- Email: [email protected]
Long-term follow-up after endoscopic sphincterotomy for bile duct stones in patients younger than 60 years of age
0016-5107/96/4406-064355.00 + O GASTROINTESTINAL ENDOSCOPY Copyright © 1996 by the _4merican Society for Gastrointestinal Endoscopy
Long-term follow-up after endoscopic sphincterotomy for bile duct stones in patients younger than 60 years of age Jacques J.G.H.M. Bergman, MD, Suzanne van der Mey, Eric A.J. Rauws, MD, Jan G.P. Tijssen, PhD Dirk-Jan Gouma, MD, Guido N.J. Tytgat, MD, Kees Huibregtse, MD Amsterdam, The Netherlands
Background: Little is known about the long-term effects of endoscopic biliary sphincterotomy. Methods: We retrospectively evaluated the rate of late complications after endoscopic sphincterotomy (EST) for bile duct stones. Patients had to meet the following inclusion criteria: (1) treated between 1976 and 1980, (2) complete stone removal after EST, (3) prior cholecystectomy or elective cholecystectomy within 2 months after EST, and (4) 60 years old or younger at the time of ERCP. A total of 100 patients were identified. Information was obtained from general practitioners and patients by telephone. Patients completed a postal questionnaire and a blood sample was obtained for liver function tests. Results: Information was obtained for 94 patients (in the majority of cases [87%] from multiple sources). There were 26 men and 68 women with a mean age of 51 years at the time of ERCP (range, 23 to 60 years). Early complications (<30 days) occurred in 14 patients (15%). One patient died of a retroperitoneal perforation secondary to EST. During a median period of 15 years (range, 3 to 18 years), 22 patients (24%) developed a total of 36 late complications. There were 21 patients with symptoms of recurrent bile duct stones and one patient with biliary pancreatitis. Other late complications, such as recurrent ascending cholangitis or malignant degeneration, were not observed. An ERCP was performed in 20 of the 22 patients with late complications and demonstrated bile duct stones in 13, combined with stenosis of the EST opening in 9 patients. Late complications were initially managed endoscopically and/or conservatively. One patient underwent surgery after failed endoscopic treatment and one patient died of cholangitis before she could undergo an ERCP. Twelve other patients died of unrelated causes during follow-up. Conclusions: After EST for bile duct stones, late complications occur in a significant proportion of patients. Stone recurrence remains the most important problem, but can in general be managed endoscopicaUy. (Gastrointest Endosc 1996;44:643-9.) Received November 20, 1995. For revision January 10, 1996. Accepted April 1, 1996. From the Department of Gastroenterology, Department of Clinical Epidemiology and Biostatistics, and Department of Surgery, Academic Medical Center, University of Amsterdam, The Netherlands. Presented in part at the annual meeting of the American Society for Gastrointestinal Endoscopy, May 1995, San Diego, California (Gastrointest Endosc 1995;41:389). Reprint requests: J. Bergman, MD, Gastroenterology, Academic Med Center, Meibergdreef 9, 1105AZ Amsterdam, The Netherlands. 37/1/74020
VOLUME 44, NO. 6, 1996
Since its introduction in 1974, endoscopic sphincterotomy has become a well established therapeutic modality for bile duct stones. 1, 2 Several series have been published describing the efficacy and early morbidity of the procedure.3v Little is known, however, about the long-term effects of endoscopic cleavage of the biliary sphincter. After endoscopic sphincterotomy, the biliary sphincter is rendered permanently insufficient,s The loss of this physiologic barrier between duodenum and biliary tract results in duodeGASTROINTESTINAL ENDOSCOPY 643
Table 1. Selection criteria for retrospective study of long-term complications after endoscopic sphincterotomy for bile duct stones Criteria Inclusion criteria Treated between 1976 and 1981 Stones demonstrated at ERCP Endoscopic sphincterotomy performed Complete clearance of the bile ducts Age <60 years Postcholecystectomy or elective cholecystectomy <2 months after EST Total number of patients included Exclusion of criteria Previous biliodigestive anastomosis (4) Previous gastric surgery (4) Malignancy of bile ducts, pancreas, or liver (2) Intrahepatic stones (2) Chronic pancreatitis (1) Primary sclerosing cholangitis (1) Referred from abroad (3) Total number of patients excluded Total number of patients in retrospective cohort
Totals
PATIENTS AND METHODS Patient selection
117
17 100
nocholedochal reflux and bacterial colonization of the biliary tract. 9, 10 The presence of bacteria in the biliary system, which under physiologic conditions is sterile,ll, 12 might lead to late complications after endoscopic sphincterotomy. These complications may include recurrence of bile duct stones from deconjugation ofbilirubin by bacterial enzymes, 13 inflammatory changes of the biliary and/or hepatic system, 14, 15 recurrent ascending cholangitis, 16 and even malignant degeneration.15, 17 Thus far, follow-up studies after endoscopic sphincterotomy have shown that late complications occur in 5% to 13% of patients during a mean follow-up of 2 to 8 years, e, 18-23 However, because the mean follow-up reported in literature is less than 8 years, late effects of loss of sphincter function, i.e., occurring 15 to 20 y e a r s a f t e r a s p h i n c t e r o t o m y , still r e m a i n u n k n o w n . Endoscopic s p h i n c t e r o t o m y w a s originally introduced for elderly a n d frail p a t i e n t s w h o w e r e unfit for surgery. B e c a u s e of t h e i r r e l a t i v e l y s h o r t life expectancy, t h e s e p a t i e n t s w e r e not a t h i g h r i s k for developing l a t e complications a f t e r endoscopic sphincterotomy. I n r e c e n t y e a r s t h e indication for endoscopic s p h i n c t e r o t o m y h a s b r o a d e n e d to also include y o u n g e r p a t i e n t s w i t h c o m m o n bile duct stones. F u r t h e r m o r e , t h e a d v e n t of t h e laparoscopic cholecystectomy h a s c a u s e d a n a d d i t i o n a l i n c r e a s e in t h e n u m b e r of y o u n g e r p a t i e n t s r e f e r r e d for endoscopic stone r e m o v a l . 24 Bec a u s e of longer life expectancy, t h e s e y o u n g e r p a t i e n t s a r e m o r e a t r i s k for developing l a t e complications aft e r endoscopic s p h i n c t e r o t o m y . C o m b i n e d w i t h t h e i n t r o d u c t i o n of endoscopic stone r e m o v a l w i t h o u t t h e 644 G A S T R O I N T E S T I N A L E N D O S C O P Y
need of sphincterotomy, 25, 26 the issue of long-term sequelae after endoscopic sphincterotomy has become increasingly important. We have, therefore, conducted a retrospective study to investigate the incidence of late complications after endoscopic sphincterotomy for removal of common bile duct stones in patients younger than 60 years of age.
Patients had to meet the following inclusion criteria: (1) ERCP performed between January 1976 and January 1981, (2) stones demonstrated at ERCP, (3) endoscopic sphincterotomy performed, (4) complete clearance of the bile ducts accomplished and documented on cholangiography (either endoscopically or percutaneously via T-drains), (5) 60 years of age or younger at the time of the initial endoscopic procedure, and (6) prior cholecystectomy or elective cholecystectomy within 2 months after endoscopic sphincterotomy. To ensure that all eligible patients were included in this study, patients were selected from three independent data bases. From a computerized data base with data from more than 2000 ERCPs performed between 1976 and 1981, 92 patients were selected. This selection was subsequently matched with an independent data base of the Department of Radielegy of our center, which yielded an additional 21 patients. Finally, this combined cohort was matched with a third independent data base in which all patients treated at our department between 1976 and 1981 were categorized according to diagnosis. This resulted in 4 additional eligible patients. Therefore, a total number of 117 patients met the inclusion criteria mentioned above. We excluded 17 patients because of various reasons (all baseline characteristics, Table 1). Consequently, the total number of patients in this study was 100. Sources of information
Information concerning the initial endoscopic treatment was obtained from endoscopy reports of our unit and from medical charts of the referring hospitals. Because follow-up after the initial endoscopy comprised a period up to 18 years, information concerning late complications was obtained from multiple sources to ensure complete and accurate data collection. First, we contacted the general practitioners of all patients by telephone and asked them to check the patient's medical chart for biliary-related complications. Second, all patients alive at the time of follow-up were contacted by telephone for any medical problems since their first treatment. Subsequently, these patients were asked to complete a postal questionnaire inquiring about recurrence of biliary symptoms and related diagnostic procedures or therapeutic interventions. Nonresponders were recontacted by telephone. Third, all patients were invited to have a blood sample taken and checked for the following liver function tests: bilirubin (conjugated and deconjugated), alanine aminotransferase, aspartate aminotransferase, alkaline phosphatase (AP), and gamma-glutamyl transferase (GGT). Again, nonresponders were contacted by telephone. In case the patient had died, the general practitioner and the paV O L U M E 44, NO. 6, 1996
tient's family were contacted for both the cause of death and the patient's medical history after the endoscopic treatment. If one or more of the above mentioned sources of information suggested that biliary-related complications had occurred, the attending physician was contacted and local medical charts were inspected. In case questionnaires and/or liver function tests indicated possible biliary pathology, patients were invited to undergo a repeat ERCP.
Initial endoscopic treatment ERCP and endoscopic sphincterotomy were performed using Olympus side-viewing endoscopes and pull-sphincterotomes based on the original Demling-Classen design. 1 After the sphincterotomy, spontaneous passage of the stones was awaited (mainly during the period between 1976 and 1978) or attempts were made to actively extract the stones after sphincterotomy (increasingly during 1979 and 1980). Complete stone extraction was confirmed in all patients by either T-tube cholangiography or retrograde cholangiogra-
phy. Patient classification On the basis of their medical history prior to the ERCP, patients were classified as having had primary bile duct stones or secondary bile duct stones at the time of the ERCP. The group of patients with primary bile duct stones consisted of all patients who had undergone cholecystectomy and/or bile duct exploration more than 2 years prior to the ERCP. This group of patients was thought to represent patients in whom the stones had formed de novo in the bile duct in the interval since the original operation. The group of patients with secondary bile duct stones consisted of all patients who presented with a gallbladder in situ or who had their gallbladders removed less than 2 years prior to the sphincterotomy. In these patients the gallstones were considered to have originated in the gallbladder and subsequently have passed into the bile duct. The majority of these patients had their ductal stones detected in the immediate postoperative period by T-tube cholangiography.
Classification of complications Early complications, occurring within 30 days after endoscopic sphincterotomy, were classifted according to the 1991 consensus guidelines. ~ Complications during follow-up were classified according to the clinical presentation and the findings at ERCP. Cholangitis was defined as the combination of fever, chills, colicky pain, cholestatic liver function tests, and/or dilatation of the bile ducts. Patients were clinically classified as having had symptoms of stone recurrence if they had at least three of the following symptoms: jaundice, colicky pain, cholestatic liver function tests, or dilatation of the ducts on radiologic imaging. The majority of ERCPs for late complications were performed at referring centers. In general, the size of the sphincterotomy opening was not documented in the endoscopy reports. Therefore, patients were considered to have had stenosis of the sphincterotomy opening if this was either explicitly mentioned in the endoscopy report or if a repeat sphincterotomy was performed for extraction of recurrent stones. VOLUME 44, NO. 6, 1996
Table 2.
Early complications (<30 days) after endoscopic sphincterotomy and stone extraction; 14 of 94 patients developed early complications Early complications
N = 94
Fatal
Fever/cholangltis Bleeding Pancreatitis Perforation Overall
7 2 3 2 14 (15%)
1 1 (1%)
RESULTS I n f o r m a t i o n was available on 94 p a t i e n t s (94%). In the m a j o r i t y of cases (87%) i n f o r m a t i o n was obtained from both the general practitioner a n d the p a t i e n t (or patient's family). I n f o r m a t i o n was not available on 6 patients because n e i t h e r the p a t i e n t nor the general practitioner could be contacted. O f the patients alive at time of follow-up, 88% completed the questionnaires a n d 89% h a d a blood sample taken. T h e r e were 26 m e n and 68 w o m e n with a m e a n age of 51 y e a r s a t the time of E R C P (range, 23 to 60 years). At the t i m e of the endoscopic sphincterotomy, 9 p a t i e n t s h a d t h e i r gallbladder still in situ. In the rem a i n i n g 85 patients (90%), the m e d i a n i n t e r v a l from p r i m a r y s u r g e r y to s p h i n c t e r o t o m y was 17 m o n t h s (range 8 days to 39 years). The surgical history in this group comprised cholecystectomy with choledochal exploration in 41 p a t i e n t s a n d w i t h o u t such exploration in 37. Seven p a t i e n t s h a d u n d e r g o n e one or more reoperations for bile duct stones. E a r l y complications (within 30 days after endoscopic sphincterotomy) occurred in 14 patients (15%). F e v e r after E R C P was the m o s t f r e q u e n t early complication (Table 2) a n d m a i n l y occurred in p a t i e n t s in w h o m spontaneous passage of the stones was awaited after endoscopic sphincterotomy. One p a t i e n t died 25 days after endoscopic s p h i n c t e r o t o m y of a pleural emp y e m a secondary to a r e t r o p e r i t o n e a l perforation. Therefore, 93 p a t i e n t s were available for long-term follow-up. The 9 p a t i e n t s w i t h a gallbladder in situ at the t i m e of endoscopic s p h i n c t e r o t o m y u n d e r w e n t elective cholecystectomy after a m e d i a n period of 22 days (range 9 to 53).
Long-term follow-up N i n e t y - t h r e e p a t i e n t s were followed-up for a median period of 15 y e a r s ( m e a n 14 years, r a n g e 3 to 18 years). F i g u r e 1 shows the K a p l a n - M e i e r plot for the absence of complications of these 93 patients, e.g., after 10 y e a r s of follow-up 82% of p a t i e n t s were free of complications since the endoscopic s p h i n c t e r o t o m y and stone extraction. T h i r t e e n of the 93 p a t i e n t s (14%) died d u r i n g follow-up. T h e r e was one biliary-related death: a 64year-old w o m a n died of cholangitis 4 years after the GASTROINTESTINAL ENDOSCOPY
645
Table 3. Late complications after endoscopic sphincterotomy for bile duct stones; 36 complications occurred in 22 out of 93 patients
100-
80-
Clinical classification
Patients Stonesor No. of No. of with stenosis episodes patients ERCP on ERCP
¢~ 60-
~'~ 40-
20-
2
4
6
8
10
12
14
16
18
Time (yrs)
Figure 1, Kaplan-Meier plot for the absence of late complications in 93 patients who underwent endoscopic sphincterotomy for removal of bile duct stones. Complications with a fatal outcome are marked with an asterisk. Example: Of the patients alive after 10 years of follow-up, 82% had not developed any late complication since the endoscopic sphincterotomy. initial endoscopic treatment. She presented in extremis with cholangitis and sepsis and died before she could undergo an ERCP. Autopsy was not performed. Overall, 22 patients (24%) developed a total of 36 late complications (range 1 to 6). The median time from endoscopic sphincterotomy to late complications was 3.5 years (range 4 months to 15 years). Although m a n y complications occurred in the first 4 years after endoscopic sphincterotomy, a significant number of complications occurred after 10 years or more (Fig, 1). In 21 of the 22 patients, late complications consisted of episodes with symptoms suggestive of recurrent bile duct stones (Table 3). An ERCP was performed in 20 patients and demonstrated recurrent stones in 13 patients, combined with stenosis of the sphincterotomy opening in 9 patients (Table 3). All patients were initially managed with endoscopic interventions and/or conservative treatment. A repeat sphincterotomy was performed in 9 patients. No serious adverse effects of endoscopic treatment occurred, but stone extraction failed in 1 patient who subsequently underwent choledochoduodenostomy.
Questionnaires and liver function tests Of the 80 patients alive at the time of follow-up, 70 (88%) returned the questionnaire. Four patients reported recent attacks of colicky pain in the right upper abdomen and subsequently underwent repeat ERCP. Two of these patients also had mild liver function abnormalities. Blood samples were obtained in 71 patients (89%) and cholestatic liver function tests were found in 16 (AP and/or GGT >3 times upper limit, 5 patients; AP and/or GGT <3 times upper limit, 11 646
GASTROINTESTINAL ENDOSCOPY
Cholangitis* 16 9 8 8 Symptoms of stone 16 9 9 5 recurreneet Episodes of colicky 3 3 2 0 pain Pancreatitis 1 1 1 0 Total 36 22 20 13 *Cholangitis:combinationof fever,chills,colickypain, cholestatic liver functiontests, and/or dilatationof the bile ducts. ?Symptoms of stone recurrence: at least three of the following: jaundice, colickypain; cholestaticliver functiontests, dilatationof the ducts on radiologicimaging. patients). In 5 patients these abnormalities were considered secondary to medication or unrelated comorbidity. Five other patients refused further investigation. The remaining 6 patients underwent repeat ERCP.
Repeat ERCP A total of 8 patients underwent repeat ERCP at the time of follow-up. The indications for repeat endoscopy were colicky pain attacks (n = 2), cholestatic liver function tests (n = 4), or both (n = 2). At ERCP the sphincterotomy opening was found patent in all patients and permitted easy passage of a 10 mm extraction balloon. In one patient brown pigment stones and sludge were removed. The other patients showed no abnormalities. To avoid bias, patients who underwent repeat ERCP for either abdominal pain or mild cholestatic liver function tests only were not classified as having had late complications. Two patients who underwent repeat ERCP for recent colicky pain attacks, dark urine, and transient cholestatic liver function tests were considered to have had recurrence of symptoms.
Relationship between patient classification and long-term follow-up Of the 93 patients available for long-term follow-up, 38 patients were considered to have had primary bile duct stones at the time of sphincterotomy and 53 patients fulfilled the criteria for secondary bile duct stones. Two patients could not be classified because information was missing on the exact date of the prior cholecystectomy. Table 4 shows patient characteristics and the cumulative incidence of late complications in these two groups. Of the 38 patients with primary bile duct stones, 11 patients (29%) developed late complications whereas late biliary complications occurred in 11 of the 53 patients (21%) with secondary VOLUME 44, NO. 6, 1996
bile duct stones (chi squared = 0.43, p = 0.51, chisquared test with Yates' correction).
DISCUSSION This series reports the rate of late complications after endoscopic sphincterotomy for removal of bile duct stones in 93 patients aged 60 years or younger at the time of initial endoscopic treatment. During a median follow-up of 15 years, 24% of the patients developed late complications. This complication rate is higher t h a n that known from other follow-up studies after endoscopic sphincterotomy (Table 5). 6, t8-23The higher complication rate in the present series could be due to a reduced size of the sphincterotomy, inasmuch as all patients were treated between 1976 and 1980, at which time the optimal size of the sphincterotomy incision may have been less established. However, in the patients who underwent repeat endoscopy 15 years after the initial ERCP, the sphincterotomy opening was found widely patent, which suggests t h a t adequate sphincterotomies were performed. In our opinion the higher complication rate in our series is more a reflection of the small number of patients lost to follow-up, the data collection from multiple sources, and the considerably longer period of observation. Although the cumulative incidence of complications was high (24%) and 44% of complications consisted of episodes of cholangitis, the related mortality rate was low (1%) and only one patient required surgical treatment. The remaining patients were effectively managed endoscopically or conservatively. Furthermore, more serious complications such as recurrent ascending cholangitis, secondary biliary cirrhosis, or malignant degeneration were not observed. However, the power ofthis study to detect an increased relative risk for such complications may be too small, and Kurumado et al. 15 have recently suggested that malignant degeneration needs at least I0 years to develop. The vast majority of complications consisted of episodes with symptoms suggestive of recurrent bile duct stones. How plausible is the theory that loss of sphincter function after endoscopic sphincterotomy leads to formation of recurrent bile duct stones? Several authors have demonstrated that the biliary tree becomes infected with bacteria after endoscopic sphincterotomyg, I0 and there is substantial evidence that bacteria play an essential role in the formation of brown pigment stones. Some bacterial species, especially E s c h e r i c h i a coli, 27 produce enzymes like ~-glucuronidase t h a t are known to precipitate bilirubin and calcium, 13 the main components of brown pigment stones. 2s Furthermore, electron microscopy studies have demonstrated t h a t bacteria are present in the core of brown pigment stones whereas they are absent in the cholesterol and the black pigment stones. 16, 29, 30 Unfortunately, the appearance of recurrent stones at VOLUME 44, NO. 6, 1996
Table 4. Patient characteristics and rate of late complications in patients who underwent endoscopic sphincterotomy for treatment of primary and secondary bile duct stones* Primary bile Secondarybile duct stones, duct stones, (N = 38) (N = 53) Men/women 13/25 12/41 Median age in y (range) 56 (31-60) 52 (23-60) Cholecystectomy 38 44 T-tube in situ -31 Elective cholecystectomy -9 after ERCP Number of patients with 11 (29%) 11 (21%) complications Number of patients with 7 6 stones/stenosis on ERCP Number of episodes of 19 17 complications Time from EST until 35 (12-133) 57 (4-175) complication(mo) *Fordefinitionsof primaryand secondarybileductstonessee text. repeat ERCP was not consistently documented in our study. Others, however, have demonstrated that recurrent stones in general have the appearance of brown pigment stones. 3°-32 Apart from bacterial infection, stasis of bile is thought to be an important factor in the pathogenesis of recurrent bile duct stones, la, a3 Although endoscopic sphincterotomy leads to bacterial contamination of bile, it also effectively drains the bile duct, which enables early passage of stones formed de novo. Recurrent stones should, therefore, develop only in the event of stenosis of the sphincterotomy opening. There are, however, several reports that demonstrate that stone recurrence can actually occur in a well-drained bile duct.16, 19, 34 In our series the size of the sphincterotomy opening was not documented in patients who underwent ERCP for late complications. The fact that four patients with recurrent stones had their stones extracted without the need of repeat sphincterotomy indicates that stenosis of the sphincterotomy opening is not a prerequisite for stone formation. Are recurrent stones caused by secondary effects after endoscopic sphincterotomy or may the underlying stone disease be held responsible? A drawback of this study is the lack of a control group t h a t would enable us to ascertain the relative influence of the underlying stone disease on the rate of late complications. The ideal control group would consist of patients who underwent treatment for bile duct stones without sphincter ablation, i.e., patients who underwent surgical bile duct re-exploration. 35 Several cohort studies have been published in the literature describing late complications after surgical treatment for bile duct GASTROINTESTINAL ENDOSCOPY 647
Table 5. Studies describing late complications after endoscopic sphincterotomy for bile duct stones Reference
Follow-up
N (%) Riemann is 340 (?) RSsch19 248 (79) Kullmann 2° 118 (100) Hawes21 115 (71) Testoni22 96 (100) Escourrou 6 96 (68) Jacobsen 23 52 (100) Present study 94 (94) *Mean, median not given.
Follow-up in months
Biliary problems
Median (range) 54 (24-108) ? (0-84) 57 (24-101) 96* (78-132) 48* (24-96) 24 (6-60) 73 (8-103) 180 (36-216)
N (%) 60 (18) 32 (13) 7 (6) 15 (13) 14 (15) 5 (5) 4 (8) 22 (24)
stones. However, c o m p a r i n g the outcome of these surgical studies w i t h the results of the p r e s e n t series is difficult. A p a r t from problems deriving from differences in d a t a collection a n d d u r a t i o n of follow-up, the m a j o r c a v e a t is the difference in p a t i e n t selection of t h e s e studies. 35 Most surgical follow-up studies concern p a t i e n t s t r e a t e d for p r i m a r y bile duct stones. In our series, 29% of p a t i e n t s w i t h p r i m a r y duct stones developed late complications d u r i n g a m e d i a n follow-up of 15 years. W h e n corrected for differences in l e n g t h of follow-up, these results seem to t a k e a n int e r m e d i a t e position b e t w e e n the results of duct re-exploration w i t h o u t s p h i n c t e r o t o m y (18% to 33% after 5 to 10 y e a r s of follow-up) 31, 34, 36, 37 a n d surgical drainage p r o c e d u r e s (surgical s p h i n c t e r o t o m y or choledochoduodenostomy, 0% to 7% after 3 to 12 y e a r s of follow-up). 36-4° This suggests t h a t for p a t i e n t s with prim a r y bile duct stones, the loss of s p h i n c t e r function after s p h i n c t e r o t o m y is not a major pathogenetic factor in the f o r m a t i o n of r e c u r r e n t stones. This is also ill u s t r a t e d b y the fact t h a t in p a t i e n t s w i t h p r i m a r y bile duct stones, r e c u r r e n t stones t h a t develop after sphinct e r o t o m y are of t h e same type as the original stones: b r o w n p i g m e n t stones. 3°32 In our series, 21% of p a t i e n t s with secondary stones developed late complications after endoscopic sphincterotomy. Surgical series on long-term follow-up after t r e a t m e n t for secondary bile duct stones are e i t h e r incomplete 41 or outdated. 42 Malet et al.43 showed t h a t ductal stones found at the same t i m e as, or w i t h i n several m o n t h s after, cholecystectomy were e i t h e r cholesterol stones or black p i g m e n t stones t h a t probably m i g r a t e d from t h e gallbladder (comparable w i t h the s e c o n d a r y bile duct stones in our study). However, r e c u r r e n t bile duct stones t h a t developed in these pat i e n t s were shown to be brown p i g m e n t stones. 44 T h e fact t h a t r e c u r r e n t stones t h a t develop after sphinct e r o t o m y in p a t i e n t s w i t h secondary bile duct stones are totally different in visual a p p e a r a n c e a n d chemical composition from the original stones 32 points 648 G A S T R O I N T E S T I N A L E N D O S C O P Y
Patients with ERCP 121 70 7 45 14 5 4 20
Stones/stenosis on ERCP N (%) 26 (8) 10 (4) 7 (6) 5 (4) 12 (13) 5 (5) 2 (4) 13 (14)
towards the s p h i n c t e r o t o m y as a possible causative factor in the f o r m a t i o n of r e c u r r e n t stones. The results of this s t u d y show t h a t a significant proportion of p a t i e n t s t r e a t e d with endoscopic sphinct e r o t o m y for bile duct stones at the age of 60 years or y o u n g e r develop r e c u r r e n t problems d u r i n g long-term follow-up. The efficacy of endoscopic s p h i n c t e r o t o m y in removing stones, the low r a t e of serious early complications, and the fact t h a t late complications in general can be m a n a g e d with endoscopic a n d conservative m e a s u r e s , m a k e the endoscopic s p h i n c t e r o t o m y still the t r e a t m e n t of choice for p a t i e n t s with p r i m a r y bile duct stones. However, for y o u n g e r p a t i e n t s with seco n d a r y bile duct stones, s p h i n c t e r o t o m y m a y be causally r e l a t e d to a h i g h e r r a t e of stone recurrence. In these p a t i e n t s endoscopic stone r e m o v a l w i t h o u t ablation of the biliary sphincter (i.e., after endoscopic balloon dilation 25, 26) m a y lead to a lower r a t e of stone r e c u r r e n c e a n d o t h e r late complications.
REFERENCES 1. Classen M, Demling L. Endoskopische sphincterotomie der papilla vateri und steinextraktion aus ductus choledochus.Dtsch Med Wochenschr 1974;99:496-7. 2. Kawai K, Akasaka Y, Murakami K, Tada M, Nakajima M. Endoscopic sphincterotomy of the ampulla of Vater. Gastrointest Endosc 1974;20:148-51. 3. Cotton PB, Lehman G, Vennes G. Endoscopic sphincterotomy, complications and their management: an attempt at consensus. Gastrointest Endosc 1991;37:383-93. 4. Vaira D, Ainley C, Williams S, et al. Endoscopicsphincterotomy in 1000 consecutive patients. Lancet 1989;2:431-4. 5. Leese T, Neoptolemos JP, Carr-Locke DL. Successes, failures, early complications and their management following endoscopic sphincterotomy: results in 394 consecutive patients from a single centre. Br J Surg 1985;72:215-9. 6. Escourrou J, Cordova JA, Lazorthes F, Frexinos J, Ribet A. Early and late complications after endoscopic sphincterotomy for biliary lithiasis with and without the gall bladder "in situ." Gut 1984;25:598-602. 7. Freeman M, Nelson D, Sherman S, et al. Complications of endoscopic sphincterotomy (ES): a prospective, multicenter, 30day outcome study [abstract]. Gastroenterolog3~1994;106:A338. 8. Geenen JE, Toouli J, Hogan WJ, et al. Endoscopic sphincterotomy: follow-up evaluation of effects on the sphincter of Oddi. Gastroenterolog3r1984;87:754-8. V O L U M E 44, NO. 6, 1996
9. Gregg JA, De Girolami P, Can" Locke DL. Effects ofsphincteroplasty and endoscopic sphincterotomy on the bacteriologic characteristics of the common bile duct. Am J Surg 1985;149: 668-71. 10. Sand J, Airo I, Hittunen KM, Mattila J, Nordback I. Changes in biliarybacteria after endoscopic cholangiographyand sphincterotomy. Am Surg 1992;58:324-8. 11. Flemma RJ, Flint LM, Osterhout S, Shingleton WW. Bacteriologic studies of biliary tract infection. Ann Surg 1967;166:56372. 12. Scott AJ, Khan GA. Origin of bacteria in bile duct bile. Lancet 1967;2:790-4. 13. Maki T. Pathogenesis of calcium bilirubinate gallstone: role of E. coli, beta*glucuronidase and coagulation by inorganic ions, polyelectrolytes and agitation. Ann Surg 1966;164:90-100. 14. Greenfield C, Cleland P, Dick R, Masters S, Summerfield JA, Sherlock S. Biliary sequelae of endoscopic sphincterotomy. Postgrad Med J 1985;61:213-5. 15. Kurumado K, Nagai T, Kondo Y, Abe H. Long-term observations on morphological changes of choledochal epithelium after choledochoenterostomy in rats. Dig Dis Sci 1994;39:809-20. 16. Goldman LD, Steer ML, Silen W. Recurrent cholangitis after biliary surgery. Am J Surg 1983;145:450-4. 17. Kinami Y, Ashida Y, Seto K, Takashima S, Kita I. Influence of incomplete bile duct obstruction on the occurrence of cholangiocarcinoma induced by di-isopropanolnitrosamine in hamsters. Oncology 1990;47:170-6. 18. Riemann JF, Lux G, Forster P, AltendorfA. Long-term results after endoscopic papillotomy. Endoscopy 1983;15:165-8. 19. Rosch W, Riemann JF, Lux G, Lindner HG. Long-term follow-up after endoscopic sphincterotomy. Endoscopy 1981;13: 152-3. 20. Kullman E, Borch K, Liedberg G. Long term follow up after en~ doscopic management of retained and recurrent common duct stones. Acta Chir Scand 1989;155:395-9. 21. Hawes RH, Cotton PB, Vallon AG. Follow-up 6 to 11 years after duodenoscopic sphincterotomy for stones in patients with prior cholecystectomy. Gastroenterology 1990;98:1008-12. 22. Testoni PA, TittobelloA. Long-term efficacyofendoscopic papillo-sphincterotomy for common bile duct stones and benign papillary stenosis. Surg Endosc 1991;5:135-9. 23. Jacobsen O, Matzen P. Long-term follow-up study of patients after endoscopic sphincterotomy for choledocholithiasis. Scand J Gastroenterol 1987;22:903-6. 24. Cotton PB. Endoscopic retrograde cholangiopancreatography and laparoscopic cholecystectomy. Am J Surg 1993;165:474-8. 25. Mac Mathuna P, White P, Clarke E, Lennon J, Crowe J. Endoscopic sphincteroplasty: a novel and safe alternative to papillotomy in the management of bile duct stones. Gut 1994;35:127-9. 26. Bergman JJGHM, Rauws EAJ, Tytgat GNJ, Huibregtse K. A prospective randomized trial comparing endoscopic sphincter-
VOLUME 44, NO. 6, 1996
27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40.
41. 42. 43. 44.
otomy (EST) with endoscopic balloon dilatation (EBD) for removal of common bile duct stones (CBDS); initial report [abstract]. Gastrointest Endosc 1994;40:P99. Tabata M, Nakayama F. Bacteria and gallstones: etiological significance. Dig Dis Sci 1981;26:218-24. Stewart L, Smith AL, Pellegrini CA, Motson RW, Way LW. Pigment gallstones form as a composite of bacterial microcolohies and pigment solids. Ann Surg 1987;206:242-50. Kaufman HS, Magnuson TH, Lillemoe KD, Frasca P, Pitt HA. The role of bacteria in gallbladder and common duct stone formation. Ann Surg 1989;209:584-91. Smith AL, Stewart L, Fine R, Pellegrini CA, Way LW. Gallstone disease: the clinical manifestations of infectious stones. Arch Surg 1989;124:629-33. Braasch JW, Fender HR, Bonneval MM. Refractory primary common bile duct stone disease. Am J Surg 1980;139:526-30. Cetta F. The possible role of sphincteroplasty and surgical sphincterotomy in the pathogenesis of recurrent common duct brown stones. HPB Surg 1991;4:261-70. Madden JL. Common duct stones: their origin and surgical management. Surg Clin North Am 1973;53:1095-113. Saharia PC, Zuidema GD, Cameron JL. Primary common duct stones. Ann Surg 1977;185:598-604. Cotton PB. Endoscopic management of bile duct stones (apples and oranges). Gut 1984;25:587-97. Allen B, Shapiro H, Way LW. Management of recurrent and residual common duct stones. Am J Surg 1981;142:41-7. Lygidakis NJ. Surgical approaches to recurrent choledocholithiasis: choledochoduodenostomy versus T-tube drainage after choledochotomy. Am Surg 1983;145:636-9. Escudero-Fabre A, Escallon A Jr, Sack J, Halpern NB, Aldrete JS. Choledochoduodenostomy: analysis of 71 cases followed for 5 to 15 years. Ann Surg 1991;213:635-42. Cuschieri A, Wood RAB, Metcalf MJ, Cumming JGR. Longterm experience with transection choledochoduodenostomy. World J Surg 1983;7:502-4. Thomas CG Jr, Nicholson CP, Owen J. Effectiveness of choledochoduodenostomy and transduodenal sphincterotomy in the treatment of benign obstruction of the common duct. Ann Surg 1971;173:845-56. Cetta F. Do surgical and endoscopic sphincterotomy prevent or facilitate recurrent common duct stone formation? Arch Surg 1993;128:329-36. Larson RE, Hodgson JR, Priestley JT. The early and long term results of 500 consecutive explorations of the common duct. Surg GynecoI Obstet 1966;122:744-50. Malet PF, Dabezies MA, Huang GH, Long WB, Gadacz TR, Soloway RD. Quantitative infrared spectroscopy of common bile duct gallstones. Gastroenterology 1988;94:1217-21. Whiting MJ, Watts JM. Chemical composition of common bile duct stones. Br J Surg 1986;73:229-32.
G A S T R O I N T E S T I N A L E N D O S C O P Y 649
Long-term follow-up after endoscopic sphincterotomy for bile duct stones in patients younger than 60 years of age Jacques J.G.H.M. Bergman, MD, Suzanne van der Mey, Eric A.J. Rauws, MD, Jan G.P. Tijssen, PhD Dirk-Jan Gouma, MD, Guido N.J. Tytgat, MD, Kees Huibregtse, MD Amsterdam, The Netherlands
Background: Little is known about the long-term effects of endoscopic biliary sphincterotomy. Methods: We retrospectively evaluated the rate of late complications after endoscopic sphincterotomy (EST) for bile duct stones. Patients had to meet the following inclusion criteria: (1) treated between 1976 and 1980, (2) complete stone removal after EST, (3) prior cholecystectomy or elective cholecystectomy within 2 months after EST, and (4) 60 years old or younger at the time of ERCP. A total of 100 patients were identified. Information was obtained from general practitioners and patients by telephone. Patients completed a postal questionnaire and a blood sample was obtained for liver function tests. Results: Information was obtained for 94 patients (in the majority of cases [87%] from multiple sources). There were 26 men and 68 women with a mean age of 51 years at the time of ERCP (range, 23 to 60 years). Early complications (<30 days) occurred in 14 patients (15%). One patient died of a retroperitoneal perforation secondary to EST. During a median period of 15 years (range, 3 to 18 years), 22 patients (24%) developed a total of 36 late complications. There were 21 patients with symptoms of recurrent bile duct stones and one patient with biliary pancreatitis. Other late complications, such as recurrent ascending cholangitis or malignant degeneration, were not observed. An ERCP was performed in 20 of the 22 patients with late complications and demonstrated bile duct stones in 13, combined with stenosis of the EST opening in 9 patients. Late complications were initially managed endoscopically and/or conservatively. One patient underwent surgery after failed endoscopic treatment and one patient died of cholangitis before she could undergo an ERCP. Twelve other patients died of unrelated causes during follow-up. Conclusions: After EST for bile duct stones, late complications occur in a significant proportion of patients. Stone recurrence remains the most important problem, but can in general be managed endoscopicaUy. (Gastrointest Endosc 1996;44:643-9.) Received November 20, 1995. For revision January 10, 1996. Accepted April 1, 1996. From the Department of Gastroenterology, Department of Clinical Epidemiology and Biostatistics, and Department of Surgery, Academic Medical Center, University of Amsterdam, The Netherlands. Presented in part at the annual meeting of the American Society for Gastrointestinal Endoscopy, May 1995, San Diego, California (Gastrointest Endosc 1995;41:389). Reprint requests: J. Bergman, MD, Gastroenterology, Academic Med Center, Meibergdreef 9, 1105AZ Amsterdam, The Netherlands. 37/1/74020
VOLUME 44, NO. 6, 1996
Since its introduction in 1974, endoscopic sphincterotomy has become a well established therapeutic modality for bile duct stones. 1, 2 Several series have been published describing the efficacy and early morbidity of the procedure.3v Little is known, however, about the long-term effects of endoscopic cleavage of the biliary sphincter. After endoscopic sphincterotomy, the biliary sphincter is rendered permanently insufficient,s The loss of this physiologic barrier between duodenum and biliary tract results in duodeGASTROINTESTINAL ENDOSCOPY 643
Table 1. Selection criteria for retrospective study of long-term complications after endoscopic sphincterotomy for bile duct stones Criteria Inclusion criteria Treated between 1976 and 1981 Stones demonstrated at ERCP Endoscopic sphincterotomy performed Complete clearance of the bile ducts Age <60 years Postcholecystectomy or elective cholecystectomy <2 months after EST Total number of patients included Exclusion of criteria Previous biliodigestive anastomosis (4) Previous gastric surgery (4) Malignancy of bile ducts, pancreas, or liver (2) Intrahepatic stones (2) Chronic pancreatitis (1) Primary sclerosing cholangitis (1) Referred from abroad (3) Total number of patients excluded Total number of patients in retrospective cohort
Totals
PATIENTS AND METHODS Patient selection
117
17 100
nocholedochal reflux and bacterial colonization of the biliary tract. 9, 10 The presence of bacteria in the biliary system, which under physiologic conditions is sterile,ll, 12 might lead to late complications after endoscopic sphincterotomy. These complications may include recurrence of bile duct stones from deconjugation ofbilirubin by bacterial enzymes, 13 inflammatory changes of the biliary and/or hepatic system, 14, 15 recurrent ascending cholangitis, 16 and even malignant degeneration.15, 17 Thus far, follow-up studies after endoscopic sphincterotomy have shown that late complications occur in 5% to 13% of patients during a mean follow-up of 2 to 8 years, e, 18-23 However, because the mean follow-up reported in literature is less than 8 years, late effects of loss of sphincter function, i.e., occurring 15 to 20 y e a r s a f t e r a s p h i n c t e r o t o m y , still r e m a i n u n k n o w n . Endoscopic s p h i n c t e r o t o m y w a s originally introduced for elderly a n d frail p a t i e n t s w h o w e r e unfit for surgery. B e c a u s e of t h e i r r e l a t i v e l y s h o r t life expectancy, t h e s e p a t i e n t s w e r e not a t h i g h r i s k for developing l a t e complications a f t e r endoscopic sphincterotomy. I n r e c e n t y e a r s t h e indication for endoscopic s p h i n c t e r o t o m y h a s b r o a d e n e d to also include y o u n g e r p a t i e n t s w i t h c o m m o n bile duct stones. F u r t h e r m o r e , t h e a d v e n t of t h e laparoscopic cholecystectomy h a s c a u s e d a n a d d i t i o n a l i n c r e a s e in t h e n u m b e r of y o u n g e r p a t i e n t s r e f e r r e d for endoscopic stone r e m o v a l . 24 Bec a u s e of longer life expectancy, t h e s e y o u n g e r p a t i e n t s a r e m o r e a t r i s k for developing l a t e complications aft e r endoscopic s p h i n c t e r o t o m y . C o m b i n e d w i t h t h e i n t r o d u c t i o n of endoscopic stone r e m o v a l w i t h o u t t h e 644 G A S T R O I N T E S T I N A L E N D O S C O P Y
need of sphincterotomy, 25, 26 the issue of long-term sequelae after endoscopic sphincterotomy has become increasingly important. We have, therefore, conducted a retrospective study to investigate the incidence of late complications after endoscopic sphincterotomy for removal of common bile duct stones in patients younger than 60 years of age.
Patients had to meet the following inclusion criteria: (1) ERCP performed between January 1976 and January 1981, (2) stones demonstrated at ERCP, (3) endoscopic sphincterotomy performed, (4) complete clearance of the bile ducts accomplished and documented on cholangiography (either endoscopically or percutaneously via T-drains), (5) 60 years of age or younger at the time of the initial endoscopic procedure, and (6) prior cholecystectomy or elective cholecystectomy within 2 months after endoscopic sphincterotomy. To ensure that all eligible patients were included in this study, patients were selected from three independent data bases. From a computerized data base with data from more than 2000 ERCPs performed between 1976 and 1981, 92 patients were selected. This selection was subsequently matched with an independent data base of the Department of Radielegy of our center, which yielded an additional 21 patients. Finally, this combined cohort was matched with a third independent data base in which all patients treated at our department between 1976 and 1981 were categorized according to diagnosis. This resulted in 4 additional eligible patients. Therefore, a total number of 117 patients met the inclusion criteria mentioned above. We excluded 17 patients because of various reasons (all baseline characteristics, Table 1). Consequently, the total number of patients in this study was 100. Sources of information
Information concerning the initial endoscopic treatment was obtained from endoscopy reports of our unit and from medical charts of the referring hospitals. Because follow-up after the initial endoscopy comprised a period up to 18 years, information concerning late complications was obtained from multiple sources to ensure complete and accurate data collection. First, we contacted the general practitioners of all patients by telephone and asked them to check the patient's medical chart for biliary-related complications. Second, all patients alive at the time of follow-up were contacted by telephone for any medical problems since their first treatment. Subsequently, these patients were asked to complete a postal questionnaire inquiring about recurrence of biliary symptoms and related diagnostic procedures or therapeutic interventions. Nonresponders were recontacted by telephone. Third, all patients were invited to have a blood sample taken and checked for the following liver function tests: bilirubin (conjugated and deconjugated), alanine aminotransferase, aspartate aminotransferase, alkaline phosphatase (AP), and gamma-glutamyl transferase (GGT). Again, nonresponders were contacted by telephone. In case the patient had died, the general practitioner and the paV O L U M E 44, NO. 6, 1996
tient's family were contacted for both the cause of death and the patient's medical history after the endoscopic treatment. If one or more of the above mentioned sources of information suggested that biliary-related complications had occurred, the attending physician was contacted and local medical charts were inspected. In case questionnaires and/or liver function tests indicated possible biliary pathology, patients were invited to undergo a repeat ERCP.
Initial endoscopic treatment ERCP and endoscopic sphincterotomy were performed using Olympus side-viewing endoscopes and pull-sphincterotomes based on the original Demling-Classen design. 1 After the sphincterotomy, spontaneous passage of the stones was awaited (mainly during the period between 1976 and 1978) or attempts were made to actively extract the stones after sphincterotomy (increasingly during 1979 and 1980). Complete stone extraction was confirmed in all patients by either T-tube cholangiography or retrograde cholangiogra-
phy. Patient classification On the basis of their medical history prior to the ERCP, patients were classified as having had primary bile duct stones or secondary bile duct stones at the time of the ERCP. The group of patients with primary bile duct stones consisted of all patients who had undergone cholecystectomy and/or bile duct exploration more than 2 years prior to the ERCP. This group of patients was thought to represent patients in whom the stones had formed de novo in the bile duct in the interval since the original operation. The group of patients with secondary bile duct stones consisted of all patients who presented with a gallbladder in situ or who had their gallbladders removed less than 2 years prior to the sphincterotomy. In these patients the gallstones were considered to have originated in the gallbladder and subsequently have passed into the bile duct. The majority of these patients had their ductal stones detected in the immediate postoperative period by T-tube cholangiography.
Classification of complications Early complications, occurring within 30 days after endoscopic sphincterotomy, were classifted according to the 1991 consensus guidelines. ~ Complications during follow-up were classified according to the clinical presentation and the findings at ERCP. Cholangitis was defined as the combination of fever, chills, colicky pain, cholestatic liver function tests, and/or dilatation of the bile ducts. Patients were clinically classified as having had symptoms of stone recurrence if they had at least three of the following symptoms: jaundice, colicky pain, cholestatic liver function tests, or dilatation of the ducts on radiologic imaging. The majority of ERCPs for late complications were performed at referring centers. In general, the size of the sphincterotomy opening was not documented in the endoscopy reports. Therefore, patients were considered to have had stenosis of the sphincterotomy opening if this was either explicitly mentioned in the endoscopy report or if a repeat sphincterotomy was performed for extraction of recurrent stones. VOLUME 44, NO. 6, 1996
Table 2.
Early complications (<30 days) after endoscopic sphincterotomy and stone extraction; 14 of 94 patients developed early complications Early complications
N = 94
Fatal
Fever/cholangltis Bleeding Pancreatitis Perforation Overall
7 2 3 2 14 (15%)
1 1 (1%)
RESULTS I n f o r m a t i o n was available on 94 p a t i e n t s (94%). In the m a j o r i t y of cases (87%) i n f o r m a t i o n was obtained from both the general practitioner a n d the p a t i e n t (or patient's family). I n f o r m a t i o n was not available on 6 patients because n e i t h e r the p a t i e n t nor the general practitioner could be contacted. O f the patients alive at time of follow-up, 88% completed the questionnaires a n d 89% h a d a blood sample taken. T h e r e were 26 m e n and 68 w o m e n with a m e a n age of 51 y e a r s a t the time of E R C P (range, 23 to 60 years). At the t i m e of the endoscopic sphincterotomy, 9 p a t i e n t s h a d t h e i r gallbladder still in situ. In the rem a i n i n g 85 patients (90%), the m e d i a n i n t e r v a l from p r i m a r y s u r g e r y to s p h i n c t e r o t o m y was 17 m o n t h s (range 8 days to 39 years). The surgical history in this group comprised cholecystectomy with choledochal exploration in 41 p a t i e n t s a n d w i t h o u t such exploration in 37. Seven p a t i e n t s h a d u n d e r g o n e one or more reoperations for bile duct stones. E a r l y complications (within 30 days after endoscopic sphincterotomy) occurred in 14 patients (15%). F e v e r after E R C P was the m o s t f r e q u e n t early complication (Table 2) a n d m a i n l y occurred in p a t i e n t s in w h o m spontaneous passage of the stones was awaited after endoscopic sphincterotomy. One p a t i e n t died 25 days after endoscopic s p h i n c t e r o t o m y of a pleural emp y e m a secondary to a r e t r o p e r i t o n e a l perforation. Therefore, 93 p a t i e n t s were available for long-term follow-up. The 9 p a t i e n t s w i t h a gallbladder in situ at the t i m e of endoscopic s p h i n c t e r o t o m y u n d e r w e n t elective cholecystectomy after a m e d i a n period of 22 days (range 9 to 53).
Long-term follow-up N i n e t y - t h r e e p a t i e n t s were followed-up for a median period of 15 y e a r s ( m e a n 14 years, r a n g e 3 to 18 years). F i g u r e 1 shows the K a p l a n - M e i e r plot for the absence of complications of these 93 patients, e.g., after 10 y e a r s of follow-up 82% of p a t i e n t s were free of complications since the endoscopic s p h i n c t e r o t o m y and stone extraction. T h i r t e e n of the 93 p a t i e n t s (14%) died d u r i n g follow-up. T h e r e was one biliary-related death: a 64year-old w o m a n died of cholangitis 4 years after the GASTROINTESTINAL ENDOSCOPY
645
Table 3. Late complications after endoscopic sphincterotomy for bile duct stones; 36 complications occurred in 22 out of 93 patients
100-
80-
Clinical classification
Patients Stonesor No. of No. of with stenosis episodes patients ERCP on ERCP
¢~ 60-
~'~ 40-
20-
2
4
6
8
10
12
14
16
18
Time (yrs)
Figure 1, Kaplan-Meier plot for the absence of late complications in 93 patients who underwent endoscopic sphincterotomy for removal of bile duct stones. Complications with a fatal outcome are marked with an asterisk. Example: Of the patients alive after 10 years of follow-up, 82% had not developed any late complication since the endoscopic sphincterotomy. initial endoscopic treatment. She presented in extremis with cholangitis and sepsis and died before she could undergo an ERCP. Autopsy was not performed. Overall, 22 patients (24%) developed a total of 36 late complications (range 1 to 6). The median time from endoscopic sphincterotomy to late complications was 3.5 years (range 4 months to 15 years). Although m a n y complications occurred in the first 4 years after endoscopic sphincterotomy, a significant number of complications occurred after 10 years or more (Fig, 1). In 21 of the 22 patients, late complications consisted of episodes with symptoms suggestive of recurrent bile duct stones (Table 3). An ERCP was performed in 20 patients and demonstrated recurrent stones in 13 patients, combined with stenosis of the sphincterotomy opening in 9 patients (Table 3). All patients were initially managed with endoscopic interventions and/or conservative treatment. A repeat sphincterotomy was performed in 9 patients. No serious adverse effects of endoscopic treatment occurred, but stone extraction failed in 1 patient who subsequently underwent choledochoduodenostomy.
Questionnaires and liver function tests Of the 80 patients alive at the time of follow-up, 70 (88%) returned the questionnaire. Four patients reported recent attacks of colicky pain in the right upper abdomen and subsequently underwent repeat ERCP. Two of these patients also had mild liver function abnormalities. Blood samples were obtained in 71 patients (89%) and cholestatic liver function tests were found in 16 (AP and/or GGT >3 times upper limit, 5 patients; AP and/or GGT <3 times upper limit, 11 646
GASTROINTESTINAL ENDOSCOPY
Cholangitis* 16 9 8 8 Symptoms of stone 16 9 9 5 recurreneet Episodes of colicky 3 3 2 0 pain Pancreatitis 1 1 1 0 Total 36 22 20 13 *Cholangitis:combinationof fever,chills,colickypain, cholestatic liver functiontests, and/or dilatationof the bile ducts. ?Symptoms of stone recurrence: at least three of the following: jaundice, colickypain; cholestaticliver functiontests, dilatationof the ducts on radiologicimaging. patients). In 5 patients these abnormalities were considered secondary to medication or unrelated comorbidity. Five other patients refused further investigation. The remaining 6 patients underwent repeat ERCP.
Repeat ERCP A total of 8 patients underwent repeat ERCP at the time of follow-up. The indications for repeat endoscopy were colicky pain attacks (n = 2), cholestatic liver function tests (n = 4), or both (n = 2). At ERCP the sphincterotomy opening was found patent in all patients and permitted easy passage of a 10 mm extraction balloon. In one patient brown pigment stones and sludge were removed. The other patients showed no abnormalities. To avoid bias, patients who underwent repeat ERCP for either abdominal pain or mild cholestatic liver function tests only were not classified as having had late complications. Two patients who underwent repeat ERCP for recent colicky pain attacks, dark urine, and transient cholestatic liver function tests were considered to have had recurrence of symptoms.
Relationship between patient classification and long-term follow-up Of the 93 patients available for long-term follow-up, 38 patients were considered to have had primary bile duct stones at the time of sphincterotomy and 53 patients fulfilled the criteria for secondary bile duct stones. Two patients could not be classified because information was missing on the exact date of the prior cholecystectomy. Table 4 shows patient characteristics and the cumulative incidence of late complications in these two groups. Of the 38 patients with primary bile duct stones, 11 patients (29%) developed late complications whereas late biliary complications occurred in 11 of the 53 patients (21%) with secondary VOLUME 44, NO. 6, 1996
bile duct stones (chi squared = 0.43, p = 0.51, chisquared test with Yates' correction).
DISCUSSION This series reports the rate of late complications after endoscopic sphincterotomy for removal of bile duct stones in 93 patients aged 60 years or younger at the time of initial endoscopic treatment. During a median follow-up of 15 years, 24% of the patients developed late complications. This complication rate is higher t h a n that known from other follow-up studies after endoscopic sphincterotomy (Table 5). 6, t8-23The higher complication rate in the present series could be due to a reduced size of the sphincterotomy, inasmuch as all patients were treated between 1976 and 1980, at which time the optimal size of the sphincterotomy incision may have been less established. However, in the patients who underwent repeat endoscopy 15 years after the initial ERCP, the sphincterotomy opening was found widely patent, which suggests t h a t adequate sphincterotomies were performed. In our opinion the higher complication rate in our series is more a reflection of the small number of patients lost to follow-up, the data collection from multiple sources, and the considerably longer period of observation. Although the cumulative incidence of complications was high (24%) and 44% of complications consisted of episodes of cholangitis, the related mortality rate was low (1%) and only one patient required surgical treatment. The remaining patients were effectively managed endoscopically or conservatively. Furthermore, more serious complications such as recurrent ascending cholangitis, secondary biliary cirrhosis, or malignant degeneration were not observed. However, the power ofthis study to detect an increased relative risk for such complications may be too small, and Kurumado et al. 15 have recently suggested that malignant degeneration needs at least I0 years to develop. The vast majority of complications consisted of episodes with symptoms suggestive of recurrent bile duct stones. How plausible is the theory that loss of sphincter function after endoscopic sphincterotomy leads to formation of recurrent bile duct stones? Several authors have demonstrated that the biliary tree becomes infected with bacteria after endoscopic sphincterotomyg, I0 and there is substantial evidence that bacteria play an essential role in the formation of brown pigment stones. Some bacterial species, especially E s c h e r i c h i a coli, 27 produce enzymes like ~-glucuronidase t h a t are known to precipitate bilirubin and calcium, 13 the main components of brown pigment stones. 2s Furthermore, electron microscopy studies have demonstrated t h a t bacteria are present in the core of brown pigment stones whereas they are absent in the cholesterol and the black pigment stones. 16, 29, 30 Unfortunately, the appearance of recurrent stones at VOLUME 44, NO. 6, 1996
Table 4. Patient characteristics and rate of late complications in patients who underwent endoscopic sphincterotomy for treatment of primary and secondary bile duct stones* Primary bile Secondarybile duct stones, duct stones, (N = 38) (N = 53) Men/women 13/25 12/41 Median age in y (range) 56 (31-60) 52 (23-60) Cholecystectomy 38 44 T-tube in situ -31 Elective cholecystectomy -9 after ERCP Number of patients with 11 (29%) 11 (21%) complications Number of patients with 7 6 stones/stenosis on ERCP Number of episodes of 19 17 complications Time from EST until 35 (12-133) 57 (4-175) complication(mo) *Fordefinitionsof primaryand secondarybileductstonessee text. repeat ERCP was not consistently documented in our study. Others, however, have demonstrated that recurrent stones in general have the appearance of brown pigment stones. 3°-32 Apart from bacterial infection, stasis of bile is thought to be an important factor in the pathogenesis of recurrent bile duct stones, la, a3 Although endoscopic sphincterotomy leads to bacterial contamination of bile, it also effectively drains the bile duct, which enables early passage of stones formed de novo. Recurrent stones should, therefore, develop only in the event of stenosis of the sphincterotomy opening. There are, however, several reports that demonstrate that stone recurrence can actually occur in a well-drained bile duct.16, 19, 34 In our series the size of the sphincterotomy opening was not documented in patients who underwent ERCP for late complications. The fact that four patients with recurrent stones had their stones extracted without the need of repeat sphincterotomy indicates that stenosis of the sphincterotomy opening is not a prerequisite for stone formation. Are recurrent stones caused by secondary effects after endoscopic sphincterotomy or may the underlying stone disease be held responsible? A drawback of this study is the lack of a control group t h a t would enable us to ascertain the relative influence of the underlying stone disease on the rate of late complications. The ideal control group would consist of patients who underwent treatment for bile duct stones without sphincter ablation, i.e., patients who underwent surgical bile duct re-exploration. 35 Several cohort studies have been published in the literature describing late complications after surgical treatment for bile duct GASTROINTESTINAL ENDOSCOPY 647
Table 5. Studies describing late complications after endoscopic sphincterotomy for bile duct stones Reference
Follow-up
N (%) Riemann is 340 (?) RSsch19 248 (79) Kullmann 2° 118 (100) Hawes21 115 (71) Testoni22 96 (100) Escourrou 6 96 (68) Jacobsen 23 52 (100) Present study 94 (94) *Mean, median not given.
Follow-up in months
Biliary problems
Median (range) 54 (24-108) ? (0-84) 57 (24-101) 96* (78-132) 48* (24-96) 24 (6-60) 73 (8-103) 180 (36-216)
N (%) 60 (18) 32 (13) 7 (6) 15 (13) 14 (15) 5 (5) 4 (8) 22 (24)
stones. However, c o m p a r i n g the outcome of these surgical studies w i t h the results of the p r e s e n t series is difficult. A p a r t from problems deriving from differences in d a t a collection a n d d u r a t i o n of follow-up, the m a j o r c a v e a t is the difference in p a t i e n t selection of t h e s e studies. 35 Most surgical follow-up studies concern p a t i e n t s t r e a t e d for p r i m a r y bile duct stones. In our series, 29% of p a t i e n t s w i t h p r i m a r y duct stones developed late complications d u r i n g a m e d i a n follow-up of 15 years. W h e n corrected for differences in l e n g t h of follow-up, these results seem to t a k e a n int e r m e d i a t e position b e t w e e n the results of duct re-exploration w i t h o u t s p h i n c t e r o t o m y (18% to 33% after 5 to 10 y e a r s of follow-up) 31, 34, 36, 37 a n d surgical drainage p r o c e d u r e s (surgical s p h i n c t e r o t o m y or choledochoduodenostomy, 0% to 7% after 3 to 12 y e a r s of follow-up). 36-4° This suggests t h a t for p a t i e n t s with prim a r y bile duct stones, the loss of s p h i n c t e r function after s p h i n c t e r o t o m y is not a major pathogenetic factor in the f o r m a t i o n of r e c u r r e n t stones. This is also ill u s t r a t e d b y the fact t h a t in p a t i e n t s w i t h p r i m a r y bile duct stones, r e c u r r e n t stones t h a t develop after sphinct e r o t o m y are of t h e same type as the original stones: b r o w n p i g m e n t stones. 3°32 In our series, 21% of p a t i e n t s with secondary stones developed late complications after endoscopic sphincterotomy. Surgical series on long-term follow-up after t r e a t m e n t for secondary bile duct stones are e i t h e r incomplete 41 or outdated. 42 Malet et al.43 showed t h a t ductal stones found at the same t i m e as, or w i t h i n several m o n t h s after, cholecystectomy were e i t h e r cholesterol stones or black p i g m e n t stones t h a t probably m i g r a t e d from t h e gallbladder (comparable w i t h the s e c o n d a r y bile duct stones in our study). However, r e c u r r e n t bile duct stones t h a t developed in these pat i e n t s were shown to be brown p i g m e n t stones. 44 T h e fact t h a t r e c u r r e n t stones t h a t develop after sphinct e r o t o m y in p a t i e n t s w i t h secondary bile duct stones are totally different in visual a p p e a r a n c e a n d chemical composition from the original stones 32 points 648 G A S T R O I N T E S T I N A L E N D O S C O P Y
Patients with ERCP 121 70 7 45 14 5 4 20
Stones/stenosis on ERCP N (%) 26 (8) 10 (4) 7 (6) 5 (4) 12 (13) 5 (5) 2 (4) 13 (14)
towards the s p h i n c t e r o t o m y as a possible causative factor in the f o r m a t i o n of r e c u r r e n t stones. The results of this s t u d y show t h a t a significant proportion of p a t i e n t s t r e a t e d with endoscopic sphinct e r o t o m y for bile duct stones at the age of 60 years or y o u n g e r develop r e c u r r e n t problems d u r i n g long-term follow-up. The efficacy of endoscopic s p h i n c t e r o t o m y in removing stones, the low r a t e of serious early complications, and the fact t h a t late complications in general can be m a n a g e d with endoscopic a n d conservative m e a s u r e s , m a k e the endoscopic s p h i n c t e r o t o m y still the t r e a t m e n t of choice for p a t i e n t s with p r i m a r y bile duct stones. However, for y o u n g e r p a t i e n t s with seco n d a r y bile duct stones, s p h i n c t e r o t o m y m a y be causally r e l a t e d to a h i g h e r r a t e of stone recurrence. In these p a t i e n t s endoscopic stone r e m o v a l w i t h o u t ablation of the biliary sphincter (i.e., after endoscopic balloon dilation 25, 26) m a y lead to a lower r a t e of stone r e c u r r e n c e a n d o t h e r late complications.
REFERENCES 1. Classen M, Demling L. Endoskopische sphincterotomie der papilla vateri und steinextraktion aus ductus choledochus.Dtsch Med Wochenschr 1974;99:496-7. 2. Kawai K, Akasaka Y, Murakami K, Tada M, Nakajima M. Endoscopic sphincterotomy of the ampulla of Vater. Gastrointest Endosc 1974;20:148-51. 3. Cotton PB, Lehman G, Vennes G. Endoscopic sphincterotomy, complications and their management: an attempt at consensus. Gastrointest Endosc 1991;37:383-93. 4. Vaira D, Ainley C, Williams S, et al. Endoscopicsphincterotomy in 1000 consecutive patients. Lancet 1989;2:431-4. 5. Leese T, Neoptolemos JP, Carr-Locke DL. Successes, failures, early complications and their management following endoscopic sphincterotomy: results in 394 consecutive patients from a single centre. Br J Surg 1985;72:215-9. 6. Escourrou J, Cordova JA, Lazorthes F, Frexinos J, Ribet A. Early and late complications after endoscopic sphincterotomy for biliary lithiasis with and without the gall bladder "in situ." Gut 1984;25:598-602. 7. Freeman M, Nelson D, Sherman S, et al. Complications of endoscopic sphincterotomy (ES): a prospective, multicenter, 30day outcome study [abstract]. Gastroenterolog3~1994;106:A338. 8. Geenen JE, Toouli J, Hogan WJ, et al. Endoscopic sphincterotomy: follow-up evaluation of effects on the sphincter of Oddi. Gastroenterolog3r1984;87:754-8. V O L U M E 44, NO. 6, 1996
9. Gregg JA, De Girolami P, Can" Locke DL. Effects ofsphincteroplasty and endoscopic sphincterotomy on the bacteriologic characteristics of the common bile duct. Am J Surg 1985;149: 668-71. 10. Sand J, Airo I, Hittunen KM, Mattila J, Nordback I. Changes in biliarybacteria after endoscopic cholangiographyand sphincterotomy. Am Surg 1992;58:324-8. 11. Flemma RJ, Flint LM, Osterhout S, Shingleton WW. Bacteriologic studies of biliary tract infection. Ann Surg 1967;166:56372. 12. Scott AJ, Khan GA. Origin of bacteria in bile duct bile. Lancet 1967;2:790-4. 13. Maki T. Pathogenesis of calcium bilirubinate gallstone: role of E. coli, beta*glucuronidase and coagulation by inorganic ions, polyelectrolytes and agitation. Ann Surg 1966;164:90-100. 14. Greenfield C, Cleland P, Dick R, Masters S, Summerfield JA, Sherlock S. Biliary sequelae of endoscopic sphincterotomy. Postgrad Med J 1985;61:213-5. 15. Kurumado K, Nagai T, Kondo Y, Abe H. Long-term observations on morphological changes of choledochal epithelium after choledochoenterostomy in rats. Dig Dis Sci 1994;39:809-20. 16. Goldman LD, Steer ML, Silen W. Recurrent cholangitis after biliary surgery. Am J Surg 1983;145:450-4. 17. Kinami Y, Ashida Y, Seto K, Takashima S, Kita I. Influence of incomplete bile duct obstruction on the occurrence of cholangiocarcinoma induced by di-isopropanolnitrosamine in hamsters. Oncology 1990;47:170-6. 18. Riemann JF, Lux G, Forster P, AltendorfA. Long-term results after endoscopic papillotomy. Endoscopy 1983;15:165-8. 19. Rosch W, Riemann JF, Lux G, Lindner HG. Long-term follow-up after endoscopic sphincterotomy. Endoscopy 1981;13: 152-3. 20. Kullman E, Borch K, Liedberg G. Long term follow up after en~ doscopic management of retained and recurrent common duct stones. Acta Chir Scand 1989;155:395-9. 21. Hawes RH, Cotton PB, Vallon AG. Follow-up 6 to 11 years after duodenoscopic sphincterotomy for stones in patients with prior cholecystectomy. Gastroenterology 1990;98:1008-12. 22. Testoni PA, TittobelloA. Long-term efficacyofendoscopic papillo-sphincterotomy for common bile duct stones and benign papillary stenosis. Surg Endosc 1991;5:135-9. 23. Jacobsen O, Matzen P. Long-term follow-up study of patients after endoscopic sphincterotomy for choledocholithiasis. Scand J Gastroenterol 1987;22:903-6. 24. Cotton PB. Endoscopic retrograde cholangiopancreatography and laparoscopic cholecystectomy. Am J Surg 1993;165:474-8. 25. Mac Mathuna P, White P, Clarke E, Lennon J, Crowe J. Endoscopic sphincteroplasty: a novel and safe alternative to papillotomy in the management of bile duct stones. Gut 1994;35:127-9. 26. Bergman JJGHM, Rauws EAJ, Tytgat GNJ, Huibregtse K. A prospective randomized trial comparing endoscopic sphincter-
VOLUME 44, NO. 6, 1996
27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40.
41. 42. 43. 44.
otomy (EST) with endoscopic balloon dilatation (EBD) for removal of common bile duct stones (CBDS); initial report [abstract]. Gastrointest Endosc 1994;40:P99. Tabata M, Nakayama F. Bacteria and gallstones: etiological significance. Dig Dis Sci 1981;26:218-24. Stewart L, Smith AL, Pellegrini CA, Motson RW, Way LW. Pigment gallstones form as a composite of bacterial microcolohies and pigment solids. Ann Surg 1987;206:242-50. Kaufman HS, Magnuson TH, Lillemoe KD, Frasca P, Pitt HA. The role of bacteria in gallbladder and common duct stone formation. Ann Surg 1989;209:584-91. Smith AL, Stewart L, Fine R, Pellegrini CA, Way LW. Gallstone disease: the clinical manifestations of infectious stones. Arch Surg 1989;124:629-33. Braasch JW, Fender HR, Bonneval MM. Refractory primary common bile duct stone disease. Am J Surg 1980;139:526-30. Cetta F. The possible role of sphincteroplasty and surgical sphincterotomy in the pathogenesis of recurrent common duct brown stones. HPB Surg 1991;4:261-70. Madden JL. Common duct stones: their origin and surgical management. Surg Clin North Am 1973;53:1095-113. Saharia PC, Zuidema GD, Cameron JL. Primary common duct stones. Ann Surg 1977;185:598-604. Cotton PB. Endoscopic management of bile duct stones (apples and oranges). Gut 1984;25:587-97. Allen B, Shapiro H, Way LW. Management of recurrent and residual common duct stones. Am J Surg 1981;142:41-7. Lygidakis NJ. Surgical approaches to recurrent choledocholithiasis: choledochoduodenostomy versus T-tube drainage after choledochotomy. Am Surg 1983;145:636-9. Escudero-Fabre A, Escallon A Jr, Sack J, Halpern NB, Aldrete JS. Choledochoduodenostomy: analysis of 71 cases followed for 5 to 15 years. Ann Surg 1991;213:635-42. Cuschieri A, Wood RAB, Metcalf MJ, Cumming JGR. Longterm experience with transection choledochoduodenostomy. World J Surg 1983;7:502-4. Thomas CG Jr, Nicholson CP, Owen J. Effectiveness of choledochoduodenostomy and transduodenal sphincterotomy in the treatment of benign obstruction of the common duct. Ann Surg 1971;173:845-56. Cetta F. Do surgical and endoscopic sphincterotomy prevent or facilitate recurrent common duct stone formation? Arch Surg 1993;128:329-36. Larson RE, Hodgson JR, Priestley JT. The early and long term results of 500 consecutive explorations of the common duct. Surg GynecoI Obstet 1966;122:744-50. Malet PF, Dabezies MA, Huang GH, Long WB, Gadacz TR, Soloway RD. Quantitative infrared spectroscopy of common bile duct gallstones. Gastroenterology 1988;94:1217-21. Whiting MJ, Watts JM. Chemical composition of common bile duct stones. Br J Surg 1986;73:229-32.
G A S T R O I N T E S T I N A L E N D O S C O P Y 649