Laparoscopic Common Bile Duct Exploration After Unsuccessful Endoscopic Stone Removal

Introduction

Since the introduction of endoscopic retrograde cholangiopancreatography (ERCP) and endoscopic sphincterotomy by gastroenterologists from Germany and Japan in the 1970s, these methods have become the treatments of choice for managing choledocholithiasis. ERCP is a useful modality that combines endoscopic detection and treatment of common bile duct (CBD) stones,^1,2 but it is not indicated for all patients with respect to risk and potential complications, such as perforation, bleeding, and pancreatitis. Difficult cannulation of the ampulla is a well-known risk factor for post-ERCP pancreatitis or perforation. Although stone removal by endoscopy is superior to other options, the treatment can be difficult in some cases (i.e., large stones, history of gastric bypass surgery, duodenal diverticulum, and Mirizzi's syndrome). In these cases, CBD exploration is the obligatory procedure for surgeons, and most surgeons still choose the open technique. It has been suggested that laparoscopic CBD exploration (LCBDE) is difficult, risky, and time consuming. ³ However, the advantages of laparoscopic management have been demonstrated in many recent studies with large series.^3–5 When ERCP fails to remove CBD stones, it remains unclear whether laparoscopic treatment is an alternative or not. The purpose of this study is to investigate the result of LCBDE following unsuccessful CBD stone removal by endoscopy.

Subjects and Methods

Results

Of 82 patients, 44 in Group 1 underwent LCBDE after unsuccessful endoscopic CBD stone removal. Among these 44 patients, 24 experienced failed cannulation due to anatomical difficulty. Six had history of previous Billroth II gastrectomy, and 18 had a periampullary diverticulum. In these cases, routine magnetic retrograde cholangiopancreatography provided accurate diagnosis of CBD stone. Failure of stone extraction was noted for 20 patients. Among these patients, 12 underwent endoscopic lithotripsy, which was not successful because of multiple, large, or impacted stones. In 8, stone removal failed because of stones >15 mm. In these 20 cases of extraction failure, the endoscopist inserted an ENBD to decompress the bile duct and prevent cholangitis (Table 1).

There were 23 men and 21 women averaging 65.4 years of age in Group 1. Nine patients had a co-morbidity such as cerebrovascular accident or cardiopulmonary disease. There were no significant differences in demographics between the groups. Stone clearance was nearly complete in both groups after LCBDE. Patients in whom stone removal failed during surgery were included in each group, respectively: the failure of CBD stone removal by laparoscopy was managed by conversion to laparotomy in 1 case due to severe adhesion of the hepatoduodenal ligament and in 1 case in which intraoperative clearance of multiple large CBD stones by laparoscopy was not successful and remnant stone was detected by choledochoscopy. A T-tube was inserted in this patient, and conservative management was performed. The mean operative time was 182.6±1.9 minutes, and there was one conversion in Group 2.

Postoperative biliary decompression was maintained in every patient. In Group 1, a T-tube, ENBD, or internal stent was placed in 13, 20, and 11 patients, respectively. In Group 2, a T-tube or internal stent was placed in 14 and 13 patients, respectively (P<.001).

Regarding postoperative complications, 2 patients in Group 1 experienced bile leakage, and 1 developed pancreatitis. In Group 2, 1 patient had bile leakage, there was 1 case of ileus, and 2 episodes of pancreatitis occurred. All complications were treated conservatively and resolved within 5 days after surgery. Mean hospitalization days were 6.9±3.7 and 6.2±5.9, respectively, in the two groups. Recurrent CBD stones were found in 10 patients in whom mean time to recurrence was 45.6±13.2 months (range, 6–120 months). Among them, 9 patients underwent ERCP, and the CBD stone was successfully removed in 7. One remaining patient without ERCP and another with unsuccessful ERCP were observed because of advanced age and asymptomatic presentations. Additional procedures were needed in 1 patient with failed ERCP due to large stones. That patient was a 79-year-old man who had successful repeat LCBDE performed 4 years after the primary LCBDE. There was no significant difference between the groups for operative time, hospital days, conversion rate, recurrence rate, and complications. The perioperative data of the two groups are summarized in Table 2.

Discussion

In the event of failed CBD cannulation during ERCP, a reported 50%–60% of patients will have further therapeutic radiological, surgical, or repeat endoscopic procedures.^7,8 If further CBD visualization is warranted only for diagnosis after failed ERCP, radiologic approaches such as percutaneous cholangiography can be recommended. In cases of therapeutic CBD stone removal, the next step of therapy is repeat ERCP or surgery. One recent study has reported that lower-risk patients who failed ERCP without other complications can be successfully and safely managed by repeat ERCP if referred to a specialized, high-volume endoscopy unit. ⁹

If CBD stones are difficult to remove by ERCP, such as in patients with previous Billroth II gastrectomy, periampullary diverticulum, or a large stone, the surgeon is consulted. Institutional protocols for CBD stone management may vary depending on institutional experience and affinity for procedures. In our institution, LCBDE has been performed since the late 1990s by a single surgeon. Here, in patients with unsuccessful ERCP lithotripsy, LCBDE has been favored as the next therapeutic step.

Transductal choledochotomy is technically more complex than the transcystic approach. ¹⁰ However, in contrast to Western countries, in Korea, most CBD stones are multiple, large, and pigmented, making the transcystic method almost always impracticable. In an Indian study, ¹¹ recommendations for a CBD stone are similar to the Korean recommendations. Here, we perform routine transductal choledochotomy during LCBDE to maximize the safety and convenience of stone removal. Another reason that routine transductal choledochotomy is favored in our institution is that it facilitates routine stenting of the CBD for decompression, as well as allowing us to use a convenient suture device. Accordingly, we have had a low incidence of bile leakage after LCBDE in comparison with many previous reports. Kelly ¹² has mentioned that as long as the surgeon has mastered flexible endoscopy and laparoscopic suturing, there is nothing technically demanding about LCBDE. Choledochoscopy has various advantages over intraoperative cholangiography: there is no need for fluoroscopy, lithotripsy can be performed under direct vision, and the incidence of stone retention is low. ¹³

Many surgeons have advocated the advantages of choledochoscopy based on their experiences. Chander et al. ¹¹ preferred intraoperative choledochoscopy in all patients to visualize and remove the CBD stone under vision and again to check for the completeness of removal at the end of the procedure. Topal et al. ⁵ endorsed the superiority of choledochoscopy over intraoperative cholangiogram in regard to conversion rate and operative time. We support the advantage of intraoperative choledochoscopy based on our experiences. CBD visualization using choledochoscopy is routinely performed following Fogarty lithotripsy, making it possible to check for remnant stone. In this study, the clearance rate by LCBDE after failed ERCP is nearly complete. Regarding morbidity after surgery, high rates of pancreatitis (7.3%) and bile leakage (14.6%) following LCBDE with choledochotomy have been reported. ¹⁴ In our study, bile leakage occurred only in 3.7% of cases, and pancreatitis occurred in 3.7% of cases when we evaluated the safety and feasibility of LCBDE in patients with unsuccessful ERCP in comparison with patients in Group 2. Moreover, most published articles do not mention late complications, but there are a few reports that mention stone recurrence in 1.8%–6.7% of patients.^15,16 Our recurrent stone rate was 12.2%. The requirements to achieve a high success rate with LCBDE include adequate training and standardization of the surgical technique. ¹⁷ We propose that the higher success rate of stone clearance in our study is due to the standardization of the procedure. We performed transductal LCBDE using a Fogarty catheter, choledochoscopy, routine bile duct decompressive stenting, and a convenient intracorporeal suture device. It is well understood that LCBDE is not currently the generally conducted procedure for treatment of CBD stones because of its difficulty and the reliance on ERCP/endoscopic sphincterotomy. However, LCBDE can be a very safe and convenient procedure and treatment modality for CBD stones with unsuccessful removal by ERCP if the surgeon is skillful and the procedure is standardized, even in especially difficult cases such as large multiple stones or after repeated or various incomplete and unsuccessful endoscopic approaches for lithotripsy. Moreover, repeated aggressive endoscopic trials can increase cholangitis, pancreatitis, bleeding, and duodenal perforation rates.^18,19 In that situation, LCBDE is an acceptable and promising procedure for CBD stone removal.

In conclusion, our surgical outcomes between the two groups after LCBDE did not show any difference. On the basis of our surgical outcomes, we confirmed the efficacy and safety of LCBDE in both groups. The endoscopic removal remains the main treatment of CBD stones. LCBDE is an acceptable alternative treatment, even when ERCP is impossible or stone removal is incomplete.