Endoscopic Ultrasound-Guided Cyanoacrylate Injection To Prevent Rebleeding In Hepatocellular Carcinoma Patients With Variceal Hemorrhage

Background and Aim Secondary prophylaxis (SP) of variceal rebleeding was reported to improve outcomes of hepatocellular carcinoma (HCC) patients, but the optimal endoscopic approach is not well defined. We compared outcomes in HCC patients who underwent SP by endoscopic ultrasound-guided cyanoacrylate obturation (EUS-CYA)versusno SP. Methods Between 2014 and 2018, 30 consecutive patients with inoperable HCC and recent endoscopically controlled variceal bleeding were prospectively recruited. Twenty-seven patients with persistent varices >= 3 mm on endoscopic ultrasound underwent EUS-CYA for SP. Thirty-three HCC patients treated by esophagogastroduodenoscopy-guided CYA obturation (EGD-CYA) alone for acute variceal bleeding between 2009 and 2013 were identified from a prospective gastrointestinal bleed registry as standard of care controls for comparison. Outcome measures were death-adjusted cumulative incidence of rebleeding, bleeding-free survival, technical success, and procedure-related adverse events of EUS-CYA. Results The majority of patients in both groups had advanced HCC, portal vein thrombosis, and Child-Pugh B cirrhosis. EUS-CYA was successful in all 27 patients with no radiographic evidence of cyanoacrylate-lipiodol embolization. Significantly lower 30- and 90-day death-adjusted cumulative incidence of rebleeding (14.8%vs42.4%,P = 0.023 and 18.5%vs60.6%,P = 0.002, respectively) and significantly higher variceal bleeding-free survival at 3 and 6 months (51.9%vs21.2%,P = 0.009, 40.7%vs15.2%,P = 0.010, respectively) were observed in the EUS-CYA group when compared with standard of care group. Conclusions Secondary prophylaxis by EUS-CYA reduced rebleeding rate and improved variceal bleeding-free survival in patients with inoperable HCC and variceal bleeding when compared with no SP. Randomized studies are needed to confirm the benefits of EUS-CYA for this difficult-to-treat population.