Bile Acid Model Helps Predict Recurrent C Difficile Infections


There was no significant difference in performance between the model with only bile acids compared to models with bile acids and clinical variables.

Jessica Allegretti, Brigham and Women's Hospital

Jessica Allegretti, MD

A new model incorporating clinical variables with bile acids could be a predictor of the risk of developing recurrent Clostridioides difficile infections (CDI).

In data presented during the 2021 Digestive Disease Week (DDW) Virtual Meeting, a team, led by Jessica R. Allegretti, Division of Gastroenterology, Hepatology and Endoscopy, Brigham and Women's Hospital, investigated novel biomarkers as potential predictors for C difficile recurrence.

Commensal gut microbiota can metabolize primary bile acids into secondary bile acids. This can inhibit C difficile growth and germination.

One issue is current prediction tools for CDI recurrence do not incorporate microbiota-derived metabolites.

Data Collection

In the prospective study, the researchers examined 59 patients experiencing a first CDI episode, diagnosed by toxin immunoassay (EIA) or polymerase chain reaction (PCR), and were undergoing treatment.

The researchers collected stool samples serially for 8 weeks following the completion of anti-CDI therapy if there was no recurrence reported or until the point of recurrence, defined as diarrhea with positive stool toxin B EIA.

The team also performed liquid chromatography-mass spectrometry to profile fecal bile acids and used the week 1 and 2 post-antibiotic time point to identify potential predictors.

They also performed student’s T test on continuous variables and constructed multivariable logistic regression models.

Two Models

The researchers started with a bile acid only model using a univariate screen and performed stepwise selection with a stay-criteria of 0.10.

The second model incorporated clinical variables previously shown to be predictive of CDI recurrent.

To test the nested area under the curves of these 2 new models, the researchers used methods by DeLong.


Of the 59 first episode CDI patients included in the study, 34% (n = 20) during the 8 weeks of follow-up. The average time to recurrence was 1.9 weeks and at week 1 and 2, there were several predictive bile acids metabolites identified between re-currers and non-re-currers including isolithocholic acid (P = 0.05), lithocholenic acid (P = 0.04), murocholic acid (P = 0.03), glycoursocholanic acid (P = 0.02), and 5-beta-cholanic acid (P = 0.05).

“Logistic regression model with these co-variates with 2 selected to stay after the stepwise procedure: 5-beta-cholanic acid (OR, 0.98; 95% CI, 0.97-1; P = 0.05) and glycoursocholanic acid (OR, 3.08; 95% CI, 1.43-6.61; P = 0.004),” the authors wrote.

The area under the curve after combining the 2 bile acids was 0.823.

In addition, 2 clinical factors previously identified as predictive of c difficile recurrence were also included—use of metronidazole vs. vancomycin and diagnosis with EIA vs. PCR.

This resulted in an area under the curve of only 0.838.

Overall, there was no significant difference in performance between the model with only bile acids compared to models with bile acids and clinical variables (P = 0.64).

“In this cohort, higher relative abundance of glycoursocholanic acid was associated with recurrence and higher relative abundance of 5-beta-cholanic acid was associated with non-recurrence,” the authors wrote. “The BA analyses may be best utilized as a marker for microbiota recover. Further independent validation of these potential novel biomarkers is required to assess their role as predictors.”


Related Videos
Prashant Singh, MD | Credit: University of Michigan
Noa Krugliak Cleveland, MD | Credit: University of Chicago
Ali Rezaie, MD | Credit: X
Remo Panaccione, MD | Credit: University of Calgary
Francisca Joly, MD, PhD | Credit: The Transplantation Society
Paul Feuerstadt, MD | Yale School of Medicine
| Image Credit: LinkedIn
Oriana Damas, MD | Credit University of Miami
Ali Rezaie, MD | Credit: X
Tailoring Chest Pain Diagnostics to Patients, with Kyle Fortman, PA-C, MBA
© 2024 MJH Life Sciences

All rights reserved.