OBJECTIVE In about 30% of cases, the etiology of acute recurrent pancreatitis remains unexplained, and the term "idiopathic" is currently used to define such disease. We aimed to evaluate the long-term outcome of patients with idiopathic recurrent pancreatitis who underwent endoscopic cholangiopancreatography (ERCP) followed by either endoscopic biliary (and seldom pancreatic) sphincterotomy or ursodeoxycholic acid (UDCA) treatment, in a prospective follow-up study. METHODS A total of 40 consecutive patients with intact gallbladder entered the study protocol after a 24-month observation period during which at least two episodes of pancreatitis occurred. All patients underwent diagnostic ERCP, followed by biliary or minor papilla sphincterotomy in cases of documented or suspected bile duct microlithiasis and sludge, type 2 sphincter of Oddi dysfunction, or pancreas divisum with dilated dorsal duct. Patients with no definite anatomical or functional abnormalities received long-term treatment with UDCA. After biliary sphincterotomy, patients with further episodes of pancreatitis underwent main pancreatic duct stenting followed by pancreatic sphincterotomy if the stent had proved to be effective. RESULTS ERCP found an underlying cause of pancreatitis in 70% of cases. Patients were followed-up for a period ranging from 27 to 73 months. Effective therapeutic ERCP or UDCA oral treatment proved that occult bile stone disease and type 2 or 3 sphincter of Oddi dysfunction (biliary or pancreatic segment) had been etiological factors in 35 of the 40 cases (87.5%) After therapeutic ERCP or UDCA, only three patients still continued to have episodes of pancreatitis. CONCLUSIONS Diagnostic and therapeutic ERCP and UDCA were effective in 92.5% of our cases, over a long follow-up, indicating that the term "idiopathic" was justified only in a few patients with acute recurrent pancreatitis.