A multidimensional approach towards studying recurrent Clostridium difficile infection

Clostridium difficile infection (CDI) is an infection of the gastrointestinal tract causing symptoms ranging from mild diarrhoea to life-threatening toxic megacolon. Between 10-30% of patients suffer a recurrent episode (rCDI) after an initial episode. Some patients develop multiple recurrent episodes, leading to unpleasant cycles of disease and antimicrobial therapy. This thesis utilises a multidimensional approach to study rCDI.
In Chapter 2, previously generated clinical data is used to assess the effect of treatment delay on two outcomes; diarrhoeal duration and risk of recurrence. It was hypothesised that delays initiating treatment result in increased symptom duration and recurrence risk. Logistic regression models highlighted treatment delay has no significant effect on diarrhoeal duration or recurrence risk. The only significant variable associated with risk of recurrence was previous CDI (P<0.001). These findings suggest clinicians should not be overly concerned by treatment delays in mild/moderate CDI.
In Chapter 3, the germination and thermotolerance properties of five strains of C. difficile spores were investigated. In the nosocomial environment spores may be reingested by the patient, germinate and initiate fulminant disease. Additionally, spores can persist in the gastrointestinal tract and germinate in response to stimulatory cues. C. difficile spore recovery was optimised by using variety of media and supplements. The ribotype (RT) 078 strain germinated more efficiently in the absence of additional supplementation. RT 027/078 strains were more thermotolerant. Intrinsic differences in spore germination characteristics between clades could facilitate the increased ability of some strains to cause rCDI.
In Chapter 4, an in vitro gut model was used to simulate rCDI. Previous research has characterised changes in the microbiota that occur in response to antibiotics. In this study a metaproteomic approach was utilised to study the overarching metabolic processes occurring during simulated rCDI. Although dysbiosis was evident, the metaproteome remained fairly constant throughout simulated infection.