SARS-CoV-2 sniffer dogs also detect samples from long-term COVID patients

A variety of dogs can be used as SARS-CoV-2 detection dogs, such as Malinois, German Shepherds, Labrador Retrievers or Cocker Spaniels. / Alexey, Stocadopicom

HANNOVER – Trained sniffer dogs can only identify people who are severely infected with SARS-CoV-2. They are also identifying samples from post-COVID-19 patients whose PCR or antibody testing can no longer detect the original infection.

This is the result of a research team led by the University of Veterinary Medicine in Hannover (TiHo). The pilot study was conducted with 9 dogs to detect SARS-CoV-2 in frontiers in medicine Published (2022; DOI: 10.3389/fmed.2022.877259).

Dogs don’t smell the viruses themselves, but rather volatile organic compounds (VOCs) that are produced by metabolic processes during a viral infection. This may indicate the presence of persistent infection or long-term metabolic changes in post-COVID-19 patients.

As part of the study, 9 trained dogs that identified samples from severe COVID-19 patients were faced with more than 700 samples of saliva, urine, and sweat in 3 test scenarios: In the first test scenario, researchers provided 8 sniffer dogs with at least 50 samples per patient A severe COVID-19 patient and a long-term COVID-19 patient.

The sensitive It indicates how successful the test is in correctly identifying patients with a disease (true positive). The Quality On the other hand, it refers to the ability of the test to correctly identify non-disease patients (true negative).

subordinate predictive predictive value In turn, it makes a statement about the likelihood that a patient is actually sick if the test result is positive (PPV) and healthy if the result is negative (NPV). For this purpose, the prevalence of the disease is taken into account, which can vary according to age or other factors. This additional variable can lead to significantly different risk values ​​in diagnostic tests.

Dogs detected COVID-19 samples with an average sensitivity of 86.7% (95% confidence interval (CI): 75.4-98.0%) and specificity of 95.8% (95% CI: 92.5-99.0%). The positive predictive value (PPV) was 82.9% and the negative predictive value (NPV) was 97.3% (see box).

Dogs don’t always recognize the long COVID

In this test scenario, the dogs recognized the long COVID samples much worse. Of the more than 400 dogs, only 18 were recognized by dogs.

In a direct comparison, dogs can hardly distinguish between acute infection and the long-running coronavirus, or: In a direct comparison, dogs are more likely to show the samples with the smell of the severe COVID-19 disease for which they have been trained.

These results indicate that the acute odor specific to COVID-19 is still present in most of the Long-COVID samples, but is likely to be at a lower concentration.

“We think it is a calibration result. When dogs are presented with post COVID samples alongside acute SARS-CoV-2 samples, they are indicative of acute samples. If you do that compared to negative controls, you can see traces of VOCs that are still present in some samples. After COVID,” explained recent author Holger Falk, Head of TiHo’s Small Animal Clinic.

The sniffer dogs recognized the long COVID samples better in the second scenario. Here 3 dogs were encountered with about 20 long-term COVID-negative control samples. They detected long COVID samples with a mean sensitivity of 94.4% (95% CI: 70.5-100.0%) and specificity of 96.1% (95% CI: 87.6-100, 0%).

By comparison, when presented with at least 24 acute SARS-CoV-2 positive samples and negative control samples, mean sensitivity 86.9 (95% CI: 55.7-100.0%) and specificity 88,1% (95% CI: 82.7-93.6). %) verification.

Beagle studies support the long COVID hypothesis: viral persistence

In contrast, in a study recently published from France, dogs identified only 51.5% of long-term COVID patients when presented alongside healthy individuals (DOI: 10.1101/2022.011.21268036).

Researchers from Hannover attribute the poor detection compared to our results (51.5% vs. 92.86%) to the difference in sample quality. Because the French study samples were collected at home and mailed without uniform freezing or refrigeration.

Both studies support the hypothesis that VOCs are present in post-COVID-19 patients long-term after initial infection. VOCs are released from body cells infected with SARS-CoV-2 as the disease progresses. “Additional studies using medical detection dogs in the pathophysiology of Long COVID should include the formation and time course of specific VOC patterns,” Falk said.

One of the French study’s authors, Emilie Seyrat, added: “Dogs were not used to detect long-term COVID itself.” Presumably, they detected VOCs released by proteins resulting from the activity and duplication of SARS-CoV2 mRNA.

The researchers therefore hypothesize that sniffer dogs detect the presence and ongoing recurrence of SARS-Cov-2. So dogs can only recognize a previous infection if they are still active.

In other words, Long COVID is COVID. Emily Sirat, Covid patient, co-author

First author Dominique Grandjean of the National Veterinary School in Alfort emphasized: If dogs recognize a sample, it means that the virus is still active, even if a small amount is sufficient. The French research team therefore supports the hypothesis that virus persistence may be the mechanism behind the prolonged COVID illness. “In other words, a long COVID is COVID,” Sirat said. Various hypotheses about the long mechanisms of COVID have already been discussed in nature medicine as well as in German medical journal (DE) mentioned.

A comparison of long COVID samples to samples of non-severe SARS-CoV-2 infected people without COVID has not been done for a long time – so it’s still clear whether sniffer dogs will recognize this difference.

There are several studies that have shown that dogs can be trained to distinguish samples from severely infected patients with SARS-CoV-2 from healthy controls. The Hanover team published the first of these studies in 2020 (DOI: 10.5281/zenodo.3950074). However, trained sniffer dogs can also detect various types of cancer, malaria and some bacterial and viral infections with high diagnostic sensitivity and specificity.

Routine use so far only outside Germany

According to Volk, dogs have not yet been used in Germany, despite some promising field studies. However, other countries such as Dubai, Abu Dhabi, Mexico, the United States and Rwanda routinely use sniffer dogs.

“The dog is never used on humans directly, but a sample is created that is given to the dog,” explains the Hannover vet.

Training to become a Bulldog takes from 2 to 6 weeks, depending on the dog. A variety of dogs can be used, such as Malinois, German Shepherds, Labrador Retrievers or Cocker Spaniels. “Every dog ​​can be trained purely theoretically if it enjoys learning and smelling,” Folk Deem said. DE. © gie / aerzteblatt.de

Leave a Comment