| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 Department of Body MRI, Mallinckrodt Institute of Radiology and Washington
University in St. Louis, Washington University Medical Center, 510 S
Kingshighway Blvd., St. Louis, MO 63110-1076.
2 Siriraj Hospital, Mahidol University, Bangkok 10700, Thailand.
3 Siemens Medical Solutions, Malvern, PA 19355.
4 University of Texas Health Science Center at San Antonio, San Antonio, TX
78229-3900.
Received March 4, 2005;
accepted after revision April 18, 2005.
Address correspondence to R. Tongdee
(tongdeer{at}mir.wustl.edu).
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. Fifty-three patients (99 legs) were scanned with a 1.5-T MR system equipped with a dedicated bilateral lower extremity phased-array coil. First, high-resolution 3D CE-MRA images of the calves and feet were obtained using two separate sagittal slabs with parallel imaging, with a resulting voxel size of 1.4 x 1.0 x 1.0 mm3. Second, standard bolus chase 3D CE-MRA was performed from the abdomen and pelvis station to the calf-foot station. Images were interpreted by two radiologists. The calf-foot arterial trees were divided into 12 segments. Each segment was characterized as diagnostic or nondiagnostic. The degree of venous contamination was assessed as interfering with the diagnosis or not. Paired Student's t test and Wilcoxon's signed rank test were used to test for statistically significant differences between the techniques.
RESULTS. For the left leg (n = 48), the mean number
(± SD) of diagnosed arterial segments for HyPer 3D CE-MRA was 9.2
± 2.3 and for bolus chase 3D CE-MRA, 7.1 ± 4.2 (p 
0.0004). For the right leg (n = 51), the corresponding values were
9.4 ± 2.2 and 7.6 ± 3.5 (p 0.0005), respectively.
For bolus chase 3D CE-MRA, venous contamination interfered with the diagnosis
in 24 of 99 legs, whereas with HyPer 3D CE-MRA, there was no interference.
Selective analysis of the dorsalis pedis arteries showed that the number of
diagnostic vessels was 62 (62.6%) of 99 for HyPer 3D CE-MRA and 13 (13.1%) of
99 for bolus chase 3D CE-MRA.
CONCLUSION. HyPer 3D CE-MRA is an alternative method for time-resolved high-resolution peripheral CE-MRA in evaluating the trifurcation and feet vessels with no venous contamination.
Keywords: calf • extremities • foot • MR angiography • MR arteriography • MRI • parallel imaging • peripheral vascular disease
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The 3D CE-MRA examination of the pelvis and lower extremities is commonly performed using a three-station moving-table bolus chase technique. With this technique, the timing of the bolus is optimized for the first station (abdomen-pelvis), and then imaging is performed as rapidly as possible to try to keep up with the flow of gadolinium down the peripheral vasculature. Although the MRA image quality is generally excellent for evaluation of the first two stations (abdomen-pelvis and thigh), quite often it is inadequate for assessment of the last station (calf and foot). Imaging of the lower station is typically degraded by venous contamination [10-12]. In addition, bolus chase 3D CE-MRA offers suboptimal spatial resolution because of restrictions caused by the imaging time and imaging volume.
To accurately characterize a stenosis, resolution in all planes should not be less than approximately one third of the vessel diameter [13]. Thus, inadequate spatial resolution is a problem in the evaluation of the small vasculature in the calf and foot vessels. The popliteal artery is a frequent site for insertion of a surgical bypass graft. Visualization of popliteal and infrapopliteal vasculature is critical before planning any revascularization procedure [14]. For this reason, further optimization of the 3D CE-MRA runoff technique is needed to improve infrapopliteal arterial depiction and to eliminate venous contamination.
|
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Patients
From August 2003 to October 2003, 53 consecutive patients (25 males, 28
females; mean age ± SD, 65 ± 13 years; range, 16-89 years) were
included in the present study; the patients were evaluated with combined
sagittal acquisition MRA of the calves and feet and three-station moving-table
bolus chase technique (HyPer 3D CE-MRA technique). Forty-seven patients had
clinical symptoms of lower extremity arterial occlusive disease, including
intermittent claudication, rest pain, discoloration, and ischemic ulcer. Of
the remaining six patients, two patients had clinical suspicion of
arteriovenous malformations, two patients had clinical suspicion for deep vein
thrombosis, one patient had popliteal artery aneurysms, and another was being
evaluated preoperatively for a fibula free-flap graft. Seven of these 53
patients had undergone prior unilateral lower extremity amputation. Thus, a
total of 99 lower extremities were evaluated for the study.
To minimize a patient's time on the MR table, preparations of the patient were performed in the recovery bay area. The preparations included establishing IV access, coaching the patient in breath-holding, considering whether oxygen supplementation would be needed, and discussing the procedure with the patient.
In the MRI suite, the patient was placed in a supine position feet-first on the MR table. A multichannel torso phased-array coil was placed over the pelvis. A dedicated bilateral lower extremity phased-array coil was used for coverage of the thighs, calves, and feet.
MRI Technique
All examinations were performed with a 1.5-T MR system (Magnetom Symphony,
Siemens Medical Solutions) equipped with quantum gradients (30 mT/m gradient
strength). The procedure was composed of two sessions and, hence, named
"hybrid" (Figs. 1A,
1B).
First, 3D CE-MRA of the calf and pedal vessels was performed using the "sagittal slabs" technique first described in this investigation. Second, three-station automated moving-table bolus chase 3D CE-MRA was performed using the "coronal slabs" technique previously reported by Ho et al. [6, 16, 17] and Meaney et al. [18], the technique that is used as a standard at most institutions.
First session: sagittal acquisition 3D CE-MRA of the calf and pedal vessels—For this session, the patient was first centered at the level of the knee joints and multiplane localizer images were acquired. A timing bolus using 2 mL of gadodiamide (Omniscan, Nycomed Amersham) was administered at a rate of 1 mL/sec, and a turbo fast low-angle shot (FLASH) sequence with automatic in-line subtraction (TR/TE, 1,000/1.6; inversion time, 500 msec; flip angle, 20°; subtrahend factor, 3; and acquisition speed, 1 image per second) was performed in the transverse plane at the level of the popliteal arteries. The scan delay time for evaluation of the calves was calculated as follows:
(time to contrast peak + subtrahend factor) + 2 sec
After this, the patient was recentered for imaging the calves and feet, which is equivalent to station III for the bolus chase 3D CE-MRA technique. Two sagittal slabs were positioned to cover the calves and feet bilaterally. The images of the right and left lower legs were obtained simultaneously using four receiver coils. The scanning time was reduced using an integrated parallel acquisition technique (iPAT, Siemens Medical Solutions) with a generalized autocalibrating partially parallel acquisitions (GRAPPA) reconstruction algorithm with an acceleration factor of 2. A single measurement with this technique was used to acquire the unenhanced subtraction mask images. The sequence parameters are shown in Table 1. The field of view was 500 x 157 mm2, resulting in a voxel size of 1.4 x 1.0 x 1.0 mm3. The acquisition time was 16 sec for two sagittal slabs or 8 sec per slab. The bandwidth used was 610 Hz/pixel.
|
Subsequently, the same sequence was repeated twice back-to-back after injection of gadolinium chelate (18 mL at a rate of 1 mL/sec). The scan delay, which was calculated as described earlier, ensured that the bulk of contrast agent arrived at the mid calves as the center of the k-space was being acquired during the sequence. For each leg, both sets of contrast-enhanced images were subtracted from unenhanced images, and sagittal maximum intensity projections (MIPs) were generated for both subtracted image sets.
Second session: standard bolus chase 3D CE-MRA of the abdomen, pelvis, and lower extremities—The second session is standard bolus chase MRA and was performed immediately after the first session. Initially, multiplane scout images of the calves (station III), thighs (station II), and lower abdomen and pelvis (station I) were acquired. Then, both coronal breath-hold and axial non-breath-hold true fast imaging with steady-state free precession images of the abdominal aorta were acquired to provide anatomic assessment of the outer wall of the vessel and as an additional set of localizers to position the subsequent CE-MRA slab. A timing bolus sequence (turbo FLASH; 1,000/1.6; inversion time, 500 msec; flip angle, 20°; subtrahend factor, 3; and acquisition speed, 1 image per second) of the abdominal aorta was performed in the sagittal plane using a test injection of gadolinium chelate (2 mL at a rate of 1.5 mL/sec). The scan delay for this session was again calculated using the following formula:
|
|
|
|
|
|
where Tcp is the time to contrast peak, SF is the subtrahend factor, and Tks is the time to the center of the k-space.
Three rapid sequential 3D FLASH MRA scans (mask images) were acquired in the coronal plane with the automatic moving table at the abdomen and pelvis level (station I), the thigh level (station II), and the calf and foot level (station III). The imaging parameters for station III are shown in Table 1. The k-space acquisition was linear for stations I and II, but elliptic-centric reordered for station III. The station I scan was obtained while the patient was breath-holding, whereas stations II and III scans were obtained while the patient was breathing freely. The acquisition times for station I was 18 sec; station II, 16 sec; and station III, 23 sec. The field of view for station III was 500 x 357 mm2, resulting in a voxel size of 1.6 x 1.1 x 1.5 mm3. The bandwidth used was 300 Hz/pixel. All three of these scans were obtained without contrast material, to be used next as subtraction masks.
The same three rapid sequential 3D FLASH MRA scans were obtained with the moving table while gadolinium chelate was injected in a biphasic manner (18 mL at a rate of 1.5 mL/sec and 16-20 mL at a rate of 0.4 mL/sec) followed by injection of 15-25 mL of normal saline at a rate of 0.4 mL/sec. The scan delay calculated earlier ensured that the bulk of the contrast agent arrived at the abdominal aorta as the center of the k-space was acquired during the sequence. For each of the three stations, the contrast-enhanced images were subtracted from the unenhanced images, and coronal MIP images were generated from each subtracted image set.
Image Evaluation
We evaluated the MRA images of the calf and foot station (station III)
acquired using the standard bolus chase technique, and we compared and
contrasted those images with dedicated 3D sagittal acquisitions of the calf
and pedal vessels using the HyPer 3D CE-MRA technique. The vessel clarity and
degree of venous contamination were assessed (Figs.
2A,
2B,
2C,
2D,
2E,
2F,
3A,
3B and
3C).
|
|
|
Analysis of venous contamination was based on a 3-point scale; 1, none; 2, present but does not interfere with diagnostic assessment; and 3, present and interferes with diagnostic assessment.
Statistical Analysis
To assess the diagnostic value of dedicated 3D sagittal acquisition of the
calf and pedal vessels using the HyPer 3D CE-MRA technique, the number of
arterial segments (n = 12) with a score that indicated diagnostic
image quality (i.e., score of 3 or 4) was summed separately for each leg. The
resulting sums were compared between both techniques using the paired
Student's t test and Wilcoxon's signed rank test to determine any
significant differences.
For the evaluation of venous contamination, the number of legs that had venous contamination (score = 1 or 2) and the number of legs for which the diagnoses were compromised by venous contamination (score = 3) were calculated for both the HyPer 3D CE-MRA and the standard bolus chase techniques.
Statistical analysis was performed with JMP statistical software (SAS Institute).
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Arterial Evaluation
In the following analysis, vessels that were scored more than 2 were
defined as diagnostic. Of the 53 patients, 99 lower extremities (1,188
arterial segments) were evaluated. We found that the HyPer 3D CE-MRA technique
yielded diagnostic-quality images of significantly more vessels than the
standard bolus chase technique for both the right and left legs.
For the left lower extremities (n = 48), the mean number ±
SD of diagnosed arterial segments was 9.2 ± 2.3 for HyPer 3D CE-MRA and
7.1 ± 4.2 for bolus chase 3D CE-MRA (p 0.0004). For the
right lower extremities (n = 51), the mean number ± SD of
diagnosed arterial segments was 9.4 ± 2.2 for HyPer 3D CE-MRA and 7.6
± 3.5 for bolus chase 3D CE-MRA (p 0.0005).
Selective analysis of the dorsalis pedis arteries showed that the number of diagnostic vessels (score = 3 or 4) was 62 (62.6%) of 99 for HyPer 3D CE-MRA and 13 (13.1%) of 99 for bolus chase 3D CE-MRA.
Venous Evaluation
With the standard bolus chase technique, 83 of 99 lower extremities (43
right, 40 left) had venous contamination and the diagnosis was compromised in
24 of 99 lower extremities (12 right, 12 left). Using the HyPer 3D CE-MRA
technique, four of 99 lower extremities (one right, three left) had minimal
venous contamination and none had compromised diagnosis.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To achieve high-spatial-resolution peripheral MRA without venous contamination, investigators have previously proposed several approaches. Korosec et al. [25] introduced 3D time-resolved imaging of contrast kinetics (TRICKS), which enables rapid reconstruction of 3D data sets and eliminates the need for timing acquisitions or triggering methods. Initial reports of sensitivity and specificity for the detection of occlusion and stenosis were high compared with conventional angiography, and diagnostic quality was higher than that of single-injection bolus chase MRA images [12, 26]. However, with the 3D TRICKS technique, there is still some reduction of spatial resolution in exchange for temporal resolution when compared with standard 3D acquisitions [26, 27]. This is because fewer high-spatial-frequency data steps are sampled, resulting in edge information compromise. Subsequently, Vigen et al. [28] introduced the undersampled projection reconstruction time-resolved imaging of contrast kinetics (PR-TRICKS) and Du et al. [29] reported PR-hyperTRICKS MRA of distal runoff vessels. These techniques are all modifications from 3D TRICKS to increase spatial resolution by acquiring more high-frequency data without increasing the acquisition time. However, none of those techniques allows near isotropic resolution. More recently, Jeffrey et al. [30] introduced WakiTrak (wide aperture kinematic table imaging with isotropic resolution) in which 3D CE-MRA is performed with parallel imaging (sensitivity encoding [SENSE]) in the upper station and relatively long imaging acquisition in the calf and foot station, with submillimeter isotropic resolution.
Kalle et al. [15] introduced the hybrid technique: a two-stage procedure in which MRA images of the calf and foot segments were obtained in the coronal plane after a first injection followed by acquiring images of the aortoiliac and femoral segments after a second injection. This technique allows high-resolution imaging of the calves and feet (voxel size=1.4 x 1.0-1.4 x 0.7-1.3 mm3) with low evidence of venous contamination. However, because of the relatively long acquisition time (16 sec in calves and 24 sec in feet), only one image set can be obtained during preferential arterial enhancement. This might be problematic in patients who have differential flow between both legs.
In this study, we describe HyPer 3D CE-MRA using sagittal acquisition with parallel imaging in the calf and foot station. Our data show that this technique offers higher diagnostic quality than the bolus chase 3D CE-MRA technique because it provides higher spatial resolution in the calf and foot station, eliminates venous contamination, and allows full visualization of the pedal arteries.
To perform time-resolved CE-MRA, spatial resolution is generally traded for high temporal resolution. In this technique, the inherent trade-off is overcome by using parallel imaging (iPAT, Siemens Medical Solutions). With parallel imaging, it is technically possible to acquire high-resolution images in a short acquisition time by using multiple receiver coils and acquiring fewer lines of k-space. In our study, we used four dedicated peripheral coils and scanned both calves and feet simultaneously with two sagittal slabs. Because the sagittal slab acquisition is more anatomically suitable for calves and feet than the coronal slab acquisition, the phase field of view of the calves and feet station can be significantly reduced. In addition, sagittal orientation allows more visualization of the pedal vessels, particularly the dorsalis pedis artery, than does the coronal slab acquisition. The reduction in the field of view also contributes to a reduction in acquisition time and higher-spatial-resolution images. With all these mentioned factors, we were able to obtain images of near isotropic resolution (1.4 x 1.0 x 1.0 mm3) that would be of benefit for more reliable imaging of the peripheral arteries, particularly when the vessels are diseased and small in caliber.
Theoretically, with the use of parallel imaging, the signal-to-noise ratio (SNR) is inversely proportional to the square root of the acceleration factor. However, the effect of SNR loss is minimal in CE-MRA because of the intrinsically high contrast nature of the technique. The application of parallel imaging to speed up the imaging time in peripheral CE-MRA requires extra caution to avoid risk of data acquisition before peak enhancement, especially in patients who have slow blood flow or patients who have differential flow between both legs. We overcome this problem by imaging the calf and foot station twice after contrast administration to capture the best arteriographic phase without venous contamination. In instances in which the contrast material in one lower extremity peaked earlier than the other, the timing delay was calculated for the early vessel. This allowed best visualization of the early leg on the first data set and the other leg with delayed flow on the second data set.
There are two drawbacks for our technique. These are, first, increasing the
average MR table time by approximately 10 min—including patient
positioning, test bolus, and scanning time—and, second, the potential
for more contrast agent (
60 mL or triple dose) to be used than with
conventional bolus chase CE-MRA (40-60 mL or double dose to triple dose).
However, the risk of an unsatisfactory study leading to a repeated examination
is significantly reduced and is almost eliminated. In addition, the capability
to evaluate pedal vasculature is beneficial in patients with severe tibial
disease who are candidates for distal bypass graft surgery. These advantages
help provide more appropriate treatment in a timely fashion.
One limitation of our study is that conventional angiography was not performed in any of the subjects, so there was no standard of reference to validate discrepant findings. At our institution, almost all cases of peripheral vascular disease are evaluated using CT angiography or MRA on a routine clinical basis and surgical decisions are based on these findings. Hence, these patients do not undergo a conventional angiography procedure. Another limitation is that we did not compare sagittal slab HyPer 3D CE-MRA with other time-resolved 3D CE-MRA techniques such as 3D TRICKS, PR-TRICKS, or WakiTrak in this study. Keep in mind that at the time of this study, 3D TRICKS, PR-TRICKS, and WakiTrak remained vendor-specific.
In conclusion, HyPer 3D CE-MRA is an alternative method for time-resolved high-resolution peripheral CE-MRA using sagittal slabs to provide images of near isotropic resolution of the calves and feet without venous contamination, thereby allowing full visualization of the pedal vasculature.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  H. Ersoy and F. J. Rybicki MR Angiography of the Lower Extremities Am. J. Roentgenol., June 1, 2008; 190(6): 1675 - 1684. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Bosch, K.-F. Kreitner, M. F. Peirano, S. Thurner, K. Shamsi, and E. C. Parsons Jr. Safety and Efficacy of Gadofosveset-Enhanced MR Angiography for Evaluation of Pedal Arterial Disease: Multicenter Comparative Phase 3 Study Am. J. Roentgenol., January 1, 2008; 190(1): 179 - 186. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
