GASTROENTEROLOGY 2013;144:341 345 CLINICAL BILIARY Randomized Trial of Endoscopic Sphincterotomy With Balloon Dilation Versus Endoscopic Sphincterotomy Alone for Removal of Bile Duct Stones ANTHONY YUEN BUN TEOH, 1 FRANCES KA YIN CHEUNG, 2 BING HU, 3 YA MIN PAN, 3 LARRY HIN LAI, 4 PHILIP WAI YAN CHIU, 1,4 SIMON KIN HUNG WONG, 1 FRANCIS KA LEUNG CHAN, 4 and JAMES YUN WONG LAU 1,4 1 Department of Surgery, Prince of Wales Hospital, and 4 Institute of Digestive Diseases, Faculty of Medicine, Chinese University of Hong Kong, Hong Kong SAR; 2 Department of Surgery, Pamela Younge Nethersole East Hospital, Hong Kong SAR; and 3 Department of Endoscopy, Eastern Hepatobiliary Hospital, Second Military Medical University, Shanghai, China This article has an accompanying continuing medical education activity on page e14. Learning Objective: Upon completion of this CME activity, successful learners will be able to identify the indications and advantages of combined limited sphincterotomy with balloon dilation (ESBD) for stone retrieval during endoscopic retrograde cholangiopancreatography. BACKGROUND & AIMS: Limited endoscopic sphincterotomy with large balloon dilation (ESBD) is an alternative to endoscopic sphincterotomy (ES) for removing bile duct stones, but it is not clear which procedure is most effective. We compared the 2 techniques in removal of bile duct stones. METHODS: Between September 2005 and September 2011, 156 consecutive patients with suspected of having, or known to have, common bile duct stones were randomly assigned to groups that underwent ES or ESBD. Patients in the ESBD group underwent limited sphincterotomy (up to half of the sphincter) followed by balloon dilation to the size of the common bile duct or 15 mm, and patients in the ES group underwent complete sphincterotomy alone. Stones were then removed using standard techniques. The primary outcome was percentage of stones cleared, and secondary outcomes included procedural time, method of stone extraction, number of procedures required for stone clearance, morbidities and mortality within 30 days, and direct cost. RESULTS: There was no significant difference between groups in percentage of stones cleared (ES vs ESBD: 88.5% vs 89.0%). More patients in the ES group (46.2%) than the ESBD group (28.8%) required mechanical lithotripsy (P.028), particularly for stones 15 mm (90.9% vs 58.1%; P.002). Morbidities developed in 10.3% of patients in the ES group and 6.8% of patients in the ESBD group (P.46). The cost of the hospitalization was also significantly lower in the ESBD group (P.034). CONCLUSIONS: ESBD and ES clear bile stones with equal efficacy. However, ESBD reduces the need for mechanical lithotripsy and is less expensive; ClinicalTrials.gov number, NCT00164853. Keywords: Cost Efficacy; Bile Duct; Surgery; Treatment. During endoscopic retrograde cholangiopancreatography (ERCP), endoscopic sphincterotomy (ES) is the standard method of enlarging the papillary orifice before stone retrieval. However, in patients with large stones, barrel-shaped stones, and tapered distal common bile ducts (CBDs), the use of other methods including mechanical lithotripsy (ML) may be required. 1 Ersoz et al first reported the use of large balloon dilation after sphincterotomy for large CBD stones and achieved a high stone clearance rate of up to 89% to 95% without the use of ML. 1 Since then, a number of case series have also suggested that endoscopic sphincterotomy with balloon dilation (ESBD) might be useful in reducing the need for ML during retrieval of large bile duct stones. 2 7 However, of the 3 randomized trials comparing ESBD with ES, only one study showed that ESBD reduced the need for ML in retrieval of large CBD stones. 8 10 Furthermore, design inadequacies of these trials also limited their capacity in completely addressing the clinical question and the benefits of ESBD remained controversial. Thus, the aim of the current study was to evaluate the benefits and risks of ESBD versus ES in removal of common bile duct stones in a randomized setting. Patients and Methods This was a prospective, randomized, comparative study conducted in the Prince of Wales Hospital and the Eastern Hepatobiliary Hospital between September 2005 and September 2011. We hypothesized that ESBD facilitates stone retrieval without added morbidities when compared with ES. The study protocol was approved by the Joint Chinese University of Hong Kong - New Territories East Cluster Clinical Research Ethics Committee. Informed consent was obtained from all patients, and all authors had access to the study data and reviewed and approved the final manuscript. The study was registered with ClinicalTrials.gov (NCT00164853). Abbreviations used in this paper: CBD, common bile duct; ERCP, endoscopic retrograde cholangiopancreatography; ES, endoscopic sphincterotomy; ESBD, limited endoscopic sphincterotomy with large balloon dilation; ML, mechanical lithotripsy. 2013 by the AGA Institute 0016-5085/$36.00 http://dx.doi.org/10.1053/j.gastro.2012.10.027
342 TEOH ET AL GASTROENTEROLOGY Vol. 144, No. 2 Patients Consecutive patients aged 18 years or older who were scheduled for ERCP for known or suspected CBD stones were invited to participate. Randomization was performed after bile duct access was gained, and a cholangiogram confirmed the presence of CBD stones in a dilated CBD of 13 mm in maximum diameter. Patients with stents or nasobiliary drains in situ without sphincterotomy and those with previous small sphincterotomy requiring further sphincter ablation were also included. Patients with the following conditions were excluded: septic shock, coagulopathy (international normalized ratio 1.3, partial thromboplastin time greater than twice that of control), platelet count 50,000 10 3 / L, suspected or confirmed malignancy, and requirement of precut sphincterotomy for bile duct access. Interventions ERCP was performed using a side-viewing duodenoscope (TJF-240; Olympus Co Ltd, Tokyo, Japan). An electrosurgical unit (UES-30; Olympus Co Ltd) with a blended current and a power setting of 40 W was used. Six consultant specialists performed all of the procedures. All patients were sedated with 0.1 mg/kg intravenous Diazepam emulsion (Diazemuls, Pfizer, New York, NY) and 0.5 mg/kg pethidine (meperidine), and antibiotic prophylaxis was given before the procedure. Patients were randomized to the ES or ESBD groups after CBD access was gained and a cholangiogram confirmed the presence of stones in a dilated CBD 1.3 cm in diameter. After the procedures, the stones were retrieved with a dormia basket (Tetracatch; Olympus Co Ltd) or extraction balloon (Multi-3 extraction balloon; Olympus Co Ltd). Mechanical lithotripsy (BML-110A-1; Olympus Co Ltd) was used to avoid stone impaction or when the stones were deemed too large to be removed through the papilla with standard methods. An occlusion cholangiogram was obtained at the end of the procedure and plastic stents or nasobiliary drains were inserted for drainage if stones were not completely removed. ES After deep cannulation was achieved, a complete sphincterotomy was performed with a 25-mm pull-type sphincterotome (Clever Cut 3; KD-V411M, Olympus, Tokyo, Japan) and the sphincter was divided up to the transverse duodenal fold. A complete sphincterotomy was defined by the free passage of a fully bowed sphincterotome and the presence of spontaneous bile drainage. In patients with prior sphincterotomies that were deemed inadequate in size, the sphincterotomy would be extended by a pull-type sphincterotome to the junction between the papillary roof and the duodenal wall. ESBD A limited sphincterotomy measuring up to one-third to one-half of the size of the papilla was first performed. This was followed by dilation of the sphincter with a 5.5-cm-long controlled radial expansion balloon (Boston Scientific, Natick, MA) over guidewire. The balloon was centered at the sphincter and gradually inflated to the size of the bile duct and held for 30 seconds until waisting was abolished. The size of the balloon used was gauged by the size of the CBD at the level of the papilla, and the maximum diameter of the balloon used was 15 mm. In patients with prior sphincterotomy, balloon dilation was the only procedure. Post-ERCP Management After the procedure, the patients were kept in the ward to monitor for any complications. Serum amylase level was checked if the patient reported abdominal pain or there was clinical suspicion of post-ercp pancreatitis. No gabexate, nonsteroidal anti-inflammatory drugs, somatostatin, or octreotide was prescribed. The patients were discharged when the acute condition was settled, and they were interviewed by telephone 30 days after the procedure for assessment of complications. Both the patient and assessors were blinded to the treatment group. Outcome Measurements The primary outcome was the stone clearance rate at the index ERCP. Secondary outcomes included procedural time, method of stone extraction, number of ERCPs required for complete stone clearance, 30-day morbidities and mortalities, and direct cost. Complete stone clearance was defined as the absence of filling defects on occlusion cholangiogram as noted by the endoscopist. The procedural time was the time between randomization and occlusion cholangiogram. Morbidities were defined and graded according to the modified 1991 consensus guidelines by Cotton et al. 11 The direct cost included the total cost for the entire admission, which comprised costs of hospital stay, performed procedures, consumables, and management of major complications. Sample Size Calculation and Statistical Analyses Assuming a 15% difference in the stone clearance rate at index endoscopy, with a type I error of 0.05 (2 sided) and a power of 0.8, a minimum of 152 patients were required (76 in each group). Randomization codes were generated by the computer on a 1:1 basis in blocks of 30 and placed in a concealed envelope. Stratification of randomization was performed according to the participating consultant endoscopist. The differences between groups were compared using Student t test for parametric data, Mann Whitney U test for nonparametric data, and Fisher exact test or 2 test for comparison in the differences in proportions. Predictors of ML were analyzed by multivariate logistic regression using the stepwise method. A 2-sided P value of.05 was considered significant. Statistical analyses were performed using SPSS 20 statistical software (SPSS Inc, Chicago, IL). Results During the study period, 156 patients were recruited to the study (128 at Prince of Wales Hospital and 28 at Eastern Hepatobiliary Hospital). Four patients with precut sphincterotomy and one patient with suspected pancreatic cancer were recruited by mistake and excluded from the analysis (Supplementary Figure 1). There were no differences in the background demographics and indications for ERCP between the 2 groups (Table 1). The procedural details of the 2 groups of patients are shown in Table 2. There were no differences in cannulation time, time to complete the procedure, mean CBD size, number of stones, and maximum size of stones. There were also no significant differences in the stone clearance rates between the 2 groups (Table 3). A total of 88.5% of patients in the ES group and 89% of patients in the ESBD group had stones cleared at index endoscopy
February 2013 ESBD VERSUS ES IN REMOVING BILE DUCT STONES 343 Table 1. Comparison of Background Demographics Between the 2 Groups of Patients ES (n 78) ESBD (n 73) Age (y), mean (SD) 72.97 (13.42) 71.62 (14.80) Sex (male/female), n 40/38 32/41 American Society of 15/44/19 18/34/21 Anesthesiologists class I/II/III, n Prior ERCP, n 24 28 Prior sphincterotomy, n 16 23 Without prior 8 5 sphincterotomy, n International normalized 1.04 (0.11) 1.05 (0.12) ratio, mean (SD) Indications for ERCP, n Acute cholangitis 43 29 Acute pancreatitis 1 1 Obstructive jaundice 17 16 Deranged liver enzymes 3 9 Radiologic abnormalities a 13 16 Others 1 2 a Radiologic abnormalities included findings of CBD stones or dilated CBD on ultrasonography, magnetic resonance cholangiopancreatography, or computed tomography. (P.279). At second ERCP, 100% and 97.2% of patients had stones cleared, respectively. For stones that were 15 mm, similar rates of stone clearance were also observed (P.570). Regarding the methods of stone extraction, significantly more patients in the ES group required the use of ML (46.2% vs 28.8%, P.028); this was particularly observed for stones that were 15 mm (90.9% vs 58.1%, P.002). Use of the dormia baskets and extraction balloons was similar between the 2 groups (P.850 and P.998, respectively). Morbidities occurred in 10.3% of patients in the ES group and 6.8% of the patients in the ESBD group (P.456). The majority of cases were graded as mild to moderate in severity. None of the patients died after procedures. The details are shown in Table 4. The rates of post-ercp bleeding and the presence of endoscopic Table 3. Comparison of Methods of Stone Retrieval and Clearance Rates ES (n 78) ESBD (n 73) P value Stone clearance rates, n(%) At index ERCP 69 (88.5) 65 (89).279 At second ERCP 9 (11.5) 6 (8.2) Stones 15 mm at 28/33 (84.8) 26/31 (83.9).570 index ERCP Methods of stone extraction, n (%) ML 36 (46.2) 21 (28.8).028 a Stones 15 mm 6/45 (13.3) 3/42 (7.1).344 Stones 15 mm 30/33 (90.9) 18/31 (58.1).002 a Dormia basket 62 (79.5) 49 (67.1).850 Extraction balloon 47 (60.3) 44 (60.3).998 NOTE. Percentages shown are percentages within the intervention group. A P value.05 indicates significance. bleeding that did not satisfy the consensus criteria were also recorded and were not significantly different (P.483 and P.612, respectively). Endoscopic bleeding was controlled by adrenaline spray in 13 patients, adrenaline injection in 3 patients, and balloon tamponade in one patient. Two patients in the ES group had suspected guidewire perforation. Amylase levels were checked in 44 patients, and 5 patients had post-ercp pancreatitis. In the ESBD group, 5 patients who underwent ES in the same session as balloon dilation experienced complications, whereas none of the patients who underwent prior sphincterotomy experienced morbidities (P.065). The direct cost of the procedures was also compared between the 2 groups. A significant reduction in overall cost was noted in the ESBD group (USD $6005 [interquartile range, $4462 $5441] vs $5025 [interquartile range, $4140 $5235]; P.034). The predictors of the need for ML during stone retrieval were further analyzed by multivariate logistic regression. Both the maximum stone size and undergoing Table 2. Comparison of Procedural Details Between the 2 Groups ES (n 78) ESBD (n 73) P value Deep cannulation 7.12 (8.9) 7.28 (7.22).665 time (min), mean (SD) Total procedural 27.20 (16.90) 24.30 (12.87).509 time (min), mean (SD) CBD size (mm), 15 (13 50) 15 (13 30).555 median (range) No. of stones (%).211 1 35 (44.9) 29 (39.7) 2 8 (10.2) 12 (16.4) 3 10 (12.8) 3 (4.1) 3 25 (32.1) 29 (39.7) Maximum size (mm) of stones (range) 13.26 (5 40) 12.47 (5 35).994 Table 4. Comparison of Morbidities and Mortalities Between the 2 Groups ES (n 78) ESBD (n 73) P value Morbidities 8 (10.3) 5 (6.8).456 Mild 2 (2.6) 3 (4.1) Moderate 3 (3.8) 2 (2.7) Severe 3 (3.8) 0 (0) Bleeding (%) 0 (0) 1 (1.4).483 Endoscopic bleeding a 10 (12.8) 7 (9.6).612 Perforation (%) 2 (2.6) 0 (0).497 Pancreatitis (%) 3 (3.8) 2 (2.7) 1.000 Cholangitis (%) 2 (1.3) 1 (1.4) 1.000 Cholecystitis (%) 1 (1.3) 1 (1.4) 1.000 Mortality 0 0 1.000 NOTE. All values are expressed as number (percentage) unless otherwise noted. Percentages shown are percentages within the intervention group. a These were patients with presence of endoscopic bleeding but no clinical evidence of bleeding.
344 TEOH ET AL GASTROENTEROLOGY Vol. 144, No. 2 ES were independent predictors of the need for ML (P.001 and P.014, respectively). Discussion This study showed that performance of ESBD or ES for retrieval of bile duct stones resulted in high stone clearance rates in patients with a CBD 13 mm in diameter. The morbidity rates in both arms were low and comparable, and most patients who experienced morbidities had cases of mild to moderate severity. In addition, the use of ESBD resulted in a decreased need for use of ML for retrieval of CBD stones, particular for stones that were 15 mm in diameter. The treatment group and the stone size were also identified as independent predictors for the use of ML. Furthermore, the cost of hospitalization was significantly less in the ESBD group. The current study was unique in several aspects. Firstly, we only included patients with a minimum CBD diameter of 13 mm. This was designed to ensure that recruited patients would receive a minimum balloon dilation of the papilla of up to 13 mm (gauged according to the size of the CBD) for retrieval of larger CBD stones. The patients in this group are most likely to benefit from this technique. Secondly, the majority of the patients included in this study had at least 2 CBD stones and almost one-half had stones that were more than 15 mm. On multivariate analysis, undergoing ES and stone size were identified as independent predictors of use of ML. This indicated that ESBD allowed retrieval of multiple large stones without the use of ML through a larger papillary orifice. Ersoz et al first described the use of ESBD as an alternative to ES alone. 1 The technique was used when stones were too large to be retrieved by basket or balloon extraction after ES. This combination technique resulted in an overall success rate of 93% without the need for ML and a complication rate of 15%. Thereafter, 3 comparative studies evaluated the techniques in a randomized setting, but the exact benefits of ESBD remained controversial. In the first study, 200 patients with bile duct stones were randomized to receive ESBD or ES. 8 There were no significant differences in the stone clearance rates, complication rates, and use of ML. It was thus concluded that ESBD was an effective alternative to ES. The study, however, recruited stones of all sizes and the rates of ML use were low (8% and 9%). The mean stone size was not mentioned. Among those recruited, 14 patients in the ESBD group and 8 patients in the ES group required more than one session of ERCP for complete stone clearance due to large and/or multiple stones. Thus, the study was not powered to detect differences in retrieval of large bile duct stones with ESBD. In another study, Stefanidis et al directly compared the retrieval of large bile duct stones (12 to 20 mm in size) by either ESBD or ES and ML in 90 patients. 9 The overall stone clearance rates were high and comparable between both groups (ESBD, 97.7%; ES, 91.1%). However, the study was stopped before reaching full enrollment because the interim analysis showed a significant difference in post- ERCP cholangitis favoring ESBD. Most of the patients in the ES and ML group had plastic stents inserted after the procedure. The reasons for the increased rates of cholangitis might be related to an inadequate sphincterotomy or a blocked stent. The design of the study also required all patients in the ES and ML group to undergo the procedure; hence, it was uncertain whether some of these patients could have stones removed without the use of ML. In the third study, patients with CBD stones that were 15 mm were randomized to undergo ESBD or ES. 10 However, the study only recruited a small number of patients and failed to show any significant differences between the procedures. Two other large case series also reported the outcomes of ESBD. 2,12 Itoi et al described that the rates of ML use were significantly less and the total procedure time and fluoroscopy time were also shorter with the use of ESBD. 2 In another multicenter series, the procedural success and rates of complications were similar to those for ES alone. 12 There are a few drawbacks to the current study. Firstly, the rates of ML use appeared to be higher than in previous reports. 1 6,8,10,12 On one hand, this may be attributed to the size of the stones in recruited patients. On the other hand, it may also indicate that endoscopists were more cautious about stone impaction during retrieval of large CBD stones and had more liberal use of ML. The size of the sphincterotomy is likely to be adequate because it was standardized according to the protocol. Nevertheless, significantly less ML use was still observed in the ESBD arm. Secondly, although only patients with a minimum CBD diameter of 13 mm were included, the presence of a tapered CBD is another factor that may also influence the difficulty of retrieving stones despite adequate balloon dilation of the sphincter. 3 The degree of tapering was not assessed in this study, because there is no universally accepted method of measuring this parameter and the measurements may be subjected to significant biases. Also, because the size of the CBD was chosen as the inclusion criteria, some patients with small stones that could be removed by ES alone may have been included in the ESBD group. Furthermore, the maximum size of balloon dilation selected for this study was 15 mm. The decision was based on the initial report of a high complication rate of 33% in patients with tapered bile ducts when dilating up to 20 mm. Such rates were not reported in later series. 8,9 In addition, the endoscopist performing the procedure interpreted the final occlusion cholangiogram after completion of the ERCP, and this may be a potential source of bias. Finally, although increased morbidities were observed in patients undergoing ESBD in the same session as compared with balloon dilation alone, there is a potential for beta error because only a small number of patients had antecedent ERCP. In conclusion, similar stone clearance rates were achieved with ESBD and ES. However, ESBD decreased the need for ML and was less expensive.
February 2013 ESBD VERSUS ES IN REMOVING BILE DUCT STONES 345 Supplementary Materials Note: To access the supplementary material accompanying this article, visit the online version of Gastroenterology at www.gastrojournal.org, and at http:// dx.doi.org/10.1053/j.gastro.2012.10.027. References 1. Ersoz G, Tekesin O, Ozutemiz AO, et al. Biliary sphincterotomy plus dilation with a large balloon for bile duct stones that are difficult to extract. Gastrointest Endosc 2003;57:156 159. 2. Itoi T, Itokawa F, Sofuni A, et al. Endoscopic sphincterotomy combined with large balloon dilation can reduce the procedure time and fluoroscopy time for removal of large bile duct stones. Am J Gastroenterol 2009;104:560 565. 3. Misra SP, Dwivedi M. Large-diameter balloon dilation after endoscopic sphincterotomy for removal of difficult bile duct stones. Endoscopy 2008;40:209 213. 4. Maydeo A, Bhandari S. Balloon sphincteroplasty for removing difficult bile duct stones. Endoscopy 2007;39:958 961. 5. Minami A, Hirose S, Nomoto T, et al. Small sphincterotomy combined with papillary dilation with large balloon permits retrieval of large stones without mechanical lithotripsy. World J Gastroenterol 2007;13:2179 2182. 6. Espinel J, Pinedo E, Olcoz JL. Large hydrostatic balloon for choledocolithiasis [in Spanish]. Rev Esp Enferm Dig 2007;99:33 38. 7. Kim TH, Oh HJ, Lee JY, et al. Can a small endoscopic sphincterotomy plus a large-balloon dilation reduce the use of mechanical lithotripsy in patients with large bile duct stones? Surg Endosc 2011;25:3330 3337. 8. Heo JH, Kang DH, Jung HJ, et al. Endoscopic sphincterotomy plus large-balloon dilation versus endoscopic sphincterotomy for removal of bile-duct stones. Gastrointest Endosc 2007;66:720 726; quiz 768, 771. 9. Stefanidis G, Viazis N, Pleskow D, et al. Large balloon dilation vs. mechanical lithotripsy for the management of large bile duct stones: a prospective randomized study. Am J Gastroenterol 2011;106:278 285. 10. Kim HG, Cheon YK, Cho YD, et al. Small sphincterotomy combined with endoscopic papillary large balloon dilation versus sphincterotomy. World J Gastroenterol 2009;15:4298 4304. 11. Cotton PB, Lehman G, Vennes J, et al. Endoscopic sphincterotomy complications and their management: an attempt at consensus. Gastrointest Endosc 1991;37:383 393. 12. Attasaranya S, Cheon YK, Vittal H, et al. Large-diameter biliary orifice balloon dilation to aid in endoscopic bile duct stone removal: a multicenter series. Gastrointest Endosc 2008;67: 1046 1052. Received April 15, 2012. Accepted October 5, 2012. Reprint requests Address requests for reprints to: Anthony Yuen Bun Teoh, FRCSEd (Gen), Department of Surgery, Prince of Wales Hospital, Chinese University of Hong Kong, Hong Kong SAR, China. e-mail: anthonyteoh@ surgery.cuhk.edu.hk; fax: (852) 2637-7974. Conflicts of interest The authors disclose no conflicts.
345.e1 TEOH ET AL GASTROENTEROLOGY Vol. 144, No. 2 Supplementary Figure 1. The CONSORT flow chart. Ca, cancer; Sphx, sphincterotomy; Misc, miscellaneous; IHD, intrahepatic duct.