| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Neuroradiology |
1 From the Department of Radiology, Universidad de Antioquia, Hospital Universitario San Vicente de Paul, Calle 64 x Kra. 51D, Medellin, Colombia. Received July 12, 1999; revision requested August 25; revision received November 10; accepted December 6. Supported in part by Quimica Schering, Bogotá, Colombia. Address correspondence to F.M. (e-mail: fmunera@epm.net.co).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: A prospective study was conducted during 24 months in 60 patients with penetrating neck trauma who were referred for conventional angiography owing to clinical suspicion of arterial injury. In the patient population, 146 arteries (77 carotid, 69 vertebral) were studied by means of conventional angiography. In all patients, conventional angiography and helical CT angiography were completed within 6 hours. Two radiologists interpreted helical CT angiographic studies by means of consensus. Conventional angiography was the standard of reference for determining the sensitivity and specificity of helical CT angiography.
RESULTS: Conventional angiograms showed arterial injuries in 10 (17%) of 60 patients. Conventional angiographic findings were arterial occlusion (n = 4), arteriovenous fistula (n = 2), pseudoaneurysm (n = 3), pseudoaneurysm with arteriovenous fistula (n = 1), and normal arteries (n = 136). Nine of 10 arterial injuries and all normal arteries were depicted adequately at helical CT angiography. Sensitivity of helical CT angiography was 90%, specificity was 100%, positive predictive value was 100%, and negative predictive value was 98%.
CONCLUSION: The sensitivity and specificity of helical CT angiography are high for detection of major carotid and vertebral arterial injuries resulting from penetrating trauma.
Index terms: Carotid arteries, angiography, 172.12113, 172.1245, 904.1222 • Carotid arteries, CT, 904.12916, 904.12917, 904.12919 • Carotid arteries, injuries, 904.411, 904.494, 904.73 • Computed tomography (CT), angiography, 904.12916 • Computed tomography (CT), clinical effectiveness, 904.12916 • Gunshot injuries, 904.411, 904.494, 904.73 • Vertebral arteries, CT, 1751.12113, 1751.12116, 901.12916, 901.12917 • Vertebral arteries, injuries, 1751.1243
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
Conventional angiography is considered the standard of reference for detection of vascular injury (15). In the neck, arteriography usually is performed if an abnormality is suspected clinically and immediate surgical intervention is not indicated. Conventional angiography is not recommended in patients who are hemodynamically unstable (15). This test provides the information regarding the site and type of injury that is necessary for surgical planning and thereby minimizes blood loss and prevents complications (3).
An added benefit of arteriography is that it offers the possibility of performing therapeutic interventions such as bleeding control through embolization or stent placement and temporary balloon occlusion of a major artery for determination of possible neurologic effects of arterial ligation (15–21). However, conventional arteriography is an invasive procedure that may result in severe complications (22,23). Alternative, noninvasive methods for detection of arterial injuries have been proposed for patients with penetrating neck trauma. These include ultrasonography (US) with color Doppler (8,24,25) and magnetic resonance (MR) angiography (26,27).
Helical computed tomographic (CT) angiography has been shown to be useful for detection of atherosclerotic occlusive disease of the carotid arteries (28–33). However, the role of this test in the examination of patients with possible carotid and vertebral arterial injuries in the setting of penetrating trauma has not been evaluated thoroughly. The purpose of this investigation was to determine the sensitivity and specificity of helical CT angiography for detection of carotid and vertebral arterial injuries caused by penetrating trauma to the neck, with the use of conventional arteriography as the standard of reference.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
Angiography sometimes is performed in patients who are asymptomatic and have injuries in close proximity to major vasculature. Patients who are hemodynamically unstable and have penetrating injuries are taken directly to surgical exploration. The number of neck vessels studied by means of conventional angiography is guided by the referring surgeons and depends on the location and direction of the injury and the clinical status of the patient. We excluded patients with history of adverse reactions to iodinated contrast agents and patients with known diabetes; hypertension; and cardiac, peripheral vascular, or renal disease.
During the study period, 64 patients with penetrating trauma were referred for conventional angiography for suspected lesions of neck arteries. Four of them were excluded from the study owing to coexistent medical conditions that might have increased the risk of nephrotoxicity as a result of the dual injection of iodinated contrast agent: diabetes (n = 2), cardiac disease (n = 1), and hypertension (n = 1). These four patients were examined successfully by means of conventional angiography. Therefore, the study population comprised 60 patients (55 male, five female; age range, 15–69 years; mean age, 27 years). Distribution according to the mechanism of injury was gunshot wound (n = 58) and stab wound (n = 2). Informed consent was obtained from all patients. When the patient's clinical condition was critical, a close relative signed the consent form. Our institution's investigation review board approved the study.
Imaging Studies
In all patients, helical CT angiographic studies were completed before and within 6 hours of conventional angiography. Helical CT angiographic examinations were performed with a CT unit (Prospeed; GE Medical Systems, Milwaukee, Wis). Contrast agent was injected through an 18-gauge catheter placed in an antecubital vein. We administered 100 mL of iopamidol (Iopamiron [300 mg of iodine per milliliter]; Schering, Bogotá, Colombia) at a rate of 4.5 mL/sec by using a power injector. Transverse sections were scanned at 3-mm collimation and a table speed of 4 mm/sec (pitch factor, 1.33) from the inferior endplate of the seventh cervical vertebra to the base of the mandible. Other scanning parameters were a tube current of 250 mAs and 120 kVp, a 20-cm field of view, and a 36-second exposure at one image per second. The images were reconstructed by using a 180° linear interpolation algorithm.
To determine scan delay, a test dose of 20 mL of iopamidol was injected at 4.5 mL/sec in the first 10 patients enrolled in the study. For this test injection, transverse sections were scanned at the level of the bifurcation of the carotid arteries, starting 8 seconds after the beginning of the injection of contrast agent, with an interscan delay of 1 second; we scanned 12 transverse sections per patient. The attenuation of opacified blood in the carotid artery was measured by using circular or oval regions of interest. This information was used to create time-attenuation curves. Scan delay was chosen as the minimum time required for the blood to achieve an attenuation coefficient of 120 HU. The scan delays determined by using this approach showed little variation among patients (10–13 seconds). Therefore, we decided to use a fixed scan delay of 11 seconds in the remaining 50 patients included in the study.
Once acquired, the volumetric data set was reconstructed retrospectively at 1-mm intervals. The complete set of reconstructed source images was transferred to an independent computer workstation (Advantage Windows; GE Medical Systems) for postprocessing. Postprocessing of the source data was performed by using software provided with the workstation by the manufacturer. Standard shaded-surface display, maximum intensity pixel projection, multiplanar reformatting, and volume rendering algorithms were used to produce two- and three-dimensional renderings. Reformations were generated by one of two investigators (D.P., S.M.V.) who were not involved in image interpretation and who were unaware of the clinical status of patients and of the results of conventional angiography.
For reformations obtained by means of maximum intensity pixel projection, 10 projections were generated at 10° increments from the true coronal projection to the true lateral projection. For shaded-surface display images, only the frontal (coronal) and lateral (sagittal) projections were created and saved. Thus, for each patient, 12 two- and/or three-dimensional reformations were created and saved at the workstation for interpretation. The total time required for postprocessing was approximately 15 minutes per patient. Independent reformation images of both carotid arteries were obtained in all patients. The vertebral arteries were evaluated exclusively with the transverse source images and with the curved-linear option of the multiplanar reformatting algorithm.
Conventional angiography was performed by using selective catheterization and serial imaging with film hard-copy or digital subtraction techniques. Only the vessels suspected of being injured, as determined by the referring surgeons, were catheterized and studied. The distribution of neck vessels that were evaluated by means of conventional angiography was as follows: carotid arteries in 60 patients (21 right only, 22 left only, and 17 both right and left) and vertebral arteries in 53 patients (19 right only, 18 left only, and 16 both right and left). Therefore, the total number of arteries studied by means of conventional angiography was as follows: 77 carotid arteries (38 right and 39 left) and 69 vertebral arteries (35 right and 34 left). A minimum of two planes, anteroposterior and oblique, was scanned in every patient. The mean number of injections per patient was four (range, 2–8). The mean volume of contrast agent used for conventional angiography was 70 mL (range, 30–100 mL).
Interpretation of Helical CT Angiographic Studies
Two radiologists (F.M., J.A.S.) interpreted the helical CT angiographic studies by means of consensus. The interpreting radiologists were not involved in generating the two- and three-dimensional reformations and had no knowledge of clinical data or conventional angiographic results. Both radiologists were fellowship trained in body imaging and had experience with interpretation of helical CT angiographic studies. Interpretation was performed at the computer workstation in an interactive fashion. The complete set of source images and two- and three-dimensional renderings obtained previously were available for interpretation. The interpreting radiologists were given the option to acquire additional projections when necessary.
The interpreting radiologists initially were asked to assess helical CT angiographic images for overall quality, especially for the presence of artifacts that resulted from patient motion or metallic fragments. On the basis of image quality, the studies were judged as being adequate or not adequate for interpretation. If the radiologists considered that the studies were of diagnostic quality, the carotid and vertebral arteries were evaluated for the presence of abnormalities. These were determined as being absent (normal examination) or present (abnormal examination). These data were used to calculate the overall sensitivity and specificity of helical CT angiography.
When an abnormality was detected, the radiologists were asked to classify the findings as follows: partial obstruction or complete obstruction (occlusion), pseudoaneurysm (extravascular collection of contrast agent), arteriovenous fistula (early filling of venous structures), and intimal flap (intraluminal linear filling defect with contrast agent on both sides). The radiologists were asked to look for fractures, soft-tissue hematoma, or other nonarterial findings on helical CT images that, in their opinion, potentially could be useful for patient care.
The standard of reference used for determination of the sensitivity and specificity of helical CT angiographic studies was the official interpretation of the conventional angiogram provided by the angiographer who performed the examination. This angiographer was the radiologist in charge of performing emergent conventional arteriography at the time of patient admission. The angiographer was asked to classify abnormal angiographic findings by following the same classification used for helical CT angiographic studies: partial obstruction or complete obstruction (occlusion), pseudoaneurysm (extravascular collection of contrast agent), arteriovenous fistula (early venous filling with contrast agent), and intimal flap (intraluminal linear filling defect with contrast agent on both sides). We derived 95% CIs (34). For these determinations, only those vessels that were studied by means of conventional angiography were considered. Vessels studied only by means of helical CT angiography were excluded from statistical analysis.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
Arterial Findings
Ten (17%) of the 60 patients had arterial injuries demonstrated at conventional angiography: seven occurred in carotid arteries and three in vertebral arteries. There were no patients in whom more than one artery had injury demonstrated at conventional angiography. Helical CT angiography properly demonstrated six of the seven carotid arterial injuries, so there were six true-positive helical CT angiographic examinations. The lesions demonstrated at both helical CT angiography and conventional arteriography were as follows: occlusion of the internal carotid artery (n = 3) (Figs 1, 2), pseudoaneurysm of the common carotid artery (n = 2), and pseudoaneurysm at the carotid arterial bifurcation with arteriovenous fistula communicating between the common carotid artery and the internal jugular vein (n = 1) (Fig 3). In one patient, conventional angiography showed a small pseudoaneurysm at the origin of the common carotid artery that was not detected at helical CT angiography. This false-negative helical CT angiographic examination resulted from not including the lower neck in the region studied.
|
|
|
|
|
|
|
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
Conventional arteriography is considered the standard of reference for diagnosis of arterial injuries. Unfortunately, there is a risk of procedure-related complications that is inherent to the invasive nature of arteriography. These complications include hematoma at the puncture site, thrombosis of the vessel used to access the arterial system, distal embolization of atheromatous plaques and thrombi, arterial spasm and ischemia, and intimal dissection (36). The reported frequency of these complications varies 0.16%–2.00% (36). In the major arteries of the neck, these complications may lead to ischemic events in the central nervous system, sometimes with catastrophic results and permanent neurologic sequelae.
For purposes of surgical management, the neck has been divided into three anatomic zones: I, below the sternal notch; II, between the sternal notch and the angle of the mandible; and III, from the angle of the mandible to the base of the skull (35). Angiography is indicated in patients with penetrating injuries in zones I and III of the neck. Some groups also have used arteriography in patients with zone II injuries without obvious findings at physical examination (9,21).
Owing to the high rate of conventional angiograms with negative results, various groups have proposed alternative noninvasive tests, such as color Doppler US, as methods for diagnosis or as initial screening modalities in patients with penetrating neck injuries (8,22,29–31,37). However, color Doppler US has not gained widespread acceptance in this setting. Limitations that decrease the use of this method for this indication include the dependence on well-trained and careful operators who may not always be available at some institutions and the technical difficulties for US evaluation of neck arteries in the presence of large hematoma or subcutaneous emphysema.
MR angiography also has been used to detect arterial injuries (31). However, there are several impediments for routine use of MR angiography as a method for diagnosis in trauma patients. Patients in critical condition are difficult to monitor properly in the MR imager. These patients often have equipment and tubes that are incompatible with MR imaging. Furthermore, in many institutions, the MR imaging system is not close to the emergency department, and MR imaging is not always readily available.
Many of the restrictions that make the use of MR imaging and MR angiography in trauma patients impractical can be overcome with helical CT angiography. Most trauma centers, as ours, have helical CT scanners always available and located close to the emergency department. Helical CT angiography has been used as a noninvasive alternative to conventional angiography in various applications and with good results (38,39,40). In the neck, most helical CT angiographic studies have focused on the diagnosis of arterial occlusive disease (28–33). The reported sensitivity and specificity for carotid arterial stenosis and dissections ranges from 87% to 100% (28–33,41,42). Helical CT angiography has been proved to have a high diagnostic performance in the detection of traumatic lesions in large arteries of the extremities (43).
In our study, we assessed the diagnostic performance of helical CT angiography for detection of arterial injuries in arteries of the neck by using conventional angiography as the standard of reference. The diagnostic performance of helical CT angiography in this setting was high, with a specificity of 100% and a sensitivity of 90%. The single false-negative helical CT angiographic examination resulted from not including the origin of the right common carotid artery in the region evaluated at helical CT. This pitfall can be avoided by scanning routinely from the top of the aortic arch to the base of the skull.
In addition to the great availability of helical CT scanners, there are other advantages of helical CT angiography that make it a useful examination in acutely ill patients such as trauma patients. Since the acquisition time of helical CT angiographic studies is short (1 minute or less), patients usually remain in the CT suite for no longer than 10 minutes. Therefore, care of patients is not delayed by the angiographic examination. Although postprocessing of the raw data is more time-consuming (15–20 minutes per patient in our study), in most patients the information provided by the transverse images alone is sufficient to make initial decisions about therapy, such as the need for immediate surgery. In addition, patients with penetrating trauma may require examination of a different anatomic region with CT and, therefore, the total amount of time that the patient spends in the radiology department for diagnostic examinations is reduced; this improves patient care by allowing faster diagnosis and treatment.
Another advantage of helical CT angiography is that technical parameters for acquisition of data are fixed, and the ability or level of training of the operator does not affect the quality of the studies. An additional advantage of helical CT is that it can be used to diagnose associated injuries. In our study, we found 21 patients with abnormalities not related to the arteries.
There are disadvantages of helical CT angiography, as compared with conventional angiography, that could limit the use of this technique in the setting of penetrating trauma. These include the requirement for a higher volume of contrast agent, lower spatial resolution, potential degradation of image quality from imaging artifacts, and the fact that therapeutic interventions cannot be performed immediately following diagnosis. Although the volume of contrast agent required for helical CT angiography is higher than that typically used with conventional angiography (100 mL vs a mean of 70 mL in our study), we believe that the risks associated with the volume excess are not substantial in most patients. However, this should be taken into account in patients who may require additional injections of contrast agent for other diagnostic procedures. In patients who require multiple injections of contrast agent, conventional angiography may be preferable, especially if there are medical conditions that increase the risk of adverse reactions from iodinated contrast agents.
The spatial resolution of helical CT angiography is lower than that of conventional angiography. This may limit detection of subtle abnormalities, such as isolated contour irregularities or intimal flaps. To our knowledge, although there is some evidence to suggest that most of these "minimal" injuries heal spontaneously over time and that no specific therapy is required (20,35), the natural history of these lesions is not well known yet. This potential limitation of helical CT angiography should be addressed in future studies.
Artifacts caused by metal, such as bullet fragments or dental fillings, may obscure arterial segments. To avoid the potentially devastating consequences of not diagnosing a clinically important arterial injury, we recommend that conventional angiography be performed whenever the integrity of the whole length of the clinically suspicious artery or arteries cannot be assessed fully by means of helical CT angiography.
Another important limitation in the small group of patients with arterial injuries is the impossibility of performing therapeutic interventions immediately following diagnosis. Although in the past most patients with arterial injuries required surgical treatment, transcatheter therapy of lesions such as pseudoaneurysm and arteriovenous fistula is now feasible at many centers (16–21). Therefore, if nonsurgical therapy is an option, helical CT angiography may not be required in patients with signs and symptoms usually associated with pseudoaneurysm or arteriovenous fistula, such as pulsatile hematoma, bruit, or thrill, since angiography still will be required prior to transcatheter therapy. However, since not all patients who present with these signs are found eventually to be candidates for transcatheter therapy, helical CT angiography may be helpful as a triage examination to direct patients to surgical or nonsurgical therapy.
A limitation of our study was the small number of patients with abnormal findings; however, the frequency of 17% is similar to that reported by other large trauma centers (8). This low prevalence is partially a result of the fact that many patients with clinically obvious arterial injuries are taken to the operating room immediately following initial resuscitation in the emergency department (8,35). We believe that the low rate of positive angiographic examinations underscores the importance of using an accurate, noninvasive diagnostic method as an alternative for initial evaluation. In this patient population, a negative helical CT angiographic examination could prevent further imaging or intervention and possibly decrease diagnostic work-up costs, since helical CT angiography is less expensive than conventional angiography. However, given the small number of patients with abnormalities that we found in our study, additional studies are necessary to validate these results before helical CT angiography can replace conventional angiography as the initial diagnostic modality.
In summary, in this study we report our initial experience with helical CT angiography in the examination of patients suspected of having injuries to cervical arteries. Our study results indicate that the sensitivity and specificity of helical CT angiography are high for detection of arterial occlusion, pseudoaneurysm, and arteriovenous fistula. Furthermore, since it is a rapid, noninvasive technique and is less expensive than conventional angiography, helical CT angiography may prove to be a good alternative to conventional angiography for the initial examination of patients with penetrating trauma to the neck. Although we believe our results are encouraging, larger clinical trials are required to determine if helical CT angiography can replace conventional angiography in this setting.
|
FOOTNOTES |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  R Toledano, J Corres, A Culebras, E Riva, and J Masjuan Isolated Horner's syndrome caused by intraoral gunshot Emerg. Med. J., December 1, 2006; 23(12): e65 - e65. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. J. Soose, J. P. Simons, and D. L. Mandell Evaluation and management of pediatric oropharyngeal trauma. Arch Otolaryngol Head Neck Surg, April 1, 2006; 132(4): 446 - 451. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. W. Stuhlfaut, G. Barest, O. Sakai, B. Lucey, and J. A. Soto Impact of MDCT Angiography on the Use of Catheter Angiography for the Assessment of Cervical Arterial Injury After Blunt or Penetrating Trauma Am. J. Roentgenol., October 1, 2005; 185(4): 1063 - 1068. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. E. A. Saad, W. E. A. Saad, M. G. Davies, D. L. Waldman, P. J. Fultz, and D. J. Rubens Pseudoaneurysms and the Role of Minimally Invasive Techniques in Their Management RadioGraphics, October 1, 2005; 25(suppl_1): S173 - S189. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. B. Nunez Jr, M. Torres-Leon, and F. Munera Vascular Injuries of the Neck and Thoracic Inlet: Helical CT-Angiographic Correlation RadioGraphics, July 1, 2004; 24(4): 1087 - 1098. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. L. Wolf Invited Commentary RadioGraphics, July 1, 2004; 24(4): 1099 - 1100. [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]
|
|
|
|
|
|
R. A. Willinsky, S. M. Taylor, K. terBrugge, R. I. Farb, G. Tomlinson, and W. Montanera Neurologic Complications of Cerebral Angiography: Prospective Analysis of 2,899 Procedures and Review of the Literature Radiology, May 1, 2003; 227(2): 522 - 528. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Munera, J. A. Soto, D. M. Palacio, J. Castaneda, C. Morales, A. Sanabria, J. E. Gutierrez, and G. Garcia Penetrating Neck Injuries: Helical CT Angiography for Initial Evaluation Radiology, August 1, 2002; 224(2): 366 - 372. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. MacFarlane and C. A. Benn Penetrating neck injury: a review Trauma, April 1, 2002; 4(2): 79 - 90. [Abstract] [PDF]
|
|
|
|
 |
|
 V. H. Gracias, P. M. Reilly, J. Philpott, W. P. Klein, S. Y. Lee, M. Singer, and C. W. Schwab Computed Tomography in the Evaluation of Penetrating Neck Trauma: A Preliminary Study Arch Surg, November 1, 2001; 136(11): 1231 - 1235. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
