Abstract Background On 12/11/23, our academic medical center changed from primarily GDH antigen/toxin A/B combination testing with reflex to C. difficile PCR in discordant samples (Figure 1) to PCR (either direct or from a multiplex gastrointestinal pathogen panel (GIPP)) as initial screening with positive specimens reflexed to antigen/toxin testing (Figure 2). After the algorithm change, we evaluated treatment rates for patients with positive PCR results who tested negative for C. difficile toxin. Methods New guidance for C. difficile infection (CDI) was posted on the Antimicrobial Stewardship website and education was provided to clinicians on the algorithm. Data was collected for inpatients ≥ 19 years old who were toxin negative, but antigen and PCR test positive for C. difficile from 10/1/22-3/31/24. GIPP could be ordered first during the entire period, but only in the first 3 hospital days. Treatment and recurrence rates for patients having a PCR test first were analyzed compared to rates for those with initial antigen/toxin tests. Treatment was counted if initiated within 2 days of test and continued ≥ 5 days, according to proposed NHSN definitions, and included oral vancomycin, fidaxomicin, or metronidazole if vancomycin was contraindicated. Based on estimated treatment rates, a sample size of 177 patients was targeted to detect a change from 80% to 60% treatment. Results Out of 179 discordant CDI results prior to the testing change, treatment occurred in 151 (85.4%) cases. After the change, treatment was prescribed 26 out of 46 (56.5%) times, a 34% reduction (p< 0.01). Baseline characteristics were similar in each group, and 56.9% of tests were ordered in the first 3 days of hospital stay. Overall, when an antigen test was performed first and a positive PCR reported last, treatment occurred in 84 of 91 (92.3%) cases. When a PCR test was performed first and a negative toxin resulted last, treatment was prescribed in 93 of 134 (69.4%) cases (p< 0.01). Treatment was the same in each group (99% vancomycin). Recurrence rates were similar regardless of testing order, 5.4% with antigen first vs. 4.4% with PCR test first. Conclusion Treatment rates for patients with discordant CDI results were significantly lower after the order of testing changed and when a negative toxin result was presented last. Disclosures Trevor C. Van Schooneveld, MD, FSHEA, FIDSA, AN2 Therapeutics: Grant/Research Support|BioMerieux: Grant/Research Support|BioMerieux: Honoraria|Thermo-Fisher: Honoraria
Read full abstract