Answers are grounded in this paper and cite their sources.
BACKGROUND: Accurate endoscopic diagnosis of colorectal lesions is essential for clinical management. This prospective multicenter study assessed the impact of magnification endoscopy on real-time decision-making for colorectal lesions. METHODS: We included 419 patients with 876 colorectal polyps and assessed using three approaches: single (white light endoscopy, WLE), dual (WLE + narrow-band imaging with magnification endoscopy, NBIME), and triple (WLE + NBIME + pit pattern analysis using magnification chromoendoscopy or acetic acid-enhanced NBIME). Expert endoscopists provided histological predictions and confidence. Predictions were further classified into three management categories: surveillance (hyperplastic polyps), endoscopic resection (conventional adenomas, sessile serrated lesions [SSLs], traditional serrated adenomas [TSAs], T1a carcinoma), and surgical resection (T1b carcinoma). RESULTS: Lesions included 24 hyperplastic polyps, 654 conventional adenomas, 85 SSL/TSAs, 90 T1a carcinomas, and 23 T1b carcinomas. Histological prediction accuracy was 84.9%, 85.3%, and 87.4% (p = 0.35), and clinical diagnosis accuracy was 94.6%, 95.5%, and 96.5% (p = 0.17) for single, dual, and triple modalities, respectively, showing numerically higher values with additional modalities. High-confidence rates significantly increased with additional modalities (73.4%, 80.0%, 83.6%; p < 0.05), correlating with improved accuracy. In high-confidence cases, histological prediction accuracy was 93.3%, 92.7%, and 93.2%, compared to 66.5%, 60.6%, and 67.4% in low-confidence cases (p < 0.05). A similar trend was observed for clinical diagnosis, with accuracy of 98.3%, 97.9%, and 97.8% versus 85.0%, 85.7%, and 89.6% (p < 0.05). Confidence gain was associated with larger size in both single-dual and dual-triple transitions (p < 0.05), particularly ≥ 10 mm lesions. CONCLUSION: Magnification endoscopy has minimal impact on diagnostic accuracy but significantly enhances diagnostic confidence, aiding clinical decisions.
No takes yet. Share an insight, caveat, or question.
Shibagaki et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: