This one’s bound to attract a lot of attention, and that can be both good and bad. A tiger at New York’s Bronx Zoo has tested positive for SARS-CoV-2. The tiger was sick (cough, decreased appetite) and was presumed to have been infected by a caretaker who was infected but asymptomatic. Three other tigers and three African lions also had a dry cough, but only one tiger was tested because sample collection in these animals requires general anesthesia.  All the affected big cats are expected to recover.  The zoo has been closed to the public since March 16, but the first signs of illness weren’t seen until March 27.

On one hand, it’s not too surprising. If domestic cats are susceptible, wild cats should be too.  On the other, there are some noteworthy aspects:

  • Zoo cats generally don’t have much direct contact with their caretaker (though this varies by zoo). Infection of these cats was therefore presumably through very transient direct, or indirect, contact.
  • Given the very limited testing of domestic cats to date, this yet again raises the question of how commonly domestic cats have been infected by owners in North America, given the large number of infected people and relatively high percentage of households with cats, and the much closer contact between people and pet cats (compared to zoo cats).

As always, don’t over-react. This just reinforces the message I’ve been saying since the January (which, despite all our talk about “One Health” led to a lot of criticism): We need to consider and investigate the potential animal aspects of this disease, such as human-to-pet transmission, in order to help prevent pets from contributing to human disease, and to help avoid creating a (domestic) animal reservoir for this virus. This is still a very much a predominantly human disease, but we can’t completely ignore the role of animals even if it’s small. We can use simple, practical and basically free measures to reduce the presumably low risks to and from animals, and we should:

  • If you’re infected, limit contact with animals.
  • If someone in your house is infected, keep your animals away from other animals and people.
  • Social distancing includes your pets. Keep your pets away from other people and animals outside your household, just like you should be doing with yourself.

Questions about mask use keep filling my inbox. I’ve written before about when mask are most useful, and my general advice hasn’t changed much. However, some recent developments have make me think more about when and where more routine mask use might be of benefit.

Why are we re-thinking when to use masks?  What’s changed?

  • There’s more SARS-CoV-2 transmission in the community in some areas. As that happens, simply identifying high risk people (e.g. travellers, those with known exposure) is less reliable for avoiding contact with the virus. The odds of anyone being infected get higher.
  • People can be infectious for a short time before they get sick.
  • There is increasing evidence that a fairly large percentage of infected (and infectious) people have asymptomatic infections (i.e. they don’t get sick but can still spread the virus).

These factors decrease our ability to confidently say “no, you’re probably not infected” about anyone; the more that happens, the more community measures such as routine mask use might make sense.

Key points to remember about masks and how they work:

  • The most important point is that in most scenarios masks mainly help protect others from the wearer if the wearer is infected. Masks reduce the release of droplets (i.e. slightly larger liquid particles) and aerosols (i.e. very small liquid particles) when the wearer breathes, coughs or talks. (Both droplets and aerosols can can contain virus, but droplets don’t travel as far.)   Masks are generally less effective at protecting the wearer.
  • The biggest benefit to the wearer might be preventing hand-to-mouth/nose contact, since we inadvertently do that all the time, and certainly increases the risk of exposure to whatever is on the person’s hands.  However, in some cases mask use can have the opposite effect if a person is constantly touching or readjusting their mask if doesn’t quite fit right or is uncomfortable.
  • Surgical and N95 masks are in short supply in many areas. We need to conserve these supplies to make sure they are available where they’re truly needed (especially for front line medical workers who are at the highest risk).

The CDC has now recommended “wearing cloth face coverings in public settings where other social distancing measures are difficult to maintain (e.g., grocery stores and pharmacies), especially in areas of significant community-based transmission.” Other major public health organizations haven’t followed suit (yet). Whether that will follow or they will hold off because of the lack of clear benefit, and lack of mask supplies, remains to be seen.

What about routine mask use in veterinary clinics?

Most veterinary clinics are doing a great job of social distancing from owners. It’s tougher to maintain social distance between staff within a clinic, but there are some things we can do.  Unfortuantely the nature of veterinary care requires us to be in close proximity to others at times , because most of our patients require some kind of restraint for many routine procedures (e.g. drawing blood, placeing an intravenous catheter). This has led to a lot of discussion about staff routinely wearing masks in clinics. Again, masks mainly help protect others when the wearer is (either knowingly or unknowingly) infected.  Those known to be infected of course should not be in the clinic at all, but the concern is there may be more and more people who are unknowingly infected and not showing any signs.  Wearing a mask may make such a person less likely to infect others. For people like me who are frequent face-touchers (my own face… not others… just to be clear), there might be some additional personal protection. So, there are some valid reasons to have a mask use policy for situations where people have to get close, BUT masks should never be used as a crutch or a substitute for physical distancing whenever feasible, or proper attention to hand hygiene.

Nonetheless, I’m hesitant to burn through our limited supply of surgical masks when we (and human healthcare workers) might need them more later in higher-risk situations. Here’s where cloth masks might be useful. Reusable cloth masks are a reasonable option in lower-risk situations where the 2 m/6 ft barrier has to be broken, and they help conserve supplies of N95 and surgical masks for higher risk situations, as well as surgical procedures.

Do cloth masks work?

Maybe. The literature is pretty sparse and mixed, with some showing evidence of at least some level of protection, one study showing an increased risk of flu-like disease, and other studies showing the considerable variation in the ability of different materials to actually prevent the dispersal of droplets and aerosols.

A recent paper in Nature Medicine (Leung et al. 2020) showed that surgical masks significantly reduced detection of SARS-CoV-2 RNA in aerosols from exhaled breath of infected people (see graph below). There wasn’t a significant reduction in virus from larger droplets, which is a bit counter-intuitive, since masks should be better at retaining larger particles. However, the results were pretty close to being statistically significant and they didn’t actually detect any virus in droplets or aerosols from mask users. (The small sample size probably limited their ability to find a significant difference.) Whether these results apply to cloth masks is hard to say and is probably highly dependent on the material type (and how well the mask fits). However, it provides support for the potential protective effect when infected people wear good quality, well-fitted masks.

 So masks may be somewhat effective at reducing the risk of virus transmission between people in certain situations. On the other hand, if we’re not wasting needed healthcare supplies by using cloth masks, then there’s limited downsides if we also use some common sense about it.

Should veterinary staff routinely wear (cloth) masks in the clinic?

  • It’s worth considering. From our veterinary infection control listserve, it seems like optional or mandatory mask use is becoming common in academic veterinary hospitals.
  • Whether to make routine mask use mandatory or elective is an interesting question. Saying “go ahead and wear one if you want” is the easy approach, but that’s not as useful for something that’s designed to protect others from the wearer.  In other situations, if someone doesn’t follow a particular recommendation, they just put themselves at risk, but in this case the person not wearing the mask may be increasing the risk to others as well. If a facility is truly concerned (or is trying to allay staff concerns), it might need to be mandatory for everyone in order to avoid conflicts within the clinic between those who do and those who do not want to wear a mask.
  • If mask use is going to be implemented for routine close contact procedures, cloth masks are a good way to preserve supplies. Rationing and reusing surgical masks has become a common approach, but that still depletes supplies pretty quickly, even if staff only get one mask per day or even every few days.

Even cloth masks are, not surprisingly, getting harder and harder to find. With the change in US CDC’s recommendation regarding routine use of cloth face coverings, I suspect they will become even more scarce.  Some clinics (and even community groups) are now making their own. There are lots of sewing patterns available online now, and the CDC even provides instructions for both sew and no-sew mask options.  I spent the evening playing around with designs and materials myself (see below).  FYI you can’t iron disposable surgical gown material – it will melt all over your iron.

Important reminder: Masks are not a cure-all. Wearing masks all the time while becoming lax at other critical practices (e.g. social distancing, hand hygiene) is counterproductive. Masks should be approached as something to use when social distancing isn’t possible, not as a means to avoid social distancing.

Here is an updated version of the decision tree for screening of owners / patients coming to veterinary clinics, which I originally posted a few weeks ago, to help with identifying higher risk situations. We also have a Russian translation courtesy of Dr. Varvara Solovyeva.

You can use the links below to download the pdfs:

Disclaimer #1 is my standard “COVID-19 is almost exclusively a human disease. Information about this virus in animals is important to investigate and consider, but don’t over-react.”

Disclaimer #2 is that this post is about another paper that’s available as a pre-print, meaning it hasn’t yet undergone peer review. However, it provides some useful new information that’s worth mentioning at this stage.

The study (Zhang et al. 2020) investigated the prevalence of antibodies  to SARS-CoV-2 in cats in Wuhan, China where the COVID-19 outbreak began.  One of the things we want to know is how often animals (cats, in this case) may get infected when they’re exposed to infected people.  Antibodies in the bloodstream indicate the immune system has responded to the pathogen, which generally means the animal was infected at some point.  Whether an infected animal can pass on the virus to someone else is a related, but separate question. This study didn’t look at the implications of infection of cats (e.g. clinical signs or transmission), just how often it occurred.

Blood samples were collected from 102 cats in animal shelters and veterinary clinics in Wuhan from January to March 2020 (i.e. during the COVID-19 outbreak). A set of 39 samples from cats that was collected from March to May 2019 (prior to the outbreak) was also studied.

  • Antibodies against SARS-CoV-2 were found using an ELISA test in 15 (14.7%) of samples. Eleven of those were also positive on a second type of antibody test (virus neutralization test).
  • Three of the positive cats belonged to owners known to have had COVID-19, and they had the highest antibody levels. Six were stray cats, and 6 were sampled at veterinary clinics but had no known contact with an infected owner.
  • None of the cats had positive PCR results, meaning there was no evidence that the virus was still present.
  • The limited knowledge of the cats’ contact with infected people limits interpretation of the results, but that’s a pretty high rate of seropositivity, especially among a group of cats that didn’t mainly consist of animals known to have been exposed to infected people.

All 39 blood samples taken from cats before the outbreak were negative. This is important to show that there’s no cross-reaction or non-specific reaction with other antibodies that may be present in the cats that would lead to a false positive result.

Overall, it’s not too surprising to see seropositive cats from infected households. If an infected person is present and cats are susceptible to infection (which we’ve already seen), it makes complete sense that seropositive animals would be found.

It was more surprising to see that high a number of positives from households without confirmed COVID-19 patients. Certainly, it’s possible that there were undiagnosed people in the households of some of them.

The strays are another interesting group. Were they pet cats that were caught as strays? Abandoned pets? In those situations, they may have been exposed to an infected owner. If they were truly feral cats, where did they get infected?  Contact with people who were feeding them? Indirect contact with infected people? Other animals? It’s hard to say.  We need to do more work rather than just keep speculating.

The take home messages remain the same:

  • If you’re sick, stay away from animals.
  • Keep your animals away from other people or animals.  Social distancing applies to the whole household, not just the human members.
  • Your own pet poses virtually no risk to you. If my cat is infected, he got it from me (in which case I’m already infected) or my family (who pose a much greater risk of transmission to me than the cat). If we keep pets with us but socially distanced from others, we don’t need to worry about them as sources of infection outside of the household.

I think most people agree that the best thing we can do with pets of people with COVID-19 is to keep them in the home with the owner. However, what do we do with someone’s pet if they have to be admitted to the hospital and they live alone?  Someone needs to take care of the pet.  But who, and how? That’s a contentious topic.

The risk posed by that pet to someone outside the household isn’t clear. The risks are presumably greatest with ferrets and cats, and risks from dogs are probably pretty limited and short-term. But, since we can’t say the risk is zero, we want to keep things contained as much as is practical.  Hong Kong’s approach of isolating and testing exposed pets isn’t going to be common, so we need another plan.

The main goal is to reduce contact of the animal with new people and new environments. Additionally, we want to make sure the animal stays away from people that are high risk for serious disease (e.g elderly, people with underlying respiratory disease or compromised immune systems, diabetics, smokers). That’s not always possible, but here is my take on some options, based on what we know to date.

Continued care in the pet’s household by someone who has recovered from COVID-19 (e.g. recovered family member goes to the house every day to care for the pet)

This may be the best scenario, since it appears that once you get over COVID-19, odds of a new infection (at least in the short term) are very low. If someone who has recovered can come into the house as needed to care for the pet, that keeps it contained and away from susceptible people.

Care by someone who has recovered from COVID-19 in the caretaker’s household

This is a good option too, assuming no one else in the caretaker’s household is susceptible (i.e. any other household members have also already recovered from COVID-19). It’s more convenient and the risk posed by moving the animal into a new house with non-susceptible people is probably inconsequential.  Preferably the household doesn’t include any other pets either, or there’s a way to make sure the animals are kept in separate areas from each other.

Care by someone who has also been exposed (e.g. family member that doesn’t live in the household but had close contact with the sick person)

As the pandemic progresses, we’ll have more people who have recovered. However, not everyone will have a recovered person available to help. If a close friend or family member of the pet’s owner has already been exposed, bringing the pet into their household adds some risk (since the person hopefully hasn’t been infected) but less than bringing it into a household with no previous exposure (where the people are even less likely to be infected). From a pet containment standpoint, it would be ideal if the pet stayed in its own household and the exposed person went there to take care of it; however, the exposed person should be self-isolating and therefore not leaving their own house for 14 days. An exception might be if it was an immediate neighbour, where going next door poses no risk of exposing anyone else, but that would ideally be figured out with public health authorities.

In any situation, I would not want an exposed pet to be brought into a household where there were high risk people.

Care by someone who’s not recovered or exposed

This is less desirable since you’re bringing an unexposed person (or at least one that’s not known to have been exposed) into the situation. If that person is low risk for serious disease and doesn’t have high risk people at home, that helps. This situation increases the risks, though. The best approach would be short term, controlled visits to the pet’s household, just for basic animal care, and ideally wearing some form of personal protective equipment (especially for the first couple days – after that, at least there wouldn’t be any significant concerns with contamination of the pet’s haircoat). By keeping the duration of contact short, controlling the type of contact, and using good hygiene practices (especially hand hygiene), exposure risks can be limited.

Taking the animal into the new caretaker’s household increases the risk because the pet and the person will likely have increased duration and closeness of contact, and there’s less ability to use barriers and hygiene around any contacts. If the animal can be kept in a cage or crate all or most of the time, or at least for the first few days to negate coat contamination concerns (not ideal for the pet but reasonable in the short term), that would help.

Other options

If the “friends, family and benevolent neighbour” approach isn’t an option, then it gets trickier. Temporarily transferring the pet to a shelter would be a consideration, if the shelter is able and willing to handle potentially contaminated or infected pets, but that varies.  There’s been a lot of planning by shelters for this contingency.

Vet clinics are another option. However, it’s not ideal given the limited isolation capacity in most clinics, as well as limited staffing and operations in some. Clinics that have the facilities, personnel and comfort with handling infectious cases would be an option. (We are set up to handle potentially infectious cases like this, but even our facility can only handle a limited number of animals.)

If exposed animals have to be moved from a household where there was a person with COVID-19, a bath or disinfectant wipe of the haircoat is reasonable to try to reduce any risk from contamination of the haircoat. We have no idea if this is a real risk, but the virus would presumably survive for some time on a hair coat if deposited there by the owner. How long? We don’t know. Probably minutes to hours, but that’s unclear. That’s why bathing and some additional short term precautions (e.g. cage/crate) might help. After a couple of days, we just need to worry about whether the animal is truly infected. We still don’t understand how common that is, or whether an infected animal can infect a person. Hopefully we’ll sort that out more soon. I’m set up to do surveillance on exposed animals but it’s been very difficult getting into households to get samples, for obvious reasons. As we learn more about how commonly or rarely exposed animals are infected, we can refine our recommendations. For now, it’s a tough balancing act between being prudent and practical.

We’ll soon reach the time when I won’t bother reporting on every new instance of SARS-CoV-2 infection in an animal in contact with an infected person, but at this stage a little more discussion is probably still warranted. As part of Hong Kong’s One Health approach to COVID-19 (one that is sadly very rare), they are investigating human-to-animal transmission of this virus by testing pets of COVID-19 patients that have had to be put in quarantine when their owners were too sick to look after them. In addition to two positive dogs that were previously reported, the Hong Kong Agriculture, Fisheries and Conservation Department has now identified a cat that tested positive for SARS-CoV-2. The cat’s owner has COVID-19 and samples from the cat’s mouth, nose and rectum were positive for the virus by PCR.

This is not a surprise at all at this point.

The cat hasn’t shown any sign of illness. That’s an important thing to keep watching, as it’s unknown how commonly (if ever) cats may get sick when they are infected, but has obvious implications for cat health, as well as potential control measures and our overall understanding of the virus.

The messages remain the same:

  • If you’re sick, stay away from people AND animals (as much as possible).
  • If your pet has been exposed to someone with COVID-19, keep it away from other people. The best thing to do is simply keep the animal in the affected household if it’s been exposed. That way, if it does get infected, it can’t spread it outside the household.
  • If an exposed pet has to leave the house (e.g. to go to a veterinarian for urgent medical care), a plan should be in place to limit the risks to anyone handling the cat in transit and on arrival at the destination.

Hot on the heels of the experimental study of SARS-CoV-2 in ferrets that I discussed a couple days ago is another new study, available in pre-print, that looked at susceptibility to SARS-CoV-2 in a slightly wider range of animals, including ferrets, cats, dogs, pigs and poultry (Shi et al. 2020).  Bear in mind that the study is only a pre-print (made available by bioRxiv), meaning it is only a preliminary report and hasn’t yet been peer-reviewed.

The overall take-home was that, as previously reported, SARS-CoV-2 replicates well in ferrets, which isn’t surprising. The authors also reported that cats seem to be quite susceptible to infection. That’s not  surprising either based on cats’ susceptibility to the original SARS virus, but it raises the stakes a bit in our need to investigate natural infection in this species (i.e. in people’s pets). The good news is they didn’t find any evidence of infection in pigs or poultry. Read on for more details.


In this experiment, ferrets were infected through intranasal administration of virus, and live virus was detected from nasal washes days 2-8 after infection.  Testing for the virus is often done using PCR, a method that detects viral RNA, but can’t differentiate between “live” (infectious) virus or dead virus. Virus isolation (usually done in cell culture) is more difficult, but lets us know if live virus is present. Here, there were positive results with both techniques.

Lower levels of virus were detected in feces by PCR but live virus wasn’t recovered. The virus seems to replicate mainly in the upper respiratory tract. Unlike the previous study, disease wasn’t as consistent, with only some of the ferrets developing fever and loss of appetite. Antibodies were detected in all ferrets, consistent with infection.


Subadult (8 month old) cats were inoculated intranasally, as for the ferrets, but the reserachers didn’t collect routine nasal samples from the cats because they were too aggressive to risk handling them that often (the risk of human exposure through a bite probably played a role here). Virus was detected in feces of some cats on day 3 and all cats on day 5. Two cats were euthanized 6 days after infection and live virus was detected from the upper respiratory tract of one. In the other, viral RNA was detected by PCR in the small intestine, but live virus wasn’t detected.

To look at transmission, they also placed 3 uninfected cats in adjacent cages. Virus was detected by PCR in 1 of 3 cats and all three developed antibodies against the virus. It was concluded that the virus was transmitted to all three cats, presumably as a result of respiratory droplets moving over a short distance from one cage to the other (just like in people).

Testing was then repeated with younger cats (70 to 100-days-old). The results in the preliminary report are sparse, but they said younger cats were more permissive to infection, with  “massive lesions in the nasal and tracheal mucosa epitheliums, and lungs of both cats”.  No obvious illness was mentioned, though.


Five three-month-old beagles were inoculated with virus and housed with 2 uninfected beagles. Viral RNA was detected using PCR in feces in 2/5 inoculated dogs on day 2 and in one on day 4. Live virus wasn’t isolated. The dog was that positive on day 2 was euthanized and virus wasn’t detected in any tissues. However, 2 of the 4 remaining dogs developed antibodies against the virus, indicating they had been infected and their immune systems had responded. There was no apparent transmission to the two dogs co-housed with the inoculated dogs. These data suggest that dogs are susceptible to infection but that susceptibility is low and there is less risk of transmission from an infected dog to another dog (or person) compared to ferrets or cats.

Other species

The researchers did similar studies with pigs, chickens and ducks. None became infected. Pigs are the noteworthy species here since there was some concern about their susceptibility based on genetic analysis of the virus’ receptors. It’s a small study that needs to be replicated but that aspect was encouraging.

Take home

Overall, the ferret results are not surprising. Ferrets are clearly susceptible to this virus. Cats also appear to be susceptible but are less likely to get sick. No shock there either, though the fact that cats could transmit the virus without direct contact raises some concerns. It was a very small transmission study so we need to see more data from other studies. The dog results are encouraging and support a low risk of infection of, and from, dogs.

Ferrets and cats remain our main focus, both in terms of keeping infected people away from them (so the animals don’t get infected) and keeping exposed animals away from unexposed people. In a household with a person with COVID-19, the risk to others in the household is still mainly from that person. However, we want to make sure pets don’t track it out of the household if people don’t recognize the potential for animal infection. The size of the overall impact of this is completely unknown, but it makes sense to take simple steps to reduce exposure of pets and keep exposed pets away from others.

A couple of new guideline documents have been released.

First up, a Canadian COVID-19 FAQ for veterinarians, created by a working group of Canadian public health and animal health experts (French translation to come).

Also, there’s a revised version of animal shelter guidelines from the US: “Interim recommendations for intake of companion animals from households where humans with COVID-19 are present”, a collaboration of the American Veterinary Medical Association, University of Wisconsin-Madison Shelter Medicine Program, the Association of Shelter Veterinarians, University of California-Davis Koret Shelter Medicine Program, University of Florida Maddie’s Shelter Medicine Program, and the CDC COVID-19 One Health Team.