Valley fever testing: Right test, right time, right patient

(upbeat electronic music) - This is Lab Medicine Rounds, a curated podcast for physicians, laboratory professionals, and students. I'm your host, Justin Kreuter, the Bow Tie Bandit, a blood, a transfusion medicine pathologist at Mayo Clinic. As the weather turns cold here in Minnesota, we thought it'd be fun to travel south and connect with some of our colleagues from our other Mayo Clinic campuses over the winter months.

And today we're rounding with Dr. Grys, Associate Professor of Laboratory Medicine and Pathology at Mayo Clinic and Co-Director of Microbiology at Mayo Clinic, Arizona. And we're gonna be talking about valley fever testing and the importance of clinical context. Thank you for joining us today, Dr. Grys. - [Thomas] Of course. - I definitely wanna give you a shout out and appreciate the bow tie you've got on for everybody today.

So for those of you that are watching on YouTube, you can see the wonderful design that Dr. Grys has got on his bow tie there. - Yep, I've got a collection of bow ties that are mostly microbial or chemical structures, things like that. My wife finds them on Etsy. So I don't always wear a bow tie, but when I do it it's a nerdy one. - I love it. So let's kick into this podcast and can you get us started

with why does clinical context matter for what test clinicians order, and to remind everybody, we're talking about valley fever testing, and so, can ya kinda launch in with a little bit of the why here? - Well, I'd like to back up first and just remind ourselves that you know, despite our best efforts, not all of our lab tests are perfect. There's always some level of imprecision or inaccuracy that we can expect.

And part of validating a test is characterizing that so that we can understand (clears throat) what those weaknesses might be. And part of what makes a test good or bad is the likelihood that the patient has that condition or that analyte present when we're doing the test. Now some tests and like in chemistry, if you don't have potassium in your blood you're gonna be laying on the ground.

You can expect to find some, so they're measuring how much of something they expect to be there, that's one type of question. A fundamentally different question is, is something there or not? And that's a lot of what we face in microbiology is, is that thing really there or not, and that's a different question to set up your test for and to set quality around and then ultimately should affect when that test is being ordered.

(clears throat) So one of the things I talk about with the medical students is pre-test probability and the positive predictive value, negative predictive value of the tests. And the two good examples are both respiratory diseases, influenza, which varies by time of year and things like valley fever, which differs by geography. So if you don't have the exposure or if it's not the right season, we can expect that result to be negative because it's just not there, right?

So testing for influenza in July generally is a bad idea because it's not circulating. We know it's not out there, so any test that's positive is likely a false positive because the prevalence, and we won't go through the math, unless you're really want to. But we'll just remind our readers that the math exists, that the positive and negative predictive value is dependent on the prevalence of that disease in the population.

So during the summer months in the Northern Hemisphere, when influenza's not circulating, the positive predictive value is very low, even with a nucleic acid test. And likewise, when flu is prevalent, a negative predictive value of a negative test is quite high. So if you get a negative test when everyone around you has flu, it's probably negative. You probably have one of the other many viruses. Now in Arizona, 10 to 30% of community-acquired pneumonia is valley fever, 10 to 30%.

So that's pretty high. We have the highest prevalence of valley fever in the world right here in this county. A part of that has impacted and however many million people in the Phoenix area and we're all breathing the air. And valley fever is caused by the fungal spores of coccidioides immitis or posadasii. So we think it only takes a few Arthroconidia, the spores, to cause disease, but to have disease, you have to have exposure to the disease.

So if you've been born and raised in another part of the country, likely you don't have valley fever 'cause you haven't been here. Now, there are anecdotes of sunbirds, grandparents returning to the Midwest or the East Coast with dusty suitcases and the grandkids playing in them and being exposed or truckers driving through the valley and having their window open and breathing the air actually getting disease. If you have lungs, you can get valley fever.

So I mean, elephants, chimpanzees in the zoo, dolphins can get valley fever. I mean, any mammal that breathes is at risk. So when I'm giving a talk in Arizona here, I say, "Raise your hand, "I've noticed you're all breathing. "You're all at risk." And it's true. It's a bit of a stretch, but it is true. It doesn't take much if you breathe in the wrong bit of air and you get some spores, you could come down with the disease.

Now, most people, probably two thirds don't, there is some dramatic, they get over it just fine. But for those that do, they have a lot of non-specific symptoms, fever, cough, fatigue. Fatigue can last for months. Night sweats, various forms of rash. We often say erythema nodosum, but it can manifest in all sorts of ways. So all these things mimic influenza-like illnesses, or right now, West Nile.

We're having the biggest year on record here with West Nile, almost a thousand cases here and dozens of deaths. So it's not specific, but it's common. So we have to remember to test. Now, if you were testing someone with these symptoms in North Carolina, even Minnesota, likely they have something else. So a positive serology is likely something else. And in fact, you have your own dimorphs up there, dimorphic fungi.

So these are fungi that exist as mycelia or hyphal forms in soil in the environment and in the lung. A general, we call it a yeast form. Sometimes historically they call it parasitic form 'cause they didn't know it was fungus. So it looks different, so different morphology dimorphic. We often characterize Histoplasma and Blastomyces as thermal dimorphs 'cause you can take a plate at 25 and move it to 37 and you get yeast.

If you do that with coccidioides, you get warmer yeast, sorry, warmer mycelia. And it kind of makes sense 'cause it's often well above body temperature outside. So it's not gonna automatically turn to its yeast form in the environment unless there's other cues to do that. So that's one of the things on the research side that we're interested in and others in the field are working on because if we can stop that transition, maybe we can help prevent disease.

- You know, I really love how you've started us off with this example in showing about how really considering examples of time of year or geography and reminding us that a lot of the testing in microbiology are kinda plus/minus rather than a degree of measurement. And I'm curious, so that kind of plus/minus, that might be a physician might think about, is this on my differential? You brought up the idea of valley fever consisting of about 10 to 30% of pneumonia, I think you had said.

And so, if somebody is considering that because the geography and time and stuff makes sense,

what then determines which kind of tests to order? - Yeah, that's a great question because despite being common, it's actually not the easiest thing to diagnose. It's a dry cough, typically, so oftentimes we don't have sputum to work with. If we have sputum, we can do culture. It will grow readily, even on bacterial media. So if we start to see some gray fuzzies on the bacterial side, we have to move those to the fungal area. We can do PCR, which is about the same sensitivity as culture.

But if we don't have someone coughing up sputum, then we're really reliant on serology, and there's a number of serology tests, all of which are quite mediocre. And so, if you are considering coccidioides, you have to test for serology, not once, but probably several times because it can take, you can get antibodies within two weeks or so, as we often think about, but some patients won't make them for two months, three months until they have a positive serologic response.

So that makes it really difficult. Meanwhile, they're getting multiple courses of antibiotics sometimes, and it's not that everybody needs treatment. Many healthy, normal people will recover eventually. And for a lotta these people, giving them something like Fluconazole won't shorten their course of disease. So begs the question, why do we need to know?

Well, if we cannot give them antibiotics, which not only increases risk of community resistance of microorganisms, but also there are direct effects from some of these antibiotics that can be a bad outcome. So for instance, 4-Quinolones can give people tendonitis. People have blown Achilles heels and other things like that after a course of antibiotics. So these are not benign drugs.

So if we can withhold the drugs they don't need and stop doing additional testing, save additional visits to the doctor, then we can really have the provider and the patient allow them that conversation of what to expect, what are the warning signs? And then it's just a matter of dealing with the disease over the next, often, few months.

So if you're negative, sometimes we need to test again in a couple of weeks and it's frustrating 'cause if your doctor doesn't think of it right away, you often go down the road of assumed viral or bacterial pneumonia. If they do think of it right away, it can be negative. So our ID team here has a CME course in usually in January that it's a ID update for primary care physicians.

And they spend a chunk of time talking about valley fever, especially for those in the local community who may have trained elsewhere. They need to know about how this disease works. And if you get an IgG positive, that doesn't mean you're immune and you had it passed. It probably means you had a recent infection because unlike a lot of infectious disease serology, over time, that positive serology will wane if you have cleared the disease.

So if you have a persistent IgG, it's likely that you still have the fungus in your body, in these granulomas and things, which then can be cause for concern if the patients gonna undergo organ transplant or cancer therapy or be treated with a TNF inhibitor or something like that. Cocci can reactivate and cause disease later in life, regardless of the area of the country they live in at that time.

So, you know, serology is one of the oldest, most mediocre and best tools we still have, unfortunately. - (laughs) I love that you describe it as mediocre and best tools 'cause I think you took us through those examples, right, where it really requires in one case you gave us of where you're really going to have a battery of this test, repeat it over time as opposed to a different test. And in other contexts you might say, that's not necessary or going overboard.

And I love the fact that you brought in the CME course, 'cause then that highlights, you know?

I'm curious about how that communication happens between the lab and the clinician or maybe how should that happen in order to provide best care for the patient. Keeping in mind here that we've got, our audience is mixture of laboratory professionals, clinicians, and students.

- Right, so anytime we validated a test in the lab, part of that, like I mentioned earlier is understanding its strong points and weak points and communicating that, when you go live at the test, it's good to share those things so that the providers know when it's useful and not useful to use that test. And then, it's things like Grand Rounds or CASE conferences where you can just keep reminding people that yeah, we can expect the test with negative. It had only been two weeks.

No one did anything wrong, but then we do need to remember to test again if it's still in the differential. So it's just a matter of being persistent and engaged in the practice because, I don't know, hear something and it sticks. Like we have to keep reminding each other and our providers across all the specialties, they have a lot of their own details within their specialty to keep up on.

So it's really our job to help them, support them in their endeavors and remind them of when our lab tests are good and not good. And sometimes as a lab director, our job is to remind them of the situations where our lab tests are not so helpful because I tell the medical students too, usually, the single best source of information is sitting in the exam room with you. It's a matter of answer asking the right questions.

We had a case of HLH, which you can pronounce that 'cause I always mess it up. Histiophacytic... Lympho, yeah, HLH, Google it, for pathology. but it's the condition that can be precipitated by a number of infections. And so, we had a patient from a Pacific Northwest who had gone to Mexico on vacation and then showed up in our ED and had HLH. And we're trying to figure out what was happening. A number of consultants that had gone through and interviewed the patient, offered their recommendations.

Finally, I think it was like the third team of ID, fellow in an attending that came through. Someone asked, "Well, what did you do in Mexico?" "Oh, I went spelunking." (smacks legs) We're done. It's histoplasma. We don't even need to do any more testing. Spelunking and a histoplasma is such a tight correlation. And then with his HLH diagnosis, it's histoplasma until proven otherwise. I mean, we're done. One question, she'd been here, inpatient, for like two weeks and it was free.

So I think, again, sometimes we have to remind our providers, it is not the lab test. Sometimes, it often is, but sometimes it's not the lab test that answers the question. We help rule things in or out, but they still have to do their job. We are not yet to AI where you scan the drop of blood and do imaging and you get your answer. That can help you, but you still have to do the art of medicine.

And especially in infectious disease and microbiology, it's asking those questions about exposure, history. Do you live with a dog? You know, we had a BMT, a bone marrow transplant patient who had salmonella and serotype Arizonae, which is highly associated with reptiles. It turns out before she went in for one of her treatments, kissed her Gila monster pet. Don't kiss reptiles if you're immunocompromised.

So these sorts of exposures and things are often, they're not always as textbook as you read, but a lot of them are. And so, it's just a matter of doing that due diligence of what are the exposures, what are the symptoms? And lab tests will help confirm or exclude those differential diagnoses. But it is our job to support that art of medicine.

- Yeah, I'm really glad that you highlighted that I think for specialists that have cultivated a relationship with the microbiology lab in this case, for example, right? That connection is a little bit more direct.

But what I hear in your answer too, is that sometimes it's the physician call it picking up the phone and calling you or the laboratory and also vice versa, sometimes when we're getting tests that maybe don't make sense or that we're gonna pick up the phone and call and have that conversation. I'm curious, what do you think kinda for the future? You said a little bit, maybe a hint of what your answer here might be with AI,

but what do you think about for the future of ordering laboratory tests? - With the electronic health records, I feel like we're halfway there. Computers are here to help us or so we thought. It seems like sometimes they make more work, but when we offer a panel, or even a selection of options, it can facilitate bad practice of bad utilization, not thinking, but it can also help us remember not to forget something that is likely.

So to me, that's the balance, and it can be difficult to come up with these in a way that's gonna work for most patients. And I think each practice needs to kinda come to their own balance point of what's a good prompt and reminder, and what's facilitating bad practice because we have physicians who think that all information is good, and they just wanna keep ordering tests to keep ruling in or out things.

But if the likelihood is low and then you get a result that you weren't expecting, you paint yourself into a corner. What are you gonna do with that result? 'Cause another whole episode you should pursue is discharged testing, right? So there's a big push made a few years ago with the Mayo to decrease the amount of discharge testing 'cause a lotta people don't ever look at those again.

They're meant to be like a safety catch and just one last look, but let's say one of those things is out of range and you order CBC and a bunch of things where you're getting multiple values for all those tests, you might have 50 or 100 different numbers come out of that. And statistically speaking, given that we're all human, we're all different, something's gonna be a little bit out of range. What's your plan? Who's gonna document that? Is it gonna be worth following up on?

I mean, if they're clinically well, can we be good with that? Because a lot of times ordering more tests at that point is only gonna create problems and create costs. So if we're not prepared to take action and change therapy based on those results, then let's think about whether we really need to order that test. And so, the same is true here.

If it's plausible that someone was in Arizona 10 years ago for a week, and now they have pneumonia and they're in Ohio, yeah, they could have Coccidioides, but if they were recently cleaning out their farm, it's probably a number of other things. And you might end up testing for Cocci as a third tier, last ditch effort, if nothing else was positive, but it should not be your first go-round because if it's positive, then you're gonna look at that. Ah, it's probably not Coccidioides.

Well, you just wasted patient money and our time doing that test. But that's the hard part, it's really zeroing in on, what's first tier? What should we leave until we get the first set of results back? Because every physician is looking at the patient, saying, "Boy, I feel badly they're in this state. "I wanna help them." And so, I'd like to give them answers sooner. Then we gotta balance that with healthcare costs and what's ultimately gonna be useful or not useful in that situation.

- Wow, that's the hard part. (laughs) Thank you so much, Dr. Grys, for this.

We've been rounding with Dr. Grys on valley fever testing and the importance of clinical context. I think that you've really given our listeners a really nuanced perspective, and I love that you're hitting on the hard part of clinical medicine. And I love that we're kinda tackling these issues here and I think you've just pegged yourself for a followup episode about discharge labs. (laughs) - [Thomas] Yeah. (upbeat music) - So to all of our listeners, thank you for joining us today.

We invite you to share your thoughts and suggestions via email. Please direct any suggestions to [email protected]. If you have enjoyed Lab Medicine Rounds podcast, please follow or subscribe. Until our next rounds together, we encourage you to continue to connect lab medicine and the clinical practice through insightful conversations. (upbeat music)