Population-Based Efficacy of Tissue Acquisition for Subepithelial Lesions in the Upper Gastrointestinal Tract

GOALS: To evaluate and compare the diagnostic yield, technical success, and safety of various tissue acquisition techniques for subepithelial lesions (SELs) in the upper gastrointestinal (GI) tract. BACKGROUND: The optimal diagnostic approach for subepithelial lesions (SELs) in the upper gastrointestinal (GI) tract remains uncertain. Although endoscopic ultrasound (EUS) aids evaluation, histologic confirmation is often required, and tissue acquisition techniques vary in efficacy. STUDY: In this retrospective multicenter study from 9 hospitals, patients who underwent EUS-guided or endoscopic biopsies for SELs (1992 to 2024) were included. Techniques assessed included EUS-guided fine-needle aspiration (FNA), fine-needle biopsy (FNB), key-hole biopsy (EUS-KHB), mucosal incision-assisted biopsy (MIAB), and conventional endoscopic biopsies. Diagnostic yield, technical success, and adverse events were assessed. Predictors of diagnostic success were analyzed using multivariable regression. RESULTS: Among 1023 biopsy attempts in 638 patients, the overall diagnostic yield was 31.9%. Lesions >20 mm were more likely to yield a diagnosis (OR: 3.62, 95% CI: 2.65-4.95). EUS-KHB and MIAB outperformed other methods for both small (≤20 mm) and large lesions, with yields of 76.7% versus 67.6% (KHB) and 60% versus 50% (MIAB). In contrast, yields were lower for FNB (63.6% vs. 22.5%), FNA (47.1% vs. 21.7%), and conventional biopsies (24.8% vs. 15.1%). Repeated EUS-FNA/B attempts were associated with reduced success. Adverse events were infrequent (0.5% to 11.1%). CONCLUSIONS: In summary, EUS-FNB is effective for lesions >20 mm, while for smaller lesions (