 12% improvement from baseline is much smaller than 10% of the predicted FEV1. Adopting the 2021 BDR definition in children could help in the diagnosis of asthma in children with normal or near-normal lung function and allow reconsideration of the therapeutic approach for children with more significant airflow obstruction. Further studies may focus on different patient populations and evaluation of longitudinal clinical outcomes. Nicole Beydon: Conceptualization; investigation; writing — original draft; methodology; software; formal analysis; data curation; writing — review & editing; validation; supervision. Margaret Rosenfeld: Conceptualization; writing — original draft; methodology; validation; writing — review & editing; supervision; formal analysis; investigation. This work received no funding. The authors declare no conflict of interest. The data that support the findings of this study are available from the corresponding author upon reasonable request.