Diagnostic accuracy of non-invasive detection of SARS-CoV-2 infection by canine olfaction.

BACKGROUND: Throughout the COVID-19 pandemic, testing individuals remains a key action. One approach to rapid testing is to consider the olfactory capacities of trained detection dogs. METHODS: Prospective cohort study in two community COVID-19 screening centers. Two nasopharyngeal swabs (NPS), one saliva and one sweat samples were simultaneously collected. The dog handlers (and the dogs…) were blinded with regards to the Covid status. The diagnostic accuracy of non-invasive detection of SARS-CoV-2 infection by canine olfaction was assessed as compared to nasopharyngeal RT-PCR as the reference standard, saliva RT-PCR and nasopharyngeal antigen testing. RESULTS: 335 ambulatory adults (143 symptomatic and 192 asymptomatic) were included. Overall, 109/335 participants tested positive on nasopharyngeal RT-PCR either in symptomatic (78/143) or in asymptomatic participants (31/192). The overall sensitivity of canine detection was 97% (95% CI, 92 to 99) and even reached 100% (95% CI, 89 to 100) in asymptomatic individuals compared to NPS RT-PCR. The specificity was 91% (95% CI, 72 to 91), reaching 94% (95% CI, 90 to 97) for asymptomatic individuals. The sensitivity of canine detection was higher than that of nasopharyngeal antigen testing (97% CI: 91 to 99 versus 84% CI: 74 to 90, p = 0.006), but the specificity was lower (90% CI: 84 to 95 versus 97% CI: 93 to 99, p = 0.016). CONCLUSIONS: Non-invasive detection of SARS-CoV-2 infection by canine olfaction could be one alternative to NPS RT-PCR when it is necessary to obtain a result very quickly according to the same indications as antigenic tests in the context of mass screening.

Biomedical detection dogs for the identification of SARS-CoV-2 infections from axillary sweat and breath samples*.

A Polymerase Chain Reaction (PCR) test of a nasal swab is still the 'gold standard' for detecting a severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection. However, PCR testing could be usefully complemented by non-invasive, fast, reliable, cheap methods for detecting infected individuals in busy areas (e.g. airports and railway stations) or remote areas. Detection of the volatile, semivolatile and non-volatile compound signature of SARS-CoV-2 infection by trained sniffer dogs might meet these requirements. Previous studies have shown that well-trained dogs can detect SARS-CoV-2 in sweat, saliva and urine samples. The objective of the present study was to assess the performance of dogs trained to detect the presence of SARS-CoV-2 in axillary-sweat-stained gauzes and on expired breath trapped in surgical masks. The samples were provided by individuals suffering from mild-to-severe coronavirus disease 2019 (COVID-19), asymptomatic individuals, and individuals vaccinated against COVID-19. Results: Seven trained dogs tested on 886 presentations of sweat samples from 241 subjects and detected SARS-CoV-2 with a diagnostic sensitivity (relative to the PCR test result) of 89.6% (95% confidence interval (CI): 86.4%-92.2%) and a specificity of 83.9% (95% CI: 80.3%-87.0%)-even when people with a low viral load were included in the analysis. When considering the 207 presentations of sweat samples from vaccinated individuals, the sensitivity and specificity were respectively 85.7% (95% CI: 68.5%-94.3%) and 86.0% (95% CI: 80.2%-90.3%). The likelihood of a false-positive result was greater in the two weeks immediately after COVID-19 vaccination. Four of the seven dogs also tested on 262 presentations of mask samples from 98 subjects; the diagnostic sensitivity was 83.1% (95% CI: 73.2%-89.9%) and the specificity was 88.6% (95% CI: 83.3%-92.4%). There was no difference (McNemar's testP= 0.999) in the dogs' abilities to detect the presence of SARS-CoV-2 in paired samples of sweat-stained gauzes vs surgical masks worn for only 10 min. Conclusion: Our findings confirm the promise of SARS-CoV-2 screening by detection dogs and broaden the method's scope to vaccinated individuals and easy-to-obtain face masks, and suggest that a 'dogs + confirmatory rapid antigen detection tests' screening strategy might be worth investigating.

Identifying SARS-COV-2 infected patients through canine olfactive detection on axillary sweat samples; study of observed sensitivities and specificities within a group of trained dogs.

There is an increasing need for rapid, reliable, non-invasive, and inexpensive mass testing methods as the global COVID-19 pandemic continues. Detection dogs could be a possible solution to identify individuals infected with SARS-CoV-2. Previous studies have shown that dogs can detect SARS-CoV-2 on sweat samples. This study aims to establish the dogs' sensitivity (true positive rate) which measures the proportion of people with COVID-19 that are correctly identified, and specificity (true negative rate) which measures the proportion of people without COVID-19 that are correctly identified. Seven search and rescue dogs were tested using a total of 218 axillary sweat samples (62 positive and 156 negative) in olfaction cones following a randomised and double-blind protocol. Sensitivity ranged from 87% to 94%, and specificity ranged from 78% to 92%, with four dogs over 90%. These results were used to calculate the positive predictive value and negative predictive value for each dog for different infection probabilities (how likely it is for an individual to be SARS-CoV-2 positive), ranging from 10-50%. These results were compared with a reference diagnostic tool which has 95% specificity and sensitivity. Negative predictive values for six dogs ranged from ≥98% at 10% infection probability to ≥88% at 50% infection probability compared with the reference tool which ranged from 99% to 95%. Positive predictive values ranged from ≥40% at 10% infection probability to ≥80% at 50% infection probability compared with the reference tool which ranged from 68% to 95%. This study confirms previous results, suggesting that dogs could play an important role in mass-testing situations. Future challenges include optimal training methods and standardisation for large numbers of detection dogs and infrastructure supporting their deployment.

Can the detection dog alert on COVID-19 positive persons by sniffing axillary sweat samples? A proof-of-concept study.

The aim of this proof-of-concept study was to evaluate if trained dogs could discriminate between sweat samples from symptomatic COVID-19 positive individuals (SARS-CoV-2 PCR positive) and those from asymptomatic COVID-19 negative individuals. The study was conducted at 2 sites (Paris, France, and Beirut, Lebanon), followed the same training and testing protocols, and involved six detection dogs (three explosive detection dogs, one search and rescue dog, and two colon cancer detection dogs). A total of 177 individuals were recruited for the study (95 symptomatic COVID-19 positive and 82 asymptomatic COVID-19 negative individuals) from five hospitals, and one underarm sweat sample per individual was collected. The dog training sessions lasted between one and three weeks. Once trained, the dog had to mark the COVID-19 positive sample randomly placed behind one of three or four olfactory cones (the other cones contained at least one COVID-19 negative sample and between zero and two mocks). During the testing session, a COVID-19 positive sample could be used up to a maximum of three times for one dog. The dog and its handler were both blinded to the COVID-positive sample location. The success rate per dog (i.e., the number of correct indications divided by the number of trials) ranged from 76% to 100%. The lower bound of the 95% confidence interval of the estimated success rate was most of the time higher than the success rate obtained by chance after removing the number of mocks from calculations. These results provide some evidence that detection dogs may be able to discriminate between sweat samples from symptomatic COVID-19 individuals and those from asymptomatic COVID-19 negative individuals. However, due to the limitations of this proof-of-concept study (including using some COVID-19 samples more than once and potential confounding biases), these results must be confirmed in validation studies.