Baseline Ulcerative Colitis Endoscopic Index of Severity and Endoscopic Mayo Score Perform Similarly Predicting Response to Medical Therapy in Patients with Ulcerative Colitis

Introduction: The Ulcerative Colitis Endoscopic Index of Severity (UCEIS) and Endoscopic Mayo Score (EMS) are used as standardized tools to report endoscopic disease activity in ulcerative colitis (UC). The performance of both scores and how they compare as predictors of prognosis has not been completely determined. The purpose of this study was to compare the prognostic capability of UCEIS against the EMS in patients with active UC to determine which scoring system is more accurate in predicting disease outcomes in UC. Methods: This was a retrospective cohort study including UC patients that underwent a baseline colonoscopy/sigmoidoscopy. UCEIS, EMS and partial Mayo score were recorded before and after therapy (in our institution, all scores are collected prospectively at the time the patient undergoes a colonoscopy/sigmoidoscopy). Phenotype and therapy started for UC, laboratory chemistries, C-reactive protein (CRP) and stool calprotectin were also collected. Primary outcomes consisted of non-response to therapy after six months, defined as a need for colectomy, switching therapy, or starting steroids. Secondary outcomes such as emergency department (ED) visits and hospital admissions were also investigated. Endoscopic remission was defined as EMS of 0/UCEIS≤1. Clinical remission was defined as a partial Mayo score < 2. Results: 105 patients met inclusion criteria. 52 (49.5%) were female and the mean age was 46 years (SD=16.6). 36 patients had left-sided colitis, 10 patients had proctitis, and 59 patients had pan-colitis. 17 patients were starting new therapy. At six months, 30 patients required steroids for flares. Both UCEIS and EMS predicted response to therapy well. For steroid-free remission at 6 months, AUC for UCEIS was 0.79 (p-value <.0001) and AUC = 0.77 (p-value <.0001) for EMS. Within the UCEIS sub-scores, the AUC was 0.76 (p-value <0.001) for vascular pattern, 0.67 for bleeding (p-value =0.004) and 0.74 for erosions/ulcers (p-value <0.001). Conversely, UCEIS and EMS poorly predicted ED visits or admissions. ROC curves for 6 months showed UCEIS scoring had an AUC = 0.57 for UCEIS (p-value =0.42) and 0.57 for EMS (p-value =0.25). Conclusion: Our study suggests that the UCEIS and EMS are equivalent when used as prognostic predictive factors in UC patients. However, when comparing scores and their role in clinical practice, we need to consider other potential variables such as inter and intra-observer variability.602_A Figure 1. Baseline characteristics of the study population602_B Figure 2. Performance of endoscopic and clinical scores predicting steroid free remissions within 6 months of baseline colonoscopy602_C Figure 3. 6 month outcomes