Laparoscopic Pancreatic Surgery

The laparoscopic management of pancreatic disease is one of the most challenging in laparoscopic surgery. This is especially true when considering that of pancreatic resection. Well-trained laparoscopic surgeons have found that operating on the pancreas, like virtually all intra-abdominal procedures, is technically feasible. Laparoscopic principles suggest that the patient will probably benefit from less postoperative pain, improved wound cosmetics, quicker return to routine activities, and shorter hospital stay. Ultimately the acceptance of many laparoscopic operations will be determined by their degree of difficulty, the operating time, the cost (both hospital and societal), and patient outcomes.

In comparison with the literature available on other laparoscopic operations, the information available on pancreatic resection is too scant to draw firm conclusions. However, leaders in the field have demonstrated that pancreatic resection is feasible, and are carefully examining their outcomes to further elucidate the role of this technically demanding procedure.

Laparoscopic procedures for the pancreas fall into four main categories:

1. Laparoscopic staging of pancreatic malignancy
2. Bilioenteric or gastroenteric bypass
3. Pancreatic resection, and
4. Management of pancreatic pseudocysts

Anatomic considerations

The majority of the pancreas lies in a retro peritoneal position, transversely oriented from the second and third portions of the duodenum to the hilum of the spleen. Anterior access to the gland (body and tail) is readily obtained by division of the gastro colic omentum. This division may be performed by electrocautery, multiple individual clip applications or vascular stapling devices, or ultrasonic dissection.

Access may be obtained through the gastrohepatic ligament, although the exposure is usually less adequate. The patient is positioned in a slight head-up position to allow gravity retraction of the viscera. An oblique angle (30° or 45°) telescope is necessary for adequate visualization. Laparoscopic ultrasound is proving to be an essential tool for many aspects of pancreatic surgery

Laparoscopic staging of pancreatic malignancy

Patients with pancreatic malignancy generally present at later stages of disease. Frequently the disease is unresectable due to tumor size or tumor metastases by the time symptoms occur. Surgical resection for pancreatic cancer still offers the only reasonable chance at a cure. Historically; many patients underwent unnecessary laparotomy in an effort to assess resectability. CT scans have helped many patients avoid the morbidity of a non therapeutic laparotomy. However, even with this modality, unresectability rates at laparotomy can approach 60%. This is most often due to the presence of unrecognized peritoneal metastases (<1 cm) and tumor invasion not appreciated on CT scan. Spiral CT and magnetic resonance imaging (MRI) are more reliable for predicting unresectability, but are still not adequate in our opinion. In a large multicenter study, Megibow and coworkers  reported a sensitivity of 77%, a specificity of 50%, and an overall accuracy of 73% for dynamic CT scanning. Also in their study, they found no additional benefit from MRI.

Diagnostic laparoscopy further narrows patient selection for therapeutic laparotomy. Warshaw and coworkers found that an additional 35% of patients could avoid laparotomy with the use of diagnostic laparoscopy. Despite improving non-invasive imaging methods since Warshaw and coworker's early reports, more recent studies confirm Warshaw and co-worker's initial findings that a significantly number of patients (22% to 35%) can avoid laparotomy with the use of staging laparoscopy.

Further, the sensitivity for evaluation of unresectable disease appears further enhanced with the addition of the laparoscopic ultrasound to the laparoscopic staging procedure. Callery and coworkers use a multifrequency laparoscopic ultrasound probe to search for occult metastases and assess posterior invasion into vascular structures like the portal vein. Tumors other than pancreatic were included. Fifty patients were referred for staging laparoscopy after interpretation of conventional noninvasive imaging modalities had determined the tumor to be resectable. Laparoscopic ultrasound established unresectability in 11 patients (22%) in whom staging laparoscopy alone was negative. In another study by John and coworkers involving 40 consecutive patients with pancreatic cancer presenting for diagnostic laparoscopy, laparoscopic ultrasound found an additional 25% (10 patients) whose disease was unresectable when compared with laparoscopy alone. They found the use of ultrasound significantly improved specificity and accuracy as compared with laparoscopy alone (88% and 81 % vs 50% and 60%, respectively).

Staging laparoscopy technique

Patients generally undergo staging laparoscopy on the same day they are scheduled for resection. Patients are placed in the supine position on an electrically equipped bed (preferably). A 10-mm trocar is placed in the infraumbilical position to serve as the camera port. The abdomen is insufflated to 15 mm Hg. A 30° laparoscope is used. A second port of 5 mm is placed in the right midclavicular line several centimeters from the subcostal margin. A four-quadrant exploration is then carried out. Grasping devices, biopsy forceps, or electrocautery instruments may be alternatively introduced through the 5-mm port. Important peritoneal surfaces to visualize for areas of metastases include the undersurface of abdomen including falciform, diaphragm, and liver. The omentum must be examined thoroughly and when possible retracted superiorly to evaluate the base of the transverse colon, its mesentery, and the ligament of Treitz (this may require an additional port).

If there is evidence of unresectability, the procedure is terminated. Otherwise laparoscopic ultrasound is carried out. A second 10mm port is placed in the right midclavicular line at the level of the umbilicus. Laparoscopic ultrasound is then performed using a 9mm in diameter linear array 7.5-MHz contact ultrasound probe with Doppler flow capability. The liver is systematically scanned (anterior, lateral, inferior) at penetration depths of 7 cm for evidence of metastatic spread or extent of primary tumour invasion. Frequently; biliary and pancreatic metastases to the liver have a characteristic bulls-eye appearance with an echoic rim encircling a mixed-echo tumour center. If found, biopsy for such lesions may be attempted percutaneously with laparoscopic ultrasound guidance.

Attention is then turned to ultrasonic evaluation of the portahepatic, peripancreatic, para-aortic, and celiac axis for evidence of nodal disease. Lymph nodes greater than 10 mm may be biopsied. Laparoscopic ultrasound with Doppler flow capability is then used to help locate and assess the potential for tumor extension to surrounding peripancreatic vascular structures (primarily portal vein, but also superior mesenteric vein and artery, and celiac axis).

Bilioenteric or gastroenteric anastamosis for pancreatic malignancy

Unresectable patients might be candidates for biliary or enteric bypass. The risk and benefits of bypass must be weighed against existing palliative options, the patient's condition, existing or impending obstruction, and expected length of survival based on tumor burden. For most patients with unresectable disease, life expectancy can be expected to be less than 1 year. Proper management tailored to the individual patient's needs is important so as to offer as much quality of life free from hospitalization as possible.

Commonly; patients will present with some degree of biliary obstruction or will suffer from it during the couise of the disease. Most patients with obstructive jaundice are best treated by placing an endoscopic or percutaneous stent. The success rate is high (85%), with a low associated mortality (1% to 2%). Studies comparing open bypass with those stented endoscopically for obstructive jaundice found no advantage to the surgical approach . Morbidity from stent placement includes potentially frequent admission to hospital (occlusion, infection) and significant cost for endoscopic retrograde cholangiopancreatography (ERCP) and stent. However, repeat placement has become less necessary with the use of improved techniques and stent design. Patients may present or develop distorted duodenal anatomy that makes initial or subsequent stent placement impossible. This finding may be coupled with gastric outlet obstruction. In these patients, bypass procedures may be offered after evaluation of surgical risk or life expectancy.

The morbidity of open surgical bypass is substantial (19%). Laparoscopic biliary (cholecystojejunostomy) or gastric bypass (gastrojejunostomy) is feasible. There is potential for shorter recovery, shorter return to activity; and low morbidity, as evident in several small studies. We favor an approach, as suggested by Nathanson, that the bypass should be reserved for a later date from the diagnostic laparoscopy at such time when duodenal obstruction precludes repeat stent or there is stent failure (blockage, recurrent sepsis). For the stomach, failure would include when symptoms of gastric outlet arise. Conditions at initiallaparoscopy that might argue for immediate bypass include inability to stent the biliary system in the preoperative setting, endoscopic or radiologic evidence of impending duodenal obstruction, or laparoscopic impression of large locally advanced mass with minimal to no evidence of metastatic spread.

Biliary and gastric bypass

Cholecystojejunostomy may be carried out if the gallbladder is present and suitable for anastomosis, and the cystic duct is patent and its junction to the common bile duct (CBD) is far from the tumor. Frequently this information is available by preoperative imaging studies (ERCP or percutaneous transluminal cholangiography). If not, patency of cystic duct and its relation to primary tumour location may be obtained by performing a cholangiogram after cannulation of the gallbladder. Similarly, laparoscopic ultrasound may be used for such an assessment.

For either anastomosis, patients are positioned supine and the port placement is the same. A 10-mm trocar is placed at the inferior umbilical region and a 30° telescope is used. Additional ports and operating room personnel are positioned.

The omentum and transverse colon are elevated with instruments introduced through the epigastric and either 12-mm port. The small bowel is traced back to the ligament of Treitz. A loop of small bowel is then chosen that will comfortably reach stomach and gallbladder without tension (note that this is true once the transverse colon and omentum are allowed to return to normal position). For the biliary bypass, a cholecystotomy is performed with electrocautery on the gallbladder fundus. The biliary contents are then aspirated. An enterotomy is performed on the antimesenteric surface of the chosen small bowel loop. A 30mm endoscopic stapler is introduced through the right 12-mm port. The jaws of the stapler are opened and one arm of the stapler is inserted into the enterotomy. The jaws of the stapler are then closed to function as a large grasper. The stapler and small bowel contained within are then maneuvered adjacent to the cholecystotomy. The jaws of the stapler are opened again and the free arm of the stapler maneuvered into the cholecystotomy. Assistance is provide by a blunt grasping instrument inserted through the additional ports (epigastric). After proper alignment is assured, the stapler is fired to complete the anastomosis. The original sites may be closed with additional firings of the stapler. At this point the endoscopic stapler will be introduced through the left 12-mm port. Care must be taken not to narrow the anastamosis or the lumen of the small bowel significantly.

To fashion the gastric bypass, a dependent site is chosen along the greater curvature. The gastrocolic omentum is divided close to the greater curve within the gastroepiploic arcade for a distance of approximately 3 to 4 cm with the ultrasonic scalpel or by electrocautery. A gastrotomy is made on the greater curvature. The anastomosis will be formed along the greater curve but will extend into the posterior wall of the stomach. Typically, the stapled anastomosis will be created by introducing the stapler through the right 12-mm port. The anastomosis should consist of two firings of the 30-mm endoscopic linear cutter.

Ideally, the stapled anastomosis should be aligned to cross the greater curvature to the posterior surface (ie, through the area of divided gastrocolic omentum). If fashioned in this way, the original puncture sites will be easier to close and the anastomosis more dependent.

Laparoscopic pancreatic resection

Indications for complete or partial pancreatic resection include:

1. Aadenocarcinoma
2. Insulinoma (neuroendocrine) and
3. Chronic pancreatitis

Improved technique and postoperative care have rendered morbidity and mortality for pancreatic resection, including Whipple's procedure, to less than 5%. Laparoscopic techniques could potentially lower this rate even more or at least afford less pain and a more rapid recovery.

Laparoscopic Whipple's procedure was first carried out by Gagner in a small series of three patients with various diseases (pancreatitis, ampullary cancer, adenocarcinoma). He subsequently has reported on a pylorus-preserving technique performed in one patient with pancreatitis. The initial experience indicates that it is technically feasible, but because of its operative time, complexity, and as yet no demonstrated improvement in outcome, this procedure must be considered investigational. Hand-assisted laparoscopic surgery may make pancreatic resection more practical.

Laparoscopic pancreatic procedures involving distal pancreatectomy appear to hold more promise at present. Soper and co-workers reponed success with his technique in the pig model. Gagner and coworkers successfully performed distal pancreatectomy for a variety of disease processes including islet cell tumours, cystadenocarcinoma, and pseudocyst. The spleen was preserved in all cases and operating times ranged from 2.5 to 5 hours. Cases were managed with the patient in the left lateral position, with pancreatic division carried out with a 60-mm linear cutter. Others are reporting initial success with distal resection.

Laparoscopic management of pancreatic pseudocyst

Pancreatic psuedocysts may be defined as a collection of pancreatic secretions, serous fluid, or necrotic debris surrounded by a nonepithelialized wall made up of granulation tissue and variable degrees of fibrous tissue. Pancreatic pseudocysts must be distinguished from true cysts of the pancreas, which are characterized histologically by the presence of an epithelial lining. Pseudocyst formation is the result of a post inflammatory process arising from patients with acute or chronic pancreatitis. An understanding of the natural history of pancreatic pseudocyst is important when deciding on invasive therapy versus expectant management. Studies like those by Bradley and coworkers had a great influence in the management of pseudocystic disease. Bradley and coworkers suggested the likelihood of regression diminished and the likelihood of complications rose dramatically after a 6-week period. More recent data suggest that this patient population may be watched safely for longer periods. Yeo and coworkers followed asymptomatic patients with pseudocysts by CT scanning for 1 year (48% were successfully observed with only a 2.7% complication rate). The only predictor for intervention was size greater than 7.4 +_0.6 cm.

General asymptomatic patients with pancreatic pseudocyst may be followed up for extended periods of time. This conservative approach is more likely to be successful in patients with small 6 cm pseudocysts. Other options are available for drainage procedures (eg, percutaneous transgastric, ERCP).

Laparoscopic pseudocyst drainage

Preoperative decision making and subsequent laparoscopic operative approach should mimic that of open operative planning. The selection of procedure will depend on the anatomic location of the pseudocyst, pseudocyst size , and associated pancreatic duct or distal common bile duct abnormalities.

Reports by Newell and coworkers document that pseudocyst-gastrostomy is technically easier than pseudocyst-jejunostomy, while remaining equally efficacious. Laparoscopic pseudocyst-gastrostomy is technically easier, but cyst-jejunostomy is also technically feasible for the cyst not amenable to gastric drainage by standard surgical principles.

Laparoscopic pseudocyst-gastrostomy was first performed by Petelin in 1991. Principles of operative drainage include biopsy of cyst wall to rule out neoplasm, dependent drainage, and precise hemostatic technique to avoid hemorrhage.

The patient position and port placement are the same as described for the bypass procedure. The pseudocyst may often be seen pushing the stomach forward. A small gastrotomy is established with cautery over the most prominent portion of the pseudocyst. Ultrasound may be helpful in locating the pseudo cyst and the site of the initial gastrotomy. The gastrotomy is then extended for several centimeters with electrocautery.

A small window is developed through the posterior wall of the stomach with electrocautery. One must remember that the posterior wall of stomach and cyst capsule will be fused and that this requires a deeper dissection with cautery than may feel comfortable for the surgeon. Ultrasound may be helpful in identifying the 12cation of the pseudocyst, and in some cases to plan dissection where the stomach wall or cyst is thinnest. The window is made large enough to accommodate the endoscopic stapler. A biopsy of the wall may be carried out at this time. Two firings of the stapler are used to create a substantial anastomosis (stapler insertion through the more comfortable 12-mm pon, usually the right). Hemostasis at the staple line should be assured. The gastrotomy is closed with either sutures or staples.

CONCLUSIONS

The laparoscopic approach to pancreatic surgery has rapidly been shown to be of considerable value. For splenectomy and adrenalectomy, it has become the procedure of choice for most elective surgery. The laparoscopic approach to the pancreas has value with respect to staging, bypass procedures, and pseudocyst drainage. Pancreatic resection is feasible, but must still be considered investigational.