DOI:10.2214/AJR.06.0901
AJR 2007; 188:1246-1254
© American Roentgen Ray Society
Time-Resolved MR Angiography: A Primary Screening Examination of Patients with Suspected Pulmonary Embolism and Contraindications to Administration of Iodinated Contrast Material
Hale Ersoy1,
Samuel Z. Goldhaber2,
Tianxi Cai3,
Tuan Luu1,
Joshua Rosebrook1,
Robert Mulkern4 and
Frank Rybicki1
1 Cardiovascular Imaging Section, Department of Radiology, Brigham and Women's
Hospital and Harvard Medical School, 75 Francis St., ASB I-L1-004, Boston, MA
02115.
2 Division of Cardiovascular Medicine, Brigham and Women's Hospital and Harvard
Medical School, Boston, MA.
3 Department of Biostatistics, Harvard School of Public Health, Boston,
MA.
4 Department of Radiology, Children's Hospital and Harvard Medical School,
Boston, MA.
Received July 24, 2006;
accepted after revision October 11, 2006.
Address correspondence to H. Ersoy
(hersoy{at}partners.org).
Abstract
OBJECTIVE. The purpose of this study was to evaluate the efficiency
and reproducibility of a single-breath-hold time-resolved 3D MR angiographic
technique in the diagnosis of pulmonary embolism.
MATERIAL AND METHODS. Twenty-seven consecutively registered patients
with clinically suspected pulmonary embolism and contraindication to
administration of iodinated contrast agents underwent imaging by time-resolved
3D MR angiography at 1.5 T. Bolus timing was not required. Two reviewers
independently analyzed MR angiograms for overall image quality and evidence of
pulmonary embolism. Additional imaging techniques, including pulmonary
embolism CT angiography, ventilation-perfusion (V/Q) lung scanning, venous
duplex sonography for deep venous thrombosis, and echocardiography for right
ventricular strain, and 30-day and 3-month clinical follow-up were used to
confirm the MR angiographic findings.
RESULTS. Image quality was sufficient for diagnosis in the cases of
98% of lobar, 92-93% of segmental, and 94-95% of all vessel parts from the
main pulmonary artery though the segmental branches with excellent
interobserver agreement. Findings on MR angiography were concordant with the
anatomic distribution of abnormalities for all pulmonary embolism CT
angiographic examinations (n = 2) and four of seven V/Q lung scans.
Screening with time-resolved 3D MR angiography allowed confident exclusion or
inclusion of pulmonary embolism in 96% of patients.
CONCLUSION. Time-resolved 3D MR angiography provides high temporal
resolution (nine phases, one phase per 3.3 seconds) and consistently yields
arterial phase only images. As found with clinical follow-up, confident
diagnosis of pulmonary embolism from the main pulmonary artery through the
segmental branches can be incorporated into a clinical service as a screening
examination of patients with contraindications to the use of iodinated
contrast material.
Keywords: angiography • cardiopulmonary imaging • dynamic MRI • lung • MRI • MRI technique
Introduction
CT of the pulmonary arteries is the standard-of-care imaging test in
the evaluation of patients with clinically suspected pulmonary embolism (PE).
In a small subset of patients, however, especially those with renal
insufficiency (serum creatinine concentration > 1.5 mg/dL) and those with
severe allergic reactions, use of iodinated contrast material is
contraindicated. When such patients need imaging, most centers rely on
ventilation-perfusion (V/Q) lung scans. The results of these scans, however,
are not definitive in the diagnosis of PE. The specificity of lung scanning
can be as low as 10% [1,
2]. When used as the primary
diagnostic examination, V/Q lung scanning often leads to additional testing
that results in delay, additional cost, and the risk of unnecessary
anticoagulation treatment.
Contrast-enhanced MR angiography has been proposed as a noninvasive tool
for the diagnosis of PE, with early reports showing a sensitivity of 68-77%
and a specificity of 95-100%
[3-6].
Patients who have had reactions to iodinated contrast media are at increased
risk of contrast reactions during MRI
[7,
8]. MR angiography, however, is
performed with gadolinium chelates for contrast enhancement. The overall rate
of severe allergic reactions to gadolinium chelates (0.01%)
[9] is small compared with the
rate of severe reactions to ionic and nonionic iodinated contrast materials,
0.16% and 0.03%, respectively
[10]. Therefore, patients with
severe allergies to iodine can be imaged safely with gadolinium-enhanced MR
angiography. Moreover, the nephrotoxicity profile of 0.5 mmol/L gadolinium
chelate given at less than 0.4 mmol/kg is generally considered less than that
of iodinated contrast agents
[11,
12].
Despite its accuracy and safety profile, the routine use of MR angiography
in the evaluation of PE in a specific subset of patients has been limited by
technical and practical factors. Image degradation from respiratory and
cardiac motion is common. In general, respiratory motion is more of a problem
with MRI than it is with state-of-the-art CT scanners, because the average
breath-hold is significantly longer. High spatial resolution is necessary
because of the small diameter of the branch vessels of the pulmonary arterial
tree. High temporal resolution is required to produce arterial phase only
images, avoiding the pulmonary venous enhancement that can obscure evaluation
of the arteries. One reason that pulmonary MR angiography has not been
routinely used is the challenge of obtaining high temporal resolution while
maintaining high spatial resolution.
Just as improvements in CT technology, such as thinner slices, more slices
per gantry rotation, and faster gantry rotation times, have made CT the
mainstay in the diagnosis of PE, advances in MRI technology, in particular
faster gradients, have enabled it to become more robust as a second-line
imaging technique for PE. We evaluated the efficiency and reproducibility of a
single-breath-hold time-resolved 3D MR angiographic technique in imaging of
patients with clinically suspected PE.
Materials and Methods
Patients
Twenty-seven consecutively registered patients (14 men, 13 women; mean age,
62 years; age range, 35-92 years) with clinically suspected PE and a
contraindication to administration of iodinated contrast material were imaged.
The patient records were obtained from our cardiovascular imaging section,
which offers 3D MR angiography as a clinical service. All patients included in
this study were inpatients, including several intensive care patients, with
one or more of the following comorbid conditions: malignancy, trauma,
long-term postoperative immobilization, infection, nephritic syndrome, and
collagen vascular disease. Images and clinical information were
retrospectively reviewed. The protocol was approved by the human research
committee of our institution.
MR Angiography
All MR angiographic examinations were performed with a 1.5-T MRI system
(Signa 11.0, GE Healthcare) with gradients operating at a speed of 40 mT/m.
After placement of a 20-gauge IV angiocatheter in an antecubital vein for
administration of a gadolinium-based paramagnetic contrast agent, the patients
were placed on the imaging table in the supine feet-first position. A standard
eight-channel phased-array coil was used for signal reception. Three-plane
steady-state free precession acquisition (fast imaging with steady-state
acquisition) was used as the locator for prescription of the 3D MR
angiographic imaging volume. For time-resolved 3D MR angiography, we used
elliptic centric time-resolved imaging of contrast kinetics. This sequence is
based on a fast 3D gradient-echo pulse sequence with the specific view-sharing
and temporal interpolation scheme described in Appendix 1
[13-16].
Imaging parameters for time-resolved 3D MR angiography were as follows:
TR/TE, 3.5/1.3; receiver bandwidth, ± 62.5 kHz; number of signals
averaged, 0.5; flip angle, 35°; field of view, 340 mm; phase field of
view, 1; matrix size, 256 x 192. A coronal oblique slab was prescribed
with 30 partitions with effective thickness of 3 mm. The MR angiographic
sequence included two phases with separate breath-holds, unenhanced mask (14.7
seconds), and contrast-enhanced dynamic acquisition (41 seconds) with the same
imaging parameters. The mask acquisition was followed by dynamic imaging with
40 mL of gadopentetate dimeglumine (Magnevist, Berlex) and immediate 20-mL
saline flush at a rate of 3 mL/s. An automated injector (Medrad) was used for
contrast and saline administration. Bolus timing was not required. A standard
scan delay (3 seconds) was used between the beginning of the contrast infusion
and the start of time-resolved 3D MR angiographic acquisition.
Patients were instructed to perform a breath-hold for as long as possible,
and the beginning of imaging was synchronized to the breath-hold. Mask
subtraction was fully automated on the MRI unit. With this MR angiographic
technique, nine phases were acquired in immediate succession with a temporal
output rate of one phase per 3.3 seconds over the 41-second scanning time. The
high temporal resolution necessitated a compromise with respect to spatial
resolution. With a 340-mm field of view, the voxel size was 1.3 x 1.8
x 3 mm. After zero interpolation, the displayed spatial resolution was
0.7 x 0.7 x 1.5 mm. In the clinical service protocol we included
complementary equilibrium phase imaging with a breath-hold 3D T1-weighted fast
gradient-echo pulse sequence with fat suppression (4.6/1.1; receiver
bandwidth, ± 62.5 kHz; number of signals averaged, 0.75; flip angle,
12°; matrix size, 320 x 224) to identify thoracic abnormalities
other than PE that could explain the symptoms.
Image Interpretation and Data Analysis
Two experienced cardiovascular imagers blinded to the clinical information
independently scored the pulmonary arterial tree to the level of the segmental
branches. For analysis, the parts of the pulmonary arterial system were as
follows: main pulmonary artery, right and left pulmonary arteries, five lobar
arteries (right upper, right middle, right lower, left upper, left lower), and
18 segmental branches, for a total of 26 parts. Image interpretation was
performed with a workstation on which source images were viewed and multiplane
reformatted images and maximum intensity projections were made for each of the
nine phases. The MRI finding diagnostic of PE was defined as an arterial
filling defect throughout all phases or as abrupt cutoff of the main or lobar
pulmonary arteries.
For each patient, 26 vessel parts were separately analyzed for overall
image quality, including respiratory motion and venous contamination. Image
quality was rated on a three-point scale: 1, no artifact; 2, artifact present
so that the arterial part was seen but the findings were insufficient for
confident diagnosis or exclusion of PE; 3, artifact substantial enough to
preclude identification of the arterial part. For assessment of image quality
and the presence of PE, the rate of agreement between the two imagers was
determined with the kappa statistic with linear weight
[17]. This value was computed
for the lobar and the segmental arteries separately and for the total 26
vessel parts. A weighted kappa value greater than 0.80 indicated excellent
interobserver agreement. For each patient, both reviewers independently
identified which of the nine phases yielded the best arterial phase 3D data
set.
Clinical and imaging follow-up data were collected through review of
medical and radiologic record. Clinical follow-up included admission and
preadmission diagnoses, comorbid conditions (e.g., malignant disease,
congestive heart failure, pulmonary disease, and infection), clinical course
during the hospitalization, anticoagulant treatment, incidence of new PE, and
mortality within 30 days and 3 months after the initial MR angiographic
study.
View larger version (166K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1 —46-year-old man with retrosternal chest pain and increasing
shortness of breath for past 2 hours. Three-dimensional time-resolved
pulmonary MR angiogram (TE/TR, 3.5/1.3; bandwidth, ± 62.5 kHz; flip
angle, 35°, 30 partitions with effective thickness of 3 mm; matrix size,
256 x 192; scan time, 41 seconds) shows nine temporally resolved phases
acquired with single breath-hold. Fourth phase has best image quality, as was
true for most patients in this study.
|
|
Additional imaging of each of the 27 patients was separated into three
categories: PE CT angiography (CTA), V/Q lung scan, and duplex sonography for
deep venous thrombosis and echocardiography for right ventricular strain. Two
imaging studies of the pulmonary arterial system, PE CTA and V/Q lung scan,
were used to determine concordance with the MR angiographic findings when
these studies were performed within 24 hours of MR angiography. PE CTA is
considered the reference standard in the diagnosis of PE
[18]. All images from both PE
CTA and V/Q lung scanning were reviewed and compared with the radiology
reports. For patients who underwent PE CTA, after correction of a transient
increase in serum creatinine concentration, the MR angiographic findings were
compared with the PE CTA findings vessel part by vessel part. The pretest PE
probability of the V/Q scan was rated low (< 10%), intermediate (10-50%),
moderately high (> 50-90%), or high (> 90%)
[19]. With respect to V/Q
scans, low probability was considered concordant with negative findings on MR
angiography. Moderately high and high probability were considered concordant
with positive findings on MR angiography. V/Q scans with intermediate
probability were classified neither discordant nor concordant
[19]. For patients with
intermediate-probability or indeterminate V/Q scans and those in whom the scan
results conflicted with clinical expectation, results of further
studies—duplex sonography of the leg veins and
echocardiography—were taken into consideration before the patient was
selected for active treatment. Imaging reports from duplex sonography and
echocardiography performed within 1 week of MR angiography were reviewed. None
of the patients underwent digital subtraction angiography.
View larger version (144K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2A —54-year-old man who underwent right antecubital vein injection and
had unknown central venous thrombosis. Fifth phase of acquisition of coronal
time-resolved 3D MR angiogram shows poor enhancement of pulmonary arteries
(open arrows) due to slow venous flow from collateral veins.
|
|
Results
All patients tolerated imaging well, and there were no substantial
complications or adverse events during or after gadolinium administration. All
data sets were eligible for evaluation. Most of our patients had central lines
in place. We did not, however, notice substantial image degradation due to the
metallic susceptibility artifact caused by the central lines. In all
instances, we obtained arterial phase only images without significant
pulmonary venous overlap (Fig.
1). In one of 27 patients, pulmonary arterial enhancement was
significantly delayed until the eighth phase owing to massive central venous
thrombosis and delayed arrival of contrast material into the right atrium
through venous collaterals (Fig.
2A,
2B).
View larger version (158K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2B —54-year-old man who underwent right antecubital vein injection and
had unknown central venous thrombosis. Contrast-enhanced equilibrium phase 3D
fast gradient-echo image shows thrombosis (arrows) of central
veins.
|
|
In the cases of 26 of the 27 patients, both reviewers were in complete
agreement that either the third or the fourth phase of the acquisition was
optimal for evaluation of the arterial tree. For assessment of image quality
(Table 1), both reviewers were
in complete agreement regarding the central (main, right, and left) pulmonary
arteries and found 98% of the vessel parts to have no artifacts. Two segmental
parts were not included within the imaging volume. Image quality was rated
diagnostic by both reviewers for 127 (98%) of 129 lobar arteries. Image
quality was rated diagnostic for 423 (92%) of 462 segmental parts by reviewer
1 and for 429 (93%) of 462 segmental arteries by reviewer 2. For all parts
(main, left, right, lobar, and segmental pulmonary arteries), image quality
was rated diagnostic for 629 (94%) of 670 parts by reviewer 1 and for 635
(95%) of 670 parts by reviewer 2. Interobserver agreement was very good for
lobar arteries, segmental arteries, and all parts
(Table 1).
On the basis of MR angiographic findings, both reviewers identified PE in
four of 27 patients. There was complete agreement regarding the vessel parts
considered to have positive findings. With respect to additional imaging
(Table 2), two of the four
patients with the diagnosis of PE based on MR angiographic findings underwent
PE CTA within 24 hours of MR angiography (Fig.
3A,
3B,
3C). Both reviewers
independently identified PE in identical parts on MR angiography and PE CTA in
both patients. Seven patients underwent V/Q lung scans (four, low probability;
two, intermediate; one, indeterminate). Findings on all four low-probability
V/Q lung scans were concordant with negative MR angiographic findings. Two
patients with intermediate-probability V/Q lung scans had MR angiographic
findings positive for PE. The areas of V/Q mismatch were concordant with the
distribution of emboli diagnosed with MR angiography (Fig.
4A,
4B,
4C,
4D). In a patient with
intermediate-probability V/Q lung scans, MR angiography did not show PE. MR
angiography of this patient showed pulmonary artery dilatation, and
echocardiography showed enlargement of the right ventricle. The patient died
in the hospital 9 days after MR angiography, but an autopsy was not performed.
Two other patients died during the 3-month follow-up period (11% all-cause
mortality). Both of these patients died in the hospital, and both deaths were
attributed to comorbid conditions.
View larger version (142K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3A —49-year-old woman with hypercoagulable state, shortness of breath,
and bilateral leg swelling who underwent MR angiography followed by pulmonary
embolism CT angiography within 24 hours. Coronal 3D MR angiogram (A)
and source image from fourth phase of acquisition (B) show filling
defect in left lower lobe artery (arrow).
|
|
View larger version (131K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3B —49-year-old woman with hypercoagulable state, shortness of breath,
and bilateral leg swelling who underwent MR angiography followed by pulmonary
embolism CT angiography within 24 hours. Coronal 3D MR angiogram (A)
and source image from fourth phase of acquisition (B) show filling
defect in left lower lobe artery (arrow).
|
|
View larger version (133K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3C —49-year-old woman with hypercoagulable state, shortness of breath,
and bilateral leg swelling who underwent MR angiography followed by pulmonary
embolism CT angiography within 24 hours. Coronal reformatted image from CT
angiography confirms presence of pulmonary embolism (arrow) in
anatomic area identical to A and B.
|
|
View larger version (63K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4A —56-year-old woman with chest pain, dyspnea, and lower extremity
edema referred for pulmonary MR angiography. Intermediate-probability
ventilation (A)-perfusion (B) lung scan obtained within 24 hours
of MR angiography shows moderate ventilation-perfusion mismatch (curved
arrow, B) in superior segment of right lower lobe.
|
|
View larger version (43K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4B —56-year-old woman with chest pain, dyspnea, and lower extremity
edema referred for pulmonary MR angiography. Intermediate-probability
ventilation (A)-perfusion (B) lung scan obtained within 24 hours
of MR angiography shows moderate ventilation-perfusion mismatch (curved
arrow, B) in superior segment of right lower lobe.
|
|
View larger version (151K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4C —56-year-old woman with chest pain, dyspnea, and lower extremity
edema referred for pulmonary MR angiography. 3D MR angiographic image shows
persistent partial filling defect (arrow) in right lower lobe
pulmonary artery.
|
|
View larger version (129K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4D —56-year-old woman with chest pain, dyspnea, and lower extremity
edema referred for pulmonary MR angiography. 3D MR angiographic source image
shows abrupt cutoff of superior segmental branch of right lower lobe artery
(open arrow).
|
|
In none of the 20 patients with negative MR angiographic findings was PE
diagnosed within 30 days of the initial scan. During the 3-month follow-up
period, PE developed in one of the 20 patients. This patient was undergoing
anticoagulation for a known hypercoagulable state and atrial fibrillation.
Among the 24 patients who survived the follow-up period, 14 avoided long-term
anticoagulation. Among the 10 patients undergoing anticoagulant therapy, four
had PE, four had deep venous thrombosis, and two had atrial fibrillation.
Discussion
Although the spatial resolution of time-resolved 3D MR angiography is less
than that of PE CTA, confident diagnoses were made at the main, lobar, and
segmental levels. The accuracy of these diagnoses was confirmed with clinical
follow-up. In most cases, this clinical information, even without evaluation
of subsegmental arteries, has significant implications for patient care,
particularly when use of iodinated contrast material is contraindicated. The
70% of patients with MR angiographic findings negative for PE avoided
long-term anticoagulation therapy and the associated complications.
Time-resolved 3D MR angiography as a primary screening examination enabled
confident exclusion or inclusion of PE in 96% (26/27) of patients, emphasizing
the utility of this technique in imaging of patients who would otherwise
undergo PE CTA. In two patients who underwent PE CTA, imaging findings were
identical to those of MR angiography. In four of seven patients who underwent
V/Q lung scanning, the results were concordant with those of MR angiography.
Because of the small number of patients who underwent both V/Q scanning and MR
angiography, inferences are difficult. Because the pretest probability of an
intermediate scan is 10-50%
[19], the definition that an
intermediate-probability V/Q scan is concordant with positive or negative
findings on MR angiography is subject to debate.
Higher spatial resolution may, in theory, be achieved at the expense of
temporal resolution. Theoretic improvement in spatial resolution without
compromising temporal resolution may be achieved by incorporating parallel
imaging techniques [20,
21]. Work in the near future
will explore the benefits of incorporating parallel imaging.
View larger version (157K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 5B —57-year-old man with chest pain, shortness of breath, and known
renal cell carcinoma. Contrast-enhanced equilibrium phase 3D fast
gradient-echo image shows tumor plaque (arrow) in bronchus
intermedius as result of direct invasion through subcarinal metastatic renal
cell carcinoma. Atelectasis (open arrows) of right middle and right
lower lobes also is evident.
|
|
The MR angiography protocol described is routinely used on our clinical
service. It is designed to minimize the risk of human error on the part of the
large number of technologists needed to perform the examination. Bolustiming
technique is not used because the pulse sequence is inherently insensitive to
timing error. However, central venous obstruction (i.e., between the IV access
site and the pulmonary arterial system) can degrade image quality (Fig.
2A,
2B). In patients with known
central venous obstruction, fluoroscopic triggering can be used for
visualization of the transit of contrast material through the right atrium,
right ventricle, and main pulmonary artery.
The magnet gradient time (i.e., excluding patient preparation) was
approximately 10 minutes. This imaging time was considerably long compared
with that of PE CTA, which takes less than 1 minute for data acquisition.
Imaging time is a potential disadvantage of PE MR angiography, particluarly
due to the patient population that can be severely ill, dyspneic, and in
hemodynamically unstable condition. However, none of our patients was unable
to tolerate the study, primarily because of adequate patient preparation and
monitoring.
Gadolinium-enhanced CTA has been suggested as an alternative to PE CTA with
iodinated contrast agents. However, high-quality gadolinium-enhanced CTA
requires the use of at least 16-MDCT technology
[11]. In addition, there is
still considerable debate over the relative nephrotoxicity of gadolinium
chelates versus iodinated contrast media. For equivalent X-ray attenuating
doses, gadolinium may be more nephrotoxic than the iodinated agents. Although
gadolinium-enhanced CTA has promise, we consider its use controversial and
believe equivalent or superior images can be obtained with MR angiography
[22].
The weakness of this study was the lack of a reference standard for the
diagnosis of PE. Only four of 27 patients underwent PE CTA within 24 hours
before or after MR angiography. Three of the PE CTA studies were performed
with an iodinated contrast agent and one with a gadolinium chelate. In
addition, as is the case for PE CTA, alternative diagnoses are important. The
protocol also included a contrast-enhanced 3D T1-weighted fast gradient-echo
sequence with fat suppression. This sequence was performed after time-resolved
imaging and was primarily used for identifying other thoracic abnormalities to
explain the patient's symptoms (Fig.
5A,
5B). We recognize that
pathologic findings depicted on MRI are suboptimal in comparison with those
depicted on CT.
The typical patients referred for PE CTA or MR angiography had dyspnea and
chest pain in addition to other comorbid conditions. Patients were not
expected to perform a breath-hold for 41 seconds (the maximum time for data
acquisition). Instead, patients were instructed to performed a breath-hold for
as long as possible and then to breathe shallowly until the end of
acquisition. Because of the temporal interpolation scheme (Appendix 1), some
respiratory motion was expected. The negative effect on image quality was
minimal, if any. The arterial phase only image quality remained high in the
third and fourth phases because arterial phase images were acquired before the
cumulative effects of respiratory motion became relevant. We attribute this
phenomenon to the specific interpolation algorithm discussed in Appendix
1.
The technique of time-resolved 3D MR angiography of the pulmonary arteries
used in this study provides high temporal resolution (nine phases, one phase
per 3.3 seconds) and consistently yields pure arterial phase images without
bolus injection timing. The acquired spatial resolution enables confident
diagnosis of PE from the main pulmonary artery through the segmental branches
and can be incorporated into routine clinical workflow. This management
strategy using MR angiography as a primary screening examination allows rapid
exclusion of PE in patients with contraindications to the use of iodinated
contrast agents.
APPENDIX 1: Segmentation and View-Sharing Algorithm for Angiographic Images
In the 3D elliptic-centric time-resolved imaging of contrast kinetics
(EC-TRICKS) pulse sequence, the k-space is divided into four segments, A, B,
C, and D, all of equal area, in the phase- and slice-encoding dimensions
(Fig. 6). This sequence is a
refinement of the linear 3D TRICKS technique, in which the k-space is divided
into seven equal segments along the slice direction only with three pairs
symmetric on both sides along the phase-encoding (ky) direction
[13,
14]. The refinements allow
faster updating of image contrast enhancement represented by the low-frequency
k-space data (segment A) with less frequent sampling of the static vessel edge
detail (segments B, C, and D). Because contrast enhancement is characterized
mostly by the low spatial components, EC-TRICKS generally provides optimal
arterial-venous separation
[15,
16].
View larger version (17K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 6 —Drawing shows k-space segmentation for elliptic centric phase
ordering for 3D time-resolved imaging of contrast kinetics MR angiographic
acquisition. Segment A represents center of k-space (contrast enhancement).
Segments B, C, and D represent periphery of k-space. kz = slice-encoding
direction, ky = phase-encoding direction.
|
|
The initial mask acquisition without contrast enhancement provides complete
k-space coverage (ABCD). After contrast administration, 12 partitions of
k-space are acquired over two loops. This acquisition is specific to this
protocol. In general, more partitions and loops can be used as long as the
data acquisition memory allows. For each loop, the order of acquisition is
ABACAD. Nine temporal phases are generated from the 12 partitions. The
specific partitions used for each of the nine temporal phases are shown in the
view-sharing algorithm (Fig.
7).
In this scheme, the contribution of the second loop to the first three or
four temporal phases is largely limited to peripheral k-space data. This
limitation may be why respiratory motion during the second loop did not appear
to compromise the arterial phase only images, those important for identifying
pulmonary embolism. On the other hand, misregistration between the mask and
the contrast-enhanced data can lead to severe image degradation because this
mask is used for background signal subtraction.
References
- [No authors listed]. Value of the ventilation/perfusion scan in
acute pulmonary embolism: results of the prospective investigation of
pulmonary embolism diagnosis (PIOPED). The PIOPED Investigators.
JAMA 1990; 263:2753
-2759[Abstract/Free Full Text]
- Hull RD, Raskob GE, Ginsberg JS, et al. A noninvasive strategy for
the treatment of patients with suspected pulmonary embolism. Arch
Intern Med 1994; 154:289
-297[Abstract/Free Full Text]
- Gupta A, Frazer C, Ferguson J, et al. Acute pulmonary embolism:
diagnosis with MR angiography. Radiology1999; 210:353
-359[Abstract/Free Full Text]
- Meaney JF, Weg JG, Chenevert TL, Stafford-Johnson D, Hamilton BH,
Prince MR. Diagnosis of pulmonary embolism with magnetic resonance
angiography. N Engl J Med 1997;336
: 1422-1427[Abstract/Free Full Text]
- Oudkerk M, van Beek EJ, Wielopolski P, et al. Comparison of
contrast-enhanced magnetic resonance angiography and conventional pulmonary
angiography for the diagnosis of pulmonary embolism: a prospective study.
Lancet 2002; 359:1643
-1647[CrossRef][Medline]
- Kluge A, Luboldt W, Bachmann G. Acute pulmonary embolism to the
subsegmental level: diagnostic accuracy of three MRI techniques compared with
16-MDCT. AJR 2006;187
: 127;
[web]W7-W14[CrossRef]
- Nelson KL, Gifford LM, Lauber-Huber C, Gross CA, Lasser TA.
Clinical safety of gadopentetate dimeglumine.
Radiology 1995;196
: 439-443[Abstract/Free Full Text]
- Niendorf HP, Haustein J, Cornelius I, Alhassan A, Clauss W. Safety
of gadolinium-DTPA: extended clinical experience. Magn Reson
Med 1991; 22:222
-228[Medline]
- Li A, Wong CS, Wong MK, Lee CM, Au Yeung MC. Acute adverse
reactions to magnetic resonance contrast media: gadolinium chelates.Br J Radiol 2006; 79:368
-371[Abstract/Free Full Text]
- Cochran ST. Anaphylactoid reactions to radiocontrast media.
Curr Allergy Asthma Rep 2005;5
: 28-31[Medline]
- Remy-Jardin M, Bahepar J, Lafitte JJ, et al. Multi-detector row CT
angiography of pulmonary circulation with gadolinium-based contrast agents:
prospective evaluation in 60 patients. Radiology2006; 238:1022
-1035[Abstract/Free Full Text]
- Runge VM. Safety of magnetic resonance contrast media.
Top Magn Reson Imaging 2001;12
: 309-314[CrossRef][Medline]
- Korosec F, Frayne R, Grist T, Mistretta C. Time-resolved
contrast-enhanced 3D MR angiography. Magn Reson Med1996; 36:345
-351[Medline]
- Mistretta CA, Grist TM, Korosec FR, et al. 3D time-resolved
contrast-enhanced MR DSA: advantages and tradeoffs. Magn Reson
Med 1998; 40:571
-581[Medline]
- Zhou Y. Improving the temporal resolution of 3D time-resolved MRA
with elliptic centric view ordering: a partial Fourier approach. In:
Proceedings of the International Society of Magnetic Resonance in
Medicine. Berkley, CA: ISMRM, 2002
- Zhou Y, Carroll T, Esparza-Coss E, Grist T, Misretta C, Korosec F.
Time-resolved MRA with elliptical centric view reordering. In:
Proceedings of the International Society of Magnetic Resonance in
Medicine. Berkley, CA: ISMRM, 2001
- Fleiss JL. Statistical methods for rates and
proportions, 2nd ed. New York: Wiley,1981
- Schoepf UJ, Goldhaber SZ, Costello P. Spiral computed tomography
for acute pulmonary embolism. Circulation2004; 109:2160
-2167[Abstract/Free Full Text]
- Miniati M, Monti S, Bauleo C, et al. A diagnostic strategy for
pulmonary embolism based on standardised pretest probability and perfusion
lung scanning: a management study. Eur J Nucl Med Mol
Imaging 2003; 30:1450
-1456[CrossRef][Medline]
- Pruessmann KP, Weiger M, Scheidegger MB, Boesiger P. SENSE:
sensitivity encoding for fast MRI. Magn Reson Med1999; 42:952
-962[CrossRef][Medline]
- Fink C, Ley S, Kroeker R, Requardt M, Kauczor HU, Bock M.
Time-resolved contrast-enhanced three-dimensional magnetic resonance
angiography of the chest: combination of parallel imaging with view sharing
(TREAT). Invest Radiol 2005;40
: 40-48[Medline]
- Thomsen HS, Almen T, Morcos SK. Gadolinium-containing contrast
media for radiographic examinations: a position paper. Eur
Radiol 2002; 12:2600
-2605[Medline]
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
Top
Abstract
Introduction
