Laparoscopic Middle Pancreatectomy: How Do I Do It?

Introduction

With the widespread use of imaging studies, symptomatic or asymptomatic low-potential malignant and benign pancreatic diseases are more frequently diagnosed. ¹ Surgical resection may be indicated. At present, parenchyma-sparing pancreatectomy is the recommended surgical procedure to avoid the long-term endocrine and exocrine pancreatic insufficiency encountered with major pancreatic resections,^2,3 especially if the patient is young. Parenchyma-sparing pancreatectomy includes enucleation and middle pancreatectomy (MP), and when enucleation is contraindicated, MP becomes the procedure of choice. In the past MP was considered a complicated surgical procedure because of the extensive dissection along the mesentericoportal vein and celiac trunk collaterals and the presence of two pancreatic cut surfaces, increasing the risk of pancreatic fistula (PF) and bleeding.^4–6 Although laparoscopy is now extensively used in the field of pancreatic surgery and has become the routine approach for distal resections,^7–10 the laparoscopic approach has rarely been described for MP.^11–16 We report and illustrate our experience with a video (http://videos.liebertpub.com/preview.cfm?videoID=1653) of laparoscopic MP (LMP) for small branch duct intraductal papillary mucinous neoplasia (IPMN) diagnosed after numerous incidents of acute pancreatitis, and we describe our results in 21 patients who underwent this procedure.

Patient and Methods

Surgical technique

Our surgical technique has already been described in detail. ¹⁶ The patient is placed in the supine position with the legs spread apart and the right arm along the body. Trocar placements and surgical team position are illustrated in Figure 1. Harmonic™ shears (Ethicon Endo-Surgery, Issy les Moulineuax, France) and a bipolar coagulation device are needed. Once exploration is complete, the gastrocolic ligament is divided, and the stomach is charged with tape and tracked by the epigastric trocar. The inferior pancreatic border is freed to allow identification of the mesentericoportal vein and facilitate handling of the pancreas. The superior pancreatic border is freed, and lymphadenectomy of the hepatic artery and the celiac trunk is performed for lymph node picking and to facilitate identification of vessels. Once the borders of the pancreas have been clearly identified, the pancreas is easy to control at the neck and is sectioned with a stapler or more rarely with the Harmonic shears. If possible, elective ligation of the main pancreatic duct can be performed. The pancreas is freed from the splenic vessels and then sectioned to the left of the lesion. Although the lesion can be located by laparoscopic ultrasound, it should be mentioned that anatomical structures, such as the gastroduodenal artery on the right and the first curve of the splenic artery on the left, are important landmarks for small lesions. One-layer pancreatogastric (PG) anastomosis is performed. Omentoplasty with the great omentum can be performed to cover the dissected vessels (preventing bleeding) and to separate the two potential PF sites and facilitate postoperative management. Sites of potential PF are drained with two tubular drains and exteriorized through the lateral abdominal wall.

OPEN IN VIEWER

FIG. 1.

Open laparoscopy and five trocars were used for this procedure: trocars 1 and 2 for the assistant, trocars 3 and 4 for the surgeon, and trocar 5 for stomach traction. The specimen was removed by trocar 3, with or without enlargement.

Our case

A 45-year-old woman was diagnosed with IPMN after numerous incidents of acute pancreatitis. She had no other medical or surgical history. She had presented with at least eight episodes of nonsevere acute pancreatitis since 2012. Blood tests showed elevated pancreatic but normal liver enzyme levels, and there were no gallbladder stones. The first endoscopic ultrasound identified a 1-cm cystic lesion in the pancreatic neck. The diagnosis of branch duct IPMN was suspected based on clinical results and the typical magnetic resonance imaging features, showing a small, irregular, cystic lesion communicating with the main pancreatic duct, which was moderately dilated at 5 mm (Fig. 2A). The lesion was located 2 cm to the left of the gastroduodenal artery, and LMP was decided on (Fig. 2B). Enucleation was contraindicated because the IPMN was deep and communicated largely with the main pancreatic duct, and histological involvement of the main pancreatic duct (mixed type) cannot be eliminated.

OPEN IN VIEWER

FIG. 2.

(A) Magnetic resonance imaging showed branch duct intraductal papillary mucinous neoplasia (white arrow) with a moderate dilatation of the main pancreatic duct (dotted arrow). (B) The lesion is located 2 cm to the left of the gastroduodenal artery (white arrow).

Results

Discussion

With the rapid and widespread use of laparoscopy, low-potential and benign pancreatic diseases become ideal candidates for laparoscopic parenchyma-sparing pancreatectomy resections. We believe that LMP should be promoted compared with other types of pancreatic resection; most MP candidates can theoretically benefit from the laparoscopic approach because there are few technical or oncological contraindications. The PG is most frequently used for reconstruction, but pancreatojejunal anastomosis had also been described.^15,17 We believe that there is no need to prolong operative time with a Roux-en-Y pancreatojejunal anastomosis, especially when randomized controlled trials have not shown that pancreatojejunal is better than PG anastomosis in the case of pancreatoduodenectomy.^18–20

In our experience and compared with other studies, the shorter operative time (191 minutes versus 225–480 minutes)^11–13,15 and the low rate of conversion (5% versus 17%–22%)^11,12 are probably related to our experience with the laparoscopic approach and familiarity with this procedure. We recently published our experience with 100 cases of MP performed over a period of 12 years, ²¹ mainly performed by the open approach. There are very few larges series on LMP,^11–15 and it is difficult to draw conclusions on morbidity because of the retrospective design of these studies and the nonstandardized definition of PF. No mortalities have been reported. In our experience, the postoperative outcome after open MP and LMP was similar for mortality, overall morbidity, and PF: 3%, 72%, and 44% and 0%, 71%, and 57%, respectively. The cosmetic advantages of the laparoscopic approach should not be underestimated because in our experience most patients were young women.

Conclusions

LMP is a safe procedure for nonmalignant pancreatic diseases, and the postoperative outcome is comparable to that of the open approach. This technique can be standardized because most candidates for MP can theoretically benefit from the laparoscopic approach.