Practice Management Guidelines for the Evaluation of Blunt Abdominal Trauma: The EAST Practice Management Guidelines Work Group

I. STATEMENT OF THE PROBLEM Evaluation of patients who have sustained blunt abdominal trauma (BAT) may pose a significant diagnostic challenge to the most seasoned trauma surgeon. Blunt trauma produces a spectrum of injury from minor, single-system injury to devastating, multisystem trauma. Trauma surgeons must have the ability to detect the presence of intra-abdominal injuries across this entire spectrum. Although a carefully performed physical examination remains the most important method to determine the need for exploratory laparotomy, there is little Level I evidence to support this tenet. In fact, several studies have highlighted the inaccuracies of the physical examination in BAT. 1,2 The effect of altered level of consciousness as a result of neurologic injury, alcohol, or drugs is another major confounding factor in assessing BAT. Because of the recognized inadequacies of physical examination, trauma surgeons have come to rely on a number of diagnostic adjuncts. Commonly used modalities include diagnostic peritoneal lavage (DPL) and computed tomographic (CT) scanning. Although not available universally, focused abdominal sonography for trauma (FAST) has recently been included in the diagnostic armamentarium. Diagnostic algorithms outlining appropriate use of each of these modalities individually have been established. Several factors influence the selection of diagnostic testing: type of hospital (i.e., trauma center vs. "nontrauma" hospital); access to a particular technology at the surgeon's institution; and the surgeon's individual experience with a given diagnostic modality. As facilities evolve, technologies mature, and surgeons gain new experience, it is important that any diagnostic strategy constructed be dynamic. The primary purpose of this study was to develop an evidence-based, systematic diagnostic approach to BAT using the three major diagnostic modalities: DPL, CT scanning, and FAST. This diagnostic regimen would be designed such that it could be reasonably applied by all general surgeons performing an initial evaluation of BAT. II. PROCESS A. Identification of References A MEDLINE search was performed using the key words "abdominal injuries" and the subheading "diagnosis." This search was limited further to (1) clinical research, (2) published in English, and (3) publication dates January 1978 through February 1998. The initial search yielded 742 citations. Case reviews, review articles, meta-analyses, editorials, letters to the editor, technologic reports, pediatric series, and studies involving a significant number of penetrating abdominal injuries were excluded before formal review. Additional references, selected by the individual subcommittee members, were then included to compile the master reference list of 197 citations. B. Quality of the References Articles were distributed among subcommittee members for formal review. A review data sheet was completed for each article reviewed that summarized the main conclusions of the study and identified any deficiencies in the study. Furthermore, reviewers classified each reference by the methodology established by the Agency for Health Care Policy and Research of the U.S. Department of Health and Human Services as follows: Class I: Prospective, randomized, double-blinded study Class II: Prospective, randomized, nonblinded trial Class III: Retrospective series, meta-analysis After review by the subcommittee, references were excluded on the basis of poor design or invalid conclusions. An evidentiary table (Table 1) was constructed using the remaining 101 references: Class I, 20 references; Class II, 32 references; and Class III, 49 references. Recommendations were made on the basis of studies included in the evidentiary table (Table 1).Table 1: Euidentiary: Practice Management Guidelines for the Evaluation of Blunt Abdominal TraumaTable 1: ContinuedTable 1: ContinuedTable 1: ContinuedTable 1: ContinuedTable 1: ContinuedTable 1: ContinuedTable 1: ContinuedIII. RECOMMENDATIONS A. Level I 1. Exploratory laparotomy is indicated for patients with a positive DPL. 2. FAST may be considered as the initial diagnostic modality to exclude hemoperitoneum. B. Level II 1. When DPL is used, clinical decisions should be made on the basis of the presence of gross blood on initial aspiration (i.e., 10 mL) or microscopic analysis of lavage effluent. 2. Exploratory laparotomy is indicated in hemodynamically unstable patients with a positive FAST. In hemodynamically stable patients with a positive FAST, follow-up CT scan permits nonoperative management of select injuries. 3. Surveillance studies (i.e., DPL, CT scan, repeat FAST) should be considered in hemodynamically stable patients with indeterminate FAST results. 4. CT scanning is recommended for the evaluation of hemodynamically stable patients with equivocal findings on physical examination, associated neurologic injury, or multiple extra-abdominal injuries. Under these circumstances, patients with a negative CT scan should be admitted for observation. 5. CT scanning is the diagnostic modality of choice for nonoperative management of solid visceral injuries. 6. In hemodynamically stable patients, DPL and CT scanning are complementary diagnostic modalities. C. Level III 1. Objective diagnostic testing (i.e., FAST, DPL, CT scanning) is indicated for patients with abnormal mentation, equivocal findings on physical examination, multiple injuries, concomitant chest injury, or hematuria. 2. Patients with seat belt sign should be admitted for observation and serial physical examination. The presence of intraperitoneal fluid on FAST or CT scan in a patient with seat belt sign suggests the presence of an intra-abdominal injury that may require surgery. 3. CT scanning is indicated for the evaluation of suspected renal injuries. 4. In the patient at high risk for intra-abdominal injury (e.g., multiple orthopedic injuries, severe chest wall trauma, neurologic impairment), a follow-up CT scan should be considered after a negative FAST. 5. In hemodynamically stable patients with a positive DPL, follow-up CT scan should be considered, especially in the presence of pelvic fracture or suspected injuries to the genitourinary tract, diaphragm, or pancreas. IV. SCIENTIFIC FOUNDATION A. Diagnostic Peritoneal Lavage DPL was introduced by Root et al. in 1965 as a rapid and accurate method to identify the presence of intra-abdominal hemorrhage after trauma. 3 Subsequent studies have confirmed the efficacy of DPL in diagnosing abdominal hemorrhage as well as its superiority over physical examination alone. 4 The accuracy of DPL has been reported to be between 92% and 98%. 5–10 The high sensitivity of DPL is because of the significant false-positive rate of the technique. 11–13 Several authors have highlighted the importance of interpreting DPL results in the context of the overall clinical condition of the patient. A positive DPL does not necessarily mandate immediate laparotomy in the hemodynamically stable patient. 12,14–16 DPL has been shown to be more efficient than CT scanning in identifying patients that require surgical exploration. 17 The complication rate associated with DPL is quite low. 18 The incidence of complications is lower for open DPL compared with the closed technique. However, closed DPL can be performed more rapidly. 19–22 Studies designed to examine the ability of physicians to estimate the red blood cell (RBC) count in DPL fluid have demonstrated the poor sensitivity of visual inspection. 23–25 A positive DPL, on the basis of microscopic analysis of lavage fluid, has been defined as > 105 RBCs/mm3. It has been recommended that patients with RBC counts in the equivocal range (i.e., 25,000–75,000 RBCs/mm3) undergo additional diagnostic testing, such as CT scanning. 12 The false-positive rate for DPL is increased in patients with pelvic fractures. 26,27 To avoid sampling the retroperitoneal hematoma, a supraumbilical approach has been recommended, theoretically reducing the chances of a false-positive result. 28 The advantages of DPL for detection of hollow visceral injuries have been clearly demonstrated. 29,30 Two studies that advocate analysis of DPL fluid for amylase and alkaline phosphatase consistent with enteric injuries have been disputed. 31–33 Similarly, the utility of the DPL white blood cell count has been questioned. 34–36 DPL is sensitive for mesenteric injury and, in fact, has been shown to be superior to CT scanning for the diagnosis of this injury. 37 Thus, DPL is a safe, rapid, and accurate method for determining the presence of intraperitoneal blood in victims of BAT. It is more accurate than CT scanning for the early diagnosis of hollow visceral and mesenteric injuries, but it does not reliably exclude significant injuries to retroperitoneal structures. False-positive results may occur in the presence of pelvis fractures. Hemodynamically stable patients with equivocal results are best managed by additional diagnostic testing to avoid unnecessary laparotomies. B. Computed Tomographic Scanning Routine use of CT scanning for the evaluation of BAT was not initially viewed with overwhelming enthusiasm. CT scanning requires a cooperative, hemodynamically stable patient. In addition, the patient must be transported out of the trauma resuscitation area to the radiographic suite. Specialized technicians and the availability of a radiologist for interpretation were also viewed as factors that limited the utility of CT scanning for trauma patients. CT scanners are now available in most trauma centers and, with the advent of helical scanners, scan time has been significantly reduced. As a result, CT scanning has become an accepted part of the traumatologist's armamentarium. The accuracy of CT scanning in hemodynamically stable blunt trauma patients has been well established. Sensitivity between 92% and 97.6% and specificity as high as 98.7% have been reported in patients subjected to emergency CT scanning. 38,39 Most authors recommend admission and observation after a negative CT scan. 40,41 In a recent study of 2,774 patients, the authors concluded that the negative predictive value (99.63%) of CT scanning was sufficiently high to permit safe discharge of BAT patients after a negative CT scan. 42 CT scanning is notoriously inadequate for the diagnosis of mesenteric injuries and may also miss hollow visceral injuries. In patients at risk for mesenteric or hollow visceral injury, DPL is generally felt to be a more appropriate test. 37,43 A negative CT scan in such a patient cannot reliably exclude intra-abdominal injuries. CT scanning has the unique ability to detect clinically unsuspected injuries. In a series of 444 patients in whom CT scanning was performed to evaluate renal injuries, 525 concomitant abdominal and/or retroperitoneal injuries were diagnosed. Another advantage of CT scanning over other diagnostic modalities is its ability to evaluate the retroperitoneal structures. 40 Kane et al. performed CT scanning in 44 hemodynamically stable blunt trauma patients after DPL. In 16 patients, CT scan revealed significant intra-abdominal or retroperitoneal injuries not diagnosed by DPL. Moreover, the findings on CT scan resulted in a modification to the original treatment plan in 58% of the patients. 44 C. Focused Abdominal Sonography for Trauma In recent years, FAST has emerged as a useful diagnostic test in the evaluation of BAT. The advantages of the FAST examination have been clearly established. FAST is noninvasive, may be easily performed, and can be performed concurrently with resuscitation. In addition, the technology is portable and may be easily repeated if necessary. 45–48 In most cases, FAST may be completed within 3 or 4 minutes. 49–51 The test is especially useful for detecting intra-abdominal hemorrhage in the patient with multiple injuries or the pregnant patient. 52 A noted drawback to the FAST examination is the fact that a positive examination relies on the presence of free intraperitoneal fluid. In the hands of most operators, ultrasound will detect a minimum of 200 mL of fluid. 53 Injuries not associated with hemoperitoneum may not be detected by this modality. 49,54,55 Thus, ultrasound is not a reliable method for excluding hollow visceral injury. 47,49,56–58 In addition, the FAST examination cannot be used to reliably grade solid organ injuries. Therefore, in the hemodynamically stable patient, a follow-up CT scan should be obtained if nonoperative management is contemplated. 59 FAST compares favorably with more traditionally used diagnostic tests. In the hemodynamically stable patient with BAT, FAST offers a viable alternative to DPL. 60 DPL may also be used as a complementary examination in the hemodynamically stable patient in the presence of equivocal or negative ultrasound findings with strong clinical suspicion of visceral injury. 61,62 FAST has demonstrated utility in hemodynamically stable patients with BAT. 58,60,63 In addition, ultrasound has been shown to be more cost-effective when compared with DPL or CT scanning. 45,47,60 Overall, FAST has a sensitivity between 73% and 88% and a specificity between 98% and 100%, and is 96% to 98% accurate. 46,50,57,58,64,65 This level of accuracy is independent of the practitioner performing the study. Surgeons, emergency medicine physicians, ultrasound technicians, and radiologists have equivalent results. 46,53,64–66 D. Other Diagnostic Modalities As interest in laparoscopic procedures has increased among general surgeons, there has been speculation regarding the role of diagnostic laparoscopy (DL) in the evaluation of BAT. One of the potential benefits postulated is the reduction of nontherapeutic laparotomies. With modification of the technique to include smaller instruments, portable equipment, and local anesthesia, DL may be a useful tool in the initial evaluation of BAT. Although there are no randomized, controlled studies comparing DL to more commonly used modalities, experience at one institution using minilaparoscopy demonstrated a 25% incidence of positive findings on DL, which were successfully managed nonoperatively and would have resulted in nontherapeutic laparotomies. 67 Although its ultimate role remains unclear, another modality to be considered in the diagnostic evaluation of BAT is visceral angiography. This modality may have diagnostic value when used in conjunction with angiography of the pelvis or chest, or when other diagnostic studies are inconclusive. 68 V. SUMMARY Injury to intra-abdominal viscera must be excluded in all victims of BAT. Physical examination remains the initial step in diagnosis but has limited utility under select circumstances. Thus, various diagnostic modalities have evolved to assist the trauma surgeon in the identification of abdominal injuries. The specific tests are selected on the basis of the clinical stability of the patient, the ability to obtain a reliable physical examination, and the provider's access to a particular modality. It is important to emphasize that many of the diagnostic tests used are complementary rather than exclusionary. On the basis of the above recommendations, a reasonable diagnostic approach to BAT is summarized in Figures 1 and 2. In hemodynamically stable patients with a reliable physical examination, clinical findings may be used to select patients who may be safely observed. In the absence of a reliable physical examination, the main diagnostic choice is between CT scanning or FAST (with CT scanning in a complementary role). Hemodynamically unstable patients may be initially evaluated with FAST or DPL.Fig. 1: Evaluation of BAT: unstable patient.Fig. 2: Evaluation of BAT: stable patient.VI. FUTURE INVESTIGATIONS Recent literature is replete with studies that emphasize the many advantages of ultrasound in the valuation of BAT. Although this technology is becoming more available to trauma surgeons, for a variety of reasons, it has not become universally available in all centers. Continued research addressing the utility of FAST, with emphasis on its advantages specific to resource use, is suggested. In addition, studies should be designed to more closely evaluate the feasibility of FAST as the sole diagnostic test in hemodynamically stable patients. Perhaps safe strategies for nonoperative management of solid visceral injuries could be developed that rely on FAST alone, such that the number of CT scans could be reduced.