- Email: [email protected]
Assessment of residual bile duct stones with use of intraductal US during endoscopic balloon sphincteroplasty: comparison with balloon cholangiography
Assessment of residual bile duct stones with use of intraductal US during endoscopic balloon sphincteroplasty: comparison with balloon cholangiography Akira Ohashi, MD, Norio Ueno, MD, Kiichi Tamada, MD, Takeshi Tomiyama, MD, Shinichi Wada, MD, Takamitsu Miyata, MD, Takashi Nishizono, MD, Shigeo Tano, MD, Toshiyuki Aizawa, MD, Kenichi Ido, MD, Ken Kimura, MD Tochigi, Japan
Background: We sought to determine the utility of intraductal ultrasonography (IDUS) in detecting residual bile duct stones during endoscopic balloon sphincteroplasty. Methods: Eighty-one consecutive patients with bile duct stones who underwent IDUS during endoscopic balloon sphincteroplasty were studied. IDUS was performed with a thin-caliber ultrasonic probe (diameter 2.0 mm, frequency 20 MHz) by a transpapillary route after stone extraction. When IDUS or balloon-retrograde cholangiography suggested residual stones, the bile duct was cleared again with a Dormia basket. Extraction of the stones was confirmed by direct duodenoscopic visualization. Videotapes of IDUS and cholangiograms were reviewed retrospectively without knowledge of the results of other diagnostic modalities. Results: In 27 of 81 patients (33%), IDUS detected small residual stones not seen on cholangiography. When stones were fragmented with mechanical lithotripsy, the accuracy of IDUS in detecting small residual stones was significantly greater than that of balloon-endoscopic retrograde cholangiography (95% vs 50%, p < 0.001). When the bile duct was greater than 10 mm in diameter, the accuracy of IDUS in detecting small residual stones was significantly greater than that of cholangiography (92% vs. 56%, p < 0.001). Conclusions: IDUS is useful for detecting small residual bile duct stones during endoscopic balloon sphincteroplasty when stones are fragmented by mechanical lithotripsy or when there is evidence of a dilated bile duct (>10 mm). (Gastrointest Endosc 1999;49:328-33) Recently endoscopic balloon dilation (sphincteroplasty) without endoscopic sphincterotomy (ES) has been reported as a safe, minimally invasive, functionpreserving method for treating bile duct stones.1-5 However, balloon dilation provides only a transient sphincteroplasty effect.6,7 Therefore accurate assessment for residual stones at the time of endoscopic balloon sphincteroplasty (EBS) is of paramount importance. One approach to assessment is intra-bile
Patients
Received December 16, 1996. For revision March 30, 1997. Accepted September 12, 1998. From the Department of Gastroenterology, Jichi Medical School, Tochigi, Japan. Reprint requests: Akira Ohashi, MD, Department of Gastroenterology, Jichi Medical School, Yakushiji, Tochigi 329-0498, Japan. Copyright © 1999 by the American Society for Gastrointestinal Endoscopy. 0016-5107/99/$8.00 + 0 37/1/94557
Eighty-one consecutive patients who underwent IDUS during EBS between July 1995 and June 1997 were included in the study. There were 43 men and 38 women. The mean age was 68.8 ± 14.0 years (range 28 to 94 years). Thirty-seven patients had gallbladder stones, and 23 of them underwent laparoscopic cholecystectomy before or after EBS. Twenty-two patients had a history of cholecystectomy. The remaining 22 patients had no gallbladder stones. The mean number of bile duct stones in individuals
328
GASTROINTESTINAL ENDOSCOPY
duct US with use of a thin-caliber ultrasonic probe, termed intraductal ultrasonography (IDUS), which has been shown to be useful in the diagnosis of bile duct diseases.8-21 We describe our preliminary experience with the detection of small residual bile duct stones with IDUS during EBS. PATIENTS AND METHODS
VOLUME 49, NO. 3, PART 1, 1999
Intraductal US for residual bile duct stones during balloon sphincteroplasty
3.0 ± 2.8 (range 1 to 17) with a mean size of 10.8 ± 5.0 mm range 4 to 23 mm) (Table 1). Written informed consent was obtained from each patient before EBS. Methods Diagnostic endoscopic procedure. Endoscopic retrograde cholangiography (ERC) was performed with use of a standard duodenoscope (model JF-200, Olympus Optical, Tokyo, Japan). Dilute contrast medium (30% amidotrizoic acid) was used for ERC. Premedication included diazepam (10 mg) and timepidium bromide (7.5 mg) administered intramuscularly. Endoscopic balloon sphincteroplasty. After the diagnostic ERC was performed, a 0.035-inch guidewire (Boston Scientific, Watertown, Mass) was passed through the diagnostic cannula into the bile duct. A Max Force 5F balloontipped biliary catheter (Boston Scientific) with a length of 180 cm and balloon length of 3 cm (maximum diameter 8 mm) was passed over the guidewire, positioning the deflated balloon across the papilla. The correct position of the balloon was confirmed by endoscopic visualization and by fluoroscopy. The balloon was then inflated with dilute radioopaque contrast medium to a pressure of 4 atmospheres for 60 seconds. After deflation for 30 seconds, the balloon was reinflated for 60 seconds to a pressure of 6 atmospheres. Subsequently, 2 to 3 inflations were performed with a pressure of 8 atmospheres separated by intervals of 30 seconds. Stone extraction. After the biliary catheter and guidewire were removed, stones were extracted by Dormia baskets, retrieval balloons, or both, according to standard guidelines.22 If necessary, stones (>8 mm in diameter) were fragmented by mechanical lithotripsy (MLT) (BML-4Q, Olympus Optical) before extraction. Intraductal ultrasonography. Stone removal was achieved by dragging a retrieval balloon (15 mm maximum balloon diameter, Boston Scientific) down the bile duct several times, after which balloon-ERC and IDUS were performed to detect residual stones. A cholangiogram was obtained with use of the retrieval balloon catheter. Two experienced examiners prospectively reviewed the fluoroscopic images. Then a thin ultrasonic probe (2.0 mm in diameter, 20 MHz, Aloka, Tokyo, Japan) was inserted into the bile duct through the accessory channel of the duodenoscope along the guidewire under fluoroscopic guidance. This high-frequency transducer provides an axial resolution of 0.1 mm and a maximum penetration of approximately 20 mm. IDUS images were recorded on videotape and individual instant photographs. Two experienced ultrasonographers prospectively reviewed the IDUS images in conjunction with the cholangiographic images. When IDUS indicated an echogenic focus with acoustic shadowing in the bile duct, the focus was considered to be a stone or fragment. After the procedure a nasobiliary drainage catheter (6F, 260 cm long, Cathex, Tokyo, Japan) was inserted into the bile duct for duct decompression and follow-up cholangiography. Study design When balloon-ERC or IDUS findings suggested residual stones, the procedure for cleaning the duct with a Dormia basket was repeated. Extraction of the stones was conVOLUME 49, NO. 3, PART 1, 1999
A Ohashi, N Ueno, K Tamada, et al
Table 1. Clinical details, stone characteristics, outcome, and complications in patients who underwent EBS Patients No. Mean age (y) (SD) Range (y) Bile duct stones Mean size (mm) (SD) Size range (mm) Mean No. (SD) Range (No.) Bile duct caliber Mean size (mm) (SD) Size range (mm) Efficacy Duct clearance EBS alone EBS + mechanical lithotripsy Frequency of endoscopic lithotripsy sessions 1-2 sessions 3 or more sessions Failed clearance in first admission Complications Pancreatitis Bleeding, perforation, or cholangitis
81 68.8 (14.0) 28-94 10.8 (5.0) 4-23 3.0 (2.8) 1-17 14.7 (5.0) 6-30 78/81 (96%) 37/81 (44%) 44/81 (56%) 68/81 (84%) 13/81 (16%) 3/81 (4%) 5/81 (6%) 0 (0%)
firmed by direct visualization with the duodenoscope. Bile duct clearance was repeated until balloon-ERC and IDUS showed no residual stones. For the blinded analysis of the two modalitics, cholangiograms obtained during balloon-ERC were reviewed retrospectively by two experienced radiologists without knowledge of the results of other diagnostic modalitics. Videotapes of IDUS were reviewed retrospectively by two experienced ultrasonographers without knowledge of the results of other diagnostic modalities. Agreement was obtained in 80 of 81 cases between the two radiologists and in all cases between the two ultrasonographers. Two radiologists discussed the case in which there was disagreement and made a final diagnosis. Statistical analysis All values are presented as the mean ± SD. Categoric results were analyzed with the chi-square test or Fisher’s exact test when the few cross-classifications were compared. Because multiple comparisons were made, only p values <0.01 were considered significant. In this manner, because fewer than 50 comparisons were made, findings can be interpreted as being significant at the 0.05 level. Multivariate effects were investigated with techniques of logistic regression.23
RESULTS Stone clearance The bile duct was cleared in 78 of 81 patients (96%) with EBS (Table 1). MLT was needed in 44 of 81 patients (54%). In 68 of 81 patients (84%), the bile duct was cleared in 1 or 2 sessions. The mean number GASTROINTESTINAL ENDOSCOPY
329
A Ohashi, N Ueno, K Tamada, et al
Intraductal US for residual bile duct stones during balloon sphincteroplasty
Figure 2. IDUS shows small residual stone (4 mm in diameter, arrow) at hepatic hilum. This was not detected by balloon-ERC.
The median follow-up was 12 months (range 1 to 24 months). To date, no adverse effects associated with EBS have been reported in any of the 81 patients except for stone recurrence in 5 patients. Comparison between IDUS and balloon-ERC
Figure 1. Cholangiographic findings after stone retrieval indicate no residual stones. IDUS catheter probe (arrow) is inserted into bile duct along guidewire.
of endoscopic sessions was 1.8 ± 1.0 (range 1 to 6). The mean duration of a treatment session was 69.0 ± 21.2 (range 32 to 140 minutes). IDUS scanning time was 3 to 5 minutes per patient. We were unable to clear the stones during the first admission in 3 patients who were treated successfully by a second EBS during a subsequent admission. Sphincterotomy was not performed during any treatment. Complications Transient abdominal discomfort was noted in half the patients during balloon inflation. No procedurerelated bleeding or perforation was encountered, but pancreatitis developed in 5 patients (6%), which resolved within 48 hours with administration of intravenous fluids and pain medication (Table 1). 330
GASTROINTESTINAL ENDOSCOPY
In 45 patients (56%) neither IDUS nor balloonERC showed residual stones (Table 2). However, in 27 patients (33%) IDUS demonstrated residual stones, although balloon-ERC failed to demonstrate them (Table 3, Figs. 1 and 2). In 1 patient (1%) IDUS failed to demonstrate an intrahepatic stone (migrating stone), which was detected by balloon-ERC. When IDUS or balloon-ERC showed residual stones, they were retrieved by a Dormia basket; retrieval was confirmed by direct duodenoscopic visualization. The sensitivity, specificity, accuracy, and negative predictive value of IDUS versus balloon-ERC in detecting small bile duct stones (fragments) were 85% (35/41) versus 22% (9/41), 100% (40/40) versus 100% (40/40), 93% (75/81) versus 60% (49/81), and 87% (40/46) versus 56% (40/72), respectively. The sensitivity, accuracy, and negative predictive value of IDUS were significantly greater than those of balloon-ERC (p < 0.001). In patients who underwent stone fragmentation with MLT the accuracy of IDUS in detecting small bile duct stones (fragments) was significantly greater than that of balloon-ERC (95% vs 50%, p < 0.001) (Table 4). When the bile duct caliber was greater than 10 mm, the accuracy of IDUS was also significantly greater than that of balloonERC (92% vs 56%, p < 0.001) (Table 5). However, logistic regression analysis of diameter of bile duct, number of stones, size of stones, and use of MLT on VOLUME 49, NO. 3, PART 1, 1999
Intraductal US for residual bile duct stones during balloon sphincteroplasty
Table 2. Detection of residual bile duct stones (fragments) after EBS with IDUS and balloon-ERC IDUS+ ERC+ Final diagnosis Stones present Stones absent Total
8 0 8
IDUS– ERC– 5 40 45
IDUS+ ERC– 27 0 27
IDUS– ERC+ 1 0 1
Agreement of IDUS and balloon-ERC on presence or absence of stones (fragments) 53/81 (65%); disagreement 28/81 (35%).
Table 3. Characteristics of residual stones (fragments) detected by IDUS, missed at balloon-ERC (n = 27) Mean stone size (mm) (SD, range) Mean stone number (SD, range) Mean bile duct caliber (mm) (SD, range) Use of MLT
3.6 1.3 15.8 20/27
(1.0, 2-6) (0.6, 1-3) (4.3, 8-25) (74%)
detecting residual stones by IDUS identified use of MLT as an almost significant (p = 0.011) predictor of residual stones (detected by IDUS, “missed” by balloon-ERC). Although dilated bile duct initially appeared to be a predictor of “missed” stones, this may be explained by the fact that patients with dilated bile ducts are more likely to have stones greater than 8 mm, for which MLT is necessary. DISCUSSION ES has been the standard method for removal of bile duct stones. However, ES has been shown to be associated with major complications such as bleeding, perforation, or pancreatitis in about 6% to 10% of patients.24-28 In addition to immediate complications, there are long-term risks of sphincterotomy in young patients referred for endoscopic stone removal before or after laparoscopic cholecystectomy. Recently, as an alternative approach to sphincterotomy, endoscopic balloon dilation or sphincteroplasty (EBS) has been reported1-5 and is possible because of advances in guidewire and biliary catheter technology. However, EBS provides only a transient sphincteroplasty effect.6,7 Therefore the assessment for residual stones after stone retrieval during EBS is a very important matter. Some investigators have conducted comparative studies of ERC and EUS and have reported that small bile duct stones (<4 mm in diameter) may be missed on conventional ERC.29-31 With ES small residual stones may be cleared spontaneously from the bile duct and have not been of great concern. In this study our research group used IDUS to detect small residual stones after stone removal during EBS. VOLUME 49, NO. 3, PART 1, 1999
Table 4. Accuracy of IDUS and balloon-ERC in detecting residual stones (fragments): relation to use of MLT
Total 41 40 81
A Ohashi, N Ueno, K Tamada, et al
MLT(–) IDUS Balloon-ERC
(89%)a
33/37 27/37 (73%)c
MLT(+) 42/44 (95%)b 22/44 (50%)d
a versus c, not significant (p = 0.068); b versus d, p < 0.001.
Table 5. Accuracy of IDUS and balloon-ERC in detecting residual stones (fragments) relation to bile duct caliber Nondilated bile duct IDUS Balloon-ERC
(93%)a
14/15 12/15 (80%)c
Dilated bile duct 61/66 (92%)b 37/66 (56%)d
a versus c, not significant ( p = 0.30); b versus d, p < 0.001. Nondilated bile duct ≤10 mm in diameter; dilated bile duct >10 mm in diameter.
Recently, many investigators have reported the usefulness of IDUS. 8-21 Assessing the extent of invasion of biliary tract cancer is one of the most promising aspects of this modality. The current study concentrated on another utility of IDUS in assessing the presence or absence of residual stones during EBS. Some investigators have already suggested the utility of IDUS in assessing small bile duct stones that were not adequately demonstrated by cholangiography. 16-18 However, there have been no reports addressing the utility of IDUS in visualizing residual stones during EBS. There are several advantages of IDUS compared with other diagnostic imaging modalities in visualizing bile duct stones: 1. IDUS can produce cross-sectional images of the bile duct in contrast to the luminal projection images of ERC, and IDUS images are not influenced by body habitus of the patient, bile duct caliber, or density of contrast material, in contrast to ERC. Our results indicated that balloon-ERC could not adequately recognize the presence of small stones or fragments when the bile duct diameter was greater than 10 mm. 2. Intra–bile duct scanning with use of a highfrequency ultrasonic probe produces high-resolution images not available with conventional extracorporeal US. 3. Once the IDUS probe is inserted into the bile duct, no special technique is required to demonstrate the whole extrahepatic bile duct, in contrast to EUS. GASTROINTESTINAL ENDOSCOPY
331
A Ohashi, N Ueno, K Tamada, et al
Intraductal US for residual bile duct stones during balloon sphincteroplasty
However, IDUS has some limitations: 1. IDUS cannot fully image the intrahepatic bile ducts, in contrast to ERC, because of the technical difficulty of probe insertion to each narrow branch. 2. Intra–bile duct insertion of a probe is essential for IDUS imaging. However, EUS is a high-quality, minimally invasive method to demonstrate small bile duct stones and does not require insertion of a probe into the bile duct or injection of contrast medium.29-31 In our study the accuracy of ERC was inferior to that of other studies,29-31 which can be explained by the fact that moderate to large stones had already been crushed and extracted at the time of balloon-ERC to assess for the presence of residual stones. And although sensitivity and specificity are the characteristics of the two procedures that are compared, their positive predictive value, negative predictive value, and accuracy are, in addition to their dependence on the sensitivity and specificity, also functions of the prevalence of the disease or condition in the setting in which the study takes place. With EBS stones were fragmented with use of MLT in 44 of 81 (56%) patients. Even in patients who did not require MLT, some stones may have been fragmented spontaneously when they were retrieved through the narrow bile duct opening. Our results indicate that, in EBS, conventional ERC (balloon-ERC) may fail to detect small residual stones (fragments), especially when stones are fragmented by MLT. This fact emphasizes the importance of accurate assessment for the presence of small fragments during EBS, because EBS provides only a transient sphincteroplasty effect. When IDUS detected residual stones, they were retrieved with use of a Dormia basket, as confirmed by direct visualization, although dragging the retrieval balloon before IDUS failed to clear them. This suggests a limitation of the retrieval balloon technique during EBS. There might be a concern that the sensitivity of IDUS cannot be established because further standard extraction techniques, such as the use of an extraction balloon, were not used when IDUS was negative and recurrent stones were found during follow-up in 5 patients. This could be mitigated by the following: 1. Cholangiograms made by a nasobiliary catheter placed after the final lithotripsy session showed no residual stones. 2. In 3 of 5 cases of recurrence (all 3 patients had prior cholecystectomy) multiple brown pigment 332
GASTROINTESTINAL ENDOSCOPY
stones developed (13, 14, and 16 mm in maximum stone diameter, respectively), which had not been demonstrated by nasobiliary cholangiograms after lithotripsy. 3. The remaining 2 patients with recurrent stones had multiple gallbladder stones and some of them entered the bile duct during follow-up. However, to establish the true sensitivity and negative predictive value of IDUS, performing a followup balloon cholangiogram and dragging the duct with a balloon 2 or 3 times in an effort to obtain confirmation that the duct had been cleared of stones are necessary. In the absence of confirmation of negative findings after IDUS, those 5 patients would be considered as having false-negative results of IDUS. In our study an air bubble in the bile duct was imaged as a strong echo with multiple scattering echos at IDUS; we could easily distinguish it from a residual stone (fragment). In conclusion, IDUS is useful to detect small residual bile duct stones (fragments) during EBS when stones are fragmented by MLT or when there is evidence of a dilated bile duct (>10 mm). However, intrahepatic migrating stones may be missed on IDUS and cholangiography remains necessary. REFERENCES 1. Staritz M, Ewe K, Meyer zum Buschenfelde KH. Endoscopic papillary dilation (EPD) for the treatment of common bile duct stones and papillary stenosis. Endoscopy 1983;15: 197-8. 2. May GR, Cotton PB, Edmunds SEJ, Chong W. Removal of stones from the bile duct at ERCP without sphincterotomy. Gastrointest Endosc 1993;39:749-54. 3. MacMathuna P, White P, Clarke E, Merriman R, Lennon JR, Crowe J. Endoscopic balloon sphincteroplasty (papillary dilation) for bile duct stones: efficacy, safety, and follow-up in 100 patients. Gastrointest Endosc 1995;42:468-74. 4. Minami A, Nakatsu T, Uchida N, Hirabayashi S, Fukuda H, Morshed SA, et al. Papillary dilation vs sphincterotomy in endoscopic removal of bile duct stones. Dig Dis Sci 1995;40:2550-4. 5. Bergman JJGHM, Rauws EAJ, Fockens P, van Berkel AM, Bossuyt PMM, Tijssen JGP, et al. Randomized trial of endoscopic balloon dilation versus endoscopic sphincterotomy for removal of bile duct stones. Lancet 1997;349:1124-9. 6. Blackstone MG. Balloon sphincteroplasty vs endoscopic papillotomy for bile duct stones. Lancet 1993;342:1314-5. 7. MacMathuna P, Lennon J, Crowe J. Balloon sphincteroplasty vs endoscopic papillotomy for bile duct stones. Lancet 1994;343:486. 8. Tamada K, Ido K, Ueno N, Kimura K, Ichiyama M, Tomiyama T. Preoperative staging of extrahepatic bile duct cancer with intraductal ultrasonography. Am J Gastroenterol 1995;90:239-46. 9. Tamada K, Ido K, Ueno N, Ichiyama M, Tomiyama T, Nishizono T, et al. Assessment of portal vein invasion by bile duct cancer using intraductal ultrasonography. Endoscopy 1995;27: 573-8. 10. Tamada K, Ido K, Ueno N, Ichiyama M, Tomiyama T, Nishizono T, et al. Assessment of the course and variations of VOLUME 49, NO. 3, PART 1, 1999
Intraductal US for residual bile duct stones during balloon sphincteroplasty
11.
12.
13.
14.
15.
16. 17.
18.
19.
20.
the hepatic artery in bile duct cancer by intraductal ultrasonography. Gastrointest Endosc 1996;44:249-56. Gress F, Chen YK, Sherman S, Savides T, Zaidi S, Jaffe P, et al. Experience with a catheter-based ultrasound probe in the bile duct and pancreas. Endoscopy 1995;27:178-84.. Furukawa T, Naitoh Y, Tsukamoto Y, Mitake M, Yamada M, Ishihara A, et al. New technique using intraductal ultrasonography for the diagnosis of disease of the pancreatobiliary system. Ultrasound Med 1992;11:607-12. Kuroiwa M, Tsukamoto Y, Naitoh Y, Hirooka Y, Furukawa T, Katoh T. New technique using intraductal ultrasonography for the diagnosis of bile duct cancer. Ultrasound Med 1994;13:189-95. Yasuda K, Mukai H, Nakajima M, Kawai K. Clinical application of ultrasonic probes in the biliary and pancreatic duct. Endoscopy 1992;24:370-5. Cushing GL, Fitzgerald PJ, Bommer WJ, Andrews MW, Cronan MS, Martines-Torres GG, et al. Intraluminal ultrasonography during ERCP with high-frequency ultrasound catheters. Gastrointest Endosc 1993;39:432-5. Engstrom CF, Wiechal KL. Endoluminal ultrasound of the bile ducts. Surg Endosc 1990;4:187-90. vanSonnenberg E, D’Agostino HB, Sanchez RL, Goodacre BB, Esch OG, Easter DG, et al. Percutaneous intraluminal US in the gallbladder and bile ducts. Radiology 1992;182:693-6. Lossef SV, Garra BS, Barth KH, Zeman RK. Percutaneous extraction of biliary stones: value of endoluminal sonography. AJR Am J Roentgenol 1992;159:411-2. Waxman I. Characterization of a malignant bile duct obstruction by intraductal ultrasonography. Am J Gastroenterol 1995;90:1073-5. Thomson H, Kisslo K, Farouk M, Chung K, Saperstein LA, Meyers WC. Technique of intraluminal biliary ultrasonography
VOLUME 49, NO. 3, PART 1, 1999
A Ohashi, N Ueno, K Tamada, et al
during laparoscopic cholecystectomy. Am J Surg 1993;165:265-9. 21. Noda T, Ido K, Ueno N, Tamada K, Ichiyama M, Tomiyama T, et al. Transpapillary intraductal ultrasonography (IDUS) of the bile duct without sphincterotomy. Ultrasound Int 1995;1:141-7. 22. Bergman JJGHM, Huibregtse K. Endoscopic sphincterotomy and stone extraction. Semin Laparosc Surg 1995;2:140-50 23. Hosmer DW Jr, Lemeshow S. Applied logistic regression. New York: John Wiley; 1989. 24. Vaira D, D’Anna L, Ainley C, Dowsett J, Williams S, Baillie J, et al. Endoscopic sphincterotomy in 1000 consecutive patients. Lancet 1989;334:431-3. 25. Cotton PB, Lehman G, Vennes J, Geenen JE, Russell RCG, Meyers WC, et al. Endoscopic sphincterotomy complications and their management: an attempt at consensus. Gastrointest Endosc 1991;37:383-93. 26. Muhldorfer SM, Kekos G, Hahn EG, Ell C. Complications of therapeutic gastrointestinal endoscopy. Endoscopy 1992;24:276-83. 27. Classen M. Endoscopic papillotomy: new indications, short and long-term results. Clin Gastroenterol 1986;15:457-69. 28. Rösch W, Reiman JR, Lux G, Lindner HG. Long-term follow-up after endoscopic sphincterotomy. Endoscopy 1981;13:152-3. 29. Edmundowicz SA, Aliperti G, Middleton WD. Preliminary experience using endoscopic ultrasonography in the diagnosis of choledocholithiasis. Endoscopy 1992;24:774-8. 30. Amouyal G, Amouyal P, Levy P, Tuzet S, Pallazo L, Vilgrain V, et al. Diagnosis of choledocholithiasis by endoscopic ultrasonography. Gastroenterology 1994;106:1062-7. 31. Prat F, Amouyal G, Amouyal P, Pelletier G, Fritsch J, Choury AD, et al. Prospective controlled study of endoscopic ultrasonography and endoscopic retrograde cholangiography in patients with suspected common-bile duct lithiasis. Lancet 1996;347:75-9.
GASTROINTESTINAL ENDOSCOPY
333
Background: We sought to determine the utility of intraductal ultrasonography (IDUS) in detecting residual bile duct stones during endoscopic balloon sphincteroplasty. Methods: Eighty-one consecutive patients with bile duct stones who underwent IDUS during endoscopic balloon sphincteroplasty were studied. IDUS was performed with a thin-caliber ultrasonic probe (diameter 2.0 mm, frequency 20 MHz) by a transpapillary route after stone extraction. When IDUS or balloon-retrograde cholangiography suggested residual stones, the bile duct was cleared again with a Dormia basket. Extraction of the stones was confirmed by direct duodenoscopic visualization. Videotapes of IDUS and cholangiograms were reviewed retrospectively without knowledge of the results of other diagnostic modalities. Results: In 27 of 81 patients (33%), IDUS detected small residual stones not seen on cholangiography. When stones were fragmented with mechanical lithotripsy, the accuracy of IDUS in detecting small residual stones was significantly greater than that of balloon-endoscopic retrograde cholangiography (95% vs 50%, p < 0.001). When the bile duct was greater than 10 mm in diameter, the accuracy of IDUS in detecting small residual stones was significantly greater than that of cholangiography (92% vs. 56%, p < 0.001). Conclusions: IDUS is useful for detecting small residual bile duct stones during endoscopic balloon sphincteroplasty when stones are fragmented by mechanical lithotripsy or when there is evidence of a dilated bile duct (>10 mm). (Gastrointest Endosc 1999;49:328-33) Recently endoscopic balloon dilation (sphincteroplasty) without endoscopic sphincterotomy (ES) has been reported as a safe, minimally invasive, functionpreserving method for treating bile duct stones.1-5 However, balloon dilation provides only a transient sphincteroplasty effect.6,7 Therefore accurate assessment for residual stones at the time of endoscopic balloon sphincteroplasty (EBS) is of paramount importance. One approach to assessment is intra-bile
Patients
Received December 16, 1996. For revision March 30, 1997. Accepted September 12, 1998. From the Department of Gastroenterology, Jichi Medical School, Tochigi, Japan. Reprint requests: Akira Ohashi, MD, Department of Gastroenterology, Jichi Medical School, Yakushiji, Tochigi 329-0498, Japan. Copyright © 1999 by the American Society for Gastrointestinal Endoscopy. 0016-5107/99/$8.00 + 0 37/1/94557
Eighty-one consecutive patients who underwent IDUS during EBS between July 1995 and June 1997 were included in the study. There were 43 men and 38 women. The mean age was 68.8 ± 14.0 years (range 28 to 94 years). Thirty-seven patients had gallbladder stones, and 23 of them underwent laparoscopic cholecystectomy before or after EBS. Twenty-two patients had a history of cholecystectomy. The remaining 22 patients had no gallbladder stones. The mean number of bile duct stones in individuals
328
GASTROINTESTINAL ENDOSCOPY
duct US with use of a thin-caliber ultrasonic probe, termed intraductal ultrasonography (IDUS), which has been shown to be useful in the diagnosis of bile duct diseases.8-21 We describe our preliminary experience with the detection of small residual bile duct stones with IDUS during EBS. PATIENTS AND METHODS
VOLUME 49, NO. 3, PART 1, 1999
Intraductal US for residual bile duct stones during balloon sphincteroplasty
3.0 ± 2.8 (range 1 to 17) with a mean size of 10.8 ± 5.0 mm range 4 to 23 mm) (Table 1). Written informed consent was obtained from each patient before EBS. Methods Diagnostic endoscopic procedure. Endoscopic retrograde cholangiography (ERC) was performed with use of a standard duodenoscope (model JF-200, Olympus Optical, Tokyo, Japan). Dilute contrast medium (30% amidotrizoic acid) was used for ERC. Premedication included diazepam (10 mg) and timepidium bromide (7.5 mg) administered intramuscularly. Endoscopic balloon sphincteroplasty. After the diagnostic ERC was performed, a 0.035-inch guidewire (Boston Scientific, Watertown, Mass) was passed through the diagnostic cannula into the bile duct. A Max Force 5F balloontipped biliary catheter (Boston Scientific) with a length of 180 cm and balloon length of 3 cm (maximum diameter 8 mm) was passed over the guidewire, positioning the deflated balloon across the papilla. The correct position of the balloon was confirmed by endoscopic visualization and by fluoroscopy. The balloon was then inflated with dilute radioopaque contrast medium to a pressure of 4 atmospheres for 60 seconds. After deflation for 30 seconds, the balloon was reinflated for 60 seconds to a pressure of 6 atmospheres. Subsequently, 2 to 3 inflations were performed with a pressure of 8 atmospheres separated by intervals of 30 seconds. Stone extraction. After the biliary catheter and guidewire were removed, stones were extracted by Dormia baskets, retrieval balloons, or both, according to standard guidelines.22 If necessary, stones (>8 mm in diameter) were fragmented by mechanical lithotripsy (MLT) (BML-4Q, Olympus Optical) before extraction. Intraductal ultrasonography. Stone removal was achieved by dragging a retrieval balloon (15 mm maximum balloon diameter, Boston Scientific) down the bile duct several times, after which balloon-ERC and IDUS were performed to detect residual stones. A cholangiogram was obtained with use of the retrieval balloon catheter. Two experienced examiners prospectively reviewed the fluoroscopic images. Then a thin ultrasonic probe (2.0 mm in diameter, 20 MHz, Aloka, Tokyo, Japan) was inserted into the bile duct through the accessory channel of the duodenoscope along the guidewire under fluoroscopic guidance. This high-frequency transducer provides an axial resolution of 0.1 mm and a maximum penetration of approximately 20 mm. IDUS images were recorded on videotape and individual instant photographs. Two experienced ultrasonographers prospectively reviewed the IDUS images in conjunction with the cholangiographic images. When IDUS indicated an echogenic focus with acoustic shadowing in the bile duct, the focus was considered to be a stone or fragment. After the procedure a nasobiliary drainage catheter (6F, 260 cm long, Cathex, Tokyo, Japan) was inserted into the bile duct for duct decompression and follow-up cholangiography. Study design When balloon-ERC or IDUS findings suggested residual stones, the procedure for cleaning the duct with a Dormia basket was repeated. Extraction of the stones was conVOLUME 49, NO. 3, PART 1, 1999
A Ohashi, N Ueno, K Tamada, et al
Table 1. Clinical details, stone characteristics, outcome, and complications in patients who underwent EBS Patients No. Mean age (y) (SD) Range (y) Bile duct stones Mean size (mm) (SD) Size range (mm) Mean No. (SD) Range (No.) Bile duct caliber Mean size (mm) (SD) Size range (mm) Efficacy Duct clearance EBS alone EBS + mechanical lithotripsy Frequency of endoscopic lithotripsy sessions 1-2 sessions 3 or more sessions Failed clearance in first admission Complications Pancreatitis Bleeding, perforation, or cholangitis
81 68.8 (14.0) 28-94 10.8 (5.0) 4-23 3.0 (2.8) 1-17 14.7 (5.0) 6-30 78/81 (96%) 37/81 (44%) 44/81 (56%) 68/81 (84%) 13/81 (16%) 3/81 (4%) 5/81 (6%) 0 (0%)
firmed by direct visualization with the duodenoscope. Bile duct clearance was repeated until balloon-ERC and IDUS showed no residual stones. For the blinded analysis of the two modalitics, cholangiograms obtained during balloon-ERC were reviewed retrospectively by two experienced radiologists without knowledge of the results of other diagnostic modalitics. Videotapes of IDUS were reviewed retrospectively by two experienced ultrasonographers without knowledge of the results of other diagnostic modalities. Agreement was obtained in 80 of 81 cases between the two radiologists and in all cases between the two ultrasonographers. Two radiologists discussed the case in which there was disagreement and made a final diagnosis. Statistical analysis All values are presented as the mean ± SD. Categoric results were analyzed with the chi-square test or Fisher’s exact test when the few cross-classifications were compared. Because multiple comparisons were made, only p values <0.01 were considered significant. In this manner, because fewer than 50 comparisons were made, findings can be interpreted as being significant at the 0.05 level. Multivariate effects were investigated with techniques of logistic regression.23
RESULTS Stone clearance The bile duct was cleared in 78 of 81 patients (96%) with EBS (Table 1). MLT was needed in 44 of 81 patients (54%). In 68 of 81 patients (84%), the bile duct was cleared in 1 or 2 sessions. The mean number GASTROINTESTINAL ENDOSCOPY
329
A Ohashi, N Ueno, K Tamada, et al
Intraductal US for residual bile duct stones during balloon sphincteroplasty
Figure 2. IDUS shows small residual stone (4 mm in diameter, arrow) at hepatic hilum. This was not detected by balloon-ERC.
The median follow-up was 12 months (range 1 to 24 months). To date, no adverse effects associated with EBS have been reported in any of the 81 patients except for stone recurrence in 5 patients. Comparison between IDUS and balloon-ERC
Figure 1. Cholangiographic findings after stone retrieval indicate no residual stones. IDUS catheter probe (arrow) is inserted into bile duct along guidewire.
of endoscopic sessions was 1.8 ± 1.0 (range 1 to 6). The mean duration of a treatment session was 69.0 ± 21.2 (range 32 to 140 minutes). IDUS scanning time was 3 to 5 minutes per patient. We were unable to clear the stones during the first admission in 3 patients who were treated successfully by a second EBS during a subsequent admission. Sphincterotomy was not performed during any treatment. Complications Transient abdominal discomfort was noted in half the patients during balloon inflation. No procedurerelated bleeding or perforation was encountered, but pancreatitis developed in 5 patients (6%), which resolved within 48 hours with administration of intravenous fluids and pain medication (Table 1). 330
GASTROINTESTINAL ENDOSCOPY
In 45 patients (56%) neither IDUS nor balloonERC showed residual stones (Table 2). However, in 27 patients (33%) IDUS demonstrated residual stones, although balloon-ERC failed to demonstrate them (Table 3, Figs. 1 and 2). In 1 patient (1%) IDUS failed to demonstrate an intrahepatic stone (migrating stone), which was detected by balloon-ERC. When IDUS or balloon-ERC showed residual stones, they were retrieved by a Dormia basket; retrieval was confirmed by direct duodenoscopic visualization. The sensitivity, specificity, accuracy, and negative predictive value of IDUS versus balloon-ERC in detecting small bile duct stones (fragments) were 85% (35/41) versus 22% (9/41), 100% (40/40) versus 100% (40/40), 93% (75/81) versus 60% (49/81), and 87% (40/46) versus 56% (40/72), respectively. The sensitivity, accuracy, and negative predictive value of IDUS were significantly greater than those of balloon-ERC (p < 0.001). In patients who underwent stone fragmentation with MLT the accuracy of IDUS in detecting small bile duct stones (fragments) was significantly greater than that of balloon-ERC (95% vs 50%, p < 0.001) (Table 4). When the bile duct caliber was greater than 10 mm, the accuracy of IDUS was also significantly greater than that of balloonERC (92% vs 56%, p < 0.001) (Table 5). However, logistic regression analysis of diameter of bile duct, number of stones, size of stones, and use of MLT on VOLUME 49, NO. 3, PART 1, 1999
Intraductal US for residual bile duct stones during balloon sphincteroplasty
Table 2. Detection of residual bile duct stones (fragments) after EBS with IDUS and balloon-ERC IDUS+ ERC+ Final diagnosis Stones present Stones absent Total
8 0 8
IDUS– ERC– 5 40 45
IDUS+ ERC– 27 0 27
IDUS– ERC+ 1 0 1
Agreement of IDUS and balloon-ERC on presence or absence of stones (fragments) 53/81 (65%); disagreement 28/81 (35%).
Table 3. Characteristics of residual stones (fragments) detected by IDUS, missed at balloon-ERC (n = 27) Mean stone size (mm) (SD, range) Mean stone number (SD, range) Mean bile duct caliber (mm) (SD, range) Use of MLT
3.6 1.3 15.8 20/27
(1.0, 2-6) (0.6, 1-3) (4.3, 8-25) (74%)
detecting residual stones by IDUS identified use of MLT as an almost significant (p = 0.011) predictor of residual stones (detected by IDUS, “missed” by balloon-ERC). Although dilated bile duct initially appeared to be a predictor of “missed” stones, this may be explained by the fact that patients with dilated bile ducts are more likely to have stones greater than 8 mm, for which MLT is necessary. DISCUSSION ES has been the standard method for removal of bile duct stones. However, ES has been shown to be associated with major complications such as bleeding, perforation, or pancreatitis in about 6% to 10% of patients.24-28 In addition to immediate complications, there are long-term risks of sphincterotomy in young patients referred for endoscopic stone removal before or after laparoscopic cholecystectomy. Recently, as an alternative approach to sphincterotomy, endoscopic balloon dilation or sphincteroplasty (EBS) has been reported1-5 and is possible because of advances in guidewire and biliary catheter technology. However, EBS provides only a transient sphincteroplasty effect.6,7 Therefore the assessment for residual stones after stone retrieval during EBS is a very important matter. Some investigators have conducted comparative studies of ERC and EUS and have reported that small bile duct stones (<4 mm in diameter) may be missed on conventional ERC.29-31 With ES small residual stones may be cleared spontaneously from the bile duct and have not been of great concern. In this study our research group used IDUS to detect small residual stones after stone removal during EBS. VOLUME 49, NO. 3, PART 1, 1999
Table 4. Accuracy of IDUS and balloon-ERC in detecting residual stones (fragments): relation to use of MLT
Total 41 40 81
A Ohashi, N Ueno, K Tamada, et al
MLT(–) IDUS Balloon-ERC
(89%)a
33/37 27/37 (73%)c
MLT(+) 42/44 (95%)b 22/44 (50%)d
a versus c, not significant (p = 0.068); b versus d, p < 0.001.
Table 5. Accuracy of IDUS and balloon-ERC in detecting residual stones (fragments) relation to bile duct caliber Nondilated bile duct IDUS Balloon-ERC
(93%)a
14/15 12/15 (80%)c
Dilated bile duct 61/66 (92%)b 37/66 (56%)d
a versus c, not significant ( p = 0.30); b versus d, p < 0.001. Nondilated bile duct ≤10 mm in diameter; dilated bile duct >10 mm in diameter.
Recently, many investigators have reported the usefulness of IDUS. 8-21 Assessing the extent of invasion of biliary tract cancer is one of the most promising aspects of this modality. The current study concentrated on another utility of IDUS in assessing the presence or absence of residual stones during EBS. Some investigators have already suggested the utility of IDUS in assessing small bile duct stones that were not adequately demonstrated by cholangiography. 16-18 However, there have been no reports addressing the utility of IDUS in visualizing residual stones during EBS. There are several advantages of IDUS compared with other diagnostic imaging modalities in visualizing bile duct stones: 1. IDUS can produce cross-sectional images of the bile duct in contrast to the luminal projection images of ERC, and IDUS images are not influenced by body habitus of the patient, bile duct caliber, or density of contrast material, in contrast to ERC. Our results indicated that balloon-ERC could not adequately recognize the presence of small stones or fragments when the bile duct diameter was greater than 10 mm. 2. Intra–bile duct scanning with use of a highfrequency ultrasonic probe produces high-resolution images not available with conventional extracorporeal US. 3. Once the IDUS probe is inserted into the bile duct, no special technique is required to demonstrate the whole extrahepatic bile duct, in contrast to EUS. GASTROINTESTINAL ENDOSCOPY
331
A Ohashi, N Ueno, K Tamada, et al
Intraductal US for residual bile duct stones during balloon sphincteroplasty
However, IDUS has some limitations: 1. IDUS cannot fully image the intrahepatic bile ducts, in contrast to ERC, because of the technical difficulty of probe insertion to each narrow branch. 2. Intra–bile duct insertion of a probe is essential for IDUS imaging. However, EUS is a high-quality, minimally invasive method to demonstrate small bile duct stones and does not require insertion of a probe into the bile duct or injection of contrast medium.29-31 In our study the accuracy of ERC was inferior to that of other studies,29-31 which can be explained by the fact that moderate to large stones had already been crushed and extracted at the time of balloon-ERC to assess for the presence of residual stones. And although sensitivity and specificity are the characteristics of the two procedures that are compared, their positive predictive value, negative predictive value, and accuracy are, in addition to their dependence on the sensitivity and specificity, also functions of the prevalence of the disease or condition in the setting in which the study takes place. With EBS stones were fragmented with use of MLT in 44 of 81 (56%) patients. Even in patients who did not require MLT, some stones may have been fragmented spontaneously when they were retrieved through the narrow bile duct opening. Our results indicate that, in EBS, conventional ERC (balloon-ERC) may fail to detect small residual stones (fragments), especially when stones are fragmented by MLT. This fact emphasizes the importance of accurate assessment for the presence of small fragments during EBS, because EBS provides only a transient sphincteroplasty effect. When IDUS detected residual stones, they were retrieved with use of a Dormia basket, as confirmed by direct visualization, although dragging the retrieval balloon before IDUS failed to clear them. This suggests a limitation of the retrieval balloon technique during EBS. There might be a concern that the sensitivity of IDUS cannot be established because further standard extraction techniques, such as the use of an extraction balloon, were not used when IDUS was negative and recurrent stones were found during follow-up in 5 patients. This could be mitigated by the following: 1. Cholangiograms made by a nasobiliary catheter placed after the final lithotripsy session showed no residual stones. 2. In 3 of 5 cases of recurrence (all 3 patients had prior cholecystectomy) multiple brown pigment 332
GASTROINTESTINAL ENDOSCOPY
stones developed (13, 14, and 16 mm in maximum stone diameter, respectively), which had not been demonstrated by nasobiliary cholangiograms after lithotripsy. 3. The remaining 2 patients with recurrent stones had multiple gallbladder stones and some of them entered the bile duct during follow-up. However, to establish the true sensitivity and negative predictive value of IDUS, performing a followup balloon cholangiogram and dragging the duct with a balloon 2 or 3 times in an effort to obtain confirmation that the duct had been cleared of stones are necessary. In the absence of confirmation of negative findings after IDUS, those 5 patients would be considered as having false-negative results of IDUS. In our study an air bubble in the bile duct was imaged as a strong echo with multiple scattering echos at IDUS; we could easily distinguish it from a residual stone (fragment). In conclusion, IDUS is useful to detect small residual bile duct stones (fragments) during EBS when stones are fragmented by MLT or when there is evidence of a dilated bile duct (>10 mm). However, intrahepatic migrating stones may be missed on IDUS and cholangiography remains necessary. REFERENCES 1. Staritz M, Ewe K, Meyer zum Buschenfelde KH. Endoscopic papillary dilation (EPD) for the treatment of common bile duct stones and papillary stenosis. Endoscopy 1983;15: 197-8. 2. May GR, Cotton PB, Edmunds SEJ, Chong W. Removal of stones from the bile duct at ERCP without sphincterotomy. Gastrointest Endosc 1993;39:749-54. 3. MacMathuna P, White P, Clarke E, Merriman R, Lennon JR, Crowe J. Endoscopic balloon sphincteroplasty (papillary dilation) for bile duct stones: efficacy, safety, and follow-up in 100 patients. Gastrointest Endosc 1995;42:468-74. 4. Minami A, Nakatsu T, Uchida N, Hirabayashi S, Fukuda H, Morshed SA, et al. Papillary dilation vs sphincterotomy in endoscopic removal of bile duct stones. Dig Dis Sci 1995;40:2550-4. 5. Bergman JJGHM, Rauws EAJ, Fockens P, van Berkel AM, Bossuyt PMM, Tijssen JGP, et al. Randomized trial of endoscopic balloon dilation versus endoscopic sphincterotomy for removal of bile duct stones. Lancet 1997;349:1124-9. 6. Blackstone MG. Balloon sphincteroplasty vs endoscopic papillotomy for bile duct stones. Lancet 1993;342:1314-5. 7. MacMathuna P, Lennon J, Crowe J. Balloon sphincteroplasty vs endoscopic papillotomy for bile duct stones. Lancet 1994;343:486. 8. Tamada K, Ido K, Ueno N, Kimura K, Ichiyama M, Tomiyama T. Preoperative staging of extrahepatic bile duct cancer with intraductal ultrasonography. Am J Gastroenterol 1995;90:239-46. 9. Tamada K, Ido K, Ueno N, Ichiyama M, Tomiyama T, Nishizono T, et al. Assessment of portal vein invasion by bile duct cancer using intraductal ultrasonography. Endoscopy 1995;27: 573-8. 10. Tamada K, Ido K, Ueno N, Ichiyama M, Tomiyama T, Nishizono T, et al. Assessment of the course and variations of VOLUME 49, NO. 3, PART 1, 1999
Intraductal US for residual bile duct stones during balloon sphincteroplasty
11.
12.
13.
14.
15.
16. 17.
18.
19.
20.
the hepatic artery in bile duct cancer by intraductal ultrasonography. Gastrointest Endosc 1996;44:249-56. Gress F, Chen YK, Sherman S, Savides T, Zaidi S, Jaffe P, et al. Experience with a catheter-based ultrasound probe in the bile duct and pancreas. Endoscopy 1995;27:178-84.. Furukawa T, Naitoh Y, Tsukamoto Y, Mitake M, Yamada M, Ishihara A, et al. New technique using intraductal ultrasonography for the diagnosis of disease of the pancreatobiliary system. Ultrasound Med 1992;11:607-12. Kuroiwa M, Tsukamoto Y, Naitoh Y, Hirooka Y, Furukawa T, Katoh T. New technique using intraductal ultrasonography for the diagnosis of bile duct cancer. Ultrasound Med 1994;13:189-95. Yasuda K, Mukai H, Nakajima M, Kawai K. Clinical application of ultrasonic probes in the biliary and pancreatic duct. Endoscopy 1992;24:370-5. Cushing GL, Fitzgerald PJ, Bommer WJ, Andrews MW, Cronan MS, Martines-Torres GG, et al. Intraluminal ultrasonography during ERCP with high-frequency ultrasound catheters. Gastrointest Endosc 1993;39:432-5. Engstrom CF, Wiechal KL. Endoluminal ultrasound of the bile ducts. Surg Endosc 1990;4:187-90. vanSonnenberg E, D’Agostino HB, Sanchez RL, Goodacre BB, Esch OG, Easter DG, et al. Percutaneous intraluminal US in the gallbladder and bile ducts. Radiology 1992;182:693-6. Lossef SV, Garra BS, Barth KH, Zeman RK. Percutaneous extraction of biliary stones: value of endoluminal sonography. AJR Am J Roentgenol 1992;159:411-2. Waxman I. Characterization of a malignant bile duct obstruction by intraductal ultrasonography. Am J Gastroenterol 1995;90:1073-5. Thomson H, Kisslo K, Farouk M, Chung K, Saperstein LA, Meyers WC. Technique of intraluminal biliary ultrasonography
VOLUME 49, NO. 3, PART 1, 1999
A Ohashi, N Ueno, K Tamada, et al
during laparoscopic cholecystectomy. Am J Surg 1993;165:265-9. 21. Noda T, Ido K, Ueno N, Tamada K, Ichiyama M, Tomiyama T, et al. Transpapillary intraductal ultrasonography (IDUS) of the bile duct without sphincterotomy. Ultrasound Int 1995;1:141-7. 22. Bergman JJGHM, Huibregtse K. Endoscopic sphincterotomy and stone extraction. Semin Laparosc Surg 1995;2:140-50 23. Hosmer DW Jr, Lemeshow S. Applied logistic regression. New York: John Wiley; 1989. 24. Vaira D, D’Anna L, Ainley C, Dowsett J, Williams S, Baillie J, et al. Endoscopic sphincterotomy in 1000 consecutive patients. Lancet 1989;334:431-3. 25. Cotton PB, Lehman G, Vennes J, Geenen JE, Russell RCG, Meyers WC, et al. Endoscopic sphincterotomy complications and their management: an attempt at consensus. Gastrointest Endosc 1991;37:383-93. 26. Muhldorfer SM, Kekos G, Hahn EG, Ell C. Complications of therapeutic gastrointestinal endoscopy. Endoscopy 1992;24:276-83. 27. Classen M. Endoscopic papillotomy: new indications, short and long-term results. Clin Gastroenterol 1986;15:457-69. 28. Rösch W, Reiman JR, Lux G, Lindner HG. Long-term follow-up after endoscopic sphincterotomy. Endoscopy 1981;13:152-3. 29. Edmundowicz SA, Aliperti G, Middleton WD. Preliminary experience using endoscopic ultrasonography in the diagnosis of choledocholithiasis. Endoscopy 1992;24:774-8. 30. Amouyal G, Amouyal P, Levy P, Tuzet S, Pallazo L, Vilgrain V, et al. Diagnosis of choledocholithiasis by endoscopic ultrasonography. Gastroenterology 1994;106:1062-7. 31. Prat F, Amouyal G, Amouyal P, Pelletier G, Fritsch J, Choury AD, et al. Prospective controlled study of endoscopic ultrasonography and endoscopic retrograde cholangiography in patients with suspected common-bile duct lithiasis. Lancet 1996;347:75-9.
GASTROINTESTINAL ENDOSCOPY
333