Pancreaticoduodenectomy Following Placement of Expandable Metal Stents
John T. Mullen, Henry F. Gomez, Jeffrey H. Lee, William A. Ross, Eddie K. Abdalla, Jean-Nicolas Vauthey, Jeffrey E. Lee, Peter W. Pisters, Douglas B. Evans, University of Texas M. D. Anderson Cancer Center, Houston, TX
Background:
Contemporary treatment programs for patients with potentially resectable pancreatic cancer often involve preoperative (neoadjuvant) therapy. When the duration of preoperative therapy is greater than 2 to 3 months, there is a significant risk of biliary stent occlusion if a plastic stent is used. Metal stents have a much-improved patency but may complicate subsequent pancreaticoduodenectomy (PD).
Methods:
The rates of perioperative morbidity, mortality and stent complications were evaluated in 272 consecutive patients who underwent PD at our institution from May 2001 (date of first PD with a metal stent) to November 2004. Patients who underwent PD after placement of a metal stent were compared to all other patients.
Results:
PD was performed in 29 (11%) patients after placement of a metal stent (metal stent group) and in 243 (89%) patients without a metal stent, including 167 (61%) who had plastic stents. When comparing the metal stent group to all other patients, there were no differences in median operative time (6.6 hrs vs 7.3 hrs; p = 0.9), intraoperative blood loss (700 vs. 600 mL; p = 0.9), or length of hospital stay (11 vs. 10 days; p = 0.2). There were no perioperative deaths in the metal stent group and 3 (1.2%) deaths in the other 243 patients. The incidence of perioperative complications was similar between the two groups (34% vs. 33%), including the rates of pancreatic fistula (7% vs. 4%), intraabdominal abscess (3% vs. 4%), and wound infection (7% vs. 4%). Metal stent-related complications (both occlusions) occurred in 2 (7%) of 29 patients during a median preoperative interval of 4.1 months. In contrast, of 167 patients with plastic stents, there were 131 (78%) complications, including 114 stent occlusions during a median preoperative interval of 3.9 months. This difference in preoperative complication rate between metal and plastic stents was significant (p < 0.001).
Conclusions:
Expandable metal stents do not increase PD-associated perioperative morbidity or mortality. Importantly, there are few stent-related complications prior to PD in those patients treated with an expandable metal stent in contrast to the high complication rate associated with plastic biliary stents. Following high-quality CT imaging, an expandable metal stent is our preferred method of biliary decompression in all patients with distal bile duct obstruction secondary to a presumed periampullary cancer, including those with potentially resectable disease.
Back to 2005 Program