| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Report |
1 Department of Radiology, Weill Medical College of Cornell University, 416 E
55th St., New York, NY 10021.
2 Advanced Magnetics, Inc., Cambridge, MA 02138.
3 Department of Vascular Surgery, Weill Medical College of Cornell University,
New York, NY 10021.
Received December 18, 2002;
accepted after revision November 11, 2003.
Address correspondence to M. R. Prince.
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CONCLUSION. Blood pool MR angiography using Ferumoxytol reveals more aortic stent-graft endoleaks than does CT angiography and depicts more endoleaks 24 hr after administration than during the immediate arterial phase because of a 50-fold increase in the volume of enhancement in the aneurysmal sac outside the stent-graft.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The term "endoleak" refers to persistent blood flow into the aneurysmal sac after stent-graft placement [1]. Large endoleaks may be detected on arterial phase MR angiography or CT angiography, but small or slow endoleaks may not be visible immediately after contrast injection. Delayed CT angiography or MRI allows more time for the contrast agent to enter the endoleak [2, 3]; however, redistribution of iodine or gadolinium contrast agents into the extracellular fluid compartment and excretion of these agents degrade delayed images. Blood pool contrast agents allow delayed MR images without degradation to be obtained. Although the United States Food and Drug Administration has not yet approved blood pool contrast agents for clinical use, we had the opportunity to test one, Ferumoxytol (Advanced Magnetics), as part of a phase II study.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Iron Oxide
The blood pool contrast agent that we used was Ferumoxytol. This agent
consists of an ultrasmall (
30 nm) superparamagnetic iron oxide core
encapsulated in semisynthetic carbohydrate. This thick coating prevents
redistribution outside the vascular space. Blood elimination is via
phagocytosis by macrophages in the reticuloendothelial system. The plasma
elimination half-life in humans is approximately 10–14 hr. Iron released
after phagocytosis enters the body's iron stores.
Patients received Ferumoxytol at a dose of 4 mg of iron per kilogram of body weight. Contrast injection was performed manually at a rate of 1 mL/sec and was followed immediately by a 20-mL saline flush using a tubing set that switched automatically between contrast agent and saline flush (SmartSet, TopSpins).
Imaging Technique
All patients were imaged on a 1.5-T unit (CVI, General Electric Medical
Systems). The body coil was used for signal transmission and a phased array
coil for reception. We used a 48 x 28 cm phased array coil (ICG, Medical
Advances) on the first four patients, but this device caused significant
superior-to-inferior wraparound ghosting artifact on axial 3D gradient-echo
images. Accordingly, for the last three patients, we used a smaller phased
array coil (Torso coil, MedRad) that offered less superior-to-inferior
coverage but matched the axial 3D volume of imaging.
In the first four patients, we performed unenhanced and arterial phase 3D MR angiography in the coronal plane with the following parameters: TR/TE, 5/1; field of view, 40 cm; slice thickness, 3.0 mm with 1.5 mm of overlap; flip angle, 30°; acquisition matrix, 512 x 160–192; and elliptical centric k-space view ordering. We performed delayed imaging at 8–23 min (mean, 14 min) and 24 hr in the axial plane with a smaller field of view (34 cm); more slices (n = 40); thicker slices (8–10 mm) reconstructed at 4- to 5-mm intervals using zero interpolation; sequential ordering of k-space; and fewer phase-encoded steps, both with and without fat suppression. Imaging at 8–23 min and at 24 hr was performed with identical imaging parameters for each subject. The variation in the time of the initial delayed imaging (8–23 min) reflected the variable amount of time necessary to acquire vital signs after injection and blood sampling. The longest delay (23 min) occurred because the scanner computer failed during a study of one patient, resulting in extra time needed for rebooting and relocalization. If motion or poor breath-holding degraded image quality, we repeated imaging until we achieved satisfactory results. Fat suppression was performed with a spectrospatial inversion pulse applied once per slice loop with an inversion time of 17–20 msec.
MR Angiography Image Analysis
A radiologist unacquainted with any clinical information reviewed the MR
angiography studies on a computer workstation (Vitrea, Vital Images) to
determine the volume of the aneurysm and the possibility of endoleak. The
location and volume and the rate of the endoleak were determined on arterial
and delayed phase (8–23 min and 24 hr, respectively) 3D MR angiograms.
Volume was calculated by adding the contrast-enhanced cross-sectional area for
each slice when contrast agent was observed in the aneurysmal sac outside the
graft lumen and by multiplying that value by slice spacing. The rate of the
endoleak was calculated as the ratio of endoleak volume (in milliliters) on
equilibrium phase images divided by the time from Ferumoxytol injection to
equilibrium phase imaging. The rate of endoleak was calculated for the 8- to
23-min point in the four patients with an endoleak detected on early phase
images and for the 24-hr time point in all patients.
CT Angiography
CT angiography, including arterial phase and delayed phase imaging, was
performed on a helical CT scanner (LightSpeed Plus, LightSpeed Ultra, or
LightSpeed QX/i, General Electric Medical Systems). A 2.5-mm slice thickness
was used, and 150 mL of iohexol (Omnipaque 350 I mg/mL, Amersham) was injected
at a rate of 4 mL/sec.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The maximum diameter of the aneurysm before treatment with aortic stent-grafting was between 4.4 and 6.5 cm (mean, 5.0 cm). After stent-graft placement, the size of the aneurysm decreased in two patients and showed no significant change in three patients. In the last patient, 1-month follow-up CT angiography revealed a rapid type II endoleak that had increased the maximum diameter of the aneurysm from 4.7 to 6.1 cm. The size of the aneurysm had stabilized at 2 years of follow-up, and this patient continues to be monitored.
The period of time between the initial stent-graft placement and the most recent CT angiography examination varied from 120 to 630 days (mean, 305 days), and the mean time interval between the last CT angiography and blood pool MR angiography was 90 days (range, 5–203 days).
Endoleak volumes and rates are shown in Table 1. In two patients, CT angiography studies showed an endoleak. Iron oxide–enhanced MR angiography revealed the endoleak during the arterial phase after the iron oxide injection and showed the endoleak location to be similar to its location on CT angiography (Figs. 1A, 1B, 1C). The rates of endoleak were high—0.5 and 0.3 mL/min. For these two patients with endoleak detected on CT angiography, the endoleak appeared larger on arterial phase CT angiography than on arterial phase MR angiography. This discrepancy may relate to the relatively longer delay between initiating contrast injection and arterial phase scanning in CT angiography than in MR angiography.
|
|
|
|
In one patient with an endoleak shown on CT angiography, an endoleak at the proximal segment of the graft was seen on intraarterial digital subtraction angiography performed at the time of stent-graft placement. Although the endoleak could not be corrected with repeated balloon inflation, it was small enough that conversion to open surgical repair was not performed.
In the other patient with an endoleak shown on CT angiography, intraarterial digital subtraction angiography was performed 7 months later and revealed patent lumbar arteries at two levels (L4 and L5) serving as feeder vessels to the aneurysmal sac. Coil embolization was performed, but the endoleak persisted after the procedure because of another patent lumbar artery. CT angiography and, 5 days later, blood pool MR angiography (Fig. 1D) identified this endoleak as type II with a feeding lumbar artery.
|
In the four patients with no endoleak shown on CT angiography, blood pool agent MR angiography revealed an endoleak. In one of these patients, CT angiography was of poor quality with only minimal arterial enhancement (Fig. 2A, 2B, 2C, 2D). For two patients with negative CT angiography results, the endoleaks were detected only on delayed MR angiograms obtained 24 hr after iron oxide administration (Fig. 3A, 3B, 3C). The endoleak rate in these two patients was extremely slow— 0.003 and 0.0007 mL/min. For the remaining patient with false-negative results on CT angiography, the endoleak was small on the arterial phase acquisition but was larger at 24 hr after Ferumoxytol injection, with an endoleak rate of only 0.1 mL/min.
|
|
|
|
|
|
|
The small number of subjects did not allow systematic testing of the imaging parameters. However, we found the axial orientation to be superior to coronal for identifying the anatomic structures and detecting endoleaks, and fat saturation to be useful for eliminating wraparound ghosting artifact in the slice direction and helping to distinguish endoleak from background tissues. Finally, the reformations and subvolume maximum intensity projections performed on the computer workstation were deemed essential for adequately evaluating the endoleak, especially for identifying feeding arteries.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Iron oxide MR angiography has been shown to be useful for detecting bleeding into the bowel lumen in animal models [8–10]. Our study of six patients with nitinol aortic stent-grafts indicates that iron oxide blood pool contrast agents are useful in illuminating subtle aortic stent-graft endoleaks that neither CT angiography nor intraarterial digital subtraction angiography detected. In these cases, iron oxide MR angiography revealed the location and volume of the endoleak and its feeder vessels. Furthermore, repeated MR angiography over 24 hr produced data from which the endoleak rate could be calculated.
No gold standard for detecting endoleak has been established, so we cannot
be 100% certain of the diagnoses made on the basis of MR angiography. However,
four of the six patients who were determined to have an endoleak showed no
reduction in aneurysm size after stent-graft placement, as would be expected
if the aneurysmal sac had thrombosed. For the two patients whose aneurysm
diameter did decrease (indicating a reduction in endotension), both had slow
endoleak rates (
0.1 mL/min) and small endoleak volumes (3.3 and 5.1 mL).
With one exception, the slower endoleaks ( 0.1 mL/min) were undetected on
CT angiography, the one exception being a poor-quality CT angiography study
with poor arterial enhancement that was interpreted as showing
"negative" findings for endoleak.
In this small series, none of the cases of endoleak positively diagnosed in volunteer subjects progressed to aneurysm rupture. However, if an aneurysm enlarges, prudent management includes ongoing imaging follow-up and intervention. Determining the clinical significance of CT-occult endoleaks or the definitive MRI measurements of endoleak rate will require outcomes studies performed on a larger population.
We had wondered whether this study was inherently biased by the requirement of voluntary participation because patients who suspected an existing endoleak might have been more likely to volunteer for complimentary scanning and perhaps the incidence of endoleak in this study was greater than in the general population of stent-graft patients. However, we now think that stent-graft endoleak rates are substantially higher than the incidence predicted on the basis of CT angiography data and that blood pool MR angiography with Ferumoxytol represents a promising method for better detection and characterization of these endoleaks.
Acknowledgments
We thank Advanced Magnetics, Inc., for financial support and technical
assistance for this study.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  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]
|
|
|
|
|
|
W. Li, J. Salanitri, S. Tutton, E. E. Dunkle, J. R. Schneider, J. A. Caprini, L. N. Pierchala, P. M. Jacobs, and R. R. Edelman Lower Extremity Deep Venous Thrombosis: Evaluation with Ferumoxytol-enhanced MR Imaging and Dual-Contrast Mechanism--Preliminary Experience Radiology, March 1, 2007; 242(3): 873 - 881. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. V. Prasad Functional MRI of the kidney: tools for translational studies of pathophysiology of renal disease Am J Physiol Renal Physiol, May 1, 2006; 290(5): F958 - F974. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Eggebrecht, H. Kuhl, G. M. Kaiser, S. Aker, M. O. Zenge, F. Stock, F. Breuckmann, F. Grabellus, M. E. Ladd, R. H. Mehta, et al. Feasibility of real-time magnetic resonance-guided stent-graft placement in a swine model of descending aortic dissection Eur. Heart J., March 1, 2006; 27(5): 613 - 620. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
