Diagnosing C. diff Infections for Optimal Patient Outcomes

Nov. 17, 2019

Clostridioides difficile, better known as C. diff, is large problem. C. diff is 1 of 5 microorganisms on the Center for Disease Control and Prevention (CDC)'s Urgent Threat List, the top category of antibiotic-resistant threats to human health. With 223,900 estimated cases in hospitalized patients in 2017, and 12,800 deaths, C. diff infection is the most prevalent infection caused by the Urgent Threat List members.

November is C. diff awareness month. This makes November a great time to highlight the many research efforts going into not only treating but also diagnosing this terrible disease. We spoke with Colleen Kraft, M.D., Associate Professor of Medicine at Emory University School of Medicine. Kraft is the lead author of a recent meta-analysis study on best diagnostic tests for C. diff infection.
Clinical Microbiology Reviews: A Laboratory Medicine Best Practices Systematic Review and Meta-analysis of Nucleic Acid Amplification Tests (NAATs) and Algorithms Including NAATs for the Diagnosis of Clostridioides (Clostridium) difficile in Adults

Press play to listen to this interview

Why is C. diff such an important infection?

Colleen Kraft: C. diff is an important infection because we know it’s the number 1 cause of hospital acquired diarrhea in the United States. When someone has this infection, it doubles their risk of mortality, doubles the length of their hospital stay, and doubles the cost of the hospital stay.
C. diff is a very important infection. It’s something that can be acquired in the hospital setting as a spore; when it’s in its spore state, it’s very easy to get for people who are fragile.

How has C. diff infection changed within the last 10 or 20 years?

Kraft: C. diff infection has changed in the last 10-20 years. We know of more severe strains, such as the NAP1 strain. We’ve also seen C. diff enter the community in ways that most of us working in the hospital did not anticipate. We are seeing people who are otherwise healthy receiving antibiotics for fairly benign reasons, such as a tooth infection, come down with C. diff. This has increased dramatically in the last decade or two.

Who is at most risk of C. diff infection?

Kraft: We know that people who are medically fragile are most at risk for C. difficile because their gut microbiomes, or gut gardens, are going to be disrupted. Anybody that takes antibiotics in theory is at risk for their gut microbiome being disrupted from antibiotics and this leads to the possibility of C. diff growing up, like a weed.
In terms of at-risk individuals, we know C. diff occurs predominantly in the elderly.

Can patients do anything to prevent contracting C. diff infection? 

Kraft: We know that probiotics are not regulated and we [health care professionals] think there is inconsistency with over-the-counter probiotics—you can’t really say that replanting your gut garden by taking probiotics will completely keep you from getting C. diff—however, any time you can use dietary probiotics or try to do things that will return your gut garden back to normal, it is a good idea. It still doesn’t protect you 100% against C. diff diarrhea.

Can healthcare practitioners do anything to prevent C. diff infection in their patients? 

Kraft: The message for health care providers is the same as for safe antibiotic use: that we want to make sure we are using the antibiotic in the right patient for the right reason. If we are using unnecessary antibiotics, giving these antibiotics allows gut gardens to be disrupted and let C. diff grow up like a weed. Really what healthcare professionals can do is to not give unnecessary antibiotics.

How is C. diff most successfully treated in most patients?

Kraft: The mainstay of C. diff diarrheal and C. diff infection treatment is antibiotics. Again, I go back to my gut garden reference: that basically we are treating the weed with a weed killer. Giving antibiotics against C. diff is still the mainstay of what we do.

What is the biggest challenge in accurately addressing CDI diagnosis? 

Kraft: The biggest challenge is that in our field of clinical microbiology, we tend to do comparison studies that compare the sensitivity, specificity, and negative and positive predictive values which can be grouped as diagnostic accuracy. We look at the diagnostic accuracy of a test but we’re leaving out pre-analytical characteristics as well as the clinical outcomes of the patients we’re testing.
Why would we do that? The reason is that those are very expensive studies, those that include all types of variables. One of the things we’re learning in our field is that many people do side-by-side comparisons and forget about the clinical implications during their studies.

There are several types of diagnostic tests for C. diff. Can you explain what each test is measuring, and the pros and cons of these tests? 

Kraft: One of the things that has led us as a field to worry about how we’re diagnosing C. difficile diarrhea is one of the tests, called a nucleic acid amplification test or PCR. This test simply amplifies the pathogenicity locus of C. difficile; it’s looking for a gene in the chromosome of the bacterium. What you don’t know if you find that gene, is whether that gene is turned on and is creating toxin; it does mean that a toxogenic C. diff strain lives in that stool that you’re testing.
This result has contributed to the issue of overdiagnosis of C. diff: people may be colonized, may not be having symptoms from it, but for some reason they get their stool tested and it shows that they have C. diff, so they get treated for it. As someone who sees people who get treated over and over and over again, I’ve seen how even the treatment of C. diff can eventually perpetuate C. diff. It leads the patient into a cycle where their gut garden can never be replanted.
Another test that we looked at was the enzyme immunoassay (EIA) test for toxin. This by itself is a fairly insensitive test but it does demonstrate–in that heterogeneous matrix of stool–that there is actually toxin present. One thing that has happened is that people want both the sensitivity of that PCR test I just mentioned with the idea that the toxin can be used to determine whether the patient has an active toxin-producing infection.
The other test we looked at was only in algorithm or in combination with the other tests, and that was the GDH test, which tests for an enzyme. That enzyme test actually correlates with the C. diff PCR but the test is again not very specific, so it usually needs to be paired with something else to show that there’s active infection.

Is there a difference in time-to-result that makes these tests more or less useful? 

Kraft: Some of our gold standards are toxogenic culture or cytotoxicity assays. Those tests take many, many days to come back, so those are not actually clinically practical and haven’t been used for decades.
Of the tests we just covered, there’s not a significant difference in turnaround time. Some clinical labs batch these tests and some labs do these in sequence or one at a time, and one doesn’t have preference over the other given the turnaround time. 

How do you know if a diagnosis has been made accurately using one of these tests?

Kraft: This is a very exciting thing about being on the Committee of Laboratory Medicine Best Practices. We are basically using the knowledge of clinical microbiologists to rate the quality of the literature that we are extracting for its content. We will look at a published manuscript and examine the diagnostic accuracy, but we also have people who are experts in the field and microbiologists through ASM, who look thorough these manuscripts and tell us whether the quality of evidence is high, medium, or low. This helps us have confidence in the recommendations that we make.

In your study, you whittle more than 11,000 studies to around 68. How long did that take? 

Kraft: It took us 5 years from the start to the publish date.

Diagnostic algorithms are workflows take multiple test results into account. What algorithm did your meta-analysis study find most effective at accurately diagnosing C. diff patients? 

Kraft: We found that any algorithm that contains C. diff PCR was a good test. That correlated well with the gold standards of the field. We did not have enough literature to make the statement that PCR alone is sufficient for C. diff diagnosis, so at the step we can only say that any of them can be used.

How do you hope the diagnostic algorithm will lead to better patient outcomes?

Kraft: What I really want to come after this manuscript tis that people start to do studies that are relevant to the diagnosis of diseases from a clinical outcome standpoint and not just based on the diagnostic accuracy of the test itself. That’s what I really hope comes from this. Our work that we showed wasn’t dramatically different from what anyone else had shown; we were just able to put it all together. If we can motivate others to produce quality evidence to support testing and diagnosis of certain infections, that’s the goal of the manuscript.
My vision as the Chair of the Laboratory Medicine Best Practices Committee is that we would have all our membership, any clinical microbiologist who wants to, learn how to go through this process of systematic review and meta-analysis and participate in our guidelines. Members would thereby learn how to rate the quality of literature and design better studies. I think this could be a paradigm shift for ASM.

Be sure to read the Clinical Microbiology Review report describing the meta-analysis in greater detail, which you can find at cmr.asm.org. You can follow Colleen Kraft on Twitter for more C. diff and clinical microbiology updates.

Author: Julie Wolf

Julie Wolf
Dr. Julie Wolf is in science communications at Indie Bio, and was a former ASM employee. Follow Julie on Twitter for more ASM and microbiology highlights at @JulieMarieWolf.