An electronic nose in the discrimination of obese patients with and without obstructive sleep apnoea

Exhaled breath contains thousands of volatile organic compounds (VOCs) in gaseous form, which may be used as markers of airway inflammation and lung disease. Electronic noses enable quick and real-time pattern analysis of VOC spectra. It has been shown that the exhaled breath of patients with obstructive sleep apnoea (OSA) differs from that of non-obese controls. We aimed to assess the influence of obesity in the composition of exhaled VOCs by comparing obese subjects with and without OSA. Moreover, we aimed to identify the discriminant VOCs in the two groups. 19 obese patients with established OSA (OO; age 51.2 +/- 6.8; body mass index (BMI) 34.3 +/- 3.5), 14 obese controls without OSA (ONO; age 46.5 +/- 7.6; BMI 33.5 +/- 4.1) and 20 non-obese healthy controls (HC; age 41.1 +/- 12.6; BMI 24.9 +/- 3.8) participated in a cross-sectional study. Exhaled breath was collected by a previously described method and sampled by using an electronic nose (Cyranose 320) and by gas chromatography-mass spectrometry (GC-MS) analysis. Breathprints were analyzed by canonical discriminant analysis on principal component reduction. Cross-validation accuracy (CVA) was calculated. Breathprints from the HC group were separated from those of OO (CVA = 97.4%) and ONO (CVA = 94.1%). Breathprints from OO were moderately separated from those of ONO (CVA = 67.6%). The presence of OSA alters the exhaled VOC pattern in obese subjects. The incomplete separation of breathprints between OO and ONO may be due to the same underlying inflammation caused by obesity.