THE SAFETY OF SPHINCTEROTOMY IN ERCP AMONG PATIENTS WITH CIRRHOSIS: INSIGHTS FROM A QUATERNARY CARE TRANSPLANT CENTER.

Introduction
Endoscopic Retrograde Cholangiopancreatography (ERCP) is a mainly therapeutic procedure used for a variety of biliary and pancreatic diseases. Patients with cirrhosis may have a coagulopathic state as well as antithrombotic proteins. There is paucity of literature regarding safety of sphincterotomy in patients with cirrhosis. We aim to study the safety of sphincterotomy during ERCP among patients with cirrhosis.

Methods
We conducted a retrospective chart review on all adult (18 years and older) patients with cirrhosis who underwent ERCP at our center between 1/2015 and 11/2023. Patients with post-liver transplant status were excluded from our analysis. Data including demographics, markers of liver function, procedure indication, details including specific procedural techniques and tools, adverse events (AE), hepatic decompensations (HD) and morbidity were collected.

Results
A total of 277 patients with cirrhosis who underwent ERCP were included, 181 (65.3%) were males, 183 (66.1%) were Caucasian. Alcohol was the most common etiology of cirrhosis in 108 (38.9%) patients. The most common indication for ERCP was jaundice in 99 (35.7%) patients. Average MELD for all patients was (19.15, SD 8.4) at time of ERCP. 197 (71.1%) patients underwent a sphincterotomy, 175 (88.8%) patients underwent biliary sphincterotomy only, 16 (8.1%) patients received pancreatic sphincterotomy and 6/197 (3.0%) patients received both. A total of 43 (21.8%) patients developed immediate post sphincterotomy bleeding. There was no significant increased risk of procedural AE or HD for patients who underwent biliary compared to pancreatic sphincterotomy or both (P=0.651). There was also no increased risk of AE or HD based on length of sphincterotomy (P=0.322). There was no increased risk of AE or HD for patients who developed post-sphincterotomy bleeding compared to those who did not (P=0.111). Interestingly, patients who were given any blood products to correct coagulopathy or anemia prior to ERCP were noted to have higher rates of complications post-ERCP (13.3% vs. 3.5%, P=0.002). The most common complications included: bleeding (6/17), hepatorenal syndrome (3/17), hepatic encephalopathy (3/17), sepsis or cholangitis (2/17), ascites or SBP (2/17).

Conclusion:
Our study suggests that performing sphincterotomy in patients with cirrhosis, especially with an elevated MELD, can result in nearly a 1 in 5 risk of bleeding. Despite this bleeding risk hepatic decompensation was not noted following sphincterotomy. Further prospective studies are needed to validate these findings, but patients should be educated regarding the increased bleeding risk following sphincterotomy in cirrhosis.