Tu1405 LONG TERM OUTCOMES OF ENDOTHERAPY USING A DEDICATED PANCREATIC BASKET CATHETER FOR PANCREATIC DUCT STONES

Residual pancreatic duct (PD) stone after endoscopic retraction may cause recurrent symptom. Fragmented stones after ESWL, tortuous PD, stricture and dilated branch duct are the causes of residual stones after the endoscopic retraction with conventional basket and balloon catheter. Reforma (Piolax Medical Devices Inc., Japan), a dedicated basket for PD stone retraction with Nitinol fine mesh structure could reduce the incidence of PD stone residue in our previous report ( GIE 2013; 78: 925-929). In this study, we further investigated the long term outcomes of endotherapy using Reforma for PD stones. This is a single center, retrospective analysis of consecutive patients with symptomatic PD stones who underwent the initial endotherapy and achieved complete stone removal on pancreatogram between February 2008 and June 2017. In cases with PD stone larger than 8 mm or refractory to endotherapy underwent extracorporeal shock wave lithotripsy (ESWL). The type of basket catheter was selected at the discretion of the attending physician. Primary outcome was the incidence of pancreatic events due to recurrent PD stones such as acute exacerbation, pancreatic pain and pseudocyst. Pancreatic events were compared between those who were treated using Reforma (Reforma group) and those who were not (Non-Reforma group). Among 103 patients who underwent the initial endotherapy during the study period, complete stone removal on pancreatogram was achieved in 90 patients (87%). The median age was 58 years (range, 27 – 90), and the median size of PD stones was 10mm (range, 3 – 30). ESWL was performed in 91% (82/90). A conventional basket catheter and Refroma were used in 37% (33/90) and 30% (27/90), respectively. Reforma was used as a single device in 9, in combination with a balloon catheter in 18. During the median follow up of 27.7 (range, 0.1 – 97.1) months, the incidence of pancreatic events was 19%: Acute exacerbation in 7%, pancreatic pain in 11% and pseudocyst in 1%. When Reforma group (n = 18) and non-Reforma group (n = 55) were compared, the incidence of pancreatic events was 11% and 24% in the Reforma group and non-Reforma group (P = 0.254). In Cox proportional hazard models, Reforma group tended to be associated with reduction of pancreatic events (Hazard ratio 0.49; P = 0.301). This dedicated pancreatic basket catheter, which can potentially catch all fragmented PD stones better than a conventional one, may reduce the development of pancreatic events after complete PD stone removal. A large scale, prospective study is warranted to confirm our hypothesis.