754. A Two-step Testing Algorithm for Hospital-onset Clostridioides difficile Infection (CDI) Reduces Prescribing of C. difficile (CD) Therapy but Its Ability to Guide Treatment Decisions Remains Unclear

Abstract Background Determining true CDI versus CD colonization through CD testing is a continuing challenge. A previously introduced decision support tool at UVA Health significantly reduced inappropriate testing without adverse outcomes. More recently, our methodology changed from nucleic acid amplification test (NAAT) alone to an initial NAAT followed by ELISA for toxin to improve specificity. The purpose of this analysis was to assess provider interpretation of test results, using targeted CD therapy as a surrogate. Methods This single-center, retrospective study evaluated all patients with a positive NAAT (Cepheid Xpert® C. difficile) on day 4 or later of hospitalization following 2-step algorithm implementation from Feb 2020 through Feb 2021. Toxin negative (TOX-) test results (C. DIFF QUIK CHEK COMPLETE®) were accompanied by a comment that discordance may represent colonization or CDI and to consider ID consult. The proportion of toxin positive (TOX+) versus TOX- patients receiving ≥ 1 dose of CD therapy served as the primary outcome with partial courses considered < 10 days. Clinical outcomes were also compared. Results Ninety patients with NAAT+ results were included, of whom 58 (64%) were TOX-. Thirty-two (100%) TOX+ (median days of therapy [IQR] = 14 [11-17]) versus 51 (88%) TOX- patients (median days of therapy [IQR] = 11 [7-14]) received CD therapy (p=0.04). Treatment decisions were guided by ID physicians for 32 (63%) TOX- patients; ID recommendations to discontinue CD therapy were followed in 2 out of 9 (22%) cases. TOX- patients received partial therapy due to patient death (n=5), presumptive colonization (n=3), and provider error (n=1). Of TOX- patients receiving partial or no treatment, there were no CDI-related adverse outcomes during the admission. CDI-related colectomy occurred in 2 (6%) and 1 (2%) TOX+ and TOX- patients, respectively. Five in-hospital deaths with CDI as a contributing factor occurred in the TOX+ group. Conclusion Adoption of a 2-step NAAT plus toxin testing algorithm for hospital-onset CDI reduced the frequency with which TOX- patients received CD therapy but the vast majority were still treated. Most providers considered a positive NAAT indicative of CDI, regardless of TOX status. Disclosures All Authors: No reported disclosures

Conclusion. Despite the COVID-19 pandemic, the HO-CDI and C. diff order rates and overall SAAR decreased; however, antibiotic utilization increased in the ICU during the COVID-19 waves. The overall decrease may be multifactorial and related to increased hand hygiene compliance, isolation and personal protective equipment use and overall decreased antibiotic use and C. diff orders.
Disclosures. Rachel Kenney, PharmD, Medtronic, Inc. (Other Financial or Material Support, spouse is an employee and shareholder) Background. Determining true CDI versus CD colonization through CD testing is a continuing challenge. A previously introduced decision support tool at UVA Health significantly reduced inappropriate testing without adverse outcomes. More recently, our methodology changed from nucleic acid amplification test (NAAT) alone to an initial NAAT followed by ELISA for toxin to improve specificity. The purpose of this analysis was to assess provider interpretation of test results, using targeted CD therapy as a surrogate.
Methods. This single-center, retrospective study evaluated all patients with a positive NAAT (Cepheid Xpert® C. difficile) on day 4 or later of hospitalization following 2-step algorithm implementation from Feb 2020 through Feb 2021. Toxin negative (TOX-) test results (C. DIFF QUIK CHEK COMPLETE®) were accompanied by a comment that discordance may represent colonization or CDI and to consider ID consult. The proportion of toxin positive (TOX+) versus TOX-patients receiving ≥ 1 dose of CD therapy served as the primary outcome with partial courses considered < 10 days. Clinical outcomes were also compared.
Conclusion. Adoption of a 2-step NAAT plus toxin testing algorithm for hospital-onset CDI reduced the frequency with which TOX-patients received CD therapy but the vast majority were still treated. Most providers considered a positive NAAT indicative of CDI, regardless of TOX status. Background. Clostridioides difficile infection (CDI) rates have plateaued at historical highs in the United States since 2010 and remains a major health problem. While optimal CDI testing remains unclear, current literature recommends testing patients whose symptoms are not clinically attributable to underlying conditions, e.g., laxatives. At Moffitt Cancer Center, a soft-stop alert was implemented to alert the provider if the patient received a laxative within the previous 48 hours of CDI testing. We aim to evaluate the incidence of CDI rates with prior laxative use in immunocompromised patients, as well as, the impact of the soft-stop alert in reducing CDI testing.

Clostridioides difficile Testing and the Use of Laxatives in Immunocompromised Adults
Methods. Retrospective, single-center, review of adult patients who were tested for CDI after the implementation of the soft-stop alert from October 1, 2020 to December 21, 2020. These patients were compared to a historical cohort of patients who were tested for CDI prior to the alert implementation from October 1, 2019 to December 31, 2019. The primary outcome was the percent of patients that received a laxative within 48 hours of CDI testing pre-alert compared to post-alert. Secondary outcomes included the percent of colonization versus active infection in this immunocompromised population, number and type of laxatives administered prior to testing, and the frequency of alert and reduction of CDI tests ordered. A cost-benefit analysis was also performed.
Results. In the historical cohort (n=480), 14.8% received a laxative within 48 hours of CDI testing (Figure 1). Within patients who received a laxative in this group, 4.2% had a definitive active infection. After the alert was implemented, a total of 630 CDI tests were ordered from October 1, 2020 to December 31, 2020, and the alert was fired for 123 (19.5%) tests ordered (Figure 2). Of the tests where the alert was fired, the CDI test was removed for 42.3% and continued for 57.7% of orders resulting in savings of $3,263. In this cohort, 5.6% of patients had a definitive active CDI infection who received a laxative and testing was continued (Figure 3).