Abstract
BACKGROUND: Information is limited regarding the role of impulse oscillometry (IOS) for the detection of preserved ratio impaired spirometry (PRISm). Therefore, we aimed to study the diagnostic ability of IOS in differentiating between PRISm and healthy subjects. METHODS: This retrospective data collection was done at the Lung Health Center, Faculty of Medicine, Chiang Mai University, Thailand between July 2019 and April 2022. The potential diagnostic possibilities of difference in resistance at 5 Hz (R5) and resistance at 20 Hz (R20) (R5-R20) for PRISm detection were analyzed. RESULTS: The prevalence of PRISm was higher when using the fixed ratio (FR) criteria (FEV(1)/FVC ≥0.7 with FEV(1) < 80% of predicted value) compared to the lower limit of normal (LLN) criteria (FEV(1)/FVC ≥ LLN and FEV(1) < LLN) (10.0% vs. 4.2%). The %prediction for R5-R20 provided an acceptable area under the curve (AUC) for PRISm, defined by the LLN and the FR criteria (AuROC = 0.75 (95%CI; 0.64, 0.85) and 0.72 (95%CI; 0.63, 0.81), respectively). The cut-off value of %predicted R5-R20 ≥120% resulted in the highest sensitivity and specificity for detecting PRISm. CONCLUSIONS: The %predicted of R5-R20 ≥ 120% showed an acceptable performance for PRISm detection and PRISm may be detected by IOS.