| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 All authors: Department of Radiology, Weill Medical College of Cornell University, 416 E 55th St., Cornell MRI, New York, NY 10022.
Received May 14, 2004;
accepted after revision July 21, 2004.
Address correspondence to H. L. Zhang.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. Time-resolved MR angiography of calf and pedal arteries was performed on 59 symptomatic legs of 52 patients using the head coil and bolus injections of 6 mL of gadolinium contrast medium. Selective X-ray digital subtraction angiography was performed within 30 days after MR angiography. Calf and pedal arteries were divided into 10 segments. X-ray digital subtraction angiography and MR angiography images were retrospectively interpreted by three expert observers, who graded segments as having no significant stenosis, significant stenosis (> 50%), or occlusion. The accuracy of MR angiography interpretations was compared with the accuracy of consensus X-ray digital subtraction angiography interpretations as the standard of reference. Arterial segments with discrepant grading on X-ray digital subtraction angiography and MR angiography were reviewed again to determine the reasons for disagreement.
RESULTS. Arterial phase MR angiography images free of venous
contamination were obtained in every case. The agreement between MR
angiography and X-ray digital subtraction angiography in depicting
infrapopliteal arterial disease was fair to good (
= 0.44–0.92).
Overall sensitivity and specificity were 83% and 87%, respectively, for
detecting significant stenosis of calf and pedal arteries and 86% and 93%,
respectively, for detecting occlusions. Accuracy was higher in the larger
vessels—for example, calf (84%) compared with foot (71%). In 21%
(22/105) of the segments graded differently on MR angiography than on X-ray
digital subtraction angiography, it was believed that MR angiography was more
likely to be correct than X-ray digital subtraction angiography because of
visualization of late-filling arteries on MR angiography that did not opacify
on X-ray digital subtraction angiography.
CONCLUSION. Time-resolved 2D projection MR angiography accurately evaluates calf and pedal arteries without degradation from venous contamination.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
One way to eliminate venous contamination in the calf is with time-resolved MR angiography. This requires fast imaging repeated over and over during the arrival of the contrast bolus. Two-dimensional projection MR angiography accelerates data acquisition by eliminating slice encoding, allowing images to be obtained every 1–2 sec with high in-plane spatial resolution but with only a single projection that cannot be reformatted into other obliquities. We hypothesize that this technique may be as suitable for infrapopliteal imaging as it is for conventional X-ray digital subtraction angiography, which also is a projection technique. Furthermore, the high temporal resolution ensures that at least one pure arterial phase image is obtained for every calf without the need for bolus timing even when flow rates differ between right and left legs.
Because of these benefits, we routinely use 2D projection MR angiography for evaluating infrapopliteal arteries in our peripheral MR angiography examinations. In this study, the accuracy of 2D projection MR angiography was assessed by comparison with X-ray digital subtraction angiography in those patients who underwent selective X-ray digital subtraction angiography within 30 days of MR angiography as part of a diagnostic or interventional study.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Imaging Technique
MR angiography.—All MR angiography examinations were
performed on a 1.5-T MR scanner (Signa Horizon Echo Speed, GE Healthcare)
using a head coil for signal transmission and reception. A 2D fast
gradient-echo sequence was used with approximately 2-sec temporal resolution.
Imaging parameters were as follows: TR/TE, 9/2; flip angle, 50°; number of
excitations, 1; slice thickness, 50–160 mm; field of view, 30–32
cm; matrix, 256 x 192; and 2D in-plane spatial resolution, 1.2 x
1.6 mm.
Image acquisition began simultaneously with injection of gadolinium-based contrast medium (gadopentetate dimeglumine, Magnevist, Berlex Laboratories; or gadodiamide, Omniscan, Nycomed Amersham) at a rate of 2 mL/sec via a 20- to 22-gauge angiocatheter followed by 30 mL of saline flush at the same infusion rate. For the initial 21 patients, the dose ranged from 5 to 7 mL, according on the instructions of the radiologist supervising the case. In the subsequent 31 cases, the dose was standardized at 6 mL for every patient. A coronal slab was prescribed to cover from just above the knees to include all the major arteries in the calf. An image before contrast arrival was used as the mask for complex subtraction from the subsequent image data. Then, the head coil was moved down to image the feet in two sagittal slabs with similar parameters using another 6-mL bolus injection of gadolinium.
The inflow vessels including the aorta and iliac and femoral arteries were evaluated with 3D bolus-chase MR angiography after the 2D projection imaging of the runoff vessels [9].
X-ray digital subtraction angiography.—A 4- or 5-French catheter or sheath was positioned in the femoral artery of the leg being treated, usually via a contralateral retrograde common femoral artery puncture and occasionally using an ipsilateral antegrade access. Selective digital subtraction radiographic images were obtained of all 59 extremities from the thigh to the feet with use of multistation imaging (Integris V-3000, Philips), with a 17- to 31-cm field-of-view image intensifier and a 1,024 x 1,024 display matrix. For these selective injections, 15–30 mL of 30% diatrizoate meglumine (Reno-Dip, Bracco Diagnostics) or iohexol 300 (Omnipaque, Nycomed Amersham) was administered at a rate of 5–7 mL/sec. For antegrade femoral arterial puncture, the injection was in the superficial femoral artery as proximal as possible. For contralateral puncture, the injection was in the common femoral artery. Anteroposterior and lateral views were chosen by the attending radiologist to best depict the arteries. At the time of the procedure, the same radiologist selected several arterial phase images from each diagnostic series to be printed to film for subsequent interpretation in this study.
Image Analysis
For each extremity, three blinded expert reviewers who were experienced in
both MR angiography and X-ray digital subtraction angiography independently
analyzed 10 native arterial segments for each examination (Figs.
1A and
1B): below-knee popliteal,
upper anterior tibial, lower anterior tibial, tibioperoneal trunk or proximal
peroneal, distal peroneal, proximal posterior tibial, distal posterior tibial,
proximal dorsalis pedis, dorsalis pedis below the lateral tarsal, and lateral
plantar.
|
|
All patient-identifying information on the sheets of film was covered with thick black tape, and each case assigned a four-digit code for data collection purposes. Image quality of calf MR angiography was graded for each vascular segment as 1 (unable to evaluate adequately), 2 (less than ideal but adequate for diagnosing significant lesions), or 3 (good depiction of significant lesions or normal segment). For each segment that was not well depicted, reviewers indicated whether there was inadequate contrast medium, motion artifact, overlapped vessel, inadequate projection, or inadequate anatomic coverage. Overlap was defined as superimposition of the artery being evaluated on another opacified arterial segment, making interpretation of its luminal caliber difficult. Inadequate projection was defined as the artery not being sufficiently unfolded to be completely evaluated. The degree of stenosis was graded as no significant stenosis, significant stenosis (> 50%), or occlusion. Severity of stenoses was determined by comparing the stenosis diameter with the luminal diameter at an adjacent normal-caliber arterial segment.
X-ray digital subtraction angiography data were graded for image quality and severity of disease using the same grading system by the same blinded reviewers. An interval of at least 3 months elapsed between review of MR angiography data and subsequent review of X-ray digital subtraction angiography data to prevent recognition of cases.
After the individual and independent grading of the MR angiography and X-ray digital subtraction angiography studies had been completed by all three reviewers, all images were reviewed again by the three reviewers together and a consensus grade determined. These data were collated to determine a single consensus grade for any vascular segment that had interobserver disagreement. Arterial segments with discrepancies between X-ray digital subtraction angiography and MR angiography consensus readings were reviewed again to determine why they had been graded differently.
Statistical Analysis
Statistical analyses were performed using SPSS version 11.0 (Statistical
Package for the Social Sciences). Sensitivity and specificity (including 95%
confidence intervals) of 2D projection MR angiography for the diagnosis of
significant lesions (defined as stenosis > 50% or occlusion), and for the
diagnosis of occlusions, were determined by using the consensus X-ray digital
subtraction angiography readings as the gold standard. Unweighted Cohen's
kappa value was calculated for the correlation of consensus MR angiography
readings with consensus X-ray digital subtraction angiography for each
arterial segment.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Image Quality
Arterial phase images free of venous contamination were obtained in every
case. The popliteal artery was well depicted with MR angiography in virtually
all patients and had the highest image quality score, 2.97
(Table 2). The image quality of
the infrapopliteal artery segments in the calf and feet was substantially
lower, 2.68 and 2.45, respectively (p < 0.001 for both),
presumably reflecting the difficulty in imaging small arteries and foot
arteries, which are more prone to image degradation from motion artifacts.
This difficulty was reflected by the blinded reviewers' identifying motion
artifacts in 18–26% of pedal artery segments, compared with a maximum of
15% of calf arteries and 2% of popliteal arteries. Pedal arteries also tended
to be degraded by inadequate contrast medium (7–14%) and inadequate
projection (11–12%).
|
Accuracy of Stenosis Grading
Unweighted Cohen's kappa value was 0.44–0.92, indicating fair to good
agreement between X-ray digital subtraction angiography and MR angiography.
Interobserver disagreement occurred in 20% (n = 110) of the arterial
segments on X-ray digital subtraction angiography versus only 16% (n
= 85) on MR angiography (p < 0.05), reflecting the imperfect
nature of X-ray digital subtraction angiography as a gold standard of
reference. The sensitivity and specificity, including 95% confidence
intervals, for time-resolved 2D projection MR angiography in diagnosis of
significant stenosis and occlusion using consensus X-ray digital subtraction
angiography as the gold standard are shown in Tables
3 and
4, respectively.
Table 5 shows the accuracy of
MR angiography in diagnosis of infrapopliteal artery disease, including
arteries in the calf and foot. MR angiography accurately depicted significant
stenosis in 84% of the calf arteries and 71% of the pedal arteries.
|
|
|
A second review of images of the 105 segments that had grading discrepancies between MR angiography and X-ray digital subtraction angiography (Figs. 6A, 6B, 7A, 7B, 7C, and 7D) showed that the difference in interpretation was most commonly due to borderline lesions (including 40–60% stenoses or partial occlusions); for 43 segments, these lesions easily could have been graded either way. For these borderline lesions, MR angiography underestimated the severity of disease in 72% (31/43) of arterial segments. Additional reasons for discrepancies in grading between MR angiography and X-ray digital subtraction angiography are listed in Table 6. In 22 segments, it was believed that MR angiography was more likely to be correct than X-ray digital subtraction angiography. This was the case when X-ray digital subtraction angiography was of poor quality (n = 3) and when the bolus timing differed between MR angiography and X-ray digital subtraction angiography, with MR angiography tending to identify vessels that never opacified adequately on X-ray digital subtraction angiography (Figs. 7A, 7B, 7C, and 7D).
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Over the past decade, MR angiography has been accepted increasingly for imaging peripheral vascular disease with high accuracy and great safety because an arterial catheter is not required and because MRI contrast agents are not nephrotoxic [9–13]. Contrast-enhanced MR angiography, however, has been limited in the infrapopliteal region because of enhancement of veins that overlap and obscure the arteries [14–16]. Venous contamination has been particularly problematic in patients with diabetes and in patients with severe peripheral vascular disease with associated cellulitis and ulceration that induces early venous enhancement.
These data from 52 patients with 59 symptomatic extremities show that time-resolved MR angiography tracks the passage of gadolinium contrast medium through the infrapopliteal arteries, producing high-quality arterial phase images with a high degree of accuracy. Overall sensitivity and specificity for 2D time-resolved projection MR angiography in detecting significant stenosis was 84% and 94%, respectively, for the calf and 79% and 71%, respectively, for the foot using X-ray digital subtraction angiography as the gold standard of reference. For diagnosis of arterial occlusion, even higher sensitivity and specificity were achieved for both calf and foot (89% and 95%, respectively, for the calf and 79% and 86%, respectively, for the foot). As expected, arteries in the calf were more accurately depicted than arteries in the foot because of difficulty in resolving the small pedal arteries with MR angiography and because of greater motion in the foot, which degrades mask subtraction.
Although X-ray digital subtraction angiography is invasive, involves ionizing radiation and iodinated contrast material, and increases patient risk and recovery time, it has been considered the gold standard for evaluating infrapopliteal arteries in patients with peripheral artery disease before attempted revascularization. In this study, higher interobserver agreement for MR angiography than for X-ray digital subtraction angiography (84% vs 80%) suggests that X-ray digital subtraction angiography is not a perfect gold standard. On a second review of images for which discrepancies between MR angiography and X-ray digital subtraction angiography had been noted, it was believed that MR angiography was more likely to be the correct image in 21% of discrepant segments. MR angiography tended to have a longer bolus duration that allowed more time for filling of infrapopliteal arteries, especially when proximal occlusive disease forced arterial enhancement to occur in a retrograde fashion. This means that the real accuracy of 2D time-resolved MR angiography is higher than what is reported here. However, overall, image quality still was judged to be better on X-ray digital subtraction angiography than on MR angiography.
These data compare favorably with studies using 2D time-of-flight imaging [17] and confirm the promising results reported for earlier studies evaluating time-resolved 2D projection MR angiography [18] and 3D time-resolved imaging of contrast kinetics (TRICKS) [19]. An important advantage of 2D projection MR angiography is that it uses much faster acquisitions and automated processing, which minimize motion artifacts and speed physician interpretation. These data also compare favorably with studies using the three-station bolus-chase technique, which shows limited accuracy in the calf, commonly because of venous contamination. Ho et al. [20] showed 79–100% sensitivity and 80–88% specificity for calf arteries. Cronberg et al. [11] showed 92% sensitivity and 64% specificity for lower leg and pedal arteries with triple-dose gadolinium contrast medium and a 512 matrix. These findings emphasize the benefit of time-resolved MR angiography, which guarantees arterial phase images free of venous contamination for the infrapopliteal vessels. Additional methods of reducing venous contamination include faster imaging using sensitivity encoding [21] and application of subsystolic thigh compression using blood pressure cuffs [22–24].
Interestingly, the upper anterior tibial artery was an area of decreased accuracy, possibly reflecting its more anterior proximal trajectory, making full evaluation on a single anteroposterior projection more difficult. This area is difficult to evaluate on X-ray digital subtraction angiography as well, likely because of overlapping bone on subtractions, and likely decreasing the accuracy of X-ray digital subtraction angiography. Also, many of the 2D images of the upper anterior tibial artery were coronal. Sagittal images would likely depict the anterior tibial arteries better by unfolding the proximal segment.
The pattern of reduced accuracy in smaller arteries (especially pedal arteries) suggests that higher resolution MR angiography may improve accuracy. Additional important reasons for lower accuracy in the pedal arteries included motion artifacts and inadequate projections for optimally unfolding the complicated course of pedal arteries. For 2D projection MR angiography, improvement in postprocessing techniques that reduce or eliminate motion artifacts may be helpful. More liberal use of patient restraints and repeated acquisitions when motion is detected, as is done with X-ray digital subtraction angiography, might also improve 2D projection MR angiography pedal imaging. In addition, techniques that share peripheral k-space from multiple time points may improve accuracy for small distal vessels by increasing the resolution and extending the data set into the third dimension to allow reformatting to multiple projections. These techniques include 3D TRICKS [19, 25], vastly undersampled isotropic projection reconstruction (VIPR) [26], and 3D spiral with sliding window reconstruction [27]. It also may be useful to supplement time-resolved NRA of the foot with a longer high-spatial-resolution 3D acquisition [28] with contrast bolus timing based on 2D projection MR angiography data.
Other limitations of time-resolved 2D projection MR angiography include overestimation of the degree of stenosis and the presence of susceptibility artifacts from metallic surgical clips and stents (Figs. 5A and 5B) and motion. The overestimation is a general problem with both 2D and 3D MR angiography and is less likely to be resolved even with high-spatial-resolution imaging. The metal artifacts were minimal in our experience and possibly will lessen as improvements in gradient performance lead to shorter TEs. Motion limits the effectiveness of background subtraction. Motion was greater in the feet, which have more degrees of freedom than do the calves. Motion artifacts may be reduced by improved foot immobilization within the coil and by techniques that eliminate or correct motion-corrupted data [29].
One important limitation of this study was our practice of performing X-ray digital subtraction angiography primarily on patients with positive peripheral MR angiography findings who require an interventional procedure. For this reason, X-ray digital subtraction angiography always followed MR angiography. In addition, some bias may have been introduced by failing to include patients with only minimal disease or surgical disease that did not require a catheter-based interventional procedure.
Currently, in our institution, treatment for peripheral vascular disease is performed using MR angiography alone in almost all cases. Only for patients with a pacemaker, severe claustrophobia, or other contraindications to MRI does peripheral vascular imaging begin with a diagnostic arteriogram. Our complete MRI protocol starts with 2D projection MR angiography using 6 mL of gadolinium followed by 3D bolus-chase MR angiography. This protocol combines the benefits of both time-resolved and high-resolution 3D imaging for two opportunities at adequately visualizing infrapopliteal arteries.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  C. Y. Kim, R. A. Mirza, J. A. Bryant, E. D. Whiting, D. M. Delong, C. E. Spritzer, and E. M. Merkle Central Veins of the Chest: Evaluation with Time-resolved MR Angiography Radiology, May 1, 2008; 247(2): 558 - 566. [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]
|
|
|
|
 |
|
 M. B. Blum, M. Schmook, R. Schernthaner, G. Edelhauser, S. Puchner, J. Lammer, and M. A. Funovics Quantification and Detectability of In-Stent Stenosis with CT Angiography and MR Angiography in Arterial Stents In Vitro Am. J. Roentgenol., November 1, 2007; 189(5): 1238 - 1242. [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: A Report of the American College of Cardiology Foundation/American Heart Association/American College of Physicians Task Force on Clinical Competence and Training Developed in Collaboration With the Society of Atherosclerosis Imaging and Prevention, the Society for Cardiovascular Angiography and Interventions, the Society of Cardiovascular Computed Tomography, the Society for Cardiovascular Magnetic Resonance, and the Society for Vascular Medicine and Biology J. Am. Coll. Cardiol., September 11, 2007; 50(11): 1097 - 1114. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
