Preoperative endoscopic stent placement before pancreaticoduodenectomy: A meta-analysis of the effect on morbidity and mortality

Abstract

Background: Pancreaticoduodenectomy is the only potentially curative treatment for peripapillary pancreatic tumors. However, postoperative morbidity and mortality are high, and different approaches have been tried to improve results, such as preoperative biliary drainage in patients with jaundice. This meta-analysis investigated the effect on postoperative outcome of preoperative biliary drainage by endoscopic biliary stent placement in patients who are jaundiced and who have peripapillary pancreatic tumors. Methods: A Medline search for the period 1985 to 2001 was performed. Eight retrospective studies and 2 prospective randomized controlled trials were included. Selection criteria for the primary analysis were as follows: patients with peripapillary pancreatic cancer, endoscopic stent placement versus no stent, radical surgery, and assessment of postoperative morbidity and mortality. A secondary analysis included both radical and palliative surgery. Results: In the primary analysis, 337 patients underwent preoperative endoscopic biliary stent placement, and 412 patients had no endoscopic biliary stent placement (controls). The overall odds ratio for postoperative complications (stent vs. no stent) is estimated as 0.79: 95% CI [0.36, 1.73] and the estimated odds ratio for postoperative mortality is 0.81: 95% CI [0.33, 1.99]. In the secondary analysis, 1008 patients underwent preoperative EBS versus 720 control patients. The odds ratio for postoperative complications in this analysis was 0.93: 95% CI [0.65, 1.33] and for postoperative mortality is 1.12: 95% CI [0.62, 2.01]. Conclusion: No evidence was found of either a positive or adverse effect of preoperative endoscopic biliary stent placement on the outcome of surgery in patients with pancreatic cancer. (Gastrointest Endosc 2002;56:529-34.)