Abstract Rationale Impulse oscillometry (IOS) is considered sensitive at detecting small airway dysfunction through measurement of the difference in resistance between 5 and 20 Hz (R5-R20/R5) and post-albuterol changes in R5 and the area of reactance under the curve (AX). Current ERS/ATS guidelines propose using an R5-R20/R5 Z-score 1.64 from the healthy matched population to diagnose asthma (King et al., 2020), although the sensitivity (SS) and specificity (SP) of this cut-off has not been directly examined in children with asthma. Methods We conducted a retrospective chart review of IOS measurements performed in children 3-17 years old from September 2022-June 2025. Z-scores were calculated using established reference equations (Gochicoa-Rangel et al., 2023). Asthma diagnosis was established using pulmonary function criteria, incorporating baseline expiratory airflow, bronchodilator reversibility, and methacholine responsiveness. For preschool-aged children who could not complete lung function testing, diagnosis relied on documented clinical history. Logistic regression was used to construct models for probability of asthma, with results reported as area under the receiver operator characteristic curve (AUC) and statistical tests of AUC vs. a non-informative value of 0.5. Cut-off values were calculated using Youden’s J index using combination of optimal SS and SP. Results Out of 308 IOS measurements, 298 (97%) were performed in children with asthma and 10 (3%) without asthma. Median levels were significantly different between asthma and no asthma, respectively, for R5-R20/R5 (49.3% vs 14.8%, p 0.0001) and its Z-score (2.6 vs -1.2, p 0.0001), and for post-bronchodilator changes in R5 (20% vs 0%, p = 0.009) and AX (37.0% vs 24.5%, p = 0.03). AUC measurements for R5-R20/R5 Z-score (0.87, p 0.001) and R5-R20/R5 (0.87, p 0.001) were excellent, while those for post-bronchodilator changes in R5 (0.74, p = 0.008) and AX (0.70, p = 0.006) were acceptable. R5-R20/R5 25% (92% SS; 80% SP), R5-R20/R5 Z-score 0.93 (84% SS; 80% SP), 38% change in AX (49% SS; 90% SP), and greater than 0% change in R5 (94% SS; 40% SP) provide the best combination of SS and SP. Table 1 shows the odds ratios (OR) for each IOS parameter threshold in predicting asthma. Conclusion IOS measures of peripheral airway resistance have “excellent” and bronchodilator reversibility in AX and R5 has “acceptable” potential to distinguish asthma in children based on AUC (Hosmer et al., 2013). The cut off R5-R20/R5 Z-score 0.93 would provide greater sensitivity than the previously proposed measurement 1.64, which had 71% SS and 80% SP in our cohort. This needs to be validated in a larger cohort. This abstract is funded by: National Jewish Health, Department of Pediatrics, Division of Allergy and Clinical Immunology
Ho et al. (Fri,) studied this question.