A 47-year-old woman is the restrained driver in a high-speed motor vehicle crash. She has a prolonged extrication time and is brought to the trauma center boarded and collared. She is hemodynamically stable, but she complains of right lower extremity pain. On examination, she has an obvious deformity of the right femur and bruising on her left shoulder and anterior abdomen superior to the umbilicus. Her abdomen is soft, with mild periumbilical tenderness.
Blunt bowel or mesenteric injury (BBMI) occurs in approximately 1% of blunt traumas and is the third most common intra-abdominal injury, surpassed only by trauma to the liver and spleen. The most common site of gastrointestinal injury involves the proximal jejunum, followed by distal ileum, mid small bowel, colon, duodenum, and stomach, in decreasing order.1 The incidence of blunt small bowel injury rose steeply with the introduction of high-speed travel after WWII, and again spiked with the first seatbelt laws in the 1970s.2,3 Motor vehicle crashes account for 70% to 85% of all BBMIs, followed by other rapid deceleration mechanisms including pedestrians struck by cars, bicycle crashes, assaults, falls, blasts, and horse kicks. In pediatric cases, child abuse should also be considered.
Blunt bowel injury causing perforation carries a high mortality, which correlates directly with time to repair. Perforation repaired within 8 hours carries a mortality of 4%. A delay of as little as 8 to 12 hours, however, has a mortality of 9%, and at 24 hours it rises to 15%. Morbidity also increases, with the risk of abscess and surgical site infection quadrupled after 24 hours.4,5
Deceleration injuries inflict damage via a combination of three types of forces. A crushing injury causes direct tissue damage, including lacerations, perforations, and hematomas. This often occurs when the bowel is compressed between two fixed objects, such as the spine posteriorly and any other anterior object (steering wheel, seatbelt, or bike handlebar). Shearing forces between fixed retroperitoneal structures and free-floating bowel can also tear the bowel wall or its supplying mesentery. This most commonly occurs at the proximal jejunum near the fixed ligament of Treitz, the terminal ileum near the retroperitoneal cecum, or the transverse and sigmoid colons at their retroperitoneal attachments.6 Lastly, bursting injuries occur when intraluminal pressure exceeds the bursting strength of the small bowel wall, generally thought to be 120 to 140 mm Hg. This intraluminal pressure can easily be accomplished by relatively small external forces when the bowel lies in a closed-loop position.7
Clinical findings of BBMI include abdominal tenderness, rigidity, and absent bowel sounds. Unfortunately, these symptoms are vague, and therefore insensitive and nonspecific to bowel injury. In fact, studies have shown that abdominal tenderness and rigidity are present in less than 50% of BBMI cases.8 Furthermore, relying on physical examination alone requires mandatory hospital admission, repeated examinations by the same observer for best results, and a neurologically intact patient with minimal distracting injuries. Even in the awake evaluable patient, clinical examination will miss up to 20% of clinically significant injuries.9 Abdominal bruising, on the other hand, has higher sensitivity. The seatbelt syndrome (seatbelt use, bruising, and vertebral fractures) is well described, and is associated with BBMI in up to 21% of cases.10,11 Chance fractures and abdominal wall tears are also commonly associated. Despite this, the physical examination remains quite unreliable, and the use of it as the sole indication for surgery has led to negative laparotomies in up to 40% of cases, which carries an unwarranted morbidity of 5% to 20%.12,13
See Table 29–1.
WORKUP AND CHOICE OF IMAGING
BBMI is relatively rare and difficult to diagnose, especially in this era of increasingly conservative management. It is important for the surgeon to be aware of the relative strengths and shortcomings of the available diagnostic modalities.
Laboratory tests are unreliable in the diagnosis of BBMI. The white blood cell count is often elevated in trauma patients at presentation, regardless of the presence or absence of bowel perforation. A leukocytosis that does not correct with time can suggest injury; however, reliance upon this method can lead to a delay in diagnosis. Amylase has been reported to be significantly higher in BBMI than non-BBMI, but no reliable cutoff has been established.11
Diagnostic peritoneal lavage (DPL) or aspirate has also been used as an adjunct to BBMI diagnosis. It has a sensitivity of >90% for both hemoperitoneum and bowel perforation. Its sensitivity increases with the time from injury, and so it plays a lesser role at initial presentation. Today, DPL can help define the etiology of free fluid detected by another modality. The traditional definition of a positive DPL is fluid recovered from the lavage with a red blood cell (RBC) count >100,000/mm3; a white blood cell (WBC) count >500/mm3, amylase >100 IU/L, or the presence of bile, bacteria, or particulate matter. These criteria confer a sensitivity of nearly 100%; however, the positive predictive value is only 35%. Several attempts to refine the positive DPL criteria have been made, and results show that a ratio of WBCs to RBCs of greater than 1:150 has a sensitivity of 97% and specificity of 99% when performed greater than 3 hours from time of injury.14 Several other iterations have been proposed, including alkaline phosphatase levels and ratios comparing aspirate counts to venous blood. DPL does have several drawbacks, including the risk of bowel perforation or vascular injury, and the introduction of air and fluid into the abdomen may obfuscate subsequent imaging (Table 29–2 and Figures 29–1 and 29–2).
Illustration of an infra-umbilical approach to diagnostic peritoneal lavage. This is an open approach by which the peritoneum is identified and grasped. The catheter is directed into the most dependent area within the pelvis where fluid might be expected to pool. It is then connected to a bag of saline, which is then infused via gravity drainage. (Reproduced with permission from Brunicardi FC, Andersen DK, Biliar TR, Dunn DL, Hunter JG, Matthews JB, Pollock RE. Schwartz’s Principles of Surgery, 9th Ed. New York, NY: McGraw-Hill; 2010.)
Positive diagnostic peritoneal lavage (DPA), showing at least 10 cc of intra-abdominal frank blood. (Reproduced with permission from Brunicardi FC, Andersen DK, Biliar TR, Dunn DL, Hunter JG, Matthews JB, Pollock RE. Schwartz’s Principles of Surgery, 9th Ed. New York, NY: McGraw-Hill; 2010.)
Focused assessment with sonography for trauma (FAST) was first coined in 1971 and is now a readily available and reliable bedside diagnostic tool for trauma surgeons. Surgeon-performed FAST examinations have been validated against radiologists in multiple observational studies.15 It is most useful—when combined with chest and pelvic radiographs—in identifying the most common sources of hemodynamic instability: cardiac tamponade, hemothorax, tension pneumothorax, hemoperitoneum, or pelvic fractures with retroperitoneal bleed. Each of these immediately life-threatening injuries can be diagnosed within minutes of presentation without leaving the trauma bay. FAST has a sensitivity of 80% to 86% and specificity of 95% to 99.7% for detection of at least 50 cc of free fluid.16,17 Because of its low sensitivity, it has less of a role in the stable patient able to undergo further testing, and its role in the workup of BBMI is minimal. The character of the free fluid cannot be reliably identified, and so FAST cannot differentiate blood from enteric fluid. Its positive predictive value for BBMI is 38%.11 A negative FAST scan in no way rules out injury. For further discussion of the FAST examination, please refer to Chapter 3: Bedside Ultrasound for Surgeons, and Chapter 26: Intra-Abdominal Hemorrhage (Figure 29–3; see Figure 3–1).