Abstract

Abstract Background Surveillance of C. difficile infection (CDI), one of the most common U.S. healthcare-associated infections, is essential to inform public health response, characterize strains causing disease, and describe changes in strain prevalence over time. The U.S. CDC’s Emerging Infections Program (EIP) has conducted active CDI surveillance, including culturing of C. difficile from a convenience sample of stool samples from cases, since 2009. Cary-Blair transport medium (CBTM) is used by some clinical laboratories for stool testing, but little is known about its impact on the recovery of C. difficile from culture. Methods We performed a cross-sectional study of the recovery of C. difficile isolates from C. difficile positive stools received 10/2021 – 10/2022 from Colorado and Georgia EIP sites. Stools were collected and stored within CBTM or within no transport medium per site protocols and subsequently frozen. Frozen samples were thawed and inoculated on taurocholate-cycloserine-cefoxitin-fructose agar plates (TCCFA) for recovery. Specimens from which C. difficile was not recovered on TCCFA plates were subjected to alcohol shock to enhance the recovery rates. We compared the recovery rates of C. difficile from stools in CBTM and stools in no transport medium. Results We cultured 825 stools from Colorado and Georgia, recovering C. difficile from 81.5% of stools. Overall, 412 (49.9%) stools were in CBTM and 413 (50.1%) were stools in no transport medium. Of those with known testing data, 410/410 (100%) CBTM stools vs 371/407 (91.2%) non-CBTM stools were C. difficile positive by PCR test (p< 0.01). Colorado stools accounted for 78% of the CBTM specimens. We recovered C. difficile on TCCFA plates from 59% (244/412) of stools in CBTM compared to 84% (348/413) in no transport medium (p< 0.01). After including alcohol shock results, C. difficile was recovered from 72% of CBTM stools, and 91% from stools in no medium (p< 0.01). Conclusion Cary-Blair transport medium may limit the recovery of C. difficile from culture and impact the ability to detect emerging strains. A large parallel study is needed to determine if these findings are reproducible and if stool collection methods should be standardized to maximize the recoverability of C. difficile by culture. Disclosures Andrew M. Skinner, MD, Academy for Continued Healthcare Learning: Honoraria|American Society of Healthcare Pharmacists: Honoraria|Ferring Pharmaceuticals: Honoraria|MJH Life Sciences: Honoraria Dale N. Gerding, MD, Destiny Pharma,: Advisor/Consultant|Sebela Pharma: Advisor/Consultant

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call