EUS-Guided Coil Embolization with Absorbable Hemostatic Gelatin Sponge vs. Cyanoacrylate Injection for Treatment of Gastric Varices: Comparison of Clinical Outcomes and Health Care Utilization Presidential Poster Award

Abstract

Introduction: Gastric variceal bleeding is an uncommon complication of cirrhosis but is associated with high rates of mortality and morbidity. Traditional endoscopic treatment options involve cyanoacrylate (CYA) injection which can be technically challenging. Coil embolization and absorbable gelatin sponge (Gelfoam®) are frequently used by interventional radiology for hemostasis. Recently, we have applied this combination therapy in EUS-guided management of gastric varices (GV) with high therapeutic success. We compared the outcome and health care utilization of endoscopic coil embolization with Gelfoam therapy versus CYA injection for treatment of GV.542_A Figure 1. Demographics and Patient Characteristics. Data are presented as number of cases (percentage) OR mean ± S.D. [ ] signify range of values. Abbreviations: NAFLD = non-alcoholic fatty liver disease; MELD = Model for End-stage Liver Disease; IGV = isolated gastric varices; GOV = gastroesophageal varices; UGIB = upper gastrointestinal bleed.542_B Figure 2. Comparison of Primary and Secondary Outcomes. *Two patients passed away prior to adequate follow-up evaluation of need for re-intervention (both due to HCC progression). Data are presented as number of cases (percentage) OR mean ± S.D. [ ] signify range of values. Abbreviations: CYA = cyanoacrylate; BRTO = balloon-occluded retrograde transvenous obliteration; HCC = hepatocellular carcinoma; HD = hemodynamic.Methods:Study Design: Matched cohort study using a prospective registry of over 10 yearsMatching: 5 consecutive patients undergoing EUS coil embolization with Gelfoam therapy for bleeding GV (10/2017-6/2018) were matched in 1:4 fashion to 20 patients who underwent cyanoacrylate injection. Matching criteria included age, Child-Pugh class, variceal type, and bleeding severity on admission. Primary outcome: Need for re-intervention at 3 months Secondary outcomes: All-cause mortality at 3 months, time spent in hospital post procedure, and post-procedural transfusion requirements Statistics: Student t-test and Fisher's exact test Results: 0% (0/5) patients in the coil embolization with Gelfoam cohort required re-intervention at 3-months, as compared to 56% (10/18) in the CYA cohort (p=0.046). Mortality at 3 months was 0% (0/5) for coil with Gelfoam versus 20% (4/20) for CYA (p=0.55), with 1 death directly attributed to ineffective intervention and uncontrolled GV bleeding. Post procedural hospitalization was higher for the CYA cohort (8.6±9.9 vs. 3.4±2.2 days; p=0.26). Post procedure transfusion requirements on average was higher for the CYA cohort (1.95±2.70 vs. 0.2±0.4 units of blood; p=0.17). Conclusion: EUS guided coil embolization with absorbable gelatin sponge therapy is a novel technique that appears to be highly effective in treating bleeding gastric varices. This technique lend itself to effective and sustained hemostasis compared to cyanoacrylate injection which requires re-intervention. There was a trend toward significance with post-procedural time spent in the hospital and transfusion requirements; larger studies in the future should help elucidate additional benefits and risks from this novel approach.