Testing for C. Difficile Infections
Types of testing algorithms used in clinical practice to help clarify a diagnosis of C difficile infection.
Peter L. Salgo, MD: One of the nice things about being a moderator for panels like these is I get all these experts, and then I can expose my stupidity and get laughed at. Which is, I take a look at the CDC [Centers for Disease Control and Prevention], I see a white count of 30,000 to 40,000 per mm3. I say, “Ha, it’s C diff. I’m smarter than all these guys. I don’t have to do any more testing. I know what it is.” Right? Wrong? Help me here. Paul says yes. Yes, I’m an idiot, or yes, that’s what I can do?
Paul Feuerstadt, MD, FACG, AGAF: No, you’re thinking along the right lines, and the IDSA/SHEA [Infectious Diseases Society of America/Society for Healthcare Epidemiology of America] guidelines speak to that in terms of triaging severity of infection: white cell count greater than 15,000 per mm3, creatine greater than 1.5 mg/dL. As we approach patients with C diff, they’ll lose 1 element that we should think about. I do agree; If a patient has loose stools, they recently received antimicrobials, and let’s say they’re in the ICU [intensive care unit] and their white cell count is 40,000 per mm3, you probably know what’s going on there, right? That might be a circumstance where time is of the essence. You might certainly send off the stool assay before you start antimicrobials, but you might also want to empirically start antimicrobials because you suspect C difficile is present. That gets into the art of medicine and 1 of the things that most of us really enjoy critically thinking about.
Peter L. Salgo, MD: My residents say, “Dr Salgo”—sometimes they call me Dr Salgo when they’re talking to me and not behind my back—“How do you know this is C diff? Just because the white count is 30,000 per mm3?” I’m tempted to say, “Damn it, Jim, because I’m a doctor,” and they’re too young to know what that is from. That said, there are tests; you can test a pretoxin. There are other specific tests. Dale, could you walk us through this?
Dale N. Gerding, MD: Sure. Just to make a point about the previous question, community-onset diarrhea needs to be worked up different from someone who has health care–associated diarrhea where C diff rises at the very top of the list. In community associate diarrhea when somebody is being admitted you really need to do a multiplex PCR [polymerase chain reaction]–type assay because there’s a lot of community-acquired diarrhea that’s not C diff, even though it’s increasing in the community.
Among the commonly used tests in the United States is the PCR, which detects toxin genes in the organism itself but does not detect the toxin itself. On the other hand, the alternative test is called a specific toxin test, which can be done either with a cell cytotoxicity assay, which has become obsolete because it’s too slow and no labs know how to do it any longer, and it has been replaced by an enzyme immunoassay for toxin.
The problem is that the toxin test is ideal except that it’s not very sensitive and it tends to pick up only perhaps 80% of the positive cases. Whereas PCR is very sensitive and picks up perhaps too many cases, because sometimes people are colonized and have diarrhea for another reason, get tested, and are found to be positive but don’t have toxin detectable in the stool.
There’s a third test called the GDH, that’s a glutamate dehydrogenase enzyme, which can be used as a screening test but is found in both toxigenic and nontoxigenic strains of C diff. You always need to pair a GDH test with a toxin test or some other means to detect either the toxin genes or the toxigenic strain itself. That’s why we frequently use what are called algorithms, putting these tests together in combination. It can be done with PCR first and then followed by a toxin test, or it can be done with a GDH test followed by an enzyme immunoassay for toxin, and then arbitrating those tests that don’t agree with a PCR test. It’s 2 different algorithms, and 1 of them gives you far more cases than the other does. That’s probably causing a lot of confusion in terms of diagnosis.
Peter L. Salgo, MD: There are some acronyms out there that you didn’t mention that I want to put out there. They may be synonymous. Is there a difference between PCR and LAMP [loop-mediated isothermal amplification]?
Dale N. Gerding, MD: The LAMP and PCR testing are both mechanisms to detect the genes. They’re just a slightly different technology. But they’re both equivalent in terms of their sensitivity.
Peter L. Salgo, MD: And NAAT, that stands for nucleic acid amplification test, which you already mentioned. And the GDH, glutamate dehydrogenase, that testing is specifically for what?
Dale N. Gerding, MD: For an enzyme that’s both found in both toxigenic and nontoxigenic strains of C diff, originally developed as a toxin A-test, but unfortunately it was identifying the wrong antigen. It turned out that it’s become widely used because it was once called a latex agglutination test and got put on a new platform, and now it’s become a quite sensitive screening test.
Peter L. Salgo, MD: If you enjoyed this content, you should subscribe. We have an e-newsletter, and you can receive upcoming Peer Exchanges and other great content in your in-box—that’s right, electronically. I’ll see you next time. I’m Dr Peter Salgo. Thanks again for watching.