- Email: [email protected]
Extrahepatic biliary obstruction secondary to metastatic cancer
Extrahepatic Biliary Obstruction Secondary to Metastatic Cancer James H. Thomas, MD, Kansas City, Kansas George E. Pierce, MD, Kansas City, Kansas Charles Karlin, MD, Kansas City, Kansas Arlo S. Hermreck, MD, PhD, Kansas City, Kansas Richard I. MacArthur, MD, Kansas City, Kansas
Jaundice occurs frequently in patients with cancer and is often the result of neoplastic extrahepatic biliary tract obstruction. The recent development of sonography, transhepatic cholangiography, endoscopic retrograde cholangiography and computed axial tomography has provided relatively simple methods of assessing these patients. Many patients with obstruction of the biliary tract secondary to metastasis can obtain effective palliation for months or years with current biliary decompression procedures. The evaluation and management of 30 patients with extrahepatic biliary obstruction secondary to metastatic cancer are described herein. Material
and Methods
The University of Kansas Medical Center records of 30 patients who were found to have obstruction of the extrahepatic bile ducts secondary to metastatic cancer were reviewed to determine the site of the primary cancer, the interval between treatment of the primary cancer and admission for jaundice, the presenting symptoms and signs, the procedures used for diagnosis, the sites of biliary obstruction, the types and success of therapy, and the duration of survival. Cumulative survival rates were calculated for all patients in whom successful palliation was obtained.
Results From 1975 to 1980, 30 patients with biliary obstruction secondary to metastatic cancer were evaluated. The ages of the patients ranged from 31 to 83 years (mean 57). Sixteen of the patients were women and 14 were men. Biliary-enteric anastomosis and percutaneous transhepatic intubation of the bile ducts were the primary techniques employed for biliary decompression. Site of the primary tumor: Thirty percent (nine) of the primary tumors were colonic carcinomas. Six of the nine colonic cancers (66 nercent) were located From the Department of Surgery, University of Kansas Medical Center, Kansas City, Kansas. Requests for reprints should be addressed to James H. Thomas, MD, Department of Surgery, University of Kansas Medical Center, Kansas City, Kansas 66103. Presented at the 33rd Annual Meeting of the Southwestern Surgical Congress, Monterey, California, May 4-7, 1981.
770
in the right colon and three were in the left colon. The second most common primary site was the gallbladder (five), followed by extremity melanomas (three), pancreas (three), small bowel (three), breast (two), stomach (two), ovary (two) and non-Hodgkin’s lymphoma (one). Interval between treatment of the primary tumor and development of jaundice: The interval between treatment of the primary tumor and admission for jaundice ranged from less than 1 day to 192 months, with a mean duration of 36 f 8.4 months. Four patients, one with cancer of the gallbladder and three with colon cancer, presented with jaundice at the time of the initial diagnosis of cancer; these patients had no obstruction-free interval. The longest obstruction-free interval was 192 months in a patient with ovarian cancer. The obstruction-free interval in nine patients with colonic cancer ranged from less than 1 day to 95 months (mean 29 f 10.6 months). In five patients with gallbladder cancer the obstruction-free interval ranged from less than 1 day to 36 months (mean 15.0 f 7 months). In the 26 patients with successful palliation there was no correlation between the obstruction-free interval and the level of bilirubin, the site of obstruction or the duration of survival after decompression. Symptoms, signs and laboratory data on admission: Twenty (70 percent) of the patients had jaundice on admission. Pain was present in 10 patients (30 percent). Icterus was present on examination in 27 patients (90 percent). In the three patients without clinical jaundice the serum bilirubin level was 1.8,2.0 and 2.3 mg/dl. The serum bilirubin in the patients with clinical jaundice ranged from 6.3 to 35.4 mg/dl (mean 16.7 f 1.8). The mean serum bilirubin level was 16.7 f 1.9 mg/dl in the 15 patients with obstruction of the common bile duct, 19 f 3 mg/dl in the 6 patients with obstruction of the common hepatic duct, 18.7 f 8 mg/dl in the 3 patients with obstruction at the bifurcation of the hepatic ducts, and 8.3 f 3.2 mgldl in the 3 patients with obstruction of the left or right hepatic duct or both. The level of serum bilirubin on admission did not corre-
The American Journal of Surgery
Extrahepatic Biliary Obstruction
late with the site of obstruction and did not predict the duration of survival. Diagnostic procedures: Percutaneous transhepatic cholangiography was used for diagnosis in 19 patients, sonography was used in 12 patients, computed axial tomography of the abdomen in 7 patients and endoscopic retrograde cholangiography in 1 patient. In seven patients the diagnosis of extrahepatic biliary obstruction was made at laparotomy. Percutaneous transhepatic cholangiography successfully demonstrated the site of obstruction in all patients so studied. Obstruction of the bile ducts was suspected and confirmed preoperatively in one of the three patients with serum bilirubin concentrations of less than 2.5 mg/dl. Site of biliary tract obstruction: The common bile duct was obstructed in 18 patients (60 percent), the common hepatic duct in 6 patients (20 percent), the bifurcation of the common hepatic duct in 3 patients (10 percent), and the right or left hepatic duct, or both, in 3 patients (10 percent). No patient had multiple sites of obstruction or had a new site of obstruction after biliary decompression. Neither the level of serum bilirubin on admission nor the duration of palliation correlated with the site of obstruction. Therapy: Biliary decompression using radiologic (external and internal drainage) or surgical procedures, or both, was successful in 27 (90 percent) of the patients. Percutaneous transhepatic biliary drainage by placement of an internal stent was attempted in 14 patients and was successful in 8 (57 percent). Of the six patients (43 percent) in whom percutaneous transhepatic internal biliary drainage failed, two received no additional therapy, two underwent conversion to percutaneous transhepatic external biliary drainage, and two underwent operation. One of these two patients was managed by tube hepatostomy, and the other was managed by percutaneous transhepatic external biliary drainage after failure of operative decompression. One patient with obstruction of both hepatic ducts underwent internal drainage of the right hepatic duct and left hepaticojejunostomy. Complications of internal biliary drainage occurred in four patients (50 percent) and included catheter infection in one, catheter occlusion in two and catheter dislodgment in one. These patients were managed by antibiotics (one), catheter replacement (one) and conversion to external drainage (two). Three patients died within 30 days of successful insertion of a biliary endoprosthesis. Surgical decompression was attempted in 19 patients and was successful in 16 (84 percent). Operative decompression was successful in one patient in whom internal drainage failed, and external drainage was successful in one patient in whom surgery failed. The most common surgical procedures were Roux-Y choledochojejunostomy (five), T-tube insertion (four), choledochoduodenostomy (four), cholecys-
Volume
142, December
1961
0 30 60 90 120
150
DURATION
180
210
SURVIVAL
240
270
300
330
360
(DAYS)
Figure 1. Cumulative survival in 27 patients with extrahepatic biliary obstruction secondary to metastatic cancer.
tojejunostomy (two), tube hepatostomy (two), Utube insertion (one) and hepaticojejunostomy (one). Two patients underwent a second operation after failure of an initial procedure. Both initial operations (T-tube insertions) were followed by choledochoduodenostomy. One of the two choledochoduodenostomies functioned. In two patients only laparotomy was performed; decompression was not attempted. Two patients died within 30 days of surgery. One had obtained successful palliation, but operation failed to provide decompression in the other patient. Success of decompression and duration of palliation: The total bilirubin level was measured two weeks after biliary decompression and ranged from 1.2 to 14.6 mg/dl (mean concentration 4.9 f 0.77). The decrease from preoperative levels (6.3 to 35.4 mg/dl; mean 16.7 f 1.8) was significant (p < 0.05). After decompression mean bilirubin levels were 4.05 f 0.77 mgldl in patients treated by external and internal drainage, and 6.08 f 1.15 mgldl in patients treated by operat,ion (difference not significant). The duration of survival in patients treated successfully by radiologic or surgical procedures, or both, ranged from 30 to 700 days (mean 197 f 36). In eight patients treated by radiologic methods, the mean duration of survival was 60 f 11 days, compared with 270 f 49 days in patients treated surgically (p < 0.05). Six of the 27 patients (22 percent) who obtained successful palliation are alive, with a mean duration of survival of 175 f 35 days. Cumulative survival rates were calculated for all patients who obtained successful decompression (Figure 1). Comments Kopelson et al [I], in a review of patients with extrahepatic biliary tract obstruction from metastatic breast carcinoma, noted that little attention has been given in surgical reports to the palliation of patients with biliary obstruction secondary to metastatic solid tumors. They describe 28 patients with biliary tract obstruction who were managed by operation, radiologic placement of an endoprosthesis, chemotherapy
771
Thomas et al
and radiotherapy. Warshaw and Welch [2] reported on eight patients wtih a bile duct obstruction secondary to metastatic colonic carcinoma. Six were managed by surgery and two by irradiation. The length of survival after decompression ranged from 1 to 36 months. Likewise, Papachristou and Fortner [3] reported on 30 patients in whom jaundice developed 1 week to 16 weeks after gastric resection for carcinoma. All 30 patients underwent exploration, and palliation was achieved in 15. The length of postoperative survival averaged 6.4 months. In the review of Kopelson et al [I], metastatic gastric carcinoma was the most frequent cause of extrahepatic obstruction (50 percent), followed by colon carcinoma (20 percent). A majority of the patients in our series (nine) had colon cancer. Five patients had cancer of the gallbladder and three had melanoma. Only two patients had gastric and breast carcinoma. Although the incidence of carcinoma of the left colon is greater than that of the right, six of our nine patients with extrahepatic biliary obstruction secondary to metastatic colon carcinoma had primary lesions in the right colon. Of the eight patients with biliary obstruction secondary to colonic carcinbma described by Warshaw and Welch [2], four had cancer of the right colon. These data suggest that right colon cancers may metastasize to the periportal lymph nodes more often than those of the left colon. The most frequent site of obstruction in the present series was the common bile duct. Likewise, Popp et al [4] found that the most frequent level of biliary obstruction was the common bile duct in seven patients with biliary obstruction secondary to metastatic breast cancer, and Papachristou and Fortner [3] reported obstruction in the distal common bile duct in 19 of 30 patients with biliary obstruction secondary to metastases from gastric carcinoma. In contrast, five of the eight patients described by Warshaw and Welch [2] were obstructed at the bifurcation of the common hepatic duct. Although there was no correlation between the site of obstruction and the duration of survival, two of the four failures in our series were in patients with obstruction of the bifurcation of the common hepatic duct of either one or both hepatic ducts. Percutaneous transhepatic cholangiography was the most useful technique for confirming the diagnosis and demonstrating the site of biliary obstruction in our patients. No complications of cholangiography occurred in this series, although bleeding, sepsis, hepatic hematoma and bile leaks are well recognized complications of this procedure. Papachristou and Fortner [3], used laparotomy as their major diagnostic procedure, and in 7 of our 30 patients laparotomy was used to establish the diagnosis. Biliary decompression was successful in 90 percent of the patients in this series. The success rate was 57 772
percent for those treated by radiologic placement of an endoprosthesis, 100 percent for those decompressed by external drainage, and 84 percent for those treated by surgical decompression. The surgical success rate in this series compares favorably with that of 50 percent reported by Papachristou and Fortner [3] in patients with metastatic gastric cancer and with the 83 percent success rate in patients reviewed by Warshaw and Welch [2]. The lower success rate with percutaneous internal biliary drainage in the present series may reflect initial problems with technique and selection of patients with more advanced disease compared with those treated surgically. Recent reports [5] indicate that immediate or early success rates may approach 80 percent for percutaneous decompression in patients with malignant obstruction. A majority of reports [6,7] suggest that nonsurgical techniques afford the best approach to biliary decompression, primarily because of the reported operative mortality rates of 10 to 20 percent [8] in patients who are operated on for decompression of the biliary tract obstructed by carcinoma of the pancreas. It must be pointed out, however, that these patients have far advanced disease and that their operative mortality is probably not representative of patients with hepatic obstruction secondary to other types of malignancy. Thirty day mortality rates after nonsurgical decompression are not reported by workers who advocate percutaneous transhepatic biliary decompression. However, Pereiras et al [9] reported two early deaths (at less than 30 days) as a result of hepatic abscess in 12 patients with percutaneous placement of biliary endoprosthesis. In the present study 5 of 30 patients (16 percent) died within 30 days of treatment. There was no significant difference in mortality for surgical (two deaths) and nonsurgical decompression (three deaths). On the other hand, long-term survival was better in patients with surgical decompression (270 f 49 days) than in those with percutaneous transhepatic decompression (60 f 11 days). Again, as noted previously, the longer duration of survival noted with operation may be attributable to patient selection rather than or in addition to possible superiority of surgical therapy in achieving long-term decompression. This study as well as others reveal that very few institutions have had a large experience with extrahepatic biliary obstruction secondary to metastatic carcinoma. Therefore, conclusions concerning the superiority of one approach versus the other are at best tentative and may not apply to certain groups of patients. For example, Popp et al [4] suggested using transhepatic biliary decompression in patients with high biliary obstruction to avoid the mortality associated with intrahepatic biliary-enteric anastomosis. Cahow [IO], however, noted that high obstructions cause hepatic abscess if only one lobe is decompressed, as is often the case with percutaneous The American Journal of Surgery
Extrahepatic Biliary Obstruction
procedures. He suggested that intrahepatic cholangiojejunostomy was the preferred method of treating these patients. He used that procedure 14 times in 11 patients without serious complications and obtained a mean duration of survival of 16 months. Patients with severe jaundice present a special problem. Since a bilirubin level of greater than 10 mg/dl may be associated with increased operative mortality [II], initial decompression of the biliary tract by percutaneous transhepatic techniques should be considered. Staged definitive operation can then be performed after the improvement of hepatic function. In spite of the problems inherent in the interpretation of data from retrospective studies, this experience suggests that (1) a large number of cancer patients with jaundice have extrahepatic biliary obstruction and may benefit from biliary decompression, (2) transhepatic percutaneous cholangiography, sonography and computed tomographic scans provide highly accurate diagnostic and anatomic information with minimal patient discomfort, (3) percutaneous transhepatic catheterization provides adequate decompression in many patients with biliary obstruction secondary to cancer, and may be of value in preparing patients for subsequent operative decompression, and (4) surgical decompression can be achieved with acceptable operative mortality in selected patients and may afford better palliation than percutaneous drainage. Summary Thirty patients with extrahepatic biliary obstruction secondary to metastatic cancer were reviewed to determine the sites of the primary tumor, diagnostic methods, therapy and success of palliation. Colon carcinoma was the most common primary tumor, and the common bile duct was most often obstructed. Both percutaneous transhepatic and surgical decompression of the biliary tract were employed. Twenty-seven (90 percent) of the patients obtained successful palliation. The length of survival averaged 270 f 49 days in patients treated surgically compared with 60 f 11 days in patients who underwent decompression by radiologic techniques. Mortality was not increased in patients undergoing operative biliary drainage. Surgical decompression may be the best method for managing patients with biliary obstruction secondary to metastatic cancer. References 1. Kopelson G, Chu AM, Doucette JA, Gunderson LL. Extra-hepatic biliary tract metastases from breast cancer. Int J Radiat Oncol Biol Phys 1980;8:497-504. 2. Warshaw AL, Welch JP. Extrahepatic biliary obstruction by metastatic colon carcinoma. Ann Surg 1978;188:593-7. 3. Papachristou D, Fortner JG. Biliary obstruction after gastrectomy for carcinoma of the stomach. Surg Gynecol Obstet 1978;147:401-4. 4. Popp JW. Schapiro RI-f, Warshaw AL. Extrahepatic biliary obstruction caused by metastatic breast carcinoma. Ann Intern
Volume 142, December 1981
Med 1979;91:568-71. 5. Burcharth F, Jensen LI, Olesen K. Endoprosthesis for internal drainage of the biliary tract. Gastroenterology 1979;77: 133-7. 6. Pollock TW, Ring ER, Oleaga JA, Freiman DB, Mullen JL, Rosato EF. Percutaneous decompression of benign and malignant biliary obstruction. Arch Surg 1979;114:148-51. 7. MacCarty RL. Nonsurgical management of obstructive jaundice in the patient with advanced cancer. JAMA 1980;244: 1976-8. 8. Buckwalter JA, Lawton RL, Tidrick RT. Bypass operations for neoplastic biliary tract obstruction. Am J Surg 1965;109: 100-6. 9. Pereiras RV Jr, Rheingold OJ, Hutson D, et al. Relief of malignant obstructive jaundice by percutaneous insertion of a permanent prosthesis in the biliary tree. Ann Intern Med 197889589-93. 10. Cahow GE. Alternative methods for hepatic decompression. Surg Clin North Am 1980;60:1305-14. 11. Denning DA, Ellison EC, Carey LC. Preoperative percutaneous transhepatic biliary decompression lowers operative morbidity in patients with obstructive jaundice. Am J Surg 1981;141:61-4.
Discussion Kent C. Westbrook (Little Rock, AR): I think Dr. Thomas has given us a clear view of a difficult problem that most of us don’t like to have to deal with. First, we have to realize that our initial treatment has failed, and second, whatever we do from then on is probably going to fail because the patients will be dead within a year or 2. But palliative manipulation of patients is a worthwhile endeavor. I have a problem comparing the patients treated surgically with those treated radiologically. I go directly to percutaneous transhepatic cholangiography. Computed tomography and ultrasound have not been of much value. If cholangiography shows a low obstruction in a patient who is a good risk, I am eager to operate. What do you do when a patient who had colon resection 2 years ago has jaundice? Do you look for liver metastasis or extrahepatic obstruction first? Second, assuming you find a patient with extrahepatic obstruction and relieve the obstruction either surgically or with an endoprosthesis, then do you give chemotherapy or radiotherapy or just follow them along? William R. Schiller (Albuquerque, NM): It is difficult to know how to choose between the surgical or radiographic options. I would appreciate your comments on it. James H. Thomas (closing): We evaluate these patients using sonography and liver and computed axial tomographic scans. If the patient does not have an obstruction, percutaneous transhepatic cholangiography is performed. If the patient’s bilirubin level is greater than 10 mg/lOO ml, we use decompression for several days before operation. I agree that if a high obstruction is demonstrated, the inclination is to allow the radiologist to place an endoprosthesis. Treatment after decompression, whether chemotherapy or radiotherapy, is based on whether the primary tumor is responsive to that sort of therapy. We think at present that surgery in selected patients is probably the most reasonable alternative. If a patient has external drainage, catheter maintenance is a problem. Complications are less likely to be a problem with surgical decompression. Again, patient selection is the key.
773
Jaundice occurs frequently in patients with cancer and is often the result of neoplastic extrahepatic biliary tract obstruction. The recent development of sonography, transhepatic cholangiography, endoscopic retrograde cholangiography and computed axial tomography has provided relatively simple methods of assessing these patients. Many patients with obstruction of the biliary tract secondary to metastasis can obtain effective palliation for months or years with current biliary decompression procedures. The evaluation and management of 30 patients with extrahepatic biliary obstruction secondary to metastatic cancer are described herein. Material
and Methods
The University of Kansas Medical Center records of 30 patients who were found to have obstruction of the extrahepatic bile ducts secondary to metastatic cancer were reviewed to determine the site of the primary cancer, the interval between treatment of the primary cancer and admission for jaundice, the presenting symptoms and signs, the procedures used for diagnosis, the sites of biliary obstruction, the types and success of therapy, and the duration of survival. Cumulative survival rates were calculated for all patients in whom successful palliation was obtained.
Results From 1975 to 1980, 30 patients with biliary obstruction secondary to metastatic cancer were evaluated. The ages of the patients ranged from 31 to 83 years (mean 57). Sixteen of the patients were women and 14 were men. Biliary-enteric anastomosis and percutaneous transhepatic intubation of the bile ducts were the primary techniques employed for biliary decompression. Site of the primary tumor: Thirty percent (nine) of the primary tumors were colonic carcinomas. Six of the nine colonic cancers (66 nercent) were located From the Department of Surgery, University of Kansas Medical Center, Kansas City, Kansas. Requests for reprints should be addressed to James H. Thomas, MD, Department of Surgery, University of Kansas Medical Center, Kansas City, Kansas 66103. Presented at the 33rd Annual Meeting of the Southwestern Surgical Congress, Monterey, California, May 4-7, 1981.
770
in the right colon and three were in the left colon. The second most common primary site was the gallbladder (five), followed by extremity melanomas (three), pancreas (three), small bowel (three), breast (two), stomach (two), ovary (two) and non-Hodgkin’s lymphoma (one). Interval between treatment of the primary tumor and development of jaundice: The interval between treatment of the primary tumor and admission for jaundice ranged from less than 1 day to 192 months, with a mean duration of 36 f 8.4 months. Four patients, one with cancer of the gallbladder and three with colon cancer, presented with jaundice at the time of the initial diagnosis of cancer; these patients had no obstruction-free interval. The longest obstruction-free interval was 192 months in a patient with ovarian cancer. The obstruction-free interval in nine patients with colonic cancer ranged from less than 1 day to 95 months (mean 29 f 10.6 months). In five patients with gallbladder cancer the obstruction-free interval ranged from less than 1 day to 36 months (mean 15.0 f 7 months). In the 26 patients with successful palliation there was no correlation between the obstruction-free interval and the level of bilirubin, the site of obstruction or the duration of survival after decompression. Symptoms, signs and laboratory data on admission: Twenty (70 percent) of the patients had jaundice on admission. Pain was present in 10 patients (30 percent). Icterus was present on examination in 27 patients (90 percent). In the three patients without clinical jaundice the serum bilirubin level was 1.8,2.0 and 2.3 mg/dl. The serum bilirubin in the patients with clinical jaundice ranged from 6.3 to 35.4 mg/dl (mean 16.7 f 1.8). The mean serum bilirubin level was 16.7 f 1.9 mg/dl in the 15 patients with obstruction of the common bile duct, 19 f 3 mg/dl in the 6 patients with obstruction of the common hepatic duct, 18.7 f 8 mg/dl in the 3 patients with obstruction at the bifurcation of the hepatic ducts, and 8.3 f 3.2 mgldl in the 3 patients with obstruction of the left or right hepatic duct or both. The level of serum bilirubin on admission did not corre-
The American Journal of Surgery
Extrahepatic Biliary Obstruction
late with the site of obstruction and did not predict the duration of survival. Diagnostic procedures: Percutaneous transhepatic cholangiography was used for diagnosis in 19 patients, sonography was used in 12 patients, computed axial tomography of the abdomen in 7 patients and endoscopic retrograde cholangiography in 1 patient. In seven patients the diagnosis of extrahepatic biliary obstruction was made at laparotomy. Percutaneous transhepatic cholangiography successfully demonstrated the site of obstruction in all patients so studied. Obstruction of the bile ducts was suspected and confirmed preoperatively in one of the three patients with serum bilirubin concentrations of less than 2.5 mg/dl. Site of biliary tract obstruction: The common bile duct was obstructed in 18 patients (60 percent), the common hepatic duct in 6 patients (20 percent), the bifurcation of the common hepatic duct in 3 patients (10 percent), and the right or left hepatic duct, or both, in 3 patients (10 percent). No patient had multiple sites of obstruction or had a new site of obstruction after biliary decompression. Neither the level of serum bilirubin on admission nor the duration of palliation correlated with the site of obstruction. Therapy: Biliary decompression using radiologic (external and internal drainage) or surgical procedures, or both, was successful in 27 (90 percent) of the patients. Percutaneous transhepatic biliary drainage by placement of an internal stent was attempted in 14 patients and was successful in 8 (57 percent). Of the six patients (43 percent) in whom percutaneous transhepatic internal biliary drainage failed, two received no additional therapy, two underwent conversion to percutaneous transhepatic external biliary drainage, and two underwent operation. One of these two patients was managed by tube hepatostomy, and the other was managed by percutaneous transhepatic external biliary drainage after failure of operative decompression. One patient with obstruction of both hepatic ducts underwent internal drainage of the right hepatic duct and left hepaticojejunostomy. Complications of internal biliary drainage occurred in four patients (50 percent) and included catheter infection in one, catheter occlusion in two and catheter dislodgment in one. These patients were managed by antibiotics (one), catheter replacement (one) and conversion to external drainage (two). Three patients died within 30 days of successful insertion of a biliary endoprosthesis. Surgical decompression was attempted in 19 patients and was successful in 16 (84 percent). Operative decompression was successful in one patient in whom internal drainage failed, and external drainage was successful in one patient in whom surgery failed. The most common surgical procedures were Roux-Y choledochojejunostomy (five), T-tube insertion (four), choledochoduodenostomy (four), cholecys-
Volume
142, December
1961
0 30 60 90 120
150
DURATION
180
210
SURVIVAL
240
270
300
330
360
(DAYS)
Figure 1. Cumulative survival in 27 patients with extrahepatic biliary obstruction secondary to metastatic cancer.
tojejunostomy (two), tube hepatostomy (two), Utube insertion (one) and hepaticojejunostomy (one). Two patients underwent a second operation after failure of an initial procedure. Both initial operations (T-tube insertions) were followed by choledochoduodenostomy. One of the two choledochoduodenostomies functioned. In two patients only laparotomy was performed; decompression was not attempted. Two patients died within 30 days of surgery. One had obtained successful palliation, but operation failed to provide decompression in the other patient. Success of decompression and duration of palliation: The total bilirubin level was measured two weeks after biliary decompression and ranged from 1.2 to 14.6 mg/dl (mean concentration 4.9 f 0.77). The decrease from preoperative levels (6.3 to 35.4 mg/dl; mean 16.7 f 1.8) was significant (p < 0.05). After decompression mean bilirubin levels were 4.05 f 0.77 mgldl in patients treated by external and internal drainage, and 6.08 f 1.15 mgldl in patients treated by operat,ion (difference not significant). The duration of survival in patients treated successfully by radiologic or surgical procedures, or both, ranged from 30 to 700 days (mean 197 f 36). In eight patients treated by radiologic methods, the mean duration of survival was 60 f 11 days, compared with 270 f 49 days in patients treated surgically (p < 0.05). Six of the 27 patients (22 percent) who obtained successful palliation are alive, with a mean duration of survival of 175 f 35 days. Cumulative survival rates were calculated for all patients who obtained successful decompression (Figure 1). Comments Kopelson et al [I], in a review of patients with extrahepatic biliary tract obstruction from metastatic breast carcinoma, noted that little attention has been given in surgical reports to the palliation of patients with biliary obstruction secondary to metastatic solid tumors. They describe 28 patients with biliary tract obstruction who were managed by operation, radiologic placement of an endoprosthesis, chemotherapy
771
Thomas et al
and radiotherapy. Warshaw and Welch [2] reported on eight patients wtih a bile duct obstruction secondary to metastatic colonic carcinoma. Six were managed by surgery and two by irradiation. The length of survival after decompression ranged from 1 to 36 months. Likewise, Papachristou and Fortner [3] reported on 30 patients in whom jaundice developed 1 week to 16 weeks after gastric resection for carcinoma. All 30 patients underwent exploration, and palliation was achieved in 15. The length of postoperative survival averaged 6.4 months. In the review of Kopelson et al [I], metastatic gastric carcinoma was the most frequent cause of extrahepatic obstruction (50 percent), followed by colon carcinoma (20 percent). A majority of the patients in our series (nine) had colon cancer. Five patients had cancer of the gallbladder and three had melanoma. Only two patients had gastric and breast carcinoma. Although the incidence of carcinoma of the left colon is greater than that of the right, six of our nine patients with extrahepatic biliary obstruction secondary to metastatic colon carcinoma had primary lesions in the right colon. Of the eight patients with biliary obstruction secondary to colonic carcinbma described by Warshaw and Welch [2], four had cancer of the right colon. These data suggest that right colon cancers may metastasize to the periportal lymph nodes more often than those of the left colon. The most frequent site of obstruction in the present series was the common bile duct. Likewise, Popp et al [4] found that the most frequent level of biliary obstruction was the common bile duct in seven patients with biliary obstruction secondary to metastatic breast cancer, and Papachristou and Fortner [3] reported obstruction in the distal common bile duct in 19 of 30 patients with biliary obstruction secondary to metastases from gastric carcinoma. In contrast, five of the eight patients described by Warshaw and Welch [2] were obstructed at the bifurcation of the common hepatic duct. Although there was no correlation between the site of obstruction and the duration of survival, two of the four failures in our series were in patients with obstruction of the bifurcation of the common hepatic duct of either one or both hepatic ducts. Percutaneous transhepatic cholangiography was the most useful technique for confirming the diagnosis and demonstrating the site of biliary obstruction in our patients. No complications of cholangiography occurred in this series, although bleeding, sepsis, hepatic hematoma and bile leaks are well recognized complications of this procedure. Papachristou and Fortner [3], used laparotomy as their major diagnostic procedure, and in 7 of our 30 patients laparotomy was used to establish the diagnosis. Biliary decompression was successful in 90 percent of the patients in this series. The success rate was 57 772
percent for those treated by radiologic placement of an endoprosthesis, 100 percent for those decompressed by external drainage, and 84 percent for those treated by surgical decompression. The surgical success rate in this series compares favorably with that of 50 percent reported by Papachristou and Fortner [3] in patients with metastatic gastric cancer and with the 83 percent success rate in patients reviewed by Warshaw and Welch [2]. The lower success rate with percutaneous internal biliary drainage in the present series may reflect initial problems with technique and selection of patients with more advanced disease compared with those treated surgically. Recent reports [5] indicate that immediate or early success rates may approach 80 percent for percutaneous decompression in patients with malignant obstruction. A majority of reports [6,7] suggest that nonsurgical techniques afford the best approach to biliary decompression, primarily because of the reported operative mortality rates of 10 to 20 percent [8] in patients who are operated on for decompression of the biliary tract obstructed by carcinoma of the pancreas. It must be pointed out, however, that these patients have far advanced disease and that their operative mortality is probably not representative of patients with hepatic obstruction secondary to other types of malignancy. Thirty day mortality rates after nonsurgical decompression are not reported by workers who advocate percutaneous transhepatic biliary decompression. However, Pereiras et al [9] reported two early deaths (at less than 30 days) as a result of hepatic abscess in 12 patients with percutaneous placement of biliary endoprosthesis. In the present study 5 of 30 patients (16 percent) died within 30 days of treatment. There was no significant difference in mortality for surgical (two deaths) and nonsurgical decompression (three deaths). On the other hand, long-term survival was better in patients with surgical decompression (270 f 49 days) than in those with percutaneous transhepatic decompression (60 f 11 days). Again, as noted previously, the longer duration of survival noted with operation may be attributable to patient selection rather than or in addition to possible superiority of surgical therapy in achieving long-term decompression. This study as well as others reveal that very few institutions have had a large experience with extrahepatic biliary obstruction secondary to metastatic carcinoma. Therefore, conclusions concerning the superiority of one approach versus the other are at best tentative and may not apply to certain groups of patients. For example, Popp et al [4] suggested using transhepatic biliary decompression in patients with high biliary obstruction to avoid the mortality associated with intrahepatic biliary-enteric anastomosis. Cahow [IO], however, noted that high obstructions cause hepatic abscess if only one lobe is decompressed, as is often the case with percutaneous The American Journal of Surgery
Extrahepatic Biliary Obstruction
procedures. He suggested that intrahepatic cholangiojejunostomy was the preferred method of treating these patients. He used that procedure 14 times in 11 patients without serious complications and obtained a mean duration of survival of 16 months. Patients with severe jaundice present a special problem. Since a bilirubin level of greater than 10 mg/dl may be associated with increased operative mortality [II], initial decompression of the biliary tract by percutaneous transhepatic techniques should be considered. Staged definitive operation can then be performed after the improvement of hepatic function. In spite of the problems inherent in the interpretation of data from retrospective studies, this experience suggests that (1) a large number of cancer patients with jaundice have extrahepatic biliary obstruction and may benefit from biliary decompression, (2) transhepatic percutaneous cholangiography, sonography and computed tomographic scans provide highly accurate diagnostic and anatomic information with minimal patient discomfort, (3) percutaneous transhepatic catheterization provides adequate decompression in many patients with biliary obstruction secondary to cancer, and may be of value in preparing patients for subsequent operative decompression, and (4) surgical decompression can be achieved with acceptable operative mortality in selected patients and may afford better palliation than percutaneous drainage. Summary Thirty patients with extrahepatic biliary obstruction secondary to metastatic cancer were reviewed to determine the sites of the primary tumor, diagnostic methods, therapy and success of palliation. Colon carcinoma was the most common primary tumor, and the common bile duct was most often obstructed. Both percutaneous transhepatic and surgical decompression of the biliary tract were employed. Twenty-seven (90 percent) of the patients obtained successful palliation. The length of survival averaged 270 f 49 days in patients treated surgically compared with 60 f 11 days in patients who underwent decompression by radiologic techniques. Mortality was not increased in patients undergoing operative biliary drainage. Surgical decompression may be the best method for managing patients with biliary obstruction secondary to metastatic cancer. References 1. Kopelson G, Chu AM, Doucette JA, Gunderson LL. Extra-hepatic biliary tract metastases from breast cancer. Int J Radiat Oncol Biol Phys 1980;8:497-504. 2. Warshaw AL, Welch JP. Extrahepatic biliary obstruction by metastatic colon carcinoma. Ann Surg 1978;188:593-7. 3. Papachristou D, Fortner JG. Biliary obstruction after gastrectomy for carcinoma of the stomach. Surg Gynecol Obstet 1978;147:401-4. 4. Popp JW. Schapiro RI-f, Warshaw AL. Extrahepatic biliary obstruction caused by metastatic breast carcinoma. Ann Intern
Volume 142, December 1981
Med 1979;91:568-71. 5. Burcharth F, Jensen LI, Olesen K. Endoprosthesis for internal drainage of the biliary tract. Gastroenterology 1979;77: 133-7. 6. Pollock TW, Ring ER, Oleaga JA, Freiman DB, Mullen JL, Rosato EF. Percutaneous decompression of benign and malignant biliary obstruction. Arch Surg 1979;114:148-51. 7. MacCarty RL. Nonsurgical management of obstructive jaundice in the patient with advanced cancer. JAMA 1980;244: 1976-8. 8. Buckwalter JA, Lawton RL, Tidrick RT. Bypass operations for neoplastic biliary tract obstruction. Am J Surg 1965;109: 100-6. 9. Pereiras RV Jr, Rheingold OJ, Hutson D, et al. Relief of malignant obstructive jaundice by percutaneous insertion of a permanent prosthesis in the biliary tree. Ann Intern Med 197889589-93. 10. Cahow GE. Alternative methods for hepatic decompression. Surg Clin North Am 1980;60:1305-14. 11. Denning DA, Ellison EC, Carey LC. Preoperative percutaneous transhepatic biliary decompression lowers operative morbidity in patients with obstructive jaundice. Am J Surg 1981;141:61-4.
Discussion Kent C. Westbrook (Little Rock, AR): I think Dr. Thomas has given us a clear view of a difficult problem that most of us don’t like to have to deal with. First, we have to realize that our initial treatment has failed, and second, whatever we do from then on is probably going to fail because the patients will be dead within a year or 2. But palliative manipulation of patients is a worthwhile endeavor. I have a problem comparing the patients treated surgically with those treated radiologically. I go directly to percutaneous transhepatic cholangiography. Computed tomography and ultrasound have not been of much value. If cholangiography shows a low obstruction in a patient who is a good risk, I am eager to operate. What do you do when a patient who had colon resection 2 years ago has jaundice? Do you look for liver metastasis or extrahepatic obstruction first? Second, assuming you find a patient with extrahepatic obstruction and relieve the obstruction either surgically or with an endoprosthesis, then do you give chemotherapy or radiotherapy or just follow them along? William R. Schiller (Albuquerque, NM): It is difficult to know how to choose between the surgical or radiographic options. I would appreciate your comments on it. James H. Thomas (closing): We evaluate these patients using sonography and liver and computed axial tomographic scans. If the patient does not have an obstruction, percutaneous transhepatic cholangiography is performed. If the patient’s bilirubin level is greater than 10 mg/lOO ml, we use decompression for several days before operation. I agree that if a high obstruction is demonstrated, the inclination is to allow the radiologist to place an endoprosthesis. Treatment after decompression, whether chemotherapy or radiotherapy, is based on whether the primary tumor is responsive to that sort of therapy. We think at present that surgery in selected patients is probably the most reasonable alternative. If a patient has external drainage, catheter maintenance is a problem. Complications are less likely to be a problem with surgical decompression. Again, patient selection is the key.
773