Blasto - PHIL19914I had an interesting question the other day about blastomycosis screening.

Blastomycosis is a nasty fungal disease caused by Blastomyces dermatitidis. The infectious form of this fungus (spores) lives in the environment, and dogs (and other species, including people) can develop serious disease after inhaling fungal spores (or being otherwise inoculated with spores, e.g. through a wound).

Fortunately, blasto isn’t a concern in most regions. In Ontario, there are two main problem areas. One is around Georgian Bay, not too far from here. The bigger hotbed, though, is farther north around Kenora. Blasto rates in dogs (and people) are quite high there.  It’s still a rare disease, but they have very high rates compared to most other places in North America.

The question I received relates to a family that visits Kenora, Ontario for a few weeks every year.  It’s not an uncommon scenario for people that live in cities to vacation in areas where blasto is endemic. They previously had a dog diagnosed with a serious case of blastomycosis, and the question was essentially “should we be testing our other dog every year after coming home from Kenora?” There is a pretty good urine test that can be used to check for infection with blasto, so it’s a fairly easy test to do.

It’s a good question. In general, I’m not a fan of screening healthy animals. No test is 100% accurate, and most diagnostic tests are designed to be used on animals with signs of disease.

Here’s where a basic epidemiology term comes in: specificity.

  • Basically, the specificity of a test tells us how often we get false positive results (a positive result in an individual that does not have the disease). If the test is 95% specific, 1 in 20 positive tests will be a false positive (95% will be right, 5% will be wrong).

Another epidemiological term also comes into play: positive predictive value.

  • This is the likelihood that a positive test means that the individual really has the disease.

Specificity and the likelihood of disease in a particular animal are important for determining the positive predictive value. When the test is done on an individual that is healthy (and therefore most likely does not have disease), the likelihood that a positive result is actually false can be high. Conversely, if the test is very sensitive (meaning it is highly likely to detect any positive cases), a negative result in such an individual has a very high likelihood of being real (a high negative predictive value).

We need to consider a few things when it comes to screening:

  • How specific is the test?
  • How low would the positive predictive value be in a healthy animal?
  • What would the response be to a positive result?

The last one is key. One of the reasons that I’m not generally a fan of screening is that it can lead to bad decisions. A clear and reasonable plan for acting on the result in needed, and that plan must differ if the result is positive versus negative.

So can testing this dog for blasto be justified?

Probably, as long as there’s a plan for what to do based on the test result and everyone knows the limitations.

If the result is negative…

  • That’s great. That’s the expected and hoped-for result. If the dog is healthy and the urine test is negative, it is very unlikely this dog has blastomycosis.

If the result is positive…

  • That’s not great. That doesn’t mean the dog has blasto, but it means that more work is needed. To me, that would involve a thorough physical examination and a chest radiograph to look for the characteristic lung lesions. If those were normal, odds are it is a false positive test but we still couldn’t rule out early disease. I’d probably recommend re-doing the urine test to see if the positive result was consistent and, perhaps more importantly, whether the value increased over time. If the latter, I’d be concerned that the dog was developing disease.
  • I would not take a positive urine test and embark directly on a prolonged and expensive course of treatment in a clinically normal animal.

So, I think screening can be justified if everyone agrees on what would happen with the two possible outcomes. If that is the case, the main downside is cost, which may not be an impediment, depending on the owner.

Image: Blastomyces dermatitidis growing on a petri dish (CDC Public Health Image Library 19914)