Diagnosis and Management of Bile Duct Strictures Robert E. Hermann, MD, Cleveland, Ohio
Strictures are the result of surgical injury to the common bile duct during cholecystectomy in the majority of patients. Less often, the injury may occur during common bile duct exploration, gastric resection, or a pancreatic operation. A stricture may result from attempts to repair a previous bile duct injury or a previous stricture. Rarely, strictures may occur without prior operation as a result of inflammation from pancreatitis, common bile duct stones, or sclerosing cholangitis; or spontaneously from a congenital stenosis that progresses to stricture. We have recently reviewed our experience with biliary strictures in 100 patients [I]. In 93 of these 100 patients the stricture was the result of a previous cholecystectomy. The stricture was found to be in the upper common bile duct or at the junction of the hepatic ducts in 54 patients; it was in the central common bile duct near the origin of the cystic duct in 28 patients, and in the distal common bile duct in 18 patients. Diagnosis As noted previously, most patients have a history of an operative procedure, usually a cholecystectomy; after the operation, reexploration for either external bile drainage, bile ascites, or jaundice indicates a probable bile duct injury. Weeks or years after this second operation, the classic symptoms of bile duct stricture (intermittent chills, fever, and jaundice) begin. The initial differential diagnosis should include a stone in the common bile .duct, sclerosing cholangitis, or chronic active hepatitis. Although carcinoma of the biliary system, ampulla of Vater, or head of the pancreas may be a consideration, the presence of intermittent jaundice and chills and fever make diagnosis of a malignant lesion unlikely. From the Department of General Surgery, The Cleveland Clinic Foundation and The Cleveland Clinic Educational Foundation, Cleveland, Ohio. Reprint requests should be addressed to Robert E. Hermann, MO, Department of General Surgery, The Cleveland Clinic Foundation, 9500 Euclid Avenue, Cleveland, Ohio 44106.
Vduma 130, November 1975
Routine laboratory studies should include a complete blood count, urinalysis, and liver function studies. The serum alkaline phosphatase level is significantly elevated, and there is moderate or significant elevation in the serum bilirubin level. Other liver function tests may also be abnormal, the bromsulphalein retention will be prolonged, and the serum glutamic oxalacetic transaminase level is often moderately elevated. High elevations in the serum glutamic oxalacetic transaminase level may indicate hepatitis. Cholangiography by the indirect or intravenous method is not usually possible when there is obstructive jaundice. Recently, two technics of direct cholangiography have been developed and are of great value in visualizing the obstructed biliary system and in identifying the site of obstruction. The first and most valuable technic is that of percutaneous transhepatic cholangiography, the direct needling of the intrahepatic bile ducts, and .instillation of radiopaque dye. (Figure 1.) This procedure should be performed the morning of the operation. If operative correction of the bile duct stricture is delayed beyond six to eight hours after the procedure, the leakage of bile from the needle hole in the liver may cause bile peritonitis. This procedure is especially valuable because it visualizes the biliary obstruction above or proximal to the point of obstruction. It provides a visual guide that clearly indicates where to look for the site of obstruction. A second technic of somewhat less value is an endoscopic retrograde cholangiogram. In this procedure the distal common bile duct is cannulated through an approach by duodenoscopy. (Figure 2.) ‘This technic is of greatest value when the stricture of the bile duct is thought to be incomplete, thus permitting visualization of the upper ducts as well. If the obstruction is thought to be complete, the visualization of the distal common bile duct alone has limited value because the upper ducts must still be found at operative exploration for effective repair of the stricture to be achieved.
519
Hermann
Figure 2. An endoscopic retrograde cholangiogram showing a stricture of the middle upper common blk duct three months after a choiecystectomy.
Figure .l. A percutaneous transhepatic cholangiogram showing a high stricture (incompfete) of the common bite duct.
Operative Management The patient with a stricture of the bile ducts should be given antibiotics preoperatively, especially if fever, chills, or an elevated white blood cell count indicate the presence of cholangitis. Antibiotics should be started prior to direct cholangiography, so that the needling or manipulation of the infected biliary system does not cause bacteremia. Our choice of antibiotic is one that is excreted in or concentrated in the bile, such as ampicillin or sodium cephalosporin. Technical factors of importance in the operative procedure include: adequate exposure, meticulous identification of the bile ducts, resection of all of the scar tissue back to normal duct structures, careful reconstruction of the duct system by whatever means chosen utilizing interrupted absorbable sutures on the inner layer of the anastomosis, repair without tension and with good blood supply, the proper utilization of T tube splints, and drainage of the subhepatic space postoperatively. The choice of repair includes a duct-to-duct repair, plastic revision or dilatation of a partial or incomplete stricture, choledochoduodenostomy, or
520
choledochojejunostomy using a Roux-en-Y segment or a loop with an enteroenterostomy below. I prefer complete resection of the stricture, rather than plastic revision or dilatation, whenever possible. In the experience of most surgeons, duct-toduct repairs do not give superior results to ductintestine repairs. Therefore, if the distal common bile duct cannot be clearly identified, is too small or too scarred, or cannot be mobilized, the choice is between choledochoduodenostomy or choledochojejunostomy. I prefer choledochoduodenostomy to choledochojejunostomy, because it can usually be performed with less surgical dissection and is more physiologic since bile is returned to the duodenum. The duodenum should be adequately mobilized so that careful repair without tension can be made. When choledochojejunostomy is employed, I prefer a Roux-en-Y jejunal segment to a jejunal loop with an enteroenterostomy below the anastomosis. If the common bile duct is greatly thickened and dilated, an indwelling splint may not be necessary or advisable. However, if the duct is thin walled or small, then either a T tube, Y tube, or free floating straight catheter splint should be used. If splints are used, they should be large (size 16 to 22F), the external limb should be brought out above or below but not through the anastomosis, and the splints should remain in place from six weeks to three months. (Figures 3 and 4.) If repair of the stricture is unusually difficult, if scar tissue remains at the anastomosis, or if the stricture has recurred several times, it may be advisable to leave The American Jwmal cl Sw~
Bile Duct Strictures
Figure 3. Choledochoduodenostomy w/th a free floating straight catheter (A) and a T tube ex/t/ng from the duodenum ( 6).
the T tube splint in place for months or years. This is less than ideal, however, as the presence of the tube is a nidus for infection and for the sedimentation and precipitation of bile salts on the tube. The tube wall will eventually plug completely and need to be removed. In such a situation, with current roentgenographic technics, it is occasionally possible to replace a large T tube with a slightly smaller one that has one limb cut short (an L tube) back through the biliary intestinal anastomosis, using special instruments under image intensification, fluoroscopic control.
as a result of biliary stricture. In the 1940’s and 1950’s, reports from the Lahey Clinic and Mayo Clinic showed a 43 to 68 per cent incidence of satisfactory or good results three or more years after operative repair [2, 31. Recent reports from these institutions, as well as from others, have shown further improvement; most authors now report good or satisfactory results that approach 75 to 85 per cent three to five years after stricture repair [4-61. These improved results are probably due to a number of factors: (1) less delay between the initial bile duct injury and the first repair, and less
Results In the 100 patients undergoing operation at the Cleveland Clinic, 59 had choledochoduodenostomy, 11 had resection of the stricture and ductto-duct repair, 9 had a choledochojejunostomy, 7 had plastic revision of the stricture, and 14 had dilation and placement of a T tube splint through a difficult recurrent stricture. In this group of patients, followed for two to twenty years, excellent or good results (no further symptoms of biliary disease or mild intermittent symptoms that subside with medical treatment) were obtained in 84 patients; 16 had recurring episodes of cholangitis or jaundice or required another operation on the biliary tract. The results achieved with choledochoduodenostomy appear to be slightly superior to the results achieved by the other methods. (Table I.1 Comments The results of operative repair of strictures of the common bile duct have steadily improved through the years. At one time, it was reported that one third of all patients died within five years Volume 130, November 1975
Figure 4. ChoIedochojejunostomy using a Roux-en-Y segment with a T tube in place across the anastomosis.
521
Hermann
TABLE
I
Results of Bile Duct
Repair
in 100 Patients,
1951 to 1971 _
___Excellent to Good
Procedure Choledochoduodenostomy Resection of stricture, duct-to-duct repair Choledochojejunostomy Plastic revision of stricture Dilatation of stricture
Poor
Per
Nom-
Per
Number
Cent
ber
Cent
53 8
90 73
6 3
10 27
7 5 11
78 71 78
2 2 3
22 29 22
delay when a recurrent stricture of the common bile duct occurs prior to reconstruction of the bile ducts; (2) the use of antibiotics to decrease or control infection in the obstructed or partially obstructed biliary system; (3) improved methods of diagnosis of biliary stricture by means of percutaneous transhepatic cholangiography and endoscopic retrograde cholangiography; and (4) improved technics of operative repair. Although duct-to-duct repairs were initially favored by many authors, recent reports have shown no advantage to this type of reconstruction. A duct-intestine repair gives somewhat better results; it can be performed at a higher level with more adequate resection of the site of previous injury and scar tissue, and normal healthy tissue with a normal blood supply can be utilized for the lower segment of the repair. Although many surgeons prefer choledochojejunostomy, we prefer choledochoduodenostomy when the duodenum is mobile, because it can be performed more quickly with less operative dissection, it is more physiologically correct in that bile is returned to the duodenum, and the long-term results are excellent. As long as the anastomosis remains patent, the fears of ascending or reflux cholangitis are unfounded. In addition, with current endoscopic technics, this anastomosis can be inspected by the endoscopist and a retrograde cholangiogram obtained when there is any question of recurrent stricture formation. The length of time that splints (T tube, Y tube, or straight rubber catheters) should be left in place after bile duct reconstruction, or whether they should be used at all, is a subject of some controversy [I]. The degree of scarring found at the time of operation, the anatomic location of the stricture, the number of previous repairs, and the adequacy of the operative procedure in performing
522
the repair will all be influencing factors as to whether or not an internal splint is used, the type of splint or tube chosen, and how long it should remain in place. We have had excellent results in some patients with thickened, dilated ducts in whom splints were not used. We prefer to remove splints relatively early, from six weeks to three months after most bile duct reconstructions, believing that the splint acts as a foreign body irritant to the anastomosis as well as a potential nidus for infection. However, with some difficult repairs, or with those that have recurred several times, or with those in which dilatation of the stricture was the only method possible, we have left stents in place for two years or longer to provide internal dilatation of the repair and prevent another recurrent stricture. We are now working with our radiologists to develop technics whereby such tubes could be removed at six to eight month intervals and a new tube replaced under fluoroscopic control. Summary In the majority of patients, strictures of the common bile duct result from an injury to the duct. The earlier the injury is repaired or the more quickly the diagnosis of stricture is made and repaired, with antibiotics to prevent infection, the better are the results of the repair. Five types of repair have been utilized: (I) choledochoduodenostomy, (2) duct-to-duct repair, (3) choledochojejunostomy, (4) plastic revision of the stricture, or (5) dilatation of a previous stricture when multiple, previous operative procedures or the difficulty of operative exposure proves too great for adequate repair. Our overall results after repair of biliary stricture are good or excellent in 84 per cent of patients. We have had the best results in patients in whom choledochoduodenostomy was performed. References 1. 2. 3. 4. 5. 6.
Hertzer
NR, Gray HW, Hoerr SO, Hermann RE: The use of Ttube splints in bile duct repairs. Surg Gynecol Obstet 137: 413,1973. Cattell RB, Braasch JW: Primary repair of benign strictures of the bile duct. Surg GynecolObstet 109: 531, 1959. Walters W, Nixon JW Jr, Hodgins TE: Strictures of the common bile duct. Ann Surg 149: 781, 1959. Aust JB, Root HD, Urdaneta L, Varco RL: Biliiry stricture. Surgtwy62: 601.1967. Cosman B, Porter MR: Benign stricture of the bile duct. Ann Surg 152: 730, 1960. Warren KW. McDonald WM: Facts and fiction regarding strictures of the extrahepatic bile ducts. Ann Surg 159: 996, 1964.
The American Journal of Surgery
Strictures are the result of surgical injury to the common bile duct during cholecystectomy in the majority of patients. Less often, the injury may occur during common bile duct exploration, gastric resection, or a pancreatic operation. A stricture may result from attempts to repair a previous bile duct injury or a previous stricture. Rarely, strictures may occur without prior operation as a result of inflammation from pancreatitis, common bile duct stones, or sclerosing cholangitis; or spontaneously from a congenital stenosis that progresses to stricture. We have recently reviewed our experience with biliary strictures in 100 patients [I]. In 93 of these 100 patients the stricture was the result of a previous cholecystectomy. The stricture was found to be in the upper common bile duct or at the junction of the hepatic ducts in 54 patients; it was in the central common bile duct near the origin of the cystic duct in 28 patients, and in the distal common bile duct in 18 patients. Diagnosis As noted previously, most patients have a history of an operative procedure, usually a cholecystectomy; after the operation, reexploration for either external bile drainage, bile ascites, or jaundice indicates a probable bile duct injury. Weeks or years after this second operation, the classic symptoms of bile duct stricture (intermittent chills, fever, and jaundice) begin. The initial differential diagnosis should include a stone in the common bile .duct, sclerosing cholangitis, or chronic active hepatitis. Although carcinoma of the biliary system, ampulla of Vater, or head of the pancreas may be a consideration, the presence of intermittent jaundice and chills and fever make diagnosis of a malignant lesion unlikely. From the Department of General Surgery, The Cleveland Clinic Foundation and The Cleveland Clinic Educational Foundation, Cleveland, Ohio. Reprint requests should be addressed to Robert E. Hermann, MO, Department of General Surgery, The Cleveland Clinic Foundation, 9500 Euclid Avenue, Cleveland, Ohio 44106.
Vduma 130, November 1975
Routine laboratory studies should include a complete blood count, urinalysis, and liver function studies. The serum alkaline phosphatase level is significantly elevated, and there is moderate or significant elevation in the serum bilirubin level. Other liver function tests may also be abnormal, the bromsulphalein retention will be prolonged, and the serum glutamic oxalacetic transaminase level is often moderately elevated. High elevations in the serum glutamic oxalacetic transaminase level may indicate hepatitis. Cholangiography by the indirect or intravenous method is not usually possible when there is obstructive jaundice. Recently, two technics of direct cholangiography have been developed and are of great value in visualizing the obstructed biliary system and in identifying the site of obstruction. The first and most valuable technic is that of percutaneous transhepatic cholangiography, the direct needling of the intrahepatic bile ducts, and .instillation of radiopaque dye. (Figure 1.) This procedure should be performed the morning of the operation. If operative correction of the bile duct stricture is delayed beyond six to eight hours after the procedure, the leakage of bile from the needle hole in the liver may cause bile peritonitis. This procedure is especially valuable because it visualizes the biliary obstruction above or proximal to the point of obstruction. It provides a visual guide that clearly indicates where to look for the site of obstruction. A second technic of somewhat less value is an endoscopic retrograde cholangiogram. In this procedure the distal common bile duct is cannulated through an approach by duodenoscopy. (Figure 2.) ‘This technic is of greatest value when the stricture of the bile duct is thought to be incomplete, thus permitting visualization of the upper ducts as well. If the obstruction is thought to be complete, the visualization of the distal common bile duct alone has limited value because the upper ducts must still be found at operative exploration for effective repair of the stricture to be achieved.
519
Hermann
Figure 2. An endoscopic retrograde cholangiogram showing a stricture of the middle upper common blk duct three months after a choiecystectomy.
Figure .l. A percutaneous transhepatic cholangiogram showing a high stricture (incompfete) of the common bite duct.
Operative Management The patient with a stricture of the bile ducts should be given antibiotics preoperatively, especially if fever, chills, or an elevated white blood cell count indicate the presence of cholangitis. Antibiotics should be started prior to direct cholangiography, so that the needling or manipulation of the infected biliary system does not cause bacteremia. Our choice of antibiotic is one that is excreted in or concentrated in the bile, such as ampicillin or sodium cephalosporin. Technical factors of importance in the operative procedure include: adequate exposure, meticulous identification of the bile ducts, resection of all of the scar tissue back to normal duct structures, careful reconstruction of the duct system by whatever means chosen utilizing interrupted absorbable sutures on the inner layer of the anastomosis, repair without tension and with good blood supply, the proper utilization of T tube splints, and drainage of the subhepatic space postoperatively. The choice of repair includes a duct-to-duct repair, plastic revision or dilatation of a partial or incomplete stricture, choledochoduodenostomy, or
520
choledochojejunostomy using a Roux-en-Y segment or a loop with an enteroenterostomy below. I prefer complete resection of the stricture, rather than plastic revision or dilatation, whenever possible. In the experience of most surgeons, duct-toduct repairs do not give superior results to ductintestine repairs. Therefore, if the distal common bile duct cannot be clearly identified, is too small or too scarred, or cannot be mobilized, the choice is between choledochoduodenostomy or choledochojejunostomy. I prefer choledochoduodenostomy to choledochojejunostomy, because it can usually be performed with less surgical dissection and is more physiologic since bile is returned to the duodenum. The duodenum should be adequately mobilized so that careful repair without tension can be made. When choledochojejunostomy is employed, I prefer a Roux-en-Y jejunal segment to a jejunal loop with an enteroenterostomy below the anastomosis. If the common bile duct is greatly thickened and dilated, an indwelling splint may not be necessary or advisable. However, if the duct is thin walled or small, then either a T tube, Y tube, or free floating straight catheter splint should be used. If splints are used, they should be large (size 16 to 22F), the external limb should be brought out above or below but not through the anastomosis, and the splints should remain in place from six weeks to three months. (Figures 3 and 4.) If repair of the stricture is unusually difficult, if scar tissue remains at the anastomosis, or if the stricture has recurred several times, it may be advisable to leave The American Jwmal cl Sw~
Bile Duct Strictures
Figure 3. Choledochoduodenostomy w/th a free floating straight catheter (A) and a T tube ex/t/ng from the duodenum ( 6).
the T tube splint in place for months or years. This is less than ideal, however, as the presence of the tube is a nidus for infection and for the sedimentation and precipitation of bile salts on the tube. The tube wall will eventually plug completely and need to be removed. In such a situation, with current roentgenographic technics, it is occasionally possible to replace a large T tube with a slightly smaller one that has one limb cut short (an L tube) back through the biliary intestinal anastomosis, using special instruments under image intensification, fluoroscopic control.
as a result of biliary stricture. In the 1940’s and 1950’s, reports from the Lahey Clinic and Mayo Clinic showed a 43 to 68 per cent incidence of satisfactory or good results three or more years after operative repair [2, 31. Recent reports from these institutions, as well as from others, have shown further improvement; most authors now report good or satisfactory results that approach 75 to 85 per cent three to five years after stricture repair [4-61. These improved results are probably due to a number of factors: (1) less delay between the initial bile duct injury and the first repair, and less
Results In the 100 patients undergoing operation at the Cleveland Clinic, 59 had choledochoduodenostomy, 11 had resection of the stricture and ductto-duct repair, 9 had a choledochojejunostomy, 7 had plastic revision of the stricture, and 14 had dilation and placement of a T tube splint through a difficult recurrent stricture. In this group of patients, followed for two to twenty years, excellent or good results (no further symptoms of biliary disease or mild intermittent symptoms that subside with medical treatment) were obtained in 84 patients; 16 had recurring episodes of cholangitis or jaundice or required another operation on the biliary tract. The results achieved with choledochoduodenostomy appear to be slightly superior to the results achieved by the other methods. (Table I.1 Comments The results of operative repair of strictures of the common bile duct have steadily improved through the years. At one time, it was reported that one third of all patients died within five years Volume 130, November 1975
Figure 4. ChoIedochojejunostomy using a Roux-en-Y segment with a T tube in place across the anastomosis.
521
Hermann
TABLE
I
Results of Bile Duct
Repair
in 100 Patients,
1951 to 1971 _
___Excellent to Good
Procedure Choledochoduodenostomy Resection of stricture, duct-to-duct repair Choledochojejunostomy Plastic revision of stricture Dilatation of stricture
Poor
Per
Nom-
Per
Number
Cent
ber
Cent
53 8
90 73
6 3
10 27
7 5 11
78 71 78
2 2 3
22 29 22
delay when a recurrent stricture of the common bile duct occurs prior to reconstruction of the bile ducts; (2) the use of antibiotics to decrease or control infection in the obstructed or partially obstructed biliary system; (3) improved methods of diagnosis of biliary stricture by means of percutaneous transhepatic cholangiography and endoscopic retrograde cholangiography; and (4) improved technics of operative repair. Although duct-to-duct repairs were initially favored by many authors, recent reports have shown no advantage to this type of reconstruction. A duct-intestine repair gives somewhat better results; it can be performed at a higher level with more adequate resection of the site of previous injury and scar tissue, and normal healthy tissue with a normal blood supply can be utilized for the lower segment of the repair. Although many surgeons prefer choledochojejunostomy, we prefer choledochoduodenostomy when the duodenum is mobile, because it can be performed more quickly with less operative dissection, it is more physiologically correct in that bile is returned to the duodenum, and the long-term results are excellent. As long as the anastomosis remains patent, the fears of ascending or reflux cholangitis are unfounded. In addition, with current endoscopic technics, this anastomosis can be inspected by the endoscopist and a retrograde cholangiogram obtained when there is any question of recurrent stricture formation. The length of time that splints (T tube, Y tube, or straight rubber catheters) should be left in place after bile duct reconstruction, or whether they should be used at all, is a subject of some controversy [I]. The degree of scarring found at the time of operation, the anatomic location of the stricture, the number of previous repairs, and the adequacy of the operative procedure in performing
522
the repair will all be influencing factors as to whether or not an internal splint is used, the type of splint or tube chosen, and how long it should remain in place. We have had excellent results in some patients with thickened, dilated ducts in whom splints were not used. We prefer to remove splints relatively early, from six weeks to three months after most bile duct reconstructions, believing that the splint acts as a foreign body irritant to the anastomosis as well as a potential nidus for infection. However, with some difficult repairs, or with those that have recurred several times, or with those in which dilatation of the stricture was the only method possible, we have left stents in place for two years or longer to provide internal dilatation of the repair and prevent another recurrent stricture. We are now working with our radiologists to develop technics whereby such tubes could be removed at six to eight month intervals and a new tube replaced under fluoroscopic control. Summary In the majority of patients, strictures of the common bile duct result from an injury to the duct. The earlier the injury is repaired or the more quickly the diagnosis of stricture is made and repaired, with antibiotics to prevent infection, the better are the results of the repair. Five types of repair have been utilized: (I) choledochoduodenostomy, (2) duct-to-duct repair, (3) choledochojejunostomy, (4) plastic revision of the stricture, or (5) dilatation of a previous stricture when multiple, previous operative procedures or the difficulty of operative exposure proves too great for adequate repair. Our overall results after repair of biliary stricture are good or excellent in 84 per cent of patients. We have had the best results in patients in whom choledochoduodenostomy was performed. References 1. 2. 3. 4. 5. 6.
Hertzer
NR, Gray HW, Hoerr SO, Hermann RE: The use of Ttube splints in bile duct repairs. Surg Gynecol Obstet 137: 413,1973. Cattell RB, Braasch JW: Primary repair of benign strictures of the bile duct. Surg GynecolObstet 109: 531, 1959. Walters W, Nixon JW Jr, Hodgins TE: Strictures of the common bile duct. Ann Surg 149: 781, 1959. Aust JB, Root HD, Urdaneta L, Varco RL: Biliiry stricture. Surgtwy62: 601.1967. Cosman B, Porter MR: Benign stricture of the bile duct. Ann Surg 152: 730, 1960. Warren KW. McDonald WM: Facts and fiction regarding strictures of the extrahepatic bile ducts. Ann Surg 159: 996, 1964.
The American Journal of Surgery
