Laparoscopic distal pancreatectomy for benign or borderline malignant pancreatic disease is widely accepted, because its benefits include less postoperative pain, less blood loss, and earlier recovery than the open procedure [1,2,3,4]. The spleen preservation possible with this procedure also circumvents the risk of postoperative infectious complications after splenectomy [5, 6]. With advances in the laparoscopic technique, the splenic vessels are increasingly preserved to avoid spleen-related complications such as splenic abscess and infarction that can occur after the splenic vessels are sacrificed [7, 8]. However, the splenic-vessel-preserving method used during laparoscopic distal pancreatectomy is more technically demanding and time-consuming than the splenic-vessel-sacrificing method [9]. It requires meticulous dissection of the pancreas from the splenic vessels, and bleeding is difficult to control in the laparoscopic setting if it occurs from the small vessel branches during dissection.

In laparoscopic spleen- and splenic-vessel-preserving distal pancreatectomy, ultrasonic shears, sometimes with the application of clips, are usually used to dissect the splenic vessels and to control their small branches. However, there is a risk of bleeding from the splenic vessels during the procedure used to isolate their small branches. Furthermore, when bleeding from the splenic vessels occurs during dissection, improper manipulation of the splenic vessels and inappropriate hemostasis may impair the patency of the preserved splenic vessels, which increases the risk of left-sided portal hypertension [10, 11]. We have recently performed a blunt dissection technique using the LigaSure vessel-sealing device (Medtronics, Minneapolis, MN, USA) during laparoscopic spleen- and splenic-vessel-preserving distal pancreatectomy, as introduced by Suzuki et al. to reduce the risk of bleeding during the dissection of the splenic vessels [12]. With this technique, the small branches of the splenic vessels are safely controlled without requiring dissection, which sometimes includes the surrounding pancreatic parenchyma. The aim of this study was to compare the safety and efficacy of the blunt dissection technique using LigaSure technology with the conventional dissection technique using ultrasonic shears during laparoscopic spleen- and splenic-vessel-preserving distal pancreatectomy. We compared the clinical outcomes and postoperative patency of the preserved splenic vessels after treatment with the two surgical techniques.

Materials and methods

Patients and clinical variables

In total, 123 patients underwent laparoscopic spleen- and splenic-vessel-preserving distal pancreatectomy at Seoul National University Bundang Hospital, Seoul, Korea, between March 2003 and December 2015. Of these patients, 49 patients who were operated on by the same surgeon (Yoon YS) were included in this retrospective study to avoid any possible bias caused by different surgeons. The indication for laparoscopic spleen- and splenic-vessel-preserving distal pancreatectomy was a suspected benign or borderline malignant tumor located in the body or tail of the pancreas on preoperative radiological imaging. The patients were divided into two groups according to the use of LigaSure: the LigaSure group and non-LigaSure group. Initially, 26 patients (non-LigaSure group) underwent surgery via a conventional dissection technique using ultrasonic shears and endoclips. Since the introduction of LigaSure technology at our institution, 23 patients (LigaSure group) have undergone blunt dissection using the LigaSure vessel-sealing system. The clinical, surgical, and pathological data were retrieved from a prospectively collected database. The two groups were compared in terms of their general characteristics (age, sex, body mass index, and American Society of Anesthesiologists [ASA] class), pathological findings (final pathological diagnosis, tumor size, and length of the resected pancreas), and surgical data (operative time, intraoperative blood loss, postoperative hospital stay, and postoperative complications).

To evaluate the postoperative patency of the preserved splenic vessels, computed tomographic (CT) images taken > 3 months after surgery were compared with the preoperative CT images. One radiologist specializing in the hepatopancreatobiliary system reviewed all the images. The patency of the splenic vessels was classified into three grades according to the degree of stenosis: grade 0, intact; grade 1, partial occlusion; and grade 2, total occlusion.

This retrospective study conformed to the ethical guidelines of the Declaration of Helsinki. The Investigational Review Board and Ethics Committee of Seoul National University Bundang Hospital, Seoul, Korea, approved the study.

Surgical technique

Under general anesthesia, the patients were placed in the supine position with left-sided elevation. After a carbon dioxide pneumoperitoneum was created with a 12 mm infra-umbilical port, 3–4 additional trocars (12 and 5 mm trocars for the operator, and one or two 5 mm trocars for the assistant) were inserted. The principle of trocar placement was the same in the two groups. The positions of the two working ports for the operator varied according to the tumor location and have been described elsewhere [11]. The surgeon and laparoscopist stood on the right side of the patient, and the assistant stood on the left side of the patient.

After trocar placement, the gastrocolic ligament was divided from the midline towards the spleen using a surgical energy device. The inferior border of the pancreas was dissected until the splenic vein was exposed. The dissection then proceeded to the superior border of the pancreas, until the splenic artery was exposed. After isolating the splenic artery, a window was made between the pancreas and splenic vessels, and the pancreas was transected using an endoscopic stapler. After the transection of the pancreas, the distal pancreas was dissected meticulously away from the splenic vessels in a medial-to-lateral fashion. The small branches of the splenic vessels encountered during the dissection were divided with endoclips or ultrasonic shears in the non-LigaSure group (Fig. 1). Tiny vessel branches were divided using ultrasonic shears alone, but relatively large vessel branches were clipped after isolation using laparoscopic dissecting forceps and then divided using scissors or ultrasonic shears. In the LigaSure group, splenic-vessel branches were controlled using the LigaSure system, without requiring dissection, and sometimes also the surrounding pancreatic parenchyma when the splenic vessels were embedded in the parenchyma (Fig. 2). In the non-LigaSure group, Harmonic Ace® (Ethicon, Somerville, NJ, USA) and the GEN04 generator were used during the initial period. Thereafter, Harmonic Ace® plus (Ethicon) and the GEN11 generator or Sonosurg® X and the G2 generator (Olympus, Tokyo, Japan) were used. In the LigaSure group, the LigaSure™ Blunt Tip and the ForceTriad (Medtronics) generator was used in all procedures. Meticulous bleeding control and irrigation were performed, and fibrin glue was applied to the pancreatic resection margin.

Fig. 1
figure 1

Intraoperative images of the conventional dissection technique using ultrasonic shears during laparoscopic spleen- and splenic-vessel-preserving distal pancreatectomy. A After the pancreas is transected, a small branch of the splenic vein (red arrow) is dissected with Maryland dissecting forceps. The dissected small branch is clipped (B), and then divided with ultrasonic shears (C). D After completion of the distal pancreatectomy, the splenic vein and artery are fully exposed with several clips (arrowheads). SA splenic artery, SV splenic vein. (Color figure online)

Full size image
Fig. 2
figure 2

Intraoperative images of the blunt dissection technique using LigaSure during laparoscopic spleen- and splenic-vessel-preserving distal pancreatectomy. After the pancreas is transected, dissection around the splenic vein is performed with LigaSure. Small branches of the splenic vein (A, red arrow) and splenic artery (B, red arrow) are bluntly dissected. C Dissected small branch of the splenic vessel is divided with LigaSure without clipping. D After completion of the distal pancreatectomy, the splenic vein and artery are clearly exposed without endoclips. SA splenic artery, SV splenic vein. (Color figure online)

Full size image

The surgical specimen was retrieved in a vinyl bag and extracted through a small incision created by extending a port-site incision. A Jackson–Pratt surgical drain was placed adjacent to the pancreatic resection margin.

Statistical analysis

Continuous variables are presented as means (ranges) and categorical variables were compared with a χ2 test or Fisher’s exact test, and continuous variables were compared with the nonparametric Mann–Whitney U test. Values of P < 0.05 were considered statistically significant. SPSS version 24.0 for Mac (SPSS, Chicago, IL, USA) was used for all analyses.

Results

Baseline characteristics

The baseline characteristics of the patients in the LigaSure group (n = 23) and non-LigaSure group (n = 26) are shown in Table 1. There were no significant differences between the two groups. The pathological diagnosis after laparoscopic spleen- and splenic-vessel-preserving distal pancreatectomy was benign or borderline malignant pancreatic lesion in all the patients, with no significant difference between the two groups.

Table 1 Baseline characteristics of the LigaSure and non-LigaSure groups
Full size table

Perioperative outcomes

Table 2 presents the postoperative outcomes for both groups. The LigaSure group had a shorter operative time than the non-LigaSure group (145 vs. 231.1 min, respectively, P = 0.001) and less intraoperative blood loss (95.6 vs. 360 ml, respectively, P = 0.001). The postoperative hospital stay was also significantly shorter in the LigaSure group than in the non-LigaSure group (6.4 vs. 9.8 days, respectively, P = 0.001). The incidence of postoperative complications did not differ significantly between the two groups. Clinically relevant pancreatic fistulas (International Study Group on Pancreatic Surgery grade B/C) occurred in two patients in each group. Symptomatic intra-abdominal fluid collection developed more frequently in the non-LigaSure group. All the complications improved with conservative or radiological interventions. There was no postoperative mortality in either group.

Table 2 Postoperative outcomes in the LigaSure and non-LigaSure groups
Full size table

We evaluated whether the learning curve of the surgeon affected the postoperative outcomes in the LigaSure group, because recent patients were treated using the LigaSure device. Patients in each group were classified into three subgroups (early, mid, and late groups) according to the timing of surgery (Table 3). In general, the operative time and intraoperative blood loss decreased as the experience of the surgeon increased. The postoperative hospital stay also gradually decreased except for a slightly increased stay in the late LigaSure group. The postoperative complication rate was not significantly different among the six subgroups. In the non-LigaSure group, the operative time and intraoperative blood loss decreased markedly after the first nine patients were treated. With a marginal significance, a 50% reduction in intraoperative blood loss (P = 0.077) and a more than 1 h decrease in the operative time (P = 0.094) were found in the mid non-LigaSure group (II) compared with the early non-LigaSure Group (I). Thereafter, the operative time and intraoperative blood loss remained stable: no significant differences were observed between the mid and late non-LigaSure subgroups. After introduction of the blunt dissection technique using the LigaSure system, operative time and intraoperative blood loss were reduced: a decrease of 35 min of operative time and 101 ml of intraoperative blood loss. However, the difference was not significant. After gaining experience with the first eight patients in the LigaSure group, a significant decrease in operative time (P = 0.028) and a marginally significant decrease in intraoperative blood loss (P = 0.065) occurred.

Table 3 Variations in perioperative outcomes
Full size table

Postoperative splenic-vessel patency

Postoperative splenic-vessel patency data are presented in Table 4. The postoperative patency of the splenic artery in the LigaSure group was grade 0 in 20 patients (90.9%), grade 1 in one patient (4.5%), and grade 2 in one patient (4.5%); and in the non-LigaSure group, it was grade 0 in 22 patients (84.6%), grade 1 in one patient (3.8%), and grade 2 in three patients (11.5%). The patency rate did not differ significantly between the two groups (P = 0.681). However, the postoperative patency of the splenic vein was significantly better in the LigaSure group than in the non-LigaSure group (P = 0.011). The splenic vein patency in the LigaSure group was grade 0 in 13 patients (59.1%), grade 1 in eight patients (36.4%), and grade 2 in one patient (4.5%); whereas in the non-LigaSure group, it was grade 0 in 16 patients (61.5%), grade 1 in two patients (7.7%), and grade 2 in eight patients (30.8%).

Table 4 Postoperative splenic-vessel patency in the LigaSure and non-LigaSure groups
Full size table

Discussion

To our knowledge, this is the first study to compare the usefulness of the blunt dissection technique using LigaSure with the conventional dissection method using ultrasonic shears during laparoscopic spleen- and splenic-vessel-preserving distal pancreatectomy. Although the safety and efficacy of LigaSure has been demonstrated in other gastrointestinal surgical procedures [14,15,16,17], there have been few reports on its use in splenic-vessel preservation during distal pancreatectomy. The use of LigaSure technology for dissection has been reported as an alternative to the traditional dissection techniques in liver resection, gastric cancer resection, and esophagogastric devascularization [14,15,16,17]. In addition, recent reports demonstrated that dissection facilitated by the LigaSure device in pancreatoduodenectomy is associated with a decreased operative time compared with the traditional dissection techniques [18, 19].

In this study, we found that the blunt dissection technique using LigaSure during laparoscopic spleen- and splenic-vessel-preserving distal pancreatectomy markedly reduced the intraoperative blood loss and operative time. These favorable outcomes in the LigaSure group (operative time, 145 min; intraoperative blood loss, 96 ml) are better than those of previous reports that used the same operative procedure, in which the operative time ranged from 191 to 220 min and intraoperative blood loss volume from 200 to 378 ml [20,21,22,23]. This technique also allowed the dissection of the splenic vessels from the pancreatic parenchyma and, therefore, increased the postoperative patency rates in the preserved splenic vessels. In addition, the use of the LigaSure device was associated with a shorter postoperative hospital stay. However, this finding needs to be interpreted with caution, because there was no significant difference between the two groups in terms of postoperative complications. At our institution, an enhanced recovery program was introduced starting in 2012, when the first patients of the LigaSure group underwent surgery. This might have contributed to the shorter postoperative hospital stay in the LigaSure group compared with the non-LigaSure group.

The main concern with laparoscopic spleen- and splenic-vessel-preserving distal pancreatectomy is bleeding from the small vessel branches during the dissection of the pancreas from the splenic vessels, because the immediate control of bleeding is difficult in the laparoscopic setting once it occurs. To control the small branches of the splenic vessels, dissection with ultrasonic shears, followed by the application of clips if necessary, has been most commonly used, but requires sufficient dissection and isolation of the small vessel branches for the safe application of clips. However, this procedure entails a risk of bleeding if the small vessel branches are injured, and is challenging in some patients in whom the splenic vein is embedded within the pancreatic parenchyma. Furthermore, even when ultrasonic shears are applied to the small vessel branches, bleeding sometimes results from incomplete sealing, probably because the high-frequency vibrations of the jaws injure the small branches before the coagulation seal is created. The clips applied to occlude the small vessel branches can also sometimes slip off the splenic vessels, and troublesome bleeding may occur.

In contrast, LigaSure safely controls the splenic-vessel branches by fusing the vessels with a combination of pressure and energy, using a different mechanism from that of ultrasonic shears. It requires no prior dissection or isolation of the small vessel branches, and can also safely clamp the branches of the splenic vein together with the surrounding pancreatic parenchyma when the splenic vessels are deeply located within or strongly adherent to the pancreatic parenchyma. Even if a small amount of pancreatic parenchyma is left on the wall of the splenic vessels, it is unlikely to cause clinical problems. LigaSure can also seal blood vessels of up to 7 mm in diameter, so that clipping or suturing can be avoided. This improved hemostasis provides better visibility, so a more accurate dissection can be achieved. These advantages of the blunt dissection technique using LigaSure reduce the intraoperative blood loss by avoiding bleeding events during the complete dissection of small splenic-vessel branches and reduce the operation time required for this procedure.

Another advantage of LigaSure demonstrated in this study is that it increases the preservation rate of the splenic vessels. Our results show that the total occlusion rate of the splenic vein was statistically significantly lower in the LigaSure group than in the non-LigaSure group. When we evaluated splenic-vessel patency after spleen- and splenic-vessel-preserving surgery in a previous study, laparoscopic surgery and intraoperative blood loss were significant risk factors for poor splenic-vessel patency. We inferred that splenic-vessel patency during laparoscopic surgery was compromised by thermal damage to the vessels by the surgical energy device, the improper manipulation of the splenic vessels, and inappropriate hemostasis. In this context, LigaSure is expected to lessen the adverse effects of laparoscopic surgery on splenic-vessel patency. Although the temperature at the tip of instrument varies with the power setting and the application time, LigaSure is reported to involve a smaller increase in temperature and less-intense thermal damage into nearby tissues than other energy devices [24, 25]. Furthermore, less dissection of the small vessel branches and the use of fewer clips should reduce the risk of bleeding during the dissection of the splenic vessels.

This study has several limitations, including its retrospective nature and small number of patients. However, according to a power analysis for comparisons of the continuous variables (operative time, intraoperative blood loss, and postoperative hospital stay, P = 0.001), the power was greater than 90%, which suggests that a sufficient number of patients were analyzed. On the other hand, in a power analysis for comparison of the categorical variable (postoperative complication rate: a difference of 13% between the two groups), the power was only 42%, which was not statistically significant. Therefore, to overcome the low statistical power, more enrolled patients are needed in a future study. Another limitation is that the surgical outcomes were influenced by the surgeon’s skill and proficiency with the surgical instruments used. Using a conventional dissection technique with ultrasonic shears, expert surgeons can complete laparoscopic spleen- and splenic-vessel-preserving distal pancreatectomy in a short operative time with minimal blood loss. However, we present our experiences with the new dissection technique using the LigaSure technology, because we experienced markedly improved perioperative outcomes: a greater than 50% reduction in operative time and 90% reduction in intraoperative blood loss on the comparison between the early non-LigaSure subgroup and late LigaSure subgroups (Table 3), while maintaining good patency of the preserved splenic vessels. Another is the effect of the learning curve on the difference in operative time and intraoperative blood loss between the two surgical procedures. More-recent patients were treated with LigaSure, so the accumulation of surgical experience might have reduced the operative time, intraoperative blood loss, and success rate of splenic-vessel patency. However, when we performed a sequential analysis of perioperative outcomes in subgroups, categorized according to the timing of surgery, the operative time and intraoperative blood loss in the non-LigaSure group became consistent after experience with the first nine patients, which was before introduction of the LigaSure technology (Table 3). In addition, the clinical outcomes of operative time and intraoperative blood loss in the late non-LigaSure subgroup were not inferior to those after the learning curve had plateaued in previous reports on the learning curve of laparoscopic distal pancreatectomy [26, 27]. Nonetheless, prospective comparative studies are required to confirm the advantages of the blunt dissection technique using LigaSure over other dissection techniques during laparoscopic spleen- and splenic-vessel-preserving distal pancreatectomy.

In conclusion, the blunt dissection technique using LigaSure during laparoscopic spleen- and splenic-vessel-preserving distal pancreatectomy reduces bleeding events during the dissection of the splenic vessels and thus reduces the operation time. This technique also increases the success rate of intended splenic-vessel preservation and the postoperative patency rate in the preserved splenic vessels. However, these potential advantages of the blunt dissection technique using LigaSure during laparoscopic spleen- and splenic-vessel-preserving distal pancreatectomy must be confirmed in well-designed clinical studies that include large numbers of patients.