The current COVID-19 pandemic is caused by the SARS-CoV-2 coronavirus, which is thought to have originated in animals. This jumped species barriers to infect humans and is now showing rapid and easy transmission between them.
A new study* shows that reverse transmission is probable, with 3-4% of a large number of domestic pets showing antibodies to the infection, though no virus was recovered from any animal.
Almost from the beginning, there have been reports that the virus can infect pet cats and dogs, with some animals showing symptoms of infection. Though these fears were initially decried, sporadic cases continue to be reported.
In these pets, the respiratory or fecal specimens, or both, have tested positive for the virus by reverse transcriptase polymerase chain testing (RT-PCR). Specific antibodies against the virus have also been detected in pet sera.
Targeted experiments also show that dogs are not easily infected and mostly develop an asymptomatic infection, with low viral titers being shed. On the other hand, cats show respiratory infection and shed high titers of the virus, and spread it to other animals as well.
The study: testing pets for SARS-CoV-2
The current study aimed at a more wide-scale testing of animal infection in their natural farm or home conditions. The researchers carried out a comprehensive survey of dogs and cats in Italy, from March to May 2020, in families with cases of COVID-19 or families living in severely affected areas.
Their own vets tested all the animals in the study during routine visits, including over 900 dogs and over 500 cats.
The samples were from nasopharyngeal, oropharyngeal or other severely affected areas in humans, or from other convenient sites. This yielded approximately 300 and 180 oropharyngeal swabs, 180 and 80 nasal samples, and 55 and 30 rectal swabs from dogs and cats, respectively.
Altogether, there were 1420 swabs, including around 40 dogs and cats each that were symptomatic at the time of sampling, and about 60 dogs and cats each from families that had one or more positive cases.
However, all were negative on PCR, including those living in households with confirmed cases of COVID-19 and those with and without respiratory symptoms. This suggests the animals were not actively infected at that time.
Serum samples from around 190 and 60 dogs and cats were available along with the full history and location, and 200 and 90 cats approximately, lacking history but with known locations.
The virus was tested for by RT-PCR targeting the viral nucleoprotein and envelope protein antigens. Plaque neutralization assays for neutralizing antibodies were also carried out to find the highest dilution at which the plaque number was reduced by 80%.
This revealed the presence of specific neutralizing antibodies in 13 dogs and 6 cats, amounting to about 3% and 4% each. The titers detected ranged from 1:20 to 1:160 and from 1:40 to 1:1280 in dogs and cats, respectively. None of these animals were symptomatic at the time of testing.
COVID-19-positive household distribution
The break up among the dogs was as follows: 6/47 dogs and 1/22 cats from COVID-19 positive households, 1/7 dogs from households suspected to be positive, and 2/133 dogs, and 1/38 cats, from households negative for COVID-19.
This means that in households confirmed or suspected to have at least one case, 13% to 14% of dogs were antibody positive, as compared to 5% of cats in only confirmed positive households. In negative households, this dropped to approximately 2% of animals, whether dogs or cats.
Age- and sex-stratified distribution
When stratified by age, they found that of 423 animals of known age, none were infected among the animals below one year. About 7%, 3% and 3% of animals aged 1-3 years, 4-7 years and 8 years or more were positive.
Some important associations were made. When there were 10 or more samples available, the human case count was strongly and positively correlated with the positive tests in dogs, and also with cats, but to a smaller and less significant extent. Community sampling in humans yielded a comparable seropositivity percentage at a similar period in Europe.
The evidence supports the ability of dogs and cats to seroconvert when living as pets in a COVID-19 positive household and in regions with high burden of human disease. The greater tendency of dogs to develop neutralizing antibodies may reflect the greater susceptibility of dogs to the infection.
More male than female dogs were infected, which may be due to the physiological differences in the sexes. This is different from humans, where infection rates are similar in both even though the disease severity is greater in males.
Implications and importance
The researchers point out, “This is the largest study to investigate SARS-CoV-2 in companion animals to date. We found that companion animals living in areas of high human infection can become infected.”
As often seen in humans, none of the animals below one year of age developed infection as assessed by PCR. This agrees with earlier research findings, and also indicates that older animals should be used in experimental studies, since otherwise the true susceptibility of the animal model may not be detected.
All animals tested positive by PCR, despite the significant percentage of seroconversion. This may mean that viral shedding is very short-lived in pet animals.
This has been observed in studies showing that shedding in cats ceases by 10 days following experimental infection, and neutralizing antibodies are detected by 13 days. In dogs, fecal samples showed the presence of the virus at up to 6 days post infection, but oropharyngeal swabs were negative.
The study notes that a natural infection in a Pomeranian, among the earliest reported, was associated with positive viral RNA in nasal swabs for 13 days, albeit at low levels, but not in fecal or rectal specimens. This may indicate variation in shedding pattern between animals.
Moreover, in another experimental animal study, half the dogs who were infected had demonstrable antibodies by 14 days. This indicates the difficulty in SARS-CoV-2 detection whether in humans or animals.
In the current study, the period that elapsed from infection to seroconversion is unknown. Even if the time of sampling was known, there could well have been delays in sampling due to the difficulty of visiting the vet during the period of lockdown. Therefore, the researchers advise that pets also be sampled to understand the true incidence of infection and viral shedding in the household and the community.
The researchers say that pets are unlikely to be an important route of viral spread, but when animals are present at high density, as on mink breeding farms, the virus may spread from animals to humans more readily.
Once the human-to-human spread is terminated, contact tracing will become more important. At that point, serologic surveys of pets may help provide a broad picture of the changing disease conditions within the community and an early warning of any transmission route left open.
bioRxiv publishes preliminary scientific reports that are not peer-reviewed and, therefore, should not be regarded as conclusive, guide clinical practice/health-related behavior, or treated as established information.