358. Evaluation of Clostridium difficile (CD) Testing Using Molecular-Based (MB) and Enzyme Immunoassay (EIA) Tests in Clinical Stool Specimens from the Saskatoon Health Region (SHR), Saskatoon, Canada
Session: Poster Abstract Session: Clostridium difficile - Epidemiology, Diagnosis, Treatment, and Prevention
Friday, October 21, 2011
Room: Poster Hall B1
  • IDSA 2011 C diff Poster (FINAL).pdf (9.7 MB)
  • Background: 

    CD is the most common nosocomial infectious cause of diarrhea. SHR is the largest health region in the province, serving approximately 300,000 residents. The objective of this trial was to assess if MB tests BD GeneOhm™ (BG) and Meridian Illumigene™ (MI) were more sensitive and specific than the TechLab Tox A/B II™ (TT) EIA-based test.


    SHR stool samples submitted by a physician for CD testing between Nov '10 to Feb '11 and that met the screening criteria for processing were tested simultaneously with BG, MI and TT. Results for all samples were recorded as positive or negative for each testing platform. If all 3 results were concordant, then the true result was recorded as such. If any of the 3 results were discordant for a sample, then further testing was performed using cytotoxicity assay testing and this was used as the true result.


    A total of 301 stool samples were tested during the study period. Eleven (3.7%) samples required cytotoxicity testing for discordant results. The prevalence of disease was 10.6% and 55% of samples were from female patients. The median age of patients with positive samples was 77.5 vs. 63.0 years for those with negative results (p=0.02). TT displayed a sensitivity (SN) of 90.6% (95% CI: 73.8-97.5%), specificity (SP) 100% (95% CI: 98.2-100%), positive predictive value (PPV) 100% (95% CI: 85.4-100%) and negative predictive value (NPV) 98.9% (95% CI: 96.5-99.7%). BG and MI testing had SN 100% (95% CI: 86.7-100%), SP 97.4% (95% CI: 94.5-98.9%), PPV 82.1% (95% CI: 65.9-91.9%) and NPV 100% (95% CI: 98.2-100%). The agreement of results between the two MB tests was very strong (kappa=0.97) but weaker between the MB and EIA tests (kappa=0.84). Cytotoxicity testing had higher agreement with EIA (kappa=0.95) vs. MB testing (kappa=0.89).


    Sensitivity of MB testing was higher than EIA testing, but the reverse was true of specificity. Agreement between testing methodologies and the historical gold standard for CD was higher with EIA rather than MB testing. While unexpected, these results challenge the idea of cytotoxicity testing being the gold standard, as the two MB test results had high levels of correlation. More studies are needed to assess if MB testing should become the new gold standard for CD.

    Subject Category: D. Diagnostic microbiology

    Oscar E. Larios, MD1, Erica Sandager, MLT2, Shena Kott, MLT2, Ulana Wills, MLT2, Stephen Sanche, MD1, Sophie Robichaud, MD2 and Joseph Blondeau, MSC, PhD, RSM(CCM), SM(AAM), FCCP2, (1)Division of Infectious Diseases, University of Saskatchewan, Saskatoon, SK, Canada, (2)Royal University Hospital, Saskatoon, SK, Canada


    O. E. Larios, None

    E. Sandager, None

    S. Kott, None

    U. Wills, None

    S. Sanche, None

    S. Robichaud, None

    J. Blondeau, None

    << Previous Abstract | Next Abstract

    Findings in the abstracts are embargoed until 12:01 a.m. EST Thursday, Oct. 20 with the exception of research findings presented at IDSA press conferences.