PT - JOURNAL ARTICLE AU - Rianne de Vries AU - Yennece W. F. Dagelet AU - Sigrid Kester AU - Johannes H. H. C. In 'T Veen AU - Eric G. Haarman AU - Annelies Lucas AU - Frans H. C. De Jongh AU - Anke-Hilse Maitland-Van Der Zee AU - Peter J. Sterk TI - Late Breaking Abstract - ENose based phenotyping of newly presented patients with asthma or COPD AID - 10.1183/13993003.congress-2018.OA3577 DP - 2018 Sep 15 TA - European Respiratory Journal PG - OA3577 VI - 52 IP - suppl 62 4099 - http://erj.ersjournals.com/content/52/suppl_62/OA3577.short 4100 - http://erj.ersjournals.com/content/52/suppl_62/OA3577.full SO - Eur Respir J2018 Sep 15; 52 AB - Rationale: Clinical/inflammatory phenotypes of chronic airway disease have been obtained by eNose, irrespective of the diagnostic label [De Vries ERJ 2018]. This requires prospective validation at the point-of-care.Aim: To determine the accuracy of the eNose for classifying newly presented patients with a differential diagnosis of asthma or COPD into diagnostic labels and clinical/inflammatory phenotypes.Methods: BreathCloud is a multicentre cross-sectional study in patients with e.g. asthma and COPD. Exhaled breathprints were collected in duplicate by eNose (SpiroNose). Four weeks after inclusion all clinical diagnoses based on current guidelines were available. Correct classification into eNose phenotypes (1. obese non-atopic; 2. eosinophilic predominant; 3. non-Caucasian; 4. inflammation atopic; 5. well-controlled) [De Vries ERJ 2018] was defined as a complete clinical/inflammatory match. Data analysis involved advanced signal processing and statistics based on principal component analysis and discriminant analysis.Results: Patients with a differential diagnosis of asthma or COPD (n=257) were classified by eNose into diagnostic labels with an overall accuracy of 81% (asthma:82%, COPD:79%) and a ROC-AUC of 0.83(CI:0.81-0.86). Classification into phenotypes reached an accuracy of 89% (phenotype 1:83%, 2:92%, 3:89%, 4:92%, 5:88%) with a ROC-AUC of 0.91(CI:0.87-0.95).Conclusion: Enose technology is able to classify newly presented patients with a differential diagnosis of asthma or COPD with high accuracy into their classic diagnostic labels and clinical/inflammatory phenotypes. These data validate breathomics for stratification of patients with chronic airway diseases in the doctor’s office.FootnotesCite this article as: European Respiratory Journal 2018 52: Suppl. 62, OA3577.This is an ERS International Congress abstract. No full-text version is available. Further material to accompany this abstract may be available at www.ers-education.org (ERS member access only).