Developing an in vivo porcine model of duct‐to‐mucosa pancreaticojejunostomy (Yonsei‐PJDTM)

Abstract Laparoscopic pancreaticoduodenectomy (LPD) is technically feasible, but its safety is still controversial. Pancreas texture and the small size of the main pancreatic duct indicate laparoscopic pancreaticoduodenectomy (LPD) as a challenging procedure. Thus, LPD could be a risk factor for postoperative pancreatic fistula (POPF), longer hospital stay, and delayed adjuvant chemotherapy that affects long‐term oncologic outcome. So, it is important to promote education on LPD especially techniques for pancreaticojejunostomy. A porcine model for duct‐to‐mucosa pancreaticojejunostomy (PJ) (Yonsei‐PJDTM) was developed, and details of the model will be described in this report.


| INTRODUC TI ON
Technical feasibility and surgical experiences of laparoscopic pancreaticoduodenectomy (LPD) have been reported, but the safety of LPD is still controversial. In fact, a recent randomized control study comparing LPD and open pancreaticoduodenectomy (OPD) raised considerable safety issues about LPD. 1 Pancreaticoduodenectomy (PD) has two phases; resection and reconstruction. 2 LPD is not that easy of a procedure. For example, in periampullary cancers associated with cholangitis and pancreatitis, resection will be difficult due to severe adhesion and invasion associated with potential risk of combined vascular resection. On the other hand, reconstruction is expected to be easy due to enlarged bile duct and remnant hard pancreas with dilated pancreatic duct. However, these clinical circumstances will be the opposite in benign and low-grade malignant periampullary neoplasm. 3 In particular, managing the remnant soft pancreas with small pancreatic ducts is believed to be one of the challenging procedures during LPD. This is because it is regarded as a risk factor to clinically relevant postoperative pancreatic fistula (POPF), 4 resulting in prolonged hospital stay, increased medical costs, and even surgery-related mortality. These factors also apply to OPD. Therefore, surgical simulation is very important to successfully manage the remnant soft pancreas with small pancreatic duct during PD, especially for less experienced pancreatic surgeons. 5 It is agreed that that soft remnant pancreas with small pancreatic duct is risky, and safe pancreaticojejunostomy (PJ) is vital during PD. However, there are no appropriate ways to develop surgical skills for this challenging procedure. Surgeons hope to overcome the learning curve and become fluent in surgical techniques for safe PJ based on accumulated surgical experience. This is a critical unmet need in the clinical practice of pancreatic surgery.
In this report, a newly developed porcine model for duct-to-mucosa PJ (Yonsei-PJ DTM ) is described. This porcine model is quite similar to that for human soft remnant pancreas with a small pancreatic duct. 6 We believe that every new surgeon who needs to perform PD or prepare for PD, regardless of whether they are using the minimally invasive or open approach, can benefit from the present model. The method teaches not only the concepts of the surgical procedure, but also improves surgical skills for safe PJ.

| Model concepts
The following is background for developing a porcine PJ model. Implantation of an artificial neo-pancreatic duct is the main idea of the Yonsei-PJ DTM model. 1. In the porcine model, dissection of the pancreatic neck above the superior mesenteric vein-splenic vein-portal vein confluence is very similar to that in real patients.
2. When dividing the pancreatic neck, the left-sided porcine pancreas is soft and appropriate for simulating remnant soft pancreas with a small pancreatic duct in PD.
3. However, the actual pancreatic duct of the porcine pancreas is too small to be appropriate for duct-to-mucosa (DTM) anastomosis training. 7 4. Therefore, an artificial tube with small lumen can be implanted into the left-side porcine pancreas after division of the neck for simulating soft remnant pancreas with a small pancreatic duct.

| Preparation of the Yonsei-PJ DTM model
To facilitate neo-artificial pancreatic duct implantation into the leftside pancreas, a specially designed catheter can be applied. The end of this catheter divides into two branches; 1.5 cm from the endpoint of the catheter is a 5 mm hole at an angle of 45° so that it can F I G U R E 1 Specially designed catheter for preparation of a Yonsei-PJ DTM model. Using this specially designed catheter, the direction of the guidewire can be changed to penetrate the dorsal pancreas. The catheter can be removed with a left-side penetrated and positioned guide-wire, through which a subsequent artifical neopancreatic duct can be inserted exit above the pancreas. Further details are filed with the Korean Intellectual Property Office (Figure 1). 8 The catheter can be inserted simply using the following steps ( Figure 2).

| Surgical simulation using the Yonsei-PJ DTM model
There are various techniques for PJ during PD. 9

| D ISCUSS I ON
Managing soft remnant pancreas with a small pancreatic duct during PD is challenging. 5 This step is an Achilles heel for safe PD. Therefore, appropriate preoperative surgical rehearsal is necessary for enhancing the safety of PD. Considering the increase in minimally invasive PD, this unmet clinical need should be resolved for safe implementation of minimally invasive PD in clinical practice. 11 We developed an in vivo porcine model mimicking remnant soft pancreas with a small pancreatic duct during PD. A specialized catheter may be very helpful for preparing this model, 12  This study has several limitations. First, it is the first to construct a porcine model; there is no accumulated data to support the efficiency of this proposal for training with laparoscopic PJ. We are currently collecting data and will report it in the future. Second, this proposal has the advantage of practicing multiple PJ simulations with one experimental pig, although it is expensive to prepare the pig. Much funding will be needed for training in minimally invasive surgery, and systematic help of the local hepato-biliary-pancreas association is critical.
In conclusion, the Yonsei-PJ DTM model could be an option for training on PJ, and further study is needed.