An angulated common bile duct predisposes to recurrent symptomatic bile duct stones after endoscopic stone extraction

Surg Endosc (2006) 20: 1594 1599 DOI: 10.1007/s00464-005-0656-x Ó Springer Science+Business Media, Inc. 2006 An angulated common bile duct predisposes to recurrent symptomatic bile duct stones after endoscopic stone extraction D. Keizman, 1 M. I. Shalom, 1 F. M. Konikoff 2 1 Department of Gastroenterology, Tel Aviv Sourasky Medical Center, 6 Weizmann Street, Tel Aviv 64239, Israel 2 The Minerva Center for Cholesterol Gallstones and Lipid Metabolism in the Liver, Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv 69978, Israel, and Department of Gastroenterology and Hepatology, Meir Medical Center, 59 Tshernichovsky Street, Kfar Saba, 44281, Israel Received: 22 September 2005/Accepted: 15 February 2006/Online publication: 20 August 2006 Part of this work was presented at the Digestive Disease Week in New Orleans, 16 20 May 2004, and published in abstract form in Gastrointestinal Endoscopy 2004;59: AB197 Correspondence to: F. M. Konikoff Abstract Background: Endoscopic sphincterotomy and stone extraction are standard procedures for the removal of bile duct stones. Stone recurrence can, however, occur in up to 25% of cases. Risk factors have been poorly defined, but are believed to be related to bile stasis. This study investigated whether an angulated common bile duct (CBD) that may predispose to bile stasis influences symptomatic stone recurrence after successful endoscopic therapy. Methods: This study included 232 consecutive patients (mean age, 64.1 years; 86 men) who had undergone therapeutic endoscopic retrograde cholangiopancreatography for bile duct stones. Data from the follow-up period (36 ± 17 months) were obtained from medical records and patient questioning. Common bile duct angulation and diameter were measured from the cholangiogram after stone removal. Results: Symptomatic bile duct stones recurred in 16% of the patients (36/232). Three independent risk factors were identified by multivariate analysis: an angulated CBD (angle, 145 ; relative risk [RR], 5.2; 95% confidence interval [CI], 2.2 12.5; p = 0.0002), a dilated CBD (diameter, 13 mm; RR, 2.6; 95% CI, 1.2 5.7; p = 0.017), and a previous open cholecystectomy (RR, 2.7; 95% CI, 1.3 5.9; p = 0.0117). Gender, age, urgency of procedure, or a periampullary diverticulum did not influence the recurrence rate. Conclusions: Angulation of the CBD ( 145 ) on endoscopic cholangiography, a dilated CBD, and a previous open cholecystectomy are independent risk factors for symptomatic recurrence of bile duct stones. The findings support the role of bile stasis in stone recurrence. Further studies using these data prospectively to identify high-risk patients are warranted. Key words: Bile duct angulation ERCP Recurrent bile duct stones Endoscopic sphincterotomy (EST) with stone extraction is a well-established therapeutic procedure for the treatment of bile duct stones [6, 11, 12]. Complications after the procedure are divided into early (within 3 months after the procedure) and late (more than 3 months after the procedure) complications [1, 2, 13]. Among the late complications are recurrent symptomatic bile duct stones, which despite increasing experience and success with the procedure, occur in 4% to 24% of patients [6, 11, 12, 14]. Bile duct stones found 6 months or more after endoscopic retrograde cholangiopancreatography (ERCP) generally are considered recurrent, as opposed to retained [2, 17, 19]. Reported risk factors for recurrence include dilation of the common bile duct (CBD) [8, 15, 18, 19] and the presence of a periampullary diverticulum [6, 7, 11, 16, 19]. These factors suggest an underlying mechanism involving bile stasis. None of these known risk factors is, however, uniformly predictive of stone recurrence. A single, measurable risk factor for identifying patients at high risk for recurrence would be desirable. The CBD generally is described as deviating to the right as it descends toward the duodenum [20]. This deviation (termed angulation in the current study), clearly visible on most cholangiograms, is an important determinant for the radiologic configuration of the duct. It is, however, a feature that has attracted little investigative attention. We speculated that an angulated bile duct might contribute to bile stasis, and hence be a predictor of recurrent symptomatic choledocholithiasis. Therefore, we performed a retrospective study to

1595 examine whether angulation of the CBD is an independent predictor of the recurrence rate for symptomatic bile duct stones after endoscopic therapy. Patients and methods Patients The study group consisted of patients who underwent EST for suspected CBD stones in our gastroenterology unit between January 1997 and December 2003. Patients with lack of evidence for stones during the procedure, stenosis (benign or malignant) of the CBD, unavailability for follow-up evaluation, loss of medical (including endoscopic) information, and a retained gallbladder with stones (1 month or more after the EST) were excluded from the study. The diagnosis of CBD stones was based on radiologic and endoscopic visualization. All the patients included in the study had cholecystectomy either before the ERCP or soon (within 1 month) thereafter. Patients with a dilated CBD during the initial ERCP also had an intraoperative cholangiogram (during the cholecystectomy) to exclude retained stones. Recurrence of CBD stones was defined as development of symptomatic stones not earlier than 1 year after the complete removal of the initial CBD stones. All patients were contacted by the same investigator (D.K.) for updated clinical follow-up data at the time of this study. Because this report deals only with symptomatic (clinically relevant) recurrence of bile duct stones, the end point of recurrence was diagnosed on clinical grounds, backed by ultrasonographic and/or endoscopic verification. Patients who were symptomatic during the follow-up period underwent clinical, biochemical (liver function tests), and radiologic (sonographic, magnetic resonance cholangiopancreatography [MRCP]) evaluation for diagnosis or exclusion of CBD stones. The final diagnosis was confirmed (or excluded) for those with suspected recurrence by a second ERCP. Preprocedure data Before the procedure, each patientõs medical data (including age, sex, and prior cholecystectomy) were recorded. Endoscopic procedure Endoscopic sphincterotomy was performed by an experienced endoscopist using a side-viewing Pentax duodenoscope (ED-3440T; Asahi Optical Co, Tokyo, Japan) and a standard pull-type papillotome. All sphincterotomies were complete, resulting in the appearance of air in the biliary tree. Common bile duct stones were extracted with a retrieval balloon or basket. For large stones, a mechanical lithotripter was used. During the first ERCP, the presence of CBD stones and a periampullary diverticulum were recorded. After stone removal, contrast material was injected, and an inflated balloon catheter (diameter, 15 mm) was withdrawn along the CBD to the duodenum to confirm clearance of the biliary tree. Assessment of bile duct angulation and diameter The angulation and diameter of the CBD (in this study, the entire bile duct from the ampulla to the bifurcation) were measured from the postprocedure cholangiogram with the patient in the prone position. Angulation was measured as the sharpest angle along the CBD from 1 cm below the bifurcation to 1 cm above the papilla (Fig. 1). The CBD diameter was measured at its widest point (corrected for magnification relative to the known diameter of the endoscope). Cholangiogram analysis was performed by two independent investigators (not involved in the patientõs treatment and blinded to the data and the other investigatorsõ measurements) to ensure reproducibility of the measurements and to lower the possibility of bias. Statistical analysis Categorical variables (gender, presence of a periampullary diverticulum, and type of cholecystectomy, whether open or laparoscopic) were compared using PearsonÕs chi-square test, or FisherÕs exact test when both variables were dichotomous. Continuous variables (age, diameter, and angulation of the CBD) were analyzed using the StudentÕs t- test when the variables were normally distributed, or by the Mann Whitney U test for variables without a normal distribution. Significant predictors for bile duct stone recurrence (p values less than 0.05) identified by univariate analysis were included in a multiple logistic regression model (i.e., multivariate analysis) to determine the most significant risk factors for recurrence of bile duct stones. The logistic regression analysis focused on relative risk. All the tests were two-sided, and a p value of 0.05 or less was considered statistically significant. Curves based on the Kaplan Meier method were used to estimate the proportion of patients free of recurrence during the follow-up period. Predictors of recurrence during the follow-up period were evaluated by the log rank test. Statistical analysis was conducted using the SPSS package 11 (SPSS Inc. 233 S. Wacker Drive, Chicago, IL 60606). The study was approved by the institutional review board of our hospital, and the research was conducted in accordance with the Helsinki declaration. Results Patients During the study period, 415 consecutive patients underwent EST for suspected CBD stones in our gastroenterology unit. Of these, 183 patients were excluded from the study because of no evidence for stones during the procedure (n = 31), stenosis (benign or malignant) of the CBD (n = 43), unavailability for follow-up evaluation (n = 41), loss of medical or endoscopic information (n = 47), or a retained gallbladder with stones 1 month or more after the EST (n = 21). Altogether, a total of 232 patients (146 women, 63%) in whom sphincterotomy and bile duct clearance had been achieved were included in the analysis. The mean age was 64.1 ± 18.3 years (median, 69 years; range, 8 93 years). Most of the procedures (63%, n = 145) were performed for hospitalized patients. There was no mortality related to the procedure or CBD stones among the study population. The mean follow-up period was 36 ± 17 months (median, 34 months; range, 12 86 months). There was no significant difference in the follow-up periods between the study groups (i.e., patients with and those without recurrent stones). Stone recurrence The overall recurrence rate for symptomatic bile duct stones was 16% (36 of 232 patients). The mean length of time until recurrence was 21.6 ± 9.2 months (median, 16 months; range, 12 59 months). Risk factors for recurrence The results of univariate analysis for recurrence of bile duct stones in relation to each factor are presented in Table 1. In terms of bile duct angulation (Fig. 1), an angle of 145 or less was predictive of stone recurrence

1596 Fig. 1. A A straight common bile duct. B An unangulated common bile duct. Table 1. Univariate analysis for recurrent bile duct stones Variable Recurrent group Nonrecurrent group p Value Preprocedure data Gender: % (n) F: 50 (18) M: 50 (18) F: 65 (128) M: 35 (68) 0.2 Admission: % (n) Elective 25 (9) 40 (78) 0.14 Hospitalized 75 (27) 60 (118) Age (years) 71.7 ± 12 62.5 ± 19 <0.0001 Age > 66 years: % (n) 72 (26) 51 (99) 0.031 Previous laparoscopic cholecystectomy: % (n) 47 (17) 75 (147) 0.02 Previous open cholecystectomy: % (n) 53 (19) 25 (49) 0.02 Procedure data Periampullary diverticulum: % (n) 33 (12) 25 (49) 0.23 CBD diameter 13 mm: % (n) 64 (23) 30 (59) <0.0001 Angulated CBD (angle 145 ): % (n) 78 (28) 36 (71) <0.0001 CBD, common bile duct (Table 1). The mean cholangiographic angulation was 141.25 ± 9.6 in patients with recurrence, as compared with 152.36 ± 18.3 in patients without recurrence (p = 0.013). Patients with a sharp angle ( 145 ) had a relative risk of 5.2 for stone recurrence, as compared with those with an angle exceeding 145 (95% CI, 2.2 12.5; p = 0.0002). Dilation of the CBD also was found to be a statistically significant risk factor for recurrence, with a mean diameter of 16.5 ± 5.2 mm (range, 11 30 mm) in patients who had recurrence and 13.8 ± 5 mm (range, 11 30 mm) in patients who had no recurrence (p = 0.005). Patients with a bile duct diameter of 13 mm or more had recurrence of stones significantly more frequently than those with a bile duct diameter of less than 13 mm (relative risk, 2.6; 95% CI, 1.2 5.7; p = 0.017). Patients who had undergone a previous open cholecystectomy (as compared with laparoscopic cholecystectomy) also had a significantly higher rate of recurrence (relative risk, 2.7; 95% CI, 1.3 5.9; p = 0.0117). The mean CBD angulation in patients after open cholecystectomy (146.8 ) was not different from that in patients after laparoscopic cholecystectomy (147.5 ). An age of 66 years or older was found in the univariate analysis to be a statistically significant risk factor for recurrence. Gender, urgency of procedure, and the presence of a peripampullary diverticulum were not statistically related to recurrence of bile duct stones. Multivariate analysis Predictors found to be significant on multivariate analysis were a CBD angulation of 145 or less, a dilated CBD (diameter, 13 mm), and a previous open cholecystectomy (Table 2). These three risk factors also were significant on Kaplan Meier analysis, especially for patients with coexistence of the three (Figs. 2 4). Approximately 25% of the patients with one risk factor had recurrent stones within 2.5 years, but when the three risk factors coexisted, about 40% had recurrent stones within this time. Age ( 66 years), which univariate analysis found to be a statistically significant risk factor for recurrence, was not found to be an independent risk factor by multivariate analysis.

1597 Table 2. Independent risk factors for recurrent bile duct stones in multivariate analysis Variable Relative risk (95% CI) p Value CBD angle 145 5.2 (2.2 12.5) 0.0002 CBD 13 mm 2.6 (1.2 5.7) 0.017 Previous open cholecystectomy 2.7 (1.3 5.9) 0.0117 CBD, common bile duct Discussion In this study, we identified three independent risk factors for symptomatic bile duct stone recurrence after successful endoscopic therapy: (1) angulation of the CBD ( 145 ), (2) dilation of the CBD (diameter, 13 mm), and (3) a previous open cholecystectomy. Of these, a dilated bile duct has been associated with stone recurrence in previous studies of post-est patients [6, 8, 11, 15, 18, 19], whereas angulation of the CBD and a previous open cholecystectomy have not been previously reported in the endoscopic literature. In general, it is difficult to interpret follow-up data after EST to detect risk factors for bile duct stone recurrence, for several reasons. First, follow-up evaluation often is incomplete because most patients become free of symptoms and leave medical institutions, with the result that many are lost to follow-up evaluation [1, 11]. Second, a large number of patients must be followed for a long period for appropriate statistical analysis to be performed [2, 4]. In the current study, 232 patients were available for individual follow-up evaluation, and the mean follow-up period of 36 ± 17 months (median, 34 months; range, 12 86 months) provided an appropriate background for the analysis. Loss of medical or endoscopic information occurred only for 47 patients. Therefore, we believe that this should not introduce a major bias or invalidate the findings. Moreover, we have observed an overall recurrence rate of 16% (36 of 232 patients), which is within the range (4 24%) reported in the literature [6, 11, 12, 14]. Stone recurrence is a well-recognized and potentially serious problem limiting the overall success of endoscopic therapy for bile duct stones. It is important that an effort be made to identify patients at high risk for stone recurrence because this would permit close follow-up evaluation of these patients to prevent serious complications of bile duct stones, such as cholangitis [5, 8]. It has been suggested that these patients might benefit from prophylactic surveillance ERCP [5]. At ERCP, measurement of bile duct angulation is simple and not associated with any additional procedure or risk. A combination of risk factors (duct angulation + dilation + previous open cholecystectomy) can increase the predictive value even further. To the best of our knowledge, the current study is the first to describe an association between angulation of the CBD and stone recurrence after endoscopic treatment. In a literature search, we found a single study that refers to an association between bile duct angulation and stone formation. In this study of 126 intraoperative cholangiograms, Warren [20] found an Fig. 2. Patients free of recurrence during follow-up evaluation according to bile duct angulation. Fig. 3. Patients free of recurrence during follow-up evaluation according to bile duct dilation. Fig. 4. Patients free of recurrence during follow-up evaluation, as determined by coexistence of all three risk factors. association between choledocholithiasis and a more acute cholangiographic angulation of the CBD at cholecystectomy. The mean angulation of the duct was 135.7 in patients with cholecystolithiasis only, as compared with 103.4 in patients with concurrent choledocholithiasis. The difference between the mean duct angulations in the Warren study and the current

1598 study could be attributed to technical differences. A retracted open abdomen at surgery could have exaggerated the duct angulation. Angulation along the course of the CBD may predispose to bile stasis and thus promote stone formation and recurrence. Indeed, bile stasis is thought to be an important factor in the pathogenesis of bile duct stones [2, 3, 5, 9, 13]. Duct dilation, the other risk factor observed, also may promote stasis, and thereby support the notion of its pathogenetic importance. Assessment of bile duct angulation in the current study was on a two-dimensional plane. This actually may have led to underestimation of CBD angulation as measured on a more accurate three-dimensional plane. This could be corrected in future prospective studies for a further increase in the accuracy of bile duct angle assessment. Although different medical conditions, such as the need for intraoperative CBD exploration, may have influenced the decision to perform open cholecystectomy in our patients, this operation per se (irrespective of the reason) was found to be an independent risk factor for stone recurrence. It is not clear from our available data what caused this increased recurrence rate in patients who underwent surgery by the open method. Because CBD angulation and previous open cholecystectomy both were found to be independent risk factors, and because the mean CBD angles were equal in patients after open and laparoscopic cholecystectomies, the effect is not by way of bile duct angulation. It is conceivable that biliary manipulation could play a role in biliary stasis by introducing local adhesions. This also could explain why laparoscopic cholecystectomy, involving less trauma, is not associated with stone recurrence. In a literature search, we found a single study that refers to a relationship between previous laparotomy in the biliary region and stone formation. In this study of 522 patients, Lygidakis [10] an association between previous intervention on the biliary tree (choledochotomy) and primary CBD stones. To the best of our knowledge, the current study is the first to describe an association between previous laparotomy in the biliary region and CBD stone recurrence after endoscopic treatment. Because this report deals only with symptomatic (clinically relevant) recurrence of bile duct stones, the end point of recurrence was diagnosed on clinical grounds, backed by laboratory, ultrasonographic, and/ or endoscopic (second ERCP) verification. There may have been false-negatives (i.e., underestimation) in terms of recurrence because stone recurrence may be asymptomatic. However, we do not necessarily consider this a limitation, but an acceptable consequence of the study design. Because only true recurrence as evident by ERCP was considered a recurrence in this study, there were no false-positives. In conclusion, our findings suggest that specific risk factors, particularly an angulated and/or dilated CBD and/or a previous open cholecystectomy, can identify high-risk patients that may benefit from intensive follow-up evaluation (by clinical follow-up assessment, MRCP, endoscopic ultrasound, or ERCP) after EST for bile duct stones. Because of the current studyõs retrospective nature, future prospective studies are warranted to confirm the clinical significance of these two newly described risk factors for bile duct stone recurrence. Acknowledgments. The authors thank Drs. Y. Rattan, A. Hallak, and M. Santo for performing some of the endoscopic procedures. References 1. Ando T, Tsuyuguchi T, Okugawa T, Saito M, Ishihara T, Yamaguchi T, Saisho H (2003) Risk factors for recurrent bile duct stones after endoscopic sphincterotomy. Gut 52: 116 121 2. Bergman JJ, van der Mey S, Rauws EA, Tijssen JG, Gouma DJ, Tytgat GN, Huibregtse K (1996) Long-term follow-up after endoscopic sphincterotomy for bile duct stones in patients younger than 60 years of age. Gastrointest Endosc 44: 643 649 3. Costamagna G, Tringali A, Shah SK, Mutignani M, Zucclala G, Perri V (2002) Long-term follow-up of patients after endoscopic sphincterotomy for choledocholithiasis, and risk factors for recurrence. Endoscopy 34: 273 279 4. Donner A (1984) Approaches to sample size estimation in the design of clinical trials: a review. Stat Med 3: 199 214 5. Geenen DJ, Geenen JE, Jafri FM, Hogan WJ, Catalano MF, Johnson GK, Schmalz MJ (1998) The role surveillance endoscopic retrograde cholangiopancreatography in preventing episodic cholangitis in patients with recurrent common bile duct stones. Endoscopy 30: 18 20 6. Kim DI, Kim MH, Lee SK, Seo DW, Choi WB, Lee SS, Park HJ, Joo YH, Yoo KS, Kim HJ, Min YI (2001) Risk factors for recurrence of primary bile duct stones after endoscopic biliary sphincterotomy. Gastrointest Endosc 54: 42 48 7. Lai KH, Lin LF, Lo GH, Cheng JS, Huang RL, Lin CK, Huang JS, Hsu PI, Peng NJ, Ger LP (1999) Does cholecystectomy after endoscopic sphincterotomy prevent the recurrence of biliary complications? Gastrointest Endosc 49: 483 487 8. Lai KH, Lo GH, Lin CK, Hsu PI, Chan HH, Cheng JS, Wang EM (2002) Do patients with recurrent choledocholithiasis after endoscopic sphincterotomy benefit from regular follow-up? Gastrointest Endosc 55: 523 526 9. Lai KH, Peng NJ, Lo GH, Cheng JS, Huang RL, Lin CK, Huang JS, Chiang HT, Ger LP (1997) Prediction of recurrent choledocholithiasis by quantitative cholescintigraphy in patients after endoscopic sphincterotomy. Gut 41: 399 403 10. Lygidakis NJ (1983) Incidence and significance of primary stones of the common bile duct in choledocholithiasis. Surg Gynecol Obstet 157: 434 436 11. Pereira-Lima JC, Jakobs R, Winter UH, Benz C, Martin WR, Adamek HE, Riemann JF (1998) Long-term results (7 to 10 years) of endoscopic sphincterotomy for choledocholithiasis. Multivariate analysis of prognostic factors for the recurrence of biliary symptoms. Gastrointest Endosc 48: 457 464 12. Rienmann JF, Lux G, Forster P, Altendorf A (1983) Long-term results after endoscopic sphincterotomy. Endoscopy 15: 165 168 13. Saito M, Tsuyuguchi T, Yamaguchi T, Ishihara T, Saisho H (2000) Long-term outcome of endoscopic sphincterotomy for choledocholithiasis with cholecystolithiasis. Gastrointest Endosc 51: 540 545 14. Seifert E (1988) Long-term follow-up after endoscopic sphincterotomy. Endoscopy 20: 232 235 15. Sugiyama M, Atomi Y (2002) Risk factors predictive of late complications after endoscopic sphincterotomy for bile duct stones: long-term (more than 10 years) follow-up study. Am J Gastroenterol 97: 2763 2767 16. Sugiyama M, Suzuki Y, Abe N, Masaki T, Mori T, Atomi Y (2004) Endoscopic treatment of recurrent choledocholithiasis after sphincterotomy. Gut 53: 1856 1859 17. Tanaka M, Takahata S, Konomi H, Matsunaga H, Yokohata K, Takeda T, Utsunomiya N, Ikeda S (1998) Long-term consequence of endoscopic sphincterotomy for bile duct stones. Gastrointest Endosc 48: 465 469 18. Uchiyama K, Onishi H, Tani M, Kinoshita H, Kawai M, Ueno M, Yamaue H (2003) Long-term prognosis after treatment of patients with choledocholithiasis. Ann Surg 238: 97 102

1599 19. Ueno N, Ozawa Y, Aizawa T (2003) Prognostic factors for recurrence of bile duct stones after endoscopic treatment by Sphincter dilation. Gastrointest Endosc 58: 336 340 20. Warren BL (1987) Association between cholangiographic angulation of the common bile duct and choledocholithiasis. South African J Surg 25: 13 15