Endoscopic Differentiation of Cytomegalovirus Lesions in Patients with Inflammatory Bowel Disease and Prospects for Diagnostic Model Development
DOI:
https://doi.org/10.67076/cmj.2026.v4n2.03Keywords:
inflammatory bowel disease, ulcerative colitis, Crohn’s disease, cytomegalovirus, CMV colitis, tissue PCR, endoscopy, targeted biopsy, diagnostic model, endoscopic classificationAbstract
Cytomegalovirus (CMV) reactivation is a clinically significant complication in patients with inflammatory bowel disease (IBD), particularly in those with severe, steroid-refractory, or immunosuppression-associated disease. Because the clinical and endoscopic manifestations of CMV-associated intestinal involvement frequently overlap with active ulcerative colitis (UC) and Crohn’s disease (CD), timely diagnosis remains challenging. Although tissue polymerase chain reaction (tissue PCR) and immunohistochemistry are considered reference diagnostic methods, standardized endoscopic criteria for selecting patients who require targeted biopsy are lacking.
Aim: To identify endoscopic features associated with CMV-positive intestinal lesions in patients with IBD and to develop an integrated endoscopic diagnostic model for optimizing targeted biopsy and molecular confirmation.
Methods: This retrospective–prospective observational study included 198 adults with confirmed UC or CD who underwent ileocolonoscopy with targeted mucosal biopsy. CMV DNA was detected by tissue PCR in intestinal biopsy specimens, and patients were classified as CMV-positive (n = 72) or CMV-negative (n = 126). Ten predefined endoscopic characteristics were analyzed using univariate logistic regression with calculation of odds ratios (ORs), 95% confidence intervals (CIs), and p-values. Based on combinations of significant morphological findings, a practical endoscopic classification and a diagnostic algorithm were developed.
Results: Several endoscopic findings demonstrated a strong association with CMV-positive tissue PCR. The highest diagnostic value was observed for relatively preserved periulcer mucosa (OR 147.47; 95% CI 41.67–521.92; p<0.001). Other significant predictors included linear ulcers (OR 17.87), punched-out ulcers (OR 14.93), spontaneous bleeding (OR 12.34), deep ulcers (OR 9.96), and dense white fibrin deposits (OR 7.80) (all p<0.001). Geographic ulcers, multiple ulcers, and necrotic lesions also showed significant associations, whereas undermined ulcer margins had limited diagnostic utility. Importantly, no single endoscopic feature was pathognomonic; diagnostic performance improved substantially when morphological findings were interpreted as integrated patterns rather than isolated observations. These findings formed the basis for a four-pattern morphological classification and a practical biopsy-guided diagnostic algorithm.
Conclusions: CMV-associated intestinal involvement in IBD is characterized by reproducible combinations of endoscopic features rather than by a single specific lesion. A structured assessment integrating ulcer morphology, periulcer mucosal appearance, fibrin deposition, and lesion distribution may improve identification of patients requiring tissue PCR confirmation and facilitate earlier diagnosis of CMV-associated disease. The proposed classification and diagnostic algorithm provide a practical framework for endoscopic decision-making and warrant prospective multicenter validation.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Azerbaijan Gastroenterology and Invasive Endoscopy Society

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.