Reply

Abstract

Prevention of post-ERCP pancreatitis has been a topic of great interest in recent years because the incidence of post-ERCP pancreatitis remains high (5%–15%) despite technical improvements (N Engl J Med 1996;335:909–918). A recent meta-analysis concluded that somatostatin and gabexate were effective in the prevention of post-ERCP pancreatitis (Gastrointest Endosc 2000;51:1–7). However, the routine use of such agents is not cost-effective because a substantial number of patients need to be treated for 1 patient to benefit from the treatment. The risk factors for post-ERCP pancreatitis are now quite well established. The use of pharmacological agents in selected high-risk cases is an appealing approach, especially if a drug given “on demand” during or after the procedure proves to be effective because a number of conditions associated with post-ERCP pancreatitis are not predictable before the procedure (J Pancreas 2003;4:22–32, 2003;4:33–40). This study by our group showed that somatostatin was effective in preventing pancreatitis when given as a bolus dose after diagnostic ERCP in patients predicted to be at high risk because of the need of therapeutic procedures. Although it is still controversial which type of therapeutic procedures is associated with a high risk, the 13.3% incidence of pancreatitis in the control group did indicate that the study patient population was a high-risk group. During the study period, the incidence of post-ERCP pancreatitis in patients undergoing purely diagnostic ERCP in our unit was 5%. All of the procedures were performed by 6 endoscopists, each with a personal experience of at least 500 ERCP procedures and a yearly case volume of about 80 procedures per endoscopist. The study was performed in a surgical endoscopy unit that deals with mainly choledocholithiasis and malignant biliary obstruction, and hence no patients with sphincter of Oddi dysfunction were enrolled. Pancreatic stenting is an alternative method that has been shown to be effective in preventing post-ERCP pancreatitis in high-risk patients. However, this approach has its own deficiencies. Stent placement following biliary interventions can be difficult. In prospective studies, failure rates ranged from 5% to 10% (JOP 2003;4:58–67). Failure of stenting was associated with a high incidence of pancreatitis (up to 66%), and the resulting pancreatitis could be severe with pancreatic necrosis (Gastrointest Endosc 2004;59:8–14). Furthermore, there is a low but definite risk of complications such as guidewire perforation during pancreatic stenting (Gastroenterology 1998;115:1518–1524). Stents without proximal flaps are associated with the risk of early stent migration, whereas stents with flaps require endoscopic removal at a later date. The need of another endoscopic procedure makes it less attractive in terms of cost-effectiveness. Together with the cost of the stent itself and the extra time involved in pancreatic stenting, it is likely that the use of a single dose of somatostatin “on demand” in high-risk patients may be a more cost-effective approach, although this needs to be testified with a comparative study. Nonetheless, for pharmacological prevention of post-ERCP pancreatitis to “sell” better, further studies are needed to search for a cheaper or more effective agent, and such agent should be studied selectively in high-risk patients with an “on demand” approach similar to the design in our study.