DOI:10.2214/AJR.05.0403
AJR 2006; 187:1107-1114
© American Roentgen Ray Society
Time-Resolved Contrast-Enhanced MR Angiography of the Thorax in Adults with Congenital Heart Disease
Oliver K. Mohrs1,2,
Steffen E. Petersen3,
Thomas Voigtlaender4,
Jutta Peters5,
Bernd Nowak4,
Markus K. Heinemann6 and
Hans-Ulrich Kauczor2
1 Darmstadt Radiology, Department of Cardiovascular Imaging at Alice Hospital,
Dieburger Strasse 29-13, 64287 Darmstadt, Germany.
2 Department of Radiology, German Cancer Research Center (DKFZ), Heidelberg,
Germany.
3 University of Oxford, Centre for Clinical MR Research (OCMR), Oxford, United
Kingdom.
4 Department of Cardiology, Cardiovascular Center Bethanien (CCB),
Frankfurt/Main, Germany.
5 Department of Radiology, University of Frankfurt, Frankfurt/Main,
Germany.
6 Department of Pediatric Surgery, University of Mainz, Mainz, Germany.
Received March 7, 2005;
accepted after revision June 7, 2005.
Address correspondence to O. K. Mohrs.
(mohrs{at}radiologie-darmstadt.de;
Internet:
www.radiologie-darmstadt.de).
Abstract
OBJECTIVE. The aim of this study was to evaluate the diagnostic
value of time-resolved contrast-enhanced MR angiography in adults with
congenital heart disease.
SUBJECTS AND METHODS. Twenty patients with congenital heart disease
(mean age, 38 ± 14 years; range, 16-73 years) underwent
contrast-enhanced turbo fast low-angle shot MR angiography. Thirty consecutive
coronal 3D slabs with a frame rate of 1-second duration were acquired. The
mask defined as the first data set was subtracted from subsequent images.
Image quality was evaluated using a 5-point scale (from 1, not assessable, to
5, excellent image quality). Twelve diagnostic parameters yielded 1 point each
in case of correct diagnosis (binary analysis into normal or abnormal) and
were summarized into three categories: anatomy of the main thoracic vessels
(maximum, 5 points), sequential cardiac anatomy (maximum, 5 points), and shunt
detection (maximum, 2 points). The results were compared with a combined
clinical reference comprising medical or surgical reports and other imaging
studies. Diagnostic accuracies were calculated for each of the parameters as
well as for the three categories.
RESULTS. The mean image quality was 3.7 ± 1.0. Using a binary
approach, 220 (92%) of the 240 single diagnostic parameters could be analyzed.
The percentage of maximum diagnostic points, the sensitivity, the specificity,
and the positive and the negative predictive values were all 100% for the
anatomy of the main thoracic vessels; 97%, 87%, 100%, 100%, and 96% for
sequential cardiac anatomy; and 93%, 93%, 92%, 88%, and 96% for shunt
detection.
CONCLUSION. Time-resolved contrast-enhanced MR angiography provides,
in one breath-hold, anatomic and qualitative functional information in adult
patients with congenital heart disease. The high diagnostic accuracy allows
the investigator to tailor subsequent specific MR sequences within the same
session.
Keywords: cardiac imaging • cardiovascular imaging • congenital heart disease • conventional angiography • heart • MR angiography • thorax
Introduction
In recent years the number of adolescents and adults diagnosed with
congenital heart disease has been growing because of the improvements in
therapeutic strategies [1]. For
a substantial number of patients who underwent surgical corrections during
childhood, the complex anatomy is not fully known, surgical reports are not
available, and patients are lost to follow-up. During adolescence, nonspecific
clinical symptoms such as dyspnea, fatigue, or arrhythmias may develop. In
contrast to neonates and children, in whom a wealth of information can be
rapidly gained by echocardiography, older patients need a different diagnostic
technique to reveal the underlying and surgical anatomy.
The usefulness of MRI for anatomic imaging in patients with congenital
heart disease is well known and has also been highlighted for functional
imaging
[2-5].
It is important to know the underlying diagnosis before performing specific
functional imaging to maintain an acceptable imaging time.
One possible way to obtain the maximum information very quickly during one
single procedure could be to start the examination with time-resolved
contrast-enhanced MR angiography of the thorax. This technique was proposed
for various clinical applications and pulmonary abnormalities in the mid 1990s
and combines tomographic and projection imaging methods
[6,
7].
The purpose of this study was to evaluate the diagnostic accuracy of
time-resolved contrast-enhanced MR angiography of the thorax in adult patients
with known or suspected complex congenital heart disease.
Subjects and Methods
Study Population
Twenty consecutive adult patients (nine women, 11 men; mean age, 38
± 14 years; age range, 16-73 years) with known or suspected congenital
heart disease were enrolled for MRI after written informed consent had been
obtained. Nine had previous operations, either palliative or corrective, and
11 were new patients or known patients presenting with a new diagnostic
problem. All patients underwent time-resolved MR angiography of the thorax
followed by a specific functional investigation. The combined information of
medical (n = 20) and surgical reports (n = 6), radiograph
angiography (n = 17), echocardiography (n = 20), and chest
radiographs (n = 12) served as the clinical reference.
Table 1 contains the
demographic data and diagnoses of the study population.
MRI
MRI was performed on a 1.5-T MRI system (Magnetom Sonata Maestro Class,
Siemens Medical Solutions). For signal detection, the combination of a
six-channel body phased-array coil and a two-channel spine phased-array coil
was used.
A fast low-angle shot (FLASH) 3D sequence (TR/TE, 1.9/0.7; flip angle,
20°; matrix, 233 x 320; voxel size, 1.5 x 1.3 x 10.0
mm3; slice thickness, 10 mm; number of slices, 10; modified
sensitivity encoding [mSENSE] with a factor of 2, 6/8 partial Fourier;
bandwidth, 1,300 Hz/pixel [coronal plane]) was used for MR angiography of the
thoracic vessels and the heart. One complete 3D data set was acquired in 1
second and repeated 29 times during one breath-hold. The imaging sequence was
started simultaneously with the administration of 10 mL of gadopentetate
dimeglumine (Magnevist, Schering) followed by 30 mL of saline solution into
the left antecubital vein (to permit the diagnosis of a persisting left
superior vena cava) at a rate of 6 mL/s. The mean dosage of the contrast agent
was 0.07 ± 0.02 mmol per kilograms of body weight.
Background signal and overfolding was suppressed by the subtraction of the
first data set (baseline) from the subsequent data sets. A four-dimensional
(4D) movie of 29 consecutive 3D maximum intensity projections was created.
Image Analysis
The data sets (source images, maximum intensity projections, and 4D movies)
were evaluated by consensus of three observers experienced either in
cardiovascular radiology, pediatric heart surgery, or cardiology. The
interpretations were performed blinded to any clinical information available.
The MR image quality was graded using a five-point scale: not assessable (1
point), poor (2 points), moderate (3 points), good (4 points), and excellent
(5 points).
In every patient, 12 single parameters were analyzed in a binary manner and
considered normal or abnormal. Each parameter, if assessable, yielded one
point in case of the correct diagnosis when compared with the available
clinical information. The parameters (1-12) were summarized into three
diagnostic categories, A, B, and C.
Category A refers to the anatomy of the main thoracic vessels and includes
the following parameters: (1) persistent left superior vena cava (therefore
the contrast agent was administered into the left antecubital vein if
possible), (2) abnormalities of the right ventricular outflow tract and the
pulmonary trunk (stenosis, enlargement), (3) abnormalities of the pulmonary
arteries (stenosis, enlargement, arteriovenous malformations, atypical
position or branching), (4) abnormalities of the pulmonary veins (stenosis,
atypical position or anomalous pulmonary venous return), and (5) abnormalities
of the aorta (stenosis, enlargement, atypical position, slings, coarctation,
patent ductus arteriosus).
Category B summarizes individual parameters of sequential cardiac anatomy:
(6) position of the heart, (7) subpulmonary atrioventricular connection, (8)
ventriculopulmonary connection, (9) subaortic atrioventricular connection, and
(10) ventriculoatrial connection.
Category C contains the evaluation of (11) right-to-left shunting (defined
as an enhancement of the subaortic atrium, the subaortic ventricle, or the
aorta before the enhancement of the pulmonary veins by visual assessment) and
(12) left-to-right shunting (defined as a second peak enhancement or a plateau
enhancement of pulmonary arteries, subpulmonary ventricle or atrium in the
interval between the enhancement of the aorta, and of the jugular veins by
visual assessment).
Statistical Analysis
Image quality is shown as mean and SD. For each of the 12 single parameters
and for the three different diagnostic categories, the total amount of yielded
points and the sensitivity, specificity, and positive and negative predictive
values are given.
Results
Image Quality
All data sets were evaluated for overall image quality, even if parts of
the cardiovascular structures were not within the 3D volume and therefore not
assessable. The mean overall image quality was 3.7 ± 1.0 points on our
five-point scale. The grading for the data sets was as follows: 5 points in
four patients, 4 points in eight patients, 3 points in five patients, and 2
points in three patients.
Category A: Anatomy of the Main Thoracic Vessels
Using time-resolved MR angiography, 88% (88/100) of the individual
parameters could be analyzed. Parameters not assessable were the left superior
vena cava (n = 3) (contrast agent could not be administered from left
side), right ventricular outflow tract (n = 2), pulmonary veins
(n = 6), and aorta (n = 1) because of partial amputation at
the edge of the 3D data set. For the 88 assessable parameters in this
category, the sensitivity, specificity, and positive and negative predictive
values were all 100%.
Table 2 shows the diagnostic
values, including the statistical results of the summarizing categories and
the single parameters. Figures
1A,
1B,
1C, and
1D and Figure S1 show a normal
anatomy of the main thoracic vessels, whereas Figures
2A,
2B,
2C,
2D,
3A,
3B,
4A,
4B,
5A,
5B,
5C, and
5D and Figures S4 and S5 show
thoracic vessels with abnormalities. (Figures S1, S4, and S5 can be seen in
the data supplement to this article, available at
www.ajronline.org.)
View larger version (148K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1A —23-year-old woman referred to MRI with suspicion of
transposition of great arteries because of prominent trabeculation of
subaortic ventricle in echocardiography. Time-resolved coronal
maximum-intensity-projection angiograms (A-C) and cine short-axis view
(D). The angiograms show normal time course of enhancement of
subpulmonary atrium and ventricle and pulmonary arteries (A); and
enhancement of pulmonary veins, subaortic atrium, ventricle, and aorta
(B). Note decreasing enhancement of subpulmonary ventricle from
A to C, indicating absence of relevant left-to-right shunt flow.
This patient suffered from noncompaction myocardium, which is visualized on
time-resolved MR angiography (C) (arrow) and is shown on cine
short-axis view (D) (arrows). See also Figure S1, cine loop,
in supplemental data online
(www.ajronline.org).
|
|
View larger version (114K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1B —23-year-old woman referred to MRI with suspicion of
transposition of great arteries because of prominent trabeculation of
subaortic ventricle in echocardiography. Time-resolved coronal
maximum-intensity-projection angiograms (A-C) and cine short-axis view
(D). The angiograms show normal time course of enhancement of
subpulmonary atrium and ventricle and pulmonary arteries (A); and
enhancement of pulmonary veins, subaortic atrium, ventricle, and aorta
(B). Note decreasing enhancement of subpulmonary ventricle from
A to C, indicating absence of relevant left-to-right shunt flow.
This patient suffered from noncompaction myocardium, which is visualized on
time-resolved MR angiography (C) (arrow) and is shown on cine
short-axis view (D) (arrows). See also Figure S1, cine loop,
in supplemental data online
(www.ajronline.org).
|
|
View larger version (157K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1C —23-year-old woman referred to MRI with suspicion of
transposition of great arteries because of prominent trabeculation of
subaortic ventricle in echocardiography. Time-resolved coronal
maximum-intensity-projection angiograms (A-C) and cine short-axis view
(D). The angiograms show normal time course of enhancement of
subpulmonary atrium and ventricle and pulmonary arteries (A); and
enhancement of pulmonary veins, subaortic atrium, ventricle, and aorta
(B). Note decreasing enhancement of subpulmonary ventricle from
A to C, indicating absence of relevant left-to-right shunt flow.
This patient suffered from noncompaction myocardium, which is visualized on
time-resolved MR angiography (C) (arrow) and is shown on cine
short-axis view (D) (arrows). See also Figure S1, cine loop,
in supplemental data online
(www.ajronline.org).
|
|
View larger version (118K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1D —23-year-old woman referred to MRI with suspicion of
transposition of great arteries because of prominent trabeculation of
subaortic ventricle in echocardiography. Time-resolved coronal
maximum-intensity-projection angiograms (A-C) and cine short-axis view
(D). The angiograms show normal time course of enhancement of
subpulmonary atrium and ventricle and pulmonary arteries (A); and
enhancement of pulmonary veins, subaortic atrium, ventricle, and aorta
(B). Note decreasing enhancement of subpulmonary ventricle from
A to C, indicating absence of relevant left-to-right shunt flow.
This patient suffered from noncompaction myocardium, which is visualized on
time-resolved MR angiography (C) (arrow) and is shown on cine
short-axis view (D) (arrows). See also Figure S1, cine loop,
in supplemental data online
(www.ajronline.org).
|
|
View larger version (172K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2A —29-year-old woman with D-transposition of great arteries as
result of undergoing Mustard atrial switch procedure during childhood. Coronal
cine view (A) shows small systemic venous pathways (asterisks)
from inferior and superior vena cava to subpulmonary ventricle (morphology of
left ventricle) and pulmonary venous pathway (arrow), which is
connected to subaortic ventricle (morphology of right ventricle). Single slice
of coronal time-resolved MR angiography shows small systemic venous pathways
(B), and sagittal angiogram shows pulmonary venous pathway (C)
(arrow). On chest radiograph (D), which is gray-scale
manipulated, pacemaker probe (implanted after MRI) shows course of blood flow
(arrows) from superior vena cava after undergoing Mustard atrial
switch. Ao = aorta, SVC = superior vena cava, PT = pulmonary trunk, IVC =
inferior vena cava, spV = subpulmonary ventricle, saV = subaortic
ventricle.
|
|
View larger version (158K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2B —29-year-old woman with D-transposition of great arteries as
result of undergoing Mustard atrial switch procedure during childhood. Coronal
cine view (A) shows small systemic venous pathways (asterisks)
from inferior and superior vena cava to subpulmonary ventricle (morphology of
left ventricle) and pulmonary venous pathway (arrow), which is
connected to subaortic ventricle (morphology of right ventricle). Single slice
of coronal time-resolved MR angiography shows small systemic venous pathways
(B), and sagittal angiogram shows pulmonary venous pathway (C)
(arrow). On chest radiograph (D), which is gray-scale
manipulated, pacemaker probe (implanted after MRI) shows course of blood flow
(arrows) from superior vena cava after undergoing Mustard atrial
switch. Ao = aorta, SVC = superior vena cava, PT = pulmonary trunk, IVC =
inferior vena cava, spV = subpulmonary ventricle, saV = subaortic
ventricle.
|
|
View larger version (158K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2C —29-year-old woman with D-transposition of great arteries as
result of undergoing Mustard atrial switch procedure during childhood. Coronal
cine view (A) shows small systemic venous pathways (asterisks)
from inferior and superior vena cava to subpulmonary ventricle (morphology of
left ventricle) and pulmonary venous pathway (arrow), which is
connected to subaortic ventricle (morphology of right ventricle). Single slice
of coronal time-resolved MR angiography shows small systemic venous pathways
(B), and sagittal angiogram shows pulmonary venous pathway (C)
(arrow). On chest radiograph (D), which is gray-scale
manipulated, pacemaker probe (implanted after MRI) shows course of blood flow
(arrows) from superior vena cava after undergoing Mustard atrial
switch. Ao = aorta, SVC = superior vena cava, PT = pulmonary trunk, IVC =
inferior vena cava, spV = subpulmonary ventricle, saV = subaortic
ventricle.
|
|
View larger version (123K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2D —29-year-old woman with D-transposition of great arteries as
result of undergoing Mustard atrial switch procedure during childhood. Coronal
cine view (A) shows small systemic venous pathways (asterisks)
from inferior and superior vena cava to subpulmonary ventricle (morphology of
left ventricle) and pulmonary venous pathway (arrow), which is
connected to subaortic ventricle (morphology of right ventricle). Single slice
of coronal time-resolved MR angiography shows small systemic venous pathways
(B), and sagittal angiogram shows pulmonary venous pathway (C)
(arrow). On chest radiograph (D), which is gray-scale
manipulated, pacemaker probe (implanted after MRI) shows course of blood flow
(arrows) from superior vena cava after undergoing Mustard atrial
switch. Ao = aorta, SVC = superior vena cava, PT = pulmonary trunk, IVC =
inferior vena cava, spV = subpulmonary ventricle, saV = subaortic
ventricle.
|
|
View larger version (147K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3A —46-year-old woman with persistent left superior vena cava. MR
angiography image shows drainage of left superior vena cava (arrows)
into right atrium via coronary sinus. To diagnose a persistent left superior
vena cava, it is mandatory to inject contrast agent from left arm.
|
|
View larger version (144K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3B —46-year-old woman with persistent left superior vena cava.
This finding was verified in oblique sagittal cine imaging (arrows).
RA = right atrium, LA = left atrium, IVC = inferior vena cava.
|
|
View larger version (111K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4A —42-year-old woman with heterotaxia syndrome (levoloop
ventricle and dextroloop vessels). Time-resolved angiogram (A) shows
subaortic ventricle (saV) and course of aorta (Ao) with excellent image
quality. One of last angiograms at time of parenchymal enhancement; reflow of
contrast agent from lower body clearly shows continuity of azygos vein
(B) (arrows) draining into superior vena cava. PT = pulmonary
trunk, spA = subpulmonary atrium, spV = subpulmonary ventricle. See also
Figure S4, cine loop, in supplemental data online
(www.ajronline.org).
|
|
View larger version (145K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4B —42-year-old woman with heterotaxia syndrome (levoloop
ventricle and dextroloop vessels). Time-resolved angiogram (A) shows
subaortic ventricle (saV) and course of aorta (Ao) with excellent image
quality. One of last angiograms at time of parenchymal enhancement; reflow of
contrast agent from lower body clearly shows continuity of azygos vein
(B) (arrows) draining into superior vena cava. PT = pulmonary
trunk, spA = subpulmonary atrium, spV = subpulmonary ventricle. See also
Figure S4, cine loop, in supplemental data online
(www.ajronline.org).
|
|
View larger version (126K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 5A —21-year-old man with criss-cross heart. One of first coronal
time-resolved angiograms (A) shows enhancement of subaortic atrium,
subaortic ventricle (saV), and aorta (Ao), indicating that systemic venous
drainage (contrast agent injection from left antecubital vein) goes directly
to systemic arterial circulation. Some seconds later, angiogram shows
enhancement of subpulmonary ventricle (spV) and pulmonary trunk (PT) caused by
large ventricular septal defect (B) (asterisk). Oblique
coronal cine view (C) verifies complex anatomy. Asterisks in A-C
show connection between subpulmonary atrium and subaortic ventricle. During
childhood patient received surgical shunt between left subclavian artery and
pulmonary artery (D) that was not detected on time-resolved MR
angiography, most likely because of small prosthesis diameter. Nevertheless,
aortic course was diagnosed as abnormal using our binary approach and
therefore high-resolution contrast-enhanced MR angiography was performed,
which allowed correct diagnosis. spA = subpulmonary atrium. See also Figure
S5, cine loop, in supplemental data online
(www.ajronline.org).
|
|
View larger version (130K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 5B —21-year-old man with criss-cross heart. One of first coronal
time-resolved angiograms (A) shows enhancement of subaortic atrium,
subaortic ventricle (saV), and aorta (Ao), indicating that systemic venous
drainage (contrast agent injection from left antecubital vein) goes directly
to systemic arterial circulation. Some seconds later, angiogram shows
enhancement of subpulmonary ventricle (spV) and pulmonary trunk (PT) caused by
large ventricular septal defect (B) (asterisk). Oblique
coronal cine view (C) verifies complex anatomy. Asterisks in A-C
show connection between subpulmonary atrium and subaortic ventricle. During
childhood patient received surgical shunt between left subclavian artery and
pulmonary artery (D) that was not detected on time-resolved MR
angiography, most likely because of small prosthesis diameter. Nevertheless,
aortic course was diagnosed as abnormal using our binary approach and
therefore high-resolution contrast-enhanced MR angiography was performed,
which allowed correct diagnosis. spA = subpulmonary atrium. See also Figure
S5, cine loop, in supplemental data online
(www.ajronline.org).
|
|
View larger version (180K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 5C —21-year-old man with criss-cross heart. One of first coronal
time-resolved angiograms (A) shows enhancement of subaortic atrium,
subaortic ventricle (saV), and aorta (Ao), indicating that systemic venous
drainage (contrast agent injection from left antecubital vein) goes directly
to systemic arterial circulation. Some seconds later, angiogram shows
enhancement of subpulmonary ventricle (spV) and pulmonary trunk (PT) caused by
large ventricular septal defect (B) (asterisk). Oblique
coronal cine view (C) verifies complex anatomy. Asterisks in A-C
show connection between subpulmonary atrium and subaortic ventricle. During
childhood patient received surgical shunt between left subclavian artery and
pulmonary artery (D) that was not detected on time-resolved MR
angiography, most likely because of small prosthesis diameter. Nevertheless,
aortic course was diagnosed as abnormal using our binary approach and
therefore high-resolution contrast-enhanced MR angiography was performed,
which allowed correct diagnosis. spA = subpulmonary atrium. See also Figure
S5, cine loop, in supplemental data online
(www.ajronline.org).
|
|
View larger version (137K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 5D —21-year-old man with criss-cross heart. One of first coronal
time-resolved angiograms (A) shows enhancement of subaortic atrium,
subaortic ventricle (saV), and aorta (Ao), indicating that systemic venous
drainage (contrast agent injection from left antecubital vein) goes directly
to systemic arterial circulation. Some seconds later, angiogram shows
enhancement of subpulmonary ventricle (spV) and pulmonary trunk (PT) caused by
large ventricular septal defect (B) (asterisk). Oblique
coronal cine view (C) verifies complex anatomy. Asterisks in A-C
show connection between subpulmonary atrium and subaortic ventricle. During
childhood patient received surgical shunt between left subclavian artery and
pulmonary artery (D) that was not detected on time-resolved MR
angiography, most likely because of small prosthesis diameter. Nevertheless,
aortic course was diagnosed as abnormal using our binary approach and
therefore high-resolution contrast-enhanced MR angiography was performed,
which allowed correct diagnosis. spA = subpulmonary atrium. See also Figure
S5, cine loop, in supplemental data online
(www.ajronline.org).
|
|
Category B: Sequential Cardiac Anatomy
Using MRI, a percentage of diagnostic points of 97%, a sensitivity of 87%,
a specificity of 100%, a positive predictive value of 100%, and a negative
predictive value of 96% were yielded in this category.
False-negative results were abnormal subpulmonary atrioventricular
connection in two patients (one patient with tricuspid atresia and single
atrioventricular valve, and one patient with levoloop ventricle, dextroloop
vessels), and caval continuity via an azygos vein [patient number 12]) and
abnormal systemic atrioventricular connection in one patient (again, patient
number 12).
The percentage of diagnostic points for sensitivity, specificity, positive
predictive value, and negative predictive value were all 100% for position of
the heart, for ventriculopulmonary connection, and for ventriculoatrial
connection. Sensitivity was lower for the subpulmonic atrioventricular
connection (50%) and for the systemic atrioventricular connection (80%), but
specificity remained 100% for both. The remaining accuracy data are presented
in Table 2.
Category C: Shunt Detection
Overall, 93% sensitivity, 92% specificity, 88% positive predictive value,
and 96% negative predictive value were yielded in this category. We saw
false-positive right-to-left shunts in two patients (one patient's status
resulted from undergoing the Mustard atrial switch procedure during childhood,
and the other patient had a major atrial septal defect type II [64%
left-to-right shunt fraction]). In one postoperative status patient, a
false-negative left-to-right shunt (transposition of great arteries and small
ventricular septal defect with < 10% left-to-right shunt fraction) was
diagnosed.
For right-to-left shunting, the percentage of diagnostic points,
sensitivity, specificity, positive predictive value, and negative predictive
value were between 60% and 100%. The detailed data are presented in
Table 2.
Discussion
Time-resolved contrast-enhanced MR angiography provides rapid anatomic and
functional information in adult patients with complex congenital heart
disease. The high diagnostic accuracy allows the investigator to tailor
subsequent specific MRI sequences within the same session.
Various MRI techniques have been used to visualize and to quantify specific
congenital cardiac abnormalities
[3,
4,
8]. The major drawback of these
techniques, especially in patients with unknown anatomy, is their need to
focus on a specific finding because of time constraints. An approach providing
a fast overview at the beginning of the MRI examination, therefore enabling
further specific morphologic and functional analysis during the same session,
would be preferable.
Our approach to overcome this problem was time-resolved contrast-enhanced
MR angiography of the thorax. In the mid 1990s this technique was proposed for
various clinical applications including pulmonary abnormalities
[6,
7]. Formerly a single slab with
a thickness of at least 160 mm was used, which was based on a lower hardware
performance [9]. Time-resolved
MR angiography with a single-slab technique could then only be used as
projection angiography comparable to digital subtraction radiograph
angiography. If consisting of multiple slices, as in our study, the data sets
of time-resolved MR angiography can also be used as a tomographic method, one
of the prerequisites for correct diagnosis in patients with complex congenital
heart disease.
In our study, we could not evaluate some structures because of partial
amputation at the edge of the slab. We had to balance a sufficient temporal
resolution of the 3D data sets with the number of slices acquired. Further
developments in parallel imaging techniques might help to hasten the data
acquisition and to improve the assessment of peripheral structures.
Our findings suggest good image quality (3.7, using a five-point scale) for
the MRI protocol described, enabling a reliable diagnosis. In our opinion,
based on initial experience, the amount of contrast agent should be as small
as possible using a high flow rate to obtain a compact bolus. The compactness
of the bolus is mandatory to separate the filling times of the different
vascular or cardiac compartments. We used a mean dosage of 0.07 ± 0.02
mmol/kg/body weight of gadopentetate dimeglumine to achieve such a compact
bolus of contrast agent without a major loss of vessel visualization
[10]. In addition, a second
injection for high-resolution MR angiography with a sufficient signal-to-noise
ratio is possible if needed during the same examination.
We used a binary approach to determine whether a single parameter was
normal or abnormal. This less time-consuming approach is very easy to manage,
which contributes favorably to a first overview. Some patients showed a
typical appearance, for instance the criss-cross heart. This peculiar but
easily recognizable anatomy has already been presented in a case report using
high-resolution MR angiography
[11]. Similar cases could also
be diagnosed immediately using time-resolved MR angiography, and functional
imaging could be performed during the same session. In other cases, the single
parameters were combined and some algorithms could be established to achieve a
diagnosis. This was not a primary objective of our study, but during the MRI
interpretations it became evident that this should probably be the subject of
further investigations.
Time-resolved MR angiography for diagnosing pathologies of the thoracic
vessels is already established
[12,
13]. Therefore, it is not
surprising that we yielded excellent diagnostic accuracy for those single
parameters that were summarized into this category. Nevertheless, the
morphology of the thoracic vessels also gave important indirect signs of
functional pathology, such as vessel dilatation caused by relevant dysfunction
of the connected valve. Pulmonary artery dilatation associated with pulmonary
valve insufficiency may serve as a typical example.
Diagnostic accuracy was only slightly reduced in detection of shunt flow.
An example of one pitfall is pulmonary regurgitation, implicating a plateau
phenomenon in the subpulmonary ventricle. Phase-contrast flow mapping of
systemic and pulmonary blood flow could not be replaced by time-resolved MR
angiography and should still be performed in patients with congenital heart
disease. Nevertheless, time-resolved MR angiography could give useful
information in patients with both right-to-left and left-to-right shunt flow
because computation of systemic and pulmonary flow mapping shows only the
major shunt flow.
Hirsch and coworkers [14]
have previously shown the diagnostic benefits of MRI over transesophageal
echocardiography in adolescents and adults with congenital heart disease. A
combination of morphologic and functional MRI yielded high diagnostic accuracy
for pathologies of the pulmonary vessels and the aorta and for surgical
shunts, but poor results for atrial morphology and atrioventricular valve
structures.
The diagnostic accuracy of this much longer MRI investigation reflects our
findings using time-resolved MR angiography in a single breath-hold. Although
our approach is not able to provide quantitative functional results such as
ventricular function or flow quantification, excellent diagnostic values could
be yielded for various pathologies of thoracic vessels and for
ventriculoarterial connections and a slightly reduced diagnostic accuracy for
atrioventricular connections and for shunt detection.
Sorensen and coworkers [15]
used a free-breathing (navigator-gated) 3D steady-state free precision
sequence in complex congenital heart disease. One advantage of this method is
the possibility of reformatting multiple planes after data acquisition,
enabling a retrospective analysis of complex cardiac morphology. This method
is, however, limited by a scanning time approximately 20 times as long as our
approach; by missing dynamic and, therefore, functional information; and by
the need of sinus rhythm as a result of ECG-gating. In contrast, time-resolved
contrast-enhanced MR angiography is quick, provides qualitative functional and
anatomic information, and offers high image quality independent of
ECG-gating.
Limitations of our study included the need to balance temporal and spatial
resolution. Using these balanced sequence parameters, it is for instance, not
possible to reliably detect coronary artery anomalies. Parallel imaging
techniques may facilitate better spatial resolution without compromising
temporal resolution. Furthermore, we would recommend more efficient parallel
imaging to enlarge the 3D volume, thereby preventing amputation of important
structures at the edge of the data set.
In conclusion, time-resolved contrast-enhanced MR angiography provides
rapid anatomic and functional information in adult patients with complex
congenital heart disease in a single breath-hold. The high diagnostic accuracy
may allow the investigator to tailor subsequent specific MRI sequences within
the same session.
References
- Sutherland GR, Hess J, Roelandt J, Quaegebeur J. The increasing
problem of young adults with congenital heart disease. Eur Heart
J 1990; 11:4
-6[Free Full Text]
- Mohrs OK, Voigtlander T, Heussel CP, et al. Morphologic and
functional assessment of vascular abnormalities of the pulmonary vasculature
by breath-hold MR techniques [in German]. Rofo2002; 174:467
-473[Medline]
- Mohrs OK, Petersen SE, Erkapic D, et al. Diagnosis of patent
foramen ovale using contrast-enhanced dynamic MRI: a pilot study.
AJR 2005; 184:234
-240[Abstract/Free Full Text]
- Pereles FS, Kapoor V, Carr JC, et al. Usefulness of segmented true
FISP cardiac pulse sequence in evaluation of congenital and acquired adult
cardiac abnormalities. AJR 2001;177
: 1155-1160[Abstract/Free Full Text]
- Petersen SE, Voigtlander T, Kreitner KF, et al. Quantification of
shunt volumes in congenital heart disease using a breath-hold MR phase
contrast technique: comparison with oximetry. Int J Cardiovasc
Imaging 2002; 18:63
-68
- Wang Y, Rossman PJ, Grimm RC, et al. 3D MR angiography of pulmonary
arteries using real-time navigator gating and magnetization preparation
Magn Reson Med 1996;36
: 579-587[Medline]
- Hennig J, Scheffler K, Laubenberger J, Strecker R. Time-resolved
projection angiography after bolus injection of contrast agent.
Magn Reson Med 1997;37
: 341-345[Medline]
- Sampson C, Kilner PJ, Hirsch R, et al. Venoatrial pathways after
the Mustard operation for transposition of the great arteries: anatomic and
functional MR imaging. Radiology 1994;193
: 211-217[Abstract/Free Full Text]
- Sonnet S, Buitrago-Tellez CH, Scheffler K, et al. Dynamic
time-resolved contrast-enhanced two-dimensional MR projection angiography of
the pulmonary circulation: standard technique and clinical applications.
AJR 2002; 179:159
-165[Abstract/Free Full Text]
- Sonnet S, Buitrago-Tellez CH, Schulte AC, et al. Dose optimization
for dynamic time-resolved contrast-enhanced 3D MR angiography of pulmonary
circulation. AJR 2003;181
: 1499-1503[Abstract/Free Full Text]
- Araoz PA, Reddy GP, Thomson PD, et al. Magnetic resonance
angiography of criss-cross heart. Circulation2002; 105:537
-538[Free Full Text]
- Goyen M, Ruehm SG, Jagenburg A, et al. Pulmonary arteriovenous
malformation: characterization with time-resolved ultrafast 3D MR angiography.
J Magn Reson Imaging 2001;13
: 458-460[CrossRef][Medline]
- Eichhorn J, Fink C, Bock M, et al. Time-resolved three-dimensional
magnetic resonance angiography for assessing a pulmonary artery sling in a
pediatric patient. Circulation 202;106
: e61-e62[CrossRef]
- Hirsch R, Kilner PJ, Connelly MS, et al. Diagnosis in adolescents
and adults with congenital heart disease: prospective assessment of individual
and combined roles of magnetic resonance imaging and transesophageal
echocardiography. Circulation 1994;90
: 2937-2951[Abstract/Free Full Text]
- Sorensen TS, Korperich H, Greil GF, et al. Operator-independent
isotropic three-dimensional magnetic resonance imaging for morphology in
congenital heart disease: a validation study.
Circulation 2004;110
: 163-169[Abstract/Free Full Text]
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
|
|
 
A. Kahana, M. J. Lucarelli, A. M. Grayev, J. J. Van Buren, C. N. Burkat, and L. R. Gentry
Noninvasive Dynamic Magnetic Resonance Angiography With Time-Resolved Imaging of Contrast KineticS (TRICKS) in the Evaluation of Orbital Vascular Lesions
Arch Ophthalmol,
December 1, 2007;
125(12):
1635 - 1642.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
