Endoscopic Sphincterotomy and Risk of Malignancy in the Bile Ducts, Liver, and Pancreas

CLINICAL GASTROENTEROLOGY AND HEPATOLOGY 2008;6:1049 1053 Endoscopic Sphincterotomy and Risk of Malignancy in the Bile Ducts, Liver, and Pancreas CECILIA STRÖMBERG,* JUHUA LUO, LARS ENOCHSSON,* URBAN ARNELO,* and MAGNUS NILSSON* *Department of Clinical Science, Intervention and Technology, Karolinska Institute, and Department of Surgery, Gastrocenter, Karolinska University Hospital; and Department of Medical Epidemiology and Biostatistics, Karolinska Institute, Stockholm, Sweden Background & Aims: After endoscopic sphincterotomy (ES), an elevated long-term risk of cholangiocarcinoma has been reported. However, large population-based studies testing this hypothesis are lacking. The aim of this study was to evaluate the risk in a large population-based cohort. Methods: Data concerning all patients having had an inpatient endoscopic retrograde cholangiopancreatography (ERCP) were collected from the Swedish Hospital Discharge Register. Incident cases of malignancy were identified through linkage to the Swedish Cancer Registry. Patients with a diagnosis of malignancy before or within 2 years of the ERCP were excluded. The cohort was followed to a diagnosis of malignancy, censoring as a result of death, emigration, or end of follow-up. The risk of malignancy was calculated as standardized incidence ratio (SIR) compared with the general population, inherently adjusting for age, gender, and year of entry. Results: A total of 27,708 patients undergoing ERCP from 1976 through 2003 for benign disease were included in the cohort. ES was performed in 11,617 of these. The risk of malignancy in the bile ducts alone and in the bile ducts, liver, and pancreas together was significantly elevated in the total cohort (SIR, 3.3; 95% confidence interval, 2.3 4.5), irrespective of whether an ES was performed. The risk of malignancy diminished with increasing follow-up time. Patients ever having had a cholecystectomy had a significantly lower risk of the studied malignancies. Conclusions: The risk of malignancy in the bile ducts, liver, or pancreas is elevated after ERCP in benign disease. However, ES does not seem to affect this risk. Endoscopic retrograde cholangiopancreatography (ERCP) with endoscopic sphincterotomy (ES) was first introduced in 1974 in Germany and Japan 1,2 and in Sweden a few years later as an alternative to open surgery with choledochotomy in the treatment of common bile duct stones (CBDS). It was mainly used in elderly patients or patients with severe comorbidity, in whom mortality was reduced compared with open surgery, 3 whereas young and healthy patients were still predominantly treated with open surgical procedures. Since the introduction of laparoscopic cholecystectomy, ERCP with ES, with or without cholecystectomy, has become the most common procedure in the treatment of CBDS worldwide, during the last years only to some extent challenged by laparoscopic techniques. 4 A consequence of this alteration in therapeutic regime is that an increased number of young patients with a long life expectancy will go through an ES. There are reports of long-term adverse effects of ES. The interruption of the sphincter of Oddi has been reported to cause an overgrowth of bacteria 5 in the common bile duct, resulting in cholangitis and recurrent stone formation. 6 There have also been reports of development of cholangiocarcinoma, assumingly as a result of the bacterial overgrowth and chronic inflammation in the bile ducts 6,7 in conformity with other reports concerning the development of malignancies after open bilioenteric anastomoses 8,9 and transduodenal sphincterotomy. 10 A previous, small population-based study did, however, not provide any support for the hypothesis of an increased risk of cancer after ES. 11 The primary aim of this population-based cohort study was to evaluate the relationship between ES for benign disease and subsequent development of malignancy in the biliary tract. A secondary aim was to study the relation between severe CBDS exposure, which is valid for the entire ERCP cohort irrespective of ES, and malignancy in the biliary tract. Because of expected extensive misclassification between extrahepatic bile duct cancers and pancreatic cancer, as well as between intrahepatic bile duct cancers and liver cancer, both bile duct cancer alone and bile duct cancer together with pancreatic and liver cancers were used as outcome parameters. Materials and Methods Registries, Cohort, and Follow-up A detailed description of the methods used in this large, retrospective cohort study has been described elsewhere. 12 In brief, we used data from the Swedish Hospital Discharge Register (inpatient statistics), a register that was established in 1965, in which discharge diagnoses and surgical procedures were computerized for each hospitalization, with the patients unique national registration numbers for identification. The coverage of the Hospital Discharge Register was 60% in 1969, 85% in 1983, and included all Swedish hospitals from 1987 and thereafter. 13 From 1965 2003, we identified all unique national registration numbers with at least one in-hospital episode with a discharge procedure code for ERCP or ES (Swedish Classifica- Abbreviations used in this paper: CBDS, common bile duct stones; CI, confidence interval; ERCP, endoscopic retrograde cholangiopancreatography; ES, endoscopic sphincterotomy; ICD-7, International Classification of Diseases, seventh revision; SIR, standardized incidence ratio. 2008 by the AGA Institute 1542-3565/08/$34.00 doi:10.1016/j.cgh.2008.04.016

1050 STRÖMBERG ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 6, No. 9 tion of Operations and Major Procedures, codes 9014, UJK02, UJK05, UJK12, UJK15 for ERCP or 5388, 5394, JKE 02, JKE 12, JKE 15, JKE 18, JKE 25, JKE 98 for ES or procedures for which ES normally is a prerequisite). This total ERCP cohort was divided into 2 subgroups: (1) patients having at least 1 procedure code registration for ES or any other endoscopic biliary procedures for which an ES normally is a prerequisite (Swedish Classification of Operations and Major Procedures, codes 5388, 5394, JKE 02, JKE 12, JKE 15, JKE 18, JKE 25, JKE 98); and (2) patients in the cohort without any procedure code registration for ES or any other endoscopic biliary procedure implying ES. Follow-up of all persons identified after any ERCP procedure was performed by using the national registration numbers for cross-linkage with the 98% complete 14 Swedish National Cancer Registry for cancer outcome ascertainment and the registries of Causes of Death and Domestic and International Relocations for censoring in the event of death or emigration of a cohort member. Those patients who had a diagnosis of malignant or benign tumor in the bile ducts, liver, or pancreas (International Classification of Diseases, seventh revision [ICD-7] 155 157, 211, 230, or corresponding codes in the later classifications) at the time of the procedure or within 2 years after it were excluded from further analyses to avoid bias, because the registered ERCP in these cases might have been performed because of the tumor or as a result of symptoms caused by a tumor that was still undiagnosed. Considering the poor prognosis of malignancies in the biliary tract, liver, and pancreas, it is highly unlikely that a tumor causing symptoms would be diagnosed more than 2 years later. The cohort was then followed from entry until diagnosis of an outcome malignancy (primary malignant tumors in the liver, bile ducts including ampullary region, and pancreas, but excluding gallbladder malignancy, ICD-7 codes 155 and 157, but excluding 1551), death, emigration, or end of follow-up (December 31, 2003), whichever occurred first. To assess the general risk of malignancy in the cohort, the SIR for all malignancies (ICD-7 140 209) was performed. Moreover, the SIR for lung malignancy (ICD-7 162 163) was calculated as an indirect estimate of tobacco smoking exposure in the cohort, because smoking might be a relevant confounder, especially concerning pancreatic cancer risk analyses. The study was approved by the Regional Research Ethics Committee of Stockholm. Table 1. Size of Cohorts of Patients Having Undergone ERCP With or Without ES All ERCP ES ERCP without ES Total 54,135 30,431 25,846 Date before entry 10,085 5,734 4,764 Error in registration 3,192 3,000 329 Follow-up less than 2 years 13,150 9,068 5,368 Eligible 27,708 12,629 15,385 Statistical Analyses Several patients had ERCP or ES procedures registered at more than one point in time. As index procedure for cancer, relative risk analyses were, first of all, every first time procedure for every patient. If the first time procedure included or implied ES, the patient s person-time was only included in the ES subgroup. If, on the contrary, a patient s first time procedure was non-es or ES-implying and followed by a subsequent procedure including a code for or implying ES, this patient had 2 index procedures: one without ES, with person-time counted in the non-es subgroup from 2 years after this procedure until the subsequent second index procedure ES, after which persontime was counted in the ES subgroup from 2 years after that procedure and on. The standardized incidence ratio (SIR), the ratio of the observed to the expected number of malignancies, was used to calculate relative risk. The expected number of cancers occurring in the entire Swedish population was calculated by multiplying the observed person-time by age-specific (in 5-year groups), gender-specific, and year of entry specific cancer incidence rates. The SIRs are inherently adjusted for confounding by age at follow-up, gender, and year of entry. The 95% confidence intervals (CIs) of SIRs were calculated, assuming that the observed numbers followed a Poisson distribution. 15 Results The final cohort of patients having undergone ERCP for diagnostics or therapy of nonmalignant disease included 27,708 patients contributing to a total of 235,518 person-years of follow-up. Table 1 shows the steps of selection of patients from the total cohort including all ERCP procedures to the eligible final benign disease ERCP cohort. The sum of the ES and non-es groups is larger than the all-ercp group, because 2,141 patients had an ERCP without ES before a subsequent ERCP with ES. Of these, 306 did not have a diagnosis of malignancy at the time of the first or second index procedure and not within 2 years thereafter and were thus counted in both subgroups. General characteristics of the cohort and the number of observed malignancies 2 years or more after the index procedure are listed in Table 2. The mean age at entry was 63 years, and there was a predominance of women. The mean follow-up time was 8.5 years in the all-ercp cohort but was shorter among patients who had had an ES. The number of bile duct malignancy is low, in spite of the large cohort, because of the low incidence of this condition in the studied population. The total risk of developing malignancy was significantly increased by 20% in the all-ercp cohort compared with the general Swedish population, with inherent adjustment for age, gender, and year of entry (SIR, 1.2; 95% CI, 1.2 1.3). This increase in risk was the same for ERCP with and without ES (Table 3). The risk of lung malignancy in the total ERCP cohort was similar to the risk in the general population (SIR, 1.1; 95% CI, 0.9 1.2), and this similarity was not affected by stratification by ES (Table 3). The risk of developing malignancy in the bile ducts, liver, or pancreas, explicitly excluding gallbladder malignancy, was 3-fold increased in the all-ercp cohort compared with the general population (SIR, 3.0; 95% CI, 2.6 3.4). Likewise, it was 3-fold increased in the ES subgroup (SIR, 3.0; 95% CI, 2.3 3.8) and in the non-es subgroup (SIR, 2.9; 95% CI, 2.4 3.3) (Table 3). Similarly, the risk of bile duct malignancy alone was increased just greater than 3 times (SIR, 3.3; 95% CI, 2.3 4.5) in the all-ercp cohort and did not differ significantly with (SIR, 3.6; 95% CI, 1.8 6.5) or without ES (SIR, 2.8; 95% CI, 1.7 4.2) (Table 3). In a separate analysis not shown in the table, the risk

September 2008 ES AND RISK OF MALIGNANCY 1051 Table 2. Selected Characteristics and Number of Patients Developing Malignancy More Than Two Years After ERCP With or Without ES All ERCP (N 27,708) ES (N 12,629) ERCP without ES (N 15,385) Mean age at entry, y 63.4 66.2 61.1 Gender (% male) 42.1 40.2 43.6 Mean follow-up, y 8.5 5.9 10.4 All malignancies (ICD-7 140 209), cases 2710 805 1867 Lung malignancy (ICD-7 162 163), cases 148 39 106 Malignancy of bile ducts, liver, or pancreas excluding gallbladder (ICD-7 155, 236 66 156 157 excluding 1551), cases Malignancy of bile ducts (ICD-7 155 excluding 155.0 and 155.1) 35 11 21 of bile duct malignancy among cohort members with stricter criteria for ES, by using only explicit ES codes and not both ES codes and other codes for endoscopic biliary procedures for which ES is normally a prerequisite, the risk was significantly increased 2.7 times (SIR, 2.7; 95% CI, 1.2 5.4) compared with the general Swedish population (data not shown). In the all-ercp cohort the risk of malignancy of the bile ducts, liver, or pancreas decreased gradually by increasing duration of follow-up. The relative risk (SIR) was 3.6, with 95% CI 3.0 4.3, during the period between 2 and 4 years after the index ERCP, and the point estimates for the relative risks between 5 and 9 years and 10 years and more after the index ERCP were 2.9 and 2.2, respectively (Table 4). The analyses stratified by ES showed a similar pattern, with decreasing point estimates for relative risks with longer follow-up. There was no significant difference between the ES and non-es subgroups in the relative risk estimates by duration (Table 4). The risk of malignancy in the bile ducts, liver, and pancreas 2 years or more after ERCP was significantly lower among patients who ever had undergone a cholecystectomy (SIR, 2.3; 95% CI, 1.8 2.9 in the all-ercp-group) compared with patients who had not (SIR, 3.4; 95% CI, 3.0 4.0). This finding was not at all affected by ES exposure (Table 5). Discussion This large, population-based cohort study was designed primarily to evaluate the concern raised by previous publications 6,7 that the chronic inflammation by longstanding bacterial overgrowth in the bile ducts after ES might lead to an increased risk of development of cholangiocarcinoma. We demonstrated an approximately 3-fold significantly increased risk of malignancy in the bile ducts alone, as well as in the bile ducts, liver, or pancreas taken together, among patients 2 years or more after ERCP with or without ES. There was no significant difference in the risk of malignancy in the bile ducts alone, or in the bile ducts, liver, or pancreas together, between patients who had had an ERCP without ES and those who had had an ES. In both the ES and non-es groups, the risk of malignancy in the bile ducts, liver, and pancreas together seemed to decrease with increasing time of follow-up. Among patients in the cohort who had ever had a cholecystectomy, the risk of malignancy in the bile ducts, liver, or pancreas was significantly lower than in the group that had not had a cholecystectomy, irrespective of ES. Strengths of our study include the population-based cohortdesign, with follow-up of a vast majority of patients having undergone ERCP procedures in Sweden during the study period. Another strength is the close to complete follow-up of the patients in the high-quality data registries mentioned above, avoiding differential misclassification and thus ensuring high internal validity. Furthermore, the large size of the cohort provided adequate statistical power, which enabled us to analyze relative risks of the studied malignancies after stratification for ES. A limitation of this study might be that cancer of the bile ducts is a rare malignancy with a low expected incidence even in a large cohort, limiting the subanalyses of cancer risk by follow-up time. Moreover, the mean follow-up time was only 5.9 years in the ERCP with ES group compared with 10.4 years in the ERCP without ES group, clearly reducing the precision of long-term cancer risk analyses in the ES patient group. However, this is a flaw that the present study shares with all previously published work concerning ES and cancer, and to date the present study is larger and has better precision than its predecessors. There is also a risk of misclassification between intrahepatic cholangiocarcinoma and liver cancer, as well as extrahepatic cholangiocarcinoma and pancreatic cancer. Therefore, both bile duct malignancy alone and bile duct malignancy Table 3. SIRs and 95% CIs for Diagnosis of Malignancies Two Years or More After ERCP on a Benign Indication All ERCP (N 27,708), ES (N 12,629), ERCP without ES (N 15,385), All malignancies (ICD-7 140 209) 1.2 (1.2 1.3) 1.2 (1.1 1.2) 1.2 (1.2 1.3) Lung malignancy (ICD-7 162 163) 1.1 (0.9 1.2) 1.0 (0.7 1.3) 1.1 (0.9 1.3) Malignancy of bile ducts, liver, or pancreas excluding 3.0 (2.6 3.4) 3.0 (2.3 3.8) 2.9 (2.4 3.3) gallbladder (ICD-7 155, 157 excluding 1551) Malignancy of bile ducts (ICD-7 155 excluding 155.0 and 155.1) 3.3 (2.3 4.5) 3.6 (1.8 6.5) 2.8 (1.7 4.2)

1052 STRÖMBERG ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 6, No. 9 Table 4. Expected (E) and Observed (O) Cases and SIRs With 95% CIs for Developing Malignancy in Bile Ducts, Liver, or Pancreas After ERCP, With or Without ES, by Duration All ERCP (N 27,708) ES (N 12,629) ERCP without ES (N 15,385) Years after procedure E O E O E O 2 4 32 113 3.6 (3.0 4.3) 13 40 3.1 (2.2 4.2) 19 72 3.8 (3.0 4.8) 5 9 31 88 2.9 (2.3 3.5) 8 25 3.0 (1.9 4.4) 22 56 2.5 (1.9 3.3) 10 16 35 2.2 (1.5 3.0) 1 1 0.9 (0.02 5.0) 15 28 1.9 (1.3 2.8) together with liver and pancreatic malignancies were used as outcomes, in spite of the fact that our hypothesis primarily concerns cholangiocarcinoma of the intrahepatic and extrahepatic bile ducts. The influence of this type of misclassification is predictable and thus tolerable, given the use of both outcomes described above. All patients with the diagnosis of a tumor in the biliary tract, liver, or pancreas at the time of the index procedure or within 2 years after it have been excluded from the cohort. Patients having uncharacteristic abdominal discomfort could have been subjects to ERCP on the assumption of a present gallstone disease when the symptoms instead were caused by an undiagnosed malignancy. Considering the poor prognosis of these tumors, it would be highly unlikely that they could give symptoms more than 2 years before the time of diagnosis. Patients with gallbladder malignancy (ICD-7 155.1) were also excluded from the cohort because gallbladder carcinoma is known to be strongly associated to gallstone disease, 16 a condition likely to be very common among patients having had an ERCP without a malignant diagnosis. This study demonstrates an approximately 3-fold increase in the risk of malignancy in the bile ducts alone as well as in the bile ducts, liver, and pancreas together in a large cohort of patients followed after ES. This increased risk is not likely to be caused by the ES procedure itself for 2 reasons. First, the risk of malignancy was equally increased among ERCP patients who had not had an ES. Second, if the increase in risk of malignancy would indeed have been caused by the ES or any other part of the ERCP procedure, the risk of these tumors would be expected to be low in the beginning of follow-up and then to gradually increase over time. On the contrary, in this cohort, both in the ES subgroup and in the non-es subgroup, the risk of malignancy is greatest at the beginning of follow-up, ie, between 2 and 4 years after the index procedure, and it then gradually diminishes with longer follow-up time. The finding of increased risk of malignancy in the bile ducts alone, as well as in the bile ducts liver, or pancreas together, after ERCP, with or without ES in nonmalignant disease, was new to us. In a study of long-term consequences of ES by Tanaka et al, 6 410 patients were followed an average of 10 years after ES, and carcinoma in the biliary tract was found in 8 patients, 3 of them late, giving a marked elevated risk of malignancy both in the short and in the long run, a finding that was attributed to the ES. The results of that study differ from ours by showing a much higher risk of development of carcinoma in the biliary tract, but they are also based on a much smaller cohort with higher risk of random error. On the other hand, in a previously published populationbased study of 992 patients who underwent ES at 6 different hospitals, Karlson et al 11 found no increase in the risk of cancer in the liver, bile ducts, or pancreas at follow-up 1 year or more after the procedure. This study, also performed in Sweden, was based on a regional subgroup of our own study cohort, thus with similar cancer incidence in the background population, although a smaller cohort with proportionately inferior precision in the cancer risk analyses. The finding of decreasing risk of malignancy with increasing follow-up time indicates that the increased risk of malignancy seen in the cohort, total, ES, and non-es, is likely to be caused by some common exposure before the ERCP procedures. The most likely candidate by far is gallstone disease in general and ductal gallstone disease in particular. Other known or suspected causal agents for malignancy in the bile ducts, liver, and pancreas include tobacco smoking, primary sclerosing cholangitis, and genetic factors. Concerning tobacco smoking, we performed control analyses of the risk of lung cancer, which is well-known to be heavily associated with smoking, without any elevation of risk in the whole ERCP cohort or in the subgroups with and without ES, thus clearly dismissing smoking as the cause of the elevated risk of the studied cancers in the whole ERCP cohort. Concerning sclerosing cholangitis and genetic factors, one would not expect the observed pattern of decreasing risk of cancer with increasing time of follow-up because these exposures do not end at the time of ERCP, as is most commonly the case with gallstone disease. Overall, 61% of the patients in our study had a discharge diagnosis of gallstone disease ever, and 46% of them had it at Table 5. Expected (E) and Observed (O) Cases and SIR With 95% CIs of Malignancy in Bile Ducts, Liver, or Pancreas (Excluding Gallbladder, ICD-7 155 157 Excluding 155.1) in Patients Having Had ERCP, With or Without ES, Stratified for Cholecystectomy (CE) or Not All ERCP ES ERCP without ES CE E O E O E O Yes 30 70 2.3 (1.8 2.9) 8 16 2.1 (1.2 3.4) 22 47 2.1 (1.6 2.8) No 48 166 3.4 (3.0 4.0) 14 45 3.3 (2.4 4.4) 34 109 3.2 (2.7 3.9)

September 2008 ES AND RISK OF MALIGNANCY 1053 discharge after the ERCP hospital admittance. In the subgroup having undergone an ES, 70% had a gallstone disease discharge diagnosis ever and 55% at the time of the procedure. Unfortunately, although the validity of the Swedish Hospital Discharge Register is high concerning procedure code classification, it is of varying quality concerning codes of discharge diagnosis, which often precludes scientific use. However, in light of few highly prevalent alternatives to gallstone disease as indications for ERCP, in such a large population-based cohort in which ma- stones. ANZ J Surg 2002;72:547 552. lignancies have been excluded, it is very likely that a vast5. Gregg JA, De Girolami P, Carr-Locke DL. Effects of sphincteroplasty and endoscopic sphincterotomy on the bacteriologic char- majority of the cohort members did indeed have gallstone disease. acteristics of the common bile duct. Am J Surg 1985;149:668 671. Gallstone disease is a known risk factor for development of cholangiocarcinoma, 17,18 6. Tanaka M, Takahata S, Konomi H, et al. Long-term consequence and cholecystectomy has been shown of endoscopic sphincterotomy for bile duct stones. Gastrointest to reduce the risk of bile duct cancer, especially 10 years or more Endosc 1998;48:465 469. after surgery. 19 One possible mechanism by which ductal gall - 7. Tranter SE, Thompson MH. Comparison of endoscopic sphincterotomy and laparoscopic exploration of the common bile duct. Br J stones could cause the high risk of malignancy observed during the early part of follow-up could be that they might have started Surg 2002;89:1495 1504. a slow, initially purely inflammatory but subsequently carcinogenic process before they were removed and that the process oenteric anastomosis 14 years after pancreatoduodenectomy for 8. Shields HM. Occurrence of an adenocarcinoma at the choledoch- continues after removal of the stones. This would be consistent benign disease. Gastroenterology 1977;72:322 324. with the observation that the risk of development of a malignancy diminished with time. The risk of development of malignancy in the bile ducts, liver, or pancreas was higher among patients who had not had a cholecystectomy before or after the ERCP, also indicating that the carcinogenesis might be connected to gallstone disease. The cholecystectomy, as to some extent also ES, would act to terminate the carcinogenic exposure in the case that gallstone disease is the true carcinogenic factor. Given this reasoning, one might also have expected that the risk of the studied malignancies would have been lower in the ES subgroup than in the non-es subgroup in the main analyses without stratification for cholecystectomy. Against this stands the quite likely possibility that the ES subgroup might represent a selection of patients with more severe ductal gallstone disease, compared with the group in which ERCP was performed without ES. This is to some extent supported by the higher proportion of gallstone discharge diagnoses in the ES subgroup. In conclusion, this study shows an elevated risk of malignancy both in the bile ducts alone and in the bile ducts, liver, or pancreas together after ERCP, regardless of whether ES was performed. The risk, however, diminished over time, indicating that ES or any other part of the procedure is unlikely to be the cause, but more likely carcinogenic exposure previous to the ERCP procedure, possibly from ductal gallstone disease. Finally, cohort members who ever had a cholecystectomy had significantly lower risk of the studied tumors, irrespective of ES, which might support gallstone disease as a true causal agent behind the observed increased risk of malignancy. 2. Kawai K, Akasaka Y, Murakami K, et al. Endoscopic sphincterotomy of the ampulla of Vater. Gastrointest Endosc 1974;20:148 151. 3. Davidson BR, Neoptolemos JP, Carr-Locke DL. Endoscopic sphincterotomy for common bile duct calculi in patients with gall bladder in situ considered unfit for surgery. Gut 1988;29:114 120. 4. Ludwig K, Lorenz D, Koeckerling F. Surgical strategies in the laparoscopic therapy of cholecystolithiasis and common duct 9. Haratake J, Horie A, Takeda N. [Cancer of the intra-pancreatic common bile duct observed 9 years after choledochojejunostomy]. Gan No Rinsho 1983;29:1367 1370. 10. Tocchi A, Mazzoni G, Liotta G, et al. Late development of bile duct cancer in patients who had biliary-enteric drainage for benign disease: a follow-up study of more than 1000 patients. Ann Surg 2001;234:210 214. 11. Karlson BM, Ekbom A, Arvidsson D, et al. Population-based study of cancer risk and relative survival following sphincterotomy for stones in the common bile duct. Br J Surg 1997;84:1235 1238. 12. Nyren O, McLaughlin JK, Gridley G, et al. Cancer risk after hip replacement with metal implants: a population-based cohort study in Sweden. J Natl Cancer Inst 1995;87:28 33. 13. Fall K, Ye W, Nyren O. Risk for gastric cancer after cholecystectomy. Am J Gastroenterol 2007;102:1180 1184. 14. Ye W, Lagergren J, Weiderpass E, et al. Alcohol abuse and the risk of pancreatic cancer. Gut 2002;51:236 239. 15. Breslow NE DN. The design and analysis of cohort studies: statistical methods in cancer research. vol II. Lyon, France: IARC Science Publishers, 1987. 16. Tazuma S, Kajiyama G. Carcinogenesis of malignant lesions of the gall bladder: the impact of chronic inflammation and gallstones. Langenbecks Arch Surg 2001;386:224 229. 17. Shaib YH, El-Serag HB, Davila JA, et al. Risk factors of intrahepatic cholangiocarcinoma in the United States: a case-control study. Gastroenterology 2005;128:620Statistical methods in cancer research.626. 18. Welzel TM, Graubard BI, El-Serag HB, et al. Risk factors for intrahepatic and extrahepatic cholangiocarcinoma in the United States: a population-based case-control study. Clin Gastroenterol Hepatol 2007;5:1221 1228. 19. Ekbom A, Hsieh CC, Yuen J, et al. Risk of extrahepatic bileduct cancer after cholecystectomy. Lancet 1993;342:1262 1265. References 1. Classen M, Demling L. [Endoscopic sphincterotomy of the papilla of vater and extraction of stones from the choledochal duct (author s transl)]. Dtsch Med Wochenschr 1974;99:496 497. Address requests for reprints to: Cecilia Strömberg, Department of Surgery, Gastrocenter K53, Karolinska University Hospital, Hälsov 141, 141 52 Huddinge, Stockholm, Sweden. e-mail: cecilia.stromberg@ karolinska.se; fax: 46858586910.