The Influence of Smoking Status on Exhaled Breath Profiles in Asthma and COPD Patients
Aim: To investigate whether the eNose is suitable as a non-invasive technique to identify how patients with different smoking habits may respond to smoke exposure and whether smoking has an influence on disease classification.
Take home message: The eNose can effectively differentiate asthma and COPD patients based on their smoking history but cannot detect recent smoking (within 24 hours). Smoking history does not influence eNose measurements in healthy individuals, suggesting it is not a confounding factor.
Introduction
Asthma and Chronic Obstructive Pulmonary Disease (COPD) are heterogeneous airway diseases influenced by various clinical and environmental factors. Analyzing volatile organic compounds (VOCs) in exhaled breath using eNose technology offers a non-invasive method for characterizing these diseases. However, it is unclear if smoking status affects the accuracy of breath analysis. This study explores whether the eNose can differentiate between ever-smokers and never-smokers, and if recent smoking impacts the diagnostic performance.
Methods
The cross-sectional study analyzed data from the BreathCloud database, including patients with asthma or COPD and healthy controls. Participants were categorized based on smoking history (ever-smokers vs. never-smokers) and recent cigarette consumption (within 24 hours before breath measurement). Exhaled breath was analyzed using the SpiroNose®, a device equipped with metal-oxide semiconductor sensors to detect VOC patterns. Standardized methods were applied for data collection, and patients performed multiple breath tests after mouth rinsing. Principal component analysis (PCA) was used to summarize the eNose signals, followed by linear discriminant analysis (LDA) for classification. Model performance was assessed using the area under the receiver-operating characteristic curve (ROC-AUC).
Results
The study included 896 patients (593 ever-smokers and 303 never-smokers). The eNose was able to discriminate between ever-smokers and never-smokers with moderate accuracy (AUC: 0.74; 95% CI: 0.70–0.77). However, it could not distinguish between patients who smoked less than 24 hours prior and those who smoked earlier (AUC: 0.60). In healthy controls, the eNose did not differentiate ever-smokers from never-smokers (AUC: 0.54). These findings suggest that while eNose technology can detect a history of smoking in patients with chronic airway diseases, recent smoking does not significantly affect the breath profile.
Discussion
The eNose demonstrated potential for differentiating patients based on smoking history in chronic respiratory diseases but was less effective in identifying recent smoking exposure. The absence of discrimination in healthy controls indicates that smoking status may not confound eNose measurements in non-diseased individuals. The findings support the use of eNose technology as a diagnostic tool for asthma and COPD, though the influence of historical smoking must be considered. The study’s large sample size and standardized methods strengthen the reliability of the results, but further research is needed to explore passive smoking effects and refine VOC analysis for improved accuracy.
