| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Vascular and Interventional Radiology |
1 From the Department of Radiology, Stanford University School of Medicine, 300 Pasteur Dr, Rm S-072B, Stanford, CA 94305-5105. Received July 27, 2000; revision requested September 8; revision received March 15, 2001; accepted April 3. Supported by NIH grant R01HL58915. Address correspondence to G.D.R. (e-mail: grubin@stanford.edu).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: Twenty-four patients with symptomatic lower extremity arterial occlusive or aneurysmal disease underwent imaging with four-channel multi–detector row CT from the supraceliac abdominal aorta through the feet. Transverse sections were acquired with a 2.5-mm nominal detector width and pitch of 6.0 (3.2-mm effective section thickness) following intravenous injection of 174–185 mL of iodinated contrast medium (300 mg iodine per milliliter). In each patient, attenuation measurements were recorded in 16 arterial and 16 venous locations. In 18 patients, two radiologists assessed the detectability and stenosis degree of 21 arterial segments per patient relative to these features at conventional angiography.
RESULTS: A mean scannning time of 66 seconds was required to cover a mean of 1,233 mm, resulting in a mean of 908 transverse reconstructions. All 504 arterial segments were depicted and analyzable. Mean arterial attenuation ranged from 253 HU in the midabdominal aorta to 357 HU in the popliteal artery and 253 HU in the dorsalis pedis or posterior tibial artery measured inferior to the tibiotalar joint. Maximum mean venous enhancement (99 HU) was observed in the saphenous vein at the ankle, with all other venous stations measuring less than 74 HU.
CONCLUSION: The arteries of lower extremity inflow and runoff can be reliably depicted with minimal venous enhancement by using multi–detector row CT.
Index terms: Arteries, CT, 92.12912. 92.12915, 92.12916, 92.12917 • Arteries, extremities, 92.72, 92.73 • Computed tomography (CT), angiography, 92.12916
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
Multi–detector row CT has had a substantial effect on CT angiography, offering shorter acquisition times, lower doses of contrast medium, and improved spatial resolution for assessing smaller arterial branches (7,8). The purpose of this study was to assess the patterns of lower extremity arterial inflow and runoff opacification with four-channel multi–detector row CT angiography in a cohort of patients with disease warranting imaging of the lower extremity arterial system. We further sought to assess the magnitude of lower extremity venous enhancement, which might complicate arterial analysis by increasing the complexity of the segmentation processes for extracting the arteries prior to 3D visualization and analysis.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
Acquisition Protocols
Four-channel multi–detector row CT scans were acquired with a CT scanner (Lightspeed Qxi; GE Medical Systems, Milwaukee, Wis) under a protocol approved by our institutional review board. Informed consents were obtained. With physician guidance (G.D.R., A.J.S.), patients were positioned supine on the CT table to allow transport of the entirety of their body inferior to the xyphoid process to pass through the CT gantry. A towel was wrapped around the patient’s knees and ankles, which were subsequently fixed together with adhesive tape. No pillows or padding was placed below the patient’s legs.
An anteroposterior scout view was first acquired to encompass the entire body below the xyphoid process. Subsequently, helical CT was performed without intravenous contrast medium from the apex of the diaphragm to the lesser femoral trochanters, with a 5-mm nominal section thickness, pitch of 6.0, 30-mm rotation table feed, and a gantry rotation period of 0.8 second. The x-ray tube potential was 120 kV and the current was 80 mA. The level of the celiac axis origin was identified.
A 20–22-gauge catheter was placed into a superficial vein within the antecubital fossa, forearm, or dorsum of the hand; 15 mL of low-osmolar iodinated contrast medium (iohexol [Omnipaque 300]; Nycomed Amersham, Princeton, NJ) was injected at a flow rate of 4 mL/sec. Serial transverse sections were obtained 1–2 cm above the celiac axis origin to measure the circulation time from the injection site to the aorta and thus determine the scanning delay. Images were acquired every 2 seconds after an initial 8-second delay for a total of 15 sections. The peak of a time-attenuation curve created from a region of interest placed into the aorta by the CT technologist was selected as the scanning delay for the ensuing CT angiography.
CT angiography was performed following a target injection of 180 mL of contrast medium at a flow rate of 3.6 mL/sec. The rate of the contrast medium injection was adjusted at the discretion of the physician monitoring the scanning prescription on the basis of the quality of intravenous access. The volume of contrast medium was subsequently adjusted to provide a bolus duration of approximately 50 seconds in all but two patients whose contrast volume was adjusted to provide a bolus duration that was equivalent to the scanning duration.
Helical CT was performed by using a 2.5-mm nominal section thickness, a pitch of 6.0, a table speed of 15 mm per rotation (18.75 mm/sec), and a 0.8-second gantry rotation period. X-ray tube voltage was 120 kV, and the current was 300 mA. Patients were requested to hold their breath during the first 20 seconds of the acquisition and were allowed to breathe quietly thereafter. Transverse sections were reconstructed by using a modified 180° linear interpolation algorithm that incorporated views on the basis of a balance between section profile broadening, helical artifact, and image noise (9). Sections were reconstructed at 1.6-mm intervals that corresponded to half of the effective section thickness, which was defined as the full width at half maximum of the section sensitivity profile, which was 3.2 mm (8). The reconstruction field of view was 30–32 cm. The entire examination took 15–30 minutes, depending on intravenous access.
All CT angiograms were processed in our 3D imaging laboratory into volume renderings, maximum intensity projections, and curved planar reformations by one of two 3D imaging technologists (L.J.L., M.C.S.) by using a workstation (Advantage Windows 3.1P; GE Medical Systems). Two curved planar reformations were created at 90° intervals about the longitudinal axis of the aorta through both common and external iliac arteries and the common femoral, superficial femoral, and popliteal arteries. Two curved planar reformations were also created through each of the anterior tibial, posterior tibial, peroneal, celiac, superior mesenteric, renal, and inferior mesenteric arteries. Data from the bones were removed by using a combination of region-growing Boolean operations and region-of-interest drawings (10). The opacity-transfer function for volume rendering and the thresholds for shaded-surface displays were determined to maximize vascular visualization while minimizing nonvascular visualization. Twelve volume renderings and shaded-surface displays were created every 30° about the craniocaudal axis, and six maximum intensity projections were created 30° about the craniocaudal axis at the frontal aspect between the right and left lateral views. Images were photographed with customized window width and level settings by the technologists (L.J.L., M.C.S.) who created the reformatted images to allow clear delineation of the enhanced lumen, mural calcium, mural atheroma and/or plaque, and extravascular tissues.
Conventional Angiography
On the basis of clinical necessity, conventional angiograms were obtained within 3 months of the CT angiograms in 18 patients. The mean interval was 27 days ± 29 (SD). In 17 patients, conventional angiography was performed with a digital subtraction technique (Angiostar; Siemens Medical Systems, Erlangen, Germany). A 5-F pigtail catheter was positioned into the upper abdominal aorta by means of a transfemoral approach for anteroposterior and lateral aortograms. The catheter was repositioned just above the aortic bifurcation, and anteroposterior and bilateral 45° oblique views were obtained. Imaging of the lower extremity runoff from the femoral arteries through the feet was performed sequentially, with separate contrast medium injections at each of the four to five additional levels required to image the entirety of the lower extremities.
The image acquisition protocol consisted of 25 anteroposterior and 25 lateral abdominal projections at three per second, 25 right and 25 left anterior oblique pelvic projections at two to three per second, 20 each in the thigh, knee, and calf at one to two per second, and 30 in the feet at one to two per second. In one patient with an infrarenal aortic occlusion, transaxillary aortography was performed in the juxta renal aorta. Imaging was performed with a screen-film technique and a stepping table to follow a single injection of contrast medium.
Data Collection and Analysis
A radiologist (A.J.S.) retrospectively recorded the scanning duration (in seconds), scanning coverage (in millimeters), number of transverse sections, and nominal section thickness (in millimeters) from annotations that were automatically attached to the CT images by the scanner at the time of acquisition. The dose of iodinated contrast medium (300 mg of iodine per milliliter of a solution), volume of contrast medium, flow rate, and injection duration were recorded at the time of image acquisition by the nurse or technologist who performed the injection and were collated with the other scanning data (by A.J.S.). Attenuation measurements were made with a workstation (Advantage Windows; GE Medical Systems) by a radiologist (A.J.S.) at 16 arterial, eight deep venous, and eight superficial venous locations (Table 1) by placing a circular region of interest within the center of the vessel of interest at a position where it was patent, not severely stenotic, and closest to the middle of the longitudinal extent of the vessel.
|
The mean, SD, and 95% CIs of the measurements were calculated for each anatomic region. Statistically significant differences were indicated by a P value of less than .05 and were identified for comparisons in which the 95% CIs of the measurements did not overlap. The differences between arterial and venous measurements were assessed at each anatomic location where the measurements were available by performing one-way analysis of variance and a Tukey-Kramer multiple-range test. Regression analysis was applied to the correlation between the degree of arterial attenuation and the time after scanning initiation by fitting a quadratic equation to all values, which were normalized as a percentage of maximum arterial attenuation per patient.
All CT images, including transverse reconstructions, volume renderings, shaded-surface displays, maximum intensity projections, and curved planar reformations were reviewed by two radiologists (G.D.R., A.J.S.) independently, and each of the 21 arterial segments (aorta and bilateral common iliac, external iliac, common femoral, superficial femoral, popliteal, anterior tibial, posterior tibial above the ankle, peroneal, dorsalis pedis, and posterior tibial below the ankle) was assessed as being normal (<50% stenotic), stenotic (50%–99% stenotic), or occluded. Discrepancies were settled by means of consensus. For 18 cases, one radiologist (G.D.R.) directly compared the CT results with those of conventional angiography and the corresponding official reports. Discrepancies between CT and conventional angiographic interpretations were settled by a consultation with the radiologist who performed conventional angiography.
Estimates of whole-body effective radiation doses were calculated for CT and conventional angiography as effective dose equivalents. Calculations were performed by using National Radiological Protection Board (Christchurch, New Zealand) Techniques NRPB-SR250, Normalized Organ Doses for X-ray Computed Tomography Calculated Using Monte Carlo Techniques and NRPB-SR262, and Normalized Organ Doses for Medical X-ray Examinations Calculated Using Monte Carlo Techniques. Effective whole-body doses were calculated for the abdomen and pelvis exposures only, because lower extremity exposure contributes minimally to the whole-body effective dose.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
|
|
|
|
|
|
|
|
|
Mean venous enhancement was significantly greater (P < .001) in patients with occlusive disease than in those with aneurysmal disease and measured 71.6 HU ± 6.6 and 52.5 HU ± 3.7 (mean ± CI), respectively. With the exception of the one patient with Buerger disease in whom the venous attenuation contralateral to the arterial disease was an average of 141 HU ± 37.1 greater, there was no significant difference in the situs of venous enhancement relative to the situs of asymmetric arterial occlusive disease (P = .86). The average venous attenuation difference contralateral versus ipsilateral to arterial disease was 0.13 HU.
To understand the potential challenge of separating arteries from veins by using intensity-based segmentation, we determined the difference between arterial and venous attenuation for each location. Because deep veins course adjacent to the arteries while the superficial veins are distant, we analyzed the superficial and deep veins separately. The mean, 95% CI, maximal value, and minimal value of arteriovenous attenuation differences at each location are plotted in Figure 9. The mean attenuation differences were 239 HU ± 105 and 237 HU ± 117 for the deep and superficial veins, respectively.
|
|
Of the two patients without known active inflammation, one with severe bilateral arterial occlusive disease had excellent arterial enhancement (>210 HU), with a mean unilaterally elevated venous enhancement below the knee of 100 HU ± 45 in the deep veins and 199 HU ± 103 in the superficial veins. The other patient had poor arterial enhancement (115 HU), with a maximal unilateral saphenous venous enhancement of 78 HU.
Radiation Exposure
For CT, the dose-length product and the CT dose index were 1578 mGy x cm and 12.97 mGy, respectively. For conventional angiography, fluoroscopic time ranged between 4 and 8 minutes per case, with radiation doses of 1,284, 5,310, 1,246, 759, 411, and 347 cGy x cm2 for the abdomen, pelvis, thighs, knees, calves, and feet, respectively. The calculated effective dose indexes were 0.93 cSv and 3.62 cSv for CT and conventional angiography, respectively. Thus, the radiation exposure with conventional angiography was 3.9 times greater than that with CT angiography.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
Because of limitations in the scanning speed with single–detector row CT, imaging of the entirety of the lower extremity inflow and runoff with sufficient spatial resolution to characterize occlusive disease has not been viable. In 1995, Lawrence et al (14) published a technical report on imaging of a portion of the lower extremity arterial system, from the inguinal ligament to the proximal calf, in six patients by using 1-second gantry rotation single–detector row CT. Two consecutive 60-second helical acquisitions with 5-mm collimation and a pitch of 1.0 (5-mm/sec table speed) were used to cover 60 cm over a minimum of 129 seconds. Fifty patients subsequently underwent CT 70 cm from the groin to the midcalf with 5-mm collimation and a pitch of 2.0. The later study reported sensitivities of 94%–100% and 67%–88% for arterial occlusion and 75%–99% stenosis, respectively. The specificities ranged from 98% to 100% and from 94% to 100% for occlusion and stenosis, respectively (15). With an effective section thickness of approximately half that used in the preceding study, the sensitivity and specificity of the technique described herein should be expected to further improve on these promising results.
Beregi et al (16) studied the popliteal arteries of 26 patients with suspected popliteal arterial disease. Because the coverage was limited to the popliteal artery, 29–45-second single–detector row CT angiography was performed with 3–5 mm collimation and a pitch of 1.0–1.2. Although the use of higher pitch values would have allowed a reduction in the collimation and overall section thickness to improve longitudinal spatial resolution (17), the results of this study demonstrated improved popliteal aneurysm detection with CT, when compared with the detection at conventional angiography (100% vs 61% sensitivity). They also demonstrated the benefit of CT for characterizing arterial luminal compression relative to that of conventional angiography due to direct visualization of the popliteal artery entrapment by the gastrocnemius muscle and adventitial cysts (16).
Four-channel multi–detector row CT, which to our knowledge was introduced clinically in 1998, provides three primary advantages over single–detector row CT angiography: shorter scanning durations associated with improved contrast medium efficiency, thinner sections of entire anatomic territories such as the renal or carotid arteries within a breath hold, and greater longitudinal coverage (7,8). The latter advantage has been applied to imaging the entirety of the lower extremity inflow and runoff for the purpose of assessing arterial occlusive and aneurysmal disease (18). In comparison with the single–detector row CT scanning reported by Lawrence et al (14), the four-channel multi–detector row CT method used in this study allowed us to image at a table speed that was 3.75 times faster while acquiring images with an effective section thickness that was 56% (3.2/5.7) thinner for an overall improvement in scanning efficiency (table speed divided by effective section thickness) of 6.7. As a result, we were able to image the entire lower extremity inflow and runoff (more than twice the distance) in approximately half the time with almost twice the spatial resolution.
For arterial segments identified with conventional angiography, we found 100% concordance with CT angiography. Furthermore, CT depicted 26 additional segments that were not analyzable with conventional angiography because of improved arterial opacification distal to the occluded segments. These data do not attest to the diagnostic accuracy of CT angiography because they were generated from a direct comparison between four-channel multi–detector row CT and conventional angiography. Independent assessments of four-channel multi–detector row CT and conventional angiography by multiple readers blinded to the results of the other examination will be required to assess the diagnostic accuracy of characterizing lower extremity arterial disease with four-channel multi–detector row CT and are critical to determining the ultimate clinical utility of this technique.
Although these initial clinical results are encouraging, there are several technical challenges associated with imaging this vascular territory that bear scrutiny: adequacy of arterial opacification at CT acquisitions as long as 75 seconds, differential bilateral arterial opacification in the presence of asymmetric disease, and venous opacification. This report focuses on these technical issues. The diagnostic accuracy of four-channel multi–detector row CT relative to that of the established standard, conventional angiography, will be reported separately.
Traditionally, CT angiography is performed during breath holding in vascular territories that are degraded by respiratory-induced motion. Considering that some breathing may be allowable below the pelvic rim, CT angiographic scanning durations rarely extend longer than 40–50 seconds (7,19,20). Furthermore, to achieve adequate arterial enhancement, injection rates of at least 3.5–4.0 mL are required. Because arterial opacification from the aorta to the feet typically requires at least 15 seconds when injecting a 10-second intraarterial contrast medium bolus (21), we decided to deviate from our standard iodinated contrast medium bolus policy of establishing a bolus duration that is equivalent to the scanning duration (20,22). Instead, we chose a bolus duration of 50 seconds, because it allowed a 180-mL contrast medium volume to be delivered at a flow rate of 3.6 mL/sec. Because 15 mL is used during the preliminary circulation time determination and 5 mL is required to prime the contrast medium supply line to the site of venous access, only 180 mL of contrast material in a filled 200-mL capacity power injector is available for CT angiography.
Using this injection protocol, we achieved greater than 150 HU opacification in 97% of arterial segments from the aorta to the feet. When we examined the degree of arterial enhancement relative to the time after initiating CT angiography, we found a central peak with substantial reductions from that peak at the beginning and end of scanning (Fig 3). This is similar to the "humped" shape enhancement profile reported for uniphasic contrast medium injections of 30-second duration during a 30-second scanning (23). Although multiphasic contrast medium injection strategies have been advocated for achieving greater uniformity of enhancement throughout the scanning, they have not been investigated for scanning durations greater than 30 seconds in vivo (23–25). Although a 50-second uniphasic injection was satisfactory, it is possible that improved arterial enhancement, particularly at the terminus of the scanning, might be achieved with fixed or tailored multiphasic injection protocols.
One challenge of performing conventional angiography in the setting of lower extremity occlusive disease is that substantial variations in the bilateral circulation time from the proximal to distal regions can occur (21). When severe asymmetries exist, the diagnostic quality of the initial angiographic examination can be limited and thus necessitate repeated selective injections of contrast medium. Because the volume of contrast medium required for performing CT angiography of the lower extremities precludes repeated injections, it is critical that both legs are sufficiently opacified despite substantial flow asymmetries. Although severe asymmetric disease was present in 10 of 17 patients with occlusive disease in this study, we found that all of the CT angiograms showed sufficient opacification bilaterally (Fig 5). Right-left arterial attenuation differences greater than 40 HU occurred with significantly greater frequency in the setting of asymmetric disease, with maximum right-left differences of 75 (293 - 218) HU above the calf and 96 (245 - 149) HU below the calf. These right-left attenuation differences did not hinder interpretation or adversely affect lesion detection when compared with these applications at conventional angiography (Fig 6).
Venous opacification has long been recognized as a limitation to arterial visualization on 3D views, particularly when examining the carotid and renal arteries, where the internal jugular and renal veins, respectively, tend to opacify quickly after arterial enhancement (1,26, 27). Although lower extremity venous opacification appears to be best visualized 3.5 minutes after contrast medium administration in the setting of suspected pulmonary embolism (28), we found that some lower extremity veins opacify substantially earlier. It is interesting to note that, as a group, the patients with occlusive disease had significantly (P < .001) greater venous enhancement at imaging than did the patients with aneurysmal disease. In fact, venous enhancement greater than 100 HU was isolated to the superficial veins and occurred in only one patient with aneurysmal disease, who had bilateral common iliac artery aneurysms with venous enhancement ipsilateral to a 37-mm common iliac artery aneurysm that was associated with extensive iliac and popliteal artery stenoses.
Although venous opacification does not present a substantial problem when tracking arterial disease on stacked transverse sections, it can substantially hinder arterial visualization on 3D views when arterial venous attenuation differences are low (<50 HU). Maximum intensity projection is currently our preferred method for 3D analysis of lower extremity CT angiographic data, because in our experience, it preserves small branch vessels and stenotic regions of larger vessels better than does volume rendering or surface displays. When venous enhancement approaches arterial enhancement, distinguishing the arteries from the veins becomes difficult on maximum intensity projections.
In general, arterial venous attenuation differences were substantially greater than 50 HU, but in patients with ipsilateral inflammation, venous opacification made analysis of maximum intensity projections difficult. In these situations, curved planar reformations were useful for displaying the course of the artery, free of overlying brightly enhanced veins, but these reformations can display only a single artery (Fig 7). Editing the data prior to 3D rendering may be a useful means of removing brightly enhancing veins when they occur.
There are two basic approaches to editing or segmenting structures such as bone or brightly enhancing veins or parenchyma away from the arteries: automatic region growing and manual region drawing (22,29). Because automatic region growing requires less operator intervention, it is both more efficient and more reproducible than manual techniques and thus is preferred. However, a prerequisite to its use is that structures must be unconnected. Because the superficial veins course within the subcutaneous tissues, they may be easier to remove from the data prior to rendering than are the deep veins, which directly parallel the arteries. We did not observe a specific tendency for venous enhancement to occur in superficial versus deep veins.
While we believe that our study results have demonstrated that CT angiography of lower extremity inflow and runoff is feasible and offers robust arterial enhancement, we believe that the greatest challenge facing its routine clinical adoption is the fact that the large amount of image data generated can easily overwhelm most CT reading and 3D rendering workstations. Our results correspond to the very first generation of multi–detector row CT technology. By using a pitch of 6.0 and a 0.8-second gantry rotation, an average acquisition of 66 seconds resulted in 908 transverse sections, with 50% overlap relative to the effective section thickness. Since performing this investigation, we have used a pitch of 8.0 and a 0.5-second gantry rotation to perform 80-second CT scanning from the celiac origin through the feet that resulted in a reduction of the effective section thickness from 3.20 to 1.25 mm, and this resulted in 2,131 transverse sections reconstructed in 0.6-mm increments (Fig 10). Although consistent arterial enhancement free of venous opacification can be challenging with 80-second scanning, future generations of multi–detector row CT scanners with the capacity to simultaneously acquire data from a larger number of detector rows are likely to substantially lessen scanning durations (30).
|
Although thinner sections typically require a greater x-ray tube current to overcome a reduction in detected photons, patients undergoing diagnostic imaging to assess the lower extremity arterial system are typically beyond their reproductive years and have little hematopoetic marrow within their pelvis and long bones (31). With the exception of a 24-year-old patient with Buerger disease, the average age of our study population was 68 years ± 7. Moreover, our data suggest that CT angiography provides an almost fourfold reduction in the radiation exposure when compared with conventional angiography. The relatively high dose of radiation in conventional angiography is predominately influenced by the large number of images acquired, with fluoroscopy accounting for less than 20% of the effective whole-body dose.
An alternative noninvasive imaging technique for assessing lower extremity arterial disease that requires no ionizing radiation or iodinated contrast medium and that has been in routine clinical use for several years is magnetic resonance (MR) angiography (13). There are many compelling advantages to the use of MR imaging, which in addition to the aforementioned benefits include less primary reconstructions owing to coronal acquisitions and no interference from high-signal-intensity nonarterial structures that are analogous to bone at CT angiography.
Nevertheless, there are several reasons why CT angiography might be a compelling alternative to MR angiography. Because of an association between lower extremity atherosclerosis and coronary arterial disease, implanted pacemakers and defibrillators are common in this patient population (32) and are contraindications to MR imaging. In addition, other metallic devices such as arterial stents or joint prostheses can result in substantial artifact that hinders arterial evaluation (13). The characterization of mural calcification is not possible with MR and may have therapeutic relevance (33,34). Finally, we have achieved voxel dimensions of 0.7 x 0.7 x 1.25–3.2 mm that allow us to image the entirety of the lower extremity arterial system in 60–85 seconds with four-channel multi–detector row CT. These voxel dimensions are five to 14 times smaller than those achieved with current MR angiographic techniques (35,36) and thus result in higher spatial resolution acquisition.
Our data suggest that multi–detector row CT can be used to assess the entirety of lower extremity arterial inflow and runoff, with volumetric data that demonstrate robust arterial enhancement and minimal venous opacification. Nevertheless, multi–detector row CT cannot be considered a viable clinical alternative to conventional or MR angiography until its diagnostic accuracy and effectiveness are determined.
|
FOOTNOTES |
|---|
Abbreviation: 3D = three dimensional
Author contributions: Guarantors of integrity of entire study, G.D.R., A.J.S.; study concepts and design, G.D.R.; literature research, G.D.R., A.J.S.; clinical studies, all authors; data acquisition and analysis, G.D.R., A.J.S.; statistical analysis, A.J.S.; manuscript preparation, definition of intellectual content, and editing, G.D.R.; manuscript review, G.D.R., A.J.S.; manuscript final version approval, G.D.R.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
Related Article
This article has been cited by other articles:
|
|
 |
|
  R. Raman, B. Raman, S. Napel, and G. D. Rubin Semiautomated Quantification of the Mass and Distribution of Vascular Calcification with Multidetector CT: Method and Evaluation Radiology, April 1, 2008; 247(1): 241 - 250. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. M. Kramer, M. J. Budoff, Z. A. Fayad, V. A. Ferrari, C. Goldman, J. R. Lesser, E. T. Martin, S. Rajagopalan, J. P. Reilly, G. P. Rodgers, et al. ACCF/AHA 2007 Clinical Competence Statement on vascular imaging with computed tomography and magnetic resonance Vascular Medicine, November 1, 2007; 12(4): 359 - 378. [PDF]
|
|
|
|
 |
|
 R. Schernthaner, D. Fleischmann, F. Lomoschitz, A. Stadler, J. Lammer, and C. Loewe Effect of MDCT Angiographic Findings on the Management of Intermittent Claudication Am. J. Roentgenol., November 1, 2007; 189(5): 1215 - 1222. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. H. Heijenbrok-Kal, M. C. J. M. Kock, and M. G. M. Hunink Lower Extremity Arterial Disease: Multidetector CT Angiography Meta-Analysis Radiology, November 1, 2007; 245(2): 433 - 439. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Burrill, Z. Dabbagh, F. Gollub, and M. Hamady Multidetector computed tomographic angiography of the cardiovascular system Postgrad. Med. J., November 1, 2007; 83(985): 698 - 704. [Abstract] [Full Text] [PDF]
|
|
= deep veins. 
= superficial veins.