test

As H5N1 influenza continues to circulate in wild birds, and spillover into a wide range of other birds and mammals, it’s difficult to find the right messaging with regard to the risk of transmission to people.

  • Historically, H5N1 influenza has had a reported mortality rate of 30-50% in people, but it been a lot lower with the strains currently circulating in most parts of the world.
  • Human infections with H5N1 influenza continue to be really rare overall, but they do occur. Generally milder disease doesn’t mean always milder disease, and a small number of fatalities have been reported in the last few years.
  • Spillover of this virus into domestic animals (including poultry, cattle and cats in particular) continues to occur. These spillovers create more risk to people given the closer and more frequent human contact with domestic animals, but surveillance has been limited in household animals such as cats and dogs.

Communicating risks in situations like this is challenging, since we want to raise awareness so people recognize that there is some risk and a need to take basic precautions, but at the same time, we want to make sure there’s no panic and that the risks are dealt with in a reasonable manner. The sweet spot between awareness and paranoia is often tough to find with infectious diseases (as the current cruise ship hantavirus cluster also shows, but that’s another story).

A recent report in CDC’s Morbidity and Mortality Weekly Reports describes a case of suspected cat-to-human transmission of H5N1 influenza in California. It highlights some risks that we’ve been trying to message for a few years now.

  • The good news is the incident didn’t make the person sick (or at least not sick enough that it got reported. The bad news is that there is evidence that tranmission from the cat occurred at all.

The report describes follow-up of people in Los Angeles County who were exposed to cats that got H5N1 influenza from consuming contaminated raw milk or raw diets.

  • There were ultimately nine confirmed feline cases, and ten other suspect cases.
  • 139 people exposed to these cats were monitored for symptoms of flu; of those, 30 developed flu-like illness.
  • 33 people, including 18 of the people who had signs of upper respiratory tract infection, were tested for H5N1 flu by PCR and all were negative, while 36% tested positive for another respiratory virus (mainly human seasonal flu).

PCR testing mostly detects active infection, where the virus is still present. The challenge with this kind of surveillance is that by the time the cats were diagnosed, contacts identified and notified, and testing performed, the virus was likely to be long gone in people (if it was there at all).  Flu infections tend to have a short incubation period (from exposure to onset of illness), and people don’t shed the virus for very long. The median time from exposure to testing in these individuals was 8 days, which is stretching it for flu, no negative results are not surprising, but can’t entirely rule out transmission.

A complementary testing method is looking for antibodies in blood (serology). Antibodies take time to be produced, but they can stay in circulation for a fairly long period of time. So, someone who was infected a few weeks ago would likely be PCR-negative but antibody-positive. The presence of antibodies tells use there was exposure to the virus at some point, but not when. That’s a big limitation of serology in some situations, but in this case where we suspect the risk of exposure to H5N1 flu other than the known contact with the infected cats is very unlikely, finding antibodies in a person is a pretty solid indication that they were infected by the cat.

Twenty-five (25) of the individuals exposed to the infected cats were tested for H5N1 antibodies, and one was positive – a veterinarian, who had no other identified risk of exposure. They didn’t get sick, but they worked on an infected cat without using any respiratory or eye protection (high-risk contact). That makes it a pretty solid presumptive diagnosis of cat-to-human transmission of H5N1 influenza, even though the person was PCR-negative for H5N1 flu when tested 7 days after exposure (so likely missed the viral shedding period).

You might say “they didn’t get sick, so who cares?” At the individual level, that’s fair, but there are broader issues. This case shows that infected cats do pose some degree of risk to people (which we’ve suspected all along). While this person didn’t get sick, the next person might not be so lucky, based on the dose of virus to which they are exposed, their underlying health, and random quirks of disease.

While the H5N1 virus is still not well adapted to people, the more it’s transmitted between mammals, the greater the risk of it adapting to become better able to infect mammals (including people).

We’re also worried about someone with regular human seasonal flu getting exposed to H5N1 flu at the same time. That creates the potential for recombination, whereby the two flu viruses mix together in the same host, potentially creating a new virus strain with the worst parts of both the seasonal flu virus (i.e. easily transmitted between people) and H5N1 flu (a “new” virus to which there is limited individual and population immunity, and potentially could cause more severe illness).  That’s how pandemic viruses emerge. The odds of that happening are low, but the more the viruses mix in different species, the more that risk increases. Here, a reasonable number of people who were exposed to H5N1-infected cats had confirmed seasonal flu infections, so it’s far from a theoretical risk.

Ultimately this report doesn’t really change the current story much, but it’s documentation of something we had assumed would happen. It reinforces the need to take basic infection control precautions around sick cats that have potentially been exposed to H5N1 flu, to take steps to limit exposure of cats to this virus (like keeping them indoors when possible), and to continue surveillance in this and other species.

And as we saw with cat-to-human transmission of SARS-CoV-2 , veterinarians are at the forefront of this risk.

“Doing nothing often leads to the very best of something.” ~Winne the Pooh

I don’t think the beloved wise sage of a bear was thinking about urine when he said that, but we can nonetheless heed the guidance of Winnie the Pooh when it comes to the management of subclinical bacteriuria. Here’s why:

Subclinical bacteriuria (also called asymptomatic bacteriuria in people) is a condition in which bacteria are present in the urine without causing any disease. In the past, this has often been considered something that needs to be treated, but we now know that the bladder isn’t always a sterile environment, and having some bacteria in it isn’t necessarily bad. Bacteria cycle in an out of the bladder, and subclinical bacteriuria is a common (normal) state for some animals (and people). We also now know that it rarely needs to be treated. In humans, the main indications for treatment are pregnancy and treatment prior to undergoing a urological surgical procedure. Unfortunately, it’s often treated unnecessarily (in both people and animals), which can lead to issues with adverse effects of antimicrobials, antimicrobial resistance, unnecessary costs and unnecessary hassles and stress.

When the first edition of the ISCAID antimicrobial use guidelines for urinary tract disease in dogs and cats (2011) was released, many people pushed back at the recommendation to not treat (or even test to look for) subclinical bacteriuria. By the time the second edition of the ISCAID guidelines for diagnosis and management of bacterial urinary tract infections in dogs and cats (2019) was published, there was a lot less resistance to this recommendation, but unnecessary treatment of this condition is still relatively common. As we continue work on the newest update to these guidelines, we hope for even better uptake, but realize that old habits die hard.

Subclinical bacteriuria is still often treated because of habit or fear. We need to focus on a few concepts to move past this out-of-date practice:

  • We treat disease, not culture results.
  • Not all bugs need to die.
  • If a bacterium isn’t bothering my patient, it shouldn’t bother me (with rare exceptions).
  • Doing something isn’t always better than doing nothing (even though we’re hard wired to think we need to do something). Listen to Winnie!

A new study in the Journal of Veterinary Internal Medicine (Le Corre et al, 2026) provides more support for Pooh’s clinical sense. It’s not earth shattering, and it’s not surprising, but it’s a really important part of providing more data help convince clinicians to leave it be.  

The study is entitled Clinical outcomes and association with disease progression and survival of subclinical bacteriuria in cats with chronic kidney disease: a multicenter retrospective study. The researchers looked at 287 cats with chronic kidney disease, which is a population that’s at increased risk for subclinical bacteriuria. All the cats had urine cultures from samples collected by cystocentesis (i.e. using a needle and syringe to inserted directly into the bladder through the body wall to help reduce contamination of the sample), and none had any signs of lower urinary tract disease (e.g. straining to urinate).

  • Bacteria were isolated from the urine of 38% of cats, which is pretty much in line with prior estimates of subclinical bacteriuria in this population. Unsurprisingly, female cats were 5.3 times as likely to have bacteriuria compared to males.
  • Eschericia coli was the most common bacterial isolate, accounting for 68% of positive cultures, followed by Enterococcus (17%), Staphylococcus (5%) and Pseudomonas (5%). Twenty percent (20%) of bacteria were multi-drug resistant (ugh).
  • Eighty-five (85) of the cats with subclinical bacteriuria were treated with antibiotics, of which 41% were treated before antimicrobial susceptibility data were available.

In cats treated with antibiotics, bacteriuria was identified again later on in 62% of them.

  • That’s a key point. It demonstrates that treating the cats didn’t do anything to change their susceptibility to future episodes of bacteriuria, so it’s not surprising it happened again. This highlights the futility of treating this condition. If we eliminate the bacterium in the bladder today, odds are a new one will move in in the near future. The more we treat, the more likely we are to harm the cat (e.g. adverse events from the antimicrobial treatment), and the next time bacteriuria occurs, it may be resistant to our first line drug of choice. If the cat then develops an actual clinical infection that needs antimicrobials, we may have more limited treatment options.

When the researchers looked at 1500 day follow up in these cats, there was no association between subclinical bacteriuria and survival (see graph below). That fits with an earlier study of bacteriuria in senior cats, and with reams of data from human medicine.

Progression of chronic kidney disease was also assessed in 134 cats. There was no difference in percentage of cats that showed progression of disease in the group with subclinical bacteriuria (42%) versus controls (40%). The detection of multiple episodes of subclinical bacteriuria was also not associated with progression of kidney disease.

There was no statistically significant difference between groups in the development of bacterial cystitis, though there was a numerical difference that’s worthy of further investigation That said, an additional question would be whether cats that were treated were more likely to have other complicating factors that were a driver for cystitis. There was also no significant increased risk of pyelonephritis in untreated cats.

All this supports the notion that subclinical bacteriuria is a common and typically benign state. Treatment can sometimes eliminate bacteria, but not always, and when it does, recurrence of bacteriuria is common. It’s possible that treatment would reduce the subsequent risk of bacterial cystitis, but even if there’s a small effect, we need to consider the potential adverse events associated with treatment, especially when treating lots of cats over and over again.

This was the researchers’ conclusion:

…we found no significant association of SBU with survival and disease progression in cats with CKD. Despite antimicrobial treatment of SBU in cats with CKD, urinary sterilization was not achieved in most cases, and progression to bacterial cystitis or pyelonephritis, although infrequent remained possible. Despite the fact that no clear guidelines for the management of SBU in cats with CKD currently exist, our results do not support routine antimicrobial treatment of this condition. Further evidence on the appropriateness of withholding treatment to manage SBU in cats with CKD remain however to be demonstrated before such recommendations can be made.

Image source: https://tommccallum.com/2021/05/21/doing-nothing-often-leads-to-the-very-best-of-something/

test

Colorado State University, in partnership with the University of Bristol, is recruiting licensed veterinarians who have encountered at least one case of Feline Infectious Peritonitis (FIP) since 2019 to complete a brief (20 minute) online survey examining clinicians’ experiences and comfort levels with diagnosing FIP and using antiviral therapies for FIP in cats. Please participate if you meet this criterion!

Your participation is voluntary, and your responses will be kept confidential. Insights from this study will help inform the development of future veterinary education resources related to FIP diagnosis and treatment.

IRB#: 7613 (approval date 02/19/2026)

test

I’m on the way home from ESCMID Global, a clinical microbiology and infectious disease conference. Although the conference didn’t include much veterinary-specific content, it did include a good collection of abstracts about zoonotic diseases, including a couple about diseases in veterinarians, one of which described an infection with Brucella canis.

I’ve written about B. canis periodically on this blog. While it’s endemic in dogs in many parts of the world, it’s not well studied. Despite being quite common in some dog populations, disease can still be pretty rare. In Ontario, B. canis has been found in numerous commercial dog breeding operations (puppy mills) whose puppies are sold widely, but most of the clinical cases we see are in imported dogs (which makes me wonder if there are differences in virulence between the strains we have here and those that come from abroad).

From a human health standpoint, there’s sometimes a lot of (somewhat misplaced) concern because the Brucella species found in food animals cause a lot of disease in people in areas where they’re endemic. In contrast, B. canis in dogs often flies under the radar, because despite lots of infected dogs, human infections are rare. Last year we published a scoping review of B. canis cases in people (Weese & Weese, Can Vet J 2025), and we were only able to fine reports of 68 human infections. That’s certainly an underestimate of the true number of cases, since many affected people are likely not diagnosed (or misdiagnosed), and not all cases get published. However, when I talk to colleagues in the human health field about this disease, their response is usually “haven’t seen one” or “Brucella what-a?” In other words, Brucella from dogs really isn’t on their radars.

Nonetheless, it’s still important to keep the risk from B. canis in mind, because rare doesn’t mean never, especially for people handling high risk dogs (e.g. imported dogs, puppy mill dogs) and in high risk situations (e.g. assisting dogs with whelping and contact with the associated maternal or fetal fluids or tissues).

There was a nice abstract at ESCMID from the Royal Liverpool University Hospital about a British veterinarian who became infected with B. canis. This bug has gotten increased attention in the UK in recent years, particularly with regard to imported dogs, but it’s still probably not high on the list for most physicians even there.

  • The affected patient was a 36-year-old companion animal veterinarian who had regular contact with dogs of unknown origin and imported dogs. The lead author (Dr. Firas Maghrabi) said this person also had contact with breeding dogs, which further increases the risk of B. canis exposure.
  • Over the course of two weeks, the veterinarian developed forearm blisters, severe back pain, joint pain, and then fever, headache, chest pain, rash and conjunctivitis. After an extensive workup targeting a wide range of infectious diseases, she was ultimately diagnosed with B. canis infection through identification of the bacterium by PCR in blood and cerebrospinal fluid (CSF), and detection of antibodies against B. canis. She was treated with a combination of antibiotics for 6 months, and fortunately fully recovered in the end.

One of the biggest challenges with B. canis in people is that it’s usually not on the radar, and if you don’t test for it, you’re not likely to find it. The only way you’d find it without a targeted test is if the bacterium grew on a blood culture, which generally isn’t very sensitive for something like this. Specific testing for B. canis is needed, and it’s not widely available for people (and antibody tests for the food animal-associated Brucella species won’t detect B. canis).

Education of physicians is obviously needed, but it’s also a bit unrealistic to expect general practitioners to be read up on every oddball infectious disease that’s out there. More realistically, the focus needs to be on getting infectious disease physicians involved in these case, and ensuring those specialists have adequate awareness of B. canis.

The flip side, which is often overlooked, is patient awareness. While I’m sure physicians dread patients who come in with a “Dr. Google” diagnosis as much as veterinarians do, people can still advocate for themselves and raise issues that might otherwise be missed, particularly if they’re aware of high risk exposures that might not be queried in a typical history. People who handle imported dogs or dogs of unknown origin, particularly for any type of reproductive procedure, should be aware of B. canis. They can then play a role in getting it on the radar earlier in the diagnostic process if they’re ill.

“I work with imported dogs and assisted with a birthing last week. Is it possible I have Brucella canis?” is a clear and fair question to ask a healthcare provider. For many physicians, their answer may be “I have no idea,” but it’s the starting point to think about it, and consider whether testing or referral to a specialist is indicated.

Words matter. Inconsistency and inaccuracy with terminology can result in misinterpretation, poor communication and creates challenges when discussing cases, interpreting research and developing guidelines.

This has been particularly evident when it comes to urinary tract disease in dogs and cats. For example, “urinary tract infection (UTI)” is a very generic term that has been commonly used to describe everything from bacterial cystitis to subclinical bacteriuria, a positive urine culture, positive urine cytology, pyelonephritis and a few other conditions, or often even indistinct combinations of those. This creates a lot of challenges since those conditions can have vastly different clinical issues. When we’re not speaking the same language, or when we mix completely different diseases into the same bundle, we create the potential for error, misdirection and misunderstanding.

So as part of our ongoing revision of the International Society for Companion Animal Infectious Diseases (ISCAID) 2019 urinary treatment guidelines, we recently completed an initiative to develop consensus-based definitions for these conditions using a broad group of international participants.

  • The guideline field has advanced a lot in recent years. We’re getting past the time where you could just get a few smart people together to create a guideline based on their knowledge and experience. While those past guidelines were often quite good, and they were really important stepping stones, they lacked broad representation, involvement of stakeholders, consideration of biases and had other limitations. The same applies to developing definitions, which requires a more rigorous process to be done right.

We started off with a steering committee of 5 people from 4 countries, then had a review by the ISCAID guideline working group members (19 people), and then two broader rounds of input from 90 people from 19 countries. The goal was to engage a broad range of people with expertise, through multiple rounds of review and suggestions, to arrive at a consensus.

  • All the methodological details of how we came to our definitions are now published online (Weese et al 2026) for anyone that wants to see them. We tried to be as transparent as possible, explaining what was done each round, how and why decisions were made (including some terms/definitions that we abandoned). It wasn’t easy, but I think it yielded some solid results, with a lot of good discussion along the way.

The final definitions and associated footnote are in the images below. For more information and context (and maybe easier reading), check out the complete open-access paper.

Are these definitions perfect? Probably not.

Did we include everyone who might have had useful input? Definitely not.

That said, they’re a step along the way (in the right direction).

As we said in the paper’s conclusion: “Standardisation of terminology is a foundational component of clinical communication, clinical research, surveillance and guideline development. This initiative has allowed for the creation of terms and definitions based on broad stakeholder consensus. The study’s output should assist with ongoing and future efforts to improve the evidence base pertaining to infectious urinary tract disease and foster improved clinical guidance.”

As the weather (slowly and inconsistently) gets nicer here in Ontario, dogs start to go swimming more. My dog, Ozzie (pictured below), is an embarrassment to the Labrador breed as he will not swim, but he’ll happily wade through water, as long as his feet don’t leave the ground (but he still manages to get thoroughly soaked).

It is also the time of year when the question inevitably comes up (yet again): Do topical parasite preventatives need to be re-dosed more frequently in dogs that swim or get bathed over the summer?

No, they don’t. (An easy answer, for once!)

We have lots of options for parasite control in dogs and cats, including oral, topical and injectable products. There are lots of different pros and cons to each, and I won’t wade into that quagmire but here’s the scoop on the topicals:

  • Topical antiparasitics don’t “coat” the animal and stay on the skin or fur; topical is just the route of administration. These products are rapidly absorbed through the skin, and exert their effects through the body, not on the skin’s surface. A lot of of the absorption occurs within a couple of hours of being applied, and the drug can adhere to the skin and haircoat as the rest is absorbed, with peak serum drug levels occurring 2-7 days later. Once the drug is in the skin (even if it’s not yet in the bloodstream), it can’t be washed off.
  • A 2016 study (Kilp et al.) evaluated “repeated intensive shampooing” starting 3 days after administration of a topical parasite preventative, and there was no effect on the ability of the drug to kill fleas or ticks on the dogs. In another study, (Taenzler et al. 2016) researchers “water immersed’” dogs 3, 21, 49 and 77 days after treatment, shampooed them for 6-8 minutes or did nothing (control group). There was no apparent difference in the efficacy of the anti-parasitic treatment between the three groups.

I just checked the labels for a couple products available in Canada: One said to keep the dog out of water for 72 hours after treatment, and another said that exposure to water (including swimming) starting 60 minutes after treatment, or bathing 90 minutes after treatment, doesn’t impact efficacy.

  • I don’t know if the difference is because some products are absorbed quicker than others, some manufacturers are more conservative than others or because the products were simply studied / evaluated differently.
  • It’s important to pay attention to those product labels though. If anything, they’re probably quite conservative, so if they say that the dog can get wet X hours after treatment, I’d follow that guideline, but if the dog gets rained on earlier than that, it’s likely not a problem. If the dog goes for a swim immediately after the product is applied, I’d be concerned the dog would need to be retreated.

In the absence of specific guidance, it’s reasonable to avoid water immersion or bathing for 48-72 hours after treatment; 24 hours is probably lots in most cases, but 48-72 hours gives a bit more margin of safety.

So, while it can take a little bit of planning and attention to make sure you don’t waste a dose of your dog’s topical preventative (particularly for dogs that swim every day), they only need to stay out of the water for a few days, after that swimming or bathing won’t have any detrimental effect on the treatment.

If you’re a companion animal veterinarian practicing anywhere in Canada, please take 10 minutes to contribute to a University of Toronto–led study on antibiotic prescribing practices for common conditions like urinary tract and respiratory infections in dogs and cats!

Your anonymous responses to this survey (available in English and French) will help guide antimicrobial stewardship efforts and inform future veterinary practice across Canada.

English survey: https://www.surveymonkey.com/r/RM8SGR6

French survey: https://www.surveymonkey.com/r/PBFWW5M

The short answer: I don’t know, but probably not, and it could potentially do more harm than good.

I get asked a lot about splitting vaccines for pets, that is to say giving different vaccines at different visits instead of giving a bunch of vaccines all at the same time. The questions are often related to small breed dogs, which are more prone to adverse events following vaccination. Sometimes the questions are because a particular dog has had an adverse reaction to a vaccine in the past, and sometimes they’re because the owners are concerned about adverse reactions, even though their dog has never had one.

The theory behind the supposed benefits of splitting vaccines is that less antigenic stimulation would lead to a lower risk of adverse events. On one hand, that’s true. Data from a really large study of vaccination adverse events in dogs (Moore et al. 2023) showed that as you add more vaccines, the adverse event rate goes up (see graph below).

So, why do I say I don’t know if splitting vaccines helps avoid adverse events? Because we have to think about what happens to the dog over time, not just what happens after each vaccination visit. If we give only one vaccine today, we have a lower risk of an adverse reaction than if we gave two, but if the dog still needs the second vaccine, it will need to come back for again for another vaccination visit, that also comes with its own risk of an adverse event.

Looking at the numbers from the graph above, let’s approximate the risks for a small breed dog that is due for DA2PP (distemper, adenovirus 2, parvo, parainfluenza, which are given together as a single injection) and rabies vaccines.

  • That’s two vaccines. Based on the crude estimate from the graph, the risk of an adverse event would be roughly 27 in 10,000 if they are given at the same visit.
  • If we only give DA2PP vaccine, the risk would be lower, at around 21 in 10,000, but the dog will still need to get a rabies vaccine.
  • When the dog comes in for its second visit and we only give the rabies vaccine, the risk of an adverse event is once again 21 in 10,000.

If we look at the risk for a single vaccine visit, the risk is lower if we only give one vaccine (21 vs 27 in 10,000). However, if we look at the cumulative risk to this dog to receive both vaccines on separate days, (21+21= 42 in 10,000) it’s actually higher than if we’d given them both at once.

How solid are these numbers? It’s hard to say. The data are crude, but they are the best we have. It makes sense, though. Splitting vaccines would have to drop the risk by at least 50% to achieve a net benefit when we have to add extra vaccination events to ensure the dog gets all the vaccines it needs. This also doesn’t consider the added stress, hassle and normal general malaise the pet can get after vaccination, which happens twice if the vaccines are split. I got my flu and COVID-19 vaccines at the same time this year. I did that on purpose because I’d rather have one episode of having a sore arm and maybe feeling crappy than two episodes. The same presumably applies to our canine patients.

It’s not that there are no valid reasons for splitting vaccines. It’s possible that some dogs that are particularly reactive to vaccines would benefit more from this strategy. We just don’t know.

The main benefit of splitting vaccines is to understand which vaccine may be causing a reaction in the dog if they have one. If I give DA2PP one day and the dog is fine, then follow up with rabies vaccine and the dog has a reaction, that’s useful because it suggests we have to focus on issues with the rabies vaccination. However, even that isn’t a guarantee of the cause, as many vaccine reactions are just random, non-repeatable events that don’t indicate long term risk.

So, split away if you want, but realize that’s it’s a bit more complicated than it might seem at first glance. It might not be reducing the risk to the dog, and it may actually be increasing it.

When it comes to disease surveillance and communication, we have a tendency to throw lots of stuff at the wall to see what sticks. Some things stick around (like this blog!), others things not so much… like our first attempt at WormsAndGermsMap about ten years ago. It was a good idea, but the technology wasn’t ready for it yet (at least for a low budget operation like ours). But technology advances, diseases continue to spread, and informal disease mapping can still be useful, so we’re giving it another try with the launch of our new and improved WormsAndGermsMap.

The new site lets users report disease events (cases) directly using a very simple submission page (see screenshot). The cases are then plotted on the map, randomly within a 5 km radius of the location provided (to protect privacy). The exact locations are retained on the back end of the site it case they’re needed, for example for looking more closely at a potential local outbreak.

The map can then be filtered by animal species, disease and dates. Clicking on a point provides some basic information on the case, but we also keep a bit more information on the back end of the site, and submitters can add comments in case there’s additional relevant information we should have.

There aren’t a lot of cases on the site yet (as you can see from the screenshot above) as we’re focusing on ongoing crowd reporting, but I’m also adding some data points we have that I think will be useful.

Why are we trying this again?

Because I get emails every week along the lines of “I think I’m seeing more of this disease.” I also get lots of questions about what diseases are where. I also get questions like “we saw something unusual. To whom can we report it?”

Informal, crowd-based disease reporting and mapping can help with all of these issues, but it’s important to remember that it also has lots of limitations. It’s based on what people happen to report, so it doesn’t capture all infections, or even a structured subset collected as part of a formal surveillance program. It also relies on the accuracy of the diagnosis, since we aren’t assessing the case ourselves to confirm the disease is what the submitter says it is. We also need to keep in mind that a disease that develops / is diagnosed in one area might have actually been picked up in an entirely different region (because animals travel too).

But the limitations also shouldn’t overshadow the potential educational and surveillance value. We’re not trying to say “the incidence of this disease is X in this area” or “there’s been an increase in disease in that area.” We’re just trying to help raise awareness about regional diseases risks, identify and track high risk incidents (e.g. canine influenza outbreaks), and give people a way to report unusual observations that might warrant more investigation or communication. Most of the time, an unusual observation is a typical disease acting atypically. Somethings, though, it’s the sign of something new. We can also sometimes pick up emerging issues or local clusters of disease. Ultimately, we have to interpret the data with caution, but the data can still be useful.

We’ll see if WormsAndGermsMap sticks better this time. I think there’s a need, but it’s only useful if we can get people to use it and submit cases. Time will tell.

test

I haven’t written about raw diets lately (beyond risks from H5N1 influenza, particularly in cats), but that doesn’t mean the risks from these diets have gone away, and they are still really popular in some areas, and some messages are worth repeating.

Potential problems with raw diets for dogs and cats include infections (e.g. Salmonella, E. coli O157) in animals that eat the food or in people who handle the food or animals, nutritional imbalances (particularly problematic in growing animals), and intestinal foreign bodies (e.g. from diets containing pieces of bone). As usual, I’ll focus on the infectious disease aspects since that’s my turf.

A recent surveillance study from the UK looked at bacterial contamination of commercial raw diets for dogs and cats. The results were predictably disappointing. They tested 380 raw pet food samples (277 canine diets and 103 feline diets), and found one or more pathogenic bacteria in 35% of samples:

  • Salmonella: 21%
  • Multidrug-resistant Salmonella: 9%
  • Campylobacter: 14% (and 21% of C. jejuni were multidrug-resistant)  
  • Shiga-toxigenic E. coli (a bad one): 12%
  • Antimicrobial-resistant E. coli: 20%
  • Colistin (an antimicrobial of last resort) resistant E. coli: 1%
  • Methicillin resistant Staphylococcus aureus (MRSA): 10%

Overall, 29% of the diets were not compliant with what is allowed to be sold in the UK because of the presence of Salmonella or the level of E. coli present (over 5000 CFU/gram). Those are pretty loose limits too, as I’d be worried about Campylobacter or any level of colistin-resistant E. coli… Regardless, a large percentage of the diets were of concern based on potential human and/or animal health risks.

Another interesting finding was that 8% of samples leaked through the packaging while being defrosted. In a household, that means potential contamination of the fridge or countertops, creating more human exposure risk.

None of that is really surprising. It’s consistent with what we’ve known for years about contamination of these diets with things like Salmonella and Campylobacter, and adds to more recent evidence that raw diets are a major source of multidrug resistant E. coli. Multiple studies have now shown that eating a raw diet is a big risk factor for dogs to be shedding multidrug resistant E. coli, with potential for disease in the animal or transmission to people.

Would I like people to stop feeding raw diets to their pets? Yes.
Do I think for a second that they’ll actually all stop feeding raw diets to their pets? No.

So my focus is on situations of greatest concern, where there are high risk people or animals in the household, including the very young, elderly, pregnant or immunosuppressed. I’d also like people who are adamant about feeding raw diets to move toward using high pressure pasteurized (HPP) raw diets. The HPP process doesn’t eliminate the risk from these diets, but should decrease it considerably.

And, as always, a little hygiene (and some common sense!) goes a long way to reduce the risk of contamination and transmission of / via kitchen surfaces, food bowls and the pets themselves. More information about risk reduction with regard to raw diets can be found on the Worms & Germs Resources – Pets page.