Selective Nonoperative Management of Abdominal Injury

S. Peter Stawicki

John P. Pryor

Over the last two decades, one of the most important developments in the care of the trauma patient has been the use of selective nonoperative management (SNOM) of significant injuries. Pediatric surgeons were the first to pioneer nonoperative care for splenic and hepatic injuries in children. The success in children prompted the adaptation of selective management in adult patients. Subsequent experience and research has demonstrated that nonoperative techniques are applicable to most types of blunt injuries, and a small proportion of penetrating type injuries. Although many agree that the techniques have greatly reduced morbidity from unnecessary surgeries, new complications and management challenges have arisen that must be understood if these techniques are to be applied correctly.¹

Often nonoperative is equated with “conservative” management. However, the reality is that nonoperative management is often more resource demanding, with more complex decision making, than performing a surgery. The nonoperative approach relies heavily on the availability of trauma-trained surgeons, modern radiographic imaging such as computed tomography (CT), accurate interpretation of such high-quality radiographic images, as well as the presence of appropriate supporting infrastructure and ancillary services such as intensive care units and interventional radiology.¹

Although initially met with some resistance, SNOM of penetrating injury is being practiced in a very limited way in some experienced centers.²,³ Although clearly applicable in stab wound injury, the role of SNOM in gunshot wounding is still controversial and must be applied with the utmost care by experienced surgeons.

HISTORICAL BACKGROUND

Nonoperative management of penetrating abdominal trauma remained the standard of care throughout most of the 19th century due to the lack of anesthesia, antibiotics, and therefore poor outcomes after attempted surgical procedures. This paradigm began to change after 1887, when the American Surgical Association recommended exploration of civilian penetrating abdominal wounds.² Owing to high mortality associated with nonoperative management of penetrating abdominal injuries, a policy of routine surgical exploration evolved during World War I, and became policy during World War II.² Mandatory exploration became surgical dictum when surgeons coming back from World War II advocated mandatory surgical exploration for all gunshot wounds in the civilian setting.³ This approach appeared to be associated with significant reductions in mortality and remained the standard of care until the 1960s and 1970s when a trend toward SNOM of stab wounds to the abdomen began to emerge.² Mandatory exploration of all gunshot wounds to the abdomen remained
the standard practice until the 1990s, when trauma centers in the United States and South Africa published their experiences on SNOM of those injuries.²

The concept of nonoperative management of blunt abdominal injuries has been evolving slowly throughout this century. However, SNOM did not become popularized until the 1970s, when better imaging techniques, improvement in nonoperative interventional techniques, refinement of intensive care of the critically ill, and increasing number of clinical studies that proved this approach is safe and effective.⁴,⁵ During the 1980s and 1990s, the nonoperative management of blunt abdominal trauma (BAT) underwent a maturation process through multiple clinical investigations, trial and error, and detailed scrutiny. As a result, the modern application of SNOM is based largely on the evidence-based approach, with known benefits, risks, indications, and contraindications (see Table 1).⁴

TABLE 1 SUCCESS OF NONOPERATIVE MANAGEMENT STRATEGY FOR DIFFERENT ORGAN/ANATOMIC LOCATION INJURIES (IN ALPHABETICAL ORDER) AND MECHANISM OF INJURY (BLUNT VS. PENETRATING)

Blunt^a
Adrenal	80%^b
Aortic (selective initial medical management)	>80%
Bowel (hematoma)	10%-15%
Duodenum (hematoma)	50%-60%
Esophagus	Rarely
Kidney	85%^c
Liver	80%
Pancreas	Low-grade injuries only
Spleen	80%-90%
Urinary bladder (extraperitoneal)	>75%
Penetrating^d
Overall	20%
Back and flank	60%
Extremity	Variable, case by case
Kidney	15%
Liver	20%-30%
Neck (selective nonoperative management)	Variable, case by case
Pelvic/gluteal area	65%-70%
Spleen	<5%
^aSchwab CW. Selection of nonoperative management candidates. World J Surg. 2001;25:1389-1392. ^bStawicki SP, Hoey BA, Grossman MD, et al. Adrenal gland trauma is associated with high injury severity and mortality. Curr Surg. 2003;60:431-436. ^cHagiwara A, Sakaki S, Goto H, et al. The role of interventional radiology in the management of blunt renal injury: A practical protocol. J Trauma. 2001;51:526-531. ^dPryor JP, Reilly PM, Dabrowski GP, et al. Nonoperative management of abdominal gunshot wounds. Ann Emerg Med. 2004;43:344-353.

TABLE 2 UNNECESSARY LAPAROTOMY—STATISTICS

Percentage of unnecessary laparotomies	1.7%-38% (average 20%)
Associated early complications	8.6%-25.6%
Associated late complications	2.4%-5%
Hospital cost difference	$10,000 more for operative management

THE BENEFIT OF SELECTIVE NONOPERATIVE MANAGEMENT—PREVENTING UNNECESSARY SURGERY

One clear benefit to SNOM is avoiding an unnecessary surgery. A laparotomy is classified as unnecessary, if the exploration reveals either no pathology (negative) or a minor injury that requires no surgical treatment (nontherapeutic).² The reported incidence of unnecessary laparotomies for trauma varies from 1.7% to 38%, and depends on the experience and practice patterns of the individual trauma center.² The rates were highest during the era of mandatory exploration for penetrating injuries, as well as solid organ injuries found on CT scan.⁶,⁷,⁸,⁹,¹⁰,¹¹,¹²,¹³ In contrast, centers that practiced a policy of SNOM demonstrated a significantly lower rate of unnecessary laparotomies (3.2% to 10%).¹⁴,¹⁵,¹⁶,¹⁷,¹⁸,¹⁹

Exploratory laparotomy is associated with significant morbidity and costs (see Table 2). The reported incidence of laparotomy or anesthesia-related early complications varies between 8.6% and 25.6%.¹²,²⁰ For late complications, such as bowel obstruction or incisional hernia, the reported overall incidence is between 2.4% and 5%.¹³,²¹ In addition, the cost and hospital stay for patients undergoing laparotomy are significantly higher than for patients successfully managed nonoperatively.¹⁵,¹⁷ In one study, the mean hospital charges for patients with abdominal gunshot wounds successfully managed nonoperatively were nearly $ 10,000 less than those for patients who underwent unnecessary surgeries.¹⁵ In fact, a policy of SNOM for abdominal gunshot wounds has been shown to save both significant amount of hospital days and hospital-related charges.¹⁷

THE RISKS OF SELECTIVE NONOPERATIVE MANAGEMENT—DELAYS, MISSED INJURY, AND COMPLICATIONS

The benefits of SNOM should be weighed against the consequences of missed injuries and delayed diagnosis. Several reports have shown that when SNOM schemes are used by experienced practitioners, the risks of delay in diagnosis are minimal.²,²² The delay beyond which the
morbidity increases is not precisely known, but some have suggested a time frame of 6 to 12 hours.² The injured organ, the length of delay, and the degree of peritoneal contamination are all likely to play a role in the incidence and severity of complications related to delay in diagnosis and/or treatment.

TABLE 3 TYPES OF MINIMALLY INVASIVE PROCEDURAL INTERVENTIONS IN TRAUMA PATIENT MANAGEMENT

▪	Arteriographic interventions, including embolization, stenting, and diagnostic applications.
▪	Percutaneous drainage of bile collections, urinomas, abscesses, and hematomas.
▪	Endoscopic stenting of bile duct injuries and selected urinary tract injuries.
▪	Minimally-invasive endoscopic-guided placement of various adjunctive devices, including percutaneous gastrostomy (PEG) or percutaneous jejunostomy (PEJ) enteric access tubes, and percutaneous endoscopically-guided tracheostomy placement.

Missed injury is a significant risk when undertaking SNOM. Approximately 5% of patients with BAT managed in a nonoperative manner will have a delay in diagnosis or a missed injury.²³ Although the resultant mortality is very low, a variety of complications (abscesses, pseudocysts, bile collections, etc) necessitate a careful follow-up that extends beyond the initial postinjury period.²³

In many ways, the use of SNOM has traded one set of potential morbidities for another. For example, nonoperative management of liver injuries has clearly shown to decrease overall blood loss and transfusion requirements.²⁴ However, a number of significant injuries will develop bile leaks from the fractured surface of the organ. Therefore, the decreased blood loss and complications of the laparotomy are exchanged for the risk of undrained biloma. Fortunately, as with this example, many of the complications of SNOM are amendable to minimally invasive techniques such as percutaneous drainage and ductal decompression with endoscopic retrograde cholangiopancreatography (ERCP) (see Table 3). Surgeons who endeavor to practice SNOM must be aware of, and ready to handle these specific types of complications (see Table 4).

TABLE 4 SYNOPSIS OF COMPLICATIONS ASSOCIATED WITH NONOPERATIVE MANAGEMENT OF INJURIES

▪	Missed injuries
▪	Delay in diagnosis
▪	Delay in treatment
▪	Inadequate/delayed resuscitation
▪	Retained hematomas (at risk for infection/abscess)
▪	Iatrogenic injuries secondary to minimally invasive treatment
▪	Abdominal sepsis and/or abscess
▪	Biliary/pancreatic leaks secondary to biliary/pancreatic injury
▪	Urinary collection secondary to renal/ureteral/bladder injury
▪	Delayed rupture of pseudoaneurysms
▪	Delayed aortic rupture
▪	Complications of delayed treatment of vascular injury (thrombosis, compartment syndrome, potential limb loss)

PRINCIPLES OF SELECTIVE NONOPERATIVE MANAGEMENT

Nonoperative management of various injuries has certain unifying principles that must be followed for clinical success (see Table 5). The first tenet is that there must be hemodynamic stability to pursue SNOM. Unstable patients are assumed to have ongoing bleeding and need surgical control of the hemorrhage. Potential exceptions to this are those with lesions not amenable to surgery such as pelvic fractures; however, the number of patients like this is small compared to the many that have unyielding hemorrhage. A corollary to the first rule is that generalized peritonitis denotes the possibility of bowel perforation, and remains an indication for surgical exploration.

The next step is to define and characterize the injury that will be treated nonoperatively. This was a challenge years ago where nonspecific tests for injury such as diagnostic peritoneal lavage (DPL) were utilized. However, with excellent cross-sectional imaging capabilities, injuries can be anatomically identified and graded. The success of SNOM for many organs is directly related to the grade of injury. In addition, imaging is used to look for variants that are associated with SNOM failure and make an educated
decision about whether or not a particular patient will be successfully managed. Such a situation exists with the finding of a “blush” on CT scan indicating active extravasation, and the need for hemorrhage control.²⁵

TABLE 5 SYNOPSIS OF GENERAL PRINCIPLES OF NONOPERATIVE MANAGEMENT

▪	Clearly define the injury
▪	Assure the patient is examinable, with clear mental status
▪	Assure patient is hemodynamically stable, with no obvious surgical indications
▪	Use caution when committing to nonoperative management in multiply injured patients
▪	Insist that adequate health care team resources are available to perform frequent physical examinations and re-imaging as needed
▪	Provide an appropriate setting for nonoperative observation such as observation ward, intensive care unit, or monitored emergency department bed
▪	Be prepared to provide surgical management promptly if indicated by a change in the signs or symptoms

The third tenet is to make a decision about nonoperative management in the context of the whole patient, taking into account all of the injuries, not each one in isolation. This is particularly important in multiply injured patient where there are competing priorities. For example, an otherwise nonoperative grade III spleen injury may be treated differently if there is an associated severe pelvic fracture. In this case, it may be difficult to determine if a subsequent drop in hemoglobin is due to ongoing pelvic bleeding or hemorrhage from the spleen. Alternatively, a lower grade spleen injury may need removal if there is a concomitant severe head injury where hypotension would lead to undue neurologic morbidity. Many such conflicts can occur from the permutations of multiple injuries in different organ systems.

The keystone of SNOM is the obligation of the surgical team to admit the patient into a highly monitored setting where continuous reassessment of the clinical situation is assured. This not only requires a monitored bed but also trained nursing staff who understand the signs and symptoms of failing SNOM, and the availability of surgeons to reexamine the patient on a regular basis. It is easy to see how resources to meet these requirements can quickly disappear in a busy trauma unit where residents and surgeons are caring for multiple patients, or in smaller hospitals where physicians may not be available in-house at night. SNOM is a safe technique only if clinical deterioration is noted promptly and laparotomy is immediately available if needed.

Lastly, factors beyond the host and the injury need to be considered. It may not be plausible to treat a homeless patient with a known substance addiction with SNOM, knowing that after a few days in the hospital he/she would be released into an environment with a high risk of reinjury. Therefore, in addition to the whole patient, the entire social situation must be considered in the decisions for SNOM.

THE DECISION TO PURSUE SELECTIVE NONOPERATIVE MANAGEMENT

Hemodynamics

As stated previously, the first requirement of attempting SNOM is hemodynamic stability. Although there is some debate about what the optimal blood pressure is for a patient with acute injury, most believe that a systolic blood pressure below 90 mm Hg is concerning. The constellation of hypotension, tachycardia, and systemic signs such as pallor, diaphoresis, and mental status changes are late indications that the patient is in shock. These patients are characterized as hemodynamically unstable. It is clear from clinical experience and a vast body of literature that patients who are hemodynamically unstable have ongoing bleeding that needs control and are not candidates for nonoperative management.

There is a subset of patients who may present as initially hemodynamically unstable but rapidly improve with minimal resuscitation to a metastable state. Patients who are metastable have likely lost a volume of blood, but have perhaps stopped actively bleeding. Using nonoperative techniques in metastable patients is done with great caution, and with the understanding that if there is renewed bleeding, chances are that the patient will become unstable again.

During the course of nonoperative management, vital signs are recorded frequently. An increased temperature or respiratory rate can indicate a hollow visceral perforation or early abscess formation. Pulse and blood pressure can also change with sepsis or intra-abdominal bleeding. Adjunctive laboratory testing, such as serial determination of white blood cell count, hemoglobin and hematocrit levels, and serum lactic acid level and base deficit can also help to determine whether failure of nonoperative management is occurring.

Physical Examination

Physical examination remains the cornerstone of the triage of traumatic injuries. Peritonitis or hemodynamic instability with other signs of abdominal injury constitutes a strong indication for emergency laparotomy. Unfortunately, the physical findings of significant abdominal injury can be subtle and the diagnosis of intra-abdominal injury uncertain. Moreover, between 20% and 40% of patients with significant hemoperitoneum have a benign abdominal examination upon initial assessment.⁹

The physical examination can have significant limitations in certain situations, including the older trauma patient population, where the effect of medications such as β-blockers may effectively mask manifestations of shock.² Similarly, young patients, especially with short prehospital transport times, may not exhibit signs or symptoms of shock despite the presence of significant internal bleeding.² Patients with associated severe head or spinal injury may not have reliable examinations. Clouded sensorium due to alcohol or other substance use may further confound the accuracy of clinical assessment. Combative and intoxicated patients pose further diagnostic dilemma, not only due to the lack or reliable physical examination findings but also due to the potential danger to the health care personnel and inability to cooperate during imaging studies, which require the patient to remain still.

Especially challenging is the evaluation of a hemodynamically unstable patient with multiple injuries and competing priorities, such as a concurrent head injury, aortic injury, pelvic fractures, or extremity trauma. In patients
for whom clinical examination is not reliable, special investigations can be crucial in early and accurate triage. Lack of reliable physical examination may constitute a relative contraindication to nonoperative management of traumatic injuries in patients who fall into this “gray” zone.

Imaging

The condition of the patient and the specific pattern of injury that is suspected will determine the imaging modality most appropriate in the given traumatic injury. An armamentarium of radiographic studies exists including plain radiograms, CT, magnetic resonance imaging, as well as a wide spectrum of ultrasonographic techniques. Imaging has become the integral part of the early decision-making process as to whether to proceed to the operating theater, interventional radiology suite, the trauma-surgical floor or the intensive care unit (see Table 6).

The location of the imaging equipment and the clinical capabilities of the supporting health care team staff and facilities should always be considered. Regardless of the type of radiographic investigation, the trauma patient who is undergoing the study should always be monitored appropriately. In addition, the trauma team should carefully plan the sequence of the resuscitation in order to minimize the loss of time and avoid radiographs that are technically impaired or have low diagnostic yield.

Plain Radiography

The overall value of plain films in the evaluation of patients with BAT is limited. The chest roentgenogram is paramount in diagnosing pneumothorax, hemothorax, and widened mediastinum in the trauma bay. Rarely, the chest radiograph may also aid in the diagnosis of abdominal injuries such as ruptured hemidiaphragm or pneumoperitoneum.

TABLE 6 IMPACT OF RADIOGRAPHIC IMAGING ON NONOPERATIVE TRAUMA MANAGEMENT

▪	Plain films remain valuable in chest, pelvic, and penetrating injury
▪	High-resolution computed tomography allows accurate organ injury assessment
▪	Ultrasonography is a fast and reliable adjunct in determining intra-abdominal hemorrhage
▪	Every trauma practitioner should be intimately familiar with FAST ultrasonography
▪	Interventional radiology techniques are used to treat acute hemorrhage as well as late complications of solid organ injury
▪	Advanced image reconstructions may be helpful when dealing with suspected aortic, spinal, and diaphragmatic injuries
FAST, focused assessment with sonography for trauma.

The pelvic or chest radiograph can demonstrate fractures of the thoracolumbar spine. The presence of transverse fractures of the vertebral bodies (i.e., Chance fractures) suggests a higher likelihood of blunt injuries to the bowel. In addition, free intraperitoneal air, or trapped retroperitoneal air from duodenal or colonic injury, may be seen. Penetrating wounds should have a radiopaque marker placed over each penetration site and radiographs obtained to determine trajectory and retained bullets or fragments.

Ultrasonography

The American College of Surgeons included the use of ultrasonography in the Advanced Trauma Life Support secondary survey since 1999. The focused assessment with sonography in trauma (FAST) examination can be repeated as many times as needed, providing an excellent adjunct to serial physical examinations in the nonoperative management of traumatic injury.²⁶ In many centers, the FAST examination has virtually replaced DPL as the procedure of choice in the evaluation of hemodynamically unstable trauma patients.

The FAST examination is based on the assumption that all immediately life-threatening abdominal injuries are associated with hemoperitoneum or pericardial tamponade. The current examination protocol consists of the pericardial, perihepatic, perisplenic, and pelvic acoustic windows performed with the patient supine. The detection of free intraperitoneal fluid is based on factors such as the body habitus, injury location, presence of clotted blood, position of the patient, and amount of free fluid present. The FAST examination is interpreted as positive if fluid is found in any of the four acoustic windows and is interpreted as negative if no fluid is seen. An examination is deemed indeterminate if any of the windows cannot be adequately assessed. Studies show that as little as 30 to 70 mL of blood can be detected with ultrasonography, with a small anechoic stripe in the perihepatic space representing approximately 250 mL of fluid, and 0.5- to 1-cm stripes represent approximately 500 mL to 1 L of free fluid, respectively.²⁷,²⁸

Only gold members can continue reading. Log In or Register to continue