| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 All authors: Department of Diagnostic Imaging, University of Maryland Medical Center and R. Adams Cowley Shock Trauma Center, 22 S Greene St., Baltimore, MD 21201.
Received April 6, 2007;
accepted after revision September 9, 2007.
Address correspondence to C. W. Sliker
(csliker{at}umm.edu).
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. Retrospective review of radiology reports and prospective clinical observation identified 108 blunt trauma patients examined with either whole-body MDCT or neck MDCTA followed by angiography over a 23-month period. From this group, results from the retrospective interpretations of 77 whole-body MDCT and 48 neck MDCTA examinations were compared with the results extracted from angiography reports to estimate the accuracy of each protocol for detecting blunt cerebrovascular injuries. Fisher's exact test was used to determine any significant difference in the results of those patients scanned with both protocols.
RESULTS. Angiography confirmed blunt cerebrovascular injury in 83 patients, with 25 (30%) showing multiple sites of injury. Most injuries were detected in cervical arterial segments. The respective sensitivities of whole-body MDCT and neck MDCTA were 69% (36/52) and 64% (16/25) for cervical internal carotid artery injuries, and specificities were 82% (58/71) and 94% (49/52). Respective sensitivities for cervical vertebral artery injuries were 74% (17/23) and 68% (13/19), and specificities were 91% (60/66) and 100% (40/40). In 17 patients scanned with both protocols, the results were not significantly different (carotid arteries, p = 1.00; vertebral arteries, p = 0.68).
CONCLUSION. Whole-body 16-MDCT and neck MDCTA can be used to diagnose blunt cerebrovascular injuries with comparable accuracy. Both show high specificities for cervical arterial injury. The sensitivity of whole-body 16-MDCT is sufficiently high to serve as an initial screening examination for blunt cerebrovascular injuries.
Keywords: blunt trauma • cerebrovascular injury • MDCT • MDCT angiography • whole-body MDCT
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Angiography, the current diagnostic standard for cerebrovascular injury, is an invasive, expensive, and labor-intensive procedure, so it is not an optimal means to screen for blunt cerebrovascular injury [9, 11]. Because CT is a firmly established and widely available tool for evaluating acute trauma patients, CT angiography is an attractive alternative to angiography in evaluating patients for blunt cerebrovascular injury. With its high spatial and temporal resolution, MDCT is well suited for diagnosing and characterizing vascular injuries and the other craniofacial, spinal, thoracic, and abdominopelvic injuries that typically are seen in multitrauma patients. Given its potential advantages, neck MDCT angiography (MDCTA) has been used routinely to evaluate for blunt cerebrovascular injury since the installation of a 16-MDCT scanner at our level 1 trauma center in May 2003.
In addition to the growing number of vascular indications for MDCTA, current 16-MDCT (and higher) scanners allow contiguous scanning from the head through the pelvis with one bolus of IV contrast material. In January 2004, a whole-body MDCT protocol, similar to those reported elsewhere in the medical literature [13–15], was adopted as a routine method for scanning blunt multitrauma patients at our institution. In addition to evaluating those areas of primary clinical interest, including the cervical spine, aorta, and abdominopelvic organs, the presence of intravascular contrast material while scanning through the cervical spine provided a "free" examination of the carotid arteries (CAs) and vertebral arteries. Coinciding with the adoption of that protocol, we noted an increase in the frequency with which blunt cerebrovascular injuries were diagnosed at our institution, suggesting that routine whole-body MDCT identified clinically silent blunt cerebrovascular injuries that may not otherwise have been noted while the patient was still asymptomatic.
Although a number of recent studies discuss the accuracy of targeted neck MDCTA for diagnosing blunt cerebrovascular injury [6, 16–19], to our knowledge, no such studies address the accuracy of whole-body MDCT for detecting these injuries. If whole-body MDCT can be used to accurately detect blunt cerebrovascular injury, it could prove to be a rapid means to screen large numbers of patients, thereby identifying blunt cerebrovascular injury in high-risk patients with a minimum of diagnostic testing while also identifying many "risk factor–negative" patients harboring blunt cerebrovascular injury. In so doing, early treatment of blunt cerebrovascular injury would be facilitated while, at the same time, limiting additional diagnostic imaging.
The goal of this study was to determine whether whole-body 16-MDCT can be used to diagnose blunt cerebrovascular injury with accuracy comparable to targeted neck 16-MDCTA, with digital subtraction angiography serving as the reference standard.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Subjects were identified either retrospectively or in a prospective observational fashion. The radiology reporting system (IDX-RAD version 9.92, IDX) was retrospectively reviewed by one experienced trauma radiologist (5 years of experience) to identify 59 blunt trauma patients whose CAs and vertebral arteries were evaluated with neck 16-MDCTA or whole-body MDCT between May 2003 and November 2004. From December 2004 to March 2006, through clinical observation, 49 patients were prospectively enrolled in the study. The total study population thus included 108 patients (71 males and 37 females; mean age, 39 years; age range, 14–76 years).
The mechanisms of injury included the following: motor vehicle collision (n = 89), motorcycle crash (n = 17), pedestrian struck by motor vehicle (n = 7), fall from height (n = 3), and bicyclist struck by motor vehicle (n = 2). The following mechanisms of injury were each encountered once: airplane crash, assault, crushed by fallen tree, explosion, hanging, industrial accident, and wrestling injury.
This study was compliant with the requirements of HIPAA and was approved by our institutional review board (IRB). The IRB granted a waiver of consent for those enrolled retrospectively. For those identified prospectively, written informed consent was required and obtained for the first nine patients, and then the IRB granted a waiver of consent for the subsequent 40 patients.
Clinical Practice
During the study period, all imaging examinations for potential blunt
cerebrovascular injury were performed at the discretion of the attending
traumatologist, frequently in consultation with a trauma radiologist or
neuroradiologist. For each patient, the decision to evaluate for blunt
cerebrovascular injury and the manner in which the workup progressed reflected
accepted imaging practices and standard patient care at our institution. In
August 2004, internal institutional guidelines based on the medical literature
[9,
11,
20–24]
were disseminated in our center in an effort to facilitate identification of
patients harboring blunt cerebrovascular injury, with the following proposed
as either risk factors for or signs of blunt cerebrovascular injury: cervical
spine fracture with foramen transversarium involvement, subluxation, rotation,
or distraction mechanism; basilar skull fracture crossing the carotid canal or
cavernous sinus; severe facial fractures (LeFort I, II, or III;
nasal–orbital–ethmoid complex; facial smash); carotid or vertebral
perivascular hematoma identified by MDCT; Horner syndrome; Glasgow Coma Scale
score 
6 at 24 hours after admission; neurologic examination incongruent
with brain imaging; stroke or transient ischemic attack; or hanging attempt
with cervical hematomas or cervical spine fractures. At no time was a
diagnostic imaging examination performed strictly for study purposes.
MDCT—Scans were acquired using one of three 16-MDCT scanners (MX8000 IDT, Brilliance 16 Power, or Brilliance Big Bore, Philips Medical Systems). Two different scanning protocols (Table 1) were in routine clinical use during the study period. The neck MDCTA protocol, limited to the region between the aortic arch and circle of Willis, was used when clinical concern was limited to the craniocervical arterial system. The whole-body MDCT protocol, which was used to reconstruct diagnostic images of the cervical spine and neck arteries, was routinely used to scan multitrauma patients in whom both cervical spine and chest CT examinations were requested. IV contrast material (iohexol, 300 mg I/mL [Omnipaque 300, GE Healthcare]) was administered with a power injector (Envision, Medrad) through peripheral venous access without a subsequent saline chaser.
|
The final images were archived to a PACS (IMPax 5.2, AGFA). At the discretion of the interpreting radiologist, independent work stations (AquariusNET Viewer and Aquarius Workstation, TeraRecon or Brilliance Workspace, Philips Medical Systems) were used to review thin-section data and perform additional post processing that included off-axis reconstructed and curved multi-planar reconstructed, volume-rendered, and maximum-intensity-projection images. All supple mental radiologist-directed postprocessed images were archived to the PACS.
Cervicocerebral angiography—Digital subtrac tion angiography examinations of the CAs and vertebral arteries were performed either by or under the direct supervision of one of six fellowship-trained neuroradiologists, each of whom has a Certificate of Added Qualification. Because examinations were performed as part of standard patient care, angio graphers were aware of the MDCT results when performing cervico-cerebral angiography.
The specific techniques used were at the discretion of each neuroradiologist based on his or her preferences and the patient's clinical condition and anatomy, although standard angiographic tech niques included arterial access via a common femoral artery, use of a 5- or 6-French sheath to maintain arterial access, selection of the artery to be studied with a 5-French catheter (typically a vertebral curved catheter), and imaging performed in at least two orthogonal planes (typically antero posterior and lateral). Complete four-vessel angio grams were not universally obtained, and the vessels and vascular segments studied were at the discretion of the attending neuroradiologist based on each patient's clinical status at the time of angiography.
In standard practice, studies were initially reviewed on the angiography console with static images photographed to hard copy for final review. Network incompatibilities prevented routine archiving of all diagnostic images from the angiographic consoles to the PACS.
Data Collection and Image Analysis
MDCT—Among the 108 study patients, 125 MDCT scans were
eligible for review (77 whole-body MDCT, 48 MDCTA). Because the use of MDCT to
diagnose blunt cerebrovascular injury did not become routine at our
institution until installation of a 16-MDCT scanner in May 2003, analysis of
MDCT results obtained from reports dated early in the study period may have
arti ficially lowered estimations of MDCT accuracy. To account for any
learning curve that may have existed during the course of clinical practice, a
retrospective review of all MDCT images, in cluding advanced postprocessed
images archived to the clinical PACS, was conducted by two American Board of
Radiology–certified radiolo gists with 21 and 16 years of trauma
radiology experience. In contrast to clinical practice, thin-section data were
not available for review, and additional postprocessing could not be performed
on independent work stations. Initial reviews were conducted independently,
with discrepancies reconciled through consensus review. The final consensus
interpretations were used for statistical analyses. MDCT scans were reviewed
between 30 days and 23 months after the initial acquisition.
During the retrospective review, the radiologists were aware that the patient group included those with and without blunt cerebrovascular injury, as determined by angiography, but they were unaware of the clinical indication for each examination other than a history of blunt trauma. The following criteria [8, 25–28] were used to diagnose injuries: intimal irregularity, mural thickening as a result of intramural hematoma, abrupt lumen caliber change, raised intimal flap, intraluminal thrombus, pseudo-aneurysm, occlusion, active extravasation, and arteriovenous fistula with early venous filling. A diagnosis of injury or of no injury was rendered for each of the following arterial segments: common CA (CCA), internal CA (ICA) cervical segment, ICA intracranial segment (petrous ICA to supraclinoid ICA), vertebral artery cervical segment, and vertebral artery intradural segment. If a segment could not be satisfactorily evaluated (as a result of artifact, poor opacification, and so on), evaluation of that segment was considered nondiagnostic.
Review of the MDCT reports was conducted by a third radiologist to identify nonvascular injuries visible on the MDCT examinations that might have been considered risk factors for injury based on clinical practice at the time, thereby potentially heightening the radiologists' suspicions for blunt cerebrovascular injury during retrospective review. In addition, the mechanisms of injury and clinical indications for each MDCT were documented to gauge the pretest probability of blunt cerebrovascular injury.
Cervicocerebral angiography—One of the authors reviewed angiography reports to determine whether each arterial segment was injured. The number and site of injuries were recorded for each patient diagnosed with blunt cerebrovascular injury. Injuries were not described with standard ized terminology and so could not be recorded by the grading system in current widespread use [25, 29]. Many folders of those studies documented with hard copy were incomplete, so retrospective review of angiographic studies could not be performed.
Data Management
Data were managed using Microsoft Excel 2003 software.
Statistical Analysis
To estimate accuracy of each MDCT scanning technique, the sensitivity,
specificity, positive predictive value, and negative predictive value (with
95% CIs) for whole-body MDCT and neck MDCTA in diagnosing blunt
cerebrovascular injury and blunt vertebral artery injury were determined from
the consensus MDCT interpretations, with results from angiography reports used
as the reference standard. In addition, the positive likelihood, negative
likelihood, and diagnostic odds ratios were determined. Arterial segments
considered nondiagnostic on MDCT examination were excluded from calculations.
Proportion estimates, likelihood ratios, and diagnostic odds ratios were
calculated with a Web-based statistics calculator
[30] and 2 x 2
contingency tables. Another Web-based calculator
[31] was used to determine 95%
CIs using the exact method. Because complete four-vessel cranio cervical
angiograms were not uniformly obtained, proportion estimates, likelihood
ratios, diagnostic odds ratios, and 95% CIs could not be determined for the
complete CA or vertebral artery circulations. Rather, proportion estimates,
likelihood ratios, and diagnostic odds ratios were calculated for each
arterial segment (CCA, ICA cervical segment, and so on). For similar reasons,
proportion estimates and ratios could not be determined on a per-patient
basis.
For the group examined with both the whole-body MDCT and neck MDCTA protocols, 2 x 2 contingency table analyses and the Fisher's exact test [32] were used to determine any statistically significant differences in the results between the two groups when evaluating for blunt CA injury and blunt vertebral artery injury. Two-tailed p values < 0.05 were considered statistically significant.
|
|
|
|
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
|
|
|
At consensus review, MDCT visualization of five CA and six vertebral artery segments was considered diagnostically inadequate. In two patients scanned with neck MDCTA, both CCA segments (n = 4) and both vertebral artery cervical segments (n = 4) were inadequately visualized. In the third patient, who was scanned with the whole-body MDCT protocol, the right ICA intracranial segment (n = 1) was inadequately visualized, whereas in the fourth patient, also scanned with whole-body MDCT, both vertebral artery cervical segments (n = 2) were insufficiently shown. In all cases, segmental visualization was rendered nondiagnostic by beam-hardening artifact.
Table 2 lists the indications for both whole-body MDCT and neck MDCTA examinations.
|
Table 3 lists the imaging results, proportion estimates, likelihood ratios, and diagnostic odds ratios of MDCT for diagnosing blunt CA injury and blunt vertebral artery injury. In brief, with each scanning protocol, MDCT most commonly depicted blunt cerebrovascular injury in the cervical segments of both the ICAs and vertebral arteries. In the respective cervical segments, each technique showed low sensitivity for both blunt CA injury (whole-body MDCT = 69%, MDCTA = 64%) and blunt vertebral artery injury (whole-body MDCT = 74%, MDCTA = 68%). Specificities for cervical segment blunt CA injury (whole-body MDCT = 82%, MDCTA = 94%) and blunt vertebral artery injury (whole-body MDCT = 91%, MDCTA = 100%) were high.
|
For the 17 patients scanned with both the whole-body MDCT and neck MDCTA protocols, the protocols were not statistically different in evaluating either CAs (p = 1.00) or vertebral arteries (p = 0.68) (Fig. 3A, 3B, 3C, 3D, 3E, 3F and Table 4).
|
|
|
|
|
|
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Clinically significant blunt cerebrovascular injury can be asymptomatic for up to 240 hours [33], and screening of at-risk patients during this period may identify up to 79% of blunt CA injuries and 100% of blunt vertebral artery injuries [11]. Proposed therapeutic options include anticoagulation, antiplatelet therapy, surgery, and endovascular stenting [2, 5, 25, 33]. In addition, selected blunt vertebral artery injuries can be treated with endovascular coil occlusion [34, 35]. The best therapeutic approach remains undecided, but studies suggest that both anticoagulation and antiplatelet therapy decrease stroke rates and limit deficit in those with symptomatic injuries [3–5, 8, 11, 33].
To facilitate more efficient screening and early diagnosis, several authors have reported risk factors for both blunt CA injury and blunt vertebral artery injury [3–5, 8–10]. Some studies have shown that blunt cerebrovascular injury–related mortality due to stroke can be significantly reduced by using predefined criteria to identify candidates for screening [6, 7]. In one study [6], after a CT angiography–based blunt cerebrovascular injury screening protocol was initiated, not only did the prescreening blunt cerebrovascular injury incidence at the authors' institution rise from 0.17% to 1.4%, but also the blunt cerebrovascular injury–specific mortality rate fell significantly from 38% to 0%.
Four-vessel cervicocerebral angiography is considered the reference standard for diagnosing blunt cerebrovascular injury [9, 12]. This invasive test carries an inherent risk of complications, including stroke, but these complications are infrequent. For example, in one representative group of 727 blunt trauma patients angiographically screened for blunt cerebrovascular injury [7], only one patient (0.1%) had an angiography-related stroke and two others (0.3%) developed catheter entry site hematomas that were managed conservatively. When targeted to the high-risk subset of patients, angiography performed by qualified operators seems to be an acceptable means of screening for blunt cerebrovascular injury. However, because it is expensive, labor-intensive, and potentially difficult to perform in a timely manner where equipment or qualified personnel are limited [20], angiography is not an optimal choice as a screening modality.
MR angiography (MRA) and duplex sonography have not been proven as suitable alternatives to angiography. The value of MRA is compromised by the nature of the MRI suite, which can be a challenging environment in which to monitor the acutely injured patient. Moreover, the sensitivities of MRA for both blunt CA injury (50–75%) [11, 12] and blunt vertebral artery injury (47%) [12] are low, and its specificities are variable (67–100% and 97%, respectively) [11, 12]. Although portable, Doppler sonography is operator-dependent, and evaluation of the intracranial ICAs and vertebral arteries can be restricted by the skull and vertebrae, respectively. The sensitivity of sonography for blunt CA injury within the neck is only 86% [36], and the authors of a recent study reported an overall sensitivity for blunt cerebrovascular injury of 38.5% [1].
Contemporary MDCTA is well suited for vascular examinations, and its ample availability and low number of anatomic limitations make it an attractive diagnostic alternative to angiography. Biffl et al. [16] and Berne et al. [18] reported on 331 and 435 patients, respectively, who were screened for blunt cerebrovascular injury with 16-MDCTA. In these studies, none of those with negative MDCTA results manifested delayed complications of blunt cerebrovascular injury on clinical follow-up. However, because a combined total of only 42 patients [16, 18] were diagnosed with blunt cerebrovascular injury and neither study followed negative examinations with angiography, the comparative accuracies of MDCTA and angiography remain unclear based on the results of these examinations.
In the first study to directly compare 16-MDCTA with angiography, Eastman et al. [17] prospectively used both MDCTA and angiography to study 146 patients with suspected blunt cerebrovascular injury. Angiography revealed 46 blunt cerebrovascular injuries in 43 patients (blunt CA injury, n = 20; blunt vertebral artery injury, n = 26), thereby giving MDCTA an overall sensitivity of 97.7% and specificity of 100% for diagnosing blunt cerebrovascular injury and sensitivities of 100% and 96.1% for detecting blunt CA injury and blunt vertebral artery injury, respectively. Despite the promising accuracy reported for MDCTA, this study suffers from limitations that temper the value of its conclusions. First, as in many other studies of 16-MDCTA, the numbers of both injured patients (n = 43) and injuries (blunt CA injury, n = 20; blunt vertebral artery injury, n = 26) were quite small, thereby limiting the power of the results. At most, 6.5% of their patients (three of 46) exhibited multivessel injury, which is discordant with the results of other studies that have reported multiple injured arteries in 18–32% of blunt cerebrovascular injury patients [9, 37, 38]. The source of this discordance is unclear, although statistical variation, given the relatively small overall number of identified injuries, is a likely explanation. Limitations notwithstanding, the data reported suggest that 16-MDCTA can be used to identify clinically significant injuries with a high degree of accuracy and can play a valuable role in the diagnosis of blunt cerebrovascular injury.
Prior studies that investigated the use of 16-MDCT to diagnose blunt cerebrovascular injury focused on scanning protocols similar to the neck MDCTA protocol reported in our study. To our knowledge, none has discussed the accuracy of a whole-body MDCT protocol for diagnosing blunt cerebrovascular injury. In our study, complete four-vessel angiography was not uniformly performed for each patient, thereby limiting measures of MDCT accuracy. Also reflecting clinical practice and affecting accuracy estimates, normal whole-body MDCT and neck MDCTA examinations were not universally confirmed with angiography. Nonetheless, general estimations of the accuracy of whole-body MDCT can be made, with the greatest credence given to values obtained for the areas of greatest injury frequency, the cervical segments of both ICAs and vertebral arteries.
With these limitations considered, we found that the sensitivities of whole-body MDCT for both blunt CA injury (69%) and blunt vertebral artery injury (74%) in the neck were low, whereas the specificities of whole-body MDCT for blunt CA injury (82%) and blunt vertebral artery injury (91%) were high. The specificities of neck MDCTA for blunt CA injury (94%) and blunt vertebral artery injury (100%) were high, but the sensitivities were disappointingly low (blunt CA injury = 64%, blunt vertebral artery injury = 68%). Because the whole-body MDCT protocol necessitated a higher pitch and abduction of the patient's arms (and therefore more beam-hardening artifacts and less z axis resolution than neck MDCTA) and a reconstruction filter (i.e., C or sharp soft-tissue filter) used to evaluate both bone and soft tissues (rather than a dedicated soft-tissue filter), the estimated accuracy for neck MDCTA would be expected to be higher than that for whole-body MDCT. However, except specificities for blunt CA injury, the estimated measures of accuracy for whole-body MDCT and neck MDCTA were comparable. Moreover, when both whole-body MDCT and neck MDCTA were used to evaluate identical arterial segments, the 2 x 2 contingency table analyses indicated no statistically significant difference in scan results (blunt CA injury, p = 1.00; blunt vertebral artery injury, p = 0.68).
Although a larger number of patients studied with both protocols is needed to definitively document diagnostic equivalence of the whole-body MDCT and neck MDCTA protocols, our data suggest that the two protocols render comparable results when used to evaluate the same arterial segment.
Although direct comparison of results is difficult, the discrepancies between the estimated sensitivities for whole-body MDCT and neck MDCTA reported here and those reported for neck MDCTA by Eastman et al. [17] are striking. Limitations of this study most likely account for the lower estimated sensitivities of MDCT for both blunt CA injury and blunt vertebral artery injury. As we already mentioned, true measures of accuracy for MDCT could not be determined because normal findings on whole-body MDCT and neck MDCTA were not universally confirmed with angiography. This shortcoming impairs measurement of true accuracy for injury on a segment-by-segment basis. Also, because accuracy could not be measured on a per-patient basis (Fig. 2A, 2B, 2C, 2D), a measure of MDCT's value as a means to screen for blunt cerebrovascular injury and a comparison with Eastman and associates' results are further complicated.
Because diagnosis and management of blunt cerebrovascular injury were not protocol-based, some patients with obvious arterial injuries diagnosed on MDCT during the study period were treated accordingly and did not undergo angiography. Therefore, the study population was likely biased toward subtle injuries that may not have been easily detected on MDCT, with a consequent reduction in its sensitivity. Overcoming these shortcomings would be difficult because routine angiography is impractical given the high clinical volume of our center and, in the case of most of the normal whole-body MDCT scans, is unethical given the relative low incidence of blunt cerebrovascular injury in the general blunt trauma population.
Despite these limitations rooted in study design and the variability of clinical practice, other facets of the current study bolster the validity of the results. First, reflecting clinical practice, many risk factors for blunt cerebrovascular injury were apparent on MDCT scans (skull base fractures, cervical spine fractures, and so on), thereby heightening the radiologists' index of suspicion for blunt cerebrovascular injury during consensus review. Second, an awareness of institutional referral patterns (e.g., neck MDCTA requested for clinically suspected blunt cerebrovascular injury, clarification of whole-body MDCT results) should have biased the reviewers toward interpreting questionable abnormalities on MDCTA as injuries. However, as the data show, scanning with MDCTA and whole-body MDCT yielded similar results. Also, although consensus review by two experienced trauma radiologists who practice at a level 1 trauma center where a large number of blunt cerebrovascular injuries have been diagnosed (blunt CA injury, n = 113; blunt vertebral artery injury, n = 59 during the study period) should have resulted in high sensitivities for blunt cerebrovascular injury, this was not the case.
The current study's design precludes definitive comparisons with other studies regarding the accuracy of both whole-body 16-MDCT and neck 16-MDCTA, but the data imply that the true accuracy of MDCT may not be as high as previously reported. Although the recent studies from Biffl et al. [16] and Berne et al. [37] suggest that MDCT identifies clinically relevant blunt cerebrovascular injury with clinically acceptable accuracy, thus supporting the findings of Eastman et al. [17], without a large prospective study (a difficult and ethically challenging undertaking), the true accuracy of 16-MDCT and the significance of those injuries it misses remain uncertain.
Questions regarding the true sensitivity of MDCT notwithstanding, the data indicate that, when consistent diagnostic criteria are used, whole-body 16-MDCT and neck MDCTA protocols can yield scans with comparable diagnostic accuracy to assess for blunt cerebrovascular injury. The data point to an acceptably high specificity of 16-MDCT for blunt cerebrovascular injury. Consequently, management decisions can be reliably based on positive MDCT results. Therefore, dedicated imaging of the neck can be reserved for those patients with clinically suspected blunt cerebrovascular injury who did not undergo whole-body MDCT or whose whole-body MDCT results were nondiagnostic, equivocal, or normal. In addition, the routine use of whole-body MDCT would facilitate diagnosis and treatment of blunt cerebrovascular injury in patients (up to 20%) without typical risk factors for injuries [9, 10].
In conclusion, whole-body 16-MDCT and neck MDCTA can be used to diagnose blunt cerebrovascular injury with comparable accuracy. Given the high specificity of both whole-body MDCT and neck MDCTA for detecting blunt cerebrovascular injury, treatment can be confidently based on positive results obtained with either scanning technique. Despite data that suggest the sensitivities of both whole-body MDCT and neck MDCTA for blunt cerebrovascular injury are lower than the sensitivities of neck 16-MDCTA reported in prior studies, the sensitivity of whole-body 16-MDCT is still sufficient for it to serve as an acceptable initial means to screen multitrauma patients for blunt cerebrovascular injuries.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  J. M. Provenzale and B. Sarikaya Comparison of Test Performance Characteristics of MRI, MR Angiography, and CT Angiography in the Diagnosis of Carotid and Vertebral Artery Dissection: A Review of the Medical Literature Am. J. Roentgenol., October 1, 2009; 193(4): 1167 - 1174. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. W. Sliker Blunt Cerebrovascular Injuries: Imaging with Multidetector CT Angiography1 RadioGraphics, October 1, 2008; 28(6): 1689 - 1708. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
