ANATOMY

A complete understanding of pancreatic relational anatomy is essential for providing appropriate treatment and understanding the potential for associated injuries. The pancreas is about 15–20 cm in length, 3.1 cm wide, and 1–1.5 cm thick. The average mass is 90 g (ranging from 40 to 180 g).⁵ The inferior vena cava, aorta, left kidney, both renal veins, and right renal artery lie posterior to the pancreas. The head of the pancreas is nestled in the duodenal sweep, with the body crossing the spine and the tail resting within the hilum of the spleen. The splenic artery and vein can be found along the superior border of the pancreas. The superior mesenteric artery and vein reside just behind the neck of the pancreas and are enclosed posteriorly by the uncinate process. This process can be absent or can almost completely encircle the superior mesenteric artery and vein.

The head of the pancreas is suspended from the liver by the hepatoduodenal ligament and is firmly fixed to the medial aspect of the second and third portions of the duodenum. A line extending from the portal vein superiorly to the superior mesenteric vein inferiorly marks the division between the head and the neck of the gland. The neck of the pancreas measures approximately 1.5–2 cm in length and lies at the level of the first lumbar vertebra. It overlies the superior mesenteric vessels and is fixed between them and the celiac trunk superiorly. The body of the pancreas is technically defined as that portion of the pancreas that lies to the left of the superior mesenteric vessels. There is no true anatomic division between the body and the tail, nor is there any imaginary dividing line as in the case of the head and neck.

The main pancreatic duct of Wirsung originates in the tail of the pancreas and typically traverses the entire length of the gland and joins the common bile duct before emptying into the duodenum. Throughout its course in the tail and body, the duct lies midway between the superior and inferior margins and slightly more posterior. The accessory duct of Santorini usually branches out from the pancreatic duct in the neck of the pancreas and empties separately into the duodenum. A significant number of anatomic variants exist and must be recognized: (1) in 60% of individuals, the ducts open separately into the duodenum; (2) in 30%, the duct of Wirsung carries the entire glandular secretion and the duct of Santorini ends blindly; and (3) in 10%, the duct of Santorini carries the entire secretion of the gland and the duct of Wirsung is either small or absent. In all cases, the ducts lie anterior to the major pancreatic vessels.

The arterial and venous blood supply of the pancreas is relatively constant. The arterial blood supply of the pancreas originates from both the celiac trunk and the superior mesenteric artery. The blood supply to the head of the pancreas appears to be the greatest, with less flow to the body and tail and the least to the neck. The veins, like the arteries, are found posterior to the ducts, lie superficial to the arteries, and parallel the arteries for the most part throughout their course. The venous drainage of the pancreas is to the portal, splenic, and superior mesenteric vein.

PHYSIOLOGY

The pancreas is a compound tubuloalveolar gland with both endocrine (insulin, glucagon, somatostatin) and exocrine (digestive enzyme precursors, bicarbonate) function. The endocrine cells are separated histologically into nests of cells known as the islets of Langerhans. There are three predominant subtypes of islet cells: alpha cells (which produce glucagon), beta cells (which produce insulin), and delta cells (which produce somatostatin). Although these cells are distributed throughout the substance of the pancreas, most reside primarily within the tail. Consequently, it would seem that a distal pancreatectomy would be poorly tolerated in terms of endocrine function. However, it is well known that resection of more than 90% of the pancreas must occur before endocrine insufficiency develops, provided the remainder of the gland is normal. In fact, partial resection induces hypertrophy and increased activity of the residual islet cells. In animal studies, Dragstedt⁶ was the first to show that removal of 80% of the pancreas did not significantly alter carbohydrate or fat metabolism or the digestion and/or absorption of food, provided that the remaining gland is normal and that pancreatic secretions still have access to the upper digestive tract via the ductal system.

DIAGNOSIS

It is important to remember that whenever there is trauma to the pancreas, particular attention must be given to the possibility of a major ductal injury for this is the single most important determinant of outcome after pancreatic injury. In fact, this concept was first recognized as early as 1962.⁷ Subsequent investigators have confirmed and reemphasized the necessity of determining the status of the pancreatic duct. In fact, Heitsch et al.⁸ found that distal resection of ductal injuries significantly lowered postoperative morbidity and mortality when compared with drainage alone. This finding was confirmed over a decade later when investigators documented a drop in mortality rate from 19%–3% after pancreatic resection proximal to the site of ductal injury.⁹

Successful diagnosis of a pancreatic injury requires a high index of suspicion. The mechanism of injury, need for laparotomy, and time interval following initial abdominal insult will direct the trauma surgeon to the most appropriate procedures and tests. Those patients with need for immediate laparotomy require little or no preoperative evaluation, as the diagnosis of pancreatic injury can be made at the time of exploration. Conversely, patients without clear need for operative exploration may require extensive efforts to establish the presence of a pancreatic injury.

Pancreatic injuries typically result from high-energy transfer to the upper abdomen. In adults, motor vehicle accidents are the primary cause of pancreatic injuries, usually secondary to impact of the steering wheel. In children, the typical scenario involves a handlebar injury to the epigastrium. In any case, the energy of impact is directed at the upper abdomen (epigastrium or hypochondrium), resulting in crushing of the retroperitoneal structures. Typical findings suggestive of retroperitoneal injury include contusion/bruising to the upper abdomen with epigastric pain out of proportion to physical examination.

Elevated serum amylase is not a reliable indicator of pancreatic trauma. In fact, the use of amylase as a screening tool in blunt trauma carries a negative predictive value of 95%.¹⁰ Measurement of the pancreatic isoamylase fraction has failed to substantially improve both the sensitivity and specificity of this value as a marker of pancreatic injury.

Asymptomatic patients with elevated serum pancreatic isoamylase require observation and repeat amylase determination. Persistently elevated serum amylase or the development of abdominal symptomatology warrants further investigation and may include computed tomography (CT) scan, endoscopic retrograde cholangiopancreatography (ERCP), or operative exploration. Abdominal CT scans have a reported sensitivity and specificity as high as 80% in diagnosing pancreatic injury.¹¹ Patton and colleagues¹² reported that in 26 patients that sustained blunt pancreatic trauma, early CT scan was suspicious for injury in 15. CT failed to demonstrate injury in four patients (21%), resulting in a delay in operative intervention (mean, 3.8 days). The remaining patients had other indications for exploration.

Computed tomography findings diagnostic of pancreatic injury include parenchymal disruption, intrapancreatic hematoma, fluid in the lesser sac or separating the splenic vein and body of the pancreas, peripancreatic edema, thickened left anterior renal fascia, and retroperitoneal hematoma and/or fluid. Clearly, certain findings are more reliable than others and rarely are all present in a single patient. In fact, some of the CT signs of pancreatic injury may not be immediately apparent following injury, but rather require time to develop postinjury. It is important to remember this when evaluating the patient with worsening abdominal symptoms and an unimpressive initial CT scan.

Endoscopic retrograde cholangiopancreatography can be useful in the diagnosis of pancreatic duct rupture. In addition, it can aid in the diagnosis of and occasionally the management of the complications of missed pancreatic injuries. A report from the University of Louisville documents ERCP as a useful diagnostic tool in the evaluation of the pancreatic duct in the early postinjury period in hemodynamically stable patients with elevated amylase levels, persistent abdominal pain, and abnormal or questionable abdominal CT findings.¹³ ERCP is also extremely helpful in the evaluation of patients in whom the diagnosis of pancreatic injury was missed during the initial evaluation. It is in these patients that ERCP can aid in diagnosing the injury, planning the surgical approach if necessary, determining internal transpancreatic stent placement, and transductal drainage of a pancreatic abscess. However, ERCP may not always be available and should not delay operation in patients with progressive clinical deterioration.

Magnetic resonance (MR) imaging, specifically MRCP (magnetic resonance cholangiopancreatography), has emerged as an alternative technique for evaluating the pancreatic duct. Although primarily used in elective circumstances, MRCP has been reported as a viable option for evaluating the status of the duct in those patients with pancreatic injuries.¹⁴ However, it frequently is not practical for use in trauma patients.

In order to successfully diagnose the presence and extent of a potential pancreatic injury, the surgeon must recognize those findings associated with pancreatic injury and adequately visualize the entire gland. In addition, it is also imperative to determine the integrity of the pancreatic parenchyma and status of the major pancreatic duct. Pancreatic injuries are classified based on the status of the duct and the anatomic location of the injury within the gland. Associated injuries often complicate pancreatic evaluation. The presence of a central retroperitoneal hematoma or a hematoma overlying the pancreas, retroperitoneal saponification or bile staining mandates complete pancreatic exploration.

Once again, it must be stressed that, if possible, it is important to determine the status of the duct at the time of exploration. Most of these injuries can be diagnosed by local exploration. Injuries to the duct occur in approximately 15% of pancreatic trauma and are generally the result of penetrating injury.¹⁵ Blunt injury can also result in transection of the major duct with or without complete transection of the gland. Minor contusions and/or lacerations of the pancreatic parenchyma usually do not require further evaluation of the duct. However, an intact pancreatic capsule does not eliminate the possibility of complete transection of the pancreatic duct.⁹

The use of intraoperative observations such as direct visualization of ductal disruption, complete transection of the substance of the gland, free leakage of pancreatic fluid, lacerations involving more than one-half of the diameter of the gland, central perforations, and severe lacerations with or without massive tissue disruption can predict the presence of a major ductal injury with a high degree of accuracy. However, in those instances in which the status of the duct is uncertain, intraoperative pancreatography has been used as a technique for visualization of the main pancreatic duct. While intraoperative pancreatography may sound appealing, it is frequently impractical.

Nevertheless, pancreatography can be performed either by directly cannulating the ampulla of Vater through a duodenotomy or the main pancreatic duct through the amputated tail of the pancreas. A 5F pediatric feeding tube is used along with 2–5 ml of contrast. Cannulating the ampulla of Vater entails creating a duodenotomy unless there is an associated duodenal injury. It should be stressed that identifying the ampulla can be difficult and that resection of the tail does not always ensure visualization of the pancreatic duct.

The simplest technique is a needle cholecystocholangiogram. In this technique, a purse-string suture is placed in the gallbladder just proximal to the cystic duct. An 18-gauge angiocatheter is then introduced into the gallbladder. The remainder of the gallbladder can be excluded with a bowel clamp. Water-soluble contrast is injected into the gallbladder under direct fluoroscopy. A cholecystectomy is not necessary following this procedure.