| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Vascular and Interventional Radiology |
1 From the Department of Oncology, Transplants and Advanced Technologies in Medicine, Division of Diagnostic and Interventional Radiology (V.N., I.B., P.P., R.C., C.V., C.B.) and Department of Vascular Surgery (S.G.S., M.F.), University of Pisa, Via Roma 67, 56126 Pisa, Italy. Received November 3, 2003; revision requested January 27, 2004; final revision received March 24; accepted May 4. Address correspondence to I.B. (e-mail: irenebargellini@hotmail.com).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: From November 1998 to February 2003, 112 patients underwent EVAR. At follow-up, duplex US and biphasic multi–detector row computed tomographic (CT) angiography were performed. In 10 patients (group A), evident aneurysm enlargement was observed, with no evidence of complications, at both CT angiography and duplex US. Group A patients, 10 men (mean age, 69.6 years ± 10 [standard deviation]), underwent US after intravenous bolus injection of a second-generation contrast agent, with continuous low–mechanical index (0.01–0.04) real-time tissue harmonic imaging. Group B patients, 10 men (mean age, 71.3 years ± 8.2) with aneurysm shrinkage and no evidence of complications, and group C patients, 10 men (mean age, 73.2 years ± 6) with CT angiographic evidence of endoleak, underwent contrast-enhanced US. Digital subtraction angiography (DSA) was performed in groups A and C. Endoleak detection and characterization were assessed with imaging modalities used in groups A–C; at contrast-enhanced US, time of detection of endoleak, persistence of sac enhancement, and morphology of enhancement were evaluated.
RESULTS: In group A, contrast-enhanced US depicted one type I, six type II, one type III, and two undefined endoleaks that were not detected at CT angiography. All leakages were characterized by slow and delayed echo enhancement detected at longer than 150 seconds after contrast agent administration. DSA results confirmed findings in all patients; percutaneous treatment was performed. In group B, contrast-enhanced US did not show echo enhancement; in group C, results with this modality confirmed findings at CT angiography and DSA.
CONCLUSION: Contrast-enhanced US depicts endoleaks after EVAR, particularly when depiction fails with other imaging modalities.
© RSNA, 2004
Index terms: Aneurysm, aortic, 94.73 • Aneurysm, CT, 94.12916 • Digital subtraction angiography, comparative studies • Ultrasound (US), Doppler studies, 94.12984, 94.12988 • Ultrasound (US), harmonic study
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
Computed tomography (CT) represents the modality of choice in the follow-up of endografts (3), although several authors have pointed out the usefulness of other imaging modalities, such as magnetic resonance (MR) angiography and duplex ultrasonography (US), to depict and characterize the endoleaks with a sometimes higher accuracy than that achieved with CT (4–8). In our institution, patients undergoing EVAR are evaluated with a strict protocol for duplex US and CT angiographic follow-up.
Nevertheless, in articles that included series of patients who underwent EVAR, several researchers (9,10) reported cases of aneurysm enlargement with apparently no cause that is detectable with any imaging modality. The term endotension has therefore been introduced to address this phenomenon.
In recently published reports (11,12), investigators described the use of US with contrast agents in the detection of endoleaks after EVAR and reported good sensitivity when compared with CT angiography and conventional duplex US.
Thus, the purpose of our study was to evaluate contrast material–enhanced US for the depiction of endoleaks after EVAR in patients with aneurysm enlargement and no evidence of endoleak.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
Aneurysm diameter ranged from 40 to 80 mm (mean, 52 mm ± 10). Several types of stent-grafts were used: a full exoskeleton of diamond-shaped nitinol wire fully covered with polyester (AneuRx; Medtronic, Minneapolis, Minn), 48; a nitinol wire framework covered with polyester with a top uncovered portion provided by hooks (Vanguard; Min Tec, Freeport, Bahamas), two; a nitinol wire covered with polyester with a top uncovered stent without hooks (Talent; World Medical Manufacturing, Sunrise, Fla), 15; a unibody fully supported device (Endologix; Endologix, Irvine, Calif), 10; a combination of self- and balloon-expandable devices made with a metal frame of elgiroy and stainless steel (Lifepath; Baxter, Deerfield, Ill), one; a spiral frame of nitinol covered in and out with expanded polytetrafluoroethylene (Excluder; Gore, Flagstaff, Ariz), 17; and a dacron polyester graft with a stainless steel exo- and endoskeleton and top barbwires (Zenith; Cook, Bloomington, Ind), 19. Seven stent-grafts were straight, whereas 105 were bifurcated endografts.
The procedure was technically successful in 108 patients, and immediate surgical conversion was required in four patients. After the procedure, all patients underwent CT angiography (HiSpeed or LightSpeed Plus; GE Medical Systems, Milwaukee, Wis) at 7 days, 6 and 12 months, and annually thereafter. They also underwent color-coded duplex US with one unit (AU5; Esaote Biomedica, Genoa, Italy) at 1 and 3 months, and thereafter they underwent color-coded duplex US with CT angiography. All examinations, as well as stent-graft implantation, were performed after informed written consent was obtained.
Endoleaks were classified as directly related to the stent-graft (type I, III, or IV) or as secondary to patent aortic branches (type II), such as the inferior mesenteric artery or the lumbar arteries. At follow-up, the previously mentioned imaging modalities depicted endoleaks as follows: six cases of type I, 25 cases of type II, and two cases of type III.
In 10 patients, an evident increase of the aneurysm maximum diameter (>5 mm, compared with the diameter calculated at a previous examination) was demonstrated with CT angiography at follow-up without any evidence of stent-graft displacement or disruption or perigraft flow. Therefore, in these 10 male patients (mean age, 69.6 years ± 10), contrast-enhanced US was performed to identify a missed endoleak (group A).
Two groups (groups B and C) of patients were used as control groups. Contrast-enhanced US was also performed in group B patients, which included 10 male patients (mean age, 71.3 years ± 8.2) with no evidence of endoleak and a decrease of the aneurysm diameter at follow-up, and in group C patients, which included 10 male patients (mean age, 73.2 years ± 6) with a type II endoleak detected at both CT angiography and duplex US. Group B and C patients were selected at random from those who met the criteria for each group.
Group A and C patients underwent digital subtraction angiography (DSA) (Multistar; Siemens, Erlangen, Germany), followed by endoluminal treatment when the endoleak was confirmed.
CT Angiographic Protocol, Postprocessing, and Evaluation
CT angiography was performed from the celiac artery to the common femoral arteries both before and after intravenous administration of contrast material (Visipaque 320; Nycomed, Oslo, Norway) at a dose of 120 mL with a flow rate of 3 mL/sec. Acquisition parameters for spiral CT were 3-mm collimation, 1-mm reconstruction spacing, and variable pitch.
Since August 2001, the single-detector spiral CT equipment was replaced with a multi–detector row spiral CT scanner (LightSpeed Plus, GE Medical Systems). CT angiography was performed in all patients with the following parameters: high-speed mode capability; gantry rotation time, 0.5–0.6 second; table speed, 7.5 mm per rotation; collimation, 2.5 mm; and reconstruction section thickness, 1.2 mm. Images were acquired in both the arterial phase and a delayed phase.
Scanning delay ranged between 20 and 40 seconds, according to patient circulation time determined with an automated bolus time test (SmartPrep; GE Medical Systems) by using 25 mL of iodinated contrast medium. Scans obtained during the venous phase were acquired with the same parameters 80 seconds after contrast material injection. After contrast-enhanced US and endoleak treatment, CT angiographic follow-up was performed in group A patients and included a more delayed phase (>180 seconds after contrast material administration) to identify late endoleaks.
Images were processed with a dedicated software package at an independent workstation (Advantage Windows 3.1 and 4.1; GE Medical Systems) to generate multiplanar reformations, maximum intensity projections, and volume renderings.
Images were reviewed by one of several authors (I.B., P.P., R.C., C.V.), each with 5–15 years of experience in CT angiography, in consensus, to evaluate maximum aneurysm transverse diameter and to identify presence and origin of endoleaks (defined as presence of contrast enhancement within the aneurysmal sac in the arterial and/or venous phases). The stent-graft was fully visualized, which enabled recognition of migrations, distortions, and structural changes; for example, distal stent-graft migration was considered significant when the distance between the lowest renal artery and the proximal aortic fixation increased by at least 10 mm, compared with the distance measured at the previous examination.
Color-coded Duplex US and Evaluation
Color-coded duplex US was performed by one radiologist (V.N.), who had 16 years of experience with color-coded duplex US and was blinded to CT angiographic findings, with an abdominal phased-array transducer (2.5–3.5 MHz). The entire abdominal aorta was scanned from the diaphragm to the iliac arteries in long axis and cross-sectional views, with the patient in the supine and the right or left lateral decubitus positions (depending on the presence of the aortic deviation), through an anterior and a translumbar approach, respectively. The entire sac was analyzed to visualize possible endoleaks by using color Doppler and power Doppler US and by obtaining a spectral Doppler waveform analysis of the detected leaks. In type II endoleaks, an attempt was made to identify inflow and outflow vessels.
Contrast-enhanced US and Evaluation
At completion of color-coded duplex US, contrast-enhanced US was performed by one radiologist (V.N.), who had 4 years of experience with contrast-enhanced US, after a bolus injection of a second-generation blood pool microbubble contrast agent (Sonovue; Bracco, Milan, Italy). The agent was administered into an antecubital vein at a dose of 1.5–2.4 mL, followed by a flush of 5 mL saline solution. At the time this study was performed, to our knowledge, no published data were available regarding the optimal dose of contrast agent for vascular examinations with contrast-enhanced US, particularly those performed for follow-up after treatment with stent-grafts. Therefore, the dose was determined according to the package insert recommendation of the manufacturer, in which the optimal dose for vascular examinations was fixed at 2.4 mL, and according to our initial experience, in which a lower dose (1.5 mL) appeared to be sufficient for endoleak detection in the majority of patients. At the beginning of our study, we performed contrast-enhanced US with a bolus of 1.5 mL; afterwards, the bolus was increased to 2.4 mL, according to the package insert recommendation, to ensure leakage detectability in all patients. In this series of patients, the bolus of 2.4 mL was injected in four patients (patients 1, 5, 6, and 9). No other criteria were used in the determination of the amount of contrast medium to be injected in each patient.
The contrast-enhanced study was performed with continuous low–mechanical index (0.01–0.04) real-time tissue harmonic imaging (Contrast Tuned Imaging; Esaote Biomedica). The low mechanical index was calculated as the peak negative pressure divided by the square root of the frequency of the pulse. The entire aorta was scanned in the longitudinal and transverse planes from the diaphragm to below the iliac limb attachment sites. Scanning was maintained for 10 minutes after contrast material administration, according to the pharmaceutical indications, and the maximum circulation time of the contrast agent was set at 11 minutes. The entire examination was tape-recorded to allow later review.
Images were analyzed to assess the following: presence of contrast enhancement within the aneurysm sac, time of detection and duration of the persistence of the contrast enhancement (number of seconds after intravenous contrast material administration), origin of the endoleak, identification of inflow and outflow collateral vessels, and enhancement morphology. The latter was described as diffuse (with contrast agent spreading into the thrombus) or cavity filling (with contrast agent concentrating in a pseudocavity within the thrombosed sac).
In patients 1 and 6, a second bolus of 2.4 mL of contrast medium was required, since the initial bolus did not allow detection of perigraft flow.
DSA and Evaluation
DSA was performed in the angiographic suite by four experienced interventional radiologists (I.B., P.P., R.C., C.V.), who each had 5–25 years of experience with angiography and who were aware of the findings of previous examinations. A transfemoral arterial percutaneous access was used, with the patient receiving a local anesthetic (lidocaine 2%, Xylocaine 2%; AstraZeneca, Basiglio, Italy). Aortography was first performed by using a 5-F pigtail catheter positioned above the proximal end of the stent-graft and a bolus injection of 40 mL of the same iodinated contrast medium as was used for CT angiography. Then selective angiography was performed according to the aortographic findings by using the properly shaped catheter.
Images were evaluated in consensus to assess the presence of an endoleak (visualized as the spreading of contrast material outside the stent-graft) and to identify its origin (from aortic branches or stent-graft disconnections). When required, percutaneous treatment of the detected endoleak was performed, by means of either selective arterial coil embolization or cuff deployment.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
|
|
Group A: Color-coded Duplex US and CT Angiographic Results
In all patients, color-coded duplex US was unable to depict any endoleak within the sac both with color-coded Doppler and power Doppler modes.
CT angiographic images were reviewed twice, before and after contrast-enhanced US examination, and this review allowed analysis of unenhanced and arterial- and delayed-phase images and performance of bi- and tridimensional reconstructions at an independent workstation. Neither endoleaks nor stent-graft abnormalities (migrations, distortions, and structural changes) were detected in any patient (Fig 2). DSA findings, endoleak treatment, and follow-up imaging after treatment are summarized in Table 2.
|
|
|
|
Group C: Imaging Results
In all group C patients, the type II endoleak detected by using duplex US and CT angiography was confirmed with contrast-enhanced US (Fig 5). In this group of patients, contrast enhancement was characterized by an arterial inflow in all cases (seven lumbar arteries and three inferior mesenteric arteries), with early echo enhancement (30–60 seconds after intravenous contrast agent administration). Besides, an arterial outflow was visualized in six of 10 patients (three inferior mesenteric arteries and three lumbar arteries). Patients with detected outflow showed rapid contrast agent washout (90–120 seconds after contrast agent administration). When no arterial outflow was detected (four of 10 patients), washout was slower and persisted for longer than 150 seconds after contrast agent administration (longer than 240 seconds in two of 10 patients).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
Endoleak classification has changed over the years. The current classification includes endoleaks that result from incomplete sealing of the stent-graft at the attachment sites (type I), those determined by means of retrograde flow from aortic collateral vessels (type II), and those due to graft disruption (type III) and porosity (type IV) (14,15).
A concept has been introduced that arises from the observation that excluded aneurysms can increase with no evidence of endoleak. This phenomenon has been defined as endotension, or type V endoleak. A pressure transmission through thrombus at the attachment site has been proposed as the cause of the aneurysm enlargement (9,10). Another hypothesis is that a very low flow endoleak is present and is not depictable with standard imaging modalities, since it allows rapid blood clot formation (16). Nevertheless, the existence of this phenomenon is somehow controversial and gives rise to a few problems in regard to treatment options in these patients (17).
The most accepted imaging modality for follow-up in patients who have undergone EVAR is CT angiography, particularly biphasic scanning (3,18). Nevertheless, radiation exposure, contrast agent allergy, and nephrotoxicity, as well as cost, represent some limitations of CT angiography. Several authors have pointed out the usefulness of duplex US, because it seems to allow better identification and characterization of endoleaks, with analysis of flow direction and velocity (6,19,20). Researchers in scattered reports also focus on the use of MR angiography as an effective tool for the identification of perigraft flow (4).
In our institution, more than 100 patients underwent EVAR in a 4-year period. All patients treated are carefully followed up with both color-coded duplex US and thin-section multiphasic CT angiography (21). However, in our experience, CT angiography does not allow detection of some types of endoleaks (2), even when intrasac systemic pressure is observed (22). Therefore, an alternative imaging modality is needed.
US contrast agents seem to substantially increase US diagnostic accuracy and have been recently applied in several fields (23,24). Their effectiveness in aiding in the detection of endoleaks has also been recently investigated (12).
In this series of patients, we evaluated the role of contrast-enhanced US in 10 consecutive patients whose duplex US and CT angiographic findings demonstrated increasing aneurysm diameter but no signs of perigraft flow or graft morphologic changes. To our knowledge, this is the first study specifically dedicated to the evaluation of patients with endotension with the use of US contrast media.
In all our patients, contrast-enhanced US demonstrated the presence of an endoleak, which was characterized by slow and diffuse echo enhancement within the sac and was detected longer than 2 minutes after contrast agent administration. This observation leads us to question the existence of endotension as a true entity and introduces new and interesting perspectives in the follow-up of patients who have undergone EVAR and in the endoleak classification.
Researchers in recent published studies (8,25) demonstrated that flow hemodynamics of type II endoleaks affect endoleak persistence and treatment outcome. Therefore, a new, more detailed classification of leaks, particularly type II endoleaks, is required that is based on Doppler waveforms and flow velocities (2,26).
There might be a relationship between blood flow characteristics and CT angiography and duplex US leakage detectability. In fact, in all patients with a nonvisualized leakage at CT angiography and duplex US, the perigraft flow visualized at contrast-enhanced US was characterized by very slow flow dynamics, with diffuse and delayed aneurysm enhancement. The contrast medium did not concentrate in a confined part of the sac, but it spread into the thrombus. The delayed appearance of the leak, its spreading throughout the sac, and its very slow flow could be the key factors in the explanation for the undetectability of endoleaks at CT angiography.
It would be interesting to investigate whether performance of CT scanning with longer delays (>3–4 minutes after contrast agent administration) increases CT sensitivity in the detection of endoleaks in patients with an enlarging aneurysm and no evidence of other complications. In fact, considering the high comparability of CT angiographic and contrast-enhanced US findings in group C patients (CT angiography–depicted endoleaks), it could be supposed that the delayed enhancement observed at contrast-enhanced US could also be visualized through CT scanning with longer delays (21). In our series of patients, this finding was observed at CT angiographic follow-up in two group A patients.
In our current series of patients, the majority of endoleaks in group A patients were fed by tiny lumbar arteries, and in no patients were outflow vessels demonstrated. The absence of an outflow artery could be critical for the pressurization of the sac, and this absence determines its enlargement (27,28). In fact, all group A patients showed evident aneurysm increase at follow-up even though the endoleak was hardly detectable.
The reliability of contrast-enhanced US in our series of patients is confirmed by the demonstration that in group B patients (aneurysm reduction and no evidence of complications), no US contrast enhancement was demonstrated within the aneurysm sac. Besides, in patients with evidence of a leak at duplex US and CT angiography (group C patients), contrast-enhanced US findings were comparable to those of the other imaging modalities. In all cases, early echo enhancement was observed, and it was possible to visualize both the inflow and outflow vessels.
Similar results were obtained by Bendick et al (12) in a series of 20 patients. They proved that contrast-enhanced US increased US sensitivity. In their article, these investigators reported the detection of two endoleaks that were not visualized at CT angiography but that were confirmed at DSA.
Contrast-enhanced US has several advantages: It is noninvasive, fast, very well tolerated, reproducible, and apparently very sensitive (12). The US contrast agent has no substantial contraindications: It consists of stabilized microbubbles of sulfur hexafluoride gas, which is eliminated through the respiratory system. It is of low solubility, innocuous, isotonic to human plasma, and devoid of antigenic potential, as it does not contain any proteinaceous material. In our series, no side effects related to the contrast agent were observed. The required dose of the microbubble contrast agent is still not well defined. At the beginning of our study, we used a dose of 1.5 mL of contrast medium; afterwards, to ensure detectability of contrast enhancement, the dose was increased to 2.4 mL, although other authors reported the use of only 1 mL of contrast medium (12).
Because of its favorable features, it is possible to administer two consecutive boluses of the microbubble contrast agent, when required; in our series of patients, a second dose was injected in two patients (patients 1 and 6) in whom imaging after administration of the first bolus did not show any clear endoleak. Contrast-enhanced US can be performed for analysis of only one defined area of the aneurysm with continuous imaging. Performance of the examination without any knowledge of the site of the leak might cause problems in the immediate identification of the sac enhancement; therefore, a second bolus might be required to image the aneurysm at a different level, particularly in patients with an enlarging aneurysm in whom there are no CT angiographic or duplex US findings of other complications. Besides, administration of a second bolus can help in the clearer depiction of slight, not well defined contrast enhancement obtained after the first injection (Fig 1).
In our series of patients, no false-positive results were observed. We agree with Bendick et al (12) in stating that the use of tissue harmonic imaging is mandatory to increase contrast-enhanced US sensitivity, allow longer scanning times and refilling processes, and reduce the incidence of false-positive studies caused by the Doppler blooming artifacts that were previously described (11). In a previous study of McWilliams et al (11) that included 53 patients, increased sensitivity of contrast-enhanced US was reported when compared with that of Doppler US, yet there was an increase of false-positive studies and a relatively low negative predictive value (86%) compared with those of biphasic CT angiography. However, they used an older-generation contrast agent and power Doppler and color Doppler imaging, which caused blooming artifacts.
The main limitation of our study was the small number of patients examined; results of a study with larger series of patients and a prospective nature would be needed to confirm our findings. Further investigation is needed to assess whether CT angiography with a longer delay would help to identify these missed endoleaks as well as they are identified at contrast-enhanced US.
Contrast-enhanced US has some limitations. The required equipment, including the contrast agent, is highly specific, not yet widely available, and expensive. Obesity and bowel gas can interfere with US scanning, and patient collaboration is always required (12). The examination is operator dependent and requires specific skills and training. Finally, contrast-enhanced US can be very sensitive in the depiction of perigraft flow but does not seem to be appropriate for the evaluation of other factors, such as graft anchorage and integrity and aneurysm morphologic changes, for which CT angiography remains the first-option imaging modality (18,29).
We believe contrast-enhanced US should not be used routinely for follow-up in patients who have undergone EVAR, but it should be reserved for patients with aneurysm enlargement and no endoleak or for those with graft abnormalities depicted with the other imaging modalities.
|
FOOTNOTES |
|---|
Authors stated no financial relationship to disclose.
Author contributions: Guarantors of integrity of entire study, M.F., C.B.; study concepts, I.B., V.N., S.G.S.; study design, V.N., I.B.; literature research, I.B.; clinical studies, V.N., I.B., S.G.S., P.P., R.C., C.V., M.F.; data acquisition, V.N., I.B., S.G.S., P.P., R.C., C.V., M.F.; data analysis/interpretation, V.N., I.B.; manuscript preparation, revision/review, and final version approval V.N., I.B.; manuscript definition of intellectual content and editing, M.F., C.B.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  P Sharma and C Kyriakides Surveillance of patients post-endovascular aneurysm repair Postgrad. Med. J., December 1, 2007; 83(986): 750 - 753. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. W. Stavropoulos and S. R. Charagundla Imaging Techniques for Detection and Management of Endoleaks after Endovascular Aortic Aneurysm Repair Radiology, June 1, 2007; 243(3): 641 - 655. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Swaminathan, G. B. Mackensen, M. V. Podgoreanu, R. L. McCann, J. P. Mathew, and G. C. Hughes Spontaneous Echocardiographic Contrast Indicating Successful Endoleak Management Anesth. Analg., May 1, 2007; 104(5): 1037 - 1039. [Full Text] [PDF]
|
|
|
|
 |
|
 A. P. Hingorani, E. Ascher, and N. Marks Duplex Arteriography for Lower Extremity Revascularization Perspectives in Vascular Surgery and Endovascular Therapy, March 1, 2007; 19(1): 6 - 20. [Abstract] [PDF]
|
|
|
|
 |
|
 M. J. Dill-Macky, S. R. Wilson, Y. Sternbach, J. Kachura, and T. Lindsay Detecting Endoleaks in Aortic Endografts Using Contrast-Enhanced Sonography Am. J. Roentgenol., March 1, 2007; 188(3): W262 - W268. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Iezzi, A. R. Cotroneo, A. Filippone, F. Di Fabio, F. Quinto, C. Colosimo, and L. Bonomo Multidetector CT in Abdominal Aortic Aneurysm Treated with Endovascular Repair: Are Unenhanced and Delayed Phase Enhanced Images Effective for Endoleak Detection? Radiology, December 1, 2006; 241(3): 915 - 921. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
