- Email: [email protected]
Predictors of unsuccessful mechanical lithotripsy and endoscopic clearance of large bile duct stones
Predictors of unsuccessful mechanical lithotripsy and endoscopic clearance of large bile duct stones Pramod Kumar Garg, MD, DM, Rakesh Kumar Tandon, MD, PhD, Vineet Ahuja, MD, DM, Govind Kumar Makharia, MD, DM, Yogesh Batra, MD, DM New Delhi, India
Background: Mechanical lithotripsy is used to break large bile duct stones. This study investigated the predictors of unsuccessful mechanical lithotripsy. Methods: Consecutive patients with bile duct stones underwent endoscopic retrograde cholangiography, sphincterotomy, and basket removal of stones. Mechanical lithotripsy was performed for stones of large size (>15 mm diameter) that precluded extraction intact. Success was defined as complete clearance of the duct. Various predictive factors, including size and number of stones, stone impaction, serum bilirubin, presence of cholangitis, and bile duct diameter were analyzed in relation to the success or failure of lithotripsy. Results: A total of 669 patients underwent endoscopic retrograde cholangiography for suspected choledocholithiasis, which was found in 401 patients. Of the latter patients, 87 had large stones that required mechanical lithotripsy. Lithotripsy was successful in 69 (79%) patients. Impaction of the stone(s) in the bile duct was the only significant factor that predicted failure of lithotripsy and consequent failure of bile duct clearance. Other factors, including stone size, were not significant. Conclusions: Mechanical lithotripsy is successful in about 79% of patients with large bile duct stones. The only significant factor that predicts failure of mechanical lithotripsy is stone impaction in the bile duct. (Gastrointest Endosc 2004;59:601-5.)
Bile duct (BD) stones are found in approximately 7% to 12% of patients with cholelithiasis.1,2 Bile duct stones vary in size from small (approximately 5 mm) to extremely large (>3 cm). The preferred approach to removal of stones is endoscopic retrograde cholangiography (ERC), sphincterotomy, and basket or balloon extraction. Bile duct stones up to 1.5 cm in diameter can be extracted intact after sphincterotomy, and the rate of successful extraction declines with increasing size of the stone.3 Larger stones must be fragmented before they can be extracted. The various modalities for lithotripsy include mechanical, electrohydraulic (EHL), laser, and extracorporeal shock wave (ESWL).4-7 Of these, mechanical lithotripsy is used most commonly.8-10 Described by Riemann et al.11 in 1982, mechanical lithotripsy is successful in breaking large stones, with subsequent clearance of the BD in 80% to 90% of cases.7-10 However, the factors that govern the success or Received August 14, 2003. For revision December 11, 2003. Accepted January 27, 2004. Current affiliations: Department of Gastroenterology, All India Institute of Medical Sciences, New Delhi, India. This paper was presented at the Digestive Diseases Week, May 1821, 2003, Orlando, Florida (Gastrointest Endosc 2003;57:AB194). Reprint requests: Pramod Kumar Garg, MD, DM, Department of Gastroenterology, All India Institute of Medical Sciences, New Delhi, 110029, India. Copyright Ó 2004 by the American Society for Gastrointestinal Endoscopy 0016-5107/$30.00 PII: S0016-5107(04)00295-0 VOLUME 59, NO. 6, 2004
failure of lithotripsy are largely unknown. Logically, factors such as the size of the stone(s), the number of the stones, the degree of jaundice as reflected by the serum bilirubin level, impaction of the stone(s) and the size of the BD may influence the success of mechanical lithotripsy. Only one study, retrospective in design, has examined the variables predictive of a successful outcome of mechanical lithotripsy, and this study found that the only predictor of outcome was stone size.12 The aim of the present study was to determine the predictors of success or failure of mechanical lithotripsy and subsequent duct clearance in patients with large BD stones. PATIENTS AND METHODS All patients with suspected BD stones who presented from January 1999 to December 2002 were included in the study. Bile duct stones were suspected if a patient with cholelithiasis or with a history of cholecystectomy presented with a dilated BD (>6 mm) on US, elevated serum levels of alkaline phosphatase and bilirubin, and/or cholangitis. Cholangitis was suspected if a patient presented with the classic Charcot’s triad or a history of fever and pain, together with abnormal biochemical tests of liver function, even if clinically anicteric. Endoscopic retrograde cholangiography by using a therapeutic duodenoscope was obtained in all patients suspected to have choledocholithiasis. The diameter of the BD (at the level of the stone) and the size (maximum transverse diameter) of the stone were measured after correction for the magnification when using the known diameter of the endoscope insertion tube GASTROINTESTINAL ENDOSCOPY
601
P Garg, R Tandon, V Ahuja, et al.
Large bile duct stones: predictors of unsuccessful mechanical lithotripsy
Table 1. Demographic data for patients (n = 87) Age (mean, SD) (range) (y) Gender Jaundice Cholangitis Gallbladder in situ
48.70 (15.30) (19-80) 34 male, 53 female 48 (55%) 18 (21%) 58 (67%)
SD, Standard deviation.
Table 2. Endoscopic retrograde cholangiography data Total no. patients No. ERCs Mean no. ERC/patient No. stones (range) Size of stone (mean, SD) (range) (cm.) Bile duct diameter (mean, SD) (range) (cm.) Mean no. lithotripsies
87 185 2.13 1-4 1.77 (0.30) (1.50-3.00) 2.19 (0.38) (1.60-3.50) 1.19
ERC, Endoscopic retrograde cholangiography; SD, standard deviation.
and its apparent diameter on a radiograph. Sphincterotomy, followed by basket extraction, was performed if stones were demonstrated by ERC. If a stone was greater than 15 mm in transverse diameter and was deemed by the endoscopist to be not amenable to extraction intact through the sphincterotomy, it was fragmented by using a mechanical lithotripter (BML 4Q, Lithocrush 201 or 202Q; Olympus Optical Co., Ltd., Tokyo, Japan). An extraendoscopic lithotripter (Soehendra type; Wilson-Cook Medical Inc., Winston-Salem, N.C.) was used as a rescue lithotripter. If the stone(s) was only partially fragmented or the duct was only partially cleared, a biliary stent or nasobiliary drain was placed temporarily. Endoscopic retrograde cholangiography with a further attempt at BD clearance, with or without lithotripsy, was performed at a later session. Lithotripsy was considered to have failed if the endoscopist felt that further attempts were likely to be futile in view of repeated failure to grasp the stone properly and crush it. Success was defined as complete clearance of the bile duct. If endoscopic clearance was unsuccessful, the patient was referred for surgical choledocholithotomy. The following factors were analyzed as potential predictors of the success or failure of mechanical lithotripsy and subsequent clearance of the BD: age, gender, pre- vs. post-cholecystectomy status, serum level of bilirubin, presence of cholangitis, size of the stone(s), number of stones, impaction of the stone(s), diameter of the BD, and proper opening of the lithotripter basket around the stone. Multiple stones was defined as more than two stones. A stone was defined as impacted in the BD if it was immobile and hampered/obstructed passage of the lithotripter basket. If a stone was impacted, an attempt was made to disimpact it with the basket to grasp it. Informed written consent was obtained from all patients undergoing ERC, and the study was performed in accor602
GASTROINTESTINAL ENDOSCOPY
dance with the humane and ethical principles of research set forth in the Helsinki guidelines. Statistical analysis Data are presented as mean (standard deviation) with the range given wherever appropriate. The categorical parameters were compared by chi-square test and the continuous variables by the Student t test. Multivariate logistic regression analysis was performed to determine the significance of various predictive variables. The modeling strategy was backward elimination, and 3 candidate variables were included that were significant or borderline significant after univariate analysis and that also were considered significant clinically, i.e., impaction of stone (1, non-impacted; 2, impacted), BD size (cm), and difference between BD size and stone size (cm). The outcome variable was success or failure of lithotripsy (0, failure; 1, success). A p value <0.05 was taken as significant. The statistical analysis was performed with statistical software (SPSS 7.5; SPSS Inc, Chicago, Ill.).
RESULTS A total of 669 patients with suspected BD stones underwent ERC during the study period, 401 (60%) of whom had BD stones. Of these 401 patients, 87 (22%) (34 men, 53 women; mean age 48.7 years) had stones greater than 1.5 cm in diameter that required mechanical lithotripsy. Jaundice was present in 48 (55%) patients. Eighteen (21%) patients had cholangitis at presentation. Of the 87 patients with large BD stones, 58 (67%) had a gallbladder in situ with gallstones, whereas, 29 (33%) had undergone cholecystectomy. For the post-cholecystectomy group, the median time interval between cholecystectomy and presentation with choledocholithiasis was 7 months (interquartile range 4-36 months; range 1 month to 8 years, except one patient in whom the interval was 18 years). The demographic and biochemical profile of the 87 patients with large BD stones is given in Table 1. In total, 185 ERC procedures were performed in the 87 patients (mean 2.13/patient) (Table 2). The number of stones ranged from one to 4; size ranged from 1.50 cm to 3 cm. The diameter of the BD ranged from 1.50 cm to 3.50 cm. The stone(s) was impacted in 15 patients (Figs. 1 and 2). In 3 of these 15 patients, the stone(s) could be disimpacted and, subsequently, captured in the basket and broken. The mechanical lithotripsy basket fractured at the joint between the wires and the long metal rod that is attached to the lithotripter handle in 4 patients; in two of them, the stones were broken with the extraendoscopic lithotripter, and the BD was cleared. In the other two patients, the basket wires were maneuvered free of the engaged stones, and the procedure was abandoned after placing a biliary stent. Overall, lithoVOLUME 59, NO. 6, 2004
Large bile duct stones: predictors of unsuccessful mechanical lithotripsy
P Garg, R Tandon, V Ahuja, et al.
Figure 1. Retrograde cholangiogram showing large impacted stone in mid bile duct.
Figure 2. Retrograde cholangiogram showing large impacted stone in proximal bile duct.
tripsy was successful in 69 of 87 (79%) patients. Of the remaining 18 patients, surgical choledocholithotomy was performed in 16 and long-term stent placement was used in two patients, because significant comorbid conditions precluded surgery. Two patients had retained stones even after surgery for which ERC and stone extraction was required. Complications developed in 5 patients, including bleeding (two patients) pancreatitis (two) and perforation (one). All except the one patient with a perforation were managed conservatively. There was no mortality. The causes of failure of lithotripsy (18 patients) were the following: (1) stone impaction in 12 patients that prevented insertion of the basket proximal to the stones, as well as full opening of the basket, with failure of the wires to encircle and grasp the stone(s); (2) multiple large stones in 3 patients that prevented full opening of the basket; (3) extremely hard stones that caused the basket wires to break (see above) in two patients; and (4) an incomplete procedure because of the development of bradycardia in one patient. Comparative data for patients with successful vs. unsuccessful lithotripsy are given in Table 3. Of the various factors analyzed, impaction of the stone was the only significant factor accounting for failure of
Table 3. Comparative data for patients with successful and unsuccessful mechanical lithotripsy
VOLUME 59, NO. 6, 2004
Successful lithotripsy N = 69
Parameter Age (y) Gender (M/F) Gallbladder in situ Bilirubin (mg/dL) Cholangitis Stone size (cm) Multiple stones (>2) Bile duct/diameter (cm) Difference between bile duct-stone size (cm) Impacted stone
Failed lithotripsy N = 18
p Value
50.11 (15.09) 43.28 (15.34) 27, 42 7, 11 44 (63.7%) 14 (77.7%) 4.60 (5.36) 4.78 (7.02) 16 (23%) 2 (11%) 1.78 (0.30) 1.71 (0.27) 20 (29%) 4 (22%) 2.23 (0.39) 2.03 (0.34) 0.44 (0.25) 0.32 (0.27) 3 (4%)
12 (66.6%)
0.092 0.270 0.310 0.911 0.260 0.417 0.649 0.061 0.066 <0.001
lithotripsy and consequent failure to clear the BD (p < 0.001). Bile duct diameter and the difference between the size of the BD and the stone were greater in the successful vs. the unsuccessful lithotripsy group, but the differences were not statistically significance (p = 0.06 for both variables). Multivariate logistic regression analysis showed impaction of the stone to be the only significant predictor of failure of mechanical lithotripsy, with an estimated odds GASTROINTESTINAL ENDOSCOPY
603
P Garg, R Tandon, V Ahuja, et al.
Large bile duct stones: predictors of unsuccessful mechanical lithotripsy
ratio 43.99: 95% CI[9.66, 200.42]). Stone impaction either prevented insertion of the basket proximal to the stone or led to failure of the basket to open fully and grasp the stone. Size of the stone(s), number of stones, pre- vs. post-cholecystectomy status, serum bilirubin level, presence of cholangitis, and BD diameter were not found to be significant predictors of failure of mechanical lithotripsy. DISCUSSION Large BD stones are difficult to remove endoscopically. Disparity between the length of the sphincterotomy and the diameter of the stone mandates that large stones be fragmented before removal to achieve BD clearance. Mechanical lithotripsy, the preferred and standard modality for fragmentation, has many advantages: it is immediately applicable during ERC, it does not require special instruments, it usually eliminates any need for another procedure (e.g., ESWL) at a later date, and it is safe and less costly than some other modalities, such as laser and electrohydraulic lithotripsy.13-15 Furthermore, continuing improvements in the design and the construction materials make the mechanical lithotripter easier to use and more effective.16,17 The reported success rate of mechanical lithotripsy ranges from 60% to 94%.7-10,13,14 An obviously logical reason for failure of lithotripsy is the extremely large size of the stone. A single study retrospectively examined factors that predicted success of mechanical lithotripsy and found that stone size determined success: the success rate was greater than 90% for stones less than 10 mm and 68% for those greater than 28 mm in size.12 Another study, however, found that stone impaction was the chief reason for failure of lithotripsy.18 The importance of predictive factors for success or failure of mechanical lithotripsy cannot be overemphasized. If it can be reliably predicted at ERC that mechanical lithotripsy is unlikely to succeed, then it may be worthwhile to refer the patient for surgical choledocholithotomy (especially if there are gallbladder stones for which cholecystectomy is indicated), instead of wasting time and resources, and exposing the patient to potential complications of multiple endoscopic procedures as often occurs in such cases. However, if mechanical lithotripsy is unsuccessful, other forms of lithotripsy (EHL, ESWL) may be attempted in older patients who have undergone cholecystectomy and in those who are poor candidates for surgery because of comorbid conditions. The present study examined the various possible predictive factors for success or failure of mechanical lithotripsy. The only important predictive factor for 604
GASTROINTESTINAL ENDOSCOPY
failure was stone impaction in the BD, with either an inability to pass the basket proximal to the stone or a failure of the basket to open fully around the stone to allow it to be grasped properly. Stone size was not found to be an important predictive factor. Size alone may not be important unless considered together with the diameter of the BD. In the present study, BD diameter and the difference between the diameter of the BD and the stone were greater in patients in whom mechanical lithotripsy was successful compared with those in whom it failed, whereas stone size was similar in the two groups. Even if the stone is extremely large, but the BD also is markedly dilated with adequate space between the stone and the wall of the duct, it may not be difficult to grasp and fragment the stone. Impaction will occur if the BD does not dilate adequately to accommodate a large stone, thus compromising the success of mechanical lithotripsy. Even if a stone is not impacted and can be moved within the BD, the duct may not be dilated adequately so that there is not enough space around the stone for the basket wires to open fully and encircle the stone. Thus, stone size alone may not be important, an observation that is confirmed by the results of the present study. No other factor was found to be significant as a predictor of failure of lithotripsy. The presence and degree of jaundice, as indicated by the serum level of bilirubin, might be important, because jaundice might indicate significant obstruction by a large stone. However, the present study did not substantiate this point, possibly because bilirubin level has many determinants. Stone hardness could be another important factor. Calcium bilirubinate stones are soft and often can be crushed manually with a standard Dormia basket.18 This type of stone usually is found in Asian patients with recurrent pyogenic cholangitis.18 In contrast, large cholesterol stones can be exceedingly hard, and, usually, mechanical lithotripsy is required for fragmentation. In an earlier study, the majority of stones in our patients were found by us to be composed of cholesterol19,20 and are similar to those found in patients in Western countries.21 Stone hardness, however, was not found in the present study to be a significant predictor of the success of mechanical lithotripsy. Although the lithotripter basket wires broke in 4 patients, the stones in two of these patients could, nevertheless, be crushed with an extraendoscopic lithotripsy device. It has been reported that a mechanical lithotripsy basket with a breaking strength of greater than 125 kg has increased the overall success of mechanical lithotripsy from 85% to 92% for stones up to 25 mm in diameter and from 67.6% to 91% for stones greater than 25 mm in size.16 VOLUME 59, NO. 6, 2004
Large bile duct stones: predictors of unsuccessful mechanical lithotripsy
In the study of Cipolletta et al.,12 which retrospectively examined factors that predicted success of mechanical lithotripsy, stone hardness also was not an important factor. Other factors, such as age, gender, pre- vs. post-cholecystectomy status, presence of cholangitis, and number of stones also were not significant predictive variables in the current study. Mechanical lithotripsy was successful in about 80% of patients with large BD stones. The only significant factor that accounted for failure of mechanical lithotripsy was stone impaction. Stone size alone was not significant as a predictor for success or failure. Alternate strategies, such as surgery or other types of lithotripsy, should be considered for patients with large impacted stones. REFERENCES 1. Koo KP, Traverso LW. Do preoperative indicators predict the presence of common bile duct stones during laparoscopic cholecystectomy? Am J Surg 1996;171:495-9. 2. Hammarstrom LE, Holmin T, Stridbeck H, Ihse I. Routine preoperative infusion cholangiography versus intraoperative cholangiography at elective cholecystectomy: a prospective study in 995 patients. J Am Coll Surg 1996;182:408-16. 3. Lauri A, Horton RC, Davidson BR, Burroughs AK, Dooley JS. Endoscopic extraction of bile duct stones: management related to stone size. Gut 1993;34:1718-21. 4. Siegel JH, Ben-Zvi JS, Pullano WE. Endoscopic electrohydraulic lithotripsy. Gastrointest Endosc 1990;36:134-6. 5. Hochberger J, Ell C. Laser lithotripy: the new wave. Can J Gastroenterol 1990;4:632-6. 6. Ponchon T, Martin X, Barkun A, Mestas JL, Chavaillon A, Boustiere C. Extracorporeal lithotripsy of bile duct stones using ultrasonography for stone localization. Gastroenterology 1990;98:726-32. 7. Van Dam J, Sivak MV Jr. Mechanical lithotripsy of large common bile duct stones. Cleve Clin J Med 1993;60:38-42.
VOLUME 59, NO. 6, 2004
P Garg, R Tandon, V Ahuja, et al.
8. Chung SC, Leung JW, Leong HT, Li AK. Mechanical lithotripsy of large common bile duct stones using a basket. Br J Surg 1991;78:1448-50. 9. Hintze RE, Adler A, Veltzke W. Outcome of mechanical lithotripsy of bile duct stones in an unselected series of 704 patients. Hepatogastroenterology 1996;43:473-6. 10. Leung JW, Neuhaus H, Chopita N. Mechanical lithotripsy in the common bile duct. Endoscopy 2001;33:800-4. 11. Riemann JF, Seuberth K, Demling L. Clinical application of a new mechanical lithotripter for smashing common bile duct stones. Endoscopy 1982;14:226-30. 12. Cipolletta L, Costamagna G, Bianco MA, Rotondano G, Piscopo R, Mutignani M, et al. Endoscopic mechanical lithotripsy of difficult common bile duct stones. Br J Surg 1997;84:1407-9. 13. Siegel JH, Ben-Zvi JS, Pullano WE. Mechanical lithotripsy of common duct stones. Gastrointest Endosc 1990;36:351-6. 14. Shaw MJ, Mackie RD, Moore JP, Dorsher PJ, Freeman ML, Meier PB, et al. Results of a multicenter trial using a mechanical lithotripter for the treatment of large bile duct stones. Am J Gastroenterol 1993;88:730-3. 15. Tandon RK, Garg PK. Management of large bile duct stones. Indian J Gastroenterol 1995;14:119-21. 16. Schneider MU, Matek W, Bauer R, Domschke W. Mechanical lithotripsy of bile duct stones in 209 patients: effect of technical advances. Endoscopy 1988;20:248-53. 17. Sorbi D, Van Os EC, Aberger FJ, Derfus GA, Erickson R, Meier P, et al. Clinical application of a new disposable lithotripter: a prospective multicenter study. Gastrointest Endosc 1999;49:210-3. 18. Leung JW, Chung SC, Mok SD, Li AK. Endoscopic removal of large common bile duct stones in recurrent pyogenic cholangitis. Gastrointest Endosc 1988;34:238-41. 19. Garg PK, Venkatachallam U, Tandon RK. Cholesterol bile duct stones with no stones in the gallbladder. J Clin Gastroenterol 1995;20:296-9. 20. Kumar D, Garg PK, Tandon RK. Clinical and biochemical comparative study of different types of common bile duct stones. Indian J Gastroenterol 2001;20:187-90. 21. Bernhoft RA, Pellegrini CA, Motson RW, Way LW. Composition and morphologic and clinical features of common duct stones. Am J Surg 1984;148:77-85.
GASTROINTESTINAL ENDOSCOPY
605
Background: Mechanical lithotripsy is used to break large bile duct stones. This study investigated the predictors of unsuccessful mechanical lithotripsy. Methods: Consecutive patients with bile duct stones underwent endoscopic retrograde cholangiography, sphincterotomy, and basket removal of stones. Mechanical lithotripsy was performed for stones of large size (>15 mm diameter) that precluded extraction intact. Success was defined as complete clearance of the duct. Various predictive factors, including size and number of stones, stone impaction, serum bilirubin, presence of cholangitis, and bile duct diameter were analyzed in relation to the success or failure of lithotripsy. Results: A total of 669 patients underwent endoscopic retrograde cholangiography for suspected choledocholithiasis, which was found in 401 patients. Of the latter patients, 87 had large stones that required mechanical lithotripsy. Lithotripsy was successful in 69 (79%) patients. Impaction of the stone(s) in the bile duct was the only significant factor that predicted failure of lithotripsy and consequent failure of bile duct clearance. Other factors, including stone size, were not significant. Conclusions: Mechanical lithotripsy is successful in about 79% of patients with large bile duct stones. The only significant factor that predicts failure of mechanical lithotripsy is stone impaction in the bile duct. (Gastrointest Endosc 2004;59:601-5.)
Bile duct (BD) stones are found in approximately 7% to 12% of patients with cholelithiasis.1,2 Bile duct stones vary in size from small (approximately 5 mm) to extremely large (>3 cm). The preferred approach to removal of stones is endoscopic retrograde cholangiography (ERC), sphincterotomy, and basket or balloon extraction. Bile duct stones up to 1.5 cm in diameter can be extracted intact after sphincterotomy, and the rate of successful extraction declines with increasing size of the stone.3 Larger stones must be fragmented before they can be extracted. The various modalities for lithotripsy include mechanical, electrohydraulic (EHL), laser, and extracorporeal shock wave (ESWL).4-7 Of these, mechanical lithotripsy is used most commonly.8-10 Described by Riemann et al.11 in 1982, mechanical lithotripsy is successful in breaking large stones, with subsequent clearance of the BD in 80% to 90% of cases.7-10 However, the factors that govern the success or Received August 14, 2003. For revision December 11, 2003. Accepted January 27, 2004. Current affiliations: Department of Gastroenterology, All India Institute of Medical Sciences, New Delhi, India. This paper was presented at the Digestive Diseases Week, May 1821, 2003, Orlando, Florida (Gastrointest Endosc 2003;57:AB194). Reprint requests: Pramod Kumar Garg, MD, DM, Department of Gastroenterology, All India Institute of Medical Sciences, New Delhi, 110029, India. Copyright Ó 2004 by the American Society for Gastrointestinal Endoscopy 0016-5107/$30.00 PII: S0016-5107(04)00295-0 VOLUME 59, NO. 6, 2004
failure of lithotripsy are largely unknown. Logically, factors such as the size of the stone(s), the number of the stones, the degree of jaundice as reflected by the serum bilirubin level, impaction of the stone(s) and the size of the BD may influence the success of mechanical lithotripsy. Only one study, retrospective in design, has examined the variables predictive of a successful outcome of mechanical lithotripsy, and this study found that the only predictor of outcome was stone size.12 The aim of the present study was to determine the predictors of success or failure of mechanical lithotripsy and subsequent duct clearance in patients with large BD stones. PATIENTS AND METHODS All patients with suspected BD stones who presented from January 1999 to December 2002 were included in the study. Bile duct stones were suspected if a patient with cholelithiasis or with a history of cholecystectomy presented with a dilated BD (>6 mm) on US, elevated serum levels of alkaline phosphatase and bilirubin, and/or cholangitis. Cholangitis was suspected if a patient presented with the classic Charcot’s triad or a history of fever and pain, together with abnormal biochemical tests of liver function, even if clinically anicteric. Endoscopic retrograde cholangiography by using a therapeutic duodenoscope was obtained in all patients suspected to have choledocholithiasis. The diameter of the BD (at the level of the stone) and the size (maximum transverse diameter) of the stone were measured after correction for the magnification when using the known diameter of the endoscope insertion tube GASTROINTESTINAL ENDOSCOPY
601
P Garg, R Tandon, V Ahuja, et al.
Large bile duct stones: predictors of unsuccessful mechanical lithotripsy
Table 1. Demographic data for patients (n = 87) Age (mean, SD) (range) (y) Gender Jaundice Cholangitis Gallbladder in situ
48.70 (15.30) (19-80) 34 male, 53 female 48 (55%) 18 (21%) 58 (67%)
SD, Standard deviation.
Table 2. Endoscopic retrograde cholangiography data Total no. patients No. ERCs Mean no. ERC/patient No. stones (range) Size of stone (mean, SD) (range) (cm.) Bile duct diameter (mean, SD) (range) (cm.) Mean no. lithotripsies
87 185 2.13 1-4 1.77 (0.30) (1.50-3.00) 2.19 (0.38) (1.60-3.50) 1.19
ERC, Endoscopic retrograde cholangiography; SD, standard deviation.
and its apparent diameter on a radiograph. Sphincterotomy, followed by basket extraction, was performed if stones were demonstrated by ERC. If a stone was greater than 15 mm in transverse diameter and was deemed by the endoscopist to be not amenable to extraction intact through the sphincterotomy, it was fragmented by using a mechanical lithotripter (BML 4Q, Lithocrush 201 or 202Q; Olympus Optical Co., Ltd., Tokyo, Japan). An extraendoscopic lithotripter (Soehendra type; Wilson-Cook Medical Inc., Winston-Salem, N.C.) was used as a rescue lithotripter. If the stone(s) was only partially fragmented or the duct was only partially cleared, a biliary stent or nasobiliary drain was placed temporarily. Endoscopic retrograde cholangiography with a further attempt at BD clearance, with or without lithotripsy, was performed at a later session. Lithotripsy was considered to have failed if the endoscopist felt that further attempts were likely to be futile in view of repeated failure to grasp the stone properly and crush it. Success was defined as complete clearance of the bile duct. If endoscopic clearance was unsuccessful, the patient was referred for surgical choledocholithotomy. The following factors were analyzed as potential predictors of the success or failure of mechanical lithotripsy and subsequent clearance of the BD: age, gender, pre- vs. post-cholecystectomy status, serum level of bilirubin, presence of cholangitis, size of the stone(s), number of stones, impaction of the stone(s), diameter of the BD, and proper opening of the lithotripter basket around the stone. Multiple stones was defined as more than two stones. A stone was defined as impacted in the BD if it was immobile and hampered/obstructed passage of the lithotripter basket. If a stone was impacted, an attempt was made to disimpact it with the basket to grasp it. Informed written consent was obtained from all patients undergoing ERC, and the study was performed in accor602
GASTROINTESTINAL ENDOSCOPY
dance with the humane and ethical principles of research set forth in the Helsinki guidelines. Statistical analysis Data are presented as mean (standard deviation) with the range given wherever appropriate. The categorical parameters were compared by chi-square test and the continuous variables by the Student t test. Multivariate logistic regression analysis was performed to determine the significance of various predictive variables. The modeling strategy was backward elimination, and 3 candidate variables were included that were significant or borderline significant after univariate analysis and that also were considered significant clinically, i.e., impaction of stone (1, non-impacted; 2, impacted), BD size (cm), and difference between BD size and stone size (cm). The outcome variable was success or failure of lithotripsy (0, failure; 1, success). A p value <0.05 was taken as significant. The statistical analysis was performed with statistical software (SPSS 7.5; SPSS Inc, Chicago, Ill.).
RESULTS A total of 669 patients with suspected BD stones underwent ERC during the study period, 401 (60%) of whom had BD stones. Of these 401 patients, 87 (22%) (34 men, 53 women; mean age 48.7 years) had stones greater than 1.5 cm in diameter that required mechanical lithotripsy. Jaundice was present in 48 (55%) patients. Eighteen (21%) patients had cholangitis at presentation. Of the 87 patients with large BD stones, 58 (67%) had a gallbladder in situ with gallstones, whereas, 29 (33%) had undergone cholecystectomy. For the post-cholecystectomy group, the median time interval between cholecystectomy and presentation with choledocholithiasis was 7 months (interquartile range 4-36 months; range 1 month to 8 years, except one patient in whom the interval was 18 years). The demographic and biochemical profile of the 87 patients with large BD stones is given in Table 1. In total, 185 ERC procedures were performed in the 87 patients (mean 2.13/patient) (Table 2). The number of stones ranged from one to 4; size ranged from 1.50 cm to 3 cm. The diameter of the BD ranged from 1.50 cm to 3.50 cm. The stone(s) was impacted in 15 patients (Figs. 1 and 2). In 3 of these 15 patients, the stone(s) could be disimpacted and, subsequently, captured in the basket and broken. The mechanical lithotripsy basket fractured at the joint between the wires and the long metal rod that is attached to the lithotripter handle in 4 patients; in two of them, the stones were broken with the extraendoscopic lithotripter, and the BD was cleared. In the other two patients, the basket wires were maneuvered free of the engaged stones, and the procedure was abandoned after placing a biliary stent. Overall, lithoVOLUME 59, NO. 6, 2004
Large bile duct stones: predictors of unsuccessful mechanical lithotripsy
P Garg, R Tandon, V Ahuja, et al.
Figure 1. Retrograde cholangiogram showing large impacted stone in mid bile duct.
Figure 2. Retrograde cholangiogram showing large impacted stone in proximal bile duct.
tripsy was successful in 69 of 87 (79%) patients. Of the remaining 18 patients, surgical choledocholithotomy was performed in 16 and long-term stent placement was used in two patients, because significant comorbid conditions precluded surgery. Two patients had retained stones even after surgery for which ERC and stone extraction was required. Complications developed in 5 patients, including bleeding (two patients) pancreatitis (two) and perforation (one). All except the one patient with a perforation were managed conservatively. There was no mortality. The causes of failure of lithotripsy (18 patients) were the following: (1) stone impaction in 12 patients that prevented insertion of the basket proximal to the stones, as well as full opening of the basket, with failure of the wires to encircle and grasp the stone(s); (2) multiple large stones in 3 patients that prevented full opening of the basket; (3) extremely hard stones that caused the basket wires to break (see above) in two patients; and (4) an incomplete procedure because of the development of bradycardia in one patient. Comparative data for patients with successful vs. unsuccessful lithotripsy are given in Table 3. Of the various factors analyzed, impaction of the stone was the only significant factor accounting for failure of
Table 3. Comparative data for patients with successful and unsuccessful mechanical lithotripsy
VOLUME 59, NO. 6, 2004
Successful lithotripsy N = 69
Parameter Age (y) Gender (M/F) Gallbladder in situ Bilirubin (mg/dL) Cholangitis Stone size (cm) Multiple stones (>2) Bile duct/diameter (cm) Difference between bile duct-stone size (cm) Impacted stone
Failed lithotripsy N = 18
p Value
50.11 (15.09) 43.28 (15.34) 27, 42 7, 11 44 (63.7%) 14 (77.7%) 4.60 (5.36) 4.78 (7.02) 16 (23%) 2 (11%) 1.78 (0.30) 1.71 (0.27) 20 (29%) 4 (22%) 2.23 (0.39) 2.03 (0.34) 0.44 (0.25) 0.32 (0.27) 3 (4%)
12 (66.6%)
0.092 0.270 0.310 0.911 0.260 0.417 0.649 0.061 0.066 <0.001
lithotripsy and consequent failure to clear the BD (p < 0.001). Bile duct diameter and the difference between the size of the BD and the stone were greater in the successful vs. the unsuccessful lithotripsy group, but the differences were not statistically significance (p = 0.06 for both variables). Multivariate logistic regression analysis showed impaction of the stone to be the only significant predictor of failure of mechanical lithotripsy, with an estimated odds GASTROINTESTINAL ENDOSCOPY
603
P Garg, R Tandon, V Ahuja, et al.
Large bile duct stones: predictors of unsuccessful mechanical lithotripsy
ratio 43.99: 95% CI[9.66, 200.42]). Stone impaction either prevented insertion of the basket proximal to the stone or led to failure of the basket to open fully and grasp the stone. Size of the stone(s), number of stones, pre- vs. post-cholecystectomy status, serum bilirubin level, presence of cholangitis, and BD diameter were not found to be significant predictors of failure of mechanical lithotripsy. DISCUSSION Large BD stones are difficult to remove endoscopically. Disparity between the length of the sphincterotomy and the diameter of the stone mandates that large stones be fragmented before removal to achieve BD clearance. Mechanical lithotripsy, the preferred and standard modality for fragmentation, has many advantages: it is immediately applicable during ERC, it does not require special instruments, it usually eliminates any need for another procedure (e.g., ESWL) at a later date, and it is safe and less costly than some other modalities, such as laser and electrohydraulic lithotripsy.13-15 Furthermore, continuing improvements in the design and the construction materials make the mechanical lithotripter easier to use and more effective.16,17 The reported success rate of mechanical lithotripsy ranges from 60% to 94%.7-10,13,14 An obviously logical reason for failure of lithotripsy is the extremely large size of the stone. A single study retrospectively examined factors that predicted success of mechanical lithotripsy and found that stone size determined success: the success rate was greater than 90% for stones less than 10 mm and 68% for those greater than 28 mm in size.12 Another study, however, found that stone impaction was the chief reason for failure of lithotripsy.18 The importance of predictive factors for success or failure of mechanical lithotripsy cannot be overemphasized. If it can be reliably predicted at ERC that mechanical lithotripsy is unlikely to succeed, then it may be worthwhile to refer the patient for surgical choledocholithotomy (especially if there are gallbladder stones for which cholecystectomy is indicated), instead of wasting time and resources, and exposing the patient to potential complications of multiple endoscopic procedures as often occurs in such cases. However, if mechanical lithotripsy is unsuccessful, other forms of lithotripsy (EHL, ESWL) may be attempted in older patients who have undergone cholecystectomy and in those who are poor candidates for surgery because of comorbid conditions. The present study examined the various possible predictive factors for success or failure of mechanical lithotripsy. The only important predictive factor for 604
GASTROINTESTINAL ENDOSCOPY
failure was stone impaction in the BD, with either an inability to pass the basket proximal to the stone or a failure of the basket to open fully around the stone to allow it to be grasped properly. Stone size was not found to be an important predictive factor. Size alone may not be important unless considered together with the diameter of the BD. In the present study, BD diameter and the difference between the diameter of the BD and the stone were greater in patients in whom mechanical lithotripsy was successful compared with those in whom it failed, whereas stone size was similar in the two groups. Even if the stone is extremely large, but the BD also is markedly dilated with adequate space between the stone and the wall of the duct, it may not be difficult to grasp and fragment the stone. Impaction will occur if the BD does not dilate adequately to accommodate a large stone, thus compromising the success of mechanical lithotripsy. Even if a stone is not impacted and can be moved within the BD, the duct may not be dilated adequately so that there is not enough space around the stone for the basket wires to open fully and encircle the stone. Thus, stone size alone may not be important, an observation that is confirmed by the results of the present study. No other factor was found to be significant as a predictor of failure of lithotripsy. The presence and degree of jaundice, as indicated by the serum level of bilirubin, might be important, because jaundice might indicate significant obstruction by a large stone. However, the present study did not substantiate this point, possibly because bilirubin level has many determinants. Stone hardness could be another important factor. Calcium bilirubinate stones are soft and often can be crushed manually with a standard Dormia basket.18 This type of stone usually is found in Asian patients with recurrent pyogenic cholangitis.18 In contrast, large cholesterol stones can be exceedingly hard, and, usually, mechanical lithotripsy is required for fragmentation. In an earlier study, the majority of stones in our patients were found by us to be composed of cholesterol19,20 and are similar to those found in patients in Western countries.21 Stone hardness, however, was not found in the present study to be a significant predictor of the success of mechanical lithotripsy. Although the lithotripter basket wires broke in 4 patients, the stones in two of these patients could, nevertheless, be crushed with an extraendoscopic lithotripsy device. It has been reported that a mechanical lithotripsy basket with a breaking strength of greater than 125 kg has increased the overall success of mechanical lithotripsy from 85% to 92% for stones up to 25 mm in diameter and from 67.6% to 91% for stones greater than 25 mm in size.16 VOLUME 59, NO. 6, 2004
Large bile duct stones: predictors of unsuccessful mechanical lithotripsy
In the study of Cipolletta et al.,12 which retrospectively examined factors that predicted success of mechanical lithotripsy, stone hardness also was not an important factor. Other factors, such as age, gender, pre- vs. post-cholecystectomy status, presence of cholangitis, and number of stones also were not significant predictive variables in the current study. Mechanical lithotripsy was successful in about 80% of patients with large BD stones. The only significant factor that accounted for failure of mechanical lithotripsy was stone impaction. Stone size alone was not significant as a predictor for success or failure. Alternate strategies, such as surgery or other types of lithotripsy, should be considered for patients with large impacted stones. REFERENCES 1. Koo KP, Traverso LW. Do preoperative indicators predict the presence of common bile duct stones during laparoscopic cholecystectomy? Am J Surg 1996;171:495-9. 2. Hammarstrom LE, Holmin T, Stridbeck H, Ihse I. Routine preoperative infusion cholangiography versus intraoperative cholangiography at elective cholecystectomy: a prospective study in 995 patients. J Am Coll Surg 1996;182:408-16. 3. Lauri A, Horton RC, Davidson BR, Burroughs AK, Dooley JS. Endoscopic extraction of bile duct stones: management related to stone size. Gut 1993;34:1718-21. 4. Siegel JH, Ben-Zvi JS, Pullano WE. Endoscopic electrohydraulic lithotripsy. Gastrointest Endosc 1990;36:134-6. 5. Hochberger J, Ell C. Laser lithotripy: the new wave. Can J Gastroenterol 1990;4:632-6. 6. Ponchon T, Martin X, Barkun A, Mestas JL, Chavaillon A, Boustiere C. Extracorporeal lithotripsy of bile duct stones using ultrasonography for stone localization. Gastroenterology 1990;98:726-32. 7. Van Dam J, Sivak MV Jr. Mechanical lithotripsy of large common bile duct stones. Cleve Clin J Med 1993;60:38-42.
VOLUME 59, NO. 6, 2004
P Garg, R Tandon, V Ahuja, et al.
8. Chung SC, Leung JW, Leong HT, Li AK. Mechanical lithotripsy of large common bile duct stones using a basket. Br J Surg 1991;78:1448-50. 9. Hintze RE, Adler A, Veltzke W. Outcome of mechanical lithotripsy of bile duct stones in an unselected series of 704 patients. Hepatogastroenterology 1996;43:473-6. 10. Leung JW, Neuhaus H, Chopita N. Mechanical lithotripsy in the common bile duct. Endoscopy 2001;33:800-4. 11. Riemann JF, Seuberth K, Demling L. Clinical application of a new mechanical lithotripter for smashing common bile duct stones. Endoscopy 1982;14:226-30. 12. Cipolletta L, Costamagna G, Bianco MA, Rotondano G, Piscopo R, Mutignani M, et al. Endoscopic mechanical lithotripsy of difficult common bile duct stones. Br J Surg 1997;84:1407-9. 13. Siegel JH, Ben-Zvi JS, Pullano WE. Mechanical lithotripsy of common duct stones. Gastrointest Endosc 1990;36:351-6. 14. Shaw MJ, Mackie RD, Moore JP, Dorsher PJ, Freeman ML, Meier PB, et al. Results of a multicenter trial using a mechanical lithotripter for the treatment of large bile duct stones. Am J Gastroenterol 1993;88:730-3. 15. Tandon RK, Garg PK. Management of large bile duct stones. Indian J Gastroenterol 1995;14:119-21. 16. Schneider MU, Matek W, Bauer R, Domschke W. Mechanical lithotripsy of bile duct stones in 209 patients: effect of technical advances. Endoscopy 1988;20:248-53. 17. Sorbi D, Van Os EC, Aberger FJ, Derfus GA, Erickson R, Meier P, et al. Clinical application of a new disposable lithotripter: a prospective multicenter study. Gastrointest Endosc 1999;49:210-3. 18. Leung JW, Chung SC, Mok SD, Li AK. Endoscopic removal of large common bile duct stones in recurrent pyogenic cholangitis. Gastrointest Endosc 1988;34:238-41. 19. Garg PK, Venkatachallam U, Tandon RK. Cholesterol bile duct stones with no stones in the gallbladder. J Clin Gastroenterol 1995;20:296-9. 20. Kumar D, Garg PK, Tandon RK. Clinical and biochemical comparative study of different types of common bile duct stones. Indian J Gastroenterol 2001;20:187-90. 21. Bernhoft RA, Pellegrini CA, Motson RW, Way LW. Composition and morphologic and clinical features of common duct stones. Am J Surg 1984;148:77-85.
GASTROINTESTINAL ENDOSCOPY
605