Totally Laparoscopic Radical Cholecystectomy for Gallbladder Cancer: A Single Center Experience

Introduction

The diffusion of abdominal ultrasound associated with laparoscopic cholecystectomy has led to the detection of gallbladder cancer (GBC) at an early stage with a potential better prognosis compared with advanced cases.

In Western countries, patients with an incidental finding range between 2.25% and 3%.^1,2 In many cases, pathological diagnosis of malignancy occurred during routine cholecystectomy or postoperatively.^1,2

The incidence of incidental gallbladder cancer (IGBC), discovered after histopathological examination of the gallbladder's specimen, ranges between 0.025% and 0.030%. ³

As described for the first time by Glenn and Hays, ⁴ the only potentially curative treatment for GBC is radical resection, a surgical procedure that combined en bloc resection of the gallbladder with hepatic resection and regional lymphadenectomy.

Cholecystectomy alone is a right choice only for the earlier stages Tis (tumor in situ) and T1a (tumor invades lamina propria) although some authors reported lymph node (LN) involvement in ∼2.5% of T1a tumors. ⁵ Starting from T1b tumor stage, reoperation is necessary. ⁶

Today laparoscopy has proven to be of benefit for patients with liver cancers in terms of short- and long-term outcomes. ⁷ However, a primary laparoscopic approach for the treatment of cancers of the biliary is not popular in the surgical community.^8–12

The aim of this study is to report the short-term data of patients who underwent a totally laparoscopic radical cholecystectomy for GBC at a single center of specialized hepatobiliary surgery. We also compared two groups of patients: patients with primary GBC discovered during preoperative diagnostic examination and patients with GBC diagnosed histopathologically after simple laparoscopic cholecystectomy.

Materials and Methods

From November 2016 to January 2019, we routinely performed a laparoscopic approach for patients with primary GBC (diagnosed preoperatively) and patients with IGBC (discovered after cholecystectomy).

All patients are subjected to laparoscopic staging with both optic and ultrasound exploration to exclude unresectable conditions as peritoneal dissemination or liver metastases. In case of liver invasion, the laparoscopic procedure is not converted to laparotomy. We converted the procedure only if there was a neoplastic involvement of the cystic duct or macroscopic evidence of vascular invasion.

Patients with primary GBC or IGBC underwent, respectively, “en bloc” resection of the gallbladder together with segments IVb and V and subsegmental or wedge resection of bed of the gallbladder.

All patients underwent laparoscopic lymphadenectomy of the hepatoduodenal ligament with clearance of LNs around the common hepatic artery and around the posterosuperior region of the pancreas head.

Surgical technique is standardized for all cases. The patient is placed in supine reverse Trendelenburg's position with the possibility to tilt during operation.

Both arms are abducted at 90° angle to the body, both legs are kept apart, slightly bent, and abducted for providing stability to the patient. All patient's pressure areas are protected by soft devices.

The operating surgeon stands between the patient's legs, while the first and the second assistant surgeons are, respectively, in the patient's left and right positions.

Routinely we use 4k magnified ultra high-definition (UHD) video and laparoscopic ultrasound (LUS) for optimal treatment.

The first 10 mm optical port was placed by open techniques on the right mid-clavicular line and above the transverse umbilical line. After a panoramic exploration of the abdominal cavity, the other operating ports were placed under vision. Two trocars of 10–11 mm were inserted, respectively, on the right anterior axillary line along the same transverse line of the first trocar and on the midline above the umbilicus. Another two ports of 5 mm were placed in the left hypochondrium and in the epigastric.

Lymphadenectomy was performed by energy devices. To control the hepatic pedicle, we performed an extracorporeal Pringle maneuver. We opened the pars flaccid of the gastrohepatic ligament. A laparoscopic tool was passed behind the hepatic pedicle through the foramen of Winslow and a cotton tape was placed. Then, the ends of the cotton tape were pulled out through an operating parietal port, placed in the left flank, with the help of a grasper put inside a 32-F thoracic tube. The tourniquet is pushed inside the abdominal cavity up to the level of the hepatic pedicle, whereas the external end of the cotton tape remains outside of the patient.

For liver transection, we used laparoscopic ultrasonic dissectors and bipolar diathermy, associated with saline pulses and aspiration.

Results

Our retrospective study included 18 patients who underwent surgery with the purpose to perform a radical laparoscopic cholecystectomy for primary GBC (38.9%) or to treat incidental GBC (61.1%). Patients' characteristics are shown in Table 1. The median patient's age was 70 years with female predominance (83.3%).

Most patients had preoperative American Society of Anesthesiologists (ASA) Score 2 (77.8%), which assesses the physical status before surgery, and about 16% of them had positive personal oncological anamnesis. Conversion rate from laparoscopy to laparotomy was 28.6% and 9.1%, respectively, for primary GBC and incidental GBC, but this difference was not statistically significant (P = .28). Among IGBC conversion from laparoscopic to laparotomy, surgery occurred in only 1 patient and was due to the need to perform a partial resection of the duodenal bulb for neoplastic infiltration. Among nonincidental gallbladder cancer (NIGBC) cases, we converted the procedure in only 2 patients to perform a more extensive lymphadenectomy (Table 2). The primary cause of unresectability was peritoneal carcinomatosis, which was discovered at the time of preliminary exploration in 2 patients with primary GBC (28.6%). The rate of macroscopic residual tumor resection (R2 resection) was 9.1% for IGBC and 14.3% for NIGBC, due to the presence of macrovascular infiltration (involvement of the right hepatic artery, hepatic hilum, or celiac tripod, Tables 3 and 4). Only 3 patients had liver recurrence (27.3%) and 1 patient had liver invasion (14.3%).

According to tumor staging of the eighth edition of the American Joint Committee on Cancer (AJCC), our cases were divided into pT2a (33.3%), pT2b (33.3%), pT3 (16.7%), and pT4 (167%).

Pathological features of the tumors were stratified according to the two groups of patients, a more advance T category and TNM stage were presented in the preoperative suspicion cases (T3−T4 18.2% versus 57.1%, P = .06, stage IVA-B 9.1% versus 71.4%, P = .017). There were also a differences in terms of perineural, microvascular infiltration (IGBC 9.1% versus NIGBC 71.4%, P = .73) and macrovascular infiltration (IGBC 9.1% versus NIGBC 28.6%, P = .28) in the two groups (Table 5).

LNs metastases were present in 5 patients (33.3%). Regional lymphadenectomy was performed in 15 patients, in 73.3% the total number of lymph nodes (total LNs) retrieved was more of 7 (7–12 LNs in 66.7% of patients and >12 LNs in 6.6% of patients) (Table 6).

According to LNs ratio (LNR: total number of positive LNs/total number of LNs retrieved), there were 10 patients without positive LNs (LNR = 0, 66.7%), 2 patients with LNR cutoff value of 0.25 (13.3%), and 3 patients with an LNR >0.25 (20%) (Table 6). Distribution of total LNs and LN+ retrieved, and of the number of patients with LN+ in each station, according the Japanese Society of Biliary Surgery (JSBS) classification, was analyzed and is given in Table 7. The majority of patients presented LNs involvement of the hepatoduodenal ligament (station 12–8, 46.5%). We observed also that LN station number 8 was the most frequently involved (20%) and in 3 of 5 patients with LN+ of another station (5–9–13–14–16) had also involvement of station 8.

The mean postoperative long stay was 8 days, excluding cases who developed complication. None of the patients required blood transfusion intraoperatively or during the perioperative period, and most patients had drains removed by postoperative day 3. Postoperative complications were classified according the Clavien–Dindo classification (Table 8). Grades I–II complications occurred in 4 patients, 1 patient developed fever and was treated only with antibiotics, 2 patients had delayed canalization, finally 1 patient developed pulmonary edema who was successfully treated with diuretic therapy. Two patients developed bile leak, first patient came back to emergency services after 5 days of discharge with abdominal pain and fever. Ultrasound showed a biloma in pelvis, which was drained by placing pigtail catheter percutaneously. Patients underwent MR (magnetic resonance) of the abdomen with hepatobiliary contrast which showed a biliary leak from the hepatic parenchyma. For high bilious output of the bile leak, an endoscopic retrograde cholangiography (ERCP) and stenting of the main duct were performed. The postprocedure course was complicated by duodenal perforation, which required urgent surgery with duodenal raffia. The subsequent postoperative course is without complications. The second patient underwent laparoscopic drainage of a biloma in the right subhepatic space. MR of the abdomen with hepatobiliary contrast showed a biliary leak from the hepatic parenchyma. Patients underwent ERCP and stenting of the common bile duct. After this procedure, the patients developed septic shock, and an urgent laparotomy showed the presence of an ischemia of the common bile duct with multiple continuous lesions. Surgeons performed common bile duct resection with Roux-en Y hepatic jejunostomy reconstruction. The subsequent postoperative course was long but without surgical complications.

Discussion

In the past decades, we have witnessed the technical progress and diffusion of the minimally invasive liver surgery worldwide.^13–15 Many reports had proved the safety and the feasibility of laparoscopic surgery for primary or metastatic liver tumors, such as hepatocellular carcinoma, intrahepatic colangiocarcinoma, and colorectal liver metastases.^13–15 For these parenchymatous liver cancers, two fundamental steps should be fulfilled during hepatic resection: to detect the tumor inside the liver and to perform a correct parenchymal transaction to obtain an R0 resection. For the last one, laparoscopic approach offers some benefits over conventional operations. The magnified view of the operating field allows perfect hemostasis during parenchymal transaction.

Laparoscopy has also proven to be of benefit to the open approach in terms of short-term outcomes, reducing morbidity, and length of hospital stay, with comparable long-term oncologic outcomes. ⁷ For some hepatic resections, such as left lateral sectionectomy, laparoscopy has been proposed as the standard technique. ¹⁶

However, for the treatment of cancers of the biliary tract (Klatskin's tumor and GBC), a primary laparoscopic approach is still rejected by the surgical community.^8–12,17 In the past, the reasons were due to the oncological concerns such as intraoperative peritoneal dissemination and possible port site recurrences. Today the real concerns are correlated with the technical difficulties involved in performing a total laparoscopic procedure.

We can say that preliminary laparoscopic staging associated with intraoperative LUS is a fundamental step for the treatment of the Klatskin's tumor and primary GBC. Staging laparoscopy may detect small liver or peritoneal metastases that are undetectable preoperatively. In our case series, among primary GBCs, we discovered 2 patients (28.6%) with peritoneal carcinomatosis at the preliminary exploration. On the other hand, we believe that staging laparoscopy should not be considered as a primary modality for the detection of locally advanced disease. The judgment of resectability is derived from vascular and biliary tract's infiltration, which requires an operative dissection of the hepatic hilum and a frozen examination of the pathological tissue, such as the sampling of the cystic duct stump in the cases of IGBCs. ¹⁸

In total, laparoscopic approach for cancers of the biliary involves several complex procedures, such as laparoscopic hemihepatectomy, laparoscopic hepatoduodenal lymphadenectomy, and bilioenteric anastomosis. ¹⁹

For all these reasons, today only a few case series have shown the feasibility of laparoscopic resection for only primary GBC or IGBC at an early stage without liver involvement (T1b-T2).^8–12 However, in our case series, we also treated T3 GBCs (16.7% of patients), among them only 3 patients had liver recurrence (27.3%) and 1 patient had liver invasion (14.3%) at the reoperation time. We believe that the need to perform wedge gallbladder bed or hepatic resection of segments IVb and V cannot be considered a real limitation for the laparoscopic approach. In contrast, the need to perform bile duct resection and biliary tract reconstruction may be a real technical difficulty.

Only few authors reported a full laparoscopic approach,^8,10 among those Gumbs and Hoffman ⁸ reported a choledochojejunostomy anastomosed with a single layer of laparoscopic suture.

For the treatment of incidental cases, there are also technical difficulties involved in performing a second laparoscopic procedure in patients who already underwent cholecystectomy for gallstones or their complications as acute cholecystitis. ²⁰ These patients can present postoperative adhesions, which increases operative time of surgery and the rate of conversion. In a few patients with massive adhesions, it may be impossible to introduce the first trocar (trocar of Hasson) according to the open technique for the CO₂ inssuflection.

When we perform the lysis of adhesions, there is an increased risk of intraoperative complications such as bowel injury or bleeding of the liver capsule.

However, in our case series, previous abdominal surgery did not represent a risk factor which led to converting the procedure from laparoscopy to laparotomy for both groups of patients (IGBC and NIGBC).

Patients with primary GBC compared with patients with IGBC presented highest conversion rate from laparoscopy to laparotomy (28.6% versus 12.5%).

The reasons that led to convert the laparoscopic procedure were due to oncological concerns such as the need to perform a more extensive lymphadenectomy or to perform an additional resection for neoplastic infiltration of a contiguous organ.

Laparoscopic LNs dissection is a very difficult procedure. Routinely we use 4k magnified UHD video that helped us to perform a meticulous LNs clearance of the hepatoduodenal ligament.

There is no consensus about the correct lymphadenectomy extension. According to the recent guidelines of the AJCC (eighth edition), the N category is redefined by the number of metastatic LNs instead of the location of these LNs. ²¹ For a correct categorization of the N stage, at least six LNs must be sampled. ²¹ In our case series, the total LNs retrieved was more than 7 in 73.3% (7–12 LNs in 66.7% and >12 LNs in 6.6%).

The majority of patients presented LNs involvement of the hepatoduodenal ligament (station 12–8, 46.5%). We also observed that LN station number 8 was the most frequently involved (20%) and in 3 of 5 patients with LN+ of another station (5–9–13–14–16) had also involvement of station 8.

Conclusions

Laparoscopy can be considered a safe treatment for IGBC or primary GBC with no statistical difference in terms of conversion rate between the two groups. The T3 stage with only liver involvement was not a contraindication. The real reasons that led to convert the laparoscopic procedure were due to oncological concerns, unrelated to the liver infiltration, such as the need to perform a more extensive lymphadenectomy or to perform an additional resection for neoplastic infiltration of a contiguous organ.