BACKGROUND AND AIM: Even though gastric varices (GV) are less frequently encountered than esophageal varices, they are associated with severe bleeding. We aim to target the feeder vessel or afferent component displaying novel therapeutic approach comprising a combined application of cyanoacrylate (CYA) and coils. This approach intended to address the vascular pathology of GV with enhanced precision and efficacy.
METHODS: A single-center, prospective registry was conducted during July/2015-October/2023. Enrolled patients included adults ≥18 years old considered for primary or secondary prophylaxis of gastro-esophageal varices (GOV) II and isolated GV (IGV) I. First, the location of the feeder vessel was determined by angiography using a 10 ml water-soluble contrast. Subsequently, using a FNA needle, coils were EUS-guided deployed (10-20 mm coiled diameter, 14-20 cm straight lengths, 0.035 inches diameter), followed by CYA injection (2-Octyl-CYA). The number of deployed coils and injected CYA was documented. Outcomes were technical success, immediate disappearance, complete obliteration, recurrence rate, recurrence free survival (RFS) and recurrence management.
RESULTS: Over eight-years, 167 patients underwent EUS-guided coils and CYA: 49.7% females, median age of 63 years, 19% alcohol-related cirrhosis (Table 1). Child Pugh B and C was estimated in 65.9% and 18% of patients, respectively. Most common previous therapies included B-blockers (32.3%), band ligation (26.9%), and CYA (21.0%). GV treatment was intended as secondary prophylaxis in 88.6%. More than one feeder vessel was targeted in 38.9%. In 67.7% of cases up to three coils were required, with a median injection of 3.6 ml of CYA. A 100% technical success was achieved. Among the 167 patients, 8.9% experienced recurrence, with a median RFS of 4.9 months. When comparing non-recurrence vs recurrence patients, immediate disappearance was noticed in 69.7% vs 40.0% with white light endoscopy (p=0.039) and complete obliteration in 90.1% vs 73.3% during EUS (p=0.072), respectively. Immediate disappearance was a significant predictive factor for non-recurrence (OR 0.29; IC 95% 0.09 – 0.85; p=.026), with a 95% positive predictive value. All recurrences were managed with EUS-guided coils and CYA (9/15) and clips and CYA (6/15), with a 12/15 non-recurrence rate (Table 2). On multivariate analysis, a lower recurrence was associated with previous B-blockers therapy (OR 0.18; 95% CI 0.02 to 0.78; p=.045), while a previous CYA attempt represented a high-risk factor for recurrence (OR 10.2; 95% CI 1.08 to 110; p=.044) (Table 3).
CONCLUSION: EUS-guided coils and CYA in feeder vessels have shown high technical success and low recurrence rate in treating IGV I and GOV II. CYA attempts should be misadvised in GV. Non-immediate disappearance requires shorter follow-ups to assess for early recurrence.

Baseline characteristics

Factors associated with risk of GV recurrence.

Federico Ferber-Reyes

Appearances