| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Emergency Radiology |
1 From the Departments of Radiology (J.A.S., F.M., J.M.G., T.S.) and Surgery (C.M., A.S.), Universidad de Antioquia, Hospital Universitario San Vicente de Paúl, Calle 64 x Carrera 51D, Medellín, Colombia. From the 2000 RSNA scientific assembly. Received October 9, 2000; revision requested November 25; revision received January 29, 2001; accepted March 7. Address correspondence to J.A.S. (e-mail: jorgeasoto@aol.com).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: During 36 months, 32 patients with a penetrating stab wound to the abdomen were examined with serial US (at admission and 12 hours later) and helical CT, with contrast material administered orally, intravenously, and rectally. Presence of hemoperitoneum and integrity of solid and hollow viscera were evaluated with both methods. Sonograms were interpreted by the radiologist who performed the examination, and CT images were independently evaluated by two radiologists. Findings of both techniques were compared with clinical outcome and/or surgical findings.
RESULTS: One (3.1%) of 32 patients required surgery: Surgical findings were massive hemoperitoneum and an extensive hepatic laceration. Both US and CT depicted these abnormalities. Thirty-one (96.9%) patients were treated conservatively, without surgery, and remained asymptomatic during 28 days of clinical follow-up after discharge from the hospital. US and/or CT showed intraperitoneal abnormalities in 21 of these patients. In 11 patients, both methods showed no evidence of visceral injury or hemoperitoneum, and none of these patients required surgery.
CONCLUSION: Serial US and CT help guide treatment for stable patients with penetrating stab injuries to the abdomen.
Index terms: Abdomen, CT, 70.12114, 70.12115 • Abdomen, injuries, 70.41, 70.43 • Abdomen, US, 70.12981 • Trauma, 70.41, 71.43
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
Liberal use of computed tomography (CT) and ultrasonography (US) for the evaluation of blunt abdominal trauma has resulted in a substantial decrease in the number of negative laparotomy findings. Conservative management of many types of solid organ injuries is now possible (6–8). However, use of these tests for the evaluation of penetrating abdominal injuries has only recently been suggested (9,10), and mostly for persons with gunshot wounds. The purpose of this study was to evaluate the usefulness of CT and US for the initial assessment of penetrating abdominal stab wounds in patients who presented to the emergency department without indication for immediate laparotomy.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
During 36 months, 202 patients were admitted to the emergency department of our institution after being stabbed in the abdomen, with evidence of penetration of the peritoneal cavity. At admission, 170 (84.2%) had clinical indications for immediate laparotomy. The remaining 32 (15.8%) patients (30 male, two female; mean age, 26 years; age range, 15–48 years) fulfilled the entry criteria, and they constitute the population of this study. Location of the penetrating injury was right upper quadrant (n = 18), right lower quadrant (n = 9), left upper quadrant (n = 3), or left lower quadrant (n = 2). The mean time between injury and admission to the emergency department was 77 minutes (range, 10–600 minutes; median, 30 minutes). Twelve (36.3%) patients had stab wounds in other body regions—extremities and/or soft tissues (n = 8) and chest (n = 4)—in addition to the penetrating abdominal injury.
After admission to the emergency department for observation, all patients underwent a detailed clinical evaluation according to the criteria of the Advanced Trauma Life Support standards (11). All clinical evaluations and physical examinations were performed by the attending surgeon in charge of the emergency department at the time of patient admission. Vital signs and findings of abdominal examination were recorded. This clinical evaluation was repeated at 1, 3, 6, 12, and 24 hours after admission, according to the same criteria. If the clinical status remained unchanged during observation (ie, no tachycardia, fever, hypotension, peritoneal irritation, or gastrointestinal bleeding), patients were discharged after 24 hours. Conversely, if the clinical condition deteriorated and any of these signs developed, the patient immediately underwent laparotomy, and surgical findings were recorded.
For compliance with the investigations review board regulations of our institution, abnormalities detected at CT or US and judged to be life threatening, such as active extravasation of contrast material administered intravenously or definite bowel perforation, by the radiologists were immediately disclosed to the trauma surgeons. Other imaging findings were withheld from the physicians responsible for decisions about patient treatment; in these patients, the decision to intervene surgically was made on the basis of clinical evaluations during observation.
US Examination
Within 60 minutes after admission to the emergency department, all patients underwent real-time US of the abdomen. This evaluation was performed at the patient’s bedside, with a mobile unit (Sonoline S145D; Siemens Medical Systems, Erlangen, Germany), by using a 3.5-MHz convex transducer. This examination was performed and interpreted by the staff radiologist in charge of emergency radiology examinations at the time of admission. The radiologist was unaware of the clinical status of the patients and of results of laboratory or other imaging studies.
The radiologist examined the abdomen for the presence of free fluid. For this purpose, the peritoneal cavity was divided into six spaces: right subphrenic space, left subphrenic space, hepatorenal fossa, right paracolic gutter, left paracolic gutter, and pelvic cul-de-sac. Images of these six spaces were stored and printed. For each of the six spaces, the anteroposterior diameter of the largest fluid collection was measured in centimeters. These six measurements were added for each patient, and a free-fluid index was calculated.
Solid organs were evaluated for the presence of lacerations; special attention was given to the areas located near the site of skin entry of the wound. A laceration was defined as a focal linear alteration in parenchymal echogenicity containing fluid or echogenic material. In addition, the radiologists recorded the presence of pleural effusion when examining the subphrenic spaces. Bowel-wall and mesenteric hematomas were not specifically sought with US. Detected abnormalities were recorded by the radiologist. For all 32 patients, this US examination was repeated 12 hours after admission, with the same equipment and technique as those used in the baseline study. The radiologist performing this follow-up study was different from the one who performed the baseline examination.
Results of the first study were not disclosed to the radiologist responsible for the follow-up evaluation. We calculated the free-fluid index at 12 hours and compared it with that of the baseline study. To allow for discrepancies introduced by the variations in the operators, we considered an increase in the free-fluid index significant if it exceeded 5 cm.
Helical CT Examination
Within 60 minutes after the initial US examination, all patients underwent helical CT of the abdomen and pelvis (Prospeed SX; GE Medical Systems, Milwaukee, Wis). For CT, we administered 800 mL of iodinated contrast material (3% Gastroview; Mallinckrodt Medical, St Louis, Mo) perorally and 500 mL rectally. We also injected 100 mL of iopamidol (Iopamiron [300 mg of iodine per milliliter]; Schering, Berlin, Germany) intravenously at 2.5 mL/sec by using a power injector and a 60-second scanning delay. Acquisition parameters were 10-mm collimation, pitch factor of 1:1 to 1.5:1.0, 28–36-second exposure at one tube rotation per second, 120 kVp, and 200 mAs.
Helical CT images were interpreted independently by two radiologists (J.A.S., F.M.) who were not involved in performing or interpreting the US studies and who were both fellowship trained in abdominal imaging. Interpretation of CT images occurred at an independent workstation (Advantage Windows, GE Medical Systems). No information about the clinical condition of the patients (including location of the penetrating injury) or the results of US was provided to the interpreting radiologists.
Both radiologists recorded the following findings as evidence of bowel perforation: presence and location of free peritoneal fluid and air, pleural effusion, solid organ lacerations, active extravasation of intravenously administered contrast material, localized mesenteric or peritoneal hematoma, and presence of extraluminal oral contrast material. A laceration was defined as a focal, linear, low-attenuation lesion interrupting the continuity of solid organ parenchyma and/or surface. Laceration extent or severity was not determined. In addition, the radiologists evaluated the abdominal wall in an attempt to localize the penetrating wound. CT images were examined at soft-tissue window settings (level, 40 HU; width, 360 HU) and at lung window settings (level, -600 HU; width, 1,100 HU); the latter were used primarily for the detection of pneumoperitoneum. We calculated interobserver agreement for the presence of abnormalities at helical CT by using the 
statistic (12).
Follow-up
Findings at US and CT were compared with those of laparotomy for patients who required surgical exploration. Patients who did not undergo surgery were followed up clinically after discharge. These patients were discharged 24 hours after admission and followed up with physical examination every 48 hours during the first week and then every 7 days for 28 days after discharge. CT findings judged as not life-threatening by the interpreting radiologists were not disclosed to the surgeons in charge of these clinical evaluations at any point during the follow-up. If the clinical status did not vary during the 28 days of follow-up and patients remained asymptomatic, we assumed that no injury requiring surgical correction had been caused by the penetrating trauma.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
Findings at US and CT
At admission, US demonstrated abnormalities in 14 (43.8%) patients (Table 1). Eight patients had free fluid at admission; Table 2 lists the location and size of fluid collections and the free-fluid index in these eight patients at admission and at 12-hour follow-up. In two of these eight patients, the initial sonogram disclosed an associated solid organ injury: hepatic laceration (n = 1) (Fig 1) and renal laceration (n = 1). At follow-up US, there was variation in the abnormal findings in two patients: A hepatic laceration that was not seen with US at admission in one patient who recovered successfully with conservative treatment and an increase in the amount of free peritoneal fluid in one patient who eventually required laparotomy. No variation was noted in the follow-up US in the remaining 30 patients, including 17 of 18 patients with normal findings at admission and 13 of 14 patients with abnormalities. These 30 patients were treated successfully without laparotomy.
|
|
|
|
|
|
|
|
|
statistic, was 0.81 (very good). In the one patient who eventually required laparotomy, CT demonstrated a laceration in the left lobe of the liver and associated free fluid, but without active extravasation of intravenous contrast material. In addition, both radiologists accurately predicted the location of the penetrating injury on the basis of CT findings alone. Abnormal findings in the wound included localized free peritoneal air in all patients (Fig 1), localized hematoma in the subcutaneous fat (n = 23), and abdominal wall pseudoaneurysm (n = 1) (Fig 5). For the single patient who required laparotomy, surgical findings were used to confirm abnormalities demonstrated with helical CT: hepatic laceration and free peritoneal fluid.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
Abdominal pain is generally considered an indicator of peritoneal irritation. However, this symptom lacks specificity; in our study, for example, 24 (75%) of 32 patients complained of abdominal pain and only one eventually required surgery. Pain and localized peritoneal irritation may be caused by hemoperitoneum, a common finding in these patients (56% in our study). The high frequency of hemoperitoneum limits the diagnostic potential of peritoneal lavage, since false-positive results lead to nontherapeutic laparotomies.
The use of US and CT in persons with abdominal trauma is now commonplace (15). These modalities are widely used for guiding treatment of patients with blunt trauma (16,17). CT depicts hemoperitoneum, solid organ injuries, pneumoperitoneum, mesenteric hematomas, and diaphragmatic injuries, as well as life-threatening conditions, such as active bleeding and hypovolemic shock (18–20), with high sensitivity rates. Several groups have also reported the benefits of CT for certain patients with penetrating abdominal injuries. McAllister et al (21) demonstrated the value of CT performed with oral, intravenous, and rectal contrast material for penetrating injuries in the dorsum. Phillips et al (22) and Hauser et al (23) reported similar good results. These investigators recommend using CT for patients with penetrating trauma in the posterior and lateral regions of the abdomen. In other studies (24,25), however, the effectiveness of CT for penetrating abdominal injuries was limited.
Many groups have also shown the usefulness of performing emergency US in patients with blunt trauma while the patient is still in the emergency department (26). This limited US examination is highly sensitive for the detection of free fluid and most solid organ injuries. One of the advantages of US is that the test can be repeated frequently, allowing detection of increasing hemoperitoneum as evidence of ongoing bleeding. Relatively few studies (26–28) have used US for penetrating abdominal trauma patients. A drawback is the limited ability to demonstrate hollow viscus injuries, since sonograms may be normal or show only a minimal amount of free fluid.
In our study, we investigated how CT and US could be used to guide treatment for patients who had penetrating abdominal injuries caused by stab wounds and presented to the emergency department without clinical indication for immediate laparotomy. We used both modalities to capitalize on the relative advantages of each: the high sensitivity of CT for the detection of fluid and visceral lesions and the possibility of repeating US examinations for monitoring the amount of free fluid. As expected, we found hemoperitoneum in a large proportion of the patients in our study: 18 (56%) of 32 patients with CT and eight (25%) patients with US. CT depicted localized free peritoneal air in all patients, and this was helpful for localizing injury entry site. During follow-up, US depicted an increase in hemoperitoneum in one patient, who eventually required laparotomy. This suggests that free fluid at presentation is a nonspecific sign, but an increase over time should be considered suggestive of a substantial injury, and careful surveillance is warranted.
An increase in fluid may be detected with either repeat CT or repeat US, but we prefer the latter because of its lesser cost, lack of ionizing radiation, and greater availability. A limitation of using repeat US at 12-hour intervals for the determination of substantial variations in free-fluid volume is that the evaluations are likely performed by different operators. Variations in technique may introduce errors that decrease the validity of the test. For this reason, we used an increase in the free-fluid index of at least 5 cm as an indicator of a significant increase in fluid volume. Although this is an arbitrary number, it decreases the likelihood that differences in measurements are solely the result of changes in the operators.
CT was useful for demonstrating hepatic and renal injuries. Extensive experience in large trauma centers has proved the value of CT for detecting blunt traumatic lesions in solid viscera (29). Results of our study show that CT also depicts hepatic lacerations caused by stab wounds, even when associated with hemoperitoneum, and that many of these patients can be treated conservatively. However, this needs to be confirmed in a larger patient population. We did not attempt to grade the severity of solid organ lacerations; instead, we looked for active extravasation of intravenous contrast material as evidence of ongoing bleeding, but this finding was not seen in any patient. US demonstrated liver lacerations in four patients. Free fluid was found with US in eight patients and with CT in 18 patients. The difference is explained by the higher sensitivity of CT for detecting lacerations in solid viscera and small amounts of fluid.
In the 32 patients in our study, we found no patients with hollow visceral injuries that required surgery. This is probably related to the fact that we included exclusively patients who were going to be followed up clinically, and hollow viscus perforation manifests early in most patients, with generalized abdominal pain and peritoneal irritation. This is a limitation of our study. In four patients, CT showed localized mesenteric or pericolonic hematomas, without other evidence to suggest bowel perforation; none of these patients eventually required laparotomy. This suggests that a bowel-wall hematoma, as an isolated abnormal finding without extraluminal oral or rectal contrast material, is not an indication for laparotomy. However, as is the case with similar hematomas caused by blunt trauma, we believe that this finding should still be considered suggestive of bowel injuries and that these patients require close surveillance.
Our results also confirm that clinical follow-up during 24 hours allows adequate treatment of patients presenting without indication for immediate laparotomy. However, the use of CT and US, in combination, may reduce the period of observation. The results of our study suggest that a modification of the treatment protocol for these patients at our institution would decrease the length of hospital stay.
Patients with indication for laparotomy at admission are not subject to any imaging study. Stable patients will undergo helical CT, as described in our study. Patients with normal findings or with abnormalities limited to the entry site are discharged immediately and followed up as outpatients. Patients with isolated hemoperitoneum or with associated solid organ injury undergo baseline sonography after CT; these patients are then admitted for observation, and US is repeated 12 hours later. If the clinical condition remains stable and there is no increase in the amount of free peritoneal fluid, patients will be discharged and followed up as outpatients. Patients with CT findings of mesenteric or bowel-wall hematomas are admitted and immediately evaluated with baseline US. US is repeated at 12 and 24 hours. All patients in whom follow-up sonograms show an increase in free peritoneal fluid are treated as inpatients for at least 48 hours. Laparotomy is performed at any time if signs of peritoneal irritation develop.
Our results suggest that the use of helical CT performed with oral, intravenous, and rectal contrast material and serial US, in combination, may help guide treatment for patients who have penetrating stab wounds to the abdomen but who do not have an indication for immediate laparotomy. The length of hospital stay for many of these patients may be decreased by using the imaging algorithm described herein. However, this algorithm should be proved in a larger patient population. These imaging techniques may also prove useful for evaluating patients in whom clinical examination is difficult, such as those with prior surgical interventions or incisional hernias and those with an altered level of consciousness.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  A. Salim, B. Sangthong, M. Martin, C. Brown, D. Plurad, K. Inaba, P. Rhee, and D. Demetriades Use of Computed Tomography in Anterior Abdominal Stab Wounds: Results of a Prospective Study Arch Surg, August 1, 2006; 141(8): 745 - 752. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Rieger, A. Mallouhi, T. Tauscher, M. Lutz, and W. R. Jaschke Traumatic Arterial Injuries of the Extremities: Initial Evaluation with MDCT Angiography. Am. J. Roentgenol., March 1, 2006; 186(3): 656 - 664. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C Ertekin, H Yanar, K Taviloglu, R Guloglu, and O Alimoglu Unnecessary laparotomy by using physical examination and different diagnostic modalities for penetrating abdominal stab wounds Emerg. Med. J., November 1, 2005; 22(11): 790 - 794. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Munera, C. Morales, J. A. Soto, H. I. Garcia, T. Suarez, V. Garcia, M. Corrales, and G. Velez Gunshot Wounds of Abdomen: Evaluation of Stable Patients with Triple-Contrast Helical CT Radiology, May 1, 2004; 231(2): 399 - 405. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
