The safety and feasibility of laparoscopic common bile duct exploration for treatment patients with previous abdominal surgery

The purpose of this study was to evaluate the safety and feasibility of laparoscopic common bile duct exploration (LCBDE) in patients with previous abdominal surgery (PAS). The outcomes were compared in 139 patients (103 upper and 36 lower abdominal surgeries) with PAS and 361 without PAS who underwent LCBDE. The operative time, hospital stay, rate of open conversion, postoperative complications, duct clearance, and blood loss were compared. Patients with PAS had longer operative times (P = 0.006), higher hospital costs (P = 0.043), and a higher incidence of wound complications (P = 0.011) than those without PAS. However, there were no statistically significant in the open conversion rate, blood loss, hospital stay, bile leakage, biliary strictures, residual stones, and mortality between patients with and without PAS (P > 0.05). Moreover, compared with those without PAS, patients with previous upper abdominal surgery (PUAS) had longer operative times (P = 0.005), higher hospital costs (P = 0.030), and a higher open conversion rate (P = 0.043), but patients with previous lower abdominal surgery (PLAS) had a higher incidence of wound complications (P  = 0.022). LCBDE is considered safe and feasible for patients with PAS, including those with PUAS.

Inclusion criteria. 1. No intrahepatic bile duct stricture identified before surgery; 2. all patients diagnosed using abdominal ultrasound, CT, endoscopic retrograde cholangiopancreatography, or magnetic resonance cholangiopancreatography; 3. diameter of common bile duct (CBD) ≥ 8 mm; 4. biliary tract malformation and tumor were excluded; and 5. no cardiopulmonary disorder that would contraindicate surgery. LCBDE procedure. Patients were placed supine and in lateral position at working height. Under general anesthesia, we used three-trocar or four-trocar techniques for the working port and choledochoscope. First, an infraumbilical trocar was inserted when the patient had no previous operative scar at the umbilicus. When the patient had an operative scar at the umbilicus, the first trocar site should be as far away as possible from the GroupA (n = 327) GroupB (n = 121) P  After carbon dioxide pneumoperitoneum was created at 12 mmHg, a 30° oblique laparoscope was placed. Then, right-side 5-mm ports were placed for a coagulation hook or ultrasonic knife using blunt or sharp separation of abdominal adhesions. When the hepatoduodenal ligament was completely exposed, the CBD was carefully assessed; the anterior duct wall was incised using electrocoagulation, followed by placement of a 12# ventricular drainage tube. After saline lavage, the stones were removed with a choledochoscope and basket. After the stones were removed, a T tube was placed or the CBD underwent primary closure with barbed wire suture.
Statistical Analysis. Statistical analysis was performed with SPSS software, version 17.0 (SPSS Inc., Chicago, IL, USA). Continuous variables were expressed as mean ± SD or median (range), and categorical variables were expressed as numbers. Continuous variables were compared using Student's t-test test or the Mann-Whitney U-test, and categorical variables were compared using the χ 2 test or Fisher's exact test. P < 0.05 was considered statistically significant.  Table 1. There were no significant between-group differences in sex, age, and liver function (P > 0.05).

Results
Perioperative outcome. The surgical outcomes are shown in Table 1. The patients with PAS had longer operative times (P = 0.006), higher hospital costs (P = 0.043), and a higher incidence of wound complications (P = 0.011) than those without PAS. However, there were no significant differences in conversion rate, blood loss, hospital stay, bile leakage, biliary strictures, wound complications, residual stones, and mortality between the two groups (P > 0.05). One patient died of septic shock, but the association with the procedure was unclear. In addition, there were no significant differences in the white blood cell (WBC) count, aspartate aminotransferase (AST), alanine aminotransferase (ALT), and gamma-glutamyltransferase (GGT) (P > 0.05). These suggest that there was no difference in postoperative patient emergency status. The 139 patients with PAS were divided into 103 with PUAS and 36 with PLAS. As shown in Table 3, patients with PUAS had longer operative times (P = 0.005), higher hospital costs (P = 0.030), and a higher open conversion rate (P = 0.043), but patients with PLAS had a higher incidence of wound complications (P = 0.022).
Patients were further grouped by primary closure or T-tube placement. In primary closure cases (Table 4), the blood loss (P = 0.031) of those with PUAS was more than in those without PUAS. However, in T-tube cases (Table 5), the mean operative time (233.9 ± 72.6 min) in those with PUAS was longer than that in those (195.2 ± 62.1 min) without PUAS (P = 0.007).

Discussion
Conventional open surgery increases patient suffering, and also increases the economic burden. In addition, although endoscopic sphincterotomy (EST) can be used for treatment of primary and recurrent CBD stones 14 , it is associated with high complication and failure rates 15 . EST can lead to postoperative complications such as pancreatitis, perforation, blood loss, sepsis, and even death; EST can also cause disruption of the sphincter of Oddi, thus causing injury to the physiological barrier that prevents cholangitis due to duodenobiliary reflux 16,17 . Natsui et al. 18 reported that the incidence of biliary bacterial contamination 30 months after EST was about 78%.
Another study reported that the postoperative acute cholangitis rate was 2.4-10.3% for EST 19 . Because of the many drawbacks of EST and open surgery, LCBDE is readily accepted by the majority of patients because of the small surgical wound, less pain, rapid postoperative recovery, and fewer complications 4-6 . In the early laparoscopic era, PAS-especially biliary surgery-was considered a contraindication for laparoscopic surgery due to abdominal adhesions [20][21][22] . With rapid advances in laparoscopic equipment and technique, as well as the continued improvement in surgical skills, patients with PAS are no longer an absolute contraindication to laparoscopic   Table 5. Characteristics and perioperative outcomes of T-tube patients. GroupA, without previous abdominal surgery; GroupB, with previous abdominal surgery; WBC, blood cell count; AST, aspartate aminotransferase; ALT, alanine aminotransferase; GGT, gamma-glutamyltransferase. WanRMB, ten thousand renminbi.
surgery [10][11][12][13] . LCBDE has been performed in patients with PAS, with good clinical results, but few reports have been published. This study was carryed to evaluate the safety and feasibility of LCBDE in patients with PAS. We found that patients with PAS had longer operative times and higher hospital costs than those without PAS. These results are understandable, because additional time is needed to dissect adhesions, and more instruments and medications are needed, increasing hospital costs 23 . We divided the patients with PAS into those with PUAS and those with PLAS. We found that patients with PUAS had longer operative times and higher hospital costs than those without PAS and PLAS; however, there was no difference between patients with PLAS and those without PAS. These results are in agreement with those of Karayiannakis et al. 11 , who reported that adhesions in patients with PUAS occurred more frequently, and were more extensive and denser than those in patients with PLAS; additionally, they found that the requirement for adhesiolysis performed by laproscopy was higher for patients with upper abdominal incisions than those with lower abdominal incisions. And the operating area of LCBDE is mainly concentrated in the upper abdominal. Therefore, it was not surprising for patients with PUAS to have higher open conversion rates than patients without PAS or patients with PLAS, as seen in our study. The high conversion rate was due to bowel injury and uncontrolled bleeding 23,24 , which helped obtain adequate exposure of the critical region of interest. Previous studies 25 have also shown that the liver capsule and surgical wounds bleed easily during adhesion separation, and these factors may result in increased blood loss and a suboptimal operative field, increasing the risk of intraoperative complications. For biliary surgery, the incidence of biliary stricture, bile leakage, and residual stones were key indicators in the safety evaluation of biliary surgery 26 . Our results showed that there were no significant differences in the perioperative results between patients with and without PAS with respect to hospital stay, blood loss, and postoperative complications, which included bile leakage, biliary stricture, residual stones, and mortality; in fact, a higher complication rate was initially expected. This shows that LCBDE with PAS did not increase the amount of bleeding and postoperative complications. In addition, there were also no significant differences in the WBC, AST, ALT, and GGT levels. These indicate that there was no difference in postoperative patient emergency status. Therefore, there was no greater systemic effect with LCBDE in patients with PAS, LCBDE is safe and feasible for patients with PAS. As for the patients with PLAS had a higher incidence of wound complications than those without PAS or patients with PUAS. The possible reason for this is that the patients with PLAS were mainly female and subcutaneous fat is more abundant among females.
For LCBDE with PUAS, the most critical technical difficulty was the establishment of pneumoperitoneum and separation of abdominal adhesions [7][8][9]27 . Therefore, the approach to the abdomen must abide by the strict application of technical principles. The first trocar (umbilical observation port) site should be as far away as possible from the original surgical incision; if necessary, pneumoperitoneum should be established using the Hasson technique. That can effectively avoid bowel injury. Once the pneumoperitoneum is established, we can use direct vision and place the second trocar in the right anterior axillary line, where there are usually fewer adhesions. If the anatomy around the hilum has been destroyed by PAS, there is an increase in potential risks of surgery. Hence, it is particularly important to understand the anatomical rules of biliary surgery with PAS. The extent of intra-abdominal adhesions depends on the original incision, with adhesions being more extensive with an oblique incision under the right costal margin. We found that there were different degrees of adhesions: 1. adhesions between the liver diaphragm surface and abdominal wall; 2. the gastric antrum and omentum were adherent to the abdominal wall or were adherent to the right side of the round ligament of the liver; 3. the first and second parts of the duodenum shifted upward to block the porta hepatis; 4. the colon near the liver was moved up, leading to the disappearance of the lacunar space in the right inferior liver. Therefore, we try to separate the gastric antrum and omentum with an ultrasonic knife at the second trocar port in the right anterior axillary line, after which we can establish another port under direct vision. Then, we carefully dissect along the liver surface to restore the normal structure of the gastric pylorus, duodenum, and hepatoduodenal ligament. The duodenum with adhesions in the hilum is used as an anatomical "landmark". The CBD is usually located deep in the duodenum; when the duodenum is dissected downward from the hilum, the CBD will be exposed. This method is called the "anterior approach to the hepatoduodenal ligament". When the adhesions are difficult to dissect with this approach, we can dissect downward from the hepatic flexure along the right lateral side of the hepatoduodenal ligament to expose the right lower space of the liver and foramen of Winslow hole, to further reveal the right side of the CBD. In this way, we can determine the location of the CBD. This method is called the "right-side approach to the hepatoduodenal ligament". If the CBD is difficult to identify, puncture with a scalpel or needle will confirm the CBD. If the gallbladder area cannot be accessed because of adhesions or severe bleeding, we switch to laparotomy.
Regarding the T-tube placement or primary closure of the CBD, there was no difference in the residual stone rate between the two groups in our study and even research demonstrate that primary duct closure after LCBDE is feasible and fewer complications than T-tube placement 28 , but the rate can reach up to 5%. Therefore, we believe that primary closure of the CBD must be performed with caution and must meet the following criteria: 1. CBD diameter > 8 mm, with few stones; 2. results of intraoperative exploration should coincide with those of preoperative evaluation; 3. stones should be completely removed, and intraoperative exploration should only find slight CBD wall inflammation and edema; 4. the CBD and duodenum are patent; 5. the surgeon is skilled at suture technique.
In conclusion, this study demonstrates that LCBDE for patients with a history of abdominal surgery is feasible and safe. Therefore, when LCBDE is planned for patients with a surgical history, the laparoscopic approach can be considered a good alternative.