Interpretative Problems in Endoscopic Retrograde Cholangiopancreatography
Melvin A. Block, MD, Detroit, Michigan Bernard M. Schuman, MD, Detroit, Michigan Malcolm L. Weckstein, MD, Detroit, Michigan
The technic of endoscopic retrograde cholangiopancreatography has sufficiently been standardized in experienced hands to permit satisfactory visualization of the bile and pancreatic ducts in most patients. These studies have provided critical preoperative information for surgeons by indicating more precisely the need for operation and the anatomic abnormalities that are present. However, aside from technical factors, which make cholangiography less readily achieved than pancreatography, interpretative problems occasionally have also been evident. Material
From November 1972 to April 1974, endoscopic retrograde cholangiopancreatography was carried out in 111 patients. Standard technics, using the Olympus JFB fiberoptic side-viewing duodenoscope, were employed. In this study, results of the examination of the biliary tract were evaluated, concentrating on forty-one patients with biliary tract problems including thirty-five patients presenting with jaundice as the primary difficulty. Cannulation was not possible in seven patients, including one patient in whom the examination was stopped because of lack of cooperation. Thus, the papilla was cannulated successfully in 83 per cent of the patients. The pancreatic duct only waS visualized in eleven of these patients and both ducts were seen in twelve patients. From the Departments of General Surgery, Medicine, and Radiology, Henrv Ford Hoscital. Detroit. Michioan. Reprint requ&‘should be add&sed to Dr Melvin A. Block, Department of General Surgery, Henry Ford Hospital, Detroit, Michigan 48202. Presented at the Fifteenth Annual Meeting of the Society for Surgery of the Alimentary Tract, San Francisco, California, May 21 and 22, 1974.
Vehane 129, January 1975
Difficulties were encountered in the interpretation of findings in ten patients. In other patients, r&nor problems in interpretation were observed. These problems could be categorized on the basis of roentgenographic interpretation or endoscopic interpretation. Results Difficulties in Roentgenographic Interpretation. Incomplete filling of the bile ducts: Al-
though incomplete filling of the extrahepatic bile ducts usually indicated complete obstruction due to malignancy, this was not always true. Thus, in one patient with incomplete filling of the common bile duct by endoscopic retrograde cholangiopancreatography, operative cholangiography shortly thereafter demonstrated a normal extrahepatic biliary tract. (Figure 1.) The diagnosis of intrahepatic cholestasis was subsequently made in this patient. In another patient, the possibility arose of partial occlusion of one or both hepatic ducts by a known large hamartoma of the liver. Transient jaundice followed by persistent elevation of the serum alkaline phosphatase level and retention of bromsulphthalein required explanation. However, filling of the hepatic ducts in this patient could not be achieved during endoscopic cholangiography. The diagnosis of concomitant hepatitis was eventually established in this patient. If patients found to have mechanical obstruction as the cause of incomplete filling of the bile ducts were eliminated, visualization of the hepatic ducts was not achieved by endoscopic cholangiog-
29
Block, Schuman, and Weckstein
Figure 1. Endoscopic cholangiogram showing (A) incomplete filling of bile duct (arrow). Absence of organic obstruction of the bile duct demonstrated subseijuently by operative cholangiogram (6).
raphy in five of thirty-two patients. In three patients the extrahepatic bile ducts were demonstrated but not the intrahepatic bile ducts. It is sometimes possible to achieve filling of the intrahepatic ducts by gravity if the table is tilted head down whenever these ducts are not filled by injection alone. However, even with the use of such measures, attempts to achieve complete filling of the bile ducts occasionally may result only in greater reflux of contrast medium into the pancreatic duct. The specific reason for incomplete filling of the hepatic duct system in the absence of an organic lesion was not evident. Vennes and Silvis [I ] noted that an insufficient volume or rate of injection of the radiopaque medium may produce a radiographic picture simulating obstruction. Dickinson, Belsito, and Cramer [2] comment on the problem of differentiating technical failure from a duct obstructed by an organic lesion. It is conceivable also that kinking of the duct can result from certain anatomic positions of the body during examination. Of course, failure to cannulate the duct is occasionally due to organic obstruction of the sphincter of Oddi or ampulla of Vater. Visualization of the gallbladder was achieved in eight of fifteen patients with intact gallbladders and with no obvious anatomic reason for failure of filling this structure. Thus, failure of visualization
30
of’ the gallbladder does not necessarily indicate that it is abnormal. However, in some instances, filling of the gallbladder provided vital information that other standard studies failed to indicate, including the demonstration of gallst,ones and the identification of calcification in the right upper abdomen, seen on plain abdominal roentgenograms, to be located outside t,he gallbladder. The presence of gallstones in patients being evaluated for pancreatitis, possibly of an alcoholic etiology, was also a helpful product of endoscopic retrograde cholangiopancreatography. Specific obstructing lesion not defirwd: Endoscopic cholangiography cannot be expect.ed to provide a histologic diagnosis of an obst.ructing lesion. Complete obstruction usually indicates the presence of malignancy with the reservations previously indicat,ed herein. In our series of patients, these malignant lesions were variable and included primary carcinoma of the bile duct or pancreas, metastases to lymph nodes adjacent to the bile duct, and direct spread of retroperitoneal metastases to surround the bile duct. A pancreatic carcinoma produced only partial obstruction of the bile duct sufficient to cause jaundice in one of the patients. In a patient with jaundice associated with pancreatitis, an endoscopic pancreatogram was consistent with this diagnosis, but a choledochogram could not be obtained. ‘Thus, the specific cause of
The American Journal of Surgery
Endoscopic Retrograde Cholangiopancreatography
Figure 2. Endoscopic choiangiogram failing to demonstrate presence of minute stones found subsequently at operation. obstruction
of the bile ducts shown by endoscopic
cholangiography must be provided by other clinical information or by operation. Silvis, Rohrmann, and Vennes [3] as well as Safrany and associates [4] also noted that ductal obstruction shown by endoscopic pancreatograms did not specifically differentiate the various pancreatic diseases. Operability for malignancy involving the head at the pancreas or distal bile ducts is not indicated by endoscopic retrograde cholangiopancreatography. Failure to identify tiny stones in bile duct: Although mild dilation of the extrahepatic bile ducts was demonstrated by endoscopic cholangiography in one patient, tiny stones in the ducts and gallbladder were not evident in the ductograms but were subsequently found at operation. (Figure 2.) In another patient, endoscopic cholangiography suggested the presence of a common duct stone as a cause of jaundice but did not depict the large amount of biliary sludge and tiny stones that were actually demonstrated soon thereafter at operation. The use of a contrast material with less density (including dilution of medium) and volume may increase the diagnostic yield of small common bile duct stones [4,5 1. Overhead films should also be obtained after completion of endoscopic retrograde cholangiopancreatography until all contrast material has passed from the ductal system. However, it must be recognized that the limits of defi-
Volume 129, January 1975
Figure 3. Endoscopically demonstrated vertical proximal pancreatic duct (long arrow) potentially confused as bile duct; only partial filling obtained of obstructed bile duct with bulbous ends (short arrow). Air bubble is present in distal pancreatic duct.
nition of all forms of cholangiography can be exceeded by minute stones. Diagnostic biliary drainage provides help in some of these patients [6]. Confusing anatomic variations: In one patient, the direction and course of the pancreatic duct simulated those expected of the bile duct. (Figure 3.) In addition, the proper interpretation for the short stump of the filled bile duct with bulbous ends, resulting from obstruction by lymph nodes containing metastases, was not appreciated initially. These anatomic and pathologic changes produced confusion initially in interpretation of the films. Whether filling of a single short duct represents either an obstructed pancreatic or an obstructed bile duct may be difficult to determine from films provided by endoscopic cannulation of the duodenal papilla. Variations in the course and configuration of the pancreatic and bile ducts, as seen in radiographs observed by endoscopic cannulation, have been reported [ 7-91. Difficulties in demonstrating recurrent common bile duct strictures: After Roux-en-Y hepaticojejunostomy for stricture of the bile duct, later evaluation of the status of the bile ducts by endoscopic cholangiography is difficult from the standpoint of technical and interpretative factors. The surgeon desires to know the status of bile ducts proximal to the stricture, but this information is not always readily provided by endoscopic retro-
31
Block, Schuman, and Weckstein
the pancreatic duct does not necessarily primary disease of the pancreas. IJnreliability of biopsies tally: Biopsy of ampullary
obtained
indicate
endoscopi-
carcinomas failed to provide histologic confirmation in two patients. This probably resulted from the inability to achieve representative tissue sampling, the specimen obtained not being sufficiently deep or adequate in volume. Also, in our experience, the low grade of some carcinomas of the ampulla and duodenal papilla has made pathologists reluctant to make a positive diagnosis of carcinoma on the basis of small biopsy specimens. Comments
Flgure 4. Photograph taken endoscopically of inflamed duodenal papilla, simulating appearance of malignancy.
grade cholangiopancreatography. Available equipment does not current.ly permit the endoscopic transit of such Roux-en-Y loops, although it is possible to examine the duodenum of patients with Billroth II reconstruction after partial gastrectomy. Percutaneous hepatic cholangiography which characterizes the bile ducts proximally remains a preferable technic to evaluate bile duct strictures. Difficulties in Erxdoscopic Interpretation. Inflammation versus carcinoma of the ampulla of L’ater: Inflammatory changes subsequent to im-
paction at or passage of a stone (through the sphincter of Oddi can simulate carcinoma to the endoscopist. (Figure 4.) However, edema surrounding an ampullary carcinoma can simulate an inflammatory process, as otiserved in one of our pat,ients. Changes at the ampulla produced by carcinoma of the head of the pancreas were interpreted as a probable primary ampullary carcinoma in a third patient. Cotton and associates [IO] point out the normal variations in the shape of the duodenal papilla. In two patients with carcinoma of the ampulla of Vater, pancreatic ductograms were obtained. Thus, the capability of cannulation of the pancreat ic duct does not rule out the presence of carcinoma at the ampulla of Vater. Dilatation of the pancreatic duct was observed in two patients with stones in the common bile duct. This observation indicates that dilatation of
33
As with all diagnostic procedures, results of endoscopic retrograde cholangiopancreatography must be interpreted on the basis of the over-all clinical picture. The value of this procedure is dependent on the capability of the persons performing and interpreting the examination. With increasing experience, results do imgrove. For endoscopic cholangiography, some of the radiologist’s interpretative problems such as those related to air bubbles are similar to those for operative cholangiography [II]. However, difficulties in visualization and assessment of the bile duct at the ampulla, including flow into the duodenum, are fewer with endoscopic technic. Others have observed false-positive and false-negative endoscopic pancreatograms [12,13]. These difficulties are best considered as potential limitations rather than inherent errors of endoscopic retrograde cholangiopancreatography. In certain patients, progress studies at a later date are justified. In certain situations, studies with endoscopic retrograde cholangiopancreatography in our practice are intentionally incomplete. When the diagnosis is established by initial radiographs of a stricture or stones producing obstruction of the bile ducts, attempts to fill the biliary tract completely were not made in order to avoid the complication of cholangitis. If carcinoma of the pancreas is likely from the pancreatograms, the additional time and effort to fill the bile ducts are not expended. Endoscopic cholangiography can provide unexpected dividends in patients with biliary tract problems [5]. In the group of patients included in this study, these dividends included: (1) clarification of questionable radiographic abnormalities in the upper gastrointestinal tract observed prior to endoscopy; (2) the finding of biliary tract lithiasis
The Amerlcsn Journal of Surgery
Endoscopic Retrograde Cholangiopancreatography
in patients under treatment and evaluation for pancreatitis, recent biliary tract studies having been considered to be within normal limits or not obtainable; (3) the finding of dilatation of the pancreatic duct in patients with stones in the common bile duet, possibly indicative of an associated pancreatitis; (4) the finding of an active duodenal ulcer, gastric polyps, or other gastroduodenal lesions, not recognized on recent x-ray studies, but the explanation of the patient’s symptoms in some instances. No significant complications from endoscopic retrograde cholangiopancreatography were recognized in the patients in this study. This examination is indicated, in our opinion, in all patients with jaundice or evidence of extrahepatic biliary tract problems for which the cause is not otherwise obvious.
Summary 1. Of forty-one patients with biliary tract problems studied by endoscopic retrograde choledochopancreatography, interpretation of ductograms or endoscopic observations in ten could not be accomplished with ease for one or several reasons. 2. Although incomplete filling of the bile ducts usually signifies organic obstruction, this occasionally is not true. The specific origin and histologic character of the lesion causing bile duct obstruction are not delineated by retrograde endoscopic cholangiography. 3. Other occasional limitations of endoscopic cholangiography include difficulties in demonstrating minute stones, in recognizing confusing anatomic variations, and in defining the status of recurrent strictures. 4. Inflammatory changes in the duodenal papilla can simulate carcinoma, or vice versa, on the basis of observation by the endoscopist. The capability of achieving a pancreatogram does not rule out the presence of carcinoma at the ampulla producing jaundice. Current technics for biopsy of the papilla or ampulla are in some instances inade-
volume 129, January 1975
quate, making results of such procedures unreliable. 5. Although endoscopic retrograde cholangiopancreatography frequent,ly permits acquisition of vital information for the surgeon and occasionally provides unexpected information of significance, interpretative difficulties in some instances always necessitate the correlation of findings with other available clinical information. References 1. Vennes JA, Silvis SE: Endoscopic visualization of bile and pancreatic ducts. Gastroinfestfndosc 18: 149, 1972. 2. Dickinson PB, Belsito AA, Cramer GG: Diagnostic value of endoscopic cholangiopancreatography .JAMA 225: 944, 1973. 3. Silvis SE, Rohrmann CA, Vennes JA: Diagnostic criteria for the evaluation of the endoscopic pancreatogram. Gastrointest Endosc 20: 51, 1973. 4. Safrany L. Tari J, Barna L, Torok I: Endoscopic retrograde cholangiography: experience of 168 examinations. Gastroinfest Endosc 19: 163, 1973. 5. Blumgart LH, Cotton PB, Burwood R, Lawrie B, Salmon P, Davies GT, Beales JSM. Skirving A: Endoscopy and retrograde choledochopancreatography in the diagnosis of the jaundiced patient. Lancet 2: 1269, 1972. 6. Block MA, Brush BE, Ponka JL, Priest RJ: The diagnosis of postcholecystectomy biliary tract stones. A comparison of biliary drainage and intravenous cholangiography. Arch Surg 73: 694, 1956. 7. Classen M, Hellwig H, Rosch W: Anatomy of the pancreatic duct: a duodenoscopic-radiological study. fndoscopy 5: 14, 1973. 8. Kasugai T. Kuno N, Kobayaski S, Hattori K: Endoscopic pancreatocholangiography. The normal endoscopic pancreatocholangiogram. Gastroenterology 63: 217, 1972. 9. Kasugai T, Kuno N, Kozo M, Kobayashi S. Hattori K: Endoscopic pancreatocholangiography. II. The pathological endoscopic pancreatocholangiogram. Gastroenterology 63: 227, 1972. 10. Cotton PB, Blumgart LH, Davies GT, Pierce JM, Salmon PR, Burwood RJ. Lawrie BW, Read AE: Cannulation of papilla of Vater via fiberduodenoscope: assessment of retrograde cholangiopancreatography in 60 patients. Lancer 1: 53, 1972. 11. Hall RC, Sakiyalak P, Kim SK, Rogers LS, Webb WR: Failure of operative cholangiography to prevent retained common duct stones. AmJSurg 125: 51, 1973. 12. Braasch JW, Gregg JA: Surgical uses of peroral retrograde pancreatography and cholangiography. Am J Surg 125: 432, 1973. 13. Norton RA, Ogoshi K, Hara Y, Niwa M, Paul RE Jr, Tomas J. Kawaz K: Pancreatographic abnormalities due to pancreatic cancer. Gastrointest Endosc 20: 13, 1973.
33
Melvin A. Block, MD, Detroit, Michigan Bernard M. Schuman, MD, Detroit, Michigan Malcolm L. Weckstein, MD, Detroit, Michigan
The technic of endoscopic retrograde cholangiopancreatography has sufficiently been standardized in experienced hands to permit satisfactory visualization of the bile and pancreatic ducts in most patients. These studies have provided critical preoperative information for surgeons by indicating more precisely the need for operation and the anatomic abnormalities that are present. However, aside from technical factors, which make cholangiography less readily achieved than pancreatography, interpretative problems occasionally have also been evident. Material
From November 1972 to April 1974, endoscopic retrograde cholangiopancreatography was carried out in 111 patients. Standard technics, using the Olympus JFB fiberoptic side-viewing duodenoscope, were employed. In this study, results of the examination of the biliary tract were evaluated, concentrating on forty-one patients with biliary tract problems including thirty-five patients presenting with jaundice as the primary difficulty. Cannulation was not possible in seven patients, including one patient in whom the examination was stopped because of lack of cooperation. Thus, the papilla was cannulated successfully in 83 per cent of the patients. The pancreatic duct only waS visualized in eleven of these patients and both ducts were seen in twelve patients. From the Departments of General Surgery, Medicine, and Radiology, Henrv Ford Hoscital. Detroit. Michioan. Reprint requ&‘should be add&sed to Dr Melvin A. Block, Department of General Surgery, Henry Ford Hospital, Detroit, Michigan 48202. Presented at the Fifteenth Annual Meeting of the Society for Surgery of the Alimentary Tract, San Francisco, California, May 21 and 22, 1974.
Vehane 129, January 1975
Difficulties were encountered in the interpretation of findings in ten patients. In other patients, r&nor problems in interpretation were observed. These problems could be categorized on the basis of roentgenographic interpretation or endoscopic interpretation. Results Difficulties in Roentgenographic Interpretation. Incomplete filling of the bile ducts: Al-
though incomplete filling of the extrahepatic bile ducts usually indicated complete obstruction due to malignancy, this was not always true. Thus, in one patient with incomplete filling of the common bile duct by endoscopic retrograde cholangiopancreatography, operative cholangiography shortly thereafter demonstrated a normal extrahepatic biliary tract. (Figure 1.) The diagnosis of intrahepatic cholestasis was subsequently made in this patient. In another patient, the possibility arose of partial occlusion of one or both hepatic ducts by a known large hamartoma of the liver. Transient jaundice followed by persistent elevation of the serum alkaline phosphatase level and retention of bromsulphthalein required explanation. However, filling of the hepatic ducts in this patient could not be achieved during endoscopic cholangiography. The diagnosis of concomitant hepatitis was eventually established in this patient. If patients found to have mechanical obstruction as the cause of incomplete filling of the bile ducts were eliminated, visualization of the hepatic ducts was not achieved by endoscopic cholangiog-
29
Block, Schuman, and Weckstein
Figure 1. Endoscopic cholangiogram showing (A) incomplete filling of bile duct (arrow). Absence of organic obstruction of the bile duct demonstrated subseijuently by operative cholangiogram (6).
raphy in five of thirty-two patients. In three patients the extrahepatic bile ducts were demonstrated but not the intrahepatic bile ducts. It is sometimes possible to achieve filling of the intrahepatic ducts by gravity if the table is tilted head down whenever these ducts are not filled by injection alone. However, even with the use of such measures, attempts to achieve complete filling of the bile ducts occasionally may result only in greater reflux of contrast medium into the pancreatic duct. The specific reason for incomplete filling of the hepatic duct system in the absence of an organic lesion was not evident. Vennes and Silvis [I ] noted that an insufficient volume or rate of injection of the radiopaque medium may produce a radiographic picture simulating obstruction. Dickinson, Belsito, and Cramer [2] comment on the problem of differentiating technical failure from a duct obstructed by an organic lesion. It is conceivable also that kinking of the duct can result from certain anatomic positions of the body during examination. Of course, failure to cannulate the duct is occasionally due to organic obstruction of the sphincter of Oddi or ampulla of Vater. Visualization of the gallbladder was achieved in eight of fifteen patients with intact gallbladders and with no obvious anatomic reason for failure of filling this structure. Thus, failure of visualization
30
of’ the gallbladder does not necessarily indicate that it is abnormal. However, in some instances, filling of the gallbladder provided vital information that other standard studies failed to indicate, including the demonstration of gallst,ones and the identification of calcification in the right upper abdomen, seen on plain abdominal roentgenograms, to be located outside t,he gallbladder. The presence of gallstones in patients being evaluated for pancreatitis, possibly of an alcoholic etiology, was also a helpful product of endoscopic retrograde cholangiopancreatography. Specific obstructing lesion not defirwd: Endoscopic cholangiography cannot be expect.ed to provide a histologic diagnosis of an obst.ructing lesion. Complete obstruction usually indicates the presence of malignancy with the reservations previously indicat,ed herein. In our series of patients, these malignant lesions were variable and included primary carcinoma of the bile duct or pancreas, metastases to lymph nodes adjacent to the bile duct, and direct spread of retroperitoneal metastases to surround the bile duct. A pancreatic carcinoma produced only partial obstruction of the bile duct sufficient to cause jaundice in one of the patients. In a patient with jaundice associated with pancreatitis, an endoscopic pancreatogram was consistent with this diagnosis, but a choledochogram could not be obtained. ‘Thus, the specific cause of
The American Journal of Surgery
Endoscopic Retrograde Cholangiopancreatography
Figure 2. Endoscopic choiangiogram failing to demonstrate presence of minute stones found subsequently at operation. obstruction
of the bile ducts shown by endoscopic
cholangiography must be provided by other clinical information or by operation. Silvis, Rohrmann, and Vennes [3] as well as Safrany and associates [4] also noted that ductal obstruction shown by endoscopic pancreatograms did not specifically differentiate the various pancreatic diseases. Operability for malignancy involving the head at the pancreas or distal bile ducts is not indicated by endoscopic retrograde cholangiopancreatography. Failure to identify tiny stones in bile duct: Although mild dilation of the extrahepatic bile ducts was demonstrated by endoscopic cholangiography in one patient, tiny stones in the ducts and gallbladder were not evident in the ductograms but were subsequently found at operation. (Figure 2.) In another patient, endoscopic cholangiography suggested the presence of a common duct stone as a cause of jaundice but did not depict the large amount of biliary sludge and tiny stones that were actually demonstrated soon thereafter at operation. The use of a contrast material with less density (including dilution of medium) and volume may increase the diagnostic yield of small common bile duct stones [4,5 1. Overhead films should also be obtained after completion of endoscopic retrograde cholangiopancreatography until all contrast material has passed from the ductal system. However, it must be recognized that the limits of defi-
Volume 129, January 1975
Figure 3. Endoscopically demonstrated vertical proximal pancreatic duct (long arrow) potentially confused as bile duct; only partial filling obtained of obstructed bile duct with bulbous ends (short arrow). Air bubble is present in distal pancreatic duct.
nition of all forms of cholangiography can be exceeded by minute stones. Diagnostic biliary drainage provides help in some of these patients [6]. Confusing anatomic variations: In one patient, the direction and course of the pancreatic duct simulated those expected of the bile duct. (Figure 3.) In addition, the proper interpretation for the short stump of the filled bile duct with bulbous ends, resulting from obstruction by lymph nodes containing metastases, was not appreciated initially. These anatomic and pathologic changes produced confusion initially in interpretation of the films. Whether filling of a single short duct represents either an obstructed pancreatic or an obstructed bile duct may be difficult to determine from films provided by endoscopic cannulation of the duodenal papilla. Variations in the course and configuration of the pancreatic and bile ducts, as seen in radiographs observed by endoscopic cannulation, have been reported [ 7-91. Difficulties in demonstrating recurrent common bile duct strictures: After Roux-en-Y hepaticojejunostomy for stricture of the bile duct, later evaluation of the status of the bile ducts by endoscopic cholangiography is difficult from the standpoint of technical and interpretative factors. The surgeon desires to know the status of bile ducts proximal to the stricture, but this information is not always readily provided by endoscopic retro-
31
Block, Schuman, and Weckstein
the pancreatic duct does not necessarily primary disease of the pancreas. IJnreliability of biopsies tally: Biopsy of ampullary
obtained
indicate
endoscopi-
carcinomas failed to provide histologic confirmation in two patients. This probably resulted from the inability to achieve representative tissue sampling, the specimen obtained not being sufficiently deep or adequate in volume. Also, in our experience, the low grade of some carcinomas of the ampulla and duodenal papilla has made pathologists reluctant to make a positive diagnosis of carcinoma on the basis of small biopsy specimens. Comments
Flgure 4. Photograph taken endoscopically of inflamed duodenal papilla, simulating appearance of malignancy.
grade cholangiopancreatography. Available equipment does not current.ly permit the endoscopic transit of such Roux-en-Y loops, although it is possible to examine the duodenum of patients with Billroth II reconstruction after partial gastrectomy. Percutaneous hepatic cholangiography which characterizes the bile ducts proximally remains a preferable technic to evaluate bile duct strictures. Difficulties in Erxdoscopic Interpretation. Inflammation versus carcinoma of the ampulla of L’ater: Inflammatory changes subsequent to im-
paction at or passage of a stone (through the sphincter of Oddi can simulate carcinoma to the endoscopist. (Figure 4.) However, edema surrounding an ampullary carcinoma can simulate an inflammatory process, as otiserved in one of our pat,ients. Changes at the ampulla produced by carcinoma of the head of the pancreas were interpreted as a probable primary ampullary carcinoma in a third patient. Cotton and associates [IO] point out the normal variations in the shape of the duodenal papilla. In two patients with carcinoma of the ampulla of Vater, pancreatic ductograms were obtained. Thus, the capability of cannulation of the pancreat ic duct does not rule out the presence of carcinoma at the ampulla of Vater. Dilatation of the pancreatic duct was observed in two patients with stones in the common bile duct. This observation indicates that dilatation of
33
As with all diagnostic procedures, results of endoscopic retrograde cholangiopancreatography must be interpreted on the basis of the over-all clinical picture. The value of this procedure is dependent on the capability of the persons performing and interpreting the examination. With increasing experience, results do imgrove. For endoscopic cholangiography, some of the radiologist’s interpretative problems such as those related to air bubbles are similar to those for operative cholangiography [II]. However, difficulties in visualization and assessment of the bile duct at the ampulla, including flow into the duodenum, are fewer with endoscopic technic. Others have observed false-positive and false-negative endoscopic pancreatograms [12,13]. These difficulties are best considered as potential limitations rather than inherent errors of endoscopic retrograde cholangiopancreatography. In certain patients, progress studies at a later date are justified. In certain situations, studies with endoscopic retrograde cholangiopancreatography in our practice are intentionally incomplete. When the diagnosis is established by initial radiographs of a stricture or stones producing obstruction of the bile ducts, attempts to fill the biliary tract completely were not made in order to avoid the complication of cholangitis. If carcinoma of the pancreas is likely from the pancreatograms, the additional time and effort to fill the bile ducts are not expended. Endoscopic cholangiography can provide unexpected dividends in patients with biliary tract problems [5]. In the group of patients included in this study, these dividends included: (1) clarification of questionable radiographic abnormalities in the upper gastrointestinal tract observed prior to endoscopy; (2) the finding of biliary tract lithiasis
The Amerlcsn Journal of Surgery
Endoscopic Retrograde Cholangiopancreatography
in patients under treatment and evaluation for pancreatitis, recent biliary tract studies having been considered to be within normal limits or not obtainable; (3) the finding of dilatation of the pancreatic duct in patients with stones in the common bile duet, possibly indicative of an associated pancreatitis; (4) the finding of an active duodenal ulcer, gastric polyps, or other gastroduodenal lesions, not recognized on recent x-ray studies, but the explanation of the patient’s symptoms in some instances. No significant complications from endoscopic retrograde cholangiopancreatography were recognized in the patients in this study. This examination is indicated, in our opinion, in all patients with jaundice or evidence of extrahepatic biliary tract problems for which the cause is not otherwise obvious.
Summary 1. Of forty-one patients with biliary tract problems studied by endoscopic retrograde choledochopancreatography, interpretation of ductograms or endoscopic observations in ten could not be accomplished with ease for one or several reasons. 2. Although incomplete filling of the bile ducts usually signifies organic obstruction, this occasionally is not true. The specific origin and histologic character of the lesion causing bile duct obstruction are not delineated by retrograde endoscopic cholangiography. 3. Other occasional limitations of endoscopic cholangiography include difficulties in demonstrating minute stones, in recognizing confusing anatomic variations, and in defining the status of recurrent strictures. 4. Inflammatory changes in the duodenal papilla can simulate carcinoma, or vice versa, on the basis of observation by the endoscopist. The capability of achieving a pancreatogram does not rule out the presence of carcinoma at the ampulla producing jaundice. Current technics for biopsy of the papilla or ampulla are in some instances inade-
volume 129, January 1975
quate, making results of such procedures unreliable. 5. Although endoscopic retrograde cholangiopancreatography frequent,ly permits acquisition of vital information for the surgeon and occasionally provides unexpected information of significance, interpretative difficulties in some instances always necessitate the correlation of findings with other available clinical information. References 1. Vennes JA, Silvis SE: Endoscopic visualization of bile and pancreatic ducts. Gastroinfestfndosc 18: 149, 1972. 2. Dickinson PB, Belsito AA, Cramer GG: Diagnostic value of endoscopic cholangiopancreatography .JAMA 225: 944, 1973. 3. Silvis SE, Rohrmann CA, Vennes JA: Diagnostic criteria for the evaluation of the endoscopic pancreatogram. Gastrointest Endosc 20: 51, 1973. 4. Safrany L. Tari J, Barna L, Torok I: Endoscopic retrograde cholangiography: experience of 168 examinations. Gastroinfest Endosc 19: 163, 1973. 5. Blumgart LH, Cotton PB, Burwood R, Lawrie B, Salmon P, Davies GT, Beales JSM. Skirving A: Endoscopy and retrograde choledochopancreatography in the diagnosis of the jaundiced patient. Lancet 2: 1269, 1972. 6. Block MA, Brush BE, Ponka JL, Priest RJ: The diagnosis of postcholecystectomy biliary tract stones. A comparison of biliary drainage and intravenous cholangiography. Arch Surg 73: 694, 1956. 7. Classen M, Hellwig H, Rosch W: Anatomy of the pancreatic duct: a duodenoscopic-radiological study. fndoscopy 5: 14, 1973. 8. Kasugai T. Kuno N, Kobayaski S, Hattori K: Endoscopic pancreatocholangiography. The normal endoscopic pancreatocholangiogram. Gastroenterology 63: 217, 1972. 9. Kasugai T, Kuno N, Kozo M, Kobayashi S. Hattori K: Endoscopic pancreatocholangiography. II. The pathological endoscopic pancreatocholangiogram. Gastroenterology 63: 227, 1972. 10. Cotton PB, Blumgart LH, Davies GT, Pierce JM, Salmon PR, Burwood RJ. Lawrie BW, Read AE: Cannulation of papilla of Vater via fiberduodenoscope: assessment of retrograde cholangiopancreatography in 60 patients. Lancer 1: 53, 1972. 11. Hall RC, Sakiyalak P, Kim SK, Rogers LS, Webb WR: Failure of operative cholangiography to prevent retained common duct stones. AmJSurg 125: 51, 1973. 12. Braasch JW, Gregg JA: Surgical uses of peroral retrograde pancreatography and cholangiography. Am J Surg 125: 432, 1973. 13. Norton RA, Ogoshi K, Hara Y, Niwa M, Paul RE Jr, Tomas J. Kawaz K: Pancreatographic abnormalities due to pancreatic cancer. Gastrointest Endosc 20: 13, 1973.
33
