Background: During treatment of Clostridium difficile infection (CDI), patterns of pathogen reduction in relationship to changes in components of the normal microbiota are hypothesized to be predictive of response to treatment and subsequent sustained cure.
Methods: At a single center, subjects enrolled into phase 2 and 3 C. difficile treatment clinical trials (2003-2008) provided fecal samples to assess killing of C. difficile and changes to components of the microbiome. Quantitative bacterial cultures, measurement of C. difficile toxin titers, quantitative polymerase chain reaction of fecal samples for Bacteroidetes, Clostridium clusters XIVa and IV, and C. difficile were performed.
Results: Quantitative bacterial cultures showed a mean log10 C. difficile count (colony-forming units [CFU]) of 6.7 ± 2.0 at study entry; vancomycin treatment consistently reduced C. difficile counts to the limit of detection (2.0 log10 CFU/g), whereas metronidazole was associated with mean C. difficile counts 1.5-2 log10 higher at 10 days of treatment. In patients receiving tolevamer, C. difficile persisted in high counts during treatment; response to treatment was correlated with neutralization of toxin along with persistence of normal microbiota components. However, this was achieved in approximately half of subjects. Both vancomycin and metronidazole further suppressed microbiome components during treatment of CDI. Lactobacilli were observed to be a microbiome component that persisted during treatment of CDI.
Conclusions: Differences of pathogen clearance and microbiome perturbation during treatment of CDI appear to explain treatment outcomes. The hypothesis that probiotic microbes could help prevent onset of CDI is supported by the observation of persistence of lactobacilli during and after treatment of CDI.
Keywords: C. difficile; lactobacilli; metronidazole; tolevamer; vancomycin.
© The Author 2015. Published by Oxford University Press on behalf of the Infectious Diseases Society of America. All rights reserved. For Permissions, please e-mail: journals.permissions@oup.com.