By Richard R. Watkins, MD, MS, FACP, FIDSA, FISAC

Professor of Internal Medicine, Northeast Ohio Medical University; Division of Infectious Diseases, Cleveland Clinic Akron General, Akron, OH

Dr. Watkins reports no financial relationships relevant to this field of study.

SYNOPSIS: Investigators compared levels of inflammatory markers in patients with Clostridioides difficile infection (CDI) to those who were colonized with it. Several markers appeared to be able to distinguish true CDI, although a gold standard definition of CDI is needed.

SOURCE: Kelly CP, Chen X, Williams D, et al. Host immune markers distinguish Clostridioides difficile infection from asymptomatic carriage and non-C. difficile diarrhea. Clin Infect Dis 2020;70:1083-1093.

Diagnosing Clostridioides difficile infection (CDI) is a frequent conundrum. In clinical practice, diarrhea is a common symptom with myriad etiologies, and testing for CDI using nucleic acid amplification tests (NAATs) can be confounded by those who are asymptomatic carriers of C. difficile. Indeed, the diagnosis can be especially challenging when colonized patients have certain underlying conditions, such as inflammatory bowel disease or irritable bowel syndrome. Kelly and colleagues sought to identify specific biomarkers that differentiate a patient with CDI from one with the combination of diarrhea and C. difficile colonization.

The participants in the study were inpatients ages 18 years or older. They were divided into four cohorts that included presumed CDI, defined as new-onset diarrhea and a positive NAAT; a carrier-NAAT, defined as a positive NAAT without diarrhea; a negative NAAT with diarrhea; and a negative NAAT without diarrhea. The negative NAAT, no diarrhea cohort was originally screened as eligible for the carrier-NAAT cohort (i.e., admitted for less than 72 hours, received at least one dose of antibiotic within the preceding seven days, and did not have diarrhea in the 48 hours prior to stool sample submission), but the submitted stool sample tested negative by NAAT. Patients were excluded if they had a colostomy, received a drug to treat CDI for more than 24 hours in the preceding seven days, had been diagnosed with CDI in the previous six months, or had tested negative for CDI in the preceding seven days.

The investigators obtained discarded serum samples drawn within one day of the stool sample collection. They measured multiple analytes, including serum cytokines, antitoxin A and B levels, and stool levels of calprotectin, toxin A and B, and antitoxin B. Also, serum albumin values within five days before and two days after stool collection were recorded.

There were 122 stool samples in the CDI-NAAT cohort; 44 in the carrier-NAAT group; 44 in the diarrhea, negative NAAT group; and 50 in the no diarrhea, negative NAAT group. Two additional cohorts were the CDI-Tox20 (n = 79), which included subjects who were positive by both NAAT and single molecule array for toxin A + B, and carrier-Tox20 (n = 34). The demographic and baseline laboratory values were similar between all the cohorts, except for significantly lower median albumin levels in the CDI-NAAT and CDI-Tox20 cohorts. Interestingly, the mean white blood cell (WBC) count was highest (12.5 × 109/L) in the no-diarrhea, NAAT negative group. Median values for 11 of the analytes were significantly greater (P < 0.05) in the CDI-NAAT group compared to the carrier-NAAT group. Of these, granulocyte colony-stimulating factor (GCSF) most clearly differentiated the CDI-NAAT group from the other cohorts, with an area under the receiver operating curve of 0.842. There were no significant differences in GCSF levels when the non-CDI cohorts were compared to each other. The other analytes that distinguished the CDI-NAAT cohort from the others, albeit less strongly, included IL-6, TNF-α, and IgG antitoxin A in blood, and IgA and IgG antitoxin B in stool.


The current testing paradigm for CDI at many hospitals, wherein a positive NAAT test is followed by the more specific toxin assay, is being questioned increasingly. This is because of recognition that many hospitalized patients are asymptomatic carriers of toxigenic strains of C. difficile. Thus, it is incumbent on clinicians to recognize when patients need treatment, but importantly, when they do not. The study by Kelly and colleagues is interesting because it found several immune markers, particularly GCSF, that potentially could be developed into tests that differentiate true CDI from colonization. Notably, the patients in the carrier-NAAT cohort all were hospitalized, making it reasonable to assume they would have had elevated levels of inflammatory markers.

Yet, the study raises almost as many questions and it answers. The fundamental problem is that no gold standard exists to diagnose CDI. The authors presumed that all patients with diarrhea and a positive NAAT had true CDI, which may not have been totally accurate. Furthermore, there was not a cohort of patients with diarrhea and a positive NAAT whose diarrhea ultimately was found to be from a different etiology than CDI. Thus, we are left to wonder what their levels of inflammatory markers would have been. Another issue is the finding that serum albumin levels were lower in the CDI-NAAT cohort. How this may have affected the levels of inflammatory markers is unclear.

Developing specific laboratory tests to distinguish CDI from colonization is an important goal. Kelly and colleagues have made important progress in this regard. However, many issues remain to be elucidated, including whether the inflammatory markers they identified can be used to determine which patients will go on to develop severe, complicated, or recurrent CDI. A gold standard definition for CDI remains elusive, but we are getting closer.