New method of screening for COVID-19 disease using sniffer dogs and scents from axillary sweat samples

• Published in: Journal of public health (Oxford, England) Vol. 44; no. 1; pp. e36 - e41
• Main Authors: Sarkis, Riad, Lichaa, Anthony, Mjaess, Georges, Saliba, Michele, Selman, Carlo, Lecoq-Julien, Clothilde, Grandjean, Dominique, Jabbour, Nabil M
• Format: Journal Article
• Language: English
• Published: England Oxford University Press 07.03.2022
• Subjects: Animals; COVID-19 - diagnosis; COVID-19 - epidemiology; Dogs; Humans; Odorants; SARS-CoV-2; Sensitivity and Specificity; Sweat; Working Dogs; COVID-19; sniffer dogs; PCR; airport; axillary sweat
• ISSN: 1741-3842; 1741-3850; 1741-3850
• DOI: 10.1093/pubmed/fdab215

Abstract Background Early screening for COVID-19 is needed to limit the spread of the virus. The aim of this study is to test if the sniffer dogs can be successfully trained to identify subjects with COVID-19 for ‘proof of concept’ and ‘non-inferiority’ against PCR. We are calling this method, Dognosis (DN). Methods Four hundred and fifty-nine subjects were included, 256 (Group ‘P’) were known cases of COVID-19 (PCR positive, some with and some without symptoms) and 203 (Group ‘C’) were PCR negative and asymptomatic (control). Samples were obtained from the axillary sweat of each subject in a masked fashion. Two dogs trained to detect specific Volatile Organic Compounds for COVID-19 detection were used to test each sample. Results [DN] turned out positive (+) in all the cases that were PCR positive (100% sensitivity). On the other hand, [DN] turned positive (+) in an average of 12.5 cases (6.2%) that were initially PCR negative (apparent specificity of 93.8%). When the PCR was repeated, true specificity was 97.2%. These parameters varied in subgroups from 100% sensitivity and 99% specificity in symptomatic patients to 100% sensitivity and 93% specificity in asymptomatic patients. Conclusion DN method shows high sensitivity and specificity in screening COVID-19 patients.