P684 Multiplex Gastrointestinal Pathogen Panel Testing is Associated with Higher Rates of Inflammatory Bowel Disease Therapy Escalation in Patients Hospitalised with Flare

Background Although the role of Clostridioides difficile infection (CDI) testing in inflammatory bowel disease (IBD) flare is well established, the function of additionally testing for non-CDI enteric infections (EI) via multiplex gastrointestinal pathogen polymerase chain reaction (GI panel) stool tests remains unclear. Patients with PCR-confirmed EI are less likely to have IBD therapy escalated; however, it is unknown whether performing testing itself affects IBD-related decision-making and outcomes. This is vital given the similar clinical presentations of flare and EI. We examined differences in IBD outcomes among patients who – in addition to testing for CDI – were and were not co-tested for EI during hospitalisation for flare. Methods We conducted a retrospective cohort study of IBD patients hospitalised with flare and tested for CDI at an urban academic medical centre. We collected data on demographics, IBD disease severity, IBD therapy, and antibiotic therapy. Patients were cohorted by co-testing for EI using a GI panel. The primary outcome was escalation of IBD therapy within 24 hours of emergency department presentation. Each IBD therapy was also studied separately. Secondary outcomes were antibiotic therapy, length of stay, and colectomy. Results Of 134 patients, 66 (49.3%) were co-tested and 7 (10.6%) had an EI detected. Disease severity and CDI rates were comparable between groups (Table 1). Co-tested patients were more likely to receive a new IBD therapy (92.4% vs. 75.0%, p=0.006), specifically intravenous (IV) steroids (89.4% vs. 36.8%, p=<0.001; Table 2). Patients who were not co-tested were more likely to have their existing IBD therapy continued on admission. Co-tested patients were less likely to undergo colonoscopy during admission and had lower rates of prolonged antibiotic therapy (>10 days; 4.5% vs. 14.7%, p=0.047), although prolonged antibiotic use was comparable between groups after excluding patients treated for CDI (5.1% co-tested vs. 1.8% not co-tested, p=0.276). Colectomy rates and length of stay were comparable between groups. Conclusion Co-testing for non-CDI enteric pathogens by GI panel was associated with a greater degree of inpatient IBD flare management – especially IV steroids. Patients not co-tested were more likely to stay on their existing IBD therapy regimen despite similar flare severities at presentation. Given the low pathogen detection rates, early negative testing for EI may promote provider confidence in intensifying IBD therapy. Early clinical improvement after therapy escalation may obviate the need for colonoscopy in these patients. Use of the GI panel may affect IBD decision-making during relapse, and should be considered in all IBD patients hospitalised with flare.