A recent publication in the Journal of Veterinary Diagnostic Investigation (Haydock et al. 2022) describes an interesting but unfortunate case of tuberculosis in a dog. Published reports of rare cases like this are often of limited value, but sometimes they highlight important broader issues, and I think this one fits that category.

The patient was four-year-old mixed breed dog that was presented to the Ontario Veterinary College because she had some liver and lung masses, and fluid in her chest. She had a very extensive workup, including lots of bloodwork, radiographs, a CT scan, ultrasounds, bronchoalveolar lavage, and even exploratory surgery to get samples of the masses for testing. I was involved because an infectious cause seemed likely, but a specific cause wasn’t readily apparent despite a pretty vast array of infectious disease tests.

There’s another twist to this story based on the origin of the dog: she was living in Toronto, but had been adopted through a rescue 18-24 months earlier, originally having come from a remote community in northern Quebec. Whether it’s a dog from another country or from another region of Canada, we have to think about what different infectious diseases might be involved when we see dogs from other places. That can be a challenge since we often don’t have good data on disease risks in different areas. There were a few things we considered in this case, but nothing really fit.

As things progressed, it seemed there was a good chance the dog had a Mycobacterium species infection. Mycobacterium is a genus of bacteria that includes a lot of different species, including the causes of human (M. tuberculosis) and bovine (M. bovis) tuberculosis, along with a big group of environmental species that rarely cause disease. We divide Mycobacterium into two main groups, tuberculous mycobacteria (including M. tuberculosis and M. bovis) and non-tuberculous mycobacteria (NTM, which includes the important M. avium complex, or MAC).

At that point, I suspected the dog was infected with MAC, a fairly ubiquitous group of environmental mycobacteria that can sometimes cause severe disseminated infection.

  • I was partly correct, but that wasn’t much consolation when we got PCR results saying it indeed was mycobacterial – but it was M. tuberculosis, the cause of human TB.

I think the first thing I said in response to the result was “I suspect there’s an issue with the test. We’ve seen cross-reaction before in samples from that lab for that test. It’s probably a non-tuberculous Mycobacterium since that’s more common and the disease fits.”

  • But it turns out, that was not the case.

It was subsequently confirmed as M. tuberculosis by culture. Whole genome sequencing and MIRU-VNTU profiling showed it was a near exact match with a TB isolate from a person from Quebec, providing more support that the dog was infected before being moved to Ontario.

So, we had a diagnosis. Unfortunately, the dog deteriorated and was euthanized shortly before we got the TB result.

However, that led to a whole new issue: human exposure to TB from the dog. This dog had obviously had close contact with its owners for close to 2 years, and was cared for by numerous veterinary personnel at the referring veterinary clinic and at OVC, including some high-risk procedures (e.g. intubation for surgery, close contact in ICU). We don’t know a lot about dog-to-human transmission of TB, but there was certainly potential risk since the dog had lung lesions and could therefore have had viable M. tuberculosis in its respiratory secretions.

Public health units in both Guelph (where we are) and Toronto (where the dog lived) were informed, and coordinated contact tracing, using definitions for exposure we created for this situation (see table below). A lot of investigation and testing was required, but fortunately in the end there was no evidence anyone was infected by the dog.

TB has been reported in dogs before, but it originates from humans in these cases. Concern has been raised about importation of TB-infected dogs from areas where the disease is common in people, and this scenario both supports that risk and highlights how “importation” should really be thought of as “dog movement,” since there can be risks with dogs from other areas of the country, even if they haven’t crossed an international border.

Dogs don’t seem to be very good hosts for M. tuberculosis, but we don’t have great data about how often human-to-dog transmission occurs, in part because testing of dogs is a challenge. Typical human tests for TB (e.g. tuberculin skin test) do not work well in dogs, so there’s no quick and easy way to screen dogs who have been exposed to people with TB. Presumably, human-to-dog transmission occur sporadically but usually doesn’t result in severe disease in dogs. Transmission from dogs back to people is hard to evaluate, since most infected dogs come from infected households, so figuring out who infected who is a challenge. Scenarios like this, where an infected dog came into the household where there were no other known sources of exposure can help us figure these things out, but aren’t commonly encountered. The fact that no one got infected from this dog is encouraging, but it’s a pretty small sample size from which to draw conclusions.

This report doesn’t mean we need to think that every dog with strange disease has TB. This is a rare case. However, it should be taken as a reminder that strange things do happen, and that the origin of a dog needs to be considered when thinking about infectious diseases. We’re getting better at asking if dogs were imported. However, it’s not the act of changing countries that increases risk. It’s traveling between different risk areas, or between communities with different disease risks, even if they’re within Canada. We need to pay more attention to dog origin and dog movement (including people who take their dog on vacation), not just whether the dog was “imported.”

Image: Medical illustration of Mycobacterium tuberculosis (source: CDC Public Health Image Library 23254)