1091. Algorithmic Release of Clostridium difficile PCR Results from a Multiplex Gastrointestinal (GI) Panel in Children < 3 years old
Session: Poster Abstract Session: Diarrhea Diagnostic Dilemmas
Friday, October 5, 2018
Room: S Poster Hall
Posters
  • Hecht_C diff under 3y IDWeek poster_9.25.18.pdf (472.9 kB)
  • Background: Infants have a high rate of asymptomatic Clostridium difficile (CD) colonization, up to 37%. Given this, our laboratory does not release CD+ results from the BioFire FilmArray Gastrointestinal Panel (FGP) in patients < 3y, unless requested by a physician. We sought to validate this model by comparing results from FGP to semi- quantitative CD PCRs for toxin B and Glutamate dehydrogenase (GDH), enzyme immunoassay (EIA) for toxin A/B/GDH, and physician requests for CD results.

    Methods: Retrospective analysis of children < 3y with GI illness and FGP CD+ results between 9/16 – 4/18. CD PCRs for toxin B and GDH, CD EIA for toxin A/B/GDH were performed on convenience samples of frozen aliquots in Cary Blair. Physician request for release of CD results was used as a surrogate of possible role of CD on GI illness.

    Results: Of 5990 FGP, 2267 (38%) were in children < 3y: 619 (27%) were CD+. Of these 619, 602 (97%) were not reported per algorithm. 62% (386/619) of CD+ samples had co-pathogens detected; enteropathogenic Escherichia coli and norovirus most frequently. For CD PCRs and EIA performed in subset of 49 CD+, mean cycle threshold values (Cts) for toxin B were evaluated (Table 1). Of 48 samples with detectable CD by toxin B PCR, 14 (29%) had both GDH and toxin B detected, 24 (51%) had only GDH detected, and 9 (19%) had neither GDH nor toxin B detected.

    Conclusion: Only 3% of FGP CD results in children < 3 y were released per physician request, suggesting limited clinical significance. A co-pathogen was detected in 62% of CD+ samples that may explain illness. Among evaluable samples, only 28.6% of CD+ had both GDH and toxin detected by EIA, possibly indicating low specificity of CD PCR. Ongoing testing and prospective studies are warranted to determine the validity of our algorithm and if semi- quantitative PCR or EIA can be useful to identify when CD detection by FGP in children < 3y is clinically significant.

    Table 1: CD Toxin B PCR Ct values compared with FGP result release, co-pathogen detection, and EIA results

    Median Ct

    Median Ct

    p-value

    FGP-CD+ result released (n=1)

    20.64

    FGP-CD+ result not released (n=47)

    28.77

    NS

    FGP-CD+ only

    (n=24)

    26.83

    Co-pathogen

    (n=24)

    29.34

    NS

    EIA toxin+

    (n=14)

    23.23

    EIA toxin-

    (n=34)

    31.11

    0.0005

    Shaina Hecht, MD1, Huanyu Wang, PhD2, Kathy Everhart, B.S.3, Monica I. Ardura, DO, MSCS4 and Amy Leber, PhD3, (1)Department of Pediatrics, Infectious Diseases and Immunology, The Ohio State University and Nationwide Children’s Hospital, Columbus, OH, (2)Nationwide Children's Hospital, Columbus, OH, (3)Department of Laboratory Medicine, Nationwide Children's Hospital, Columbus, OH, (4)Pediatrics, Infectious Diseases and Immunology, Host Defense Program, The Ohio State University and Nationwide Children’s Hospital, Columbus, OH

    Disclosures:

    S. Hecht, None

    H. Wang, None

    K. Everhart, None

    M. I. Ardura, None

    A. Leber, Nationwide Children's Hospital: Research Contractor , Research support .

    Findings in the abstracts are embargoed until 12:01 a.m. PDT, Wednesday Oct. 3rd with the exception of research findings presented at the IDWeek press conferences.