Whole-genome sequencing improves discrimination of relapse from reinfection and identifies transmission events among patients with recurrent Clostridium difficile infections

https://doi.org/10.1016/j.jhin.2015.01.021Get rights and content

Summary

Background

Recurrent Clostridium difficile infection (CDI) represents a significant healthcare challenge. Patients may suffer multiple episodes of CDI with the index strain (relapse) or become infected by another strain acquired nosocomially (reinfection).

Aim

We aimed to characterize C. difficile isolates causing recurrent CDI at a tertiary referral hospital by whole-genome sequencing (WGS) to assess strain similarities at the highest level of genetic resolution and accurately detect relapse, reinfection, and putative strain transmission events.

Methods

An 18-month prospective study of recurrent CDI was undertaken. Clostridium difficile was cultured from stool samples collected longitudinally from any patients suffering ≥2 clinically defined CDI episodes. Patient demographics and clinical data were recorded, and strain relatedness investigated by both polymerase chain reaction (PCR)-based ribotyping and WGS.

Findings

Nineteen patients were identified with ≥2 clinically defined CDI episodes who cumulatively suffered 39 recurring CDI episodes (58 total episodes). Patients had a median length of stay (LOS) of 144 days and experienced between two and seven CDI episodes. Ribotyping indicated 27 apparent same-strain relapses, five reinfections and the predominance of ribotypes 078 (ST-11) and 020 (ST-2). WGS allowed characterization of relapse with increased certainty and identified emergent within-strain single nucleotide variants (SNVs) with potential functional impact on diverse genes. Shared ribotypes among 14 patients with recurrent CDI suggested 10 possible patient-to-patient transmission events. However, WGS revealed greater diversity at the sub-ribotype level, excluding all but four transmission events.

Conclusion

WGS exhibits several advantages over PCR-based ribotyping in terms of its ability to distinguish relapse from reinfection, to identify patient-to-patient transmission events, and to exact fine structure characterization of recurrent CDI epidemiology. This offers the potential for more focused infection prevention strategies to eliminate strain transmission among patients with recurrent CDI.

Introduction

Between 15% and 50% of patients who develop Clostridium difficile infection (CDI) will suffer subsequent CDI episodes, which adds to the clinical and economic burden of this disease.1, 2, 3, 4 Recurrence is defined as a CDI episode occurring within eight weeks of a previous infection.5 Accepted risk factors for recurrent CDI include older age (>65 years), prescribing of additional ‘non-CDI’ antibiotics, and cumulative time spent in the healthcare environment.1, 3, 6, 7 Recurrent clinical episodes may be categorized as relapse, when due to the original strain, or reinfection, when caused by a newly acquired strain.

Molecular typing studies of C. difficile have provided insight into the proportions of cases with relapsed CDI as opposed to reinfection.8, 9 Estimates for reinfections range from 12% to 35% of recurrent CDI episodes, within the limits of discrimination provided by conventional typing methods such as polymerase chain reaction (PCR)-based ribotyping, pulsed-field gel electrophoresis (PFGE) or multi-locus sequence typing (MLST) with the intervals to recurrence after a first episode of CDI ranging from 24 to 42 days.2, 3, 7, 10, 11, 12

The use of whole-genome sequencing (WGS) has provided evidence for a higher degree of C. difficile strain diversity than previously acknowledged.13, 14, 15 A recent study applied WGS to 1223 C. difficile strains and found 45% of all isolates investigated to be genetically distinct, suggesting a considerable reservoir of endemic C. difficile strains.15 Of the patients infected with genetically indistinguishable strains, Eyre et al. found that 38% had identifiable hospital contact with another symptomatic case and 36% had no recognizable shared epidemiology.15 This underscores the existence of unidentified C. difficile transmission routes.15, 16, 17 In a subsequent study, Eyre et al. applied WGS to recurrent CDI with the consideration that ribotyping may underestimate reinfections caused by endemic ribotypes.18 This provided improved discrimination between relapse and reinfection through comparisons of paired isolates (index versus first recurrence) and revealed that 81% of recurrences were caused by the same strain, 15% by reinfections with 4% assigned to an indeterminate category.18

We undertook prospective analysis of CDI episodes meeting clinical and microbiological criteria and identified all patients suffering recurrent CDI over an 18-month period. Strains causing index as well as first and subsequent CDI episodes were characterized using both conventional ribotyping and WGS to assess the level of concordance of these methods in view of the enhanced discriminatory power of WGS.

Section snippets

Setting

St James's Hospital (SJH) is a 1015-bed acute tertiary care hospital with some 3800 staff members and an immediate catchment population of about 350,000. Annual inpatient admissions exceed 25,000 with more than 220,000 outpatient and 46,000 emergency department visits per annum.

Study cohort

Between January 1st, 2012 and June 30th, 2013, all clinical cases of recurrent CDI were identified at St James's Hospital, Dublin, in accordance with national guidelines for recurrent CDI. In addition, any patient

Recurrent CDI prevalence and associated patient demographics

Over the 18-month study period, a total of 230 CDI episodes were documented at SJH representing a CDI rate of 0.42/1000 occupied bed-days and a recurrence rate of 10% among hospitalized patients. Although recurrent CDI is generally defined as a positive CDI result dated within the preceding eight weeks of a prior CDI infection, for the purpose of this study, we extended our definition to include any CDI episode preceded by a prior episode in the same patient over the course of the entire study

Discussion

We investigated the molecular epidemiology of recurrent CDI cases at a tertiary referral hospital comparing conventional PCR-based ribotyping and WGS analysis. Overall, the age profile of patients with recurrence was reflective of national data for adult inpatients in Ireland.25 However, our recurrent CDI cohort had an exceptional LOS which placed them in the minority (3.3%) of inpatient admissions nationally.25 Even within this category, the national mean LOS is estimated to be 65.5 days,

Acknowledgements

The authors would like to thank the staff at St James's Hospital Microbiology Department for their technical support and donation of isolates.

References (36)

  • F. Barbut et al.

    Epidemiology of recurrences or reinfections of Clostridium difficile-associated diarrhea

    J Clin Microbiol

    (2000)
  • L.C. McDonald et al.

    Recommendations for surveillance of Clostridium difficile-associated disease

    Infect Control Hosp Epidemiol

    (2007)
  • R. Fekety et al.

    Recurrent Clostridium difficile diarrhea: characteristics of and risk factors for patients enrolled in a prospective, randomized, double-blinded trial

    Clin Infect Dis

    (1997)
  • D.W. Eyre et al.

    Predictors of first recurrence of Clostridium difficile infection: implications for initial management

    Clin Infect Dis

    (2012)
  • C.W. Knetsch et al.

    Current application and future perspectives of molecular typing methods to study Clostridium difficile infections

    Euro Surveill

    (2013)
  • J.W. Marsh et al.

    Association of relapse of Clostridium difficile disease with BI/NAP1/027

    J Clin Microbiol

    (2012)
  • M. Kamboj et al.

    Relapse versus reinfection: surveillance of Clostridium difficile infection

    Clin Infect Dis

    (2011)
  • K. Oka et al.

    Molecular and microbiological characterization of Clostridium difficile isolates from single, relapse, and reinfection cases

    J Clin Microbiol

    (2012)

    • Cited by (36)

    • Prevalence of the Clostridium difficile BI/NAP1/027 strain across the United States Veterans Health Administration

      2018, Clinical Microbiology and Infection

    • Microbiologic factors affecting Clostridium difficile recurrence

      2018, Clinical Microbiology and Infection
      Citation Excerpt :

      The picture is further complicated by patients harbouring multiple C. difficile genotypes [66]. Some studies using more discriminatory techniques suggest that reinfection accounted for ∼50% of recurrent infections [67–70]. Varying rates for recurrence due to relapse have been reported in the literature, ranging from ∼52% to 88% of rCDI episodes [67,68].

    • Genomic and phenotypic diversity of Clostridium difficile during long-term sequential recurrences of infection

      2018, International Journal of Medical Microbiology
      Citation Excerpt :

      There is evidence that strains of this toxinotype do not cause colitis; as they asymptomatically colonize in the gut but do not cause symptoms upon infection using a hamster model (Geric et al., 2006). As has been shown in previous studies, whole genome sequencing of C. difficile strains has the advantage over PCR-based ribotyping methods to track patient-to-patient transmission events and to more precisely distinguish relapse from reinfection (Kumar et al., 2016a; Mac Aogáin et al., 2015; Sim et al., 2017; Stevenson et al., 2016). However, nearly no data have been published on possible in-patient genomic microevolution events during recurrent CDI episodes.

    • Determining the cause of recurrent Clostridium difficile infection using whole genome sequencing

      2017, Diagnostic Microbiology and Infectious Disease

    • Molecular types and antimicrobial susceptibility patterns of Clostridium difficile isolates in different epidemiological settings in a tertiary care center in Israel

      2016, Diagnostic Microbiology and Infectious Disease
      Citation Excerpt :

      More intriguing findings are the high MIC values of metronidazole and vancomycin that were found in 4 of the 14 isolates associated with re-infections, including the PCR-ribotype 017 isolates. These findings were not reported by previous studies (Aitken et al., 2015; Figueroa et al., 2012; Mac Aogáin et al., 2015; Marsh et al., 2012; Sisto et al., 2011), suggests at least an epidemiological connection between patients who develop recurrent CDI and isolates that are non-susceptible to antimicrobials. Hence, a detailed clinical study is warranted in order to evaluate the role of non-susceptibility to antimicrobials with recurrent infections.

    • Human and porcine transmission of Clostridioides difficile Ribotype 078, Europe

      2021, Emerging Infectious Diseases
    View all citing articles on Scopus
    View full text
    Copyright © 2015 The Healthcare Infection Society. Published by Elsevier Ltd. All rights reserved.