Donald E. Wenner, MD FACS
Paul Whitwam, MD
James Rosser, MD FACS
Syed Hashmi, MD
Donald E. Wenner III

Introduction
LCBDE (laparoscopic common bile duct exploration) has been shown to be a safe, cost effective way to treat CBD stones. Despite this, LCBDE has not gained widespread adoption by surgeons. The technique has proven difficult to master, and damage to the fragile choledochoscope by grasping forceps and passage through the port valves has been problematic. Cases involving large, impacted, and/or multiple stones have required conversion to open CBDE.

Method
The Multi-channel Instrument Guide (MIG) is introduced as a solution for these problems. The MIG is a J-shaped plastic extrusion with three lumens. It is flexible and can be straightened for insertion through a 10 mm port. The MIG facilitates insertion of a flexible 2.8–3.2 mm choledochoscope into the CBD. At the same time, additional tools such as balloon or irrigation catheters and lithotripters can be introduced into the CBD. These can be manipulated under video guidance via the choledochoscope. This procedural multi-tasking allows for a more efficient LCBDE. We describe our initial experience using the MIG in 23 patients.

Results
Of the 23 patients who had LCBDE procedures, 20 had stones present in the CBD. Multiple stones were present in 48% of the patients and impacted stones were present in 26% of the patients and large stones, greater than 1 cm, were present in 26% of the patients. A 95% stone clearance rate was achieved. Difficult cases with large, impacted and/or multiple stones were resolved using the MIG. Two choledochoscopes were damaged, one during surgery, and one during cleaning and storage.

Conclusion
The MIG has demonstrated significant advantages over previously described techniques. The device secures biliary tract access and allows procedural multitasking while at the same time protecting the delicate and expensive equipment. Subsequently a simplified technique algorithm can be followed that may encourage more surgeons to adopt the routine performance of LCBDE.

Key Words: choledocholithiasis, intraoperative cholangiogram, laparoscopic cholecystectomy, laparoscopic common bile duct exploration, choledochoscopy, Multi-channel Instrument Guide

Introduction
Laparoscopic cholecystectomy (LC) is the most common elective abdominal surgical procedure done in the United States [2]. Between 10 and 15% of patients undergoing cholecystectomy have associated common bile duct (CBD) stones [11]. As patients age, their chance of having common duct stones increases [8]. The treatment of CBD stones has changed markedly since Ludwig Courvoisier did the first common bile duct exploration (CBDE) with removal of stones in 1890 [1]. In the current era, the treatment of choledocholithiasis has followed three routes: preoperative endoscopic retrograde cholangiography (ERCP), laparoscopic common bile duct exploration (LCBDE), or postoperative ERCP. The disadvantage of ERCP followed by endoscopic sphincterotomy (ES) is its associated morbidity and mortality [18, 5]. LCBDE has been shown to be as effective, if not superior to ERCP [4, 15]. It has also been shown to decrease hospital length of stays [13]. Despite its cost advantages, widespread application of LCBDE has been limited [14]. Trans-cystic duct approaches to LCBDE have been described in detail, and these have limitations imposed by the size of the cystic duct, the anatomy of the cystic duct-CBD junction, and the difficulty in accessing the hepatic ducts [2, 13, 14]. We describe our experience with the use of the Multi-channel Instrument Guide (MIG) which we have found to be useful in LCBDE.

Methods
In this initial pilot study on the use of the MIG, we attempted to duplicate open CBDE as closely as possible using laparoscopic techniques. A choledochotomy approach to LCBDE was used in all of the patients in this study. Patients who had a trans-cystic duct approach to LCBDE during this time period are not included in this analysis, and we did not use the MIG for trans-cystic LCBDE during this initial evaluation of the MIG. All patients in whom the MIG was used between August 2001 and April 2003 are included for analysis. The MIG is supplied by LapSurgical Systems LLC, (888 527 7874), P.O. Box 3525, Roswell, NM 88202. The device is supplied as a disposable single use item packaged sterile in a peel pack. The device has received FDA clearance for laparoscopic access to the CBD. The MIG assembly consists of three parts: a 9 mm outer diameter (O.D.) introducer sheath, the MIG itself, and an occlusion plug. The introducer sheath is 25 cm in length, and will fit through a standard 10 mm laparoscopic port. The MIG itself is a three lumen plastic extrusion with an O.D. of 7 mm and a 35 cm length. The MIG has a J-shaped tip, which is cut at a 45-degree angle to aid in the insertion of the MIG into a choledochotomy. The largest lumen of the MIG has a 3.5 mm inside diameter (I.D.), which allows the passage of a flexible choledochoscope with O.D. of 2.8-3.2 mm. The other two MIG lumens are 1.9 mm I.D. and allow for the passage of balloon catheters, irrigation catheters and stone baskets (Fig. 1). Both the inside and the outside surfaces of the MIG are coated with a lubricious, hydrophilic coating that reduces friction and allows instruments to be run smoothly through the lumens. The coating also reduces friction between the extrusion and the sheath. Wetting the extrusion with sterile saline solution activates the lubricious coating. The third part of the MIG assembly is an occlusion plug for the two small lumens. This is present to prevent loss of pneumoperitoneum when the MIG is being used in conjunction with the choledochoscope, and the smaller lumens are not occupied with an instrument. This feature is useful for introduction of a choledochoscope into the cystic duct for trans-cystic LCBDE, and was not used during this initial study. The plug can be used so that one or both of the smaller lumens can be plugged. The MIG ready for action with choledochoscope, nitinol stone basket and irrigation catheter is shown in Figure 2.

Patients
Between August 2001 and April 2003, 23 patients (15 women and 8 men) underwent a LCBDE procedure using the MIG by one of three surgeons. The average patient age was 68 years (range 16–91). 65% of the patients were female. The mean preoperative bilirubin was 3.5 (range 0.5–21.7). The mean preoperative alkaline phosphatase was 324 (range 105–977). Serum amylase preoperatively ranged from 27–600 with the mean being 125. The WBC had a range of 6300-24800 with a mean WBC of 11575. In all but one patient, LCBDE was performed in conjunction with LC with dynamic fluoroscopic intra-operative cholangiography (DFIOC) being used to detect the presence of stones. The one patient who did not have a LCBDE performed with a cholecystectomy had undergone a previous LC with DFIOC and had shown no signs of CBD stones at his initial surgery. However the patient showed laboratory signs of biliary tract obstruction and a magnetic resonance cholangiogram suggested an obstructing stone. An ERCP was attempted but was not successful, thus a decision to perform a LCBDE was made.

Operative Technique
Four laparoscopic port sites were established, a 10 mm umbilical port, a 10 mm epigastric port and two 5 mm ports in the anterior axillary line on the right side, one in the sub-costal position and one at the level of the umbilicus. A DFIOC was performed using a dual lumen 4 F balloon cholangiocatheter. Based on the results of the DFIOC (anatomy and size of the cystic duct, the number, location and site of stones) patients were selected for a choledochotomy approach to LCBDE. In this study an anterior longitudinal choledochotomy was employed in all of these patients. The MIG was inserted through the epigastric 10 mm port site. A fifth port located in either the left sub-costal area in the mid-clavicular line, or in the right abdomen between the umbilical port and the lower 5 mm port site was occasionally established for additional retraction. This fifth port was especially useful in obese patients. Endoshears were used to perform the choledochotomy. A procedural algorithm (Fig. 3) was used to guide the LCBDE procedure. Once the choledochotomy was made, an attempt was made to flush the stones out with irrigation or to use a balloon catheter to remove stones. These attempts were made without choledochoscope guidance. This was done while the operating room personnel were setting up the necessary equipment (choledochoscope, video camera, pressure irrigation bags, etc.) for the LCBDE procedure. If the stones were successfully removed, the MIG was used to introduce a choledochoscope into the choledochotomy and the choledochoscope was used to confirm total clearance of the biliary tree. This was defined as clearance at stage 1. If these initial attempts were not successful in clearing the bile ducts, the MIG was used with the choledochoscope as well as with a balloon catheter and irrigation catheter. The choledochoscope was run through the large MIG lumen and the balloon catheter and irrigation catheter were run through the smaller lumens. Using video guidance provided by the choledochoscope, the balloon catheter was positioned beyond the stone and then inflated and pulled back to dislodge the stone. This was defined as clearance at stage 2. If the bile duct was still not cleared, the technique algorithm progressed to stage 3. A lithotripter was run through the working channel of the choledochoscope, an irrigation catheter (4 F) connected to a pressurized saline infusion line was then run through a small lumen of the MIG and the tip positioned adjacent to the impacted stone. The lithotripter was used to shock and subsequently fracture the stone. Irrigation was continually run through the irrigation catheter to maintain a clear field of view and flush away stone debris and blood. Once the impacted stone was sufficiently broken, a balloon catheter was run through one of the small lumens in the MIG and advanced past the stone, inflated and then pulled back out of the choledochotomy to remove any remaining stone fragments. After any one of these methods was successfully performed, the choledochoscope was used in conjunction with the MIG to verify final clearance of the biliary tree.

Both primary closure of the choledochotomy and placement of T-tubes were used during the study. The decision to perform primary closure was based on the final choledochoscope inspection of the bile duct. If on final inspection, there was a widely patent ampulla (verified by the ability to pass a choledochoscope through the ampulla into the duodenum), only mild inflammation, no infection or cholangitis, and the surgeon was sure that all of the stones were removed, primary closure of the choledochotomy was performed. If there was any question, a T-tube was placed in the duct and the choledochotomy was closed around it. If primary closure of the choledochotomy was done, a cystic duct cholangiogram was done to check that the closure was not leaking, that the Ampulla of Vater was patent, and that there were no retained stones. If a T-tube was placed, a T-tube cholangiogram was performed to document good positioning of the T-tube within the CBD. A final T-tube cholangiogram was also performed as an outpatient prior to removal of the T-tube to document final clearance of the CBD. The T-tube was left in place for a minimum of four weeks before removal.

Several procedural details on the use of the MIG are worth mentioning. We like the 70 cm x 2.8 mm choledochoscope for use with the MIG. It is long enough that it can be laid on the sterile field over the patient’s lower chest (Fig, 4). Gentle tip deflection is all that is required to manipulate the flexible tip of the choledochoscope. This can be provided by the surgical assistant or technician. The surgeon manipulates the choledochoscope and catheters in and out of the MIG (Fig. 5). When we have a large bile duct (> 1 cm), we insert the MIG into the choledochotomy, when we have a smaller bile duct we position the MIG just above the choledochotomy, and direct the choledochoscope and tools through the choledochotomy. Pressure irrigation through both the working channel of the choledochoscope and through an irrigation catheter helps to keep the bile duct distended and maintains a clear field of view within the bile duct. The MIG is best loaded with choledochoscope and tools before insertion; it is straightened by pulling back into the introducer sheath both for insertion and withdrawal. In order to change direction after completing inspection of the distal bile duct for instance, the tools and choledochoscope are simply withdrawn into the MIG and it is rotated 180 degrees, now the surgeon is ready to begin inspection of the proximal bile duct. The MIG helps the surgeon keep order given the complex array of irrigation catheters, choledochoscopes, video cameras, laparoscopes and lithotripters that this procedure requires.

We have found that the laparoscope gives a good magnified view of the CBD, and precise placement of sutures in the CBD is possible. We have found that intra-corporeal knot tying techniques work well when the choledochotomy is closed primarily. The extra-corporeal techniques are easiest when placing a T-tube; however care must be exercised to support the tissues when pulling the long suture used for extra-corporeal knot tying through the bile duct wall.

We have placed a closed suction drain in the sub-hepatic space in all LCBDE patients in whom a choledochotomy has been used.

Results
Of the 23 patients who had a MIG enhanced LCBDE procedure, 20 had stones present in the CBD. Multiple stones were present in 48% of the patients, impacted stones were present in 26%, large stones, > 8 mm, were present in 35%, 13% had fibrin embedded stones, and hepatic duct stones were present in 13% as well (Fig. 6). Seventeen of the twenty stone patients (85%) had one or more of these features. The mean operative time was 147 minutes (range 77–263). Clearance of stones was achieved at stage 1 in 48% of patients, at stage 2 in 30% of patients and at stage 3 in 22% of patients (Fig. 7). In no case did a patient with stones > 1 cm, or impacted or fibrin embedded stones achieve clearance at stage 1. All patients that progressed to stage 3 had either large stones > 1 cm or impacted stones. T-tubes were used in 19 of the cases. Four patients had LCBDE performed as an outpatient surgical procedure. There was no mortality in the study. Clinical or laboratory evidence of pancreatitis following LCBDE did not occur in any of these patients. Major complications occurred in 3 patients (13%). These included a 90 year old who developed post-operative atrial fibrillation, CHF, and an elevated CKMB of 6.4. She responded to therapy and recovered well. One patient had retained stones and required subsequent ERCP/ES. One patient who had her choledochotomy closed primarily developed bile peritonitis and required return to the OR for placement of a T-tube. We had not checked the choledochotomy closure by infusing saline through the cystic duct and doing a DFIOC. This complication changed our technique. Minor complications that required no intervention included a dislodged T-tube and prolonged bilious drainage from the T-tube tract after T-tube removal. Over the course of the study, two choledochoscopes had to be repaired due to scope damage. One was damaged during manipulation in surgery and one was damaged by closing the choledochoscope case lid on the tip of the choledochoscope. In the three patients that did not have stones causing biliary obstruction, the LCBDE was deemed a success because the MIG and the choledochoscope were successfully used to inspect the CBD and to diagnose the nature of the common bile duct obstruction. In one patient this was secondary to extrinsic compression of the CBD from an edematous pancreas, a second patient had obstruction secondary to pancreatic cancer, and the third had obstruction secondary to stenosis of the Ampulla. These three patients had their jaundice palliated by T-tube insertion. The patient with ampullary stenosis underwent a successful ERCP/ES post-operatively, the patient with pancreatic cancer was not a candidate for resection for cure and had a subsequent biliary bypass procedure, and the patient with the edematous pancreas had the T-tube removed after the pancreatic edema had subsided, and made an uneventful recovery.

Of the 20 patients who had stones present, 19 had total clearance of the CBD using the MIG enhanced LCBDE. This was a stone clearance success rate of 95%. The one patient, who did not have full clearance, had multiple impacted stones. This patient had multiple stones removed during LCBDE; however several stones were missed on final inspection. These likely were lodged in hepatic radicals. A cholangiogram done postoperatively showed retained stones and a post-operative ERCP/ES was performed. This patient had final successful clearance of all stones; however she developed post ERCP pancreatitis. Three patients in the study had preoperative ERCP. In two, the ERCP was unsuccessful because there were multiple large impacted stones. Both of these patients had subsequent successful LCBDE procedures with clearance of the common bile duct through the use of the MIG with the lithotripter and balloon catheter. The third patient had severe pancreatitis that caused swelling and distortion of the ampulla of Vater, which prevented cannulation of the CBD and thus the ERCP had to be aborted. Subsequent LCBDE failed secondary to severe adhesion formation and the patient underwent conversion to an open CBDE. This exploration showed the CBD to be clear of stones; however, it was compressed and obstructed secondary to severe pancreatic edema. We used the MIG to pass the choledochoscope during the open CBDE in this case, and placed a T-tube. Postoperatively, the patient did well.

Discussion
In this study, the MIG enhanced LCBDE has shown a 95% clearance rate. This success rate was obtained in spite of 85% (17of 20) patients having difficult stone features. These include 11 patients with multiple stones, six with large (> 1 cm) stones, six with impacted stones, three with hepatic duct stones, and three with fibrin embedded stones. The totals add up to over seventeen as several patients had several difficult features. In other studies, LCBDE stone clearance rates of 75-97% have been found [5, 13, 14, 7]. The MIG allows for an irrigation catheter, balloon catheter, stone basket and lithotripter all to be deployed at once with video guidance via the choledochoscope. This multi-tasking also allows for a more efficient LCBDE. An algorithm for using the MIG was employed that progressed from simple to more complex methodology.

The MIG is also effective in minimizing damage to the choledochoscope which occurs during LCBDE. Our personal experience is that choledochoscope damage occurs about half the time when the 2.8 or 3.2 mm choledochoscope is used during LCBDE. The choledochoscope is protected when it is inside the MIG and thus does not have a chance to get damaged by either the port or by grasping forceps. In our study, only two choledochoscopes were damaged in 23 cases. One choledochoscope was damaged during surgery, and one was damaged during cleaning and storage by closing the case lid on the tip of the choledochoscope. The small choledochoscopes have advantages over the larger 4.9 mm choledochoscopes. The small choledochoscope is more maneuverable in small bile ducts, and can be advanced across the ampulla more easily. Stones impacted within the ampulla can be reached with the small choledochoscope that are inaccessible with the larger choledochoscope. The small choledochoscope can traverse into the smaller hepatic radicals and is easier to maneuver into these smaller radicals than the larger choledochoscope.

LCBDE, using the MIG, allows for a single procedure to remove both the gallbladder and to clear stones from the biliary tree. The single-step procedure has been shown to be cost-effective [13]. We continue to be concerned about the disadvantages of ERCP/ES, especially the risk of post ERCP pancreatitis. Acute pancreatitis is a common complication of ERCP with an incidence of 8–18% [12, 3]. Despite the technical improvements in recent years and the increased expertise of operators, the incidence of pancreatitis after ERCP/ES has not decreased [5]. Recurrent bile duct stones after sphincterotomy occur in 2-7% of patients after previous bile duct clearance [10, 6]. The recurrent stones are the bilirubinate type, irrespective of the type of stones at initial treatment, suggesting that bacterial infestation due to ablation of the sphincter mechanism may have a causative role [17]. Furthermore, questions have also been raised whether there is an elevated risk of bile duct cancer after biliary enteric anastomosis or sphincterotomy [16, 9].

We have also shown that LCBDE can be done in an ambulatory setting. In fact, four of our patients had LC, LCBDE in an outpatient setting. In order for surgeons to achieve a high success rate for LCBDE they must be familiar with both trans-cystic and choledochotomy approaches. We feel that the MIG is a major advance in the era of LCBDE and will be useful for both trans-cystic and choledochotomy approaches to LCBDE. A more efficient LCBDE procedure with a rational procedural algorithm may encourage more surgeons to adopt the routine performance of LCBDE. The MIG allows the surgeon many options and tool combinations that were heretofore impossible to use. We are just beginning to understand the possibilities and utilities that this simple device provides to the expert laparoscopic biliary surgeon with advanced laparoscopic skills.

References
1. Beal JM. (1984) Historical Perspective of Gallstone Disease. Surg Gynecol Obstet 158:181-189.
2. Berci, G, Morgenstern, L. (1994) Laparoscopic management of common bile duct stones. Surg Endosc 8:1168-1175.
3. Centers for Disease Control and Prevention, National Center for Health Statistics, Division of Health Care Statistics, Hospital Care Statistics Branch. Available at: http//www.cdc.gov/nchs/data/tables/2002/02hus095.pdf. Accessed February 24, 2003.
4. Cotton P, Lehman G, Vennes J, et al. (1991) Endoscopic sphincterotomy complications and their management: an attempt at consensus. Gastrointest Endosc 37:383-393.
5. Cuscheri A, Lezoche E, Morino M. (1999) EAES multicenter prospective randomized trial comparing two-stage versus single-stage management of patients with gallstone disease and ductal calculi. Surg Endosc 13:952-957.
6. Freeman ML, Nelson DB, Sherman S et al. (1996) Complications of endoscopic biliary sphincterotomy. N Engl J Med 335:909-18.
7. Geenen D, Geenen J, Jafri F, et al. (1998) The role of surveillance endoscopic retrograde cholangio-pancreatography in preventing episodic cholangitis in patients with recurrent common bile duct stones. Endoscopy 30:18-20.
8. Giurgiu, DI, Margulies DR, Carroll BJ, et al. (1999) Laparoscopic common bile duct exploration: long term outcome. Arch Surg 134:839-844.
9. Glen F, McShervey, CK. (1975) Calculus Biliary Disease. In: Ravitch MM (ed), Current Problems in Surgery, Year Book Medical, Chicago, pp 1-38.
10. Hakamada K, Sasaki M, Endoh M, et al. (1997) Late development of bile duct cancer after sphincteroplasty: a ten- to twenty-two year follow-up study. Surgery 121:488-92.
11. Ideda S, Tanaka M, Matsumoto S, et al. (1998) Endoscopic sphincterotomy: long-term results in 408 patients with complete follow-up. Endoscopy 20:13-17.
12. National Institutes of Health Consensus Development Conference Statement on Gallstones and Laparoscopic Cholecystectomy (1993) Amer J Surg 165:390-396.
13. Paganni, AM, Feliciotti, F, Guerrieri, M, et al. (2001) Laparoscopic Common Bile Duct Exploration. Journal of Laparoendoscopic & Advanced Surgical Techniques 11:391-400.
14. Petelin, JB. (1993) Clinical Results of Common Bile Duct Exploration. End Surg 1:125-129.
15. Poon R, Yeung C, Liu C, Yuen W, Lo C, Tang A. Fan S. (2003) Intravenous bolus somatostatin after diagnostic cholangio-pancreatography reduces the incidence of pancreatitis associated with therapeutic endoscopic retrograde cholangio-pancreatography procedures: a randomized controlled trial. Gut 52:1768-1773.
16. Rhodes M, Sussman L, Cohen L, Louis MP. (1998) Randomized trial of laparoscopic exploration of common bile duct versus postoperative endoscopic retrograde cholangiography for common bile duct stones. Lancet 351:159-61.
17. Ricardi R, Islam S, Canete JJ, Arcand PL, Stoker ME. (2003) Effectiveness and long-term results of laparoscopic common bile duct exploration. Surg Endosc 17:19-22.
18. Stoker ME. (1995) Common bile duct exploration in the era of laparoscopic surgery. Arch Surg 130:265-269.
19. Strong R. Late bile duct cancer complicating biliary-enteric anastomosis for benign disease. (1999) Amer J Surg 177:472-474.
20. Tanaka M, Takahata S, Konomi H, et al. (1998) Long-term consequence of endoscopic sphincterotomy for bile duct stones. Gastrointest Endosc 48:465-9.
21. Vaira D, Ainley C, Willimas S, et al. (1989) Endoscopic Sphincterotomy in 1000 Patients. Lancet ii:431-34.