| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 Department of Diagnostic Radiology, RWTH Aachen University, Pauwelsstrasse 30,
52057 Aachen, Germany.
2 Department of Radiology, University Hospital of the Saarland, Homburg/Saar,
Germany.
3 Department of Vascular Surgery, RWTH Aachen University, Aachen, Germany.
Received May 10, 2007;
accepted after revision December 14, 2007.
Address correspondence to K. M. Ruhl
(karlruhl{at}gmx.net).
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. Twenty-one patients (five women, 16 men; mean age ± SD, 65 ± 14 years) were examined on a 3-T whole-body MR system with an 8-element phasedarray coil for preoperative evaluation of the pedal arterial system and assessment of the visualized vessels to serve as a graft touch-down site in pedal bypass surgery. Time-resolved 3D MRA of the foot was performed after automatic injection of 0.2 mmol/kg of gadobenate dimeglumine using a sagittal gradient-echo T1-weighted sequence (TR/TE, 4.2/1.6; flip angle, 30°; field of view, 290 mm; matrix, 352; 120 slices; slice thickness, 0.8 mm) with a spatial resolution of 0.8 x 0.8 x 1.6 mm reconstructed to 0.6 x 0.6 x 0.8 mm and a temporal resolution of 3.9 seconds using keyhole and sensitivity-encoding (SENSE) technology (SENSE factors: 4 in anteroposterior direction and 2 in right-left direction). Dynamic subtractions and rotating maximum intensity projections were calculated. The original image data sets were transferred to a dedicated workstation for objective signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) analysis of the arteries. Subjective image analysis regarding image quality and diagnostic findings was performed by two radiologists in consensus.
RESULTS. In all patients, images of diagnostic quality were obtained. Despite the known limitations regarding signal intensity measurements in images acquired with the use of parallel imaging technique, SNR and CNR proved to be excellent, with mean ± SD values of 294 ± 158 and 248 ± 144, respectively. Although most of the patients had diabetic foot syndrome with arteriovenous shunting, the arteries and the potential vessel for bypassing could be clearly separated from the veins in each case due to the temporal information given by our study. The ability to reliably discriminate arteries from veins is of high clinical relevance in planning pedal bypass surgery.
CONCLUSION. Fast contrast-enhanced time-resolved 3D MRA of the foot at 3 T is feasible and of high clinical value for the preoperative evaluation of the arterial supply of the foot.
Keywords: contrast-induced nephropathy • dorsalis pedis bypass • MR angiography • parallel imaging • pedal arteries • peripheral arterial disease • vascular imaging
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Advances in MR hardware and fast data acquisition techniques, such as parallel imaging or advanced k-space filling schemes, make it possible to complete a whole 3D data set in a few seconds [3, 4]. This capability allows both high spatial resolution and high temporal resolution while achieving large volume coverage. In this prospective clinical study, we evaluated the potential of submillimeter in-plane spatial resolution contrast-enhanced time-resolved 3D MRA at a high magnetic field strength of 3 T for preoperative evaluation of the pedal vasculature in patients with diabetes, peripheral arterial disease (PAD), or both.
|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
MRI
Examinations were performed on a 3-T MR system (Achieva, Philips Medical
Systems). The patients were examined in the supine position with one foot
placed in a commercial cylindrically shaped 8-element phased-array
radiofrequency coil (SENSE head, MRI Devices) for signal reception. The foot
was positioned in slight plantar flexion and immobilized with foam
padding.
Localization and Coil Sensitivity Calibration
Localization was performed using a gradient-echo sequence (TR/TE, 6.7/1.3;
flip angle, 50°; field of view [FOV], 300 x 300 mm; matrix size, 256
x 256 pixel elements; three perpendicular stacks in sagittal, coronal,
and axial orientations; scanning duration, 8 seconds). An extra reference scan
was obtained to determine the coil sensitivity profiles (4.0/0.8; flip angle,
1°; FOV, 300 x 300 mm; matrix, 64 x 64).
Time-Resolved 3D MRA (4D MRA)
MRA images were acquired in a sagittal plane using a 3D gradient-echo
T1-weighted sequence with the following parameters: 4.2/1.6; flip angle,
30°; rectangular FOV, 290 x 290 mm; matrix size, 352 x 352
reconstructed to 512 x 512; 120 slices; spatial resolution, 0.8 x
0.8 x 1.6 mm reconstructed to 0.6 x 0.6 x 0.8 mm. To achieve
a temporal resolution of 3.9 seconds per dynamic scan, 2D sensitivity-encoding
(SENSE)-based parallel imaging was performed using a net acceleration factor
(R) of 8 (R = 4 along the anteroposterior direction and
R = 2 in the right-left orientation)
[5]. To further accelerate
time-resolved 3D MRA, parallel imaging was supplemented by the half-Fourier
technique and keyhole imaging (25%) in conjunction with centrally ordered
k-space sampling (termed "contrast-enhanced time robust
angiograph" [CENTRA])
[6-8].
Centrally ordered k-space sampling was applied to enhance image contrast in
contrast-enhanced MRA (CE-MRA). For every patient, 16 dynamic scans, each with
a 3.9-second temporal resolution, were obtained to cover both arterial and
venous phases of the contrast agent passage through the pedal vessels. This
was followed by a fully sampled reference scan mandatory for the keyhole
approach, accounting for an overall scanning duration of 75 seconds (3.9
seconds for each dynamic scan and 16.5 seconds for the last dynamic scan
including the reference image).
Paramagnetic contrast agent (0.5 mmol/L of gadobenate dimeglumine [MultiHance, Bracco]) [9] was injected automatically (Spectris Solaris EP MR Injection System, Medrad) through a peripheral catheter placed in an antecubital vein. A dose of 0.2 mmol/kg of body weight was injected with a flow rate of 3.0 mL/s followed by a 30-mL saline flush at the same flow rate. Data acquisition was started manually 10 seconds after the beginning of contrast injection to ensure at least one unenhanced dynamic scan was obtained for dynamic subtraction. Rotated dynamic maxi mum intensity projections were rendered over a 180° sector with 20 reconstructions from the subtracted images. Interpretation of the images was based on both the original data set and the maximum-intensity-projection images.
The images were transferred to a dedicated workstation (View Forum, Philips Medical Systems) and to a PACS (Sectra PACS, Sectra Imtec AB). Two radiologists read the images in consensus regarding good image quality with sharply delineated pedal vessels, differentiation between arteries and veins, presence of arterio venous shunting with early venous drainage, and pronounced soft-tissue enhancement [10]. Further more, the data sets were analyzed regard ing whether a temporal resolution of 3.9 seconds, as given in this study, was necessary or whether a temporal resolution of 7.8 seconds (2 x 3.9 seconds) would have been suitable to obtain the same diagnostic outcome.
The mean time between the peak arterial enhancement and the peak venous
enhance ment was calculated in 11 MRA data sets. The other 13 data sets could
not be analyzed in that respect because either the peak venous enhancement was
not reached reliably until the last dynamic scan (eight of 24 data sets) or no
venous filling was observed during the scanning time (five of 24 data sets)
(Fig. 1). For further objective
image analysis, regions of interest (ROIs) were placed in the original data
sets to measure signal intensity in either the dorsal pedal artery or the
posterior tibial artery (SIart) and in a pedal vein
(SIven) in every dynamic scan, in the muscle (SImusc),
and in the surrounding air (SIair). Signal-to-noise ratios (SNRs)
were calculated for the dynamic scan showing the highest signal intensity in
the vessel with the following equation:
 |
 |
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
|
The pedal arch proved to be patent in 22 of the 24 angiography studies. In the remaining two angiography examinations, the pedal arch was not displayed, although the main supplying arteries of the foot were clearly visible and delineated. In one patient, we found that the pedal arch was missed because of the tremendously delayed arrival of contrast medium in the target area (first arterial enhancement in the foot was observed at approximately 55 seconds after contrast agent injection). In the other patient, both the dorsal artery of the foot and the posterior tibial artery were occluded proximal to the pedal arch.
In every patient, both anterior and posterior circulations or one of the circulation systems of the foot could be evaluated due to occlusion of its counterpart. The posterior circulation was patent in 20 of the 24 MR angiography examinations, whereas the anterior circulation could be identified in 14 of the 24 examinations.
Pronounced soft-tissue enhancement could be visualized in 17 MR angiography studies [10]. The most common locations for soft-tissue enhancement were the forefoot (10 cases), the heel (nine cases), and toes (eight cases) (Figs. 3 and 4).
|
 |
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
As new therapeutic, especially surgical, options for the therapy of the diabetic foot emerge, the radiologist is expected to provide an accurate preoperative angiogram of the foot even in the presence of severe macroangiopathic changes and impaired renal function. MRA can provide high-quality images in this setting using the described technique. Postoperatively, MRA can be used to prove the patency of the bypass graft and to document improved perfusion of the foot (Fig. 5). Most previous studies have focused on high-resolution MRA of the foot [13]. Some authors also have described the time-resolved approach for MRA of the lower leg or foot at 1.5 T with a temporal resolution of between 6 and 25 seconds but with limited spatial resolution [10, 14]. However, CE-MRA of the foot requires high spatial resolution and high temporal resolution. High spatial resolution is mandatory to provide details about the vascular anatomy, which becomes key in target areas occupied by vessels with small diameters. High temporal resolution ensures that at least one acquisition coincides with the desired arterial bolus phase and enables separation of arteries from veins. Furthermore, high-temporal-resolution images provide information about flow dynamics [15].
|
In this study, we approached the challenge of combining submillimeter in-plane spatial resolution with whole-target-organ coverage while achieving a temporal resolution of 3.9 seconds, which was enabled by the synergy of a keyhole-based time-resolved sampling strategy, half-Fourier imaging, elliptic-centric phase encoding, and 2D SENSE-based parallel imaging.
Eliminating the background signal via subtraction facilitates clear visualization of the enhanced vascular lumen. Investigators have also found that this approach enables visualization of small distal vessels, especially in regions where high fat signals are prevalent, as given in the foot [14]. Therefore, background subtraction is essential for a reliable diagnostic evaluation of pedal vasculature. The main disadvantage of subtraction techniques is susceptibility to motion, which may result in blurring; therefore, this disadvantage stresses the need for fixation of the foot in the coil to be adequate for imaging but not painful for the patient. For this purpose, we used elastic foam padding, which was tolerated well by the patients.
In a number of studies, investigators have proven the feasibility of CE-MRA of the foot at 1.5 T [10, 16], reporting its higher confidence and better quality when compared with 2D time-of-flight angiography [14] and even its supremacy over DSA [13], particularly because higher intraluminal contrast is needed in DSA, especially in patients with stenoses of the lower leg arteries. DSA in combination with standard angiography may fail to reveal patent arteries or vessel segments that are suitable for distal bypass grafting in patients with severe arterial occlusive disease [13].
At 3 T, mostly neurovascular imaging has been clinically implemented to date [17, 18]. Renal MRA at 3 T has been proven feasible for high-spatial-resolution 3D MRA [19]. To our knowledge, this study is the first to prove the feasibility of fast contrast-enhanced time-resolved 3D MRA of the foot with submillimeter spatial resolution at 3 T.
The ability to reliably discriminate arteries from veins is of high clinical relevance in planning pedal bypass surgery. In our protocol, this distinction was possible without restrictions in any case. Anterior and posterior circulations and the patency of the pedal arch could be evaluated in every patient. The key information needed before vascular surgery, which includes patency of the pedal arch, documentation of the presence and degree of collateral pathways, and accurate depiction of target vessels suitable for surgical bypass [16], were all provided by our study.
In 10 patients, either the peak venous enhancement was not reached or no venous filling at all could be visualized during 62.4 seconds for 16 dynamic scans. This is due to current hardware and MR system design constraints—that is, the maximum number of dynamic scans is limited by an intrinsic limitation of the reconstruction memory capacity, so the imaging protocol used in this study supports a maximum number of 16 dynamic scans. To overcome this constraint, 64-bit-based reconstruction hardware needs to be made clinically available, which will help to further foster the clinical potential of highly accelerated 3D CE-MRA for the venous drainage evaluation. Many element coil arrays, together with many receiver channels [20], may provide clinical options to further improve both temporal and spatial resolution.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) and venous (
) signal intensity
curves over time in 70-year-old man with peripheral arterial disease, Fontaine
stage IV, and diabetes.
