P3291 - Intravenous Steroid Duration as an Independent Predictor of Colectomy in Acute Severe Ulcerative Colitis

Eleazar E.. Montalvan-Sanchez, MD¹, Shreyak Sharma, MD¹, Dalton A. Norwood, MD², Alexander N.. Levy, MD¹
1Yale University School of Medicine, New Haven, CT; 2University of Alabama at Birmingham, Birmingham, AL

Introduction: Acute severe ulcerative colitis (ASUC) often requires hospitalization and timely escalation of therapy to avoid colectomy. While IV corticosteroids remain the first-line treatment, prolonged use without response may reflect delays in identifying steroid-refractory disease or in initiating advanced IBD therapies. We aimed to evaluate whether the duration of IV steroid use independently predicts colectomy in hospitalized ASUC patients.

Methods: This single-center retrospective study included patients hospitalized with acute severe UC from 1/1/15-8/30/23. Patients who underwent colectomy were compared to those who did not. Logistic regression identified predictors of colectomy. Malnutrition was defined as >5% weight loss over the last 6 months; labs were dichotomized using institutional cut-offs. IV steroid duration was evaluated both continuously and using a 5-day clinical cut-off. Variables with clinical relevance or univariate significance were included in stepwise selection.

Results: A total of 212 patients hospitalized with ASUC were included, of whom 48 (22.6%) underwent colectomy. As shown in Table 1, colectomy was associated with lower admission hemoglobin (11.0 vs 11.9 g/dL, p=0.037), higher BMI (27.2 vs 24.6, p=0.045), greater likelihood of IV steroid use ≥5 days (58.3% vs 30.6%, p< 0.001), and TPN requirement (10.4% vs 3.7%, p=0.063). Patients requiring colectomy were less likely to be biologic-naïve (43.8% vs 69.5%, p=0.001) and more likely to be non-responders to infliximab (59% vs 3%, p< 0.001). In multivariable logistic regression models (Table 2), Anemia (OR 4.62, 95% CI 1.66–12.85), TPN use (OR 6.05, 95% CI 1.34–27.36), and IV steroid duration ≥5 days (OR 2.95, 95% CI 1.30–6.69) were independently associated with colectomy. Similar results were observed when IV steroid duration was analyzed as a continuous variable. Biologic-naïve status was associated with reduced odds of colectomy but did not reach statistical significance. BMI and abnormal albumin were not independent predictors in either model.

Discussion: Prolonged IV steroid use (≥5 days), anemia on admission, and TPN requirement were independently associated with increased colectomy risk in hospitalized ASUC patients. Delays in identifying nonresponse to steroids and in escalating to advanced therapies may contribute to worse outcomes. These findings support the importance of early risk stratification and timely therapeutic decision-making in ASUC.


Figure: Table 1 summarizes baseline demographic and clinical characteristics of hospitalized patients with acute severe ulcerative colitis (ASUC), stratified by colectomy status. Continuous variables are presented as median (interquartile range) or mean ± standard deviation, and categorical variables as number (percentage). P-values were calculated using appropriate statistical tests (chi-square or t-test) based on variable type.
Abbreviations: BMI = body mass index; CRP = C-reactive protein; IV = intravenous; TPN = total parenteral nutrition.



Figure: Table 2 presents multivariable logistic regression models evaluating predictors of colectomy in patients hospitalized with acute severe ulcerative colitis (ASUC). Model A includes intravenous (IV) steroid exposure as a categorical variable (≥5 days vs. <5 days), while Model B treats steroid exposure as a continuous variable (in days). Odds ratios (OR) with 95% confidence intervals (CI) are shown for each covariate. Both models adjust for demographics, disease severity markers, treatment response, and nutritional status.

Abbreviations: OR = odds ratio; CI = confidence interval; ASUC = acute severe ulcerative colitis; IV = intravenous; TPN = total parenteral nutrition.


Disclosures:
Eleazar Montalvan-Sanchez indicated no relevant financial relationships.
Shreyak Sharma indicated no relevant financial relationships.
Dalton Norwood indicated no relevant financial relationships.
Alexander Levy: BMS – Consultant. Johnson & Johnson – Consultant.

Eleazar E.. Montalvan-Sanchez, MD¹, Shreyak Sharma, MD¹, Dalton A. Norwood, MD², Alexander N.. Levy, MD¹. P3291 - Intravenous Steroid Duration as an Independent Predictor of Colectomy in Acute Severe Ulcerative Colitis, ACG 2025 Annual Scientific Meeting Abstracts. Phoenix, AZ: American College of Gastroenterology.