Effect of expressing sRaw reversibility relative to baseline or predicted values in children

<bold>Background:</bold> Expressing post-bronchodilator response (BDR) in % baseline (%B) biases the results according to baseline alteration and to the size of the subject. Reversibility is expressed as % predicted (%P) for the interrupter technique and for FEV₁. <bold>Objectives:</bold> To compare the sensitivity/specificity of Raw and sRaw BDR expressed as %B or %P to distinguish children with a significant FEV₁ reversibility. <bold>Methods:</bold> Monocentre retrospective study (2021 – 2023) conducted in all subjects seen for asthma, chronic cough, or unspecific symptoms without comorbidities except for possible overweight/obesity. Salbutamol was systematically administered. <bold>Results:</bold> We retrieved 2436 files with pre and post-BD sRaw and spirometry recorded in 1623 children (females 617 (38%)), age 12.4 (±2.6) years, 1465 (90%) with asthma, 179/30 overweight/obese). Baseline obstruction was detected by FEV₁/FVC (n=2430), or sGaw <-1,645 z-score, or both in 584 (24%), 943 (39%) and 412 (17%) cases, respectively. FEV₁ BDR was >10%P in 422 (17%) patients. ROC curve analyses for Raw (n = 2408) and sRaw showed high and similar sensitivity/specificity and AUCs for BDR %P and %B (Table). <table-wrap><object-id>erj;64/suppl_68/PA1347/TB1</object-id><object-id>T1</object-id><object-id>TB1</object-id><table><colgroup><col></col><col></col><col></col><col></col><col></col></colgroup><tbody><tr><td>BDR</td><td>Raw %B</td><td>Raw %p</td><td>sRaw %B</td><td>sRaw %p</td></tr><tr><td>AUC (95% CI)</td><td>0.77 (0.74-0.79)</td><td>0.78 (0.76 - 0.81)</td><td>0.79 (0.77 - 0.81)</td><td>0.82 (0.80 - 0.84)</td></tr><tr><td>Cut-off (Sensitivity/Specificity) %</td><td>-33 (73/67)</td><td>-45 (69/74)</td><td>-35 (74/70)</td><td>-56 (75/74)</td></tr></tbody></table></table-wrap> <bold>Conclusion:</bold> We conclude that expressing Raw or sRaw BDR as %B or %P did not change the discrimination of children with positive FEV₁ BDR, but corresponds to different cut-off of which the clinical relevance needs to be further studied.