S856 Predictors of Colectomy in Acute Severe Ulcerative Colitis

Abstract

Introduction: Up to 30% of patients admitted with acute severe ulcerative colitis (UC) will require a colectomy despite advances in medical therapies. Our aim was to assess the risk factors for colectomy in patients hospitalized with acute severe ulcerative colitis using initial admission data. Methods: This is a single-center, retrospective study of all patients hospitalized with acute severe UC from 1/1/2012-11/1/2021. Patients who underwent colectomy (during the admission or after discharge) were compared with patients who did not have a colectomy. Continuous variables were analyzed using an unpaired student’s t-test. Categorical variables were analyzed using a Pearson’s chi-square test. Variables with p< 0.05 on univariate analysis were identified and incorporated as independent variables in a multivariate regression model. Results: A total of 168 patients hospitalized with acute severe UC were included (52.4% were male; mean age was 39.8 ±17.2 years). The median disease duration was 2.5 years (IQR 1-10) and 64.9% (n=109) of the cohort were biologic naïve. A total of 31 patients (18.5%) required a colectomy. On univariate analysis, the following factors were associated with a higher risk of colectomy: female sex (70.9% vs 42.3%, p=0.004), lower hemoglobin (10.5 vs 12 g/dL, p=0.002), lower albumin level (3.2 vs 3.5 g/dL, p=0.015). Patients who were on oral corticosteroids on admission had a higher risk of colectomy (70.9% vs 48.9%, p=0.02) while patients who were biologic naïve had a lower risk of colectomy (41.9% vs 70.1%, p=0.003). C-reactive protein level on admission was not associated with risk of colectomy (Table). A multivariate logistic regression model comprising the above variables showed that female sex (p=0.019) and lower hemoglobin (p=0.048) were independently associated with an increased risk of colectomy. Conclusion: In our cohort, approximately one fifth of patients with acute severe UC required a colectomy. Female sex and lower hemoglobin on admission were independently associated with a higher risk of colectomy while biologic naïve patients had a lower risk of colecomy. Further studies to assess if earlier treatment could further lower colectomy rates are needed. Table 1. - Baseline Characteristics Variables Colectomy (n=31) No colectomy (n=137) p-value Patient characteristics Age, mean (SD) 40.5 (19.7) 39.6 (16.7) 0.82 Female, n (%) 22 (70.9) 58 (42.3) 0.004 Disease activity Pancolitis, n (%) 20 (64.5) 97 (70.8) 0.75 Disease duration in years, mean (SD) 6.3 (7.9) 6.9 (9.4) 0.71 Presence of extraintestinal manifestations, n (%) 1 (3.2) 12 (8.8) 0.29 Smoking, n (%) 2 (6.5) 7 (5.1) 0.58 Concomitant CMV infection, n (%) 2 (6.5) 1 (0.7) 0.08 Concomitant C. difficile infection, n (%) 1 (3.2) 8 (5.8) 0.56 Labs at presentation Hemoglobin g/dL, mean (SD) 10.5 (2.2) 12.0 (2.6) 0.002 C-reactive protein md/dL, mean (SD) 85.7 (61.2) 80.4 (84.9) 0.73 Albumin g/dL, mean (SD) 3.2 (0.7) 3.5 (0.6) 0.015 Medications Biologic naïve, n (%) 13 (41.9) 96 (70.1) 0.003 Inpatient infliximab rescue, n (%) 16 (51.6) 51 (37.2) 0.14 On oral corticosteroid at time of admission, n (%) 22 (70.9) 67 (48.9) 0.026 Statins on admission, n (%) 4 (12.9) 9 (6.6) 0.44 Chronic outpatient opioid use, n (%) 3 (9.7) 10 (7.3) 0.65 Inpatient opioid, n (%) 17 (54.8) 72 (52.6) 0.82 On IMM on admission, n (%) 1 (3.2) 13 (9.5) 0.25 SD: standard deviation; CMV: cytomegalovirus; C. difficile: Clostridiodes difficle; IMM: immunomodulator.