Diagnostic performance of eNose technology in COVID-19 patients after hospitalization

BMC Pulmonary Medicine

Research

B. F.M. van Raaij¹  J. D. Veltman²  J. J.M. Geelhoed³  J. F. Hameete³  J. L. Stöger⁴ 

[1]Department of Internal Medicine, Section of Geriatrics and Gerontology, Leiden University Medical Centre, Albinusdreef 2, 2333ZA, Leiden, Netherlands

[2]Department of Pulmonary Diseases, Amsterdam University Medical Centre, Amsterdam, Netherlands

[3]Department of Pulmonary Diseases, Leiden University Medical Centre, Leiden, Netherlands

[4]Department of Radiology, Leiden University Medical Centre, Leiden, Netherlands

关键词: Breath analysis;    COVID-19;    Electronic nose technology;    Follow-up;   

DOI  :  10.1186/s12890-023-02407-6

 received in 2023-01-31, accepted in 2023-03-31,  发布年份 2023

来源: Springer

【 摘 要 】

BackgroundVolatile organic compounds (VOCs) produced by human cells reflect metabolic and pathophysiological processes which can be detected with the use of electronic nose (eNose) technology. Analysis of exhaled breath may potentially play an important role in diagnosing COVID-19 and stratification of patients based on pulmonary function or chest CT.MethodsBreath profiles of COVID-19 patients were collected with an eNose device (SpiroNose) 3 months after discharge from the Leiden University Medical Centre and matched with breath profiles from healthy individuals for analysis. Principal component analysis was performed with leave-one-out cross validation and visualised with receiver operating characteristics. COVID-19 patients were stratified in subgroups with a normal pulmonary diffusion capacity versus patients with an impaired pulmonary diffusion capacity (DLCOc < 80% of predicted) and in subgroups with a normal chest CT versus patients with COVID-19 related chest CT abnormalities.ResultsThe breath profiles of 135 COVID-19 patients were analysed and matched with 174 healthy controls. The SpiroNose differentiated between COVID-19 after hospitalization and healthy controls with an AUC of 0.893 (95-CI, 0.851–0.934). There was no difference in VOCs patterns in subgroups of COVID-19 patients based on diffusion capacity or chest CT.ConclusionsCOVID-19 patients have a breath profile distinguishable from healthy individuals shortly after hospitalization which can be detected using eNose technology. This may suggest ongoing inflammation or a common repair mechanism. The eNose could not differentiate between subgroups of COVID-19 patients based on pulmonary diffusion capacity or chest CT.

【 授权许可】

CC BY   
© The Author(s) 2023

【 图 表 】

40507_2023_167_Article_IEq244.gif

Fig. 10

40507_2023_167_Article_IEq233.gif