ACR Appropriateness Criteria®
Clinical Condition: Blunt Abdominal Trauma, Adults
Variant 1: Stable patient.
| Radiologic Exam Procedure |
Appropriateness Rating |
Comments |
| X-ray, chest, upright |
8 |
Computed tomography (CT) and x-rays may be appropriate. See original guideline document text for details. |
| X-ray, abdomen, supine and upright |
8 |
CT and x-rays may be appropriate. See original guideline document text for details. |
| CT, abdomen & pelvis |
8 |
Multidetector CT (MDCT) is preferable. CT and x-rays may be appropriate. See original guideline document text for details |
| INV, angiography embolization |
8 |
Not a screening procedure. Angiography is indicated to delineate and treat active bleeding or other lesions amenable to angiographic therapy, but only when this type of lesion is first detected or suspected, either by CT or by some other means. |
| US, screen for hemoperitoneum |
4 |
Low sensitivity of ultrasound to injuries that require surgery (active hemorrhage, viscus perforation) and its inability to exclude injuries that require in-hospital observation lessen its usefulness for key triage decisions. |
| US, organ sonogram |
3 |
|
| MRI, organ evaluation |
2 |
|
| MRI, diaphragm evaluation |
2 |
|
Appropriateness Criteria Scale
1 2 3 4 5 6 7 8 9
1 = Least appropriate 9 = Most appropriate
|
Note: Abbreviations used in the tables are listed at the end of the "Major Recommendations" field.
Variant 2: Unstable patient.
| Radiologic Exam Procedure |
Appropriateness Rating |
Comments |
| X-ray, chest, upright |
7 |
|
| US, screen for hemoperitoneum |
7 |
|
| X-ray, abdomen, supine and upright |
6 |
|
| US, organ |
4 |
|
| INV, angiography embolization for bleeding |
4 |
|
| CT, abdomen & pelvis |
4 |
MDCT is preferable. Clinical judgment needed on stability of patient versus need for diagnostic information. |
| MRI, organ evaluation |
2 |
|
| MRI, diaphragm evaluation |
2 |
|
Appropriateness Criteria Scale
1 2 3 4 5 6 7 8 9
1 = Least appropriate 9 = Most appropriate
|
Note: Abbreviations used in the tables are listed at the end of the "Major Recommendations" field.
Variant 3: Hematuria > 35 Red Blood Cells (RBC)/High Power Field (HPF) (stable).
| Radiologic Exam Procedure |
Appropriateness Rating |
Comments |
| X-ray, chest, upright |
8 |
CT and x-rays may be appropriate. See original guideline document text for details. |
| X-ray, abdomen, supine & upright |
8 |
CT and x-rays may be appropriate. See original guideline document text for details. |
| CT, abdomen & pelvis |
8 |
MDCT is preferable. CT and x-rays may be appropriate. See original guideline document text for details. |
| CT cystogram, abdomen & pelvis |
7 |
|
| Retrograde urethrogram |
7 |
If urethral injury is suspected. |
| IVU |
4 |
|
| X-ray, cystogram |
4 |
|
| INV, renal angiogram |
4 |
|
| US, organ |
3 |
|
| US, bladder |
3 |
|
| MRI, kidneys and bladder |
2 |
|
Appropriateness Criteria Scale
1 2 3 4 5 6 7 8 9
1 = Least appropriate 9 = Most appropriate
|
Note: Abbreviations used in the tables are listed at the end of the "Major Recommendations" field.
This review considers only the issue of blunt abdominal trauma in adults. Penetrating trauma and pediatric cases are not considered.
Category A
Hemodynamically unstable patients presenting to the emergency room with clinically obvious major abdominal trauma and with unresponsive profound hypotension need rapid clinical evaluation and immediate resuscitation with volume replacement. If such unstable patients do not respond to resuscitation (become hemodynamically stable), and if they have clear clinical evidence of abdominal injury, they should go immediately to the operating room without imaging. During resuscitative efforts if time and circumstances permit, conventional radiographs of the chest and abdomen are often obtained as part of trauma protocols. This may help identify a pneumothorax, pneumoperitoneum, or significant bone injury. Ultrasound performed by an experienced sonologist to check for intraperitoneal free fluid may quickly provide information that can support a decision to operate immediately, with the caveat that the false negative rate is at least 15%. More detailed ultrasound to check for organ injury takes too long in this setting and suffers from poor sensitivity. There is now general agreement that routine diagnostic peritoneal lavage (DPL) is obsolete because of its invasive nature, lack of specificity, and inability to predict the need for therapeutic surgery.
Category B
Hemodynamically stable patients, patients with mild to moderate responsive hypotension presenting to the emergency room after blunt abdominal trauma, and unstable patients who stabilize after initial resuscitation are in a separate category. These patients typically have a history of significant trauma and have at least moderate suspicion of intra-abdominal injury based on clinical signs and symptoms. For these patients, two decisions need to be made: (1) Is urgent therapeutic surgery or angiography needed? (2) If surgery is not needed, is a period of close observation warranted? If computed tomography (CT) is to be performed, plain films will offer little if any incremental help with those questions. Rather, the decision to proceed with urgent surgery depends on the identification of specific CT criteria that predict that the surgery will be therapeutic: active hemorrhage, parenchymal "blush" or pseudoaneurysm in the spleen, or perforation of a hollow viscus (including the pancreatic duct). In patients with active hemorrhage or pseudoaneurysm of the spleen, angiographic embolization may also be therapeutic. The decision to operate urgently does not solely depend on the identification of hemoperitoneum or the identification of parenchymal injury to the liver or spleen, because most patients in this category ultimately do not need surgery. However, accurate identification of hemoperitoneum or organ injury is important because patients with these findings require at least a period of close observation. Patients with multiple organ injury or significant active bleeding may need surgery even if they are hemodynamically stable. Conversely, stable patients with isolated organ injury may not need surgery (or may need only angiography plus embolization) even with a large amount of hemoperitoneum.
Either way, time is available in such patients to obtain chest and abdominal radiographs, a hematocrit plus blood chemistries, and a urinalysis. If a reliable abdominal exam can be performed (the patient is conscious and does not need prolonged anesthesia for other procedures) and all the above preliminary tests are unremarkable, a period of close observation may be all that is needed. However, if a reliable abdominal exam cannot be performed (patient is unconscious or prolonged nonabdominal surgery is anticipated) or if a clinical evaluation suggests organ injury, hemoperitoneum, or peritonitis, further imaging is needed.
At this point, ultrasound is not a good modality for further imaging because it misses up to 25% of liver and spleen injuries, most renal injuries, and virtually all pancreatic, mesenteric, and gut injuries. It also misses a high proportion of retroperitoneal hemorrhage and of bladder rupture. Combining the results for ultrasound in 1535 abdominal trauma patients from eight published series yields an average sensitivity for hemoperitoneum of 88% and for organ injury of 74%. Unfortunately, a negative ultrasound (absence of hemoperitoneum) does not rule out significant organ or viscus injury that might require surgery or observation.
Although ultrasound is 63% sensitive to moderate amounts of free intraperitoneal fluid (compared with computed tomography), 400 to 600 cc's are needed for ultrasound detection of fluid in the trauma setting. Almost regardless of volume, an ultrasound diagnosis of free fluid alone does not predict that surgery is needed or that surgery will be therapeutic. In addition, in the best of hands, there is at least a 15% false negative rate for detecting hemoperitoneum with ultrasound. Further, ultrasound is quite insensitive in detecting organ injury: 62% of spleen and 14% of liver injuries are missed compared with computed tomography and operative findings. Ultrasound poorly identifies active hemorrhage and also does not accurately predict the need for surgery in splenic injuries.
Ultrasound is also insensitive to perforation of gut and to pancreatic injury. For these reasons, it is not very useful in deciding when a patient needs urgent therapeutic surgery or angiography. For the same reasons, ultrasound is not an accurate modality to determine whether a patient needs a period of close observation; thus, if a negative ultrasound is the sole imaging modality used to triage a patient, for safety reasons it must be followed by a 12- to 24-hour period of in-hospital observation. It should be noted that 96% of trauma centers perform fewer than two trauma ultrasound exams per month, so there is currently little national experience with or teaching of trauma ultrasound.
In contrast, for category B trauma patients, computed tomography accurately predicts if therapeutic surgery is urgently needed by identifying active hemorrhage, splenic injury (either parenchymal contrast blush or pseudoaneurysm), gut perforation, and pancreatic injury. For these reasons, it is an excellent modality for deciding whether a patient needs urgent therapeutic surgery or is a candidate for therapeutic angiography. Because computed tomography is sensitive in detecting both hemoperitoneum and injury to the liver (sensitivity 93%) and spleen (sensitivity 95%), it is an accurate modality for deciding if a patient needs a period of close observation. The trend toward placing helical computed tomography scanners close to or in emergency departments has substantially diminished the delay in getting patients to the computed tomography scanner and has decreased actual scan time to less than 40 seconds. In most circumstances, results from a helical computed tomography of the abdomen and pelvis can be obtained faster than results from a detailed ultrasound that includes evaluation of abdominal organs and gut.
If multidetector computed tomography with rapid image display capability is available in or next to an emergency department, abdominal computed tomography can be performed in about 2 minutes - excluding time needed for patient transport, computed tomography scan setup, and archiving of images. Including all time requirements, patient turnaround with rapid-process multidetector computed tomography can be less 10 minutes for a trauma patient. For single-slice incremental computed tomography, turnaround time is somewhat longer, usually 20 minutes. Scanning multiple body regions increases these times variably.
An experienced radiologist should carefully examine images on film, picture archiving and communication system (PACS), or at the computed tomography console, where images can be altered to seek bone injury, pneumoperitoneum, or subtle organ injury. Particular care should be taken to find minimal injury of the spleen because these patients may need observation for potential delayed hemorrhage. In some instances, stable patients with more severe injuries of the liver or spleen plus hemoperitoneum may be managed conservatively with only close observation. It should be noted, however, that various schemes for using computed tomography to grade liver or spleen lacerations are not helpful in deciding whether a patient needs surgery. This decision must be based on the clinical status of the patient in combination with the image findings. If evidence of active hemorrhage is discovered on computed tomography exams, the patient may be taken to the operating room or undergo arteriography plus embolization to control the hemorrhage.
The computed tomography exam should be carefully examined for subtle signs of pancreatic injury because these patients may need immediate surgery or close observation for signs of complications. Duodenal perforation produces subtle but frequent findings on computed tomography, e.g., typically extraluminal air or contrast in the retroperitoneum or elsewhere; these findings mandate surgical intervention. Duodenal hematoma may not require surgery but does need close observation. Other gut injury or perforation produces direct or indirect findings on computed tomography in 50% to 94% of cases. However, if the computed tomography is negative for gut injury in the face of a high clinical suspicion, diagnostic peritoneal lavage, laparoscopy, or a period of observation plus repeat computed tomography may be used to further evaluate the patient.
There may be a rationale for creating a subcategory of stable patients with trivial trauma, a low clinical index of suspicion, and no signs or symptoms of intraabdominal injury. In such patients, a negative ultrasound alone may be adequate to release the patient from observation at a lower cost than if computed tomography had been used. Computed tomography is necessary, however, if there are any positive findings on ultrasound.
It may also be reasonable to use computed tomography, in conjunction with the clinical information, to decide whether to observe patients in the hospital for a day or send them home promptly at the completion of their investigation in the emergency department. The high sensitivity of computed tomography in detecting injuries that require observation in the hospital means that a negative computed tomography may be adequate to release the patient to home in selected cases. Ultrasound has a substantially lower sensitivity to the kind of injuries that must be observed in the hospital. For this reason, a negative ultrasound is not adequate to safely release the patient to home. This weakness of ultrasound is reflected in the design of many outcomes-based investigations on the use of ultrasound in trauma: all keep patients with a negative ultrasound in the hospital for a period of observation of 12 to 48 hours before release.
Category C
Patients with hematuria require some modification to the imaging workup. Patients with microscopic hematuria (less than 35 red blood cells per high power field) do not need specific urinary tract imaging. All patients with microscopic hematuria greater than 35 red blood cells per high power field, with macroscopic hematuria, or with fracture/diastasis of the symphysis pubis and its rami plus any hematuria need imaging of the urinary tract. If the urethral meatus has gross blood, if there is a floating prostate, or if a Foley catheter cannot be passed, a retrograde urethrogram should first be performed to rule out urethral injury. However, if clinical evaluation or the urethrogram indicates no urethral injury, a computed tomography cystogram should be added to the abdominal computed tomography (see appendix to the original guideline document). Computed tomography images should be examined carefully for evidence of renal perfusion, hemorrhage, or extravasation of contrast or urine from the kidney or bladder. Two studies have documented the poor ability of ultrasound to detect injuries of the kidney. All but the worst renal injuries are treated with observation; intraperitoneal bladder rupture is usually treated with surgical repair.
Abbreviations
- CT, computed tomography
- INV, invasive
- IVU, intravenous urography
- MRI, magnetic resonance imaging
- US, ultrasound
