Introduction
The laparoscopic approach for pancreatectomy is well reported. In the case of distal pancreatectomy, this method suggests clear advantages over the open approach. However, the development of this minimally invasive surgery technique remains slower than that in other gastrointestinal organs. The low acceptance of this procedure among surgeons is ascribed to various reasons, including limitations of laparoscopic surgery, such as limited degrees of freedom for manipulation, 2D imaging adaptation, and long learning curve required to master a new skill. Recently developed robotic surgical systems can overcome several of the abovementioned limitations and drawbacks regarding conventional laparoscopic approach. This feature may hasten minimally invasive surgery development of pancreatectomy. The robotic surgical system is also thought to be beneficial for spleen-preservation rate of improvement in laparoscopic distal pancreatectomy. However, few robot-assisted laparoscopic distal pancreatectomy procedures have been reported, and only some publications have compared robot-assisted laparoscopic distal pancreatectomy with conventional laparoscopic distal pancreatectomy [
1-
7]. Therefore, the current study compared the short-term outcomes between robot-assisted laparoscopic distal pancreatectomy and conventional laparoscopic distal pancreatectomy for benign and malignant pathologies.
Materials and methods
The study population comprised a consecutive series of patients who underwent minimally invasive approach of distal pancreatectomy in a tertiary referral center between July 1999 and January 2015. All data obtained in this study were retrospectively analyzed in a prospectively maintained database. All patients provided informed consent, and patient confidentiality was preserved.
A robot-assisted laparoscopic distal pancreatectomy program was initiated in our department in May 2009. The da Vinci
® S Surgical System (Intuitive Surgical, Inc., Sunnyvale, CA) was used for all robot-assisted procedures. Robotic procedures are performed by consultant surgeons who acquired a combination of advanced laparoscopic skills and extensive experience with open pancreatic surgery. All patients were informed about the detail and nature of the operation, and informed consent was obtained before surgery. The outcome measures included operating time, blood loss, length of hospital stay, and procedure-related morbidity and mortality. Drain fluid amylase level was routinely checked on postoperative day 3 or day 4. In accordance with the International Study Group of Pancreatic Surgery, postoperative pancreatic fistula was defined with a drain output of any measurable fluid volume on or after postoperative day 3 with an amylase level greater than three times of the serum amylase activity [
8,
9]. If patient was suspected to develop pancreatic fistula or show signs of persistent post-operative fever, an abdominal computed tomography (CT) scan was conducted to detect any complications. All patients were regularly followed up in our outpatient clinic. For malignant pathologies, all patients were followed up with serial carcinoembryonic antigen and carbohydrate antigen 19-9 (Ca 19-9) assays. Furthermore, abdominal ultrasonography or CT was performed every 12 months.
Laparoscopic distal pancreatectomy
For this approach, six ports were used (four 5 mm, one 10 mm, and one 12 mm). After port placement and pneumoperitoneum development, an abdominal exploration was performed. The lesser sac was subsequently entered through the greater gastrocolic omentum. The remainder of the operation was performed laparoscopically through the same technique used in open distal pancreatectomy as previously described [
10].
Robotic distal pancreatectomy
For this approach, five ports were used (three 5 mm for robotic arms, one 12 mm camera port, and the other 12 mm port for bedside surgeon). Patient was positioned in right semi-lateral position. The robot was docked into position after port placement and pneumoperitoneum development. The robot can be docked in the head or left shoulder position. In normal circumstances, an endoscopic linear stapler is used to divide the pancreas. If the pancreas is thickened, monopolar scissors are used alternatively to transect the pancreas. This technique will be useful to prevent crushing injury among patients with thick pancreatic parenchyma. The pancreatic ductal opening in the remnant was searched. A row of 3-0 absorbable or non-absorbable sutures was placed in a figure-of-eight fashion over the end of the pancreas. A separated 4/0 absorbable or non-absorbable suture was placed in a mattress fashion around the divided pancreatic duct (if any). The remainder of the operation was similar to that of the conventional laparoscopic approach.
Statistical methods
Prospectively collected data were analyzed retrospectively. Continuous variables were expressed as mean±standard deviation or mean and range. Continuous variables were compared using Student’s t-test. Categorical variables were compared using χ2 test or Fisher’s exact test. A P-value<0.05 was considered statistically significant.
Results
During the entire study period, 35 minimally invasive approaches of distal pancreatectomy with or without splenectomy were performed (robotic approach, n = 17; conventional laparoscopic approach, n = 18). A comparison of patients’ characteristics and pathology between the two groups is presented in Table 1. The mean operating time in the robotic group (221.4 min) was significantly longer than that in the laparoscopic group (173.6 min) (P = 0.026). Robot docking time was also included in the operating time measurement. The operating time of the robotic approach in the first eight patients and the last nine patients was similar (239.1±86.4 min vs. 205.6±60 min) (P = 0.362). Both robotic and conventional laparoscopic groups showed no significant difference in spleen-preservation rate (52.9% vs. 38.9%), operative blood loss (100.3 ml vs. 268.3 ml), overall morbidity rate (47.1% vs. 38.9%), and post-operative hospital stay (11.4 days vs. 14.2 days) (Table 2). All minor and major complications are summarized in Table 2. No significant difference was observed in the incidence of postoperative pancreatic fistula after robotic or conventional laparoscopic approach (n = 7, 41.2% vs. n = 6, 33.3%, P = 0.73). The numbers of grades A, B, and C in the robotic and conventional laparoscopic groups were 1, 5, 1 and 1, 5, 0, respectively. No patient presented cardiopulmonary complications in both groups. Moreover, both groups reported no perioperative mortality. The mean follow-up period of the robotic group and laparoscopic group was 27.4±18.8 and 113.5±49 months, respectively.
Discussion
Substantial evidence indicated that conventional laparoscopic distal pancreatectomy provides the advantages of less blood loss, less complication rates, and shorter length of hospital stay than open distal pancreatectomy; thus, the former procedure can be recommended as the treatment of choice for benign and noninvasive lesions when clinically indicated by experienced surgeons [
11,
12]. However, providing clear recommendations is very difficult with regard to laparoscopic resection of malignant pancreatic tumors because of limited conclusive data. Conventional laparoscopic surgery exhibits its own limitations, including reduced freedom of movement within the abdominal cavity and 2D view of a 3D operative field, reduced precision, and poor ergonomics. These limitations translate into a long learning curve, which requires longer time and more effort to develop and maintain such advanced laparoscopic skills. Therefore, since the first laparoscopic distal pancreatectomy was reported by Professor Cuschieri in 1994 and Professor Gagner in 1996, minimally invasive surgical technique remains not widely adopted in distal pancreatectomy [
13,
14]. These shortcomings of traditional laparoscopic surgery comprise the impetus behind robotic surgery development. By contrast, robotic system allows complex dissections, and this method is performed more easily and precisely.
Robot-assisted laparoscopic distal pancreatectomy is still a new and challenging procedure. With the advantages of the robotic system, the minimally invasive surgical technique may be applied to this complex operation. However, the major drawback is the cost of the robotic system. Additional evidence is also needed to validate the specific function of this method. To date, cost-effectiveness analysis of the robotic system remains underdeveloped. Therefore, as another novel surgical development, we need to study the safety, effectiveness, cost, and oncological outcomes of this technique over a period of time.
To our knowledge, four nonrandomized studies have been published to compare robotic and laparoscopic approaches for distal pancreatectomy [
4-
7]. Waters
et al. from Indiana University evaluated 77 distal pancreatectomies (robotic approach,
n = 17; laparoscopic approach,
n = 28; open approach,
n = 32) [
4]. Their study found that direct hospital costs are comparable among all groups, and their data suggested a shorter length of stay in robotic versus laparoscopic or open approaches. Additionally, spleen-preservation rates may improve with robotic approach at increased operative time. Kang
et al. from the Yonsei University College of Medicine evaluated 45 distal pancreatectomy cases (robotic approach,
n = 20; laparoscopic distal pancreatectomy,
n = 25) for benign and borderline malignant lesions [
5]. They found that robot-assisted approach was superior to conventional laparoscopic approach in spleen preservation; however, this feature is inferior with regard to cost and operative time. Daouadi
et al. from the University of Pittsburgh Medical Center evaluated 124 distal pancreatectomy cases (robotic approach,
n = 30; laparoscopic approach,
n = 94) [
6]. Robotic approach was equivalent to laparoscopic approach in nearly all measures of outcome and safety; however, this approach significantly reduced the risk of conversion to open resection. Oncological outcomes in these cases were superior for the robotic approach with higher rates of margin negative resection and improved lymph node yield for both benign and malignant lesions. Lee
et al. from the Memorial Sloan-Kettering Cancer Center evaluated 805 distal pancreatectomy cases (robotic approach,
n = 37; laparoscopic approach,
n = 131; open approach,
n = 637) [
7]. Patients in the open group demonstrated significantly higher intraoperative blood loss; these cases also required longer hospital stay. The robotic and laparoscopic approaches were comparable with respect to most perioperative outcomes, without clear advantage of one approach over the other. However, the robotic approach displayed no advantage of enhanced spleen preservation in that study. Both of these minimally invasive techniques may demonstrate advantages over open approach in well-selected patients. All the approaches achieved a similarly high rate of R0 resection for patients with adenocarcinoma.
Our current study showed that robotic approach was equivalent to laparoscopic approach in nearly all measures of outcome and safety; however, this approach required significantly longer operation time. In terms of blood loss and spleen-preservation rate, the robotic group was better, but the results were not statistically significant. This finding might be attributed to the small sample size of our patients. Based on our experiences, the robotic system facilitated dissection and control of small branches arising from splenic arteries and veins. This process might transform into spleen-preservation advantages. The limitations of the present study included small patient number in the series, as well as limited number of patients with malignant pathologies for a significant analysis.
In conclusion, robot-assisted laparoscopic distal pancreatectomy yielded outcomes similar to those of conventional laparoscopic approach. However, the robotic approach tended to have the advantages of less blood loss and shorter hospital stay. Further studies are necessary to determine the clinical position of robotic distal pancreatectomy.
Compliance with ethics guidelines
Eric C.H. Lai and Chung Ngai Tang declare no conflict of interest. Informed consent was obtained from all patients included in this study.
Higher Education Press and Springer-Verlag Berlin Heidelberg
PDF
(82KB)
