Nonoperative management of blunt and penetrating abdominal injuries

Introduction

The evaluation and management of the abdominal cavity in the blunt trauma patient has undergone radical change over the past several decades, due both to significant technological advances as well as a critical reappraisal of management techniques and their outcomes. Early and rapid diagnosis of injuries coupled with the application of modern trauma care principles has made successful nonoperative management of most blunt abdominal injuries the rule rather than the exception. Although select patients may require immediate or delayed surgical intervention, nonoperative management can now safely be extended to the majority of patients with blunt abdominal injury regardless of age or associated injuries. However, the inappropriate use of nonoperative management and the failure to adhere to strict principles and protocols are associated with major adverse outcomes, including delayed therapy, increased morbidity, and preventable death.

Incidence

The incidence of intra-abdominal injury following blunt trauma will vary widely by the patient population, mechanism of injury, and diagnostic studies employed by the particular center. Approximately 12% of all blunt trauma patients who are screened with computed tomography (CT) have one or more intra-abdominal injuries, with 46% being major injuries and 30% requiring surgical or angiographic intervention. The vast majority will be solid organ injuries to the spleen and liver, followed by injury to the kidney, mesentery, small bowel, colon, and pancreas. These injuries may be categorized as solid organ (liver, spleen, kidney), hollow viscus (stomach, duodenum, small bowel, colon, ureter, bladder), endocrine (pancreas, adrenal), or vascular. Overall, greater than 80% of blunt solid organ injuries may be managed without surgical intervention and with similar or lower complication rates compared with operative management.

Mechanism of injury

Blunt trauma may produce abdominal injuries through a variety of mechanisms, including direct transmission of energy to abdominal structures causing tissue disruption or hollow viscus blowout, shearing from rapid deceleration, direct compression of abdominal organs against the vertebral column, and puncture or laceration from associated rib fracture, spine fracture, or foreign bodies. Although there is not a linear relationship between the degree of force and the amount of abdominal injury, mechanisms involving higher velocity and forces will result in more significant and extensive injuries to the abdominal organs. Direct transmission of force to the abdomen will predominantly be absorbed by the large solid organs, such as liver, spleen, and kidney, resulting in parenchymal disruption. Rapid deceleration forces tend to affect fixed or tethered structures such as the kidneys, duodenum, and bowel mesentery, resulting in lacerations or pedicle avulsion. Although seat belt use has resulted in a decrease in traumatic brain injury and death, there is a twofold increase in the incidence of hollow viscus injuries among belted passengers. Organs and structures that are fixed in position to or in close proximity to the vertebral column may also be injured by direct compression and include the distal duodenum, pancreas, and great vessels. Fractures of the lower rib cage may directly lacerate upper abdominal structures including the diaphragm, liver, spleen, and kidneys.

Diagnosis

The diagnosis of intra-abdominal injury in the blunt trauma patient begins with the primary survey and focused examination of the abdomen. Hypotension should be assumed to be due to hemorrhage from an abdominal injury until proved otherwise. Physical examination of the abdomen may be limited by distracting injuries or depressed mental status, but should focus on the elicitation of peritoneal signs, localized tenderness, external bruising or evidence of a “seat belt sign,” and distention. Peritonitis should never be attributed to a solid organ injury, as isolated hemoperitoneum should not cause diffuse peritoneal irritation. Focused assessment with sonography in trauma (FAST) is now commonly performed as part of the initial evaluation. Although a “positive” FAST examination reliably identifies the presence of free fluid in the abdominal cavity suggestive of injury, a negative study does not exclude significant abdominal injury and should not be considered a definitive evaluation. Although ultrasound has been used to identify and grade specific organ injuries (i.e., liver and spleen), its reliability and reproducibility in this capacity has not been well demonstrated.

The utility of trauma ultrasound beyond the standard four-view FAST examination is becoming increasingly appreciated, particularly for blunt truncal trauma. An extended FAST examination (E-FAST) that includes thoracic imaging for pneumothorax and hemothorax can be performed with little additional time compared with standard FAST. The technical details of performing an E-FAST examination can be easily taught to staff and resident-level providers and has been shown to be more accurate than the chest radiograph. Additional helpful uses of ultrasound in the acute trauma setting include assessment of volume status and fluid responsiveness (using vena cava diameter and respiratory variation), as well as guidance of procedures such as percutaneous vascular access. Diagnostic peritoneal lavage (DPL) has largely been replaced by the FAST examination and CT scan and is infrequently indicated, although it may be useful in select cases in which there is suspicion for hollow viscus perforation with a compromised physical examination and equivocal CT scan findings. It may also have a role in austere or military settings or in the rare circumstance when equipment or expertise in ultrasonography are unavailable. However, a diagnostic peritoneal aspirate, looking only for the presence of gross intraperitoneal blood and foregoing any attempt at peritoneal lavage, can be very useful in the patient who is hypotensive with a negative FAST examination. A urinalysis should also be obtained on all patients, and evaluation of the complete urinary tract (kidneys, ureters, bladder) should be performed in the presence of significant hematuria.

CT has become the standard of care for the definitive diagnosis of most blunt abdominal injuries and should be used liberally. Missed intra-abdominal injuries, typically resulting from an incomplete diagnostic evaluation, represent one of the most common causes of preventable deaths from trauma. Modern-generation helical CT scanners provide excellent detailed imaging of the abdominal organs, including retroperitoneal structures and major vasculature. It has a sensitivity and specificity approaching 100% for solid organ injuries and provides anatomic detail that is invaluable for injury grading. One often underappreciated exception to this is the limited ability of CT scan to identify and classify pancreatic injuries, particularly when performed immediately following the traumatic event. A recent American Association for the Surgery of Trauma (AAST) multicenter study found a sensitivity of only 60% for modern multidetector CT scan to detect pancreatic injury, highlighting the need for a high index of suspicion and adjunctive diagnostic methods. Serial measurement of amylase or lipase levels every 6 hours over the first 48 to 72 hours is effective for diagnosing major injuries, but these values may be normal in up to 30% of minor pancreatic injuries. If there is a suspicion for a pancreatic injury either by examination or suggested on the initial CT scan, then magnetic resonance imaging can be helpful to better delineate the injury and to assess for involvement of the pancreatic ductal system.

The abdominal CT scan should usually be performed with intravenous (IV) contrast as this greatly enhances the image quality and diagnostic yield. Multiple recent studies and several meta-analyses strongly suggest that the use of modern IV contrast agents does not increase the risk of acute kidney injury and calls into question the whole concept of “contrast nephropathy.” If an abdominal CT scan is felt to be indicated, then contrast should not be withheld in the majority of patients even with some degree of baseline renal insufficiency. The omission of IV contrast significantly decreases the diagnostic utility of CT imaging for blunt abdominal trauma and may lead to missed or delayed diagnoses of clinically significant pathology. In addition to the enhanced image quality, a “contrast blush” on contrast-enhanced CT can provide evidence of active bleeding or arteriovenous fistula that may prompt immediate intervention or a significant change in management or disposition. Although some older series have characterized CT as unreliable to detect hollow viscus perforation or major duodenal/pancreatic injury, more recent experience demonstrates that a high-quality CT scan will correctly identify most of these injuries. However, repeat CT imaging (if no other indication for laparotomy is present) or DPL should be considered in those infrequent situations with a high index of suspicion for missed injury or equivocal findings on the initial CT scan. We perform the abdominal CT scan with IV contrast only, as oral contrast has been shown to add little value in the trauma setting and may create undue delay as well as risk aspiration. Oral contrast may be useful when obtaining a delayed CT scan to evaluate for hollow viscus perforation, or to better delineate known or suspected pancreatic or duodenal injuries.

Anatomic location of injury and American association for the surgery of trauma organ injury scale grading

The abdominal cavity can be divided into two main compartments, the peritoneal cavity (or intraperitoneal) and the retroperitoneum. The majority of injuries following blunt trauma are to the intraperitoneal structures such as the liver, spleen, small bowel, and mesentery and frequently result in clinical signs and symptoms such as pain, tenderness, and distention. The main retroperitoneal structures of concern to the trauma surgeon are the kidneys, duodenum, pancreas, great vessels, and retroperitoneal portions of the ascending and descending colon. Clinical signs and symptoms with retroperitoneal injuries may frequently be absent or significantly delayed, even in the presence of a severe injury. Fortunately, retroperitoneal hollow viscous injuries are relatively rare.

Abdominal injuries identified by CT should be graded according to the American Association for the Surgery of Trauma Organ Injury Scale (AAST-OIS) system ( Table 1 ). This provides a commonly understood language for discussion and study of these injuries and may be used to guide the level and duration of monitoring for nonoperative management. The AAST-OIS for spleen, liver, and kidney was recently updated in 2018 and has previously been validated through analysis of these injuries and associated outcomes in the National Trauma Data Bank. The most notable changes in the 2018 update are the characterization of imaging, operative, and pathologic findings for each injury grade, as well as the incorporation of visceral vascular injuries into the scoring system. Although higher-grade injuries are associated with higher rates of morbidity and failure of nonoperative management, the grade of injury should not be the primary factor in this decision. All grades of injury may be successfully managed nonoperatively in the appropriate clinical setting. Additional factors such as the amount of hemoperitoneum, patient age and comorbid conditions, presence of associated injuries (particularly traumatic brain injuries), and presence of a contrast “blush” should be noted and factored into subsequent management decisions.

Management

Initial management decisions in patients with a known or suspected intra-abdominal injury should be based on the clinical examination and hemodynamic status. Patients admitted with peritonitis or hemodynamic instability that persists despite adequate fluid resuscitation should undergo prompt exploratory laparotomy. Fluid resuscitation in the early evaluation period should be administered judiciously and only if necessary. Overzealous volume resuscitation with elevation of the mean arterial pressure may exacerbate hemorrhage from the injured organ or may cause an iatrogenic drop in the hemoglobin by hemodilution, which may be difficult to differentiate from active bleeding. We prefer small-volume boluses with immediate assessment of the patient’s response by an experienced trauma surgeon. There is some very controversial evidence that supports the positive resuscitation and immunomodulatory benefits of hypertonic crystalloid solutions over standard crystalloid or colloid formulas. Administration of small boluses of hypertonic fluid (100–250 mL of 3% to 7.5% saline) will result in decreased tissue edema with improved gas exchange and a decreased systemic and organ-specific inflammatory response, with an excellent safety profile. In patients with an associated traumatic head injury, hypertonic saline has the added benefit of lowering intracranial pressure while volume resuscitating the patient. However, there has been no proven significant impact of these solutions on mortality or other outcomes. For patients with major injury requiring initiation of large-volume blood products, a balanced resuscitation approximating a 1:1:1 ratio of packed red blood cells to plasma and platelets has been associated with improved survival when compared with resuscitation with crystalloid in both military and civilian populations. Recent series have suggested that the use of whole blood may be superior to a balanced component resuscitation in severely injured patients, but comparative studies are still ongoing. If cold-stored whole blood is available at a particular center, then this should be utilized as the initial resuscitation agent as it will provide an automatic 1:1:1 balanced resuscitation and reversal of any hemorrhage-related coagulopathy. In patients with associated traumatic head injury, the addition of hypertonic saline during resuscitation may be beneficial for lowering intracranial pressure while also volume resuscitating the patient. However, the need for initiation of large-volume resuscitation or blood transfusion should almost always prompt abandoning attempts at nonoperative management and proceeding to the operating room. In extremely rare cases, and under the close supervision of experienced trauma surgeons, the use of resuscitative endovascular balloon occlusion of the aorta, in conjunction with angioembolization, may be used to nonoperatively stabilize and manage hemodynamically unstable patients with solid organ injuries.

The primary components of safely managing these injuries are appropriate monitoring and frequent reassessments of the patient’s clinical examination and laboratory values. The level of inpatient care (intensive care unit vs. step-down unit versus ward), and the frequency of monitoring should be dictated by the patient’s clinical status, associated injuries, and the severity of the organ injury. All personnel caring for the patient should be made aware of the presence and type of abdominal injury, and a clear plan for monitoring and alerting the trauma team to any changes should be in place. The majority of injuries that fail nonoperative management will declare themselves within 48 hours of injury, and this should be the period of most intensive monitoring.

Additional important factors to be considered in guiding management are the age of the patient and the presence of comorbid conditions and associated injuries. Traditionally, nonoperative management of abdominal solid organ injuries was contraindicated in elderly patients and those with multiple associated injuries, particularly severe traumatic brain injuries. However, with improvements in imaging technology and monitoring capabilities, many centers are reporting favorable results of nonoperative management in these more difficult patient populations. Success rates for nonoperative management of over 90% have been reported among patients with multiple associated injuries, with similar complication rates to those with isolated injuries. This should only be attempted at centers with experience and expertise in managing complex, multisystem trauma and requires coordination and cooperation between the involved surgical services, such as neurosurgery and orthopedics. Although most comorbid conditions do not significantly impact the success of nonoperative management, the presence of cirrhosis should raise a red flag of caution following blunt abdominal trauma. Hemostasis of blunt hepatic or splenic lacerations in the presence of advanced cirrhosis may be significantly compromised by portosplenic hypertension, increased organ rigidity, hepatosplenomegaly, and systemic coagulopathy. In addition, these patients often have extremely poor physiologic reserve and have a high rate of mortality and morbidity with both nonoperative and operative management strategies. The failure of nonoperative management of splenic injuries among patients with advanced cirrhosis has been reported as high as 92%, with an associated increase in mortality rate vs. immediate laparotomy.

Spleen and liver

Patients with any identified injury of the spleen or liver should be admitted to the hospital for a minimum of 24 to 48 hours of observation. We recommend intensive care unit or intermediate-level (step-down) admission for all high-grade injuries (grades III through V, Figs. 1 and 2 ). The primary purpose of observation is to identify the presence of any associated abdominal injuries and to monitor for ongoing or recurrent bleeding from the liver or spleen. Patients with higher grade (grade IV or V) hepatic or splenic injuries, in particular, have much higher rates for failure of observation alone and are more likely to require operative intervention or angioembolization. The overall incidence of missed injuries in these patients appears to be low (around 2%) and should not influence the decision for nonoperative management. Serial physical examinations should focus on the patient’s hemodynamic status and any evidence of worsening abdominal tenderness, distention, or the development of peritonitis. Serial laboratory evaluations should include a complete blood count at the minimum. Some measure of global tissue perfusion and acidosis, such as the lactate or base deficit, may be useful in making management and treatment decisions in these patients. The timing and appropriateness of blood transfusion in these patients remain an area of controversy. Although the need for transfusion was previously used as a guideline for operative intervention, this is no longer the case. For splenic injuries, we favor a low threshold for operative or angiographic intervention if the patient requires more than 1 to 2 units of transfused blood. We accept a higher threshold for surgical intervention on liver injuries that require transfusion, typically after 4 to 6 units of transfused blood. Ideally it would be preferred if one could avoid transfusion and surgery, but the exact timing of transfusion, surgery, or both for patients with solid organ injuries remains more an art than science. It requires the expert judgment of an experienced trauma surgeon to avoid the error of delaying a needed laparotomy until the patient is on the verge of hemodynamic collapse.

Bed rest and activity restriction have traditionally been recommended for these injuries, but there are no clinical data or science to support this practice. Prolonged immobility should be avoided, and patients should be allowed to mobilize as early as possible. Although routine repeat imaging of all injuries with ultrasound or CT does not appear to be beneficial or cost-effective, we recommend reimaging in select patients such as those with high-grade injuries or those with above average activity levels, such as athletes, firefighters, military service members, and police officers. The repeat imaging can help to delineate healing of the injury and guide return to full activity levels, as well as identify late complications such as hematomas/bilomas, necrosis, and pseudoaneurysms. Age-appropriate immunizations against encapsulated organisms should be given to all patients with grade IV or V splenic injuries as they may be functionally asplenic. Immunizations should be administered prior to patient discharge to ensure compliance.