Safety of ERCP and EUS Procedures in Cirrhotic Patients

Abstract

Introduction: Therapeutic ERCP and diagnostic EUS procedures are associated with known risks. These risks may be greater in patients with hepatic cirrhosis given the sequelae associated with decreased hepatic synthetic function. We sought to examine whether the severity of cirrhosis was associated with greater incidence of adverse events. Methods: We performed a retrospective study of all patients diagnosed with hepatic cirrhosis who underwent ERCP or EUS-guided FNA/FNB from July 2016 to April 2018 at our safety net hospital. We reviewed the etiology for cirrhosis, Child-Pugh and MELD-Na scores at the time of procedure, indications for procedure, interventions completed, and 30-day post-procedural adverse events. Statistical analysis was done to assess whether Child-Pugh class and MELD-Na score were associated with greater adverse event rates and whether advanced techniques, defined as electrohydraulic lithotripsy (EHL), laser lithotripsy, or needle-knife techniques, were associated with higher rates of complications. Results: 50 procedures (40 ERCP, 10 EUS-FNA/FNB) performed on 22 patients were included. 18 cases were classified as Child's A, 22 cases as Child's B, and 10 cases as Child's C. Mean MELD-Na score was 14.1. Cirrhosis was documented as being secondary to pyogenic cholangitis (n=11), NASH (n=9), malignancy (n=2), alcoholism (n=20), HBV (n=9), HCV (n=4), PBC (n=1), AIH (n=2), and cryptogenic (n=2). Interventions included EHL (n=11), stent placement/removal (n=31), stone removal (n=24), Spyglass cholangioscopy (n=18), spincterotomy (n=11), laser lithotripsy (n=4), FNA/FNB biopsies (n=7), liver biopsy (n=3), needle-knife techniques (n=6), and papillotomy (n=2). 30-day procedurerelated adverse events occurred in 12 cases and included fever (n=3), upper GI bleed (n=4), bacteremia (n=3), septic shock (n=1), bile leak (n=1), stent migration (n=1), retroperitoneal hematoma (n=1), and death (n=3). Both Child-Pugh class and MELD-Na score were significant predictors of adverse events (p=0.045, p=0.011 respectively). Advanced techniques were not associated with increased event rates (p>0.05). Conclusion: Consistent with established literature, our study demonstrates that severity of cirrhosis, as defined by Child-Pugh class and MELD-Na score, are associated with increased rates of adverse events in invasive procedures. However, the rate of adverse events appears to be unaffected by the use of advanced techniques with ERCP/EUS procedures among cirrhotic patients overall.