Search

Background There is a paucity of data on community-based Clostridioides difficile infection (CDI) and how these compare with inpatient CDI. Aim To compare data on the populations with CDI in hospitals vs the community across 12 European countries. Methods For this point-prevalence study (July–November 2018), testing sites sent residual diagnostic material on sampling days to a coordinating laboratory for CDI testing and PCR ribotyping (n = 3,163). Information on whether CDI testing was requested at the original site was used to identify undiagnosed CDI. We used medical records to identify differences between healthcare settings in patient demographics and risk factors for detection of C. difficile with or without free toxin. Results The CDI positivity rate was 4.4% (country range: 0–16.2) in hospital samples, and 1.3% (country range: 0–2.2%) in community samples. The highest prevalence of toxinotype IIIb (027, 181 and 176) was seen in eastern European countries (56%; 43/77), the region with the lowest testing rate (58%; 164/281). Different predisposing risk factors were observed (use of broad-spectrum penicillins in the community (OR: 8.09 (1.9–35.6), p = 0.01); fluoroquinolones/cephalosporins in hospitals (OR: 2.2 (1.2–4.3), p = 0.01; OR: 2.0 (1.1–3.7), p = 0.02)). Half of community CDI cases were undetected because of absence of clinical suspicion, accounting for three times more undiagnosed adults in the community compared with hospitals (ca 111,000 vs 37,000 cases/year in Europe). Conclusion These findings support recommendations for improving diagnosis in patients presenting with diarrhoea in the community, to guide good practice to limit the spread of CDI.

Objectives Clostridioides difficile (CD) is widely reported as one of the most prevalent multi-drug resistant (MDR) organisms. Assessment of temporally disparate isolate collections can give valuable epidemiological data to further the understanding of antimicrobial resistance progression. Methods A collection of 75 CD isolates (1980–86) was characterised by PCR ribotyping, cell cytotoxicity assay and susceptibility testing with a panel of 16 antimicrobials and compared to a modern surveillance collection consisting of 416 UK isolates (2012–2016). Agar-incorporation was performed to ascertain susceptibility data for vancomycin, metronidazole, rifampicin, fidaxomicin, moxifloxacin, clindamycin, imipenem, chloramphenicol, tigecycline, linezolid, ciprofloxacin, piperacillin/tazobactam, ceftriaxone, amoxicillin, tetracycline and erythromycin. Genomes were obtained using Illumina HiSeq3000 sequencing and assembled using CLC Genomics Workbench. Resistance genes were identified using the Comprehensive Antibiotic Research Database's Resistance Gene Identifier and ResFinder3.0. Results Twenty-six known and one previously unobserved ribotype (RT) were detected. RT015 and RT020 dominated; 21.3% and 17.3%, respectively. Three moxifloxacin resistant (16–32 mg/L) RT027 isolates were recovered, pre-dating the earliest reports of this phenotype/genotype. Phenotypic resistance was observed to moxifloxacin (9.3% of isolates), ciprofloxacin (100%), erythromycin (17.3%), tetracycline (9.3%), linezolid and chloramphenicol (4.0%). Phenotypic comparisons with modern strains revealed increasing minimum inhibitory concentrations (MIC), with MIC50 elevations of one doubling-dilution for the majority of compounds, excluding clindamycin and imipenem. Moxifloxacin MIC90 comparisons revealed a two doubling-dilution increase between temporal isolate collections. Historical genomes revealed twenty different resistance determinants, including ermB (8.0% of isolates), tetM (9.3%), cfr (5.3%) and gyrA substitution Thr-82→Ile (9.3%). Seventeen isolates (22.7%) were resistant to ≥3 compounds (MDR), demonstrating ten different combinations. Intra-RT diversity was observed. Conclusions Antibiotic resistance in CD has increased since the early 1980s, across the majority of classes. Moxifloxacin resistance determinants may pre-date its introduction.