Duodenoscrotal fistula secondary to retroperitoneal migration of an endoscopically placed plastic biliary stent

Abstract

Endoscopic placement of a biliary stent for management of neoplastic biliary stricture is generally safe and relieves cholestasis in approximately 80% of cases.1Huibregtse K Katon RM Coene PP Tytgat GN Endoscopic palliative treatment in pancreatic cancer.Gastrointest Endosc. 1986; 32: 334-338Abstract Full Text PDF PubMed Scopus (164) Google Scholar, 2Speer AG Cotton PB Russell RC Mason RR Hatfield AR Leung JW et al.Randomised trial of endoscopic versus percutaneous stent insertion in malignant obstructive jaundice.Lancet. 1987; 2: 57-62Abstract PubMed Scopus (590) Google Scholar Complications such as hemorrhage and cholangitis related to the endoscopic procedure are rare. Other complications, such as intestinal perforation related to erosion through the intestinal wall by the distal end of the stent or proximal or distal stent migration, are rare but frequently lead to life-threatening problems and occasionally death. A case of duodenal-scrotal fistula by means of the retroperitoneal space secondary to perforation of the duodenum by a plastic biliary stent is reported. To our knowledge, this is the first report of such a complication.Case reportAn 86-year-old man was hospitalized with obstructive jaundice. Total bilirubin was 32.3 mg/dL (normal: <1.0 mg/dL). Transabdominal US and magnetic resonance cholangiography demonstrated dilation of the intrahepatic biliary tree with a nonspecific filling defect at the hepatic hilum that appeared to be either a neoplasm or stone. At ERCP a guidewire was negotiated through a stricture and into the proximal left intrahepatic biliary system; cannulation of the right hepatic duct was impossible. A 10F, 15-cm long plastic Amsterdam-type endoprosthesis was then inserted through the stricture to drain the left intrahepatic ducts. Cytopathologic evaluation of a specimen obtained by brushing during the endoscopic procedure revealed malignant cells compatible with cholangiocarcinoma. Percutaneous transhepatic cholangiography was performed 1 day later and an internal-external drainage catheter (8F, pigtail type) was positioned through the right biliary system.A drainage catheter was inserted through the cutaneous fistula in the scrotum and into the collection under fluoroscopic guidance. Subsequent duodenoscopy revealed that the endoscopic stent was wedged into the bowel wall. It was retrieved whereupon the fistula in the duodenal wall was clearly visible. Closure of the fistula by application of clips was unsuccessful. Thereafter, management consisted of conservative measures (nothing by mouth, intravenous fluids, antibiotics). The choice of nonoperative therapy was based on the advanced age of the patient and published data indicating that such injuries to the duodenal wall frequently close spontaneously.3Lopez Roses L Gonzalez Ramirez A Lancho Seco A Santos Blanco E Ibañez Alonso D Avila S et al.Clip closure of a duodenal perforation secondary to a biliary stent.Gastrointest Endosc. 2000; 51: 487-489Abstract Full Text Full Text PDF PubMed Scopus (50) Google Scholar A stent was not placed to drain the left lobe. No clinical evidence of complications related to the undrained liver lobe developed, and after 6 days the leukocytosis had resolved and fluoroscopy with injection of contrast by means of the percutaneous drain disclosed only a small residual fistulous tract. However, the patient became febrile on the next day and the white blood cell count rose to 56,000 mm3. CT demonstrated a retroperitoneal abscess with gas inside. The patient underwent surgical drainage but died 2 days later as a result of multiorgan failure.DiscussionDuodenal perforation is an uncommon complication of endoscopic biliary stent placement.4Ruffolo TA Lehman GA Sherman S Aycock R Hayes A Biliary stent migration with colonic diverticular impaction.Gastrointest Endosc. 1992; 38: 81-83Abstract Full Text PDF PubMed Scopus (40) Google Scholar, 5Lammer J Neumayer K Biliary drainage endoprosthesis: Experience with 201 placements.Radiology. 1986; 159: 625-629Crossref PubMed Scopus (130) Google Scholar The mechanical force exerted by the tip of a plastic endoprosthesis against intestinal mucosa can lead to necrosis of the wall and then perforation. In such cases, inflexibility of the stent and/or incorrect stent length are thought to be the main etiologic factors.6Gould J Train J Dan S Mitty HA Duodenal perforation as a delayed complication of placement of biliary endoprosthesis.Radiology. 1988; 167: 467-469Crossref PubMed Scopus (46) Google Scholar Many methods have been suggested for determining optimal biliary stent length.7Johanson JF Schmalz MJ Greenen JE Incidence and risk factors for biliary and pancreatic stent migration.Gastrointest Endosc. 1992; 38: 341-346Abstract Full Text PDF PubMed Scopus (269) Google Scholar The duodenal perforation may present as an early or a delayed complication5Lammer J Neumayer K Biliary drainage endoprosthesis: Experience with 201 placements.Radiology. 1986; 159: 625-629Crossref PubMed Scopus (130) Google Scholar, 8Kendall BJ Jutabha R So L Jamidar PA Determining required stent length in endoscopic retrograde biliary stenting.Gastrointest Endosc. 1995; 41: 242-243Abstract Full Text Full Text PDF PubMed Scopus (7) Google Scholar, 9Elder J Stevenson G Delayed perforation of a duodenal diverticulum by a biliary endoprosthesis.Can Assoc Radiol J. 1993; 44: 45-48PubMed Google Scholar, 10Coppola R Masetti R Riccioni ME Ciletti S De Franco A Detweiler M et al.Early retroduodenal perforation following endoscopic internal biliary drainage.Endoscopy. 1993; 25: 255-256Crossref PubMed Scopus (13) Google Scholar and may occur into either the intraperitoneal or retroperitoneal space. In the former instance, it can lead to life-threatening bile peritonitis; in the latter it can produce nonspecific symptoms (flank pain, fever, abdominal distention) or no symptoms, suggesting that such retroperitoneal perforations may remain contained.5Lammer J Neumayer K Biliary drainage endoprosthesis: Experience with 201 placements.Radiology. 1986; 159: 625-629Crossref PubMed Scopus (130) Google ScholarThere are case reports of perforation of the gut distal to the duodenum related to stent migration.11Schuster D Achong DJ Knox TA Fawaz KA Duodenal perforation by a biliary endoprosthesis: evaluation by hepatobiliary scintigraphy.J Clin Gastroenterol. 1992; 15: 177-179Crossref PubMed Google Scholar, 12Lenzo NP Garas G Biliary stent migration with colonic diverticular perforation.Gastrointest Endosc. 1998; 47: 543-544Abstract Full Text Full Text PDF PubMed Scopus (39) Google Scholar, 13Schaafsma RJH Spoelstra P Pakan J Huibregtse K Sigmoid perforation: a rare complication of a migrated biliary endoprosthesis.Endoscopy. 1996; 28: 469-470Crossref PubMed Scopus (35) Google Scholar The risk of perforation may be higher in patients with extensive colonic diverticulosis because entrapment of the leading end of the stent and the thin wall of the diverticular sac predispose to rupture. Proximal stent migration has also been reported.14Storkson RH Edwin B Reiertsen O Faerden AE Sortland O Rosseland AR Gut perforation caused by biliary endoprosthesis.Endoscopy. 2000; 32: 87-89Crossref PubMed Scopus (37) Google Scholar, 15Nicholson AA Martin DF Misplacement of endoscopic biliary endoprosthesis.Endoscopy. 1997; 29: 125-127Crossref PubMed Scopus (6) Google Scholar This leads to failure of biliary drainage, the proximal portion of the stent lying outside the biliary system, and persistent jaundice. More serious complications related to proximal migration, hepatic abscess formation in particular, are rare.16Tan CC Hall RI Pallan JP Irons RP Freeman JG Trans-hepatic proximal migration of percutaneous-endoscopic biliary stent presenting as an abdominal wall abscess.Gastrointest Endosc. 1996; 43: 152-154Abstract Full Text Full Text PDF PubMed Scopus (5) Google ScholarThe duodenal perforation in the present case was followed by development of a retroperitoneal duodenoscrotal fistula, with leakage of bile and purulent material from the scrotal skin. CT with injection of contrast medium by means of the percutaneous catheter and opacification of the fistulous tract demonstrated that the tract probably spread through the anterior pararenal space and continued distally to the pelvis along the anterolateral surface of the psoas muscle. The ductus deferens was the most likely path into scrotum. In retrospect, the use of a 15-cm stent was inappropriate. Furthermore, placement of a second percutaneous drain through the strictured biliary segment might have been a factor in the migration of the stent because of the friction generated by passage of one stent against the other. This might have forced the endoscopically placed stent inferiorally into the duodenum wall. Whether closure of the duodenal perforation was necessary is uncertain. Closure of this type of perforation with endoscopically placed clips has been reported.3Lopez Roses L Gonzalez Ramirez A Lancho Seco A Santos Blanco E Ibañez Alonso D Avila S et al.Clip closure of a duodenal perforation secondary to a biliary stent.Gastrointest Endosc. 2000; 51: 487-489Abstract Full Text Full Text PDF PubMed Scopus (50) Google Scholar This was unsuccessful in the present case. A catheter was placed through the fistula in the skin of the scrotum to drain the collection demonstrated by CT along the psoas muscle. Subsequent injection of contrast through the percutaneous drain demonstrated only a small residual fistulous tract. Removal of this drain may have been an error in management that possibly led to the development of the retroperitoneal infection. Although one case of “biliscrotum” as a complication of endoscopic sphincterotomy has been described,17Neoptolemos JP Harvey MH Slater ND Carr-Locke DL Abdominal wall bile staining and “biliscrotum” after retroperitoneal perforation following endoscopic sphincterotomy.Br J Surg. 1984; 71: 684Crossref PubMed Scopus (19) Google Scholar to our knowledge the present case is the first reported of a retroperitoneal duodenal perforation in association with a duodenoscrotal fistula secondary to an endoscopically placed plastic biliary stent. Endoscopic placement of a biliary stent for management of neoplastic biliary stricture is generally safe and relieves cholestasis in approximately 80% of cases.1Huibregtse K Katon RM Coene PP Tytgat GN Endoscopic palliative treatment in pancreatic cancer.Gastrointest Endosc. 1986; 32: 334-338Abstract Full Text PDF PubMed Scopus (164) Google Scholar, 2Speer AG Cotton PB Russell RC Mason RR Hatfield AR Leung JW et al.Randomised trial of endoscopic versus percutaneous stent insertion in malignant obstructive jaundice.Lancet. 1987; 2: 57-62Abstract PubMed Scopus (590) Google Scholar Complications such as hemorrhage and cholangitis related to the endoscopic procedure are rare. Other complications, such as intestinal perforation related to erosion through the intestinal wall by the distal end of the stent or proximal or distal stent migration, are rare but frequently lead to life-threatening problems and occasionally death. A case of duodenal-scrotal fistula by means of the retroperitoneal space secondary to perforation of the duodenum by a plastic biliary stent is reported. To our knowledge, this is the first report of such a complication. Case reportAn 86-year-old man was hospitalized with obstructive jaundice. Total bilirubin was 32.3 mg/dL (normal: <1.0 mg/dL). Transabdominal US and magnetic resonance cholangiography demonstrated dilation of the intrahepatic biliary tree with a nonspecific filling defect at the hepatic hilum that appeared to be either a neoplasm or stone. At ERCP a guidewire was negotiated through a stricture and into the proximal left intrahepatic biliary system; cannulation of the right hepatic duct was impossible. A 10F, 15-cm long plastic Amsterdam-type endoprosthesis was then inserted through the stricture to drain the left intrahepatic ducts. Cytopathologic evaluation of a specimen obtained by brushing during the endoscopic procedure revealed malignant cells compatible with cholangiocarcinoma. Percutaneous transhepatic cholangiography was performed 1 day later and an internal-external drainage catheter (8F, pigtail type) was positioned through the right biliary system.A drainage catheter was inserted through the cutaneous fistula in the scrotum and into the collection under fluoroscopic guidance. Subsequent duodenoscopy revealed that the endoscopic stent was wedged into the bowel wall. It was retrieved whereupon the fistula in the duodenal wall was clearly visible. Closure of the fistula by application of clips was unsuccessful. Thereafter, management consisted of conservative measures (nothing by mouth, intravenous fluids, antibiotics). The choice of nonoperative therapy was based on the advanced age of the patient and published data indicating that such injuries to the duodenal wall frequently close spontaneously.3Lopez Roses L Gonzalez Ramirez A Lancho Seco A Santos Blanco E Ibañez Alonso D Avila S et al.Clip closure of a duodenal perforation secondary to a biliary stent.Gastrointest Endosc. 2000; 51: 487-489Abstract Full Text Full Text PDF PubMed Scopus (50) Google Scholar A stent was not placed to drain the left lobe. No clinical evidence of complications related to the undrained liver lobe developed, and after 6 days the leukocytosis had resolved and fluoroscopy with injection of contrast by means of the percutaneous drain disclosed only a small residual fistulous tract. However, the patient became febrile on the next day and the white blood cell count rose to 56,000 mm3. CT demonstrated a retroperitoneal abscess with gas inside. The patient underwent surgical drainage but died 2 days later as a result of multiorgan failure. An 86-year-old man was hospitalized with obstructive jaundice. Total bilirubin was 32.3 mg/dL (normal: <1.0 mg/dL). Transabdominal US and magnetic resonance cholangiography demonstrated dilation of the intrahepatic biliary tree with a nonspecific filling defect at the hepatic hilum that appeared to be either a neoplasm or stone. At ERCP a guidewire was negotiated through a stricture and into the proximal left intrahepatic biliary system; cannulation of the right hepatic duct was impossible. A 10F, 15-cm long plastic Amsterdam-type endoprosthesis was then inserted through the stricture to drain the left intrahepatic ducts. Cytopathologic evaluation of a specimen obtained by brushing during the endoscopic procedure revealed malignant cells compatible with cholangiocarcinoma. Percutaneous transhepatic cholangiography was performed 1 day later and an internal-external drainage catheter (8F, pigtail type) was positioned through the right biliary system. A drainage catheter was inserted through the cutaneous fistula in the scrotum and into the collection under fluoroscopic guidance. Subsequent duodenoscopy revealed that the endoscopic stent was wedged into the bowel wall. It was retrieved whereupon the fistula in the duodenal wall was clearly visible. Closure of the fistula by application of clips was unsuccessful. Thereafter, management consisted of conservative measures (nothing by mouth, intravenous fluids, antibiotics). The choice of nonoperative therapy was based on the advanced age of the patient and published data indicating that such injuries to the duodenal wall frequently close spontaneously.3Lopez Roses L Gonzalez Ramirez A Lancho Seco A Santos Blanco E Ibañez Alonso D Avila S et al.Clip closure of a duodenal perforation secondary to a biliary stent.Gastrointest Endosc. 2000; 51: 487-489Abstract Full Text Full Text PDF PubMed Scopus (50) Google Scholar A stent was not placed to drain the left lobe. No clinical evidence of complications related to the undrained liver lobe developed, and after 6 days the leukocytosis had resolved and fluoroscopy with injection of contrast by means of the percutaneous drain disclosed only a small residual fistulous tract. However, the patient became febrile on the next day and the white blood cell count rose to 56,000 mm3. CT demonstrated a retroperitoneal abscess with gas inside. The patient underwent surgical drainage but died 2 days later as a result of multiorgan failure. DiscussionDuodenal perforation is an uncommon complication of endoscopic biliary stent placement.4Ruffolo TA Lehman GA Sherman S Aycock R Hayes A Biliary stent migration with colonic diverticular impaction.Gastrointest Endosc. 1992; 38: 81-83Abstract Full Text PDF PubMed Scopus (40) Google Scholar, 5Lammer J Neumayer K Biliary drainage endoprosthesis: Experience with 201 placements.Radiology. 1986; 159: 625-629Crossref PubMed Scopus (130) Google Scholar The mechanical force exerted by the tip of a plastic endoprosthesis against intestinal mucosa can lead to necrosis of the wall and then perforation. In such cases, inflexibility of the stent and/or incorrect stent length are thought to be the main etiologic factors.6Gould J Train J Dan S Mitty HA Duodenal perforation as a delayed complication of placement of biliary endoprosthesis.Radiology. 1988; 167: 467-469Crossref PubMed Scopus (46) Google Scholar Many methods have been suggested for determining optimal biliary stent length.7Johanson JF Schmalz MJ Greenen JE Incidence and risk factors for biliary and pancreatic stent migration.Gastrointest Endosc. 1992; 38: 341-346Abstract Full Text PDF PubMed Scopus (269) Google Scholar The duodenal perforation may present as an early or a delayed complication5Lammer J Neumayer K Biliary drainage endoprosthesis: Experience with 201 placements.Radiology. 1986; 159: 625-629Crossref PubMed Scopus (130) Google Scholar, 8Kendall BJ Jutabha R So L Jamidar PA Determining required stent length in endoscopic retrograde biliary stenting.Gastrointest Endosc. 1995; 41: 242-243Abstract Full Text Full Text PDF PubMed Scopus (7) Google Scholar, 9Elder J Stevenson G Delayed perforation of a duodenal diverticulum by a biliary endoprosthesis.Can Assoc Radiol J. 1993; 44: 45-48PubMed Google Scholar, 10Coppola R Masetti R Riccioni ME Ciletti S De Franco A Detweiler M et al.Early retroduodenal perforation following endoscopic internal biliary drainage.Endoscopy. 1993; 25: 255-256Crossref PubMed Scopus (13) Google Scholar and may occur into either the intraperitoneal or retroperitoneal space. In the former instance, it can lead to life-threatening bile peritonitis; in the latter it can produce nonspecific symptoms (flank pain, fever, abdominal distention) or no symptoms, suggesting that such retroperitoneal perforations may remain contained.5Lammer J Neumayer K Biliary drainage endoprosthesis: Experience with 201 placements.Radiology. 1986; 159: 625-629Crossref PubMed Scopus (130) Google ScholarThere are case reports of perforation of the gut distal to the duodenum related to stent migration.11Schuster D Achong DJ Knox TA Fawaz KA Duodenal perforation by a biliary endoprosthesis: evaluation by hepatobiliary scintigraphy.J Clin Gastroenterol. 1992; 15: 177-179Crossref PubMed Google Scholar, 12Lenzo NP Garas G Biliary stent migration with colonic diverticular perforation.Gastrointest Endosc. 1998; 47: 543-544Abstract Full Text Full Text PDF PubMed Scopus (39) Google Scholar, 13Schaafsma RJH Spoelstra P Pakan J Huibregtse K Sigmoid perforation: a rare complication of a migrated biliary endoprosthesis.Endoscopy. 1996; 28: 469-470Crossref PubMed Scopus (35) Google Scholar The risk of perforation may be higher in patients with extensive colonic diverticulosis because entrapment of the leading end of the stent and the thin wall of the diverticular sac predispose to rupture. Proximal stent migration has also been reported.14Storkson RH Edwin B Reiertsen O Faerden AE Sortland O Rosseland AR Gut perforation caused by biliary endoprosthesis.Endoscopy. 2000; 32: 87-89Crossref PubMed Scopus (37) Google Scholar, 15Nicholson AA Martin DF Misplacement of endoscopic biliary endoprosthesis.Endoscopy. 1997; 29: 125-127Crossref PubMed Scopus (6) Google Scholar This leads to failure of biliary drainage, the proximal portion of the stent lying outside the biliary system, and persistent jaundice. More serious complications related to proximal migration, hepatic abscess formation in particular, are rare.16Tan CC Hall RI Pallan JP Irons RP Freeman JG Trans-hepatic proximal migration of percutaneous-endoscopic biliary stent presenting as an abdominal wall abscess.Gastrointest Endosc. 1996; 43: 152-154Abstract Full Text Full Text PDF PubMed Scopus (5) Google ScholarThe duodenal perforation in the present case was followed by development of a retroperitoneal duodenoscrotal fistula, with leakage of bile and purulent material from the scrotal skin. CT with injection of contrast medium by means of the percutaneous catheter and opacification of the fistulous tract demonstrated that the tract probably spread through the anterior pararenal space and continued distally to the pelvis along the anterolateral surface of the psoas muscle. The ductus deferens was the most likely path into scrotum. In retrospect, the use of a 15-cm stent was inappropriate. Furthermore, placement of a second percutaneous drain through the strictured biliary segment might have been a factor in the migration of the stent because of the friction generated by passage of one stent against the other. This might have forced the endoscopically placed stent inferiorally into the duodenum wall. Whether closure of the duodenal perforation was necessary is uncertain. Closure of this type of perforation with endoscopically placed clips has been reported.3Lopez Roses L Gonzalez Ramirez A Lancho Seco A Santos Blanco E Ibañez Alonso D Avila S et al.Clip closure of a duodenal perforation secondary to a biliary stent.Gastrointest Endosc. 2000; 51: 487-489Abstract Full Text Full Text PDF PubMed Scopus (50) Google Scholar This was unsuccessful in the present case. A catheter was placed through the fistula in the skin of the scrotum to drain the collection demonstrated by CT along the psoas muscle. Subsequent injection of contrast through the percutaneous drain demonstrated only a small residual fistulous tract. Removal of this drain may have been an error in management that possibly led to the development of the retroperitoneal infection. Although one case of “biliscrotum” as a complication of endoscopic sphincterotomy has been described,17Neoptolemos JP Harvey MH Slater ND Carr-Locke DL Abdominal wall bile staining and “biliscrotum” after retroperitoneal perforation following endoscopic sphincterotomy.Br J Surg. 1984; 71: 684Crossref PubMed Scopus (19) Google Scholar to our knowledge the present case is the first reported of a retroperitoneal duodenal perforation in association with a duodenoscrotal fistula secondary to an endoscopically placed plastic biliary stent. Duodenal perforation is an uncommon complication of endoscopic biliary stent placement.4Ruffolo TA Lehman GA Sherman S Aycock R Hayes A Biliary stent migration with colonic diverticular impaction.Gastrointest Endosc. 1992; 38: 81-83Abstract Full Text PDF PubMed Scopus (40) Google Scholar, 5Lammer J Neumayer K Biliary drainage endoprosthesis: Experience with 201 placements.Radiology. 1986; 159: 625-629Crossref PubMed Scopus (130) Google Scholar The mechanical force exerted by the tip of a plastic endoprosthesis against intestinal mucosa can lead to necrosis of the wall and then perforation. In such cases, inflexibility of the stent and/or incorrect stent length are thought to be the main etiologic factors.6Gould J Train J Dan S Mitty HA Duodenal perforation as a delayed complication of placement of biliary endoprosthesis.Radiology. 1988; 167: 467-469Crossref PubMed Scopus (46) Google Scholar Many methods have been suggested for determining optimal biliary stent length.7Johanson JF Schmalz MJ Greenen JE Incidence and risk factors for biliary and pancreatic stent migration.Gastrointest Endosc. 1992; 38: 341-346Abstract Full Text PDF PubMed Scopus (269) Google Scholar The duodenal perforation may present as an early or a delayed complication5Lammer J Neumayer K Biliary drainage endoprosthesis: Experience with 201 placements.Radiology. 1986; 159: 625-629Crossref PubMed Scopus (130) Google Scholar, 8Kendall BJ Jutabha R So L Jamidar PA Determining required stent length in endoscopic retrograde biliary stenting.Gastrointest Endosc. 1995; 41: 242-243Abstract Full Text Full Text PDF PubMed Scopus (7) Google Scholar, 9Elder J Stevenson G Delayed perforation of a duodenal diverticulum by a biliary endoprosthesis.Can Assoc Radiol J. 1993; 44: 45-48PubMed Google Scholar, 10Coppola R Masetti R Riccioni ME Ciletti S De Franco A Detweiler M et al.Early retroduodenal perforation following endoscopic internal biliary drainage.Endoscopy. 1993; 25: 255-256Crossref PubMed Scopus (13) Google Scholar and may occur into either the intraperitoneal or retroperitoneal space. In the former instance, it can lead to life-threatening bile peritonitis; in the latter it can produce nonspecific symptoms (flank pain, fever, abdominal distention) or no symptoms, suggesting that such retroperitoneal perforations may remain contained.5Lammer J Neumayer K Biliary drainage endoprosthesis: Experience with 201 placements.Radiology. 1986; 159: 625-629Crossref PubMed Scopus (130) Google Scholar There are case reports of perforation of the gut distal to the duodenum related to stent migration.11Schuster D Achong DJ Knox TA Fawaz KA Duodenal perforation by a biliary endoprosthesis: evaluation by hepatobiliary scintigraphy.J Clin Gastroenterol. 1992; 15: 177-179Crossref PubMed Google Scholar, 12Lenzo NP Garas G Biliary stent migration with colonic diverticular perforation.Gastrointest Endosc. 1998; 47: 543-544Abstract Full Text Full Text PDF PubMed Scopus (39) Google Scholar, 13Schaafsma RJH Spoelstra P Pakan J Huibregtse K Sigmoid perforation: a rare complication of a migrated biliary endoprosthesis.Endoscopy. 1996; 28: 469-470Crossref PubMed Scopus (35) Google Scholar The risk of perforation may be higher in patients with extensive colonic diverticulosis because entrapment of the leading end of the stent and the thin wall of the diverticular sac predispose to rupture. Proximal stent migration has also been reported.14Storkson RH Edwin B Reiertsen O Faerden AE Sortland O Rosseland AR Gut perforation caused by biliary endoprosthesis.Endoscopy. 2000; 32: 87-89Crossref PubMed Scopus (37) Google Scholar, 15Nicholson AA Martin DF Misplacement of endoscopic biliary endoprosthesis.Endoscopy. 1997; 29: 125-127Crossref PubMed Scopus (6) Google Scholar This leads to failure of biliary drainage, the proximal portion of the stent lying outside the biliary system, and persistent jaundice. More serious complications related to proximal migration, hepatic abscess formation in particular, are rare.16Tan CC Hall RI Pallan JP Irons RP Freeman JG Trans-hepatic proximal migration of percutaneous-endoscopic biliary stent presenting as an abdominal wall abscess.Gastrointest Endosc. 1996; 43: 152-154Abstract Full Text Full Text PDF PubMed Scopus (5) Google Scholar The duodenal perforation in the present case was followed by development of a retroperitoneal duodenoscrotal fistula, with leakage of bile and purulent material from the scrotal skin. CT with injection of contrast medium by means of the percutaneous catheter and opacification of the fistulous tract demonstrated that the tract probably spread through the anterior pararenal space and continued distally to the pelvis along the anterolateral surface of the psoas muscle. The ductus deferens was the most likely path into scrotum. In retrospect, the use of a 15-cm stent was inappropriate. Furthermore, placement of a second percutaneous drain through the strictured biliary segment might have been a factor in the migration of the stent because of the friction generated by passage of one stent against the other. This might have forced the endoscopically placed stent inferiorally into the duodenum wall. Whether closure of the duodenal perforation was necessary is uncertain. Closure of this type of perforation with endoscopically placed clips has been reported.3Lopez Roses L Gonzalez Ramirez A Lancho Seco A Santos Blanco E Ibañez Alonso D Avila S et al.Clip closure of a duodenal perforation secondary to a biliary stent.Gastrointest Endosc. 2000; 51: 487-489Abstract Full Text Full Text PDF PubMed Scopus (50) Google Scholar This was unsuccessful in the present case. A catheter was placed through the fistula in the skin of the scrotum to drain the collection demonstrated by CT along the psoas muscle. Subsequent injection of contrast through the percutaneous drain demonstrated only a small residual fistulous tract. Removal of this drain may have been an error in management that possibly led to the development of the retroperitoneal infection. Although one case of “biliscrotum” as a complication of endoscopic sphincterotomy has been described,17Neoptolemos JP Harvey MH Slater ND Carr-Locke DL Abdominal wall bile staining and “biliscrotum” after retroperitoneal perforation following endoscopic sphincterotomy.Br J Surg. 1984; 71: 684Crossref PubMed Scopus (19) Google Scholar to our knowledge the present case is the first reported of a retroperitoneal duodenal perforation in association with a duodenoscrotal fistula secondary to an endoscopically placed plastic biliary stent.