- Email: [email protected]
Percutaneous-endoscopic biliary stenting in patients with occluded surgical bypass
Percutaneous-Endoscopic Biliary Stenting in Patients with Occluded Surgical Bypass TAT-KINTSANG,M.D., ARTHURR.CRAMPTON,M.D.,JOELR. JOEL A. CAHAN, M.D., EV~WO~, MOONS
BERNSTEIN,M.D.,STEPHENBUTO,M.D.,
PURPOSE:The purpose of this investigation was to test the feasibility of using a recently developed technique of placing internalized biliary stems into patients who have had reobstruction after initial surgical bypass. PATIENTSANDMETHODS: Sevenmenandthree women, 46 to 85 years of age (eight with pancreatic carcinoma, one with metastatic colon, and one with metastatic ovarian carcinoma), all had reobstruction after initial surgical bypass palliation. Subsequent attempts to place stents via endoscope failed in five patients, a pair of 7-Fr stents placed in one patient failed to drain well. Endoscopic stenting in four patients was not even attempted because of severely distorted anatomy. Nine of the 10 patients then had successful internal stent placement by a combined percutaneous-transhepatic and peroralendoscopically guided technique. RESULTS:One of these nine placeable stems failed to drain well and the patient died 8 days later with massive tumor. Seven showed a significant decrease in bilirubin levels and improved quality of life. Two of these had sepsis that responded to antibiotics. Life span ranged between 11 days and 10 months, with one patient still alive; no deaths were directly due to stents. CONCLUSION: A combined transhepatic-peroral technique of placing internalized biliary stents can be expected to result in repalliation in a majority of patients with reobstruction after earlier surgical bypass and in whom subsequent attempts at endoscopic placement of stents have failed or in whom tumor growth prevents undertaking the endoscopic approach.
e-establishing internal biliary drainage in paR tients with unresectable obstructing biliary tract tumors can be approached by either surgical bypass or prosthetic stent placement. For the patient in favorable condition, surgery has theoretic advantages marred only by morbidity and mortality rates. For others in not so favorable condition, or who refuse operation, nonsurgical drainage methods have been developed. Probably the most popular method is the external-internal drain [1,2]. Its main advantages are relative ease of initial placement and subsequent replacement if obstructed. Disadvantages are percutaneous infection, bile leaks, pull-out dislodgement, and the continuous visible and palpable presence of the device. Completely internalized stents were developed to minimize some of these drawbacks. But, since longevity of stent patency seems correlated to their diameter [3], very large (lo- to 14-French [Fr]) bore stent tracts are produced when the transhepatic placement technique is used, thus increasing the likelihood of significant trauma to the liver. The advent of modern endoscopes has led to the entirely endoscopic approach of biliary stenting, whereby major liver trauma is minimized. However, this exclusively endoscopic approach can be thwarted by the inability to traverse tight stenoses because of limited mechanical leverage; also, the endoscopic passage may be impeded by operatively altered anatomy [4-61, or tumor extension into the duodenum. Most such patients must heretofore be content with external-internal stent or external drainage. In an effort to optimize the advantages of the endoscopic technique and minimize the disadvantages of the percutaneous approach, we have developed a combined percutaneous-endoscopic procedure [7-lo]. This report studies the feasibility and clinical results of using this new percutaneous-endoscopic biliary stent (PEBS) placement technique to create a palliative stent in 10 patients who originally underwent surgical bypass for jaundice but who subsequently had reobstruction.
PATIENTS AND METHODS Patient Population Studied (Table I) From November 1985 to March 1988, 10 patients who had earlier surgical procedures for malignant biliary obstruction experienced reobstruction and recurrent jaundice. Each was approached with the intent to endoscopically place a stent across the new obstruction, but because of the severe anatomic deformities caused by tumor overgrowth and prior surgery, five of six endoscopic attempts at stent placement failed; in the sixth case, two parallel 7-Fr stents were placed. This latter patient, however, had repeated episodes of sepsis and required replacement by a single 11.5-Fr stent, using the new combined approach. Endoscopic 344
April
1990
The American
Journal
of Medicine
Volume
88
Age/ Sex
75/M
67/F
71/M
48/M
85/M
46/M
47/M
53/F
Patient
1
2
3
4
5
6
7
8
Ovarian carcinoma 1 year prior; cholangiocarcinoma 3 years prior
Colon carcinoma metastatic to porta hepatis
Pancreatic cancer
Pancreatic cancer
Pancreatic cancer
Pancreatic cancer
Pancreatic cancer
Pancreatic cancer
Primary Disease and
and
no biliary
and
and
Hepatojefunostomy with Roux-en-Y construction/44 months
Failed attempted stent at other hospital. Cholecystectomy and T-tube placement/9 months-when T-tube dislodged
Cholecystojejunostomy gas;;j;tunostomy/2
Gastrojejunostomy, drainage/7 days
Cholecystojejunostomy g$rc;sefunostomy/2
Cholecystojejunostomy gm$;Ffunostomy/6
Choledochojejunostomy and gastrojejunostomy/6 months (external biliary drainage 4 days before stent placement)
Cholecystojejunostomy g$;jsafunostomy/14
Pre-Existing Surgical Drainage Procedure before the Stent Placement/Duration between Surgery and Stenting
Failed
Failed
Not attempted
Failed
Not attempted
7
on page 346
cm/l0 Fr
25 days
6 months
11 cm/11.5
Fr
3 months
9cm/11.5 Fr
0 days: PEBS ye; to
Fr
7 cm/l0
10.5 3.9
21.54 1.2
30-
25 -
13-8
5
40
Patient died of metastatic ovarian carcinoma to the lungs
None
Admitted 9 days after stent placed for fever and leukocytosis. Resolved with ceftazidime and clindamycin
lniitial exploratory laparotomy revealed massive tumor prohibiting surgical bypass. Stent placed easily but never drained well
Readmitted 1 week prior to death with bowel obstruction and perforation from massive tumor involvement: underwent exploratory laparotomy but resection not possible
Episode of sepsis immediately after stent insertion; treated with gentamicin and piperacillin
3-l
53 days
Developed UGI bleeding shortly after stem placement, resulting in less of a decrease in bilirubin
19 - 17
11 days
9cm/11.5Fr
None
Clinical Complications and Course after PEBS Placement
3.3 + 0.9
Total Bilirubin Change (w/dL)
10 months
39 days
Fr
Duration Patency of PEBS Stent
7 cm/11.5Fr
5cm/11.5Fr
7 cm/11.5
PEBS Stent Size (length/W
Continued
2 Endoscoprc 7-Fr stents that failed to drain well were replaced with PEES
Failed
Failed
Outcome of Attempted Endoscopic Stent prior to PEBS
TABLEI ClinicalStatusof 10 PatientsWhoReceivedSubsequentPercutaneous-Endoscopic BiliaryStents(PEBS)after Initial SurgicalBiliaryDrainageFailed
25 days
6 months
3 months
8 days
60 days
39 days
Improved
Improved
Improved
Not improved
Improved
Improved
Improved slightly
Improved
10 months
11 days
General Quality of life between PEBS Stenting and Death
Patient Survival From Time of PEBS Stenting
PEBS FOR REOBSTRUCTION
/ TSANG
ET AL
approach was not even attempted in four patients because of severe tumor distortion. This group of patients consisted of seven men and three women, with a median age of 64 years (range: 46 to 85 years). Eight patients had pancreatic carcinoma. One had metastatic colon carcinoma to the porta hepatis. One had cholangiocarcinoma and a subsequent metastatic ovarian carcinoma to the porta hepatis. The interval between the earlier surgical drainage and the recurrent biliary obstructions ranged from 7 days to 44 months. Of the eight patients with pancreatic carcinomas, six had both cholecystojejunostomies and gastrojejunostomies, one had both choledochojejunostomy and gastrojejunostomy, and one had only a gastrojejunostomy because cholecystojejunostomy was prevented by tumor invasion. The patient with metastatic colon carcinoma had a failed endoscopic stent placement attempt followed by a failed and complicated attempt at external-internal stent placement at another hospital; this was followed by cholecystectomy and T-tube placement, but the T-tube dislodged after 9 months. The patient with biliary obstructing metastatic ovarian carcinoma had had a prior hepatojejunostomy and Roux-en-Y reconstruction 3 years before for cholangiocarcinoma. Technique of Percutaneous-Endoscopic Stent Placement The interventional radiologist initially obtains a standard transhepatic cholangiogram, which enables the obstructed duct to be traversed by an 0.038-inch guide wire into the duodenum, over which wire a 6-Fr Teflon catheter is placed. Next, this guiding system (wire and overlying catheter) is endoscopically snared as it is pushed into the duodenum. This guide is manufactured sufficiently long so that one end is secured externally at the cutaneous entry site while the duodenal end can be removed per OS.Then a lo-Fr or 11.5-Fr endoprosthetic stent (Cotton-Lung) is placed over the end of the guide exiting the mouth and is pushed over the guide to the ampulla and through the stenosis by a stiff lo-Fr introducer tube. Once the stent is in the’ desired position, as determined by fluoroscopic monitoring, the guide system is removed per OSwhile the introducer is held firmly against the stent to prevent dislodgement. To confirm patency of the stent, radiopaque contrast can be injected through a short, temporary, 5- or 6-Fr catheter placed percutaneously into the ducts above the stent over the wire part of the guide as it is removed per OS.Finally, both the temporary transhepatic catheter and stiff stent introducer are removed (Figure 1). RESULTS Although combined PEBS placement was attempted in all 10 patients, it could not be completed in one patient. The failure in this patient, as in virtually all the very few failures so far in our experience, was due to inability to pass the guide wire through the tumor from above even on subsequent attempts; in this situation an external drain is left in place. Also, in this patient, a subsequent attempt to surgically re-establish internal bile flow failed because of tumor progression. Of the nine patients with initially successful stent placement, poor drainage occurred in one (bilirubin 346
April
1990
The American
Journal
of Medicine
Volume
88
PEBS FOR REOBSTRUCTION
/ TSANG
ET AL
Figure 1. Sequential radiographs (Patient 4) illustratingthe problem and solution to postsurgical biliary obstruction. A, percutaneous transhepatic cholangiogram (PTC) showing obstructed cholecystojejunostomy (arrow). B, PTC catheter passed through obstruction (arrows) into distal common bile duct. C, percutaneous transhepatic-peroral guiding system (open arrows) inside biliary stent (solid arrows), which is in desired position across obstruction. D, guiding system removed. Contrast injected through temporary hepatic catheter (curved arrow) shows stent is patent and bridges obstruction between bile ducts and duodenum (arrowheads).
level increased from 25 to 40 mg/dL), but earlier surgical exploration had revealed massive tumor overgrowth prohibiting surgical bypass. He died 8 days later. Seven of the patients in whom stents were placed showed significant improvement in bilirubin values and other liver function tests. The one who developed upper gastrointestinal bleeding, unrelated to the stent, showed a “false” bilirubin response. One patient with a successfully placed stent and pancreatic tumor developed sepsis immediately after stent placement, but responded well to gentamicin and piperacillin and improved to survive 39 more days. Another patient with pancreatic carcinoma developed
sepsis 9 days after stent placement but responded to ceftazidime and clindamycin to survive without jaundice another 3 months. Two of the successfully placed stents required a 4day drainage between the initial percutaneous entry into the bile ducts and the passage of a guide through tumor into the duodenum. These were considered “delayed successes” rather than “failures,” since delay time was short and the stent was eventually placed. Seven of the eight patients with functioning stents died between 11 days and 10 months from time of stent placement, the stents remaining patent (bilirubin levels staying low) until death. The eighth patient in this group is still living, with a patent stent 7 months April 1990
The American
Journal
of Medicine
Volume
88
347
PEBS FOR REOBSTRUCTION
/ TSANG
ET AL
after placement. None of the deaths were directly related to the stent. The general quality of life as measured by decreased pain, pruritis, and jaundice and increasing appetite with weight stabilization showed improvement in seven of 10 patients.
COMMENTS Having been initially unresectable and now with failed surgical biliary diversion, these 10 cases represent a distillation of palliative problems in the management of end-stage obstructive jaundice. Our 90% large-bore stent placement rate and 30% rate of overall complications are roughly similar to the rates in larger series of patients with less advanced disease. Sepsis, although usually transient and treatable, continues to plague stent placement procedures by any of the various methods. Since the advantages of the larger-bore stent have been documented [ll], we believe the 11.5 Fr is a reasonable compromise between sufficient internal diameter (2.75 mm) assuring adequate drainage and the potentially more traumatic larger sizes. The principal advantage of the combined technique, as opposed to either the percutaneous transhepatic or the endoscopic alone, seems to be the greater mechanical advantage of having fixed, firm, hands-on control of both ends of the guiding system. Thus, most, if not all, forces exerted by the introducer (“pusher”) are exerted in overcoming the tight tumor stenoses by being transmitted axially along the guide with little off-axis or buckling deformity being allowed by the fixed guide. Buckling not only dissipates force away from desired directions, but can cut or slice sideways through normal structures. In patients with Roux-enY or Billroth II who may have unapproachable ampullae via an endoscope, the percutaneously placed guide wire coming through the ampulla can be pushed through the surgically modified gut to a point accessible to an endoscopic snare [8]. If this maneuver becomes impossible, then stenting transhepatically or percutaneous drainage would be elected. Another feature of PEBS deserves emphasis since we believe it also plays a role in management of complications. As originally conceived, the procedure was seen as a collaborative effort between interventional radiologist and endoscopic gastroenterologist. A protocol has evolved for management of the jaundiced patient whose disease has progressed beyond surgical palliation. If there is not too much tumor growth deformity of gastroduodenal anatomy, the less invasive procedure of endoscopic retrograde biliary stent placement is usually attempted first. Should this fail,
348
April
1990 The American
Journal
of Medicine
Volume
88
the interventional radiologist performs percutaneous entry into intrahepatic biliary ducts by a 6-Fr coaxial catheter and proceeds to attempt guide wire passage through the obstruction. This should be successful in more than 90% of attempts. Thus, the PEBS placement can usually be completed in 1 day. Should the guide wire not pass the obstruction on the first attempt, a 6-Fr Cope loop is placed for external drainage, and follow-up re-attempt is made 3 to 4 days later. A second attempt at guide passage is successful in well over half of these initial 10% failures, and final internal stenting is completed in well over 95%. Another important function of the “team approach” is management of stent clogging or malpositioning. The endoscopist can remove the clogged stent and accomplish replacement with relative ease, assuming the ampulla is accessible, since now the tumor tract is dilated by the preexisting stent and allows replacement endoscopitally with much less force required. This report describes the feasibility and acceptable clinical results of placing completely internalized large-bore biliary stents by a less traumatic approach. Thus, a number of terminally ill patients, even after prior surgical bypass has failed, can be offered such stents, which allow them a more normal lifestyle and improved quality of life.
REFERENCES 1. Ring EJ. Oleaga JA. Freiman DB, Husted JW, Lunderquist A: Therapeutic applications of catheter cholangiography. Radiology 1978; 128: 333-338. 2. Hoevels J, Lunderquist A, lhse I: Percutaneous transhepatic intubation of brie ducts for combined internal-external drainage in preoperative and palliative treatment of obstructive jaundice. Gastrointest Radio1 1978; 3: 23-31. 3. Leung JW. Del Favero G, Cotton PB: Endoscopic brliary prostheses: a companson of materials. Gastrointest Endosc 1985; 31: 93-95. 4. Siegel JH: Improved brliary decompressjon wjth large calrber endoscoprc prostheses. Gastrointest Endosc 1984; 30: 21-23. 5. Marks WM. Freeny PC, Kozarek RA, Ball TJ: Endoscopic retrograde brliary drarnage: an alternate method of drainage. Semin Intervent Radio1 1985; 2: 74-80. 6. Siegel JH: Endoscopic decompression of the biliary tree. In: Silvrs SE, ed. Therapeutrc gastrointestinal endoscopy. New York: Igaku-Shoin, 1985; 241-268. 7. Tsang TK, Crampton AR, Bernstern JR, Ramos SR. Wieland JM: Percutaneousendoscopic brlrary stent placement: a preliminary report. Ann Intern Med 1987; 106: 389-392. 8. Tsang TK, Crampton AR, Merselman M, Desai TK, Bernstein JR Percutaneousendoscopic biliary stent placement for Billroth II and total gastrectomy with Rouxen-Y enteroenterostomv. Gastrointest Endosc 1988: 34: 45-47. 9. Kerlan RK Jr, Ring EJ: Pogany AC, Jeffrey RB Jr: Brliary endoprostheses: rnsertion usrng a combrned peroral-transhepatic method. Radiology 1984; 150: 828-
830. 10. Brambs HJ. Billmann P, Pausch J. Holstege A, Salm R: Non-surgical biliary drainage: endoscopic conversion of percutaneous transhepatic into endoprosthetic drainage. Endoscopy 1986; 18: 52-54. 11. Speer AG, Cotton PB. MacRae KD: Endoscopic management of makgnant biliary obstruction: stents of 10 F gauze are preferable to stents of 8 F gauze. Gastrointest Endosc 1988; 34: 412-417.
BERNSTEIN,M.D.,STEPHENBUTO,M.D.,
PURPOSE:The purpose of this investigation was to test the feasibility of using a recently developed technique of placing internalized biliary stems into patients who have had reobstruction after initial surgical bypass. PATIENTSANDMETHODS: Sevenmenandthree women, 46 to 85 years of age (eight with pancreatic carcinoma, one with metastatic colon, and one with metastatic ovarian carcinoma), all had reobstruction after initial surgical bypass palliation. Subsequent attempts to place stents via endoscope failed in five patients, a pair of 7-Fr stents placed in one patient failed to drain well. Endoscopic stenting in four patients was not even attempted because of severely distorted anatomy. Nine of the 10 patients then had successful internal stent placement by a combined percutaneous-transhepatic and peroralendoscopically guided technique. RESULTS:One of these nine placeable stems failed to drain well and the patient died 8 days later with massive tumor. Seven showed a significant decrease in bilirubin levels and improved quality of life. Two of these had sepsis that responded to antibiotics. Life span ranged between 11 days and 10 months, with one patient still alive; no deaths were directly due to stents. CONCLUSION: A combined transhepatic-peroral technique of placing internalized biliary stents can be expected to result in repalliation in a majority of patients with reobstruction after earlier surgical bypass and in whom subsequent attempts at endoscopic placement of stents have failed or in whom tumor growth prevents undertaking the endoscopic approach.
e-establishing internal biliary drainage in paR tients with unresectable obstructing biliary tract tumors can be approached by either surgical bypass or prosthetic stent placement. For the patient in favorable condition, surgery has theoretic advantages marred only by morbidity and mortality rates. For others in not so favorable condition, or who refuse operation, nonsurgical drainage methods have been developed. Probably the most popular method is the external-internal drain [1,2]. Its main advantages are relative ease of initial placement and subsequent replacement if obstructed. Disadvantages are percutaneous infection, bile leaks, pull-out dislodgement, and the continuous visible and palpable presence of the device. Completely internalized stents were developed to minimize some of these drawbacks. But, since longevity of stent patency seems correlated to their diameter [3], very large (lo- to 14-French [Fr]) bore stent tracts are produced when the transhepatic placement technique is used, thus increasing the likelihood of significant trauma to the liver. The advent of modern endoscopes has led to the entirely endoscopic approach of biliary stenting, whereby major liver trauma is minimized. However, this exclusively endoscopic approach can be thwarted by the inability to traverse tight stenoses because of limited mechanical leverage; also, the endoscopic passage may be impeded by operatively altered anatomy [4-61, or tumor extension into the duodenum. Most such patients must heretofore be content with external-internal stent or external drainage. In an effort to optimize the advantages of the endoscopic technique and minimize the disadvantages of the percutaneous approach, we have developed a combined percutaneous-endoscopic procedure [7-lo]. This report studies the feasibility and clinical results of using this new percutaneous-endoscopic biliary stent (PEBS) placement technique to create a palliative stent in 10 patients who originally underwent surgical bypass for jaundice but who subsequently had reobstruction.
PATIENTS AND METHODS Patient Population Studied (Table I) From November 1985 to March 1988, 10 patients who had earlier surgical procedures for malignant biliary obstruction experienced reobstruction and recurrent jaundice. Each was approached with the intent to endoscopically place a stent across the new obstruction, but because of the severe anatomic deformities caused by tumor overgrowth and prior surgery, five of six endoscopic attempts at stent placement failed; in the sixth case, two parallel 7-Fr stents were placed. This latter patient, however, had repeated episodes of sepsis and required replacement by a single 11.5-Fr stent, using the new combined approach. Endoscopic 344
April
1990
The American
Journal
of Medicine
Volume
88
Age/ Sex
75/M
67/F
71/M
48/M
85/M
46/M
47/M
53/F
Patient
1
2
3
4
5
6
7
8
Ovarian carcinoma 1 year prior; cholangiocarcinoma 3 years prior
Colon carcinoma metastatic to porta hepatis
Pancreatic cancer
Pancreatic cancer
Pancreatic cancer
Pancreatic cancer
Pancreatic cancer
Pancreatic cancer
Primary Disease and
and
no biliary
and
and
Hepatojefunostomy with Roux-en-Y construction/44 months
Failed attempted stent at other hospital. Cholecystectomy and T-tube placement/9 months-when T-tube dislodged
Cholecystojejunostomy gas;;j;tunostomy/2
Gastrojejunostomy, drainage/7 days
Cholecystojejunostomy g$rc;sefunostomy/2
Cholecystojejunostomy gm$;Ffunostomy/6
Choledochojejunostomy and gastrojejunostomy/6 months (external biliary drainage 4 days before stent placement)
Cholecystojejunostomy g$;jsafunostomy/14
Pre-Existing Surgical Drainage Procedure before the Stent Placement/Duration between Surgery and Stenting
Failed
Failed
Not attempted
Failed
Not attempted
7
on page 346
cm/l0 Fr
25 days
6 months
11 cm/11.5
Fr
3 months
9cm/11.5 Fr
0 days: PEBS ye; to
Fr
7 cm/l0
10.5 3.9
21.54 1.2
30-
25 -
13-8
5
40
Patient died of metastatic ovarian carcinoma to the lungs
None
Admitted 9 days after stent placed for fever and leukocytosis. Resolved with ceftazidime and clindamycin
lniitial exploratory laparotomy revealed massive tumor prohibiting surgical bypass. Stent placed easily but never drained well
Readmitted 1 week prior to death with bowel obstruction and perforation from massive tumor involvement: underwent exploratory laparotomy but resection not possible
Episode of sepsis immediately after stent insertion; treated with gentamicin and piperacillin
3-l
53 days
Developed UGI bleeding shortly after stem placement, resulting in less of a decrease in bilirubin
19 - 17
11 days
9cm/11.5Fr
None
Clinical Complications and Course after PEBS Placement
3.3 + 0.9
Total Bilirubin Change (w/dL)
10 months
39 days
Fr
Duration Patency of PEBS Stent
7 cm/11.5Fr
5cm/11.5Fr
7 cm/11.5
PEBS Stent Size (length/W
Continued
2 Endoscoprc 7-Fr stents that failed to drain well were replaced with PEES
Failed
Failed
Outcome of Attempted Endoscopic Stent prior to PEBS
TABLEI ClinicalStatusof 10 PatientsWhoReceivedSubsequentPercutaneous-Endoscopic BiliaryStents(PEBS)after Initial SurgicalBiliaryDrainageFailed
25 days
6 months
3 months
8 days
60 days
39 days
Improved
Improved
Improved
Not improved
Improved
Improved
Improved slightly
Improved
10 months
11 days
General Quality of life between PEBS Stenting and Death
Patient Survival From Time of PEBS Stenting
PEBS FOR REOBSTRUCTION
/ TSANG
ET AL
approach was not even attempted in four patients because of severe tumor distortion. This group of patients consisted of seven men and three women, with a median age of 64 years (range: 46 to 85 years). Eight patients had pancreatic carcinoma. One had metastatic colon carcinoma to the porta hepatis. One had cholangiocarcinoma and a subsequent metastatic ovarian carcinoma to the porta hepatis. The interval between the earlier surgical drainage and the recurrent biliary obstructions ranged from 7 days to 44 months. Of the eight patients with pancreatic carcinomas, six had both cholecystojejunostomies and gastrojejunostomies, one had both choledochojejunostomy and gastrojejunostomy, and one had only a gastrojejunostomy because cholecystojejunostomy was prevented by tumor invasion. The patient with metastatic colon carcinoma had a failed endoscopic stent placement attempt followed by a failed and complicated attempt at external-internal stent placement at another hospital; this was followed by cholecystectomy and T-tube placement, but the T-tube dislodged after 9 months. The patient with biliary obstructing metastatic ovarian carcinoma had had a prior hepatojejunostomy and Roux-en-Y reconstruction 3 years before for cholangiocarcinoma. Technique of Percutaneous-Endoscopic Stent Placement The interventional radiologist initially obtains a standard transhepatic cholangiogram, which enables the obstructed duct to be traversed by an 0.038-inch guide wire into the duodenum, over which wire a 6-Fr Teflon catheter is placed. Next, this guiding system (wire and overlying catheter) is endoscopically snared as it is pushed into the duodenum. This guide is manufactured sufficiently long so that one end is secured externally at the cutaneous entry site while the duodenal end can be removed per OS.Then a lo-Fr or 11.5-Fr endoprosthetic stent (Cotton-Lung) is placed over the end of the guide exiting the mouth and is pushed over the guide to the ampulla and through the stenosis by a stiff lo-Fr introducer tube. Once the stent is in the’ desired position, as determined by fluoroscopic monitoring, the guide system is removed per OSwhile the introducer is held firmly against the stent to prevent dislodgement. To confirm patency of the stent, radiopaque contrast can be injected through a short, temporary, 5- or 6-Fr catheter placed percutaneously into the ducts above the stent over the wire part of the guide as it is removed per OS.Finally, both the temporary transhepatic catheter and stiff stent introducer are removed (Figure 1). RESULTS Although combined PEBS placement was attempted in all 10 patients, it could not be completed in one patient. The failure in this patient, as in virtually all the very few failures so far in our experience, was due to inability to pass the guide wire through the tumor from above even on subsequent attempts; in this situation an external drain is left in place. Also, in this patient, a subsequent attempt to surgically re-establish internal bile flow failed because of tumor progression. Of the nine patients with initially successful stent placement, poor drainage occurred in one (bilirubin 346
April
1990
The American
Journal
of Medicine
Volume
88
PEBS FOR REOBSTRUCTION
/ TSANG
ET AL
Figure 1. Sequential radiographs (Patient 4) illustratingthe problem and solution to postsurgical biliary obstruction. A, percutaneous transhepatic cholangiogram (PTC) showing obstructed cholecystojejunostomy (arrow). B, PTC catheter passed through obstruction (arrows) into distal common bile duct. C, percutaneous transhepatic-peroral guiding system (open arrows) inside biliary stent (solid arrows), which is in desired position across obstruction. D, guiding system removed. Contrast injected through temporary hepatic catheter (curved arrow) shows stent is patent and bridges obstruction between bile ducts and duodenum (arrowheads).
level increased from 25 to 40 mg/dL), but earlier surgical exploration had revealed massive tumor overgrowth prohibiting surgical bypass. He died 8 days later. Seven of the patients in whom stents were placed showed significant improvement in bilirubin values and other liver function tests. The one who developed upper gastrointestinal bleeding, unrelated to the stent, showed a “false” bilirubin response. One patient with a successfully placed stent and pancreatic tumor developed sepsis immediately after stent placement, but responded well to gentamicin and piperacillin and improved to survive 39 more days. Another patient with pancreatic carcinoma developed
sepsis 9 days after stent placement but responded to ceftazidime and clindamycin to survive without jaundice another 3 months. Two of the successfully placed stents required a 4day drainage between the initial percutaneous entry into the bile ducts and the passage of a guide through tumor into the duodenum. These were considered “delayed successes” rather than “failures,” since delay time was short and the stent was eventually placed. Seven of the eight patients with functioning stents died between 11 days and 10 months from time of stent placement, the stents remaining patent (bilirubin levels staying low) until death. The eighth patient in this group is still living, with a patent stent 7 months April 1990
The American
Journal
of Medicine
Volume
88
347
PEBS FOR REOBSTRUCTION
/ TSANG
ET AL
after placement. None of the deaths were directly related to the stent. The general quality of life as measured by decreased pain, pruritis, and jaundice and increasing appetite with weight stabilization showed improvement in seven of 10 patients.
COMMENTS Having been initially unresectable and now with failed surgical biliary diversion, these 10 cases represent a distillation of palliative problems in the management of end-stage obstructive jaundice. Our 90% large-bore stent placement rate and 30% rate of overall complications are roughly similar to the rates in larger series of patients with less advanced disease. Sepsis, although usually transient and treatable, continues to plague stent placement procedures by any of the various methods. Since the advantages of the larger-bore stent have been documented [ll], we believe the 11.5 Fr is a reasonable compromise between sufficient internal diameter (2.75 mm) assuring adequate drainage and the potentially more traumatic larger sizes. The principal advantage of the combined technique, as opposed to either the percutaneous transhepatic or the endoscopic alone, seems to be the greater mechanical advantage of having fixed, firm, hands-on control of both ends of the guiding system. Thus, most, if not all, forces exerted by the introducer (“pusher”) are exerted in overcoming the tight tumor stenoses by being transmitted axially along the guide with little off-axis or buckling deformity being allowed by the fixed guide. Buckling not only dissipates force away from desired directions, but can cut or slice sideways through normal structures. In patients with Roux-enY or Billroth II who may have unapproachable ampullae via an endoscope, the percutaneously placed guide wire coming through the ampulla can be pushed through the surgically modified gut to a point accessible to an endoscopic snare [8]. If this maneuver becomes impossible, then stenting transhepatically or percutaneous drainage would be elected. Another feature of PEBS deserves emphasis since we believe it also plays a role in management of complications. As originally conceived, the procedure was seen as a collaborative effort between interventional radiologist and endoscopic gastroenterologist. A protocol has evolved for management of the jaundiced patient whose disease has progressed beyond surgical palliation. If there is not too much tumor growth deformity of gastroduodenal anatomy, the less invasive procedure of endoscopic retrograde biliary stent placement is usually attempted first. Should this fail,
348
April
1990 The American
Journal
of Medicine
Volume
88
the interventional radiologist performs percutaneous entry into intrahepatic biliary ducts by a 6-Fr coaxial catheter and proceeds to attempt guide wire passage through the obstruction. This should be successful in more than 90% of attempts. Thus, the PEBS placement can usually be completed in 1 day. Should the guide wire not pass the obstruction on the first attempt, a 6-Fr Cope loop is placed for external drainage, and follow-up re-attempt is made 3 to 4 days later. A second attempt at guide passage is successful in well over half of these initial 10% failures, and final internal stenting is completed in well over 95%. Another important function of the “team approach” is management of stent clogging or malpositioning. The endoscopist can remove the clogged stent and accomplish replacement with relative ease, assuming the ampulla is accessible, since now the tumor tract is dilated by the preexisting stent and allows replacement endoscopitally with much less force required. This report describes the feasibility and acceptable clinical results of placing completely internalized large-bore biliary stents by a less traumatic approach. Thus, a number of terminally ill patients, even after prior surgical bypass has failed, can be offered such stents, which allow them a more normal lifestyle and improved quality of life.
REFERENCES 1. Ring EJ. Oleaga JA. Freiman DB, Husted JW, Lunderquist A: Therapeutic applications of catheter cholangiography. Radiology 1978; 128: 333-338. 2. Hoevels J, Lunderquist A, lhse I: Percutaneous transhepatic intubation of brie ducts for combined internal-external drainage in preoperative and palliative treatment of obstructive jaundice. Gastrointest Radio1 1978; 3: 23-31. 3. Leung JW. Del Favero G, Cotton PB: Endoscopic brliary prostheses: a companson of materials. Gastrointest Endosc 1985; 31: 93-95. 4. Siegel JH: Improved brliary decompressjon wjth large calrber endoscoprc prostheses. Gastrointest Endosc 1984; 30: 21-23. 5. Marks WM. Freeny PC, Kozarek RA, Ball TJ: Endoscopic retrograde brliary drarnage: an alternate method of drainage. Semin Intervent Radio1 1985; 2: 74-80. 6. Siegel JH: Endoscopic decompression of the biliary tree. In: Silvrs SE, ed. Therapeutrc gastrointestinal endoscopy. New York: Igaku-Shoin, 1985; 241-268. 7. Tsang TK, Crampton AR, Bernstern JR, Ramos SR. Wieland JM: Percutaneousendoscopic brlrary stent placement: a preliminary report. Ann Intern Med 1987; 106: 389-392. 8. Tsang TK, Crampton AR, Merselman M, Desai TK, Bernstein JR Percutaneousendoscopic biliary stent placement for Billroth II and total gastrectomy with Rouxen-Y enteroenterostomv. Gastrointest Endosc 1988: 34: 45-47. 9. Kerlan RK Jr, Ring EJ: Pogany AC, Jeffrey RB Jr: Brliary endoprostheses: rnsertion usrng a combrned peroral-transhepatic method. Radiology 1984; 150: 828-
830. 10. Brambs HJ. Billmann P, Pausch J. Holstege A, Salm R: Non-surgical biliary drainage: endoscopic conversion of percutaneous transhepatic into endoprosthetic drainage. Endoscopy 1986; 18: 52-54. 11. Speer AG, Cotton PB. MacRae KD: Endoscopic management of makgnant biliary obstruction: stents of 10 F gauze are preferable to stents of 8 F gauze. Gastrointest Endosc 1988; 34: 412-417.