A Clinical Prediction Model for Early Colectomy in Patients With Severe Ulcerative Colitis Treated With Tacrolimus

Area of Science:

• Gastroenterology
• Clinical Medicine
• Pharmacology

Background:

• Tacrolimus effectively treats refractory ulcerative colitis (UC).
• Some patients require colectomy despite tacrolimus treatment.
• Early surgical risk assessment is crucial for optimal patient outcomes.

Purpose of the Study:

• Identify predictors of colectomy within 3 months of tacrolimus initiation.
• Develop a clinically applicable prediction model for early colectomy risk.

Main Methods:

• Retrospective analysis of 114 hospitalized severe UC patients treated with tacrolimus (2011-2025).
• LASSO-penalized logistic regression with cross-validation to identify predictive variables.
• Model development using serum albumin, hemoglobin, and age at tacrolimus initiation.

Main Results:

• 21.1% of patients underwent colectomy; 14.0% within 3 months.
• The prediction model showed good discrimination (AUC 0.78).
• High-risk group identified by the model had significantly higher cumulative colectomy rates (p < 0.001).

Conclusions:

• A clinical prediction model accurately estimates early colectomy risk based on baseline factors.
• The model can guide surgical timing and treatment strategy decisions.
• This tool aids in managing patients with refractory ulcerative colitis.