A recent paper in the Journal of Clinical Pharmacy and Therapeutics entitled “A doggy tale: Risk of zoonotic infection with Bordetella bronchiseptica for cystic fibrosis (CF) patients from live licensed bacterial veterinary vaccines for dogs and cats” (Moore et al. 2021) discusses (as the title suggests) human health risks from commonly-used B. bronchiseptica vaccines for pets.

Bordetella bronchiseptica is just one of a few different bugs that causes “kennel cough” in dogs (more accurately called canine infectious respiratory disease complex (CIDRC)). A variety of vaccines against B. bronchiseptica are available, including both oral and intranasal formulations that contain “modified live” bacteria, and injectable formulations that contain killed bacteria. Modified live vaccines (MLVs) contain attenuated (weakened) forms of the bacterium or virus in question that are not supposed to be able to cause disease, but induce a more natural immune response. So MLVs aren’t completely innocuous, and therefore generally aren’t used in immunocompromised individuals, because of the chance that even a modified/weakened bug could cause disease in such a person.

Bordetella bronchiseptica causes disease in a number of different animal species, but seems to be a rare cause of disease in people (unlike it’s cousin, Bordetella pertussis which causes whooping cough). However, infection with B. brochiseptica can occur in people, and those with diseases like cystic fibrosis (CF) are presumably at higher risk.

The authors of the paper state that patients with CF “should avoid exposure to live veterinary bacterial vaccines and seek animal vaccination utilising non- live vaccines.

  • I agree with point #1. High-risk individuals should avoid direct exposure to live vaccines, which can occur during vaccination of the animal, as the vaccine is squirted into the dog’s mouth or nose (and sometimes splattered elsewhere). Ideally, high-risk owners should not be in the room when such a vaccine is given. That’s a very practical, very easy and probably the most effective preventive measure.
  • I’d argue against point #2. Injectable killed vaccines for B. brochiseptica are inferior to MLVs, and that has relevance to the exposure and health of a high-risk owner too.

Here is my thought process when is comes to this situation:

  • No vaccination or less effective vaccination increases the risk of disease in the pet.
  • Bordetella bronchiseptica can cause disease in high-risk people, so we don’t want the pet to be infected.
  • Disease probably also increases the risk of exposure of people to this bacterium and others (from coughing/sneezing pets).
  • Disease also increases the risk that the pet may need to be treated with antibiotics, leading to an increased risk of antibiotic resistance in other bacteria carried by the pet, and some of those bugs can also be transmitted to people.

Millions of doses of these MLVs have been given to dogs with little to no clear evidence of risk to people. The main reference to which the authors point is a report about a mild infection in a boy who was squirted directly in the eye with a vaccine. That’s a lot different in terms of exposure than having contact with a recently vaccinated dog.

The issue of residual modified live bacteria from the vaccine being present in the dog’s nose or mouth for a while after vaccination is usually raised. That’s fair, to some extent, but it ignores the big picture. Yes, there is a very minimally risk that the modified live bug might be present in the dog’s nose/mouth, but there are lots of other (and more dangerous) bacteria in the nose/mouth of every dog. The risk is basically no different from a dog that was recently vaccinated and one that has not been vaccinated, because it’s the more common bacteria found in both dogs that I’m most worried about.

The statement that vaccination “requir[es] a period of CF patient exclusion from the shedding dog,” is not supported by anything I’ve ever seen and doesn’t make sense to me given the above thought process.

Like most things, we need to consider the cost-benefit in each situation.

What’s the human health risk of using MLVs for B. brochiseptica in dogs?

  • Exceptionally low.

What’s the benefit of using MLVs for B. brochiseptica in dogs?

  • Improved animal health, and I could argue reduced human health risks from decreased exposure to sick animals (because we have to think beyond just the risk from the vaccine).

There’s also a statement in the paper that “CF pharmacists, hospital pharmacists and community pharmacists are important custodians of vaccine-related advice to people with CF, who are frequently consulted for such advice. “

  • Very true. However, I’d add the need for a One Health approach. Veterinary input is needed for a proper risk assessment, and to put the issues into context for the individual pet/pet owner. It would be nice to see papers like this written in collaboration with veterinary experts, and for pharmacists and veterinarians to engage more with each other in situations like this. Connections between pharmacists (and many other human healthcare professionals) and veterinarians tend to be pretty poor.

“Kennel cough” (now more conventionally termed “canine infectious respiratory disease complex’)  is a fairly common problem in dogs that can be caused by an array of bacteria and viruses. We commonly see it in outbreaks, often linked to kennels, but sometimes we see higher levels of disease in the broader community. What we’re more concerned about is new problems , new patterns or more severe disease.

We may be seeing an increase in respiratory disease activity in dogs in a few parts on Ontario at the moment. It’s always hard to say for sure because it’s based on information from different sources, and whether it’s a true increase, an increase in reporting of the normal amount of disease, or just a misperception is hard to say.

We don’t want to over-react, but we also don’t want to miss the start of something important, so we’re paying attention to the information that’s coming in and trying to make sense of it.

An important limitation to the available data is the amount of diagnostic testing that gets done. Only a small percentage of dogs with “kennel cough” get tested to try to determine which viruses and bacteria are actually involved.

Should all dogs with “kennel cough” be tested?

  • No.  Since a lot of pathogens can cause the clinical signs we see with this syndrome and we can’t test for them all, the test results rarely impacts how we treat an individual dog. It’s nice information to have but it’s usually hard to justify the cost for an average household pet. This recommendation is also part of the 2017 ISCAID  treatment guidelines for respiratory disease in dogs and cats.

When is testing more rewarding?

  • Testing is more useful when something is unusual about the scenario or the patient. From the patient standpoint, testing can be more useful when it involves a kennel or shelter, since the result could affect the infection control response. It also can help differentiate “vaccine breakthrough” from the presence of a bug we can’t vaccinate against.
  • By “unusual scenario” I mean something different in the incidence, distribution or severity of disease. If we think we’re seeing more disease, testing is useful to see if disease is mainly caused by one bug, whether we have a mix of causes, or whether the cause can’t be identified (suggesting something different/new might be present).

When do I really want to test dogs with respiratory disease?

  • When I’m concerned about a foreign disease like canine flu, I definitely want to get testing done. Figuring out when to worry about that comes down to two big factors: high attack rates and links to imported dogs, especially from Asia. When most dogs in a group get sick, I worry about something new like flu and want to test them, so that we find out as early as possible if flu is present and we can take measures to contain it, like we successfully did last year. A high attack rate was what led to identification of the biggest flu cluster we had when canine flu hit Ontario in 2018.

So, what about now in Ontario?

  • I’d like more information but don’t have any money for testing, so I’m relying on information that comes in from various sources. It’s always a fine balance between raising awareness and causing paranoia, so it’s important to put things in perspective. We’re on the lookout for respiratory disease in dogs and want to learn more, but we’re far from panicking about the situation.
  • The average dog owner doesn’t need to do anything more than good routine care and using common sense. However, we’d like to figure out if something new or interesting is going on.

More to come (hopefully).

Kennel cough, also (and more properly) referred to as canine infectious respiratory disease complex (CIRDC), has been in the news lately. This condition is a syndrome, not a specific disease, being potentially caused by a range of bacteria, viruses and Mycoplasma, including canine parainfluenza virus, canine influenza virus, canine respiratory herpesvirus, canine adenovirus, distemper virus, Bordetella bronchiseptica and Streptococcus zooepidemicus. Regardless of the cause, it’s still a highly infectious disease characterized by a hacking cough. Serious illness, including deaths, can occur but is uncommon.

Here are a couple of kennel cough issues have hit the press lately:

  • Mandatory kennel cough vaccination is now required for dogs competing in the Iditarod Trail Sled Dog Race. It’s easy to see how this disease is a concern in these sled dogs, given the stress and rigours of competing and the mixing of many dogs from different areas. Kennel cough vaccination doesn’t prevent all cases, since it only protects against Bordetella bronchiseptica +/- parainfluenza virus, but it’s a useful infection control tool in high risk populations. The requirement has been implemented in part due to cases of kennel cough that were encountered in the 2011 race, along with the publicity that was generated (including the attention of PETA).
  • In Rocklin, California, a dog park was closed for two days because of a kennel cough outbreak. It seems the closure was in response to the diagnosis of kennel cough in two dogs, and it’s an unusual move given the apparently low number of cases. The issue isn’t the park environment itself being biohazardous – rather, the park provides an opportunity for dog-dog transmission. Given that, it’s a questionable control measures since it’s unlikely that people will keep their dogs at home. Rather, they’ll probably just go to another park, where the same risks will be present. It’s a bit like the debate around school closures with pandemic influenza. On the surface, it seems like a good idea, since kids won’t pass around flu at school. However, in reality, what happens is kids congregate at the mall and other places if schools are closed, so it just moves the site of transmission somewhere else and probably doesn’t have any net benefit. Here, a better response would probably be an educational campaign to get people to keep sick dogs at home, have people keep their dog away from other dogs at the park and encourage vaccination of high risk dogs (which would include those that go to a park and interact with other dogs).
  • A kennel cough outbreak was reported in Bozeman, Montana, with veterinarians asking owners to be on the lookout for disease. Local veterinarians reported a spike in the number of cases, with one clinic reporting  around 20 cases in the past month, which is a pretty remarkable number for your average vet clinic.
  • And locally… nothing specific, but I keep getting reports of clusters of respiratory disease in dogs. We often don’t get a chance to investigate small clusters to figure out the cause, since information often gets to me after the fact, but it’s a recurrent problem in Ontario. Most of the reports are rather poorly defined clusters of sick dogs, with occasional severe outbreaks involving fatalities (including one I’m dealing with at the moment).

An article from NBCMontana.com describes a kennel cough outbreak in dogs in Bozeman, Montana. It’s a pretty basic article that outlines a rather typical presentation of kennel cough (now largely referred to as canine infectious respiratory disease complex – a respiratory infection that can be caused by a range of viruses, bacteria and Mycoplasma).

As part of the story, they state that if you have a sick dog, the "best course of action is to call your local veterinarian and get medication." I realize it’s a quick statement, perhaps tossed in without much consideration, but there are some important issues to consider.

Should someone call a veterinarian and get medication, or should a veterinarian actually see the dog?

  • Sometimes dogs just need to be given time and rest. Viruses are often the cause of this condition, and it just takes time for the infection to resolve (just like person with a cold virus). If that’s the case, a little over-the-phone veterinary advice might be fine. If drugs are needed, then the dog needs to go to a veterinarian. Affected dogs might need something to control cough, which need to be given by prescription, and occasionally antibiotics are needed, but in either case a veterinarian needs to see the dog first. If the dog is sick enough that it needs additional treatment above and beyond this, then of course it needs to be seen by a veterinarian.

Are there any problems with a dog like this going to the veterinarian?

  • Here’s where the ball often gets dropped. The last thing we want to see is someone walking through the from door with a hacking, biohazardous dog who goes nose-to-nose with other dogs in the waiting room, breathes on half of the surfaces in the room, sits there for ten minutes while waiting for the appointment, and gets handled by every staff member before they realize the dog might be infectious. A situation like that can turn a veterinary clinic into a source of infection for many other dogs, and help an outbreak spread.

A very basic but well coordinated approach can greatly reduce the risk of dogs infecting other dogs in the clinic. These would include:

  • Not taking a biohazardous dog into the waiting room. The owner can call from the car upon arrival or come in without the dog to let the clinic know they’re there.
  • The dog can be admitted directly into isolation or an exam room, thereby avoiding contact with other animals in the waiting room or elsewhere in the clinic.
  • Veterinarians and techs that are going to work with the dog can know in advance and come in prepared, wearing appropriate protective outerwear (e.g. gloves and a labcoat or gown that they use for only that appointment) to prevent contamination of their clothing or body.

Very easy to do. Probably very effective too, but often not done.

It’s very common for kennels to require dogs be vaccinated against "kennel cough" before they are allowed in. There are two main reasons for this:

  1. Reducing the risk that a dog will bring kennel cough into the facility and spread it to other dogs.
  2. Reducing the likelihood that a dog will acquire kennel cough if someone else brought it in.

Overall, it’s a sound policy, but it’s far from 100% effective and it needs to be part of an overall kennel infection control program to work. Relying solely on vaccination to prevent kennel cough is a weak approach that can ultimately fail, particularly if other infection control practices are poor or if vaccination protocols are illogical.

Why isn’t it 100% effective?

1) Kennel cough is a syndrome, not a specific disease. It can be caused by many different viruses and bacteria, often in combination. Kennel cough vaccines are typically targeted against Bordetella bronchiseptica +/- canine parainfluenza, two important causes of kennel cough, but not the only causes.

2) No vaccine is 100% effective. Vaccines help reduce the risk of illness, but they don’t completely eliminate it. Some vaccines are better than others, and some animals respond better to vaccines than others.

3) Timing is another issue. One of the weak points of many kennel protocols is the requirement that the dog be vaccinated "before entry," or within a certain number of weeks or months. The problem with this is vaccines are not immediately effective. What often happens is people decide to board their animal at the last minute or realize the night before that they need their dog vaccinated, so the vaccine gets given a day (or less) before kenneling. The intranasal kennel cough vaccine (squirted up the nose) takes a few (3-5) days to be effective, and the injectable vaccine takes even longer (a week or more). Vaccination very soon before boarding, particularly for a dog that has never been vaccinated against kennel cough before, is unlikely to result in protection from infection by the time of boarding.

Requiring vaccination before boarding makes sense, but it’s important to remember that:

  • It’s not 100% effective.
  • It doesn’t negate the need for a good infection control program.
  • It needs to be given at an appropriate time to be effective.

I had a call from a colleague in Ottawa (Ontario) the other day, asking if I’d seen an increase in kennel cough in dogs lately. Kennel cough is a respiratory infection of dogs that can be caused by a variety of different viruses and bacteria, or combinations thereof, but is often associated with the bacterium Bordetella bronchiseptica. Apparently, this colleague’s clinic has seen a large number of cases compared to normal, and he was wondering if the trend was more widespread and/or if there’s something new out there to be concerned about.

Informal reports like this are often the key to identifying new problems. There are only a few reportable diseases of companion animals (such as rabies), and existing federal and provincial public health and animal health agencies tend to have little mandate regarding non-reportable infectious diseases of companion animals. That means that there is no centralized reporting or investigation for all these other diseases (in other words: we’re on our own).

Most often, reports of higher disease rates or suspected outbreaks don’t end up leading to anything. Things tend to revert back to baseline fairly quickly without any explanation of what happened. Sometimes, however, reports like this are the first in a series that can flag the emergence of a new disease or a change in existing disease patterns.

Is anything actually going on with kennel cough in Ottawa? It’s hard to say. A report like this could be due to:

  • A focal outbreak caused by exposure at a single kennel, park or event.
  • A local outbreak of "run-of-the-mill" kennel cough that is being spread from multiple sources, but which involves the normal kennel cough bacteria and viruses.
  • Increased reporting of the normal baseline rate of disease, with increased awareness leading to the appearance of an outbreak.
  • A new disease (either a brand new disease or, more likely, the first instance of an existing disease in the area).

Whenever I hear reports like this in Ontario, I think about canine influenza. This virus is present in dogs in many regions of North America, but we have yet to identify it in Ontario (at least from the last data I have. We also couldn’t find any evidence of canine influenza virus in a surveillance study we did a while ago). It is certainly possible that this virus could make it to Ontario, and I would not be surprised at all if canine flu caused a readily detectable cluster of respiratory disease cases when it arrived.

Should dog owners in Ottawa be worried? No.

Should dog owners and veterinarians in Ottawa pay attention? Sure. It’s always good to be aware of things that are happening locally. Dog owners need to be aware of the risk of exposure to a variety of infectious diseases. Veterinarians should consider testing for canine influenza (and dog owners need to be willing to pay for the testing) if they see changes in respiratory disease patterns in their area.

How can dog owners reduce the risk of exposure of their dogs to respiratory diseases? Common sense. The more dogs that a dog meets, the closer they get to them and the less vaccination in the population, the greater the risk. Kenneling and other situations where many dogs get together increase the risk, and preemptive kennel cough vaccination should be considered in such cases. This vaccine doesn’t protect against all causes of respiratory infection, but it can protect against some of the most likely causes. People should keep their dogs away from other dogs that look sick (especially dogs that are coughing), and if they have a sick dog, they should keep their dog away from other dogs for a few weeks.

(click image for source)

A large whooping cough (pertussis) outbreak has been ongoing in people California in 2010. This bacterial infection, caused by Bordetella pertussis, is a highly transmissible disease that can result in serious problems (including death) in young infants. At last report, there were over 6000 cases of whooping cough, making this the largest outbreak in 60 years. Over 200 infants have been hospitalized, and there have been at least 10 deaths. Nine of the 10 deaths were in infants less than two months of age.  Infants in this age group have little to no immunity to the disease because they haven’t been vaccinated, and they are more prone to severe complications.

Bordetella pertussis is a human bacterium. It does not infect animals and animals are not direct sources of infection. (Actually, experimental infection of neonatal puppies with large doses of B. pertussis can result in shedding of the bacterium by a small percentage of dogs, but that’s not particularly relevant to the normal household situation). Therefore, people don’t need to worry about infecting their pets and pets passing the infection on to other people. However, it’s not impossible that pets could play an indirect role in transmission. A pet’s haircoat could possibly become contaminated with the pertussis bacterium from someone coughing around it, or touching it with contaminated hands. The bacterium could survive on the haircoat for a while (probably days), and someone could potentially get the bacterium on their hands by petting it, and subsequently become infected.

What are the odds of this happening? Who knows. It’s not something that anyone has investigated, as far as I know.

Could dogs and cats be important sources of pertussis in households? Probably not. I assume that if there is a person with whooping cough in a household, that person is more likely to be the source of infection for other people than a pet. 

Could pets spread pertussis outside the home? That might be a more realistic concern. People with pertussis might keep themselves away from others and stay at home, but if they contaminate their dog’s coat and the dog meets people on a walk or at the park (or at a veterinary clinic, or anywhere else), I have to wonder whether there could be the potential for spread of the disease.

What should we do about this? Common sense should prevail, and itt’s important for pertussis as well as other diseases. If someone in the household has an infectious disease that is transmissible and for which a pet could potentially be a vector, some basic precautions should be taken. Good attention to hygiene might help reduce contamination of the pet’s haircoat. This includes regular handwashing (especially after coughing and before petting an animal), avoiding coughing close to the pet and not letting the pet sleep close to the person’s head. Keeping the pet away from people outside the house, or at least limiting it’s contact with high-risk people might also be useful. In particular, keeping pets that might have been contaminated away from infants would be wise.

Overall, the risks are very low. We don’t need to fear dogs and cats as potential pertussis vectors. However, in the absence of proof that there’s no risk, and with a highly transmissible and potentially serious disease, use of some simple infection control measures makes sense. https://youtube.com/watch?v=KZV4IAHbC48%3Ffs%3D1%26hl%3Den_US

Kennel cough is a highly infectious respiratory disease in dogs. The disease got its name because infection and outbreaks often occur in kennels, where many dogs from various backgrounds are mixed together.  Some of the dogs in kennels may be carrying infectious diseases, and other dogs may be very susceptible these diseases – putting them all together in what can be a stressful environment for any dog creates a recipe for infection.

Kennel cough itself is more of a syndrome than a specific disease. It can be caused by a few different bacteria and viruses (and combinations thereof) that produce the same type of clinical signs. These pathogens include Bordetella bronchiseptica, Mycoplasma, canine parainfluenza virus, canine adenovirus-2, canine distemper virus and canine herpesvirus. Of these, only Bordetella bronchiseptica is a potential concern in terms of transmission to humans.  Bordetella bronchiseptica can cause respiratory infections in people, but this is probalby quite rare and largely confined to high-risk individuals, like those with a weakened immune system, who have had their spleen removed, who already have underlying respiratory disease of another kind, and pregnant women. The evidence of transmission of B. bronchiseptica from pets to people is relatively weak and circumstantial – it is not clear whether the human Bordetella infections in these cases were truly due to contact with a pet.

In households with individuals with a weakened immune system (e.g. HIV/AIDS, transplant or cancer patients), some measures that can be taken to help reduce the risk of transmission of Bordetella from pets include:

  • Avoid boarding dogs at kennels or veterinary clinics. If boarding cannot be avoided, ensure that dogs have been properly vaccinated against Bordetella bronchiseptica, and that the kennel or clinic requires all other dogs boarding there to be vaccinated as well.
  • Avoid obtaining a dog directly from from an animal shelter.
  • Wash your hands regularly after handling any dog, particularly if you’ve touched the dog’s  nose or mouth.
  • Don’t allow dogs to lick your face or hands.

The Bordetella vaccine for dogs is a modified live vaccine, meaning that a live but less virulent (pathogenic) form of the bacterium is administered to "prime" the immune system against regular Bordetella. Concern has been raised by some people that exposure to the vaccine strain could actually cause disease in high-risk individuals, whose immune systems might be unable to fight off even this "weaker" version of the bacteria.  It is unclear whether exposure to the vaccine strain actually poses any risk.  Nonetheless, it is reasonable to pay extra attention to avoiding contact with the dog’s face for a day or two after vaccination. Also, it is probably wise for immunocompromised owners to not hold the dog when it is being vaccinated, in order to reduce the risk of exposure to the vaccine. 

What we know about SARS-CoV-2 in pigs hasn’t changed a lot since the first version of this post. It’s still a fairly “good news” situation, but one that could also use some more investigation.

Are pigs susceptible to SARS-CoV-2?

Kind of, but not really.  There are somewhat conflicting experimental data, but the debate is really whether pigs have very little vs no susceptibility to the virus. There is obviously a difference between “no” and “yes, but only a little,” but from a big picture standpoint, we don’t currently have evidence that there are issues for pig health, pigs as a source of infection in people, or pigs as a potential reservoir for emergence of new virus variants.

Why did we talk a lot about pigs initially?

At the start of the pandemic, we were worried about the potential for this virus to infect pigs because of their susceptibility to the original SARS virus, and because it was predicted that they would be quite susceptible based on their ACE2 receptor. ACE2 is the receptor that the virus uses to enter the cells of the host. If the virus can’t enter cells, it can’t infect them. Different animals have slightly different ACE2 receptors on their cells, and the pig ACE2 receptor is quite similar to the human ACE2 receptor, suggesting there could be similar susceptibility in both species. Looking at ACE2 receptors has been interesting, but we’ve seen that it doesn’t always link up with what actually happens, as was the case in pigs.

Trying to grow the virus in cell lines from an animal species can provide some additional information about potential susceptibility. In one study, the SARS-CoV-2 virus was grown in 2/3 pig cell types, but did not damage those cells. In another study, the virus grew in the pig cells and caused some cell damage. Similar results were reported in another cell line study.

These all contributed to the concerns about the susceptibility of pigs; however, there are limitations to what in vitro studies can tell us. To get the real story, we need to look at actual pigs. So far, all the information we have about the virus is live pigs is from a couple of experimental studies.

So, are pigs susceptible to SARS-CoV-2?

In one study,  5 pigs were experimentally infected and mixed with 3 other pigs. In another study, 9 pigs were infected and then mixed with 3 other pigs. A third study infected 9 pigs and added 6 uninfected pigs.

  • Nothing remarkable happened in any of these studies. None of the pigs got sick, and all samples collected were negative for the virus. Antibodies against the virus weren’t found in any of the pigs. This all indicated that the pigs were not infected, and there was a big sigh of relief as it appeared that concerns about pigs were unnecessary.

In yet another study, pigs were exposed to the virus via the nose, the trachea and by injection. All the pigs stayed healthy and the virus wasn’t detected in any samples from the pigs, but antibodies against the virus were found in pigs that were injected with the virus. That shows the body responded to the virus, but since it was injected, it’s not really relevant to the natural situation.

However, leave it to Canadians to be disruptive – another experimental study in pigs changed the story a little bit.  It didn’t raise major concerns, but it suggested things are not quite as clear cut as we’d hoped.

  • In the Canadian study, 16 pigs were exposed to a higher dose of the virus, and nothing remarkable happened. Some developed mild discharge from the eyes for a few days. One had a slight cough and was mildly depressed for a few days.
  • Low levels of virus were detected by PCR in respiratory samples from two of the sixteen pigs, but live virus could not be isolated.
  • The virus was isolated from a lymph node of one pig, and antibodies were detected in the blood of two pigs, supporting some level of true infection.
  • Two pigs were added to the exposed pigs 10 days after inoculation, and they didn’t become infected.
  • So, this study showed some degree of susceptibility in pigs, but with infrequent mild disease and no evidence that pigs are infected to the degree that they would be able to pass on the virus to another animal (or person).

Another similar study involving inoculation of pigs with SARS-CoV-2 by different routes (blood, trachea, nose) also found none of the pigs got sick. Viral RNA was detected from oral, nasal or rectal swabs by PCR in some inoculated pigs, but virus wasn’t isolated and transmission to in-contact pigs wasn’t observed. These two studies are still consistent with a “don’t worry” narrative – if there was human-to-pig transmission of SARS-CoV-2, the pig would not likely get sick and would not likely be able infect other pigs or people.

Have any pigs outside of a lab been infected with SARS-CoV-2?

There are no reports of any naturally infected pigs, but I’m also not aware of any actual testing of pigs on farms. (“We don’t think there’s anything to investigate” is often stated with an unspoken “we don’t really want to know.”)

Field data are always useful because experimental studies don’t tell the full story. So, some data about pigs exposed to infected farmers would be useful to have, to round out the story. There have to have been large numbers of pigs exposed to infected people, especially on some large farms in areas where COVID-19 has run rampant. The fact that we haven’t heard rumblings of problems is good. However, without formal surveillance, it only tells us we don’t have evidence of a significant pig health issue. We can’t rule out the potential that pigs get infected but don’t get sick. That’s why we really should have more active surveillance, looking at pigs that have potentially been exposed.

What’s the recommendation when it comes to SARS-CoV-2 and pigs?

The same as for other animal species. If we keep infected people away from animals, we don’t need to worry about human-to-animal transmission, or any subsequent animal health or animal-to-human transmission issues. While the odds of someone infecting a pig are very low, it’s best to avoid exposing pigs to infected people whenever possible. That may not be an option on small farms run by one person or a family, but the more we can keep infected people away from animals (of all kinds), the better.

What about new variants of SARS-CoV-2 in pigs?

That’s the wild card for all our animal discussions. Experimental studies were done early in the pandemic and used the original strain of the virus. The SARS-CoV-2 strains we’re seeing now are quite different, at least in humans. Odds are low that delta or other variants would be much more able to infect pigs, but we simply don’t know. As we see new variants, we need to realize that what we know from earlier work isn’t necessarily still the case. It’s another reason ongoing surveillance would be good, but I won’t hold my breath on that.

There’s not as much to update about SARS-CoV-2 in dogs as there was in cats. We have more numbers than we did before, but the overall issues in dogs and our understanding of them haven’t really changed.

Spoiler alert: dogs and owners can both relax, as the risks are very limited.

Are dogs susceptible to the SARS-CoV-2 virus?

Yes, but… not very… maybe.  It depends on what you mean by “susceptible.”

Nice and clear, eh?

There’s a difference between getting infected and getting sick. Yes, dogs can clearly be infected. However, they don’t seem to be as susceptible as cats, and it’s debatable whether they get sick from the infection… I’m still a bit on the fence about that (more below).

How often do dogs get infected?

This is where we’ve gotten the most information recently.  Studies that have looked at dogs living with COVID-19-infected people have generally identified impressive rates of human-to-dog transmission. In these studies, researchers either look for evidence of the virus itself in the dogs (usually using a PCR test) or they look for antibodies against the virus in dogs’ blood. The problem with PCR testing is that there’s a very narrow window of virus shedding in this species, so it’s easy to miss the window (in which case the test comes back negative even though the dog was infected) based on sampling logistics and timing. With good antibody tests, we can get a good idea of whether dogs were infected in the past (although there are potential issues there too) because the antibodies hang around for a lot longer.

Early in the pandemic, the virus was identified by PCR in dogs from infected households, setting the scene for further studies. The best initial effort was from Hong Kong, where they identified the SARS-CoV-2 virus in nasal, oral and rectal swabs from 2/15 dogs that were quarantined because their owner was infected. Neither dog had signs of infection, both developed antibodies to the virus, and gene sequencing showed that the viruses in the dogs were the same as the viruses in their respective owners. Subsequent work has found similarly, fairly low rates of PCR-positivity among dogs with household exposure to infected people (e.g. Hamer 2021).

Serological studies looking for antibodies against SARS-CoV-2 in dogs have shown that transmission is actually much more common than this, with rates of up to 46% in dogs from infected households (e.g. Stevanovic 2021Hamer 2021). The results from our (hopefully soon to be submitted) Canadian study were similar, with about 43% of dogs with household exposure testing positive for antibodies to the virus.

Some studies have tested blood from undefined populations of dogs, for example by testing leftover samples of blood submitted by veterinarians to diagnostic labs. These studies tell us very little, because there’s no accompanying info about the dogs, particularly whether dogs were exposed to anyone with COVID-19. Not surprisingly, low rates of antibody detection (0.2-3.4%) have been found in these stuies (e.g. Ito et al. 2021,  Patterson 2020, Smith 2021). Whether these positives represent infected dogs from households with infected people vs false positive results isn’t clear.

Do dogs get sick when they are infected?

That’s still unclear. In small experimental studies, dogs could be infected but didn’t show any signs of disease (e.g. Shi et al. 2020, Bosco-Lauth et al. 2020). Field studies are harder to evaluate because there’s nothing specific about the clinical signs we’d expect to see from SARS-CoV-2 infection in a dog (e.g. lethargy, coughing, sneezing, decreased appetite and other flu-like signs). These non-specific signs can be caused by lots of diseases in dogs, so if we find a dog that has evidence of previous SARS-CoV-2infection (i.e. antibodies) and it was reported to have been sick, it’s hard to say whether it was sick because of SARS-CoV-2 or whether it was sick because of something else and had an incidental SARS-CoV-2 infection. Larger and better designed studies are needed to figure that out.

In our preliminary work, we found an association between antibodies in dogs and the owner reporting that the dog was acting sick around the same time the owner was infected. That is to say, this finding was significantly more common in dogs that had antibodies than in those that didn’t, suggesting infection with SARS-CoV-2 could have caused (or contributed)  to illness in (at least some of) the dogs. However, the signs that were reported were very mild and often vague (e.g. “the dog was a little quieter”), so while it might suggest that dogs can get sick, it would seem any illness is generally really mild.

Serious disease from SARS-CoV-2 has been identified in cats, and there has been some chatter about a very small number of dogs getting really sick or dying. However, when you consider the massive number of people that have been infected and the apparent high rate of transmission to dogs, if there was a true serious disease issue, I think we’d see more evidence of it by now. Also, we’re going to find incidental infection in some dogs with serious diseases or that die for other reasons, just because of the large number of infected dogs.

My take-home message on this at this point – with the variants that are currently circulating – is that SARS-CoV-2 poses very little health risk to dogs. More work on that is underway, though.

Can dogs infect other animals or people with SARS-CoV-2?

Probably not. Dogs are probably much lower risk than cats, and the even the risk from cats is still unclear. The fact that the virus has been isolated from dogs (e.g. Hamer 2021) is a concern, because if there was live virus in the dog’s nose, you have to assume there was some risk of exposure to individuals in-contact with that dog. Whether there was enough virus being shed to actually infect someone is completely unknown, and it’s probably exceptionally rare for a dog to be shedding enough virus to pose a risk. Experimentally, dog-to-dog transmission has not been seen. That’s not a guarantee that it can’t happen, since these experimental studies were conducted in an artificial environment with very small numbers of animals, but it provides more support of limited risk.

Overall, I’d say the risk of SARS-CoV-2 infection from dogs is very low. I don’t think we can say it’s zero (we can’t guarantee much with this virus), but I think it’s very unlikely that a dog would pose a realistic risk to a person or another animal.

That said, why chance it? If a dog is infected or at risk of being infected (e.g. living in a household with an infected person), it should be kept away from other people and dogs. Dogs interact nose-to-nose and nose-to-bum a lot, and have a lot of contact with their faces, so keeping exposed dogs under control and away from others is a reasonable precaution. We’ve also seen transmission of other respiratory viruses between neighbouring dogs through fence-line contact, so this should be avoided as well, just in case.

Could dogs be an important reservoir of the SARS-CoV-2 virus once it’s controlled in people?

No, dogs are not susceptible enough to the virus to serve as a reservoir. To be a reservoir, the virus would have to be able to keep spreading dog-to-dog. That’s not going to happen because of the low susceptibility and short shedding time in this species. You’d need a very large number of dogs in regular close contact to even begin to have a risk, and then only IF dogs were able to effectively transmit the virus.

What about variants of concern (VOCs) in dogs?

Variants of SARS-CoV-2 such as alpha have been reported in dogs (e.g. Barroso-Arevalo 2021 and Hamer 2021). That’s expected as different variants become dominant in people, because people are the source of infection in dogs. Unless a human variant has more or less affinity for dogs than the original strains (possible, but not very likely), we expect the strains infecting dogs to be a reflection of the strains infecting in humans. I assume that large numbers of dogs have been infected with the delta variant as it now dominates in people as well.

Could new variants of concern emerge in dogs?

Almost certainly not. Variants develop by chance during viral replication. The more transmission, the more replication, the greater the risk of a variant emerging through random mutation. Since dogs are not going to be involved in sustained transmission of the virus, there’s pretty much no chance we’d see a new variant emerge in dogs and spread back to people. Yes, it just takes one replication error and transmission event for a variant to emerge, but the odds of it happening from a human-to-dog transmission AND the dog then infecting a person are pretty much zero.

Could dogs be a bridge to infecting wildlife with SARS-CoV-2?

Probably not, or at least much less likely than cats. Their low susceptibility, short period of infection, limited (if any) infectivity to others, and limited direct contact with susceptible wildlife mean the odds of dogs being infected by their owners and then infecting wildlife are very low.

So, we shouldn’t worry about SARS-CoV-2 in dogs?

Worry, no.  But we still need to pay attention to it.

What should be done with dogs?

Do the same as for cats:

  • If you are infected, try to stay away from animals… all animals, human and otherwise.
  • If your dog has been exposed to a person with COVID-19, keep it inside and away from others.

The risks to and from dogs are exceptionally low, but precautions are common sense and easy… a few short term mild hassles for some peace of mind.

Ultimately, dogs are part of the family – so if your family is isolating, the dog should be included in that too.