November 8, 2017 Total Running Time: 23:39 (Original 26:47)
-- Soundtrack “Upbeat Happy Indie Pop Track” courtesy of AudioBlocks
Text Intro to Accompany Podcast:
In the 17th installment of the Open Forum: Infectious Diseases (OFID) podcast, Editor-in-Chief Paul Sax, MD, faces off with Rebeca Plank, MD, MPH, once again—this time to draft the best team of diagnostic tests in infectious diseases. The picks are inventive, ranging from blood cultures to HIV antibody tests. The pair also delves into their least favorite tests and the new diagnostic capabilities they anticipate on the horizon.
Dr. Plank is an infectious disease specialist with a particular interest in HIV/AIDS and HCV. She is employed at Merck Research Laboratories, where she works on late-stage clinical trials, and is a part-time associate physician at Brigham and Women’s Hospital and part-time assistant professor of medicine at Harvard Medical School.
Podcast Transcript to Accompany Podcast:
Hi, this is Paul Sax. I’m the Editor-in-Chief of Open Forum Infectious Diseases—just a reminder, that’s O-F-I-D, not “Ofid.” And I’m delighted to say that joining me today on this podcast is my friend, my colleague, and proud resident of Jamaica Plain, MA, Dr. Rebeca Plank.
Rebeca, welcome.
Thanks so much for having me back, Paul.
Well, you know, Rebeca, the last time you were on this podcast, we did a draft. We chose our favorite antibiotics. Any thoughts about how that turned out?
You know, I think it was well received by clinicians and some even non-clinicians, and I got a lot of feedback supporting Doxycycline.
Yes, well, Doxycycline was an excellent choice. I mean, you went first last time, and I’m not sure that was fair, but I agree; Doxycycline was a great choice.
Well, today, we’re going to do something different—we’re going to pick “Diagnostic Tests in Infectious Diseases.” We’re each going to choose our top 3 favorites, followed by our least favorite. Then something aspirational, something not available now but we wish were. And importantly, we have not shared our picks with each other ahead of time, so sky is the limit.
Yeah, and I think, to make it fair, since I got to choose first last time, you should choose first this time.
That is very, very kind of you. And before we start, I just want to thank our sponsor.
Who is?
Actually, we don’t have a sponsor, so I don’t need to thank our sponsor. But if there’s anyone out there who wants to sponsor this podcast, just reach out to our colleagues at Oxford University Press. So I’m going to choose first. And I think people may have pegged me for choosing this, but I don’t care. I’m going to choose the HIV antibody test, or at least the diagnostic test for HIV. And let me tell you why. This test has incredibly high stakes, always has. Finding out whether you’re HIV positive or negative is so important. And imagine if this test were not accurate, but it is sensationally accurate. It has, right from the beginning, had a diagnostic accuracy of 99 percent or greater, and now it’s even more accurate. And the whole process by which we do a two-step verification has turned out to be extraordinarily accurate, so much so that it’s really, really rare to have a false positive. And also, increasingly, it’s really, really rare to have a false negative. So, I’m going to go with diagnostic test for HIV. What do you think, Rebeca?
Well, that’s on my list. I think that it’s good for two related reasons. One of them is that people who have HIV need proactive management, because it puts them at risk for other infections. And the other thing is that sometimes people are acutely ill with HIV. And even though this isn’t the ideal test for hyper-acute illness, it can pick up, as you said, illness in relatively earlier stages now with the antigen being included in the test.
That’s right. The 4th-generation test now picks up people with acute infection because there is the antigen component. Even so, if you clinically suspect the acute HIV, you should actually send an HIV RNA or a viral load test. And I’m kind of including that in the package. Is that allowed? I mean, I’m including all diagnostic tests for HIV.
I’ll allow it.
Okay, thank you. I want to comment again that it’s really amazing to contemplate, from a policy perspective, that we would like everyone tested for HIV, even people who don’t have any risk factors, because the test is so darn accurate that we’re not going to mislabel lots of people as being positive who really aren’t, and that’s unusual in a diagnostic test. So, that’s my choice number one—diagnostic test for HIV.
So, that was my number two choice. My number one choice is blood cultures. I think that this is the real mainstay of infectious disease workup. And sometimes we’re trying to decide what this person has, and we draw blood cultures looking for that, but sometimes we know what they have, we’re just not sure what’s causing it. So, for example, someone has a vegetation on their valve, you have to get blood cultures to determine how you’re going to best treat them. And sometimes they’re febrile—for example, fever in a returning traveler, it’s important to get blood cultures. You don’t actually know what you’re going to find.
I’d say there’s barely any ID consult in a febrile patient that we don’t recommend blood cultures. In fact, you could almost make a rubber stamp—if we still had paper charts—that said, “blood cultures times two.” You want to comment a little bit about some of the limitations of blood cultures?
Well, there’s always the possibility of contaminants, and I’ll say that a few things we never consider to be contaminants, like Candida, or Staph aureus or Pseudomonas. I think that the grayer zone is when someone has an indwelling device, and—let’s say—they have a coag-negative Staph bacteremia or it’s picked up on the blood cultures, it’s not necessarily a bacteremia. So, that becomes challenging. And I think that that’s why we’re not going to be able to get away from clinical judgment in infectious disease.
No, of course not. It’s such a powerful test. I’ll share with you a clinical anecdote. We were taking care of this young guy who unfortunately had opiate use disorder who injected drugs, and he developed a spine infection. And it was one of these cases where it wasn’t quite clear whether he needed to have an operative drainage of an epidural abscess or whether he should be managed medically. I had a conversation with a neurosurgeon and I could kind of tell—even though it was a phone conversation—that he wasn’t really that keen to do an operation. Understandably, these are higher-risk cases. But then I said to him, “You know, we haven’t been able to clear his blood cultures.” And boy, that got his attention. People who have persistent bacteremia, almost always Staph aureus, they rise right to the top of the list, and those you have to be extra careful with.
I have to say, blood cultures are such an important test that it was actually number two on my list. So I’m going to go to my choice number three, which is going to be now my choice number two, and I’m going a little bit more exotic; I’m going to go with the E-test.
That’s an interesting favorite.
Yes, taking us outside of general medicine and now going to a test that maybe only ID geeks know about, and it’s the Epsilometer test. It’s sort of fascinating. When you think about how we do antibiotic susceptibility testing, it all starts with putting discs with antibiotics down on a plate and seeing the zone of inhibition, and saying, “Oh look, big zone. Therefore, it’s susceptible.” And what the makers of the E-test have done is they really cleverly decided to have this gradient of antibiotic concentrations on this long strip. And then you put that strip down on the plate. And then you see the zone of inhibition is bigger around one end and then smaller around the other. Where the zone of inhibition crosses the E-strip is the MIC [minimum inhibitory concentration]. It’s a really remarkably ingenious way of kind of combining various ways of doing susceptibility testing.
Can you tell us when you would use that now?
There are certain new antibiotics that are coming out where you don’t have standardized or machine ways of doing susceptibility testing. So, you really have to get the new antibiotics that have E-tests that are associated with them. It’s also really, really useful with commonly used antibiotics and common bugs to know, “Boy, this is a ceftriaxone-susceptible Strep species. I can treat this patient with endocarditis with ceftriaxone or with penicillin.” There’s really a lot of serious infections where the E-test is absolutely critical. So, going with the E-test, whoever thought of it should get an award, he or she. I don’t know what awards you give in diagnostic microbiology, but why not? Don’t you think?
I think so. I’m sure they have some award ceremony where, maybe, we’re not invited.
Exactly. You know, I would say that the people who don’t know about the E-test, and now have learned something new, they might also confuse it with the D-test, which definitely is not something that’s going to be in my list. Just a quick parenthesis, this is a way of diagnosing inducible resistance in Staph aureus isolates to clindamycin. And now, it’s kind of standard in all labs. And the reason it’s called “D” is because the zone of inhibition between the erythromycin disc and the clindamycin disc is irregular and forms a “D”—little known fact.
It’s not round all the way around; it’s flat on one side.
It’s flat on one side and it looks like the letter “D.” Of course, it’s key that you put the clindamycin disc on the right and the Erythromycin disc on the left, or else it’s not a “D” at all. It’s just some funny shape.
I’ll encourage people to Google that if they have any questions about what that looks like.
All right, Rebeca.
Well, E-test, I have to say it was not on my list.
I’m not surprised.
What’s on my list is lumbar puncture.
Wait, wait, wait, wait, wait. I’m talking about diagnostic tests in infectious disease. Lumbar puncture is a CSF [cerebrospinal fluid] exam.
I love the lumbar puncture. I love it.
If you need to have your head examined. Maybe you need a lumbar puncture?
You know, you focused way down in the infectious disease on this E-test. I’m going to take the 30,000-foot view because I think that people come with a fever and a headache, and lumbar puncture can be so helpful. I especially love it when it’s totally normal.
Yes, very reassuring.
You shared an anecdote about a case that you saw—I think that, the lumbar puncture, if you can put together all the data points that it can provide you, really, it’s so helpful diagnostically. And I’ll tell you that, you may remember this case; we presented it to you in Morning Report one year.
Oh, yes.
And yes, it was somebody who came in overnight with a biphasic illness; that’s not how she described it, but she had severe headache, phonophobia, photophobia and the night float resident did the LP, and this person had a very high white blood cell count and otherwise looked like aseptic meningitis totally; and based on that was able to diagnose LCMV [lymphocytic choriomeningitis] overnight.
Yes, really remarkable clinically.
It was very remarkable. The comment from one of my colleagues at the time was, “I wouldn’t have even sent that test.” And I think that there’s something to be said about the lumbar puncture with aseptic meningitis can sometimes say, “Well, aseptic meningitis will just run its course,” but I think that it also prevents further work up that may be unnecessary. If there is something like HSV [herpes simplex virus], you can treat it.
I agree. It is clinically useful and useful both as a negative and useful when it gives you a positive. I guess, you know, it is very general, but not so general that it really doesn’t apply. I mean, you know, you didn’t choose cardiac catheterization for example. That’s not on your list, so—
No. It’s further down, anyway.
At least you chose something that is commonly used by infectious disease doctors or commonly recommended. I’d say that we are second specialists to the neurologists in recommending LPs.
Yes. People request an ID consult, and they are always unhappy when one of the recommendations is lumbar puncture.
Yeah, this actually reminds me of a very common primary care interaction you get where a primary care clinician might send a syphilis test to work up mild dementia, and then the syphilis test comes back positive, and the query inevitably is: “Do I really have to do a CSF exam?” And sadly, the answer is yes, unless of course you have a clear alternative explanation for that person’s dementia. And then you did a test for neurosyphilis and it came back positive, so you’ve got move to the next step, which is the CSF exam, which would usually be normal.
I’m going to stay with the exotic ID tests and I’m going to take, as my next choice, the MALDI-TOF test. The MALDI-TOF test is something that has entered our world relatively recently. And by the way, for those listening, this stands for matrix assisted laser desorption, ionization, time-of-flight mass spectrometry. That’s pretty good. I’m glad they call it MALDI-TOF because that’s a lot to say. That’s even more to say than trimethoprim-sulfamethoxazole. But anyway, this is a way that the micro lab now can identify microorganisms so accurately. And we now have a blood culture result that comes back and, hey, speciating microorganisms to such a fine degree is really powerful. We’ve been learning about all these new organisms, Gram-negatives, Gram-positives, you can use it now for identification of fungal infections. And it takes away a lot of the ambiguous biochemical tests that we used to rely on. So, I think MALDI-TOF is a real advance. Kind of a cutting-edge one, but my choice number three.
Does it save time?
Well, it does save time. Once you have the machine up and running it definitely saves time. The other thing that is great about it is that the database of characteristic readings in the MALDI-TOF machine are constantly being updated so that you can learn more and more and more all the time. The more people who have the MALDI-TOF machine, the better the identification of the organisms will be. That’s my next choice. I agree that it’s kind of a little bit exotic, but I have a bunch of runners-up I’m going to mention too.
So, as far as I understand, it saves the lab people time, but you still have to grow the organism first.
Yes, you need actual microorganisms.
Like a colony or a positive blood culture.
Exactly. I’m not a laboratory scientist by a long shot; in fact, I’ve never done a single second of laboratory research in my life, but I do wander through the micro lab, and it’s kind of cool to see these machines, and it’s also kind of cool to Google how these tests are actually done because it’s really remarkable. So, that’s my choice number three.
Okay, well, I’m going to do the same thing I did last time where you are drilling down on ID and I’m taking another 30,000-footview. I’m going to go with the pap smear.
The pap smear. The good thing about these drafts is that we don’t really set a lot of rules ahead of time.
Oh, I thought you were going to say we don’t really set policy or guidelines. “These are the best ones. Only use these.”
Exactly. Yeah, I think maybe we should ask John Bartlett to be our official scorekeeper.
Arbiter, yeah.
Or our commissioner.
That sounds good. So, the reason I like the pap smear is because it is actually a preventive diagnostic test.
Rebeca, before you go on, remember these are diagnostic tests in infectious disease.
Yeah. It’s a diagnostic and a preventive at the same time, and the infection is HPV and especially high-risk human papilloma viruses that can cause cervical cancer. I think that it’s an incredibly powerful way of detecting disease early and preventing worse outcomes. And either you can do HPV screening on its own, you can do it with cytology, you can do reflex testing. Do I want to include HPV in there? But because it’s caused by this virus, I’m putting it in diagnostic test for infectious disease.
All right, well, you convinced me on the lumbar puncture, but on this one, I’m not so sure. Let me just also counter by mentioning the incredible advance in human health with the HPV vaccine. I mean, it’s amazing. We already have seen declines in cervical cancer incidence in some parts of the world, and you could envision a time not so far in the future, Dr. Rebeca Plank, when the pap smear is not even necessary.
I hope so. I hope that that will be very soon, but in many parts of the world they don’t have the vaccine yet. So, we do still need to do screening and prevention.
One other interesting thing about pap smears, and you as a person who spent some time in Africa will know about this, is that occasionally, people will diagnose genitourinary schistosomiasis on a pap smear. There is a characteristic quality to the cervix when people have that. Do you remember that?
Yeah, they have the calcified eggs on the cervix, and you scrape it with a spatula, and it makes a little sandpaper noise.
Exactly. And not only that, it turns out to be a way of increasing risk of HIV acquisition. So, a little-known fact.
Now you’ve come around. You love the pap smear.
I’d say you’ve made some valid points. So, just to summarize, my top three choices were HIV test, the E-test, and MALDI-TOF.
And mine were blood cultures, lumbar puncture, and pap smear.
Okay, so, let’s talk about some of the runners-up now. Do you have any that you had on your list that you were considering that you didn’t get to use?
Well, as I said, HIV antibody was high on my list, but you chose first, so you get that on your team. The other thing I think is great is flu PCR [polymerase chain reaction]. One of the reasons I like this test is because it can help you treat people that need to be treated, and it also helps you isolate people who should be isolated to prevent contagion. And it’s come a long way since the rapid antigen. I will say that the important thing about it is making sure that you get a good sample because just swabbing a nostril is not going to make the diagnosis.
One other test I had on my list that’s sort of related to flu PCR is the HSV CSF PCR. Imagine you had a person with encephalitis in the old days when you didn’t have HSV PCR. This is the most common, sporadic cause of encephalitis, and now we can make the diagnosis noninvasively. And so, we don’t need to do a brain biopsy anymore. So, that was one I considered on my list.
You might need to do an LP, though.
Yes. One I actually also considered on my list was the serum cryptococcal antigen as a way of excluding meningitis because, remember, virtually everybody who has cryptococcal meningitis, it got to the meninges via the blood. And so, therefore, they have a positive serum cryptococcal antigen. So, those are two that I considered using, but ultimately were further down in my list.
So, we’re going to shift now and instead of talking about our three favorite ID diagnostic tests, we’re going to talk about our least favorite ID diagnostic test. We are each going to choose one. And Rebeca, just because I’m generous and I went first at the start of this, you’re going to go first now.
All right. So, mine is a culture from a nonsterile site.
Ah, excellent choice. The bane of the ID doctor’s existence. Give me some examples.
One of them is tracheal aspirate in an ICU patient. And I think that one of the things that makes that really hard is that any of the things that grows in that aspirate could or might not be causing a lower respiratory tract infection. And it’s hard to put all the pieces together. Also, if you have Stenotrophomonas or an Acinetobacter that’s extremely drug-resistant, we’re really in a corner as far as whether or not we’re going to pull the trigger on a big gun antibiotic.
It’s very tough. There are ID specialists who have made their whole career on this diagnosis of pneumonia in the ICU patient, and it’s not so simple as just sending a tracheal aspirate.
It’s not simple. And then, the other clinical specimen that I think is really hard is a swab from a diabetic foot. And I think that one of the things that makes that challenging is the same—the things that are on the swab may not be the things that are causing infection lower down. And they don’t always correlate. So, I think that you have something very resistant on the surface or something very sensitive, it still makes you nervous about what you’re not treating underneath.
Absolutely. I’ve got a third one for you. How about the culture from the abdominal drain?
When the drain has been in for two weeks.
Yes. Yes. Frequently, multi-pathogen, which one is causing the problem…
And Candida.
And Candida along for the ride. Okay, excellent choice.
All right. So, my least favorite ID test is the Lyme serology. I mean, this is a really difficult test on so many fronts. First, we know that it’s negative when people show up with the erythema migrans. They show up with early Lyme disease, you don’t even need to send the test. You just look at the rash and make a diagnosis. And the test is very likely to be negative in those contexts. Second is, once the antibody turns positive, which can take several weeks, and usually it’s positive when people have disseminated disease or Bell’s palsy or one of the late manifestations, you then treat the patient. And the positive test may or may not go away, and may or may not correlate with whether the patient gets 100 percent better from the treatment. So, it’s not good as a test of cure either.
I think what we really need for Lyme disease testing is we need a good PCR, a good antigen test. We just don’t have that, and I hope some smart people out there are working on it because, boy, it can’t come soon enough. What do you think?
Building on that, there are a lot of developments that are promising, as far as testing for even DNA or RNA of pathogens, I think that we have to be careful that if we know what we’re looking for, that’s going to be very useful. If we are pan scanning people as it were, I think that can be a little bit problematic because again, it comes down to the problem of what’s a true pathogen.
Yeah, sure. But I think that we would agree that if you were evaluating someone, for example, for an unexplained neurologic syndrome and they had a very highly predictive test for Lyme PCR or antigen, that would be very useful.
It would be very useful. And I think that the other thing is people who, for example, had an infection two years ago, and now they have new symptoms, you know is this another infection because it’s hard to rely on antibodies, as you said.
Okay, so now, the last section of this scintillating podcast, we are going to have a category that we’re calling “aspirational,” aspirational diagnostic tests and ID. Not aspiration, like aspiration pneumonia, but aspirational, like things we would like to have. I think it’s my turn to go first, and the one I’m going to choose is a way of amplifying the findings of PET CT scans to help distinguish between infection and neoplasm. All of us who do ID consultations will be asked to see a patient who, for example, has a spot on his or her lung that has increased uptake in a focal avidity or however the radiologists say it, and there’s this bright spot on the PET Scan, and it’s consistent with either neoplasm or infection. I would like them to be able to wave a magic wand over that spot to then tell us, “Yes, this is infection,” or, “No, this is not infection, but it’s cancer,” or something like that. Right now it’s an area of uncertainty, and we have a lot of trouble distinguishing the two.
That would be great. I support that.
Okay, excellent.
So, mine actually goes back to what you were saying about Lyme disease, which I wish that we had a way of diagnosing pathogens quickly; for example, instead of having to grow the pathogen before we can identify it, if there were some way of using genetic material information to make a diagnosis.
Like PCR or 16S. Those things are coming.
They are coming, exactly. I don’t understand why they are not yet mainstream, maybe you can say.
Yeah, there are many scientific/diagnostic companies that are working on making these a reality. And the big challenge, of course, is validation—making the positive test be correlated with an actual disease. And that’s where they fall down. So, there are tons of these outfits out there who want to do this kind of testing that you describe, and it’s very exciting because I think it’s coming. I would guess, within 5 to 10 years, we will have this in all labs. But right now, nothing has been so validated that we can then replace it with culture. But that’s exactly what we need. You know, we have often situations where we can’t identify an organism, and we would love to have it before culture, and we would love to have some molecular amplification test that gives us the answer, but we just don’t have it yet. But stay tuned.
Rebeca, I just want to thank you again for joining me on this OFID podcast. We were drafting today, ID diagnostic tests. Thanks for joining me.
Thank you, Paul.