Interpretation of the ratio of lung function parameters is independent of population-based reference equations. This study evaluated the diagnostic accuracy and optimal cut-off of oscillometry-derived ratios for identifying z-score–based small airway dysfunction.

Pre-bronchodilator oscillometry records of 2453 adults aged ≥18 years were retrospectively examined. Resmon™ Pro FULL V3 was used for oscillometry. Peripheral airway resistance ratios were calculated as (R5-19/R5) × 100, and reactance ratios were calculated as (X5/AX) × 100. Receiver operating characteristic (ROC) curves were constructed to evaluate the diagnostic accuracy of peripheral airway resistance and reactance ratios, using z-score thresholds of R5–19 > 1.64 and X5 < –1.64 as reference standards. Optimal cutoffs were identified using Youden’s index.

The mean age of the cohort was 45.5 ± 16.3 years, and 55.2 % were male. ROC curve analysis demonstrated that the AUC for peripheral resistance was 0.98 (95 % CI, 0.98–0.99; p < 0.001). The optimal cutoff value for the resistance ratio was 19.8 %, which yielded a sensitivity of 91.3 % and a specificity of 95.6 %, with corresponding positive and negative predictive values of 97.3 % and 86.5 %, respectively. The AUC of the reactance ratio was 0.91 (95 % CI: 0.90–0.92; p < 0.001). The optimal cutoff for the reactance ratio was 16.2%, with a sensitivity of 76.1%, a specificity of 94.7%, and positive and negative predictive values of 96.7% and 66.3%, respectively.

This study suggests that oscillometry-derived ratios, particularly the resistance ratio, are reliable markers for detecting small airway dysfunction, independent of normative reference equations.