Primary prophylaxis of gastric variceal bleeding: endoscopic obturation, radiologic intervention, or observation?

BACKGROUND No single effective method has yet been established for the primary prophylaxis of bleeding from gastric varices (GV). METHODS We retrospectively analyzed liver cirrhosis patients with GV who had undergone either endoscopic variceal obturation (EVO) or balloon-occluded retrograde transvenous obliteration (BRTO) as prophylactic treatments, comparing them with those who were observed without any procedural intervention. The endpoints were GV bleeding rate and complete eradication rate. RESULTS 72 patients in EVO, 41 patients in BRTO, and 97 patients in the clinical observation groups were enrolled. No difference was observed in baseline characteristics. As the primary endpoint, 14 (19.4%) patients in the EVO group and 3 (7.3%) in the BRTO group bled from GV after prophylactic treatment, and 34 (35.1%) patients bled in the observation group during the median follow-up of 35 months (p = 0.001). Patients who received EVO or BRTO developed less bleeding from GV than those who received observation only, with no difference between EVO and BRTO (EVO vs. observation, p = 0.038; BRTO vs. observation, p = 0.001; EVO vs. BRTO, p = 0.089). As secondary endpoints, GV disappeared completely in 33 patients (45.8%) in the EVO group and 31 patients (75.6%) in the BRTO group (p = 0.003). By multivariate analysis, complete eradication of GV was the sole determinant for predicting GV bleeding. CONCLUSIONS EVO and BRTO are effective and safe primary prophylactic treatments for preventing bleeding from GV. In particular, BRTO is better than EVO in complete eradication of GV.