Abstract
OBJECTIVE: Assessment of lung function variability is recommended for the diagnosis of asthma, but its specificity in separating asthmatic from COPD subjects is low. This study aimed to test the hypothesis that the day-to-day variability of respiratory resistance depends on the respiratory phase considered and observation time. METHODS: Respiratory resistance was measured daily by oscillometry at 5 Hz in 47 mild asthmatics, 20 moderate-to-severe COPD, and 35 healthy subjects. The coefficient of variation was calculated over multiple time scales using full breaths, inspiratory phase, or mid-inspiratory phase. RESULTS: The coefficient of variation of mid-inspiratory resistance was significantly higher in asthmatic than healthy and COPD groups at time scales >7 days, but not different between healthy and COPD. The accuracy of the 14-days coefficient of variation of mid-inspiratory resistance in separating asthmatic from the other groups, calculated as the area under the receiver-operating characteristic curve, was 0.86, with 73% sensitivity and 83% specificity at the optimal cutoff of 10%. Moreover, the coefficient of variation was significantly higher in asthma than COPD despite an increased mean resistance in the latter. CONCLUSION: When expressed as the day-to-day coefficient of variation of mid-inspiratory oscillometric resistance, the variability of lung function does not appear related to the presence or degree of airflow obstruction. Two-week assessment of day-to-day variability of mid-inspiratory resistance provides accurate separation of asthmatic from both healthy and COPD subjects. These findings demonstrate that simple, self-administered daily oscillometry can provide useful clinical information, supporting more accurate asthma diagnosis in real-world settings. Clinical and Translational Impact-The coefficient of variation of mid-inspiratory resistance computed over 14-days separated asthmatic from healthy and COPD subjects with 73% sensitivity and 83% specificity. Daily self-administered oscillometry can support asthma diagnosis.