| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2001;72:495-502
© 2001 The Society of Thoracic Surgeons
a Department of Surgery, University of Maryland School of Medicine, Baltimore, Maryland, USA
b Departments of Medicine and Anesthesia, University of Maryland School of Medicine, Baltimore, Maryland, USA
c Section of Surgery, R. Adams Crowley Shock Trauma Center, Baltimore, Maryland, USA
d Section of Radiology, R. Adams Crowley Shock Trauma Center, Baltimore, Maryland, USA
Address reprint requests to Dr Downing, Division of Cardiac Surgery, University of Maryland Medical Center, Suite N4W94, Baltimore, MD 21201
e-mail: sdowning{at}smail.umaryland.edu
Presented at the Forty-seventh Annual Meeting of The Southern Thoracic Surgical Association, Marco Island, FL, Nov 9–11, 2000.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
Methods. Retrospective review of 54 patients undergoing aortic repair from July 1994 to December 1999. Spiral CT was the initial diagnostic study in 52 patients. Pseudoaneurysm or aortic wall defect in the presence of mediastinal hematoma was considered diagnostic. Angiography, initially routine, was later performed only when requested by the surgeon, and for all "nonnegative" studies (periaortic hematoma without detectable aortic injury).
Results. Twenty-six patients underwent angiography before operation (group 1). Nineteen group 1 spiral CTs were unequivocally diagnostic; 7 were nonnegative and angiography was required. Twenty-eight other patients underwent repair based on spiral CT alone (group 2). There was one false-positive result in both groups. There were no unexpected operative findings. Mean time from admission to diagnosis was 5.7 ± 3.4 hours for group 1 and 1.7 ± 1.7 hours for group 2 (p < 0.01).
Conclusions. Operating on the basis of a diagnostic spiral CT is safe and expeditious. Aortography may be reserved for those with equivocal studies.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
Aortography has historically been the diagnostic standard for TAR at most centers [2–4]. In some institutions, transesophageal echocardiography (TEE) plays the same role [5]. However, both TEE and aortography are invasive, time-consuming, and expensive [6]. In addition, despite its position as the gold standard, aortography is not free of false-positive or false-negative results [3, 7, 8].
Contrast-enhanced computed tomographic (CT) scanning is an appealing alternative to aortography and TEE. Computed tomographic scanning is fast, relatively inexpensive, noninvasive, and readily available at all trauma centers [9]. In addition, a large number of blunt trauma victims are already undergoing CT to evaluate other injuries [10]. The role of CT in the evaluation of TAR has been controversial, however. This is due in part to earlier experience with conventional CT scanning, which was plagued by low sensitivities and specificities, and also by the perceived mandatory need for aortography before operation regardless of CT findings [2, 9, 11–13] .
Spiral CT scanning (S-CT) is the new standard for CT scanning. The S-CT can image an entire patient in seconds. The vascular images that are generated are much superior to conventional CT images, and are frequently comparable to aortography [9]. Experience from our institution in more than 1,100 blunt chest trauma patients has shown S-CT to be 100% sensitive, based on clinical follow-up, and 99.7% specific, with an 89% positive and 100% negative predictive value and an overall diagnostic accuracy of 99.7% [8]. Other researchers have confirmed similar results [3, 8, 9, 14]. Despite evidence that S-CT can be as accurate and effective as angiography, few cardiothoracic surgeons will operate on the basis of a positive S-CT alone, and most authorities recommend angiography or TEE to make the diagnosis of TAR before proceeding to operation [2, 10, 12].
At our institution S-CT has been the center of the algorithm for evaluating patients with suspected TAR since mid-1994. On the basis of our experience, we have evolved from using S-CT as a screening tool to identify those patients who require angiography [6, 9, 15] to operating on the basis of the S-CT alone.
We reviewed our experience to test the hypothesis that it is safe and expeditious to forgo angiography in patients with a clearly positive S-CT.
|
Material and methods |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
Group 2 was composed of patients who proceeded to the operating room on the basis of S-CT alone. The time from admission to diagnosis was calculated from the time of admission to the Trauma Bay and the time of completion of the final diagnostic study. This time period was chosen for review, instead of time from arrival to operation, to eliminate the effect of other diagnostic and therapeutic procedures, which occasionally took priority over treatment of the aortic injury. Time data were compared by unpaired t test (Excel; Microsoft, Redmond, WA).
The protocol for evaluating a potential TAR is outlined in Figure 1. Patients suspected of TAR by mechanism of injury, physical examination, or chest radiograph underwent S-CT as described by Mirvis and colleagues [8]. In brief, the S-CT scans were performed using a Siemens Somatom Plus 4 (Siemens Medical Systems, Iselin, NJ). The routine technique is 8 mm/s table movement (pitch = 1), with a scan delay 25 seconds after a bolus of 100 mL of Omnipaque 24 (iohexol; Nycomed, Princeton, NJ) for imaging the chest alone, or 150 mL of Omnipaque 240 for the chest and abdomen–pelvis. Scans encompassed the thoracic inlet to the upper abdomen, including the aortic hiatus. All injuries were detected on standard cross-sectional images. Angiography was performed on a Toshiba Digital angiography System (Toshiba Medical Systems, Tustin, CA) and was obtained using a 1,024 by 1,024 matrix in both left anterior oblique and anteroposterior projections. Forty milliliters of Omnipaque 350 (Nycomed) were injected at 20 mL/s using a 5F pigtail catheter. Additional supplemental views were taken as indicated.
|
|
|
Results |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
In 26 patients (group 1), the final decision to proceed to operation was based on the results of aortic angiography. Twenty-four patients underwent protocol screening S–CT followed by angiography. Two patients were taken directly to angiography (due to equipment availability). Nineteen patients in group 1 had clear evidence of aortic rupture on S-CT (group 1A), and on the confirmatory aortogram. Seven patients had suspicious, but not diagnostic CT scans (group 1B), and required angiography to complete the diagnosis. Of the patients in group 1B, 3 patients had mediastinal hemorrhage with no obvious aortic defect, 3 had mediastinal hemorrhage with questionable aortic irregularities, and 1 patient had an aortic filling defect with a modest mediastinal hematoma.
In 28 patients (group 2) the final decision to proceed to operation was based on the S-CT findings alone. There was one false-positive aortogram in group 1 (no screening CT in that patient), and one false-positive S-CT in group 2. Both of these patients had large ductal diverticuli and mediastinal hemorrhage at the time of operation, but no aortic rupture. There were no known false-negative studies in either group based on clinical follow-up during patients’ hospitalization.
During the study period, S-CT evolved from a screening modality into the principle diagnostic study as demonstrated in Figure 3. The most common signs of TAR detected on S-CT are summarized in Table 1. Mediastinal hemorrhage was present on 100% of the scans considered diagnostic (groups 1A and 2). A clearly detectable pseudoaneurysm was present in 86% of the patients in group 2 and 76% of those in group 1A.
|
|
The sensitivity and specificity for aortography in this selected group of patients was 100% (26 of 26 patients) and 96% (25 of 26 patients), respectively. Including all group 1 patients, the protocol of S-CT with selected angiography for patients with mediastinal hemorrhage alone or equivocal S-CT findings of aortic injury had a sensitivity of 100% (52 of 52 patients) and a specificity of 96% (50 of 52 patients).
No patient in our series underwent preoperative TEE. Eleven patients had intraoperative TEE, primarily to assist in patient management. In 3 of 11 patients (27%), the area of interest on the aorta was not visualized well enough to confirm or refute the diagnosis of TAR.
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
Screening for TAR
A plain chest radiograph is the most common initial study used to screen for TAR. Signs of TAR on chest radiograph include: a widened mediastinum, aortic knob abnormalities, tracheal or nasogastric tube shift, apical capping, wide paraspinal lines, a depressed left mainstem bronchus, and suspicious rib injuries [16]. However, no single or combination of radiographic signs demonstrates sufficient sensitivity to detect all cases of traumatic aortic rupture [16]. In addition, because of the high number of false-positive studies, relying on chest radiograph to select patients for aortography is inefficient and leads to the performance of large numbers of normal angiograms [6, 8, 9].
The use of conventional, contrast-enhanced CT scanning as a screening tool to select patients for subsequent angiography is well supported [9, 17]. Mirvis and other investigators advocate contrast-enhanced CT scanning for all patients with suspected TAR [6, 8–10]. Patients with mediastinal hemorrhage or signs of aortic injury (pseudoaneurysm, intimal flap, aortic contour abnormality, intraluminal thrombus, or pseudocoarctation) are recommended to undergo aortography to definitively diagnose aortic injury. Patients with no evidence of mediastinal hemorrhage or aortic injury on CT do not need further evaluation [6, 8, 9]. Used as a screening tool, conventional CT has a 100% sensitivity, 88% specificity, 100% negative predictive value, and 25% positive predictive value [6]. A negative contrast-enhanced CT effectively rules out TAR [2, 6, 8, 9].
The excellent sensitivity and strong negative predictive values of CT, as well as its ubiquity, speed, and noninvasive nature make CT scanning an ideal screening tool.
However, the specificity of conventional CT is lower than is desirable for a definitive diagnostic modality due to image imperfections arising from volume averaging (particularly in the aortopulmonary window), inconsistent vascular opacification, and cardiovascular and respiratory motion artifacts [9].
Some researchers still challenge the role of a screening CT in the diagnosis of TAR [2, 4]. They argue that if angiography is still required to confirm the diagnosis, eliminating CT and proceeding directly to angiography is more efficient. This approach, however, is significantly more costly, subjects many patients to unnecessary angiography, and has never been shown to be quicker or safer [6, 8, 9, 18].
Definitive diagnosis of TAR
Among the definitive diagnostic studies for TAR, aortography remains the gold standard [2, 3]. However, aortography is invasive, expensive, time-consuming, and is not free of false-positive or false-negative results [7, 8].
The TEE has gained in popularity as a primary diagnostic study. Sensitivities and specificities as high as 100% and 98% have been reported [5]. However, TEE is an invasive, operator-dependent study [5] that is not readily available at many trauma centers. In addition, the distal ascending aorta and portions of the proximal descending aorta may not be visualized [5, 19]. There are also potential concerns with airway and cervical spine management during TEE [19]. Our own experience with TEE for aortic trauma is limited to its use in the operating room on patients who have already been diagnosed by another modality. In this group, the aortic injury could not be visualized in 3 of 11 patients.
Computed tomographic scanning is an ideal diagnostic method for aortic injury due to its relatively low cost (as compared with angiography), nearly universal availability in emergency rooms, and its lack of operator dependence [9, 17]. In addition, at most trauma centers, CT scanning is already an integral part of the diagnosis and management of serious blunt injury [10, 18]. In the experience reported here, 52 of 54 TAR patients had indications for and underwent simultaneous CT scanning of another body part to evaluate potential injuries.
Due to technical limitations, conventional CT has not been reliable enough to function as the sole definitive diagnostic modality [2, 8, 11, 20]. During conventional CT scanning, the x-ray source rotates 360o around the patient to collect a single transverse image. It must then "unwind" back to the starting position before obtaining the next image slice. During the unwinding period the x-ray source also has time to cool. Because the imaging period for conventional CT encompasses several cardiac cycles and often several pulmonary cycles, there is motion artifact in cardiac, pulmonary, and vascular structures. This leads to blurring of vessel walls, and volume averaging in curving areas that are not perpendicular to the imaging plane, such as the aortopulmonary window. In addition, the required scanning time makes it difficult or impossible to precisely coordinate the intravenous contrast bolus with the actual scan of the aortic region of interest.
The S-CT is a newer, more effective CT technology. It uses slip-ring technology that eliminates the need for the stop–start unwinding motion, and incorporates roentgenogram sources that are capable of sustaining higher power levels for long periods of time without the need to stop and cool [9]. The x-ray source spirals continuously around as the patient advances through the machine (see Fig 4). Instead of multiple single images, a true volume of data are acquired. The speed of spiral technology allows S-CT to image the entire aorta while it is at the peak of enhancement after a bolus of intravenous contrast. A complete data set can be obtained during a single breath hold. These factors translate into less motion artifact, less respiratory misregistration, and improved definition of vascular structures.
|
The false-positive rates we observed were essentially identical for both groups examined (1 of 26 patients for aortography versus 1 of 28 patients for S-CT). Both of these false-positive results were in patients with mediastinal hemorrhage and ductal diverticuli (which are notoriously difficult to discern from a pseudoaneurysm using any diagnostic study) [3, 7].
As would be expected, our results demonstrate that eliminating aortography quickens the diagnostic process. Despite the promising reports of delayed therapy for TAR [3], speed is still important [4, 10]. Rupture from TAR appears to have a bimodal pattern; some patients rupture within 4 hours of admission, occasionally during the diagnostic studies [4, 10], whereas other patients appear to have more stable disease. During the study period, 1 patient who had a pseudoaneurysm detected on S-CT did rupture immediately after confirmatory angiography and did not survive operation.
The ability to make a rapid diagnosis is also useful because it allows for the prompt diagnosis and management of other concurrent life-threatening injuries. Many patients in our series underwent therapy for intracranial, intraabdominal, or pelvic injuries before the repair of their TAR. Although we do not advocate routine delayed therapy, we do triage injuries and treat the most life-threatening injuries first.
Some caution should be used in analyzing the sensitivity and specificity data. This study was only designed to determine the safety and efficacy of operating on the basis of a positive S-CT. As mentioned, studies designed to determine the sensitivity and specificity studies for S-CT have already been published [8]. Our data represent a very select group comprised only of patients who underwent operation. In addition, the "final" radiograph readings were frequently completed in batches, when the results of all studies were sometimes known to the radiologists. However, this is not relevant for group 2 patients, all of whom proceeded directly to operation based on the initial S-CT interpretation. In practice, the decision for operation is usually made initially by the cardiothoracic surgery resident, on advice from a senior radiology resident, with the cardiothoracic surgery attending reviewing the study before proceeding.
Eighty-six percent of patients who proceeded directly to operation had mediastinal hemorrhage and an aortic pseudoaneurysm on S-CT (see group 2; Table 1). This suggests that this combination of signs is the easiest for most readers to confidently detect. The interpretation of the CT for aortic injury is no more difficult than most other concepts of CT anatomy, and is currently taught to radiology residents in the first-year as a routine part of the curriculum. A grossly positive study is not difficult to read, and many patients fall into that category. The results of this study should be easily translatable to the general community as their experience with S-CT grows. In addition, these results were obtained using a single slice scanner (which is widely available throughout the United States). Multiple-slice scanners are now becoming available. Technical improvements in these devices have even further improved resolution making it even easier for radiologists and nonradiologists to interpret the studies.
It should be emphasized that we use S-CT for two purposes. The S-CT is a first screening tool. Patients who have no mediastinal hematoma are considered free of aortic injury. Patients who have a periaortic mediastinal hematoma without evidence of aortic wall abnormalities are categorized as having an "equivocal" or "nonnegative" study and receive further evaluation. Only patients who have a clearly detectable aortic injury (ie, pseudoaneurysm or wall defect) in addition to a mediastinal hematoma are considered to have a positive study. This is the group that proceeded directly to operation without angiography.
In conclusion, we have had excellent results with S-CT as the core of our diagnostic evaluation for TAR. It is an ideal screening study for aortic injury, and frequently is simultaneously indicated to evaluate potential concurrent head, abdominal, or pelvic injuries. When there are clear findings of TAR on S-CT, the patient may proceed safely to the operating room without other aortic studies. This approach streamlines the diagnostic process and eliminates the need for other unnecessary invasive studies.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
This is an important subject because most thoracic surgeons do not have a wide experience with this lesion; it is still uncommon except in the very large trauma centers and even fewer radiologists have much experience with it. There are so many diagnostic tests, including standard CTs, transesophageal echoes, and intravascular ultrasound, that it is a little difficult to choose the best diagnostic technique.
The first slide shows that from July 1994 until July of this year we did screening helical CTs of the chest, abdomen, and pelvis on approximately 15,000 consecutive blunt-injuried patients. One hundred ninety-nine of the helical CTs were suspicious for a traumatic ruptured aorta. Eighty-seven of these were confirmed by aortogram. Therefore, by our calculations the helical CT was in agreement with the presenters, 100% sensitive; it was 99.8% specific; however, its positive predictive value was only 44% and its negative predictive value, of course, was 100%. So the fact that it has a low positive predictive value leads us to think that although it is an excellent screening technique, it has much to be desired in terms of total specificity or accuracy. We would recommend screening with a helical CT as opposed to all the other techniques that have been mentioned.
Now, we look at the patients who had an aortography for suspected traumatic rupture. There were 199 that were suspicious on helical CT, 189 of these had an aortography, of which 112 were clearly negative by both the radiologists’ and the thoracic surgeons’ readings (they are all read by both); however, there were 5 patients who were initially negative by aortography, and because there was something suspicious on the CT or intravenous ultrasound, we repeated the aortogram; all 5 patients turned out to have minimal but definite aortic injury.
Therefore we conclude that the two techniques are complementary, not competitive. One cannot depend entirely on the helical CT for accuracy of diagnosis, although it is 100% sensitive as a screening technique. It should be done first and aortogram when the helical CT is suspicious.
Thank you.
DR NORMAN J. SNOW (Chicago, IL): I enjoyed this paper very much because it confirms my bias about this particular subject, which is about a 180-degree turn from 5 years ago. The CT scan is inexpensive, it is readily available, and it provides a great deal of additional information about chest trauma patients that you might not get otherwise. And I think that one of the most important things that is underemphasized is, as Dr Pate indicated, it is nearly perfect in its negative predictive value, and that can be very important, especially in the multiple trauma patient.
At a trauma center there may be many people who are transferred in, and I was very surprised by Tim Fabian’s study a number of years ago that indicated the total time from injury to operating room can be as long as 7 to 10 hours for these patients, which amazed me until I went back and looked at our own figures, and they are very, very similar. A lot of these people will come with "negative conventional CT scans" who do not have a spiral CT with contrast, and we have to guard against accepting those negative scans as the true state of the patient, and we may have to repeat the scan according to our own protocols.
I was interested that you defined your time of delay as the time of admission to the trauma bay to completion of diagnostic studies rather than time to the operating room, which really reflects the risk time to the patient for rupture. Were there no transfers in other than those terribly long delays of 90 hours that might have changed your statistics? And why do not you use time to the operating room? What other studies or procedures might be done between the time of the end of the work-up and the time to the operating room for the traumatic aortic injury itself? The algorithm that you propose is very logical, although there might be some slight differences among other trauma centers.
I have some questions for you. Do you advocate routine antihypertensive therapy for your patients with suspected aortic rupture while they have a chest radiograph although it is only a couple of hours according to your statistics? Did any patients of yours die or rupture preoperatively, that is, before the clamps were in place? Is there concern for the total dye load with a head, chest, and abdominal CT in these patients? Did any of your patients have trouble with renal failure, etc., because of the extra dye load? And do you think that the aortogram or the transesophageal echo, given a good echocardiographer, is the better choice for patients with an equivocal CT scan of the chest?
I enjoyed the paper very much.
DR DOWNING: Let me answer a couple of these things before I lose track. First, related to Dr Pate’s comments, I think that if you examine the literature it is very clear that CT scan, particularly spiral CT, is a great screen. If you do not have a mediastinal hemorrhage, you are very unlikely to have an aortic injury.
The specificity of the CT depends on what you use as your diagnostic criteria. We consider a mediastinal hematoma without other signs of aortic injury to be an indication for an aortogram. We do not consider mediastinal hematoma to be diagnostic. If you consider hematoma a positive finding, the specificity will appear low because a lot of patients who get an aortogram for hematoma alone do not have an aortic rupture. If you use clearly positive findings such as a big pseudoaneurysm, or an obviously visible contour abnormality like the ones I showed you on our slides, I think that your specificity will be very high; if you use general criteria for any irregularity of the aortic wall, it will be low.
Regarding the antihypertensive therapy, we do use antihypertensive therapy in patients who are stable and we suspect have an aortic injury.
We have had 1 patient rupture between angiography and the operating room, and that was a patient who had a positive finding on the spiral CT.
Regarding the dye load, I think that CT is going to give you less dye than the aortogram will, and I think that would be a lower risk. There are many factors that account for renal failure postoperatively and I would be hard pressed to say the contrast from any one of our studies did or did not contribute.
Regarding echo, for patients who have an equivocal study, we have relied on angiography. Our own experience with echo is modest. We have only used it in 11 patients, and those are patients who were already in the operating room. In 3 of these patients we could not visualize the area of concern in the aorta. Perhaps if you had an extensive experience or a better result with your echo, you might go to that instead of the aortography.
Your last question was about the time. The reason I chose the time from admission to completion of the final diagnostic study was that was the one data point we could be most accurate with. First, it was electronically documented in the radiology system. Second, we were interested in what is the time it takes to make the diagnosis. Once you make the diagnosis, you may go straight to the operating room and repair the aorta, or then you may say, "We have that diagnosis down, now we are going to fix the bleeding abdomen." So for us the important time to comparing studies was, how long does it take to make the diagnosis. There are many factors that would pollute the data if you looked at time after the study until time in the operating room, including things like how long it took to get the operating room ready and what did ortho do to the knee and so forth. So I think that just gives us a pure idea of the time.
DR THORALF SUNDT (St. Louis, MO): Steve, I enjoyed your paper very much. Your group, along with the group in Memphis, of course, continues to define the management of this condition. Fortunately, as you know, I work at an institution that sees considerably less trauma than yours, and I am happy with that.
It seems to me that there are two questions afoot here: one is making the diagnosis of whether or not there is an aortic injury, and the second is the precise site of that injury. Now, although the vast majority of such injuries certainly occur at the site of the ligament, for those of us who use some form of circulatory support, and I recognize that that is not universal, the site of injury may impact on our choice among perfusion techniques. Could you comment on going ahead to the operating room and making those decisions without an angiogram?
Again, I very much enjoyed your paper. Thank you.
DR DOWNING: It has not really affected our choice of technique. We use femoral vein to distal aorta heparin-bonded partial bypass without systemic heparin, and relying more on spiral CT has not changed our approach. You can see the location of the injury fairly well if you run through the CT scans, including giving yourself a rough idea of how far below the arch it may or may not be. We did have 3 patients with arch injuries proper that I did not include in this data. All of those patients had angiography because of the limited experience we have with CT as the sole study for injuries in the arch.
DR SUNDT: Did you examine these studies retrospectively? If you had applied your algorithm to them, would they have undergone CT alone? Would there have been any surgical difficulties?
DR DOWNING: We could readily see all of the injuries, but based on 3 patients I really would hate to say it was okay to just use the CT scan alone.
|
References |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  W. T. Lau, D. K.M. Wong, and H. W. Louie Patent Ductus Arteriosus Masquerading as Aortic Transection in a Trauma Victim Ann. Thorac. Surg., December 1, 2008; 86(6): e5 - e6. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Kaewlai, L. L. Avery, A. V. Asrani, and R. A. Novelline Multidetector CT of Blunt Thoracic Trauma1 RadioGraphics, October 1, 2008; 28(6): 1555 - 1570. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Sammer, E. Wang, C. C. Blackmore, T. R. Burdick, and W. Hollingworth Indeterminate CT Angiography in Blunt Thoracic Trauma: Is CT Angiography Enough? Am. J. Roentgenol., September 1, 2007; 189(3): 603 - 608. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Rubin, O. N. Pages, A. Poncet, and B. Baehrel Endovascular Treatment of an Acute Subdiaphragmatic Aortic Rupture Ann. Thorac. Surg., December 1, 2006; 82(6): 2276 - 2278. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. A. Bruckner, D. J. DiBardino, T. C. Cumbie, C. Trinh, S. H. Blackmon, R. G. Fisher, K. L. Mattox, and M. J. Wall Critical Evaluation of Chest Computed Tomography Scans for Blunt Descending Thoracic Aortic Injury Ann. Thorac. Surg., April 1, 2006; 81(4): 1339 - 1346. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 The effect of changing presentation and management on the outcome of blunt rupture of the thoracic aorta. J. Thorac. Cardiovasc. Surg., March 1, 2006; 131(3): 594 - 600.
|
|
|
|
 |
|
 R. A Lancey, A T. Pezzella, and D. A Phillips Factors Affecting Early Exsanguination and Death in Blunt Thoracic Aortic Trauma Asian Cardiovasc Thorac Ann, September 1, 2004; 12(3): 202 - 207. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C E O'Conor Diagnosing traumatic rupture of the thoracic aorta in the emergency department Emerg. Med. J., July 1, 2004; 21(4): 414 - 419. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Wong, M. B. Gotway, A. D. Sasson, and R. B. Jeffrey Periaortic Hematoma at Diaphragmatic Crura at Helical CT: Sign of Blunt Aortic Injury in Patients with Mediastinal Hematoma Radiology, April 1, 2004; 231(1): 185 - 189. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. R. Tai and K. D Boffard Thoracic trauma: principles of early management Trauma, April 1, 2003; 5(2): 123 - 136. [Abstract] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
