Laparoscopic pancreaticoduodenectomy for ampullary adenocarcinoma – a case report

Summary

Ampullary adenocarcinoma (AA) is rare disease, accounting for less than 1% of all digestive cancers, although it is the second most common cancer arising from the periampullary region (after pancreatic cancer). Treatment of ampullary neoplasia varies from endoscopic resection, transduodenal surgical ampullectomy to a pancreaticoduodenectomy. At the time of diagnosis, the curative surgical resection rate of AA is higher than that of pancreatic cancer (80% to 20%) and the 5-year survival rate is also significantly longer. Laparoscopic procedures in colorectal surgery became the standard and the benefit of early post-operative outcomes is well proven. In this article, we present a 68-year-old patient with AA treated by laparoscopic pancreaticoduodenectomy (LPD). We consider the pros and cons of LPD and open pancreaticoduodenectomy, discuss possible treatment options of AA (including endoscopy treatment), and present a current review of literature.

Key words:

adenocarcinoma of the ampulla of Vater – pancreaticoduodenectomy – surgery – periampullary tumors – pancreatic cancer – laparoscopy – endoscopy

Introduction

Ampullary adenocarcinoma (AA) is quite rare, accounting for less than 1% of all digestive cancers [1,2], with a reported incidence of fewer than 1 person per 100 000 [3,4]. However, it is the second most common cancer arising from the periampullary region, comprising 6–20% of periampullary tumours [5,6]. Premalignant lesions of ampulla of Vater are generally treated by endoscopic resection. According to some authors, endosco-pic treatment is also feasible in selected patients with early carcinoma limited to mucosa [4,7]. The first endoscopic papillectomy was performed by Suzuki in 1983 [8]. Transduodenal surgical ampullectomy (TSA) is considered if premalignant lesions cannot be endoscopically resected [9,10] and in the early stage of cancer in polymorbid patients [11,12]. American surgeon, William Halsted, first performed this procedure in 1898 [11,13]. Nevertheless, in AA pancreaticoduo-denectomy (PD) is the preferred therapeutic method. PD can be performed by open surgery, which was first performed by an Italian surgeon, Alessandro Codivilla, in February 1898 (although he never published his surgery) [14]. A German surgeon, Walter Kausch, performed and first published PD for AA in 1909 [14]. Later the PD procedure was improved by an American surgeon, Allen Whipple, in 1935 [15] and is often called Whipple’s procedure. With the current increasing role of mini invasive procedures, PD is also possible to carry out laparoscopically. Laparoscopic pancreaticoduodenectomy (LPD) was first described and performed almost 100 years later after an open procedure by Canadian surgeons Gardner and Pomp in 1994 [16]. From that time laparoscopic technology and instruments have been improved, however, LPD remains one of the most advanced and challenging laparoscopic surgeries. PD is an extensive surgical procedure with high morbidity and mortality even in high volume centres. Surgical post-operative complications as well as non-surgical post-operative complications appear after PD. After laparoscopic colorectal surgery, a lower incidence of respiratory infections is observed compared to open procedures. Improved post-operative outcomes, including lower morbidity, are reasons why minimally invasive surgery is also more expandent in hepatopancreatobiliary surgery.

Case report

Our patient is a 68-year-old woman. Her case history includes hypertension and cholecystectomy in 2006 (solitary cholecystolithiasis). In 2014, the patient underwent endoscopic retrograde cholangiopancreatography (ERCP) for liver enzyme elevation and suspicion of choledocholithiasis with findings of slight stenosis of papilla (histologically chronic inflammation with fibrosis). There were no gallstones present in the bile duct and papillosphincterotomy was performed. In 2018, the patient was examined after a weight loss, liver enzyme and bilirubin elevation (alanine aminotransferase 4.9 µkat/L, aspartate transaminase 3.4 µkat/L, alkaline phosphatase 15.0 µkat/L, gamma-glutamyltransferase 21.7 µkat/L, bilirubin 33.1 µmol/L, amylase 0.34 µkat/L, carbohydrate antigen 19-9 6.7 U/ml). Due to a dilation of the biliary tract and ultrasound suspicion of a choledocholithiasis, ERCP was indicated. ERCP revealed a stenosis of ampulla of Vater and there were no lithiasis present in the biliary tract. A biopsy from the papilla and the papillosphincterotomy was performed. The biopsy confirmed adenocarcinoma. The staging CT scan has added findings of dilation of the bile and pancreatic duct (double duct sign) (Fig. 1 A, B). Due to a double duct sign we suspected invasion of the ducts. The patient was presented at a multidisciplinary committee and was indicated for surgery.

Surgery procedure

The operation started with a staging laparoscopy. After exclusion of liver metastasis, six ports for the laparoscopy procedure were placed. First, the lesser sac was opened through the gastrocolic ligament and then the lower edge of the pancreas was released; the superior mesenteric vein (SMV) was found and at its level the pancreas body was liberated from the SMV surface. After this, the pancreas was interrupted by Thunderbeat® (Fig. 2A) and the drain was inserted into the dilated pancreatic duct. Near the upper pancreatic edge, the common hepatic artery was located and a lymphadenectomy from her detachment from the celiac trunk to the hepatoduodenal ligament was performed. The gastroduodenal artery was ligated and the common dilated hepatic duct was interrupted (Fig. 2B). The duodenum was cut below the pylorus, released from retroperitoneum, followed by a dissection of the uncinate process from SMV and a radical resection was finished by dissection from superior mesenteric artery. The reconstruction phase differs from the colorectal laparoscopic procedures as most of the anastomoses are possible to carry out by stapler. In this procedure all anastomoses were done by laparoscopic suture (however, there are some authors, who prefer to perform the reconstruction phase from minilaparotomy). Our department prefers to construct the pancreatico-gastric anastomosis (PGA) to the poste-rior wall of the stomach, as was also done in this laparoscopic procedure. First, the posterior portion of PGA was performed with single stitches between the remnants of the pancreas and posterior gastric wall (Fig. 2C,D), followed by anterior and posterior gastrotomy. The pancreatic body was pulled through the posterior gastrotomy into the stomach. The anterior portion of PGA was finished by single stitches (through the anterior gastrotomy) (Fig. 2E) and finally the anterior gastrotomy was sutured (Fig. 2F). The reconstruction phase was completed by Y-Roux hepaticojejunal anastomosis (HJA) and gastrojejunal anastomosis. The specimen was extracted in a protective bag by a Pfannenstiel incision. Two drains were placed into the abdomen cavity (in proximity to HJA and PGA). PGA and HJA were augmented by autologous platelet-rich fibrin sealant Vivostat® (Fig. 2G). The patient was discharged 10 days after the surgical procedure without any post-operative complications. Definitive histology (Fig. 3A) established a diag-nosis of AA (grade 2) with invasion of duodenal wall and pancreas (T3). Resection margins, evaluated according to a Leeds pathology protocol [17,18], were negative, 2 of 11 removed lymph nodes were positive (N1) (Fig. 3B) and perineural invasion was present. Considering these risk factors we prescribed adjuvant treatment with gemcitabine for 6 months.

Discussion

Endoscopic papillectomy is globally recognised as the first method of choice in premalignant ampullary lesions [10,19]. The possibility of endoscopic therapy of early adenocarcinoma of the ampulla of Vater bounded to mucosa, without a deeper invasion to submucosa, is stated by some authors [4,7,20]. In cases where the premalignant lesions cannot be resected endoscopically [9,10] or in select patients within the early carcinoma stage (without a lymph node invasion and in polymorbid patients) [11,12,17], the TSA should be considered. However, PD remains a method of choice in AA. Due to its anatomical location, AA usually presents early with jaundice or pancreatitis [21] and due to an earlier clinical presentation, resection rates (possibility of resection) for most patients are much higher than for pancreatic cancer. Curative surgery is possible in up to approximately 80% of cases at the time of diagnosis of ampullary cancer compared to pancreatic adenocarcinoma, where the resectability rate is around 20% [2,22,23]. Easier feasibility of laparoscopic resection in the case of ampullary tu-mours compared to pancreatic cancer can be due to its earlier clinical manifestation and so, at time of diagnosis, the surrounding organs, arteries (superior mesenteric artery, celiac trunk) and veins (portal vein) are not infiltrated by tumour. In comparison with other periampullary cancers, AA has better surgical outcomes in treatment and long-term survival rate. The 5-year survival rates following resection range from 38% to 67% [24,25]. Choi et al. [26] in 2011 reported a 59.9% overall 5-year survival rate after PD (R0/R1 resection). Sudo et al. [24] even published a 64% overall 5-year survival rate after PD resection for AA. The reason seems to be due to earlier clinical diagnosis and so a higher resectability rate and less aggressive tumour characteristics compared to other periampullary cancers (pancreatic cancer, distal bile duct carcinoma) [2,24,25]. The advantages and benefits of laparoscopy for patients have been well described in other kinds of abdominal surgery (primarily in colorectal and inflammatory bowel disease surgery), such as shortened hospital stay, quicker recovery, lower morbidity, better cosmetic outcome and lower percentage of incisional hernias [27–31]. However, in hepatopancreatobiliary surgery, it still remains a challenge for surgeons and represents one of the most difficult laparoscopic procedures. What also must be considered is that in oncologic curative surgery, the radicality and complete resection of the tumour is the primary goal. Many studies have been comparing the surgical oncologic results and survival rate after laparoscopic and open pancreaticoduodenectomy (OPD). Long-term overall survival after LPD is comparable to and also appears to be true for pancreatic cancer as well [32–35]. Studies that also compared pathology results did not demonstrate any significant difference and achieved similar negative resection margins and number of lymph node harvest in LPD and OPD (in AA and also pancreatic cancer) [32,34,36–38]. PD is associated with high mortality and morbidity rate even in high volume centres. The post-operative mortality rate is between 3 and 5% [39,40]. The morbidity ranging from 40 up to 60%, primarily due to pancreatic fistula, post-operative bleeding and delayed gastric emptying [11,39,41]. Just Dokmak et al. [39] reported higher morbidity in cases of LPD. Otherwise, the same morbidity rate has been recorded in comparison of LPD with OPD [32,35,37,38,42] and some studies and systematic meta-analysis have even reported a lower occurrence of post-operative complications in LPD [33,43,44]. Adjuvant oncologic therapy should be considered after surgery resection if risk factors occur (positive lymph nodes, positive resection margins, unfavourable grading, perineural, angio or lymphovascular invasion) [17,22,45]. As presented in our case, because of positive lymph nodes and perineural invasion, the adjuvant therapy was indicated. Typical advantages of laparoscopy were confirmed in LPD, such as less abdomen pain, less blood loss, shorter hospital stay and faster post-operative recovery [32,38,42,43]. Comparison of incisions and cosmetic outcome between LPD and OPD can be seen in figure 4. On the contrary, the length of the operative time has been described as shorter in OPD in some comparative studies [35,42–44], but it is dependent on the surgeon’s learning curve of this highly advanced technical laparoscopic procedure [46]. It has to be said that most studies performed LDP on select patients. Inclusion criteria in cohorts were tumours which had not spread to the main vessels or smaller tumour size (up to 2–3 cm); excluded were patients with high cardiopulmonary morbidity or relative exclusion criterion for laparoscopy with previous history of major surgery [32,36,37,47].

Conclusion

We present a 68-year-old woman with adenocarcinoma of ampulla of Vater that successfully underwent LDP. Patients with periampullary adenocarcinoma treated by LPD have a similar long-term survival rate compared to OPD and similar or lower post-operative complications, with shorter length of hospital stay and faster post-operative recovery (and other more positive surgical outcomes). However, future prospective studies are needed because most of the current published studies are retrospective. LPD is performed optimally on select patients (smaller tumour, no vessels infiltration, as mentioned above) due to the procedure’s technical difficulties. LPD is an extremely challenging and difficult procedure with a slow learning curve and should be performed by experienced laparoscopic surgeons. Due to this, AA has a higher resectability rate than pancreatic carcinoma, and in the early stage of AA where there is usually no spreading of the tumour to mesenteric vessels and portal vein, it seems to be suitable for a minimally invasive procedure.

Supported by Ministry of Health, CZ – conceptual development of research organisation (FNBr, 65269705). Submitted/Doručeno

The Editorial Board declares that the manuscript met the ICMJE „uniform requirements“ for bio­­­­medical papers.

The authors declare they have no potential conflicts of interest concerning drugs, products, or services used in the study.

Submitted: 10. 9. 2018

Accepted: 16. 10. 2018

MUDr. Vladimír Procházka, Ph.D.

Department of Surgery, University Hospital

Brno Jihlavská 20 625 00 Brno

prochazka.vladimir@fnbrno.cz