The bronchodilator response in childhood asthma, a comparison of ATS/ERS and GINA criteria in a register-based cohort.

<bold>Background:</bold> The ATS/ERS (A/E) definition of a positive bronchodilator response (BDR) is different from the definition used by GINA (G). We investigated concordance between BDR-A/E and BDR-G. <bold>Methods:</bold> Pre- and post-bronchodilator (BD) FEV1 (Liter), Asthma Control Test (ACT) and asthma treatment were extracted from the Swedish National Airway Register from 3301 asthmatic children (age 6-17 years). GLI reference equations were used. BDR-A/E was calculated using ((postBD – preBD) / predicted value) x 100 and >10% were considered positive. BDR-G was calculated using ((postBD– preBD) / preBD) x 100 and >12% and at least 200 ml e were considered positive. Concordance was estimated with Cohens kappa and Chi-square and Students t-test were used to compare differences. <bold>Results:</bold> A positive BDR-A/E was identified in 15.2% (n=501), a positive BDR-G in 11.9% (n=394) and 11% (n=360) were positive according to both tests. The concordance between BDR-A/E and BDR-G was 0.79 (substantial agreement). A positive BDR-A/E but negative BDR-G was found in 4% (n=137). A negative BDR-A/E but positive BDR-G was found in 0.9% (n=30) and these children were older (p<0.001), had lower ACT (p=0.067), more treatment(p<0.001) and more often FEV1<LLN both pre and post (p<0.001) BD compared to children with positive BDR-A/E and negative BDR-G. <bold>Conclusion:</bold> Almost 1/3 of the children with BDR-A/E or BDR-G have discordant results. Children with negative BDR-A/E but positive BDR-G have poorer asthma control, more treatment and more impaired lung function compared to children with positive BDR-A/E and negative BDR-G. BDR-A/E does not seem to be appropriate for children.