Two-layer duct-to-mucosa pancreaticojejunostomy with a half purse-string suture: A novel anastomosis technique for soft pancreas

Background: Pancreaticoduodenectomy (PD) is a routine method in pancreatic surgery. To date, postoperative pancreatic stula (POPF) remains the most common complication and is also the major cause to death after pancreaticoduodenectomy. In order to reduce the incidence of POPF, we established a new anastomosis technique where we use a half purse-string suture on the basis of two-layer duct-to-mucosa pancreaticojejunostomy (PJ) technique and also assessed the effectiveness and safety of this approach in this study. Methods: To evaluate this new approach, 80 patients who received the new PJ technique were included in this study during 2017–2018. Meanwhile, 195 patients who underwent traditional duct-to-mucosa PJ were collected. We also introduced this new surgical approach in detail and analyzed various risk factors for postoperative complications to verify its safety and advantages. Results: First of all, there were no signicant differences in patients' preoperative characteristics. Besides, patients in the new PJ group received a less operative time (175.2±45.8 vs. 161.3±41.0, p=0.022) while no differences were found in blood loss, length of hospital stay and cost. The incidence of POPF in the new PJ group was 19%, which was much lower than those in the traditional PJ group (36%) (p=0.007). More importantly, a much lower incidence of grades B POPF (14% vs. 4%, p=0.026) was found in the new technology group, which would clinically benet patients a lot. Univariate and multivariate regression analysis also veried that this new PJ procedure was effective to improve postoperative POPF. Conclusions: Results demonstrate that this new technique is easy accomplished, safe and effective compared to traditional approach, which showed satisfactory outcomes especially at resulting in a lower POPF incidence.


Introduction
Pancreaticoduodenectomy(PD) is a widely used procedure for malignant and benign tumors of the pancreatic head and the ampullar region (1). Standard PD is a complex procedure including anastomosis of multiple organs and reconstruction of digestive system. Although it is reported quite safe and the mortality related to PD has decreased to approximately 2% in some big centers (2,3), the overall operative morbidity after PD has remained between 30% and 50%, as reported (4,5). Postoperative pancreatic stula (POPF) is considered the most common complication associated with patients morbidity and mortality, which results from leakage of secretions at the anastomotic stoma with a high rates between 10% and 28% (6). Studies demonstrate that patients with POPF have a worse overall survival, as well as higher complications, longer hospital stay (7,8). Well recognized risk factors related to POPF include gender, diabetes, BMI, pancreaticojejunal anastomosis technique, the diameter of the pancreaticduct and pancreatic texture (9) . As pancreaticojejunostomy(PJ) is the most crucial step, numerous surgical groups have investigated several methods to build a stable PJ for reducing the incidence of POPF(2, 5, 10), including pancreaticogastrostomy (11), duct-to-mucosa anastomosis (12), pancreatic duct occlusion, the use of biologic adhesive (13-15), Chen's U-stitch approach (16) and end-to-side suspender PJ (17). Here, in our institution, we studied a novel anastomosis technique where we use the half purse-string suture on the basis of two-layer duct-to-mucosa pancreaticojejunostomy technique, which effectively invaginated the pancreas into the jejunum in addition to allowing a duct-to-mucosa anastomosis. Operative Technique

Materials And Methods
In the distal pancreas, a 1.0-1.5cm long stump of pancreas was freed. The pancreaticojejunal anastomosis was covered by the jejunal wall and could be securely xed well within the pouch formed from the intestinal wall with our half purse-string suture invagination technique. The half purse-string suture technique made the pancreaticojejunal anastomosis successful invagination without suturing the pancreatic parenchyma. This procedure was performed as follows. The pancreatic parenchyma was clearly transected with electrocautery, and then we used a scalpel transect main pancreatic duct when approaching pancreatic duct position. Hemostasis procedures included suture ligatures or electrocautery.
The cutting end of the pancreatic stump was isolated for a distance of 1.0-1.5cm. After identifying the main pancreatic duct, we inserted a silicone tube into the main duct to avoid pancreatic duct occlusion and also xed the tube with suturing thread. The jejunum was positioned in the lesser sac through a rent in the transverse mesocolon. Make a half purse-string suture near the posterior mesenteric border of the jejunum with 2-0 silk suture. The length of the half purse-string suture should be a little more than the width of the pancreatic stump (picture 1-a). We tightly laced the silk thread around pancreatic stump at about 1.0 cm from the cut margin. Thus, jejunum was then wrapped the posterior of the pancreatic stump (picture 1-b). Then we made the two-layer end-to-side duct-to-mucosal pancreatico-jejunal anastomosis. The rst posterior layer was sutured between the pancreatic parenchyma of the stump and the seromuscular layer of the jejunum using interrupted silk thread. Then we made a small hole on the antimesenteric border of the jejunum which was opposite to the pancreatic duct. Next, the second layer was performed between the pancreatic duct and all layers of the jejunal wall also by interrupted silk thread sutures (picture 1-c). The silicon tube placed into the main pancreatic duct before was then inserted in jejunal lumen after the posterior side of the second layer was completed. The anterior two layers of the pancreatico-jejunal anastomosis were anastomosed in the reverse order and in the same manner (picture 1-d).The last layer was stitched together with interrupted silk thread sutures between the seromuscular layer of the anterior border of the jejunum and the silk thread used for half purse-string  p=0.007). What's more, patients in T-PJ group received a much lower incidence of "clinically relevant" stulas (grades B 4% vs. 14%, p=0.026), which would bene t patients a lot. There were also no differences in other complications including delayed gastric emptying and infection (Table 1). What's more, we then performed univariate logistic regression model to analyze clinical risk factors for POPF. Results veri ed that soft pancreatic texture (p=0.002) and T-PJ producer (p=0.008) were signi cant risk factors for the incidence of POPF (Table 2). Further multivariate analysis of these two factors showed that soft texture and surgical procedure were the signi cant risk factors of POPF (Table 2). Taken together, all these results con rmed that this new PJ technique decreases POPF importantly.

Discussion
With advances in technology, PD is reported to be safer than before. Although the postoperative mortality rate of patients after pancreatic surgery has decreased, the incidence of postoperative POPF is still high at big volume centers (4). POPF remains the most common and serious clinical complication after pancreaticoduodenectomy, which may worsen both early and long-term outcomes for patients. Among those de nite risk factors for POPF (include patient factor, operation factor and pancreas factor (18,19)), soft pancreatic texture may be the most important factor. Besides, the degree of pancreatic brosis is another important factor (20). One of the reasons is that the shear forces applied during xing the tube with suturing thread will lacerates the gland and a pancreatic leakage may easily occur from the area of the suture placement. Therefore, a reliable way to better place the tube into the main pancreatic duct and carefully x this structure is needed (21). Hence many surgical procedures have been used in the management of the pancreatic remnant including pancreaticojejunostomy, pancreaticogastrostomy and other new PJ technology (16,17). Besides, recent studies found that the smear positivity of drainage uid on POD 3 was one of the independent risk factors for POPF (22,23).
In our study, we introduced a new technique: the half purse-string suture technique, which was detailed introduced in the article. This new technique showed favorable results in our 80 patients. No signi cant differences were found in demographics outcomes. Besides, compared to traditional PJ, patients with this half purse-string suture technique received a less operative time (161.3±41.0min vs. 175.2±45.8min, p=0.022). Furthermore, only 19% patients developed POPF (36% in control group, p=0.007). Among these patients, rate of grade B-type POPF sharply decreased (4% vs. 14%, p=0.026). Further regression analysis con rmed that T-PJ technology could reduce the incidence of POPF. Therefore, this new surgical technique is safe and effective.
Our half purse-string suture technique has several advantages by making the pancreaticojejunal anastomosis successful invagination without suturing the pancreatic parenchyma. With this invagination technique, the pancreaticojejunal anastomosis is covered by the jejunal wall and can be securely xed well within the pouch formed from the intestinal wall. It is a good method to protect the pancreaticojejunostomy. We can avoid parenchymal tearing at the outermost layer of invaginating sutures with no suture in the pancreas. The possibility of pancreatic juice leakage has been greatly reduced with this technique for soft pancreas. The possibility of hemostasis of the cut end of the pancreatic remnant has also been reduced with this technique. It is necessary to insert a suitable silicon tube into the main pancreatic duct in this technique. We believe that this will help to avoid occlusion of the main pancreatic duct.
One limitation of our study is that this new T-PJ is carried out by surgeons from one department.
Surgeon's experience is reported to be a signi cant factor for incidence of POPF (9,18). Besides, a multicenter randomized control trial is needed to further evaluate the safety and effectiveness of this technology. We hope that this technique will be accepted more widely in the near future. Besides, laparoscopic and robot pancreaticoduodenectomy are widely used worldwide (24-26), which provides more challenges for surgeons.

Conclusions
Results demonstrated that our half purse-string suture technique is safe and easy to perform. More important is that this technique could effectively reduce the incidence of POPF.

Declarations
Ethical approval and consent to participate The study was approved by Shanghai Changhai Hospital Ethics Committee. Written informed consent was obtained from all participants.

Consent for publication
Not applicable

Availability of data and materials
The datasets used and analyzed during the current study are available from the corresponding author on reasonable request.

Competing interest
The authors declare that they have no competing interests.   Figure 1