A Randomized Controlled Trial of Valdecoxib and Glyceryl Trinitrate for the Prevention of Post-ERCP Pancreatitis

Abstract

Efforts to prevent post-endoscopic retrograde cholangiopancreatography (ERCP) pancreatitis have been largely unsuccessful. Cyclo-oxygenase-2 enzyme-mediated inflammatory pathway has been suggested in the pathophysiology of acute pancreatitis. Glyceryl trinitrate (GTN) might prevent post-ERCP pancreatitis by relaxing the sphincter of Oddi. To evaluate the efficacy of valdecoxib, a cyclo-oxygenase-2 inhibitor, and GTN transdermal patch for the prevention of post-ERCP pancreatitis. Patients undergoing first ERCP procedure from October 2003 to August 2005 were randomized to receive either 20 mg intravenous valdecoxib or GTN patch (10 mg/h) at the start of ERCP, or assigned to control group. The study followed CONSORT guidelines. Primary outcome measure was frequency of post-ERCP pancreatitis in the 3 groups. A total of 380 patients were randomized; 121 patients in valdecoxib (group 1), 124 in GTN (group 2), and 126 in the control arm (group 3) were analyzed. There was no difference in the frequency of post-ERCP pancreatitis between the groups (12 each in groups 1 and 2, and 13 in group 3; P=0.986). None of the patients had severe pancreatitis. The frequency of post-ERCP pain and amylase levels were also similar in the 3 groups (P=0.769 and P=0.947, respectively). Pancreatic duct cannulation, cholecystectomy, difficult cannulation, and pre-cut were risk factors for pancreatitis on univariate analysis. On multivariate analysis, pancreatic duct cannulation was the only independent risk factor for pancreatitis (P≤0.001; odds ratio 5.67; 95% confidence interval: 2.76-11.63). Valdecoxib and GTN were not effective for the prevention of post-ERCP pancreatitis.