DOI:10.2214/AJR.05.0143
AJR 2006; 186:1737-1745
© American Roentgen Ray Society
Optimization of CT Angiography of the Carotid Artery with a 16-MDCT Scanner: Craniocaudal Scan Direction Reduces Contrast Material-Related Perivenous Artifacts
Cécile de Monyé1,
Thomas T. de Weert1,
William Zaalberg1,
Filippo Cademartiri1,
Dorine A. M. Siepman2,
Diederik W. J. Dippel2 and
Aad van der Lugt1
1 Department of Radiology, Erasmus University Medical Center, Dr. Molewaterplein
40, Rotterdam, The Netherlands, 3015 GD.
2 Department of Neurology, Erasmus University Medical Center, Rotterdam, The
Netherlands.
Received January 28, 2005;
accepted after revision April 27, 2005.
Address correspondence to A. van der Lugt
(a.vanderlugt{at}erasmusmc.nl).
Abstract
OBJECTIVE. The objective of our study was to compare the effect of a
caudocranial scan direction versus a craniocaudal scan direction on arterial
enhancement and perivenous artifacts in 16-MDCT angiography of the supraaortic
arteries.
SUBJECTS AND METHODS. Eighty consecutive patients (51 men; mean age,
62 years; age range, 28-89 years) underwent scanning in the caudocranial
direction (group 1; n = 40) or the craniocaudal direction (group 2;
n = 40). All patients received 80 mL of contrast material followed by
a 40-mL saline chaser bolus, both administered IV at 4 mL/sec. Bolus tracking
was used. Attenuation inside the arterial lumen was measured at intervals of 1
sec throughout the data set. Attenuation in the superior vena cava (SVC) was
measured. Contrast material-related perivenous artifacts were graded on a
scale of 0-3 (none to extensive).
RESULTS. Attenuation in the ascending aorta, carotid bifurcation,
and intracranial arteries was slightly lower in group 2 versus group 1 (231
± 64 H, 348 ± 52 H, and 258 ± 48 H vs 282 ± 43 H,
381 ± 73 H, and 291 ± 77 H, respectively; p < 0.05).
Maximum and mean arterial attenuations were slightly lower in group 2 versus
group 1 (369 ± 58 H and 303 ± 48 H vs 401 ± 71 H and 334
± 58 H; p < 0.05). Attenuation in the SVC was much lower in
group 2 versus group 1 (169 ± 39 H vs 783 ± 330 H; p
< 0.001). Mean streak artifact score was much lower in group 2 versus group
1 (1.3 ± 0.9 vs 2.5 ± 0.6; p < 0.001).
CONCLUSION. Use of a craniocaudal scan direction results in slightly
lower attenuation of the carotid artery and much lower attenuation of the SVC.
Streak artifacts are significantly reduced. This technique allows better
evaluation of the ascending aorta and supraaortic arteries.
Keywords: arteries • cardiovascular disease • cardiovascular imaging • CT angiography • CT technique • perivenous artifacts
Introduction
Acute ischemic neurologic symptoms are related to small-vessel
disease of the intracranial perforating arteries, thromboembolism from
atherosclerotic disease in the supraaortic arteries, and cardiac embolism
[1]. The most common source of
thromboembolism is atherosclerotic disease of the carotid bifurcation.
However, atherosclerotic lesions in the aorta, the origin of the supraaortic
arteries, the common carotid artery (CCA), the internal carotid artery (ICA)
distal to the bifurcation, and the vertebrobasilar circulation can cause
transient ischemic attack or ischemic stroke due to thromboembolism
[2,
3]. In the evaluation of
patients with cerebrovascular disease, complete vascular imaging from the
aorta to the circle of Willis must be performed before therapeutic decision
making can be undertaken.
With the introduction of MDCT, particularly 16-MDCT scanners, CT
angiography (CTA) has become an attractive diagnostic method in the care of
patients with cerebrovascular symptoms
[4]. With bolus tracking, CTA
scanning can be optimally synchronized with the passage of contrast material
in the arteries [5]. The CTA
scanning usually starts before the injection of contrast material ends. With
this method, the presence of undiluted contrast material in the subclavian
vein, brachiocephalic vein, and superior vena cava (SVC) produces artifacts
that project over the ascending aorta and the origin of the supraaortic
arteries [6,
7]. Such artifacts can obscure
adjacent structures and thus hide or suggest stenosis or occlusion of the
proximal supraaortic arteries (Figs.
1A,
1B,
1C and
1D). Addition of a chaser bolus
to the main contrast bolus may reduce the frequency of these artifacts by
clearing the veins of contrast material. However, the timing of the CTA scan
is not altered by the addition of a chaser bolus, and artifacts do occur. Use
of a craniocaudal scan direction, opposite to the direction of blood flow, may
reduce the number of artifacts caused by delayed scanning of the apex of the
thorax (Figs. 2A,
2B,
2C and
2D).
View larger version (81K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1A —CT angiograms of supraaortic arteries in 74-year-old woman scanned
in caudocranial direction with left-sided injection of contrast material.
Coronal maximum intensity projection (15 mm). High-density contrast material
in left subclavian vein and reflux of contrast material in neck veins give
rise to artifacts over origin of supraaortic vessels.
|
|
View larger version (111K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1B —CT angiograms of supraaortic arteries in 74-year-old woman scanned
in caudocranial direction with left-sided injection of contrast material.
Axial image at level of origin of left vertebral artery (arrow).
|
|
View larger version (37K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1C —CT angiograms of supraaortic arteries in 74-year-old woman scanned
in caudocranial direction with left-sided injection of contrast material.
Axial image at level of proximal part of left common carotid artery
(arrow) and left subclavian artery (arrowhead).
|
|
View larger version (24K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1D —CT angiograms of supraaortic arteries in 74-year-old woman scanned
in caudocranial direction with left-sided injection of contrast material.
Axial image at level of first 1 cm of brachiocephalic trunk (long
arrow) and left common carotid artery (short arrow). Evaluation
of atherosclerotic disease is hampered by streak artifacts.
|
|
View larger version (58K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2A —CT angiograms of supraaortic arteries. Four maximum intensity
projections (30 mm) in coronal plane in four patients. CT angiographic scans
in caudocranial direction with right-sided (72-year-old man, A) and
left-sided (74-year-old woman, B) injection of contrast material. Very
high density of contrast material in subclavian vein and superior vena cava
hides origin of supraaortic arteries.
|
|
View larger version (57K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2B —CT angiograms of supraaortic arteries. Four maximum intensity
projections (30 mm) in coronal plane in four patients. CT angiographic scans
in caudocranial direction with right-sided (72-year-old man, A) and
left-sided (74-year-old woman, B) injection of contrast material. Very
high density of contrast material in subclavian vein and superior vena cava
hides origin of supraaortic arteries.
|
|
View larger version (64K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2C —CT angiograms of supraaortic arteries. Four maximum intensity
projections (30 mm) in coronal plane in four patients. CT angiographic scans
in craniocaudal direction with right-sided (48-year-old man, C) and
left-sided (54-year-old man, D) injection of contrast material. High
density of contrast material is not left in veins, and all arteries are
clearly depicted.
|
|
View larger version (57K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2D —CT angiograms of supraaortic arteries. Four maximum intensity
projections (30 mm) in coronal plane in four patients. CT angiographic scans
in craniocaudal direction with right-sided (48-year-old man, C) and
left-sided (54-year-old man, D) injection of contrast material. High
density of contrast material is not left in veins, and all arteries are
clearly depicted.
|
|
The purpose of this study was to compare the effects of a caudocranial scan
direction versus a craniocaudal scan direction on enhancement of the carotid
artery and on the presence of perivenous artifacts in CTA with a 16-MDCT
scanner.
Subjects and Methods
Study Population
Between November 2002 and August 2003, 80 consecutive patients (51 men and
29 women; mean age, 62 years; age range, 28-89 years) who underwent CTA of the
carotid artery were enrolled in the study. The indication for CTA was
suspected atherosclerotic disease of the carotid or vertebrobasilar vascular
system in patients who had had a transient ischemic attack or minor ischemic
stroke. Exclusion criteria were previous allergic reaction to iodine contrast
medium, renal insufficiency (serum creatinine concentration > 100 mmol/L),
pregnancy, and age less than 18 years. Patients with occlusion of the carotid
artery also were excluded. The institutional review board approved the study,
and patients gave informed consent in writing.
Each patient was allocated to one of two groups with different scan
protocols. The first 40 patients (group 1) underwent scanning in the caudal to
cranial direction, the second 40 patients (group 2) underwent scanning in the
cranial to caudal direction. Age, sex, and body weight were recorded for each
patient.
Scan Protocol
The patients underwent CTA of the carotid artery with a 16-MDCT scanner
(Sensation 16, Siemens Medical Solutions). Patients were positioned supine on
the CT table with the arms along the chest. A lateral scout view including the
thorax, neck, and skull was acquired. The CTA scan range reached from the
ascending aorta to the intracranial blood vessels (2 cm above the sella
turcica). Scan parameters were as follows: number of detectors, 16; individual
detector width, 0.75 mm; table feed per rotation, 12 mm (pitch of 1); gantry
rotation time, 0.5 sec; 120 kV; 180 mAs; and scanning time, 10-14 sec,
depending on patient's size and anatomic features. The entire examination took
approximately 15 min.
Contrast material (iodixanol 320 mg I/mL [Visipaque, Amersham Health]) was
injected with a double-head power injector (Stellant, Medrad) through an 18-
to 20-gauge IV cannula (depending on the size of the vein) in an antecubital
vein. The right antecubital vein was preferentially used because it provides
the shortest path for contrast material through the venous system and
therefore the least dilution. When venous access could not be achieved on the
right side, the left antecubital vein was used. The saline chaser bolus was
injected through the second head of the power injector immediately after
injection of contrast material was completed. All patients received 80 mL of
contrast material and a 40-mL saline chaser bolus, both at an injection rate
of 4 mL/sec.
Synchronization between the passage of contrast material and data
acquisition was achieved with real-time bolus tracking. The arrival of the
injected contrast material was monitored in real time with a series of dynamic
axial low-dose monitoring scans (120 kV, 20-40 mAs) at the level of the
ascending aorta at intervals of 1 sec. The monitoring sequence started 5 sec
after initiation of administration of contrast material. The CTA scan was
triggered automatically by means of a threshold measured in a region of
interest (ROI) set in the ascending aorta. The size of the ROI in the
ascending aorta for the bolus triggering was adjusted to the size and
composition of the ascending aorta but was always greater than 5 mm in
diameter. The trigger threshold was set at an increase in attenuation of 75 H
above baseline attenuation (
150 H in absolute H value). When the
threshold was reached, the table was moved to the start position while the
patient was instructed not to swallow. Breath-hold instructions were not given
to the patient. CTA data acquisition was started automatically 4 sec
(caudocranial scan direction) or 6 sec (craniocaudal scan direction) after the
trigger threshold was reached. All bolus timing procedures and CTA scans were
successfully completed. No significant adverse reactions to contrast material
or other side effects occurred.
Data Collection and Analysis
Images were reconstructed with an effective slice width of 1 mm,
reconstruction interval of 0.6 mm, field of view of 100 mm, and convolution
kernel B30f (medium smooth). The images were transferred to a stand-alone
workstation and evaluated with dedicated analysis software (Leonardo, Siemens
Medical Solutions).
For each patient, site of injection and scan delay (in seconds) were
recorded. Axial images were used for the attenuation measurements. On each
40th slice (1-sec interval), beginning with the most caudal slice, an ROI was
drawn throughout the data sets in two regions: the ascending aorta to the
right ICA and the ascending aorta to the left ICA. Measurements were recorded
at the following locations: ascending aorta, aortic arch, proximal CCA (first
two measurements in the CCA), distal CCA, proximal ICA (first two measurements
in the ICA), distal ICA, carotid siphon, and intracranial part of the ICA.
Measurements in the brachiocephalic trunk were considered measurements in the
CCA.
Two observers measured and recorded all data. Attenuation was measured by
drawing a circular ROI in the center of the vessel lumen. The ROI was drawn as
large as the anatomic configuration of the lumen allowed in the axial slice.
The mean value of the measurements on the left and the right side at each time
point was calculated. Time-attenuation curves were generated for each patient,
and mean, minimum, and maximum attenuations were assessed. Because attenuation
greater than 200 H was considered optimal, the number of measurements less
than 200 H was counted. The analysis resulted in one to three measurements per
location depending on the size of the patient. For analysis of attenuation,
the measurements obtained in these locations were averaged.
Contrast material-related perivenous artifacts were graded on a four-point
scale adjusted from Rubin et al.
[6] and Vogel et al.
[8] (Figs.
3A,
3B,
3C and
3D). A score of 0 indicated no
streak artifacts and clear anatomic detail; 1, minimal streak artifacts
without notable obscuration of adjacent arteries; 2, moderate streak artifacts
partially obscuring adjacent arteries; and 3, extensive streak artifacts
completely obscuring adjacent arteries. The artifact score was assessed by two
observers blinded to scan protocol. In case of a lack of congruence, consensus
was reached. Attenuation in the SVC was measured in the most caudal slice.
Reflux of contrast material in the veins of the neck was measured in
centimeters on a coronal maximum-intensity-projection image.
View larger version (80K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3A —Contrast material-related perivenous artifacts graded on four-point
scale in four different patients. Score of 0 indicates no streak artifacts and
clear anatomic detail in 39-year-old man.
|
|
View larger version (84K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3B —Contrast material-related perivenous artifacts graded on four-point
scale in four different patients. Score of 1 indicates minimal streak
artifacts without notable obscuration of adjacent arteries in 40-year-old
woman.
|
|
View larger version (93K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3C —Contrast material-related perivenous artifacts graded on four-point
scale in four different patients. Score of 2 indicates moderate streak
artifacts partially obscuring adjacent arteries in 59-year-old man.
|
|
View larger version (116K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3D —Contrast material-related perivenous artifacts graded on four-point
scale in four different patients. Score of 3 indicates extensive streak
artifacts completely obscuring adjacent arteries in 63-year-old man.
|
|
Statistical Analysis
Baseline characteristics, attenuation parameters, and artifact parameters
in the two groups were compared using Student's t test, chi-square
test, or Mann-Whitney test. In the pair-wise comparison of attenuation
parameters, a linear regression model was used to adjust for weight and age.
The Spearman's rank correlation test was used for assessment of the relation
between artifact parameters. The software used for statistical analysis was
SPSS 11.5 (SPSS). A p value of < 0.05 was considered statistically
significant.
Results
Patients and Procedures
Patient demographics were not significantly different in the two groups
(Table 1). Scan delay was
significantly higher in group 2 (p < 0.01) because of the extra 2
sec necessary after the threshold was reached for the table to move to a
cranial start position in comparison with a caudal start position. After
correction for those 2 sec, there was no significant difference in scan delay
between the two groups.
Caudocranial Scan Direction
Mean arterial attenuation for the caudocranial scan direction was 334
± 58 H (Table 2). The
minimum and maximum arterial attenuations were 255 ± 50 H and 401
± 71 H. Fifteen measurements in five of the 40 patients (513
measurements) had an attenuation less than 200 H. These measurements were
obtained in the ascending aorta (n = 2), aortic arch (n =2),
CCA (n = 3), ICA (n = 1), carotid siphon (n = 2),
and intracranial arteries (n =5).
Craniocaudal Scan Direction
The mean arterial attenuation per patient was 303 ± 48 H
(Table 2). The minimum and
maximum arterial attenuations per patient were 212 ± 49 H and 369
± 58 H. Sixty-nine of the measurements in 16 of the 40 patients (493
measurements) had an attenuation less than 200 H. These measurements were
obtained in the ascending aorta (n = 23), aortic arch (n =
12), CCA (n =15), ICA (n = 8), carotid siphon (n =
5), and intracranial arteries (n =6).
Comparison of Attenuation Curves
Group 2, in whom the craniocaudal scan direction was used, had lower mean,
maximum, and minimum attenuations (p < 0.05) than group 1, in whom
the caudocranial scan direction was used (Tables
2 and
3, Figs.
4 and
5). After adjustment for age
and weight, no significant difference was found in mean (p = 0.07) or
maximum (p = 0.26) attenuation. Minimum attenuation remained
significantly different (p < 0.01). In both patient groups,
maximum attenuation was reached in the proximal ICA.
View larger version (9K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4 —Time-attenuation curves show intraluminal attenuation at slice
number from caudal to cranial. Slightly lower attenuation is evident for
craniocaudal scan direction in comparison with caudocranial scan
direction.
|
|
View larger version (8K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 5 —Intraluminal attenuation of group 1 (caudocranial scan direction)
and group 2 (craniocaudal scan direction) at different locations from
ascending aorta (asc ao) to circle of Willis. Maximum attenuation was reached
in proximal internal carotid artery (ICA) in both groups. Prox = proximal, CCA
= common carotid artery, Dist = distal, Intracran = intracranial arteries.
|
|
The craniocaudal scan direction resulted in significantly lower attenuation
in all locations (p < 0.05), except for the proximal CCA
(p = 0.39). After adjustment for age and weight, no significant
difference was found for the aortic arch (p = 0.07), proximal CCA
(p = 0.87), distal CCA (p = 0.17), proximal ICA (p
= 0.21), or distal ICA (p = 0.16). Attenuation in the ascending
aorta, carotid siphon, and intracranial arteries remained significantly
different (p <0.05).
Comparison of Artifacts
Attenuation in the SVC was much higher in group 1 (caudocranial) than in
group 2 (craniocaudal): 782 ± 330 H (range, 183-2,083 H) and 169
± 39 H (range, 102-288 H) (p < 0.001)
(Table 4). The mean artifact
scores were 2.5 ± 0.6 and 1.3 ± 0.9 for groups 1 and 2,
respectively (p < 0.001). All but one of the patients in group 1
had an artifact score of 2 or more. Extensive streak artifacts completely
obscuring adjacent arteries (score of 3) were seen in 21 (53%) of the patients
in group 1 and in three (8%) of the patients in group 2. Reflux of contrast
material in the neck veins measured 2.9 ± 2.4 cm (range, 0-14.4 cm) and
1.2 ± 1.5 cm (range, 0-5.1 cm) in groups 1 and 2 (p <
0.001). Reflux of contrast material in the neck veins measuring more than 2 cm
was seen in group 1 in 24 (60%) of the patients and in group 2 in 11 (28%) of
the patients (p < 0.001).
Comparison of Left and Right Injection Sides
In group 2 (craniocaudal scan direction) there was more reflux and a higher
artifact score with left-sided injection than with right-sided injection,
although the difference was significant only for reflux (p < 0.01)
and almost significant for artifact score (p = 0.08)
(Table 5). Attenuation of the
SVC was higher in patients who received a right-sided injection, although the
difference was not significant (p = 0.19).
Relationships Among Artifact Parameters
There was a significant relation between attenuation of the SVC and
artifact score (Spearman's r = 0.62, p < 0.001) (Figs.
6A,
6B). A clear cutoff point was
seen at an attenuation of the SVC of ± 200 H. Above this level,
artifacts interfered with evaluation of the arteries. There was also a
significant relation between reflux in the neck veins and artifact score
(Spearman's r = 0.52, p <0.001).
View larger version (11K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 6A —Box -and-whisker plots of attenuation in superior vena cava (SVC)
and of reflux of contrast material in neck veins according to artifact score.
Plot shows clear cutoff point at ± 200 H attenuation of SVC. Above this
level artifacts interfered with evaluation of arteries. Circle indicates
outlier.
|
|
View larger version (11K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 6B —Box -and-whisker plots of attenuation in superior vena cava (SVC)
and of reflux of contrast material in neck veins according to artifact score.
Plot shows greater amount of reflux is associated with higher artifact score.
Stars indicate extremes.
|
|
Discussion
Artifacts caused by the inflow of undiluted high-density contrast material
in the SVC during thoracic helical CT have been described
[5-7,
9]. These artifacts can cause
obscuration of enlarged lymph nodes, incomplete characterization of axillary
and mediastinal masses, and obscuration of vascular lesions. During CTA,
artifacts can especially obscure the ascending aorta and proximal supraaortic
arteries and thus hide or suggest stenosis or occlusion of the origin of the
supraaortic arteries. Perivenous artifacts are frequently seen when the start
of a CT scan occurs before injection of contrast material ends. In case of
delayed start of data acquisition, after the end of contrast material
injection, stasis of contrast material in the major thoracic veins can cause
artifacts.
Rubin et al. [6], who
performed helical CT with an injection duration of 40 sec and a scan delay of
25 sec, found that 3:1 dilution of contrast material resulted in diminished
perivenous artifacts. To keep the same total injected iodine dose with a 3:1
dilution, the total injected volume and the injection rate have to be
increased four times. In CTA, however, dilution of contrast material is not an
option because the already high injection rate should be increased to more
than 10 mL/sec in maintaining the injected iodine dose.
Haage et al. [5] tested the
effect of the addition of a chaser bolus to the main contrast material bolus
and subsequent reduction of contrast material. They found a reduction in
perivenous artifacts. This result can be explained by the reduction in total
iodine concentration and shorter contrast material injection time rather than
use of a chaser bolus after the main bolus.
In theory, to prevent perivenous artifacts, a chaser bolus is useful in CTA
only when the scan starts after injection of contrast material ends. In our
study, the optimal scan delay was within the injection period of 20 sec in
most of the patients. Although use of a chaser bolus in CTA of the supraaortic
arteries leads to optimal use of contrast material
[4], it does not decrease
perivenous artifacts.
To synchronize data acquisition relative to optimal arterial enhancement,
the scan direction in CTA usually is in the direction of blood flow
[10]. With 16-MDCT, scanning
time in CTA of the supraaortic arteries has decreased to less than 15 sec.
This change may allow reversal of the scan direction without a compromise in
vascular attenuation. In addition, perivenous artifacts may decrease because
of the delay in scanning of the apex of the thorax when a craniocaudal scan
direction is used.
Our study showed that a craniocaudal scan direction resulted in slightly
lower attenuation of the carotid artery, although attenuation remained high
enough for good evaluation, in comparison with a caudocranial scan direction.
With both scan directions, peak attenuation was at the level of the proximal
ICA, which is the most relevant site. After adjustment for age and weight, no
significant difference in mean or maximum attenuation was found for the two
scan directions. Minimum attenuation, however, was significantly lower for the
craniocaudal scan direction. The explanation is that the scan direction is the
opposite of the direction of blood flow, and to have maximal enhancement at
the halfway point (the level of the proximal ICA), the scan is at the cranial
level a little too early and at the caudal part a little too late for optimal
enhancement. Therefore, attenuation at the beginning and at the end of the
craniocaudal scan, and thus minimum attenuation, is lower than for a
caudocranial scan. This factor is also reflected in the significantly lower
attenuation, after adjustment for age and weight, at the ascending aorta,
carotid siphon, and intracranial arteries and the lack of difference in
attenuation at locations in between.
We found that a craniocaudal scan direction resulted in much lower
attenuation of the SVC, which resulted in fewer perivenous artifacts. By the
time the craniocaudal scan reaches the apex of the thorax, injection of
contrast material has ended, and contrast material has been flushed from the
veins by the chaser bolus.
There was a tendency to increased artifacts with left-sided injection
compared with right-sided injection, although this difference was not
significant (p = 0.08 for the craniocaudal scan direction). The
explanation is that venous return in the left subclavian vein, because of its
transverse course into the SVC, may be more likely to be affected by changes
in intrathoracic pressure and to be compressed by normal structures, such as
the aorta. These effects can cause more pooling of contrast material in the
subclavian vein and more reflux in the neck veins with left-sided injection
than occurs with right-sided injection
[11]. We found more reflux in
patients in the craniocaudal scan direction group who received a left-sided
injection (p < 0.01) and a significant relation between reflux and
artifacts (p < 0.01).
A limitation of our study was that the groups were not randomly allocated.
Baseline characteristics, however, were not significantly different. Another
limitation was that our results probably will not apply when bolus triggering
is not used to optimize the timing of data acquisition. The timing for a
craniocaudal scan direction has to be more precise than for a caudocranial
scan direction. Without bolus triggering, data acquisition may be too early
for good arterial attenuation in the intracranial arteries and too late for
good attenuation in the aorta.
In conclusion, we advocate the use of a craniocaudal scan direction in
16-MDCT angiography of the supraaortic arteries. Right-sided injection is
preferred over left-sided injection. This protocol results in good arterial
attenuation, low attenuation of the SVC, and few perivenous artifacts and
facilitates evaluation of the ascending aorta and supraaortic arteries.
References
- Warlow C, Sudlow C, Dennis M, Wardlaw J, Sandercock P. Stroke.
Lancet 2003; 362:1211
-1224[CrossRef][Medline]
- Ersoy H, Watts R, Sanelli P, et al. Atherosclerotic disease
distribution in carotid and vertebrobasilar arteries: clinical experience in
100 patients undergoing fluoro-triggered 3D Gd-MRA. J Magn Reson
Imaging 2003; 17:545
-558[CrossRef][Medline]
- Rouleau PA, Huston J 3rd, Gilbertson J, Brown RD Jr, Meyer FB,
Bower TC. Carotid artery tandem lesions: frequency of angiographic detection
and consequences for endarterectomy. Am J Neuroradiol1999; 20:621
-625[Abstract/Free Full Text]
- de Monyé C, Cademartiri F, de Weert TT, Siepman DA, Dippel
DW, van der Lugt A. CT angiography of the carotid artery with a
16-multidetector-row CT scanner: comparison of different volumes of contrast
material with and without bolus chaser. Radiology2005; 237:555
-562[Abstract/Free Full Text]
- Haage P, Schmitz-Rode T, Hubner D, Piroth W, Gunther RW. Reduction
of contrast material dose and artifacts by a saline flush using a double power
injector in helical CT of the thorax. AJR2000; 174:1049
-1053[Abstract/Free Full Text]
- Rubin GD, Lane MJ, Bloch DA, Leung AN, Stark P. Optimization of
thoracic spiral CT: effects of iodinated contrast medium concentration.
Radiology 1996;201
: 785-791[Abstract/Free Full Text]
- Loubeyre P, Debard I, Nemoz C, Minh VA. High opacification of hilar
pulmonary vessels with a small amount of nonionic contrast medium for general
thoracic CT: a prospective study. AJR2002; 178:1377
-1381[Abstract/Free Full Text]
- Vogel N, Kauczor HU, Heussel CP, Ries BG, Thelen M. Artefact
reducing in diagnosis of lung embolism using spiral CT with saline bolus [in
German]. Rofo 2001;173
: 460-465[Medline]
- Nakayama M, Yamashita Y, Oyama Y, Ando M, Kadota M, Takahashi M.
Hand exercise during contrast medium delivery at thoracic helical CT: a simple
method to minimize perivenous artifact. J Comput Assist
Tomogr 2000; 24:432
-436[CrossRef][Medline]
- Fleischmann D. Use of high concentration contrast media: principles
and rationale—vascular district. Eur J Radiol2003; 45[suppl 1]:S88
-S93
- Sakai O, Nakashima N, Shibayama C, Shinozaki T, Furuse M.
Asymmetrical or heterogeneous enhancement of the internal jugular veins in
contrast-enhanced CT of the head and neck.
Neuroradiology 1997;39
: 292-295[Medline]
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
Top
Abstract
Introduction
