Pancreatic and duodenal injuries

Introduction

Injuries to the pancreas and the duodenum are some of the most feared injuries in trauma. Because of the location of the pancreatoduodenal complex (PDC) and the surrounding structures, even simple injuries are complicated and have the potential of major long-term consequences (see Chapter 2 ). In this chapter, we will review the history of pancreatic and duodenal injuries, as well as current recommendations for diagnosing and managing injuries through a variety of approaches, ranging from nonoperative to damage control techniques. Throughout this discussion, it is imperative to remember that success in managing injuries to the PDC depends on having a high index of suspicion and using safe operative principles.

History

Injuries to the PDC were originally described in autopsy observations. The first reported blunt pancreatic injury was published by Travers in 1827. By the early 1900s, there were only 45 reported cases, but this quickly grew in the 1920s after Walton described distal pancreatectomy (see Chapter 117B ). Although many modifications have been made to this procedure over the last 100 years, the principles hold true: adequate exposure, hemostasis, and wide drainage.

As often is true in trauma surgery, we did see a reduction in pancreatic mortality through the experience of war. Only a handful of reports are available for pancreatic injuries in the Civil War and World War I, but by World War II, there were 62 reported duodenal injuries with an associated 56% mortality. In the Korean War, the number of cases decreased to 9 cases, but the mortality improved to 22%. Duodenal injuries have been notoriously difficult to manage throughout history. The earliest reported series of traumatic duodenal ruptures came from Berry and Giuseppi, who reported 29 patients with duodenal injuries, all of whom died. The largest series of military-related duodenal injuries to date came from World War II, which identified 118 cases and described an associated 56% mortality.

Despite high levels of motor vehicle collisions and penetrating trauma, pancreatic and duodenal injuries remain rare. A review of the United Kingdom’s Trauma Audit and Research Network database found that PDC trauma occurs in 0.32% of cases, with a male-to-female ratio of 2.5:1. Unfortunately, these cases are often lethal, with a reported mortality of approximately 20% in both pancreatic and duodenal trauma, irrespective of the mechanism. This high rate of mortality is often due to exsanguination secondary to injuries of other organs. One-third of patients with an injury to the pancreas or the duodenum will have a concomitant injury, with 58% having an injury to the stomach, 57% having an injury to the liver (see Chapter 113 ), and 35% having a vascular injury. Variables associated with predicting mortality include increased age, increased injury severity score, and hemodynamic compromise

Surgical anatomy of the pancreas and duodenum: A trauma surgeon’s perspective

The pancreas and the duodenum are uniquely situated in that they are both surrounded by critical structures and both are not immediately accessible. It is the complexity of their structures that requires a detailed anatomic understanding of both the pancreas and duodenum (see Chapter 2 ) separately, and as the combined PDC.

The duodenum is a mostly retroperitoneal structure that begins at the pylorus and extends to the ligament of Treitz. The second portion of the duodenum contains the ampulla of Vater, which is where the pancreatic ducts enter the duodenum to allow the flow of pancreatic fluid and bile. The proximal pancreas consists of the head, uncinate process, and neck. The head of the pancreas is held within the C loop of the duodenum. The uncinate process extends to the left behind the superior mesenteric vessels. The distal pancreas consists of the body and tail; the body of the pancreas rests over the aorta, and the tail of the pancreas extends to the splenic hilum.

The vasculature of the pancreas and duodenum is robust (see Chapter 2 ). The PDC is fed by the gastroduodenal, superior mesenteric, and splenic arteries, with substantial collateralization. The gastroduodenal artery branches into the superior pancreaticoduodenal artery, which further divides into anterior and posterior branches. The inferior pancreaticoduodenal artery is a branch from the superior mesenteric artery, and it also subdivides into anterior and posterior branches. There are multiple variations to the blood supply (see Chapter 2 ), with 15% to 20% of patients having a replaced right hepatic artery from the superior mesenteric artery. The splenic artery runs posterior to the pancreas along the length of the body and tail. All of these vessels originate in close proximity to each other, which can make controlling hemorrhage difficult in the setting of a devastating trauma.

Diagnosis of pancreatic and duodenal trauma (imaging and grading)

The initial workup in any trauma situation begins with the primary and secondary assessment as described in the American College of Surgeons Advanced Trauma Life Support (ATLS) course. Even with the trauma evaluation algorithm taught in ATLS, it is essential to have a high level of suspicion when assessing for pancreatic and duodenal injuries. If the patient is hemodynamically unstable, and the source of distress based on physical examination is intraabdominal, the patient should be explored immediately. If the patient is hemodynamically stable or the source of distress is unclear, it is reasonable to consider additional imaging. In the trauma setting, the first imaging available will often be a chest x-ray or a focused abdominal sonography for trauma (FAST) examination. The FAST examination is an ultrasound that looks for free fluid in the right upper quadrant, left upper quadrant, and pelvis and signs of pericardial effusion or tamponade. The FAST will often be of limited value for diagnosing pancreatic and duodenal injuries because of their retroperitoneal location. If the imaging is grossly positive and the patient is hemodynamically unstable, this should be taken as sufficient evidence to proceed to the operating room for exploration. Additionally, penetrating abdominal trauma that violates the fascia also requires exploration.

In patients who are hemodynamically stable and have experienced blunt trauma, it is reasonable to pursue additional imaging. Missed injuries may result in significant morbidity and mortality. Findings suggestive of a duodenal injury include wall thickening greater than 4 mm, lack of wall continuity, periduodenal fluid, fluid in the right anterior pararenal space, diminished bowel-wall enhancement, and extraluminal air or contrast. Multidetector computed tomography (MDCT) imaging is 98% accurate for stomach and duodenal perforations. The FAST examination has mostly replaced diagnostic peritoneal lavage (DPL), but some do consider DPL a helpful adjunct in an otherwise stable trauma patient with unexplained findings on computed tomography (CT). In this situation, DPL can be as high as 100% sensitive for bowel injuries. More commonly, diagnostic laparoscopy or exploratory laparotomy are used in these cases.

Pancreatic injuries are more complicated to identify. Overall, CT imaging has a sensitivity of 70% to 95% for detecting pancreatic injuries. With MDCT, imaging accurately identified the grade of pancreatic injury in 91% of patients. MDCT imaging has the greatest sensitivity in the detection of main duct injuries and is 97.9% sensitive in the parenchymal phase and 100% sensitive in the portal venous phase at identifying pancreatic duct injuries. This suggests that the portal venous phase is the best contrast phase to assess the pancreas in the setting of blunt trauma (see Chapter 17 ).

The management of pancreatic injuries is primarily based on pancreatic duct involvement. Consequently, it is imperative that the status of the duct be assessed in patients where there is a concern for injury, such as those with parenchymal lacerations with moderate peripancreatic fluid. In patients who are hemodynamically stable, endoscopic retrograde cholangiopancreatography (ERCP) or magnetic resonance cholangiopancreatography (MRCP) (see Chapters 17 and 30 ) may be of value. ^, ERCP allows for an intervention to be completed at the same time as diagnosis; however, ERCP is an invasive procedure with its own risk of complications and it may miss injuries to nearby structures. MRCP is noninvasive and may be used to assess additional structures, but it does not offer a means to intervene. The sensitivity of MRCP can be enhanced with the addition of secretin stimulation.

Some patients will require surgical intervention before imaging is available. In this case, it is important to assess the duodenum and pancreas intraoperatively. Later, how to best visualize the duodenum and pancreas will be discussed, but the principal goals are to mobilize the duodenum to facilitate a circumferential evaluation and to fully visualize the pancreas to assess for any obvious ductal injury. If there is any concern for a ductal injury, it is reasonable to proceed with an intraoperative fluoroscopic cholecystocholangiograph. This may be performed by inserting an angiocatheter into the gallbladder and injecting 20 to 30 mL of water-soluble contrast. The visualization may be made easier by mixing methylene blue with the contrast material. Intravenous morphine can also be used to promote contraction of the sphincter of Oddi to help visualize the pancreatic ducts. An intraoperative ERCP also may be considered if the patient is hemodynamically stable. Historically, pancreatograms were obtained by creating an enterotomy in the duodenum to cannulate the ampulla of Vater; however, this has fallen out of favor and is not recommended by the American College of Surgeons (ACS). However, one scenario in which the pancreatogram is still supported by the ACS is if the distal pancreas is already injured, in which case is reasonable to cannulate the distal duct for a contrast or methylene blue study.

The American Association for the Surgery of Trauma (AAST) has provided a grading system for the pancreas and the duodenum. This grading system is helpful in operative decision making, as well as prognostication ( Tables 114.1 and 114.2 ). Serial laboratory tests may be of value for individuals with potential pancreatic injuries but for whom imaging was not definitive and a reliable examination cannot be followed. In these cases, serum amylase and lipase should be obtained at admission and trended every 6 hours. A normal initial serum lipase has a negative predictive value of 99.8%, but it is greatly limited in its ability to predict injuries, with a positive predictive value of only 3.3%. Persistently elevated or rising serum amylase and lipase levels, however, have a combined 100% specificity and 85% sensitivity for predicting pancreatic injury and may be of some value.