RISK FACTORS FOR THE PERITONITIS ASSOCIATED WITH ENDOSCOPIC ULTRASOUND-GUIDED HEPATICOGASTROSTOMY

Introduction: Endoscopic ultrasound-guided hepaticogastrostomy (EUS-HGS) has been performed as an alternative drainage to endoscopic retrograde cholangiopancreatography (ERCP) for advanced cancer. However, Peritonitis, which is less likely to occur with ERCP, is observed in 3.6-9.4% after EUS-HGS. Peritonitis can lead to prolonged hospitalization and delays the time until the start of chemotherapy.
Aims: The aim of this study was to determine the factors associated with peritonitis after EUS-HGS.
Methods: This retrospective study included 188 consecutive patients who underwent the initial EUS-HGS between July 2016 and November 2022. We evaluated patient characteristics, procedural factors, and early adverse events. Finally, we elucidated the factors affected occurring peritonitis after EUS-HGS. Peritonitis was defined as the patient with fever or the elevation of inflammatory markers in blood test, new-onset abdominal pain, and new fluid collection in the abdominal cavity on computed tomography within 7 days after EUS-HGS resulted in prolonged hospitalization or additional procedures.
Results: Among the 188 patients, 181 malignant biliary obstruction, while 7 had benign conditions. Preprocedural cholangitis was diagnosed in 116 patients, and duodenal invasion was observed in 119 patients. Needle gauges of 19 and 22 were employed in 159 patients and 29 patients, respectively. Biliary duct puncture sites B2 and B3 were selected for 83 and 105 patients, respectively. The median diameter of punctured bile duct was 6 mm with the median distance to hepatic parenchyma of 28 mm. The median bile amount aspirated during procedure was 25 mL. Antegrade stenting with HGS was performed in 57 patients. Metallic stent was placed for HGS in 128 patients. The median procedural time was 32 minutes. The rate of early adverse events was 14%, including peritonitis (7%), pancreatitis (3%), cholangitis (3%), and cholecystitis (1%). Multivariable logistic regression analysis revealed that the diameter of punctured bile duct ≤ 3 mm (odds ratio, 4.2; 95% confidence interval, 1.1-16; p = 0.032) and duodenal invasion (odds ratio, 4.8; 95% confidence interval, 1.0-23; p = 0.049) independently increased the risk of peritonitis after EUS-HGS.
Conclusion: A small diameter of punctured bile duct and duodenal invasion may serve as risk factors of peritonitis after EUS-HGS.