Full text of DOI:10.1177/000331970005101102

Evaluation ofElectron
Beam
Tomographic
with
Coronary
Arteriography
in
Three-Dimensional Reconstruction
Subjects
Healthy
Bin
MD*
Lu,
Dai,
MD*
RuPing
HuaBai, BS*
Sha MS*
He,
MD*
BaoLian
Jing,
NanZhuang, MD*
MD†
MD†
MD†
RunLin
YueJin
JiLin
Gao,
Yang,
Chen,
and MatthewJ.
MD‡
Budoff,
and
CHINA
CALIFORNIA
TORRANCE,
BEIJING,
ABSTRACT
In this
the authors evaluated the
characteristics ofcontrast-enhanced
study,
performance
electron-beam
with
in
the
three-dimensional reconstruction
(EBT)
tomography
lumen in
defining
with normal
coronary artery
healthy subjects. Thirty patients
coronary arteriography
coronary
selective
underwent contrast-enhanced
(SCA)
angiograms by
EBT
examination. Measured included
of luminal
enhancement,
parameters
degree
intravascular contrast-to-noise ratio
and diameter and
of
visualized
(CNR),
aortic blood and
length
attenuation
lumen. Ventricular
were found
cavity,
pool,
coronaryartery
to be
different before and after intravenous
of contrast material
within each
vessel
significantly
injection
segments
CNR decreased from
to distal
in the
0.001).
0.001),
(p <
(p
proximal
occurring
<
with
a
of 11.2 &plusmn;2.3
left anterior
peak
proximal
descending
Luminal
to
a
low of
2.0in the distal left circumflex
4.8
(LCX).
(LAD)
coronary
artery
diameters visualized
EBT
EBT
difference with that of SCA
had no
0.05).
by
(p >
significant
with three-dimensional reconstruction allows for noninva-
Therefore,
sive
angiography
arteries in
of the
coronary arteriography revealing long
major coronary
segments
normal
subjects.
From
of
Division
and Chinese
Academy
and
Harbor-UCLA Research
FuWai
Medical
Education
the
of
and internal
Medicine,
Cardiology,
of
’<Department
Radiology
Cardiovascular Institute and
UnionMedical
Hospital, Peking
College,
and the *Division of
California.
Sciences,
China;
Beijing,
Institute, Torrance,
Westminster
Cardiology,
Glen Cove
708
Glen
NY
U.S.A.
11545,
&copy;2000
Inc.,
Avenue,
Head,
Publications,
895
Downloaded from ang.sagepub.com at UNIV MASSACHUSETTS AMHERST on April 1, 2015

’
’
mmwith no
ECG
ness was 1.5
intersectional
gap.
Introduction
the predicted R-Rinterval of
at 80%of
heart beat was used.
triggering
The rawdata ofserial
a
every
Electron beam
with electro-
beenused for
sectional
were reconstructed
kernel and
&dquo;normal&dquo; reconstruction
both
(EBT)
using
tomography
images
has
(ECG)
cardiographic
triggering
algorithm.
calcifi-
ac-
and
cations
(CAC)
The
were used for
field of view of
artery
years.1
image
detecting
coronary
quantifying
than
following settings
the
15
matrix size of
for more
10
cm;
Being
quisition
:
x
x
the heart
fastest noninvasive
method of
512
and
size of
mm2.
0.29
0.29
512,
imaging
pixel
with few
or
artifacts from cardiac
motion
Patients underwent examinationheadfirst in
pulsation
stan-
EBT
a
is established as
a
in
a
and were scanned
cranio-
trained and
&dquo;gold
respiration,
supine position
dard&dquo; for CAC
CAC scores
caudal
were
detection.2Although
direction. The
patients
well with the total atherosclerotic bur-
correlate
den 3,4 and
to hold their breath in
requested
deep inspiration
future cardiac
that the amount of
it has
calcifica-
before the examination. The first
ization of the heart
was local-
step
events,’
predict
been demonstrated
scan.
position using preview
tion does not correlate well with the
of
an-
The second
was determination of the con-
severity
step
transit
stenosis of
giography
a
lesion.6 Selective
the
mode
thickness
time;
given
coronary
trast-agent
ofthe flow
multiple-slice
remains the
standard&dquo; for
but it is
was
with
a
slice
(SCA)
the
used,
&dquo;gold
study
sec-
of mmand
a
of mm
other
after
7
4
lumen,7,11
visualizing
coronary
artery
gap
every
invasive and
tion.
time was 50 msec for two si-
expensive.
Acquisition
multaneously acquired
(Ultravist300,
Recent studies
have used
media
sections. Contrast
was
contrast-enhanced,
EBT
three-dimensional
for detect-
ECG-triggered,
Schering, Germany)
injected
and
different
and several
into the antecubital vein at
a
rate of 8
mL/sec
stenosis,9
grading coronary
ing
three-dimensional
reconstruction
and
the
tained. The
dose was
mL. Adelay of
from
7 sec
35
proto-
study
totally
cols have been used and
Our
charac-
the
to the first
main-
was
reported. 10-13
injection
acquisition
was
to assess and
CT time was determined
designed
compare
peak
using
delay
teristics ofEBT
with
differ-
the
curve
the flow scan
coronaryangiography
time-density
plus
ent three-dimensional
andto further as-
time
scan
20
time.
and this was used as the
7
(mean
s),
protocols
the
sess
and limitations of EBTfor visual-
ability
delay
The third step was
mensional data
the lumina of
arteries.
coronary
ofthe three-di-
acquisition
izing
which consisted of axial
breath hold in
sec-
set,
obtained
tions
during
deep inspi-
A
ration.
were
total of 50 to 60 cross-sectional
images
with an
to 50 sec. The contrast
time of about
acquired
acquisition
was
administered
flow rate of
4
40
Materials andMethods
Subjects
agent
with
the antecubital
a
vein,
though
mL/sec
and
a
total dose of 100-150
the
mL,
using
time.
determined scan
delay
previous
We
32-73
with
lective
studied 30 cases
3
(27
male,
female,
age range
Image Processing
transferred to
53.5
9.6
old)
old,
years
normal
average
years
coronaryarteries by
conventional se-
The axial
workstation
mensional
maximum
a
were
images
special
All the sub-
Three-di-
(SCA).
Canada).
ISG,
coronary angiography
had normal cardiac
(Allegro,
reconstructions
function without
was
were
jects
heart disease. EBT
formedwithin 11.6 5.0
any
per-
performed using
and
curved
which was used to
(MIP)
coronary
angiography
intensity
projection
before or after SCA.
reconstruction
(CPR),
days
planar
reconstruct
interest ofthe
of 50-100 Hu
of
levels
ar-
regions
coronary
teries. Window
width was 500
measure
(window
Data
of
Acquisition
were used to observe and
Hu)
Beam
Electron
Tomography
the visualized
lumen.
coronary artery
The
electron
mode of the
in
a
contrast-enhanced
Reconstructions
rendered
curved
were
single-slice
plane
scan-
beam
that followed the
course of each coro-
(CT)
12.34
computed tomography
projected
version
was used. Slice thick-
software
left
main
set;
the data
ner,
South San
(Imatron,
C-150LXP,
(LM)
coro-
nary artery through
and left
anterior
CA)
Francisco,
(LAD),
descending
right
896
Downloaded from ang.sagepub.com at UNIV MASSACHUSETTS AMHERST on April 1, 2015

EBT
results of
were
The
were
three-dimensional
with
and left circumflex
(LCX)
(RCA),
image.
naryartery
rendered
in
a
blinded fashion
the axial
The third
was shaded surface
with CPR
using
angiography
where
compared
separately
<
0.05
the Student’s
t
cross sections as reference
three-dimensional method
which had been
test,
differences.
p
signi-
images.
using
fied statistical
in
other
SSDwas
(SSD),
reported
display
of
threshold
studies.9 Thereconstruction
in our
55.2
10.3 Hu
group.
Results
Data
Analysis
all
well tolerated
enhancement
30 sub-
EBT scan was
The
by
Both on
con-
and
Measurement.
Attenuation
plain
ofthe heartwasper-
contrast at-
Contrast
jects.
axial EBT
the attenua-
of aortic root
trast-enhanced
sections,
without
side effects. The
LV
formed
any
ascending
Hounsfield
tion values
lumen
ocardium
(in
units)
(LV),
of
tenuation of
aorta,
cavity, Myo
lateral
left ventricle
LV
(AO),
my-
ofcoro-
and RCAwere
LVwall, and
lateral
29.2-54.3
101.0-244.3
LM, LAD, LCX,
and
proximal segments
(Myo),
Hubefore contrast administration and
were measured. Theresults be-
and enhanced scanwere com-
arteries
(CA)
nary
after contrast administration
Attenuation in aortic
and RCAincreased to exceed 200
Hu
tweenunenhanced
the
<
1).
0.001)
root,
(p
(Figure
t
test.
paired
pared using
LV
LM,
cavity,
Luminal contrast
whereas LAD and LCXincreased close to
190
Contrast-to-Noise Ratio
(CNR).
Hu,
Hu,
increased to the
and
attenuation
was defined as the difference between the mean
an oval of interest and the
myocardial
lowest
CT value of
(101 Hu).
degree
region
LAD
dominant
or
The CNRwas
in the
mean CT value of the
such as
tissue,
highest
proximal
adjacent
in the distal LCX
and lowest
C
(4.8
(11.3 3.0)
vein, ie,
coronary
fat,
myocardium,
differences of
is
There were no
C
is the luminal
2.0).
(where
contrast, DL
significant
DL - DST
and
the CNR
and RCA
than
the
is the
ofvisualized
of
LAD,
LM,
lumen,
DST
among
proximal
coronary
density
but all three were
LCX
The stan-
tissue).
higher
decreased
The CNR
density
surrounding
(p > 0.05),
in
deviation of attenuation
measured
dard
0.001).
systematically
within each
to distal
points
(p <
was taken as
a
measure of
the CNRwas defined as the contrast
noise
from
(N).
So,
proximal
coronary
image
divided
2).
0.001)
(Figure
by
artery (p
<
in
mid-
There were
CNR
and distal
The CNR
CIN.
segments
with the
noise, ie,
correlations amongimage
proximal
positive
noise
and
surface area
(BSA);
of the
arteries
(N),
dle,
body
coronary
weight,
r
r
and
0.53.
was
t
test and the cor-
correlation values were
0.60
compared
paired
had
CNR
correlation
-0.51 and r =
and
relation
noise, CNR,
However,
with
image
negative
among image
weight,
and BSA
surface areas
were evaluated.
(r
(BSA)
-0.47);
weight
body
the
or
was
thus the
the
the lower
BSA,
(Figure
higher
weight
EBT
ofthe
3).
(in
image quality
of Electron
and Selective
Beam
Comparison
Tomography
The
luminal diameter
millime-
apparent
Coronary
Angiography
of the
of each
ters)
proximal portions
coronary
images
EBT
on different three-dimensional
The
were reconstructed and
two
images
analyzed
artery
(SSD,
0. 5,
and
4.4
were
4.1
4.0
CPR)
MIP,
and
and 3.6
LM,
0.6,
by
radiologists
experienced
independently
The
3.3
3.3
angiographyresults.
whowereblinded to the
0. 4; LAD,
0.4,
0.7,
and
obtained in all
3.1
3.2
3.3
LCX,
0.6,
4.0
0.8,
0.6;
subjects
coronary angiograms
and 3.8
two
were
and evaluated
3.9
0. 8,
0. 8; RCA,
0. 9,
0.9,
performed
by
experi-
of
of which were
different than
The diameter and
none
enced
length
significantly
cardiologists.
the visualized normal
arteries were av-
selective
3.7
results
(4.3
0.05.
correlation value
whereas SSD had
0.7,
The
coronary
the standard devi-
coronaryangiography
>
and
3.3
mean
or minus
4.0
0.5,
0.9,
1.1 ),
p
eraged by
plus
at
a
had the
was measured
cm from the
CPR method
compared
and the diameter
ation,
point
highest
0.90),
0.87).
with
SCA
of each
1-1.5
(r
origin
approximately
The
of
r
value
the lowest
branch downto the distal
(r
length
segments.
was then
lumina
Three-dimensional
could reveal the
the visualized
imated
ous
approx-
images
coronary
the
of
ofthe
lumen
ar-
length ofmanycontigu-
long segments
coronary
the arteries
seen with EBT
major
bymeasuring
each different
on
teries. The
of
length
straight-line segments
897
Downloaded from ang.sagepub.com at UNIV MASSACHUSETTS AMHERST on April 1, 2015

1.
Contrast enhancementvalues measured
left ventricle
aorta
(LV),
Figure
using
left anterior
ascending
(LAD),
left
(AO),
circumflex
ventricular
left main
left
(LM),
myocardium
(Myo),
right coronary artery
and after contrast
descending
and
The contrast attenuation
was
(LCX),
(RCA).
(Hu)
significantly
<
different before
administration
(Con-)
(Con+)
0.001).
agent
(p
2.
noise and
contrast-to-noise ratios
for
(CNR)
vessel
Figure
Image
coronary
within each
segments. Image
noise did not
increase from
to distal
significant
proximal
coronary
segments
artery
0.001).
>
<
whereas CNRdecreased
from
to distal
0.05),
left main
(p
LM
systematically
(p
proximal
descending coronary artery,
segments
LAD left anterior
LCX left
coronary artery,
circumflex
RCA
P
segment, M middle
proximal
coronary artery,
segment, D distal
right
coronary
artery,
segment.
898
Downloaded from ang.sagepub.com at UNIV MASSACHUSETTS AMHERST on April 1, 2015

area
correlation with
and
surface
3.
noise has
a
(A)
(B),
weight
body
Figure
whereas
Image
positive
has
a
correlation.
contrast-to-noise ratios
image
body
negative
BSA
surface
CNR contrast-to-noise
Hu Hounsfield units.
ratio,
area,
<
for
im-
and
and
difficult
for
was shorter than
SCAvisualized
than
horizontal,
reconstruction,
0.001),
(p
with
this method. The
was
MIP was
and CPR
SSD
MIP reconstruction
better
evaluating
possible
protocols
LAD
could
ofthe
of the
lumina
4).
(Figure
length
significantly
(p < 0.05)
than those revealed onLCXand
be-
reveal about
LAD,
ofthe lumen
segments
78.7%
RCA,
greater
the RCA.
of
cause LADwas
andwithout
65.6% of
the
and
64.3%
LCX,
posterior
mainlyperpendicular
motion artifacts
is
lumen
the
of
The
5).
(Figure
artery
descending
significant
899
Downloaded from ang.sagepub.com at UNIV MASSACHUSETTS AMHERST on April 1, 2015

Visualized luminal
4.
evaluated
conventional
and
Figure
length
by
coronary angiography
EBTthree-dimensional
of the
EBTthree-dimensional
but the shown were shorter than in
segments
0.001),
could reveal the lumen of
images.
images
long
arteries,
segments
major coronary
<
<
in all vessels
MIPwas better than SSD and
CPRprotocols
0.05).
angiography
CPR
(p
(p
MIP maximum
curved
EBT electron beam
reconstruction,
SSD shaded surface
planar
projection,
tomography,
intensity
display.
the thinnest collimation
and fastest
to obtain the
Discussion
(1.5
mm),
volume scan time
(100 msec/slice)
temporal
and
resolution
highest spatial
possible.10
Reduction of section thickness is an effective
Electron
in
beam
method of
and
reducingvolume-averagingeffects
tomographycoronaryangiography
wasfirst
Chernoff
or contrast-to-noise
healthy subjects
reported by
decreasing signal-to-noise
in
in 11
The contrast enhance-
1997
ratio. We utilized an
at an
80%
in which the
ECG
heart
subjects.14
triggering
CNRmeasured on EBT axial
coro-
R-R interval of
ment,
images,
every
cycle,
and visualized luminal
three-dimensional SSD
end
of
have minimal heart
diastole should
diameter,
nary artery
stage
motion.14
measured on EBT
were evaluated. No
formed between different
construction methods and
A
recent
demonstrated less car-
length
study
was
diac
at
motion
to 50% of the cardiac
45%
images
comparison
per-
cycle,
three-dimensional re-
which
Section
now
for these
smaller
is
studies.ll
preferred
with
a
fromthe
of
unintended -
angiography.
acquisition
delay
R-R
wave reduces the likelihood
with variations in the R-Rin-
systolic
terval,
quisition
tigated
acquisition
and Its Effects
Image Quality
of
of
but the effect of the
ac-
timing
image
The
tocols
EBT
three-dimensional
has not been
on motion artifacts
inves-
image quality
pro-
on its two-dimensional
are
dependent
systematically
cross-sectional
are so
Because
arteries
on EBT
coronary
Goodcontrast enhancement
images.
coronary
and movewith heart
con-
for
three-dimen-
is
small,
tortuous,
angiography
very important
we selected
EBT
scan
the
with
sional
Contrast attenuation ofthe
reconstruction.
traction,
protocol
900
Downloaded from ang.sagepub.com at UNIV MASSACHUSETTS AMHERST on April 1, 2015

5.
Selective
/~
Figure
coronary
showed
a
man
arteriography
with normal LAD
52-year-old
LCX
(arrow),
(arrowhead)
and normal RCA
EBTreconstruction
(A)
(B).
of the same case showed normal LAD
(small
(wide
LCX
and
RCA
(arrow
arrow),
arrow)
head),
(C).
on MIP
LADwas
image
reconstructed on CPR
and the
(D),
image
LAD
RCA
SSD
showed
ofthe
(E)
image
patency
(arrow
LCX
head),
CPR curved
and
(small
arrow),
(wide arrow).
reconstruction,
planar
EBT electron beam
LAD left anterior
tomography,
LCX left
descending,
MIP
maximum
circumflex,
RCA
intensity projection,
right
coronary
SSD shaded surface
artery,
display.
901
Downloaded from ang.sagepub.com at UNIV MASSACHUSETTS AMHERST on April 1, 2015

onthe
and
dose,
coronary
artery depends
density,
of
Three-Dimensional Protocols
Accuracy
rate of
contrast
there is no
agent injection;
signif-
icant
in the reformatted two-
correlation with
resolution
and BSA
correlation with
(r
image
0.12),
noise.
High
spatial
weight
but there
a
dimensional
were needed for
is
axis)
(x,
y
strong
images
and
for
a
CNRof EBT
three-dimensional reconstruction
z
So,
can
(x,
given subject,
y,
high
images
can avoid
be obtained
in the
resolution
z
axis
mis-
and dose of
axis).
byusinghigh density
High
contrast
minimize
accurate
volume
visualization of
three-dimensional
effect,18
registration,
agent.
Other factors that
partial
EBT
and
effect
image
improve
coronary
quality
are
and
lumens.
arrhythmias (including tachycardias,
artery
Among
proto-
visualize
atrial
to
and
who are unable
and SSD can
fibrillation),
cols, MIP, CPR,
patients
accurately
the
maintain breathhold. Breathhold-
diameter as
with
adequately
coronary artery
compared
before examination is
the standard
results.19
The CPR
ing training
necessary.
angiographic
Schmermund 15
travenous
that
administered
in-
of
0.4-1.0
method was the most accurate method for visu-
reported
mg
to
was
the
luminal diameter or stenosis of
because no
Atropine
alizing
patients
prox-
density
<
with rest heart rates
the
imal
in-
mean
75
beats/min;
arteries,11
coronary
heart
81
rate increased from
10
to
formation is lost in CPRreconstruction. For coro-
69
beats/min
13
0.001 vs baseline heart
and total
luminal
for
MIP
the best
is
beats/min
visualization,
nary
protocol
(p <
test),
34
length
rate
t
voxial
identification,
using paired
breathholding
high-density
time
decreased to
5
sec
25 to
because contrast
artery lumenwith
enhance the
agents coronary
42).
(range
faster heart rate
increases the
than that ofthe
However,
higher density
coronary
motionboth in
and
and
It
is
more
for
artery
speed
scope,
surrounding myocardium.
impossible
coronary
motion
times to
artifacts occurred.
CPRto
all
the branches of
Reduction of scan
render
arter-
the
of 30-50
motion
ies
of
because the tortuous course of the
msec is
to
vessels,
range
coronary
necessary
thus
eliminate
CT with current
500 msec will
arteries well.
affect
for
cause of
the
information is lost be-
and,
artifact 14;
SSD,
density
spiral
scan
times of 250-
acquisition
thresholding.
be unable to
and
Schmermund
the
that
image
coronary
modalities
Moshage
reported
steno-
SSD was better for
Image
postprocessing
identifying
coronary
two- and three-dimensional
but overestimation of the
stenosis and in-
sis,
image quali-
raw CT data
should be recon-
to differentiate between the visualized
therefore,
ty ;
ability
structed
a
smaller field of view
lumen
wall
calcifications were
or
and
(12.7
x
using
coronary
15.0
matrix size
&dquo;cone-beam&dquo;
reconstruction threshold for
the three-dimensional
Achenbach
that CPR
cm),
(512
512),
bigger
problematic.9>15
suggested
and
was
a
method for
cal-
kernal,
&dquo;sharp&dquo;
algorithms.
good
identifying coronary
Window level and
cifications and
cifications for
From our
the
effects from cal-
diminishing
evaluating
the luminal
stenosis.11
reconstructing
image
should be chosen
to luminal
MIP
has both the advan-
attenua-
according
experiences,
tion of the
vessels.
of
and
CPR
and should be
considered
three-di-
coronary artery
SSD,
target
tages
as the first
The measured CNRwas
in
the
choice for
mensional reconstruction.
higher
ofthe
vessel lumenis
proxi-
ar-
mal
than in
the distal
portions
proximal
volume
coronary
because the
with less
teries,
larg-
effects. Also
motion
er,
partial
Values and Limitations ofthe
Protocols
artifacts
due to cardiac
and
increase
ofeachvessel are
pulsation may
Several authors
that EBT with
distally,
usually
nective
proximal
portions
recently reported
surrounded
low-attenuation
con-
different
ods can
three-dimensional reconstruction meth-
by
fatty
whereas mid and distal
attenuation
to visualize the
be used
tissue,
portions
routinely
major
are often near
vessels of the norma114 or stenotic human coro-
higher
myocardium,
or atrium. 14
arteries should use
ofdistal
coro-
tree.9-13 The
and
vein,
coronary
nary
bolus with
LCX has been
to
Imaging
nary artery
specificity,
diagnostic sensitivity,
a
continuous
contrast
contrast
50%
(for >
stenosis)
accuracy
The
agents.
be more difficult
have
94%,
to
to
74%
been
and
be
to
79%
higher
89%,
density
reported
demonstrated to
87%,
respectively.
visualize than the LAD
LM in
and
with other
ul-
modalities,
previous
imaging
and
Compared
studies
CNRfound in
EBT
the LCX
The lower
CT can-
lumen.10
has
nuclear
medicine,
the
using
trasound,
angiography.11,13
may
spiral
account for the de-
not well
visualize
resonance
coronary artery
creased
to
visualize
this vessel.
(MRA)
ability
Magnetic
angiography
902
Downloaded from ang.sagepub.com at UNIV MASSACHUSETTS AMHERST on April 1, 2015

the
of attenua-
calcified
tion for contrast-filled
for
been
range
lumina.14
with
overlaps
plaques
coronary artery
research
diagnosing
reported
In re-
but it is still in the
EBT
coronary
disease,
technologic
kind of
was
EBT
three-dimensional
not suited to
be
to
cent
can
of
studies,
any
protocol
angiography
performed
stage .20,21
angiography
reconstruction
arteries
visualize
coronary
segments
longer
se-
conventional
but could
EBT
MRA.22,23
for
than that
has
that of MRA.14
The
for old
solved.9-14
limited
hold and
ial. Other
tion for
completelyreplace
previously
reported
lective
someinvasive
resolutions than
and
replace
coronary arteriography
spatial
higher temporal
for
example, follow-up
transluminal coro-
procedures,
after
limitations of20-slice
percutaneous
or
scanning
procedures
previous
EBT scanners has been
(PTCA)
(CABG) .25,26
coronary artery by-
nary angioplasty
generation
The current scanner can
a
un-
perform
pass graft
breath
during
single
coverage
spatial
of contrast mater-
pass
the first
during
limitations include low
resolu-
spatial
or small branches of
distal
visualizing
(from
arteries,
misregistration
respira-
with the
coronary
Acknowledgments
_
tory movementor
poorsynchronization
heart
resolution
which is
and motion artifact
cardiac
(from
cycle),
Theauthors
fromthe
assistance
the clinical
and
sec-
low
ms
caused
time
contraction)
acknowledge
support
by
temporal
of
Fu
or
scan
(100
cardiology,
image),
per
department
long
Wai
of
catheter
and LCX arteries.24
in the RCA
tion
most
lab,
Hospital,
coronary
China.
apparent
reconstruction
For three-dimensional
of
methods,
Beijing,
or even MIP studies
the
SSD,
CPR,
knowledge
rendering
of cardiac
sections of each
some
requires
because the cross
to be identified
anatomy,
coronary
MatthewJ.
Harbor-UCLA
Education
FACC
Research and
Budoff, MD,
on all axial
observed window
has
The
artery
correctly
Institute
West Carson
of
interest,
images.
region
RB-2
threshold should be se-
and reconstructed
suitably.
1124
Street,
level,
CA
lected
90502
Torrance,
calcifications have been
Coronary artery
E-mail:
in some
to
budoff@flash.net
attempts
identify
opacified
of attenuation for
proven problematic
evaluate the stenosis of the
or
because the
lumen,10,11
range
References
Goodman
et al: Predictive
arteries:
coronary
Detec-
et al:
5. Arad
value
19-month
Circulation
1. Tanenbaum
Kondos
Veselik
Y,
LA,
K,
the
KE,
arteries
coronary
SR,
GT,
in
Spadaro
of electron beam CT of
of
ul-
with
tion of calcific
trafast
by
deposits
and correlation
1173
follow-up
asymptomatic subjects.
computed tomography
Am
1996.
J
Cardiol
63:870-872,
1989.
93:1951-1953,
angiography.
2. Wexler
Froenlich
et al:
6. Mautner
Mautner
SL,
GC,
J,
Coronary
al:
et
Crouse
J,
L,
B,
Coronary
Brundage
assessment with electron beam
correlation.
calcification
artery
CT and
calcification:
epidemiology,
artery
Pathophysiology,
Radiology
A
state-
histomorphometric
1994.
and clinical
methods,
imaging
implications.
192:619-623,
the American
for health
From
ment
professionals.
Heart Association. Circulation
1996.
94:1175-1192,
Judkins MP: Selective
7.
coronary arteriography:
transfemoral
A
Radiology
technique.
percutaneous
89:815-824, 1967.
et al:
Breen
artery
Schwartz
RS,
3.
4.
JF,
PF,
Coronary
Sheedy
detected with ultrafast CT as an
disease.
calcification
indication of
185:435-439,
Radiology
8. Marcus
estimates of
battered
Skorton
Johnson
et al: Visual
MR,
artery
coronary
1992.
ML,
DJ,
diameter
"A
stenosis:
percent
coronary
Am
Coll Cardiol
11:882-
standard."
J
gold
et al: Ultrafast
calcium re-
Janowitz
G,
WR,
detected
AS,
Kaplan
Agatston
885, 1988.
coronary
computed tomography
Achenbach
stenosis: Three-dimensional
Seese
et
al:
of
extent
arterial ath-
1994.
9.
the
flects
S,
B,
WE,
Moshage
artery
Coronary
with elec-
coronary
angiography
erosclerosis. Am
J Cardiol
74:1272-1274,
imaging
903
Downloaded from ang.sagepub.com at UNIV MASSACHUSETTS AMHERST on April 1, 2015

contrast
en-
to the evaluation ofcarotid
trocardiographically triggered,
agent
angiography: Application
stenosis.
electron-beam CT.
hanced,
1993.
196:707-714,
et al:
Evaluation of
M,
189:211-219,
Radiology
artery
Radiology
1995.
18. Thomas
beam CT
Ritman EL: Anelectron
PJ,
approach
CH,
McCollough
10. Nakanishi
Ito
Imazu
for transvenous
arteri-
1995.
T,
K,
stenoses
coronary
19:383-389,
electron beam
Assist
and
CT
ography. J ComputAssist
coronaryartery
multiplanar
21:121-127, 1997.
using
Tomogr
reformation.
J
Tomogr
Comput
Brown
Bolson
et al: Lumendiam-
EL,
19.
JT,
BG,
Dodge
of
eter
normal humancoronary arteries: Influence of
11. Achenbach
et
al: Curved
reconstructions for the evaluation of
S,
W,
D,
Ropers
anatomic
and left ventricular
Moshage
sex,
variation,
age,
hy-
con-
or dilation. Circulation
multiplanar
1992.
86:232-246,
pertrophy
trast-enhanced electron beamCTof the
ar-
coronary
20. Pennell
onance
Firmin
et al:
res-
DJ,
imaging
J,
DN,
Keegan
of
Magnetic
teries. Am
J
Radiol
170:895-899,
1998.
arteries:
and
coronary
results. Br Heart
Technique
12.
Chernoff
stenoses: Assessment with
Adams
DM,
et al:
JR,
contrast-enhanced
GP,
Reddy
artery
Coronary
J
70:315-326, 1993.
preliminary
21.
Li
et al:
WJ,
W,
NG,
Boyte
Manning
electron-beam CT and axial reconstructions. Radiol-
1998.
Fat-suppressed
breath-hold
resonance
magnetic
coronary angiogra-
208:167-172,
ogy
Circulation
87:94-104, 1993.
phy.
13. Budoff
Oudiz
Zalace
RJ,
et al:
Intravenous
con-
MJ,
CP,
22. Duerinckx
MR
UrmanMK: Twodimensional
three-dimensional
trast
AJ,
coronary
coronary angiography using
of initial clinical results.
enhanced electron beam
angiography: Analysis
computed tomography.
1999.
1994.
Am
J
193:731-738,
Paschal
MBM,
Cardiol
Radiology
83:840-845,
23. Hofman
Li
et
MRI of coro-
al:
14. Chernoff
Ritchie
CJ,
CT
CB: Evaluation of
in
CB,
D,
DM,
Higgins
2Dbreath-hold
naryarteries:
vs 3Drespiratory-gated
electron beam
coronary angiography
healthy
J ComputAssist Tomogr
1995.
Am
J
1997.
19:56-62,
acquisition.
Radiol 169:93-99,
subjects.
Ritchie
CJ,
speeds
Radiology
Godmin
24.
Crawford
et
al: Minimum
15.
16.
17.
Schmermund
et
al: Intra-
coro-
JD,
CR,
ofmotion
A,
BJ,
computed
segmental
PF,
Sheedy
Rensing
scan
for
in
CT.
venous
artifacts
electron-beam
for
suppression
tomographic
of
coro-
185:37-42, 1992.
nary
angiography
analysis
stenoses. Am
J
Coll Cardiol
31 :1547-
nary artery
25. Achenbach
high-grade
BachmannK: Detection
W,
of
S,
Moshage
1554, 1998.
restenosis after
PTCA
contrast-en-
using
Diederichs
hanced electron beamCT. Circulation
1997.
et al:
G,
Effects of
colli-
CG,
spiral
DP,
96:2785-2788,
Keating
Glatting
Blurring
of vessels in
mation
CT
angiography:
and
width,
maximum
pitch, viewing
intensity projection.
1996.
plane,
windowing
Comput
26. Achenbach
D:
S,
W,
Noninvasive,
Moshage
Ropers
in
J
Assist
three-dimensional visualization of
coronary artery
20:965-974,
Tomogr
electron
1997.
beam
AmJ
bypass grafts by
Cardiol
tomography.
Dillon
VanLeeuwen
Fernandez
et al: CT
EH,
MS,
MA,
79:856-861,
904
Downloaded from ang.sagepub.com at UNIV MASSACHUSETTS AMHERST on April 1, 2015