Current Role of Minimally Invasive Radical Cholecystectomy for Gallbladder Cancer

Background. For Tis and T1a gallbladder cancer (GbC), laparoscopic cholecystectomy can provide similar survival outcomes compared to open cholecystectomy. However, for patients affected by resectable T1b or more advanced GbC, open approach radical cholecystectomy (RC), consisting in gallbladder liver bed resection or segment 4b-5 bisegmentectomy, with locoregional lymphadenectomy, is considered the gold standard while minimally invasive RC (MiRC) is skeptically considered. Aim. To analyze current literature on perioperative and oncologic outcomes of MiRC for patients affected by GbC. Methods. A Medline review of published articles until June 2016 concerning MiRC for GbC was performed. Results. Data relevant for this review were presented in 13 articles, including 152 patients undergoing an attempt of MiRC for GbC. No randomized clinical trial was found. The approach was laparoscopic in 147 patients and robotic in five. Conversion was required in 15 (10%) patients. Postoperative complications rate was 10% with no mortality. Long-term survival outcomes were reported by 11 studies, two of them showing similar oncologic results when comparing MiRC with matched open RC. Conclusions. Although randomized clinical trials are still lacking and only descriptive studies reporting on limited number of patients are available, current literature seems suggesting that when performed at highly specialized centers, MiRC for GbC is safe and feasible and has oncologic outcomes comparable to open RC.


Introduction
The role of laparoscopic surgery in the management of digestive tract tumor is increasingly accepted worldwide [1,2]. However, although laparoscopic cholecystectomy (LC) began the era of laparoscopic surgery and is one of the most frequently performed mini-invasive procedures, the use of laparoscopic surgery is skeptically considered in the management of gallbladder cancer (GbC) [3]. GbC represents the most aggressive malignancy of the biliary tract and is characterized by an extremely poor prognosis. While increasing evidence shows that, for Tis and T1a GbC with clear margins and unbroken gallbladder, simple cholecystectomy, either laparoscopic or open, can be curative [4][5][6][7], for patients affected by resectable T1b or more advanced GbC, radical cholecystectomy (RC), consisting in liver resection (liver bed resection or segment 4b-5 bisegmentectomy) with locoregional lymphadenectomy, is the only available treatment to positively affect the prognosis [8][9][10][11][12].
Minimally invasive RC (MiRC) is skeptically considered by the majority of HPB surgeons, mainly due to the fear of tumor dissemination during laparoscopy, to difficulty in achieving adequate lymphadenectomy, and to complexity of laparoscopic liver resection.
The aim of this study is to review the available literature on feasibility and postoperative and oncologic outcomes of MiRC for patients affected by T1b or more advanced GbC.

Materials and Methods
Clinical case studies reporting on patients affected by GbC who underwent MiRC, meaning that laparoscopic or robotassisted approach was used for both liver resection and locoregional lymph nodes excision, were included in the  [13,14] for all years to June 2016 (last PubMed search was performed on June 10, 2016). Initially, searches employing MeSH terms were performed for keywords and text (title or abstract). As shown in Table 1, search terms were organized in three main groups (search number: 3, 11, and 12), which were further combined with each other finally resulting in the identification of 242 manuscripts. In addition, a "manual" research using the "related articles" function was used in order to "explode" research, and results were supplemented by further searches of reference lists of other articles, resulting in the identification of additional 24 manuscripts. Titles, abstracts, and full texts of resulting 266 manuscripts were independently reviewed by two authors (GZ and AM) to assess whether the studies met the eligibility criteria. Contrasting results between GZ and AM were discussed case by case, until an agreement was found. Included articles could be classified in case reports and series of MiRC for GbC and casecontrol studies comparing results of MiRC versus open radical cholecystectomy. An intention-to-treat analysis was performed; consequently, cases converted to open procedures were included in the analysis.
Of 168 patients included in the 13 aforementioned studies, 16 underwent a LC according to study protocol [22,27]. Of the remaining 152 patients who underwent an attempt of MiRC, minimally invasive approach was laparoscopic in 147 patients and robotic in 5 patients. Overall, 15 (10%) patients were converted to an open procedure according to study protocol in one case [17], due to postcholecystectomy adhesions which made peritoneal laparoscopic exploration unfeasible in 11 cases [17], due to intraoperative portal bleeding in one case [21], and due to intraoperatively detected persisting bile leak from the liver bed in the remaining case [17]. MiRC was attempted as an elective procedure for a preoperative suspicion of GbC in 110 patients and as a completion procedure following diagnosis of GbC on a cholecystectomy specimen in the remaining 42 patients. MiRC included at least a resection of the liver bed in all but 5 patients who, according to the corresponding study protocol [23], underwent simple cholecystectomy because the GbC was located on the peritoneal side of the gallbladder. Liver resection consisted in liver bed resection in 98 patients (with a liver bed thickness ranging between 2 mm [22] and 3 to 5 cm [18,19]), segment 4b-5 resection in 49 patients [15,24,26], and extended right hepatectomy in 2 patients [21]. Laparoscopic intraoperative ultrasonography was performed during MiRC in seven studies. Locoregional nodal excision was performed in all patients and represented the initial resective procedure in 7 studies. Port site excision was performed in 8 out of 42 (19%) patients who underwent completion MiRC (Table 3).
Mean intraoperative blood loss was 150 cl (range: 0-1500 cl); mean operation duration was 235 minutes (89-490 min). Overall, 15 patients (10%) out of 152 who underwent an attempt of MiRC experienced postoperative morbidity, while postoperative mortality was nil. Following MiRC, mean length of hospital stay was 5 (2 to 19) days. Final pathology data were available for 144 patients and revealed T0-1a, T1b, T2, and T3 GbC in 9, 36, 81, and 18 cases, respectively. Mean retrieved lymph node number ranged between three and 13; lymph node status was N0 in 115 patients, N1 in 21, and NX in the remaining 8. R0 resection was obtained in all patients undergoing MiRC. Long-term survival outcomes were reported in 11 studies: after a mean follow-up duration ranging between 11 and 84 months, 14 patients experienced disease recurrence, whose location was specified in 10 cases. No port site recurrence was observed during follow-up (Table 4).

Discussion
In the current review, we retrospectively analyzed perioperative and oncologic outcomes of 152  [7]. In contrast, for patients affected by T1b or higher stage tumor necessitating RC, a minimally invasive approach is questioned by majority of HPB surgeons.
∧ Liver bed resection not performed if tumor located on the peritoneal side of the gallbladder.  about a cautionary note about the use of laparoscopy in patients in whom GbC was suspected, making GbC a formal contraindication for laparoscopy. Additional reports correlated the occurrence of port sites and peritoneal implantation to the association of CO 2 pneumoperitoneum effect and imprecise handling of gallbladder during laparoscopy leading to accidental perforation of gall bladder [36,[40][41][42]. Port site and peritoneal recurrence may occur through direct and indirect implantation of tumor cells, during the laparoscopic procedure [43]. Direct implantation may depend on seeding of exfoliated malignant cells during tumor extraction without a protective bag or on contact with instruments contaminated with tumor cells. Indirect contamination may be related to pneumoperitoneum, based on an "aerosol" effect with dissemination of exfoliated tumor cells to the port sites during the turbulence of insufflation or to a "chimney" effect with tumor cells wound implantation during desufflation [44]. However, increasing evidence highlights the role of gentle manipulation of gallbladder and of the use of plastic bag for specimen extraction in reducing the rate of port site and peritoneal tumor implantation [23,24,40], while pneumoperitoneum per se seems to have a limited role in the development of such complications. Consistent with this hypothesis, we currently found that, among 10 patients whose disease recurrence following MiRC was specified in detail, no peritoneal or port site recurrence occurred, further supporting the hypothesis that laparoscopic approach is not directly responsible for increasing the risk of peritoneal and port site dissemination, provided that gallbladder wall is not damaged and GbC is not exposed during MiRC. RC is a complex procedure, consisting in liver bed or more extended anatomic liver resection, associated with hepatic pedicle lymph node dissection, eventually extended to peripancreatic and para-aortic lymph nodes, and, eventually, in selected cases, to common bile duct resection and reconstruction. Intrinsic technical difficulty of performing such procedures by minimally invasive approach represents an additional factor slowering MiRC acceptance. Concerning liver resection, the technique of laparoscopic liver resection has already been established [45], and radical laparoscopic surgeries for liver diseases have been demonstrated to show outcomes equal to those of open surgeries [46,47]. Although the thickness of liver parenchyma to resect during RC remains a matter of debate, current evidence supports safety and feasibility of a minimally invasive approach for liver bed resection, as well as for segmentectomy 4b-5.
GbC has a high tendency to lymphatic invasion; thus, an adequate lymphadenectomy, commonly identified as the retrieval of at least six lymph nodes, is required to obtain a proper tumor staging [10,38,48]. Hepatic portal pedicle is a complex structure, containing important and delicate elements whose damage during lymphadenectomy may result in uncontrollable bleeding or injury to bile duct. This has brought about question of safety and adequacy of laparoscopic lymphadenectomy for GbC, representing an obstacle to the advancement of MiRC for GbC. However, current evidence suggests that laparoscopic lymphadenectomy may yield outcomes similar to those following open approach RC, with a mean number of dissected lymph nodes ranging between three [18,19] and 13 [23,27] among 13 studies analyzed in the current review. In addition, two studies [26,27] comparing results of MiRC and of open RC did not show statistical difference between two approaches concerning the mean number of lymph nodes retrieved.
Finally, concerning bile duct resection and reconstruction, this can be indicated during RC only if the cystic duct margin is positive and although in some of the analyzed studies such situation represented an indication to open RC [22,27], it does not represent an absolute contraindication to MiRC. Two of the 13 studies reviewed in the current analysis report on overall 3 patients who underwent a MiRC associated with common bile duct resection and biliojejunal reconstruction, none of them developing postoperative complications related to the procedure.

Conclusions
In conclusion, available data on MiRC for GbC show encouraging results in terms of perioperative and oncologic outcomes. However, the limited number of descriptive studies reporting on patients undergoing MiRC, as well as the absence of randomized clinical trials comparing MiRC and open RC, does not allow for recommending MiRC outside highly specialized centers with adequate experience in hepatobiliary and laparoscopic surgery.