2361. Evaluation of a 2-Step Testing Algorithm for Clostridioides difficile Infection

BACKGROUND: Clinical data describing use of a multistep algorithm for diagnosis of Clostridioides difficile infection (CDI) is limited. In June 2018 we implemented a 2-step testing algorithm in which PCR testing (Aries® assay) is performed for all specimens followed by EIA toxin testing (TOX A/B QUIK CHEK® assay) when PCR is positive. We sought to describe outcomes for patients with PCR+/EIA+ vs. PCR+/EIA− results. Outcomes evaluated included frequency of CDI treatment, retesting and retreatment within 3 months, and investigator determined categorization of C. difficile results by an investigator blinded to the EIA result. METHODS: A retrospective cohort study was performed on a random sample of 85 unique patients with a PCR+ stool sample from July 2018 through December 2018. Demographic and clinical data were abstracted from the medical record during the index encounter and for 3 months thereafter. Based on predetermined criteria, index encounter results were categorized as representing probable, possible, unlikely, or indeterminate cases of symptomatic CDI. RESULTS: For the 85 study patients, 42%, 27%, and 31% were tested in the inpatient, outpatient, and ED/urgent care settings. Twenty-seven patients (32%) were EIA+, all of whom received CDI treatment. Fifty-eight (68%) were EIA-, of which 79% received treatment. Of the 12 EIA- patient who did not receive treatment two had retesting within 3 months; one of whom subsequently tested EIA+ and was treated and the other tested PCR-. At least 1 C. difficile test was repeated within 3 months in 48% of EIA+ and 33% of EIA- patients. Based on repeat testing CDI treatment was prescribed for 12% of EIA+ subjects and for 11% of EIA- subjects. For the EIA+ patients, 70%, 19%, 7%, and 4% were classified as probable, possible, unlikely and indeterminate cases of symptomatic CDI when compared with 38%, 34%, 22%, and 5% for EIA- patients. CONCLUSION: During the first 6 months of a 2-step testing algorithm, we found that patients with EIA- test results were frequently treated for CDI and that 72% of EIA- cases were classified as probably or possibly having symptomatic CDI. Further study is needed to determine whether patients with EIA- results categorized with probable or possible symptomatic CDI would improve without CDI treatment. DISCLOSURES: All authors: No reported disclosures.