Sixteen cases of laparoscopic central pancreatectomy for benign or low‐grade malignant tumours in the pancreatic neck and proximal body

The purpose of this study is to examine and analyse the outcomes and patient experiences associated with laparoscopic central pancreatectomy.


Introduction
Currently, the mainstream surgical modality for pancreatic neck or proximal body tumours is laparoscopic pancreaticoduodenectomy or laparoscopic distal pancreatectomy.Although the number of laparoscopic central pancreatectomy reported in recent years has increased, and the feasibility of its application in early-stage ductal adenocarcinoma of the pancreatic neck has been confirmed, central pancreatectomy is still an uncommon option for pancreatic neck or proximal body tumours. 1,2In order to discuss and summarize the relevant experience, the minimally invasive hepatobiliary-pancreatic team of Ningbo Medical Center Lihuili Hospital (Xingning Branch and Eastern Branch) performed 16 cases of laparoscopic central pancreatectomy from January 2017 to July 2023, which are reported as follows.

Data sources
All the perioperative data of 16 patients who underwent laparoscopic central pancreatectomy between September 2017 to July 2023 were from Ningbo Medical Center Lihuili Hospital (Xingning Branch and Eastern Branch).The patients' preoperative examination includes blood routine, urine routine, faecal routine, biochemical routine, coagulation function, tumour markers, hepatitis B trilineage and pre-S1 antigen, pre-transfusion test, chest radiograph, electrocardiogram, enhanced CT, enhanced MRI and MRCP, etc. Inclusion criteria: (1) Benign or low-grade malignant tumours in the pancreatic neck and proximal body (preoperative routine EUS sampling confirmation), such as neuroendocrine tumours, intraductal papillary mucinous neoplasm, serous cystic neoplasms, mucinous cystic neoplasms, solid pseudopapillary neoplasms, etc.; (2) The diameter of the tumour is ≤5 cm, and at least 5 cm of normal pancreatic tissue can be retained at the distal end to ensure that the surgical resection margin is negative and the purpose of radical resection is achieved; (3) The tumour is close to the main pancreatic duct, and simple pancreatic tumour enucleation cannot be performed; Exclusion Criteria: (1) severe cardiovascular disease unable to tolerate laparoscopic surgery; (2) is using anticoagulant drugs; (3) ASA score ≥ 4 points; (4) Preoperative suspicion of malignant tumour or distant metastasis of malignant tumour; (5) Intraoperative rapid freezing pathology suggests malignant tumours, leading to conversion to other surgical methods; (6) atrophy of distal pancreas; (7) diffuse chronic pancreatitis; (8) blood supply to distal pancreas solely from the transverse pancreatic artery (left branch of the dorsal pancreatic artery).The perioperative data of patients, such as operative time, intraoperative bleeding volume, postoperative biliary leakage, and pancreatic fistula were retrospectively analysed.This study was approved by the Ningbo Medical Treatment Center of Lihuili Hospital ethical committee (Approval number: KY2022SL213-01).Informed consent was obtained from all patients.All methods were performed in accordance with the relevant guidelines and regulations.

Surgical method
All the surgery was performed by using the '5-hole method'.We access the abdominal cavity by open method if patients had a history of previous lower abdominal surgery.As for patients without previous surgical history, we used a directly pneumoperitoneum needle to establish a pneumoperitoneum.Pneumoperitoneum was established by connecting the pneumoperitoneum machine and maintaining the abdominal pressure at 14 mmHg.A 10 mm trocar (Kangji, Hangzhou, China) was placed 1 cm below the umbilicus as the observation hole.After the laparoscope was placed, we explored firstly the abdominal cavity and selected the location of the remaining four trocars according to the size of the intra-abdominal space of the patient.In general, two 5 mm trocars were placed separately 3 cm below the costal margin in the left and right axillary midline, a 12 mm trocar was placed 5 cm adjacent to the right of the umbilicus, and a 5 mm trocar was placed 5 cm adjacent to the left of the umbilicus.After disconnecting the gastrocolic ligament, the assistant pulled the stomach upward to expose the anterior margin of the pancreas.Intraoperative ultrasound explored the pancreas to clarify the approximate location of the tumour and the line of resection.Then we opened the gap between the posterior margin of the pancreas and the superior mesenteric vein (Fig. 1).After the assistant lifted the pancreas, the main surgeon started to use an ultrasound knife (Ethicon, USA) to dissect the pancreas and the incisal margin was usually 1 cm away from the tumour.The surgical specimen was removed and sent for intraoperative rapid freezing pathological examination, and if the tumour was confirmed to be malignant, we would adopt laparoscopic pancreaticoduodenectomy or laparoscopic distal pancreatectomy instead.The head stump of the pancreas was sutured continuously with a 3-0 gastrointestinal thread (Ethicon, USA), which is kept as far as possible on the left of the superior pancreaticoduodenal artery during the procedure (Fig. 2).The tail stump of the pancreas was anastomosed by the traditional duct-to-mucous Roux-en-Y pancreaticojejunostomy.

Postoperative follow-up
Outpatient clinic, telephone, or WeChat follow-up was used to understand the basic situation of the patient after discharge.Patients who have removed drainage tubes after discharge were regularly reviewed every 2 weeks in the outpatient clinic for blood routine, biochemical routine, ultrasound of the abdomen, and every 3 months for the whole abdominal enhanced CT or MRI.Patients discharged with the drainage tube were regularly reviewed every week in the outpatient clinic for blood routine, biochemical routine, drainage fluid amylase, and every month for abdominal nonenhanced CT until the drainage tube was removed.After that, the whole abdominal enhanced CT or MRI was reviewed every 3 months.

Patient characteristics
Five cases in males and eleven cases in females with ages 26-79 years old (average 47.4 years old).Among them, eight cases were accidentally found by physical examination, four cases had abdominal pain with abdominal distension, and four cases were admitted to the hospital for other reasons.In all patients, the tumours were located in the neck or proximal body of the pancreas.All the preoperative imaging data suggested benign tumours, and the tumours were basically located in the neck or proximal body of the pancreas.

Intraoperative data
In all patients, the operations were successfully completed without intraoperative conversion to open surgery.The operative time was 160-360 min (average 281.75 min), and the estimated intraoperative bleeding volume was 50-300 mL (average 113.75 mL).Among them, intraoperative cholangiography was performed in two cases, and intraoperative fluoroscopic laparoscopic assistance was performed in four cases.All the intraoperative rapid freezing pathology were confirmed as benign tumours with a negative resection margin.

Postoperative data
Postoperative pathology confirmed that two cases were intraductal papillary mucinous neoplasm, six cases were serous cystic neoplasms, one case was mucinous cystic neoplasms, five cases were solid pseudopapillary neoplasms and two cases were neuroendocrine tumours.The maximum tumour diameter ranged from 3.0 to 5.0 cm (average 3.67 cm).There was no one case of postoperative biliary stenosis or biliary leakage.There were no cases of perioperative and 30-day post-discharge deaths.Among them, there were five cases without detected pancreatic fistula, six cases of biochemical leakage, three cases of grade B pancreatic fistula, and two cases of grade C pancreatic fistula.Among two patients with grade C pancreatic fistula, one patient underwent pancreatic-intestinal anastomotic stoma removal, distal pancreatectomy and splenectomy due to the peri-anastomotic abscess during hospitalization and splenic artery haemorrhage.One patient underwent DSA overlay stenting due to gastroduodenal artery bleeding caused by pancreatic fistula.Because of the poor hemostatic result, the patient accepted again exploratory laparotomy to stop bleeding during hospitalization.Two patients with unplanned reoperations recovered well and were discharged after surgery.

Follow-up information
Among the 16 patients, 2 patients were discharged with drainage tubes, and 14 patients had all drainage tubes removed during hospitalization.Of the two patients discharged with drainage tubes, one had the drain removed within 2 weeks of discharge, and one of them within 4 weeks of discharge.During long-term follow-up, 16 patients did not show poor glycemic control, diarrhoea, gastroparesis, or pancreatic internal or external secretory insufficiency (Table 1).

Discussion
The specific extent of the neck or proximal body of the pancreas is not clearly defined but is generally considered to be the pancreatic tissue in front of the superior mesenteric vein.For benign tumours or borderline tumours in the neck or proximal body of the pancreas, the main surgical modalities are pancreaticoduodenectomy with or without pyloric preservation, pancreatic segment resection, and simple pancreatic tumour enucleation. 3,4With the concept of preserving pancreatic function, the operation process of pancreaticoduodenectomy is complicated and there are many postoperative complications, so it is not the best surgical method for pancreatic neck or proximal body tumours.Although simple pancreatic tumour enucleation can preserve more normal pancreatic tissue, this procedure is generally suitable for tumours <3 cm in diameter that is not close to the main pancreatic duct. 4In this study, the average pancreatic tumour diameter of the patient was 3.67 cm, and the majority of the tumour was located in the pancreatic neck tissue in front of the superior mesenteric vein area, so central pancreatectomy was used first.Some systematic reviews and meta-analyses [5][6][7] showed that central pancreatectomy could decrease the risk of exocrine failure and impairment of endocrine function.In order to preserve pancreatic function, we generally performed central pancreatectomy for benign tumours or low-grade malignant tumours in the neck or proximal body of the pancreas with a diameter ≤5 cm, which is consistent with the surgical indications suggested in the relevant studies. 8,9However, some studies relaxed the exclusion criteria for tumour size to 7 cm, 10 and a few cases reported resection of larger tumours. 11In pancreatic cystic disease, preoperative imaging is not able to completely differentiate benign and malignant lesions, which brings a challenge to the standardization of laparoscopic central pancreatectomy.In general, the malignant potential is increased with tumour size.Therefore, in the absence of pancreatic puncture pathology demonstrating benign lesions, central pancreatectomy for larger diameter pancreatic masses needs to be performed with caution.On the one hand, central pancreatectomy may make difficulties in ensuring negative margins and achieving radical resection when malignant tumours cannot be excluded, and on the other hand, it is often difficult to preserve sufficient distal pancreatic ducts during surgery due to the large size of the lesion, which may increase the incidence of postoperative complications.Based on our experience, we believe it is necessary to adhere to the tumour size exclusion criteria of ≤5 cm tumour diameter.
The gastroduodenal artery descend anterior to the right margin of the pancreatic neck, which is to the left of the pancreatic segment of the common bile duct (Fig. 3).Theoretically, when determining the right incisal margin of the pancreatic neck, it is sufficient to position the pancreatic incisal margin to the left of the vertical line of the gastroduodenal artery to ensure that the pancreatic segment of the common bile duct is not damaged when the right incisal margin of the pancreas is dissected (Fig. 3).It has been shown that about 3% of the gastroduodenal arteries are travelled on the right side of the common bile duct. 12Therefore, in practice, it is not completely certain whether the common bile duct injury and stenosis are actually caused.Our experience is: (1) according to the patient's CT and MRI before surgery, it is necessary to carefully identify: a. the travel of the common bile duct in the pancreas, b. measuring preliminarily the distance between the gastroduodenal artery and the common bile duct, c. identifying the approximate location of the tumour to determine the distance between the right incisal margin of the tumour and the common bile duct and whether it will exceed the vertical line of the gastroduodenal artery, d. whether there is variation in the gastroduodenal artery; (2) If the right incisal margin of the tumour may exceed the vertical line of the gastroduodenal artery, we generally use intraoperatively fluorescence imaging (indole green 2.5 mg, intravenous bolus) to assist the visualization of the common bile duct (Fig. 4), and then proceed to dissect the pancreas after the common bile duct is completely visualized.According to the principle that fluorescence imaging can penetrate approximately 5-10 mm of tissue, 13 if fluorescence appears during the dissection of the pancreas, it indicates that the incisal margin may be close to the common bile duct, and this would remind us of avoiding damaging the common bile duct during the dissection process.If there is extravasation of the contrast medium, it indicates that there may be a rupture in the common bile duct or even that the common bile duct is severed; (3) In the process of the dissection of the right margin of the pancreas, although there is no damage to the common bile duct, it may cause stenosis of the lower section of the common bile duct due to the heat conduction of the ultrasonic knife.In uncertain cases, we will use intraoperative cholangiography to determine whether the common bile duct is intact (Fig. 5).Of course, this method is not suitable for every operation, and it is only carefully chosen when there is a high suspicion of common bile duct stenosis, after all, the cholangiography process accompanied by the puncture and suture of the common bile duct which may also cause some degree of damage to the common bile duct.
The International Study Group of Pancreatic Fistula (ISGPS) has classified the degree of postoperative pancreatic fistula into three categories since 2016: (1) biochemical leakage: although the drainage fluid amylase is increased, it has no clinical significance, and it no longer specifically refers to pancreatic fistula; (2) grade B pancreatic fistula: need to change the postoperative treatment, continuous drainage for >3 weeks, percutaneous or endoscopic puncture and drainage, angiographic intervention for haemostasis, no infection that caused organ failure; (3) grade C pancreatic fistula: second surgery, organ failure, and death. 14In our study, the rate of grade B and above pancreatic fistula was 31.3%, which is similar to the incidence of postoperative pancreatic leakage in central pancreatectomy reported in the relevant literature, and slightly higher than in pancreaticoduodenectomy 6,15 and distal pancreatectomy. 2,5,76][7] The probability of postoperative pancreatic fistula after central pancreatectomy is higher than that of other surgical procedures is mainly because the partial resection of the neck and proximal body of the pancreas resulted in there are two stump in the pancreas.The currently recognized prognostic factors associated with postoperative pancreatic fistula are soft texture of the pancreas, small pancreatic duct diameter (<5 mm), high intraoperative blood loss (>400 mL) and high-risk pathological types (diseases other than pancreatic cancer or chronic pancreatitis). 16In our study, patients undergoing central pancreatectomy had mainly benign tumours, so the anastomosis was usually performed on a soft pancreas with an nondilated main pancreatic duct (mean pancreatic duct diameter of 3.82 mm), which may have increased the incidence of postoperative pancreatic fistula.Our general approach was to use the continuous double-layer suture method to close the head stump of the pancreas with 3-0 gastrointestinal thread and the tail stump of the pancreas was anastomosed by the traditional duct-to-mucous Roux-en-Y pancreaticojejunostomy, and the silicone support tube was placed before anastomosis according to the size of the patient's pancreatic duct.The drainage tubes were placed separately below the pancreatic-intestine anastomotic stoma and the head stump of the pancreas, and the wound was covered with the large omentum as much as possible so that it would not be too extensive even if the pancreatic fistula occurred after surgery.However, whether the method of covering with large omentum can really play a clinical role needs further study.On the postoperative day, the 'somatostatin 3mg q12h micropump' was used to suppress pancreatic secretion (5-7 days), and then adjust the dose of somatostatin for '2mg q12h micropump' to prevent transient diarrhoea (2-3 days) according to the volume of drainage fluid and amylase levels of the patient.Before removing the drainage tube, the whole abdomen CT is routinely reviewed to assess whether the patient has encapsulated fluid and if the patient does not have obvious encapsulated fluid in the operative area and one of the following conditions is met, the drainage tube can be removed: (1) the drainage volume of the abdominal drainage tube is <10 mL/d for 3 consecutive days, and the amylase of the drainage fluid is <500 U/L; (2) if the drainage volume of the abdominal drainage tube is <10 mL/d for 3 consecutive days, but the amylase of the drainage fluid is >500 U/L, the drainage tube should be pulled out for 1-2 cm daily until the lateral hole of the drainage tube is exposed on the patient's skin surface; (3) If the drainage volume of the abdominal drainage tube is >10 mL/d and the amylase of the drainage fluid is >500 U/L, the patient should be discharged from the hospital with the tube and self-record the drainage volume daily and review weekly in the outpatient clinic.When one of the first two conditions is met, the drainage tube can be removed.
In conclusion, laparoscopic central pancreatectomy is safe and feasible after careful preoperative planning and combined with the aid of intraoperative fluoroscopic laparoscopy and cholangiography.But central pancreatectomy may increase the probability of postoperative pancreatic fistula and this operation needs to be  performed by an experienced surgeon.Although the occurrence of pancreatic fistula has been increased compared to other pancreatic surgeries, conservative treatment was successful in the majority of patients and there were no serious complications due to pancreatic fistula.At the same time, central pancreatectomy effectively reduces the incidence of pancreatic insufficiency.Taking into account its long-term advantages in reducing the risk of pancreatic exocrine and/or endocrine insufficiency, we believe that central pancreatectomy remains feasible for the treatment of benign or low-grade malignant tumours in the pancreatic neck or proximal body.

Fig. 3 .
Fig. 3. Anatomical location of the gastroduodenal artery and common bile duct; Blue line: the ideal right incisal margin of the pancreatic neck.

Fig. 4 .
Fig. 4. Intraoperative fluorescence imaging for the visualization of the common bile duct (white arrow).

Table 1
Perioperative data of patients