The purpose of this clinical trial is to evaluate whether Endoscopic Ultrasound-Guided Selective Variceal Devascularization (EUS-SVD) is effective in preventing the re-bleeding of gastric varices. It will also investigate the safety of EUS-SVD. The main questions it aims to answer are:

Does EUS-SVD lead to a greater reduction in the one-year gastric variceal re-bleeding rate among participants compared to conventional Endoscopic Gastric Variceal Obturation (GVO)?

What medical problems (adverse events) do participants experience when receiving EUS-SVD treatment?

Researchers will compare EUS-SVD with GVO (a currently commonly used first-line treatment method) to see if EUS-SVD can be a superior option for treating gastric varices.

Participants will:

Be randomly assigned to undergo either the EUS-SVD or the GVO procedure.

Receive treatment in the hospital and be hospitalized for observation (approximately 5 days for the EUS-SVD group, 3 days for the GVO group).

Return to the hospital for a follow-up gastroscopy or endoscopic ultrasound examination at 3 months after the procedure.

Receive telephone or outpatient follow-ups at 6 months and 12 months after the procedure to report their health status and any bleeding episodes.

Throughout the study period, record and report any re-bleeding events, need for re-intervention, and other complications.

Inclusion Criteria:

1. Chinese population aged 18-80 years;
2. High-risk gastric varices confirmed by gastroscopy (for primary prevention) or active bleeding due to gastric varices or a history of acute gastric variceal bleeding within the past 1 month (for secondary prevention);
3. Subjects who prefer endoscopic treatment and voluntarily sign the informed consent form.

Exclusion Criteria:

• (1) Subjects with a history of secondary preventive treatment for gastric varices or shunt surgery; (2) Subjects complicated with hepatorenal syndrome, hepatic encephalopathy grade III/IV, or severe jaundice (serum bilirubin >10 mg/dl); (3) Subjects complicated with multiple organ failure or advanced malignant tumors; (4) Subjects who are pregnant, in gestation, or lactating; (5) Subjects suspected of having splenic vein or portal vein thrombosis; (6) Subjects with coagulation dysfunction, platelet count <50,000/mL or INR ≥2; (7) Subjects with esophageal stenosis or a history of esophageal or gastric surgery; (8) Subjects deemed unsuitable for this study by physician evaluation.