Objective: The tumors of the head of the pancreas are one of the leading causes of cancer-related death in Western countries. The current gold standard for these tumors is a Whipple procedure. This procedure did not change in its surgical steps since when it was initially introduced in 1935. More recently, a laparoscopic approach with similar outcomes has been described. The aim of this paper is to describe the laparoscopic surgical technique performed in our unit, reporting single center postoperative outcomes.
Patients and methods: From the 1st January 2013 to the 31st December 2015 a database was created. Data about patients who underwent a laparoscopic pancreaticoduodenectomy (LPD) were collected prospectively. All patients were preoperatively assessed with blood samples, tumor markers, CT chest abdomen and pelvis and/or MRI pancreas. Only patients with specific characteristics were considered eligible for an LPD: performance status 0, body mass index (BMI) less than 30 kg/m2, a small neoplastic lesion (< 3.5 cm) confined to the pancreas, the absence of infiltrated organs and/or blood vessels (T1 or T2). Postoperative data and complications were recorded and described according to the Clavien-Dindo classification and the international study group of pancreatic surgery definitions.
Results: In a time interval of 36 months, 31 patients with an initially considered resectable pancreatic cancer were referred. 11 patients were found to have metastasis during the preoperative workout. Only 10 patients were considered eligible for a LPD. Six of them were men (60%). The mean BMI was 25.01 kg/m2 (19.6-29.8). 5 patients, who underwent to LPD did not have any comorbidities. An overall 50% of all patients were jaundice at the time of diagnosis with a mean bilirubin level of 181.3 µmol/L (119.7-307.8). All patients with a direct bilirubin greater than 250 µmol/L underwent a preoperative percutaneous biliary drainage. In the majority of the LPD performed (50%), the histology reported a pancreatic adenocarcinoma. Other postoperative histology described were: IPMN (20%), ampullar neoplasia (20%) and neuroendocrine tumor (10%). Neo-adjuvant chemotherapy was never considered indicated. The reported postoperative complications were: 1 anastomotic bleeding, 2 pancreatic fistula, 1 infected intra-abdominal collection and 1 delay gastric emptying. The pancreatic fistulas were considered grade A and grade B. One fatality after LPD occurred because of an uncontrollable, diffuse severe hemorrhagic gastritis associated with a GJ anastomosis bleeding in the POD 25. The mean hospital stay was 12.3 days (8-25). The mean operative time was 224 min (170-310). There were no intraoperative complications. The main intraoperative blood loss was 220 ml (180-400) and intraoperative blood transfusions were not required. The resection margins were negative (R0) in 100% of cases and the mean lymph nodes harvested were 24 (18-40). The LPD is still a not common practice. Our results are comparable with those reported in literature about the open technique. These remarkable surgical outcomes are probably related to the extremely careful preoperative patient selection performed. The indication for a laparoscopic vs. an open pancreaticoduodenectomy was based on a CT scan pancreas performed less than 30 days before the planned date of surgery and a careful preoperative assessment. A low complication rate and a relative short stay in hospital were associated to a good quality of life in the early postoperative period and an early referral for postoperative chemotherapy. Good clinical outcomes were associated with outstanding oncological results.
Conclusions: Laparoscopic pancreaticoduodenectomy is a feasible surgical procedure. Remarkable oncological and surgical outcomes can be achieved with a morbidity and mortality rate in line with the data reported by the large series of open procedures.