Impulse oscillometry and spirometry measurements relative to personal best values at the time of acute exacerbations of cystic fibrosis in adults.

BACKGROUND Diagnosis of acute exacerbation (AE) of cystic fibrosis (CF) must be precise because both under- and over-prescription of antibiotics may be detrimental. How lung function tests contribute to diagnose AE is unclear. We aimed to describe variation of spirometry and oscillometry measurements, at Stable state and at AE in adults with CF. METHODS Patients were included in a retrospective single-centre study when both spirometry (FEV1, FVC) and oscillometry (X5, R5, R5-R20 and AX) data were available for at least one Stable and one AE visit between December 2016 and July 2019. For each visit, we calculated variation (Δ) in spirometry and oscillometry indices in comparison with personal best values. Measurements were expressed as % of predicted values and Z-scores when applicable. Areas under ROC curves (AUC) were computed. RESULTS Forty-two patients (28 ± 9 years, FEV1 64 ± 21%) were included; 80 AE and 104 Stable visits were analysed. FEV1 (L, %pred and Z-score) and FVC (%pred and Z-score) varied significantly between AE and Stable visits (p < .05), although differences were small (80 ml/2.7%pred for FEV1). Among oscillometry indices, X5 (kPa.s.L-1 ), R5-R20 (kPa.s.L-1 ) and AX (kPa/L) varied significantly. The AUCs for the variation in spirometry indices ranged from 0.601 (ΔFVC L) to 0.635 (ΔFEV1%pred). They were not significantly different from the AUCs for ΔX5 (0.589), ΔR5-R20 (0.649) and ΔAX (0.598). CONCLUSIONS Performance of both spirometry and oscillometry to discriminate AE from Stable state was poor. Variation of oscillometry indices (X5, R5-R20, AX) may be helpful when spirometry is unreliable or uncomfortable.