Exhaled breath profiling by electronic nose enabled discrimination of allergic rhinitis and extrinsic asthma.

Abstract

Background: The analysis of volatile organic compounds (VOCs) by an electronic nose is an innovative technique that has shown its potential in the non-invasive diagnosis of several respiratory diseases, including asthma. We postulated that an electronic nose could discriminate between subjects affected by allergic rhinitis with and without concomitant extrinsic asthma, as well as from healthy controls, in terms of exhaled VOCS profile. This pilot study sought to explore the above mentioned hypothesis. Methods: 14 patients with Extrinsic Asthma and Allergic Rhinitis (AAR), 14 patients with Allergic Rhinitis without asthma (AR) and 14 healthy controls (HC) participated in a cross-sectional study. Exhaled breath was collected by a widely-used method and sampled by an electronic nose (Cyranose 320). Raw data were reduced by Principal component analysis and analysed by canonical discriminant analysis. Cross-validation accuracy (CVA) and ROC-curves were calculated. External validation in newly recruited patients (7 AAR, 7 AR and 7 HC) was tested using the previous training set. Results: Breathprints of patients with AR clustered from those with AAR (CVA = 85.7%), as well as HC (CVA = 82.1%). Breathprints from AAR were also separated from those of HC (CVA = 75.0%). External validation confirmed the above findings. Conclusions: An electronic nose can discriminate exhaled breath from well-characterized subjects with allergic rhinitis with and without extrinsic asthma, which represent two different diseases with partly overlapping features. Our results support the view of using exhaled molecular profiling to help diagnosing asthma also in patients with allergic rhinitis.