Key points are not available for this paper at this time.
BACKGROUND: Patients with inflammatory bowel disease (IBD) are at higher risk for Clostridium difficile infection (CDI). Disruption of gut microbiome and interaction with the intestinal immune system are essential mechanisms for pathogenesis of both CDI and IBD. Whether genetic polymorphisms associated with susceptibility to IBD are also associated with risk of CDI is unknown. AIMS: To use a well-characterised and genotyped cohort of patients with UC to (i) identify clinical risk factors for CDI; (ii) examine if any of the IBD genetic risk loci were associated with CDI; and (iii) to compare the performance of predictive models using clinical and genetic risk factors in determining risk of CDI. METHODS: We used a prospective registry of patients from a tertiary referral hospital. Medical record review was performed to identify all ulcerative colitis (UC) patients within the registry with a history of CDI. All patients were genotyped on the Immunochip. We examined the association between the 163 risk loci for IBD and risk of CDI using a dominant genetic model. Model performance was examined using receiver operating characteristics curves. RESULTS: The study included 319 patients of whom 29 developed CDI (9%). Female gender and pancolitis were associated with increased risk, while use of anti-TNF was protective against CDI. Six genetic polymorphisms including those at TNFRSF14 Odds ratio (OR) 6.0, P-value 0.01 were associated with increased risk while 2 loci were inversely associated. On multivariate analysis, none of the clinical parameters retained significance after adjusting for genetics. Presence of at least one high-risk locus was associated with an increase in risk for CDI (20% vs. 1%) (P = 6 × 10⁻⁹). Compared to 11% for a clinical model, the genetic loci explained 28% of the variance in CDI risk and had a greater AUROC. CONCLUSION: Host genetics may influence susceptibility to Clostridium difficile infection in patients with ulcerative colitis.
Ananthakrishnan et al. (2013) studied this question.