What question did this study set out to answer?

June 13, 2026

Endoscopic variceal ligation versus cyanoacrylate injection for acute variceal bleeding from the cardia to the lesser curvature

YCYu‐Fu ChenNational Yang Ming Chiao Tung University TYTsung‐Chieh YangNational Yang Ming Chiao Tung University PHPing‐Han HsiehNational Yang Ming Chiao Tung University

Key Points

The aim is to compare the effectiveness of endoscopic variceal ligation (EVL) and endoscopic cyanoacrylate injection (ECI) for treating acute variceal bleeding from the cardia to the lesser curvature.
Retrospective analysis of 129 patients with acute variceal bleeding treated with EVL or ECI from 2006 to 2024.
Classified varices as type 1 gastroesophageal varices-cardiac varices (GOV1-CVs) or conventional GOV1s and analyzed treatment outcomes using Kaplan–Meier and Cox regression.
Outcomes included 5-day treatment failure, 6-week rebleeding, and 6-week survival.
High hemostasis rates with no significant differences in treatment failure (3.4% EVL vs. 4.3% ECI, p=1.000) or 6-week survival (76.3% EVL vs. 77.1% ECI, p=0.956).
EVL demonstrated lower 6-week rebleeding for GOV1-CVs (9.8% vs. 40.0%, p=0.011) compared to ECI.
Independent predictors of rebleeding included portal vein thrombosis (HR 5.4, p=0.020) and clinically significant bleeding.

Abstract

Background: Current guidelines recommend endoscopic variceal ligation (EVL) and endoscopic cyanoacrylate injection (ECI) for esophageal and fundal variceal bleeding, respectively. However, the optimal strategy for treating acute variceal bleeding (AVB) from the cardia to the lesser curvature remains undefined. This study compared the efficacy of EVL and ECI for AVB at this region. Methods: We retrospectively analyzed patients with AVB from the cardia to the lesser curvature treated with EVL or ECI. Varices were classified as type 1 gastroesophageal varices-cardiac varices (GOV1-CVs; 0–2 cm below the gastroesophageal junction) or conventional GOV1s (2–5 cm below). Outcomes including 5-day treatment failure, 6-week rebleeding, and 6-week survival were compared using Kaplan–Meier analysis. Independent predictors of rebleeding and mortality were identified via Cox regression, and subgroup analyses by variceal location. Results: We enrolled 129 patients (EVL: 59; ECI: 70) between 2006 and 2024. Both modalities achieved high hemostasis rates, with no significant differences in 5-day treatment failure (3.4% vs. 4.3%, p = 1.000), 6-week rebleeding (10.2% vs. 12.9%, p = 0.630), or 6-week survival (76.3% vs. 77.1%, p = 0.956). On multivariable analysis, portal vein thrombosis (HR 5.4, p = 0.020) and clinically significant bleeding (HR 4.4, p = 0.016) predicted 6-week rebleeding, whereas Child–Pugh class C (HR 4.3, p 1.3 mg/dL (HR 2.5, p = 0.024) predicted 6-week mortality. EVL demonstrated lower 6-week rebleeding than ECI for GOV1-CVs (9.8% vs. 40.0%, p = 0.011), but not GOV1s ( p = 0.408). In the GOV1-CVs subgroup, EVL (HR 0.2, p = 0.030) was the only independent predictor of rebleeding. Conclusion: EVL and ECI demonstrated comparable efficacy for AVB from the cardia to the lesser curvature. Notably, EVL is superior to ECI in reducing 6-week rebleeding in GOV1-CVs, supporting EVL as the preferred approach and underscores the importance of precise variceal classification.

Ask AI

Helpful

Bookmark

View Full Paper