A decade of Clostridioides Difficile infection – An in-depth analytical retrospective review of a single centres experience

Background: Clostridioides difficile infection (CDI) is the leading cause of hospital-acquired infectious diarrhoea. CDI is potentially preventable and can disrupt hospital services. The Beaumont Hospital Department of Clinical Microbiology provides daily onsite C. difficile laboratory testing and has a comprehensive CDI prevention and control programme for over a decade. This has included a CDI surveillance programme which has not been appraised since its development. 

Objectives:

1. To investigate CDI trends using an observational retrospective cohort study of hospitalised patients attending Beaumont Hospital over ten years and to investigate patient factors predicting first episode of CDI and recurrence. 

2. To evaluate the attributes and continued feasibility of the current CDI surveillance approach in Beaumont hospital.

Methods: C. difficile is a standing agenda item at the weekly infection prevention and control team meeting with multi-disciplinary input and reporting to key stakeholders in real time. C. difficile data from 1st January 2012 to 31st December 2021 was extracted. Data included patient demographics, admission details, C. difficile testing and CDI case details, C. difficile ribotyping (RT) and CDI outbreak data. Data on antimicrobial exposure before CDI onset and specific antimicrobial CDI treatment were captured since 2016. All laboratory samples tested for C. difficile, irrespective of case type were analysed initially. Subsequent analysis explored the sub-groups of new onset inpatient using ꭓ2 analyses with unadjusted odds ratios (ORs) presented, new and recurrent CDI by Poisson regression for trends in rates and predicting factors. A Cox proportional hazards regression model for time to recurrence. 

Results: Data associated with 1975 samples from 1437 patients was reviewed. This included 904 patients with 1,045 CDI episodes, 151 (8%) patients with 490 samples who initially had CDI but the subsequent episode did not meet the case definition and 382 (19%) patients with 440 samples who never met the CDI case definition. When RT were reviewed by CDI origin and case type, no difference was observed between healthcare associated (HA) and community associated (CA) groups for the commonest RTs. CDI testing was requested by clinicians in only 22.4% cases. The majority of CDI was HA (82.2%) with more females affected than males (OR 2.3 p<0.01). The rate of CDI recurrence was 9.1%. Fidaxomicin as first line CDI treatment significantly reduced the hazard ratio associated with time to recurrent CDIs. Over the decade, the quarterly CDI rate did not vary significantly, however, community-associated (CA)-CDI increased. HA CDI was highest in January, March and July.  There was no year without a CDI outbreak. RT 014 dominated as well as RT 078, 005 and 015 in both HA and CA CDI. The average LOS differed significantly between the HA (67.1 days) and the CA (14.6 days). 

Discussion and Conclusion: Despite rigorous CDI IPC programmes HA-CDI rates have not changed, even when more sensitive testing methods were introduced, whereas CA-CDI is at its highest in a decade. These findings underpin the importance of using prospective, multidisciplinary surveillance to inform local decision making. The convergence of ribotypes between HA and CA CDI as well as the onset of CDI symptoms in the community question the traditional view of CDI as a hospital issue. Further research to trace circulating RTs between healthcare facilities and the community is needed.