Clinical Implications of the New Criteria for a Positive Bronchodilator Response in Children With Asthma

Abbreviations

BD

bronchodilator

FEV₁

forced expiratory volume in one second

FVC

forced vital capacity

Spirometry aids in diagnosing and monitoring asthma by assessing airflow limitations and bronchodilator (BD) responses [1]. The American Thoracic Society and European Respiratory Society recently updated BD response criteria, now defining a positive response as a > 10% increase in FEV₁ (post‐BD) relative to the predicted value. Previously, it required a > 12% and ≥ 200 mL increase from pre‐BD values [1]. This change aims to reduce biases related to age, sex, and height, but is based on adult data and unvalidated in children. Paediatric studies have indicated that the new BD response could lead to overdiagnosing asthma [2]. One study reported that children who tested positive, based on the new criteria, had a higher baseline FEV₁, forced vital capacity (FVC), and FEV₁/FVC [3].

We used data from a real‐world paediatric, specialist care asthma cohort to explore the concordance between the old and new BD response criteria and the clinical characteristics of the children identified by them. The study complied with the Declaration of Helsinki and was approved by the Swedish Ethical Review Authority (2019‐04915). Informed consent was waived.

Data on asthma diagnoses were extracted from the Swedish National Airway Register [4]. Spirometry data were assessed using pre‐ and post‐bronchodilator FEV₁ and FEV₁/FVC and are presented as percentages below the lower limit of normal (z‐score < −1.645) using references from the Global Lung Initiative [5]. Old BD responses were calculated using the previous definition [1]: ((post‐bronchodilator_FEV1 (L) − pre‐bronchodilator_FEV1 (L))/pre‐bronchodilator_FEV1 (L)) × 100. A change of > 12% and ≥ 200 mL was considered positive. New BD responses were calculated using the updated defintition [1]: ((post‐bronchodilator_FEV1 (L) − pre‐bronchodilator_FEV1 (L))/predicted value (L)) × 100. A change of > 10% was considered positive. Age was categorised as 5–11 and 12–17 years. The body mass index, calculated as weight (kilograms)/height² (meters), was categorised as normal weight or overweight/obesity using age‐dependent cut‐offs. An asthma control test score of ≤ 19 denoted uncontrolled asthma. Treatment for asthma included bronchodilators, inhaled corticosteroids, leukotriene receptor antagonists, and biological treatments. Treatment was categorised as step 1–2 or step 3–5, as defined by the Global Initiative for Asthma. Statistical analyses were conducted using SPSS Statistics version 29 (IBM Corp, New York, USA). Chi‐square tests were used for between‐group comparisons, and statistical significance was p < 0.05. Cohen's kappa estimated the agreement between old and new BD responses.

The study comprised 3301 children aged 5–17 years with complete pre‐ and post‐bronchodilator spirometry data. Both old and new BD responses were negative in 2770 cases (84%) whereas the other 531 (16%) had a positive test result (Table 1). Using the new BD responses yielded more positive results than old BD responses. Almost a third of the children who were positive in one or both tests had discordant results. The kappa value for agreement between the old and new BD responses was 0.79. Positive old BD responses, but negative new BD responses, were found in 30 children (0.9%). These children were older and more likely to have had FEV₁ < LLN and airway obstruction both pre‐ and post‐bronchodilators, compared with the 137 children with negative old BD responses and positive new BD responses (Table 1). Old BD responses were positive in 8.5% with pre‐bronchodilator FEV₁ > LLN and new BD responses in 13.1% of that group.

TABLE 1.

Characteristics of 3301 children based on their bronchodilator responses according to the old and new guidelines.

BD negative and positive responses	Old BD– new BD–	Old BD+ new BD+	Old BD+ new BD –	Old BD– new BD+	p a
	n = 2770	n = 364	n = 30	n = 137
Female, n (%)	1081 (39.0)	124 (34.1)	10 (33.3)	61 (44.5)	0.26
Age, 12–17 years, n (%)	1367 (49.4)	127 (34.9)	23 (76.7)	24 (17.5)	< 0.001
Overweight/obese, n (%)*	637 (23)	81 (22.3)	8 (26.7)	27 (19.7)	0.40
ACT ≤ 19, n (%)**	631 (25.5)	102 (30.9)	10 (34.5)	35 (28.5)	0.52
Exacerbations ≥ 2, n (%)	265 (9.6)	54 (14.8)	4 (13.3)	9 (6.6)	0.21
Step 3–5 treatment, n (%)***	998 (43.8)	128 (41.6)	20 (76.9)	45 (40.2)	< 0.001
Pre‐BDFEV1 < LLN, n (%)	377 (13.6)	133 (36.5)	27 (90.0)	8 (5.8)	< 0.001
Pre‐BDFEV1/FVC <LLN, n (%)	339 (12.2)	194 (53.3)	16 (53.3)	28 (20.4)	< 0.001

Note: BD, bronchodilator; *body mass index (BMI) missing, n = 3; **Asthma Control Test (ACT) missing, n = 345; ***Treatment missing, n = 576.

^a Chi‐square tests were used for comparisons between Old BD+ new BD– vs. Old BD– new BD+.

Two studies have evaluated the impact of the new BD responses on children. A population‐based study [2] assessed the spirometry results of 2293 children, including 14.4% with current asthma. New BD responses were positive in 36% with asthma and 25% without, compared to 17% and 10% for the old BD responses. The effects of changing the BD responses were also evaluated using data from a pulmonary function laboratory on 1224 children referred for cough or asthma [3]. The new BD response was positive in 18% and the old BD response in 15%. Children with a positive new BD response had higher baseline FEV₁, forced vital capacity (FVC) and FEV₁/FVC than those with a positive old response.

Our results extend previous findings. They show that children with positive old, but negative new, BD responses differed from children with negative old, but positive new, BD responses. They were older, prescribed more treatment, and had more impaired lung function. When the new BD response was used, children with less asthma morbidity were classified as having a positive bronchodilator response, and a few children with more morbidity were classified as having no bronchodilator response. However, the bronchodilator response is not the only factor used to evaluate troublesome asthma in clinical settings, and current treatment also impacts bronchodilator responses. Nevertheless, the findings underline that bronchodilator responses from adults cannot be extrapolated to children.

Interestingly, both the old and new BD responses were positive in a significant proportion of children with pre‐bronchodilator FEV₁ above the lower limit of normal. This challenges clinical guidelines that only recommend bronchodilator responses in children with pre‐bronchodilator airway obstruction.

The study's strengths included the large cohort and before and after bronchodilator lung function measurements. All the children were diagnosed with asthma and the results are not necessarily generalisable to healthy children or those with suspected asthma. The limitations included no knowledge of adherence to prescribed medications or whether the data originated from new referrals or follow‐up visits. Some data on body mass index, asthma control tests, and asthma treatment were also missing. In Sweden, spirometry is generally performed by specialised paediatric nurses and many have national licences to perform this. We can consider that the quality of spirometry was good, even though we did not have access to the original data.

In conclusion, switching from the old to new BD responses meant that children with less asthma morbidity were classified as having a positive bronchodilator response. A few children with more morbidity were classified as having no bronchodilator response. This underlines that bronchodilator response results from adults cannot be extrapolated to children. Cut‐off values balance sensitivity and specificity, and BD responses should be interpreted with other markers and clinical judgement.

Author Contributions

Caroline Stridsman: conceptualization, methodology, project administration, resources, writing – review and editing, funding acquisition. Helena Backman: conceptualization, methodology, writing – review and editing. Lowie E. G. W. Vanfleteren: conceptualization, methodology, writing – review and editing. Anna Asarnoj: writing – review and editing, methodology, conceptualization. Henrik Ljungberg: conceptualization, methodology, writing – review and editing. Anne Lindberg: conceptualization, methodology, writing – review and editing. Apostolos Bossios: conceptualization, methodology, writing – review and editing. Jon R. Konradsen: conceptualization, methodology, formal analysis, project administration, writing – original draft, funding acquisition, resources.

Conflicts of Interest

The authors declare no conflicts of interest.