Related Experiment Video
Updated: Aug 18, 2025
05:07
Multimodality Diagnosis of Mesenteric Ischemia
Published on: July 21, 2023
687
"MURAL" model to predict bleeding from mural-based lesions in potential small bowel bleeding may improve diagnostic
Julajak Limsrivilai1, Thanaboon Chaemsupaphan1, Sipawath Khamplod1
1Division of Gastroenterology, Department of Medicine, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok, Thailand.
Medicine
|December 9, 2022
Summary
A new "MURAL" model predicts small bowel bleeding from mural-based lesions, improving diagnostic decisions. This tool enhances efficiency and lowers costs compared to traditional methods like video capsule endoscopy.
Area of Science:
- Gastroenterology
- Medical Imaging
- Health Economics
Background:
- Small bowel bleeding (SBB) diagnosis is challenging, with video capsule endoscopy (VCE) excelling at mucosal lesions and computed tomography enterography (CTE) at mural-based lesions.
- Current diagnostic strategies lack a clear method to differentiate lesion types, potentially leading to suboptimal investigation choices.
Purpose of the Study:
- To develop and validate a predictive tool, the "MURAL" model, to identify patients with SBB originating from mural-based lesions.
- To assess the cost-effectiveness of integrating the MURAL model into diagnostic strategies for SBB.
Main Methods:
- Retrospective development and validation of a logistic regression model (MURAL) using patient data.
- The model incorporates parameters including age, atherosclerosis, chronic kidney disease, antiplatelet use, and serum albumin.
- Cost-effectiveness analysis comparing diagnostic strategies: VCE, CTE, and MURAL model.
Main Results:
- The MURAL model demonstrated strong predictive performance with an area under the receiver operating characteristic curve of 0.778 (derivation) and 0.821 (validation).
- At a 24.2% cutoff, the model achieved 70% sensitivity and 83% specificity for detecting mural-based lesions in the validation cohort.
- Cost-effectiveness analysis showed the MURAL model strategy had a comparable missed lesion rate, a lower missed tumor rate, and reduced costs versus the VCE strategy.
Conclusions:
- The MURAL model effectively predicts SBB from mural-based lesions, aiding in the selection of appropriate diagnostic investigations.
- Implementing the MURAL model can enhance diagnostic efficiency and reduce healthcare costs in managing SBB.
- This predictive tool offers valuable guidance for clinical decision-making in SBB workup.
