AJR 2002; 179:911-916
© American Roentgen Ray Society
Noninvasive Visualization of Coronary Arteries Using Contrast-Enhanced Multidetector CT: Influence of Heart Rate on Image Quality and Stenosis Detection
Tom Giesler1,
Ulrich Baum2,
Dieter Ropers1,
Stefan Ulzheimer3,
Evelyn Wenkel2,
Maria Mennicke1,
Werner Bautz2,
Willi A. Kalender3,
Werner G. Daniel1 and
Stephan Achenbach1
1 Department of Internal Medicine II, Universität Erlangen-Nuernberg,
Ulmenweg 18, D-91054 Erlangen, Germany.
2 Institute of Diagnostic Radiology, Universität Erlangen-Nuernberg,
D-91054 Erlangen, Germany.
3 Institute of Medical Physics, Universität Erlangen-Nuernberg, D-91054
Erlangen, Germany.
Received February 20, 2002;
accepted after revision April 3, 2002.
Supported by a grant from the ELAN Program, University of
Erlangen-Nuernberg, Erlangen, Germany.
Address correspondence to T. Giesler.
Abstract
OBJECTIVE. Although multidetector CT (MDCT) with retrospectively
ECG-gated image reconstruction has been shown to permit noninvasive
visualization of the coronary arteries, the 125-250 msec required for image
acquisition frequently causes motion artifacts. We investigated the influence
of a patient's heart rate on the presence of motion artifacts and on accuracy
of stenosis detection on contrast-enhanced MDCT.
MATERIALS AND METHODS. In 100 patients, MDCT was performed, and
ECG-gated cross-sectional images were retrospectively reconstructed. From the
10 data sets obtained for each patient (reconstructed at 0-90% of the cardiac
cycle in increments of 10%), we chose the best data set for every coronary
artery. The images of the arteries were evaluated for occurrence of artifacts
and the presence of high-grade stenosis (diameter reduction exceeding 70%) or
occlusions. MDCT results were compared with coronary angiographic
findings.
RESULTS. Of the 400 coronary arteries, 115 (29%) could not be
evaluated because of motion artifacts (n = 84) or other reasons
(n = 31). Overall, 51 (49%) of 104 stenoses were revealed on MDCT.
For detecting stenosis in those arteries that we could evaluate, MDCT had a
sensitivity of 91% (51 of 56 stenoses detected) and a specificity of 89%. As
the heart rate increased, the number of arteries that could be evaluated
decreased, and overall sensitivity for stenosis detection decreased from 62%
(heart rate 
70 beats per minute) to 33% (heart rate > 70 beats per
minute).
CONCLUSION. MDCT can reveal coronary stenoses, but the usefulness of
MDCT as an aid in accurately evaluating stenoses decreases as a patient's
heart rate increases.
Introduction
Conventional invasive coronary angiography constitutes the clinical gold
standard for detection of coronary artery stenoses. However, the risk of
potentially serious adverse effects and the associated costs have led to an
intensive search for noninvasive alternatives. It has recently been shown that
multidetector CT (MDCT) in combination with retrospective ECG-gating permits
visualization of the coronary artery lumen
[1,2,3]
and detection of coronary artery stenoses
[4,5,6,7,8,9,10,11,12].
The latest generation of MDCT scanners acquires images in up to four parallel
slices with a gantry rotation time of 500 msec. Through the use of the
simultaneously recorded ECG data, images can be reconstructed at any second of
the cardiac cycle with a data acquisition window of approximately 125-250
msec, depending on a patient's heart rate
[13,14,15].
However, the temporal resolution frequently does not completely eliminate
motion artifacts
[9,10,11,12].
We studied 100 patients to assess the influence of heart rate on the presence
of motion artifacts and on accuracy in detecting coronary artery stenoses.
Materials and Methods
One hundred consecutive patients who had been referred for invasive
coronary angiography were studied (77 men and 23 women; age range, 39-83
years; mean age ± SD, 63 ± 10 years; weight range, 58-123 kg;
mean weight, 81 ± 13 kg). Only patients with sinus rhythm were
included. Patients who were pregnant or who had an unstable clinical
condition, severe renal failure, previous allergic reactions to iodinated
contrast agent, thyroid disease, or any circumstances that would not allow the
patient to lie in the supine position were excluded. Patients who had coronary
stents or who had undergone previous bypass surgery were also excluded. All
patients gave written informed consent, and the study protocol was approved by
the institutional review board. Data from the first 64 patients in our series
were included in a prior analysis of the performance of MDCT in revealing
coronary artery stenoses
[9].
Multidetector CT (MDCT)
MDCT was performed using a Somatom Volume Zoom scanner (Siemens, Forchheim,
Germany) according to a previously described protocol
[1]. Patients were examined in
the supine position, and all image acquisitions were performed during
inspiratory breath-hold. To reduce the heart rate, a ß-blocker (50 mg of
metoprolol) was administered orally 2 hr before the examination in the last 36
patients to provide comparison data between this group and the first 64
patients. After determining the contrast agent transit time
[1], we acquired MDCT data
during an IV injection of 160 mL of the iodinated contrast agent iopromid
(Ultravist 370; Schering, Berlin, Germany) at a rate of 4 mL/sec. The
following scanning protocol was used: collimation, 4 x 1 mm; gantry
rotation time, 500 msec; and table feed, 1.5 mm/rotation. The tube current was
150 mA at 140 kV to keep the radiation dose within a reasonable range. A delay
corresponding to the measured contrast agent transit time was kept between the
initiation of the contrast agent injection and the start of the acquisition.
The breath-hold time was between 30 and 40 sec, depending on the scanning
volume.
For image reconstruction, we processed the raw CT data on a separate
workstation (ImpactIR; VAMP, Möhrendorf, Germany). Using retrospective
ECG-gating and the 180° multislice cardiac interpolation algorithm
[13], we reconstructed
cross-sectional images with a slice thickness of 1.2-1.4 mm in 1-mm intervals
to obtain image acquisition windows of 125-250 msec (full width at one-tenth
maximum of the phase-contribution profile)
[13], depending on a patient's
heart rate [1]. The field of
view was 180 mm with an image matrix of 512 x 512 pixels. For each
patient, we created 10 data sets evenly spaced throughout the cardiac cycle
(0-90% of the R wave—to—R wave interval). We inspected all data
sets and, for each coronary artery, selected the data set that contained the
fewest motion artifacts for further evaluation.
Using an off-line workstation (NetraMD; ScImage, Los Altos, CA), an
independent reviewer who was unaware of the clinical condition or the invasive
coronary angiographic findings of the patients evaluated the MDCT images. The
coronary arteries were classified as "assessable" or
"nonassessable" on the basis of the cross-sectional images,
sliding thin-slab maximum intensity projections
[16], multiplanar
reconstructions, and three-dimensional reconstructions. The images of the
arteries that could be assessed were examined for the presence of high-grade
stenoses (diameter reduction exceeding 70%) and occlusions. Results were
documented separately for the four major epicardial arteries (left main, left
anterior descending, left circumflex, and right coronary artery). Side
branches with a diameter equal to or greater than 2 mm were included in the
analysis of the respective coronary artery (e.g., a stenosis detected in a
diagonal branch would be documented as stenosis of the left anterior
descending artery).
Quantitative Coronary Angiography
Coronary angiograms were obtained in all patients 1 to 3 days after MDCT.
Lesion severity was evaluated using a digital analysis system (Quant-Cor.QCA
version 2.0; Siemens) that was based on the Cardiovascular Angiography
Analysis System II (CAAS II; Pie Medical Imaging, Maastricht, The
Netherlands). The automated edge detection of the system has been validated
and described elsewhere [17,
18]. The tip of the 5-French
or 6-French angiography catheter was used as a calibration device. The luminal
edges were detected using a weighted sum of the first- and second-derivative
functions of the brightness profile perpendicular to the center line of the
vessel, and the vessel diameter was determined by computing the shortest
distance between the right and left contours. The percentage of diameter
stenosis and the interpolated reference diameter were averaged across two
orthogonal projections. Lesions with a mean diameter reduction of 70% or more
were considered high-grade stenoses. In addition, the reference diameter of
every lesion was documented because only lesions in vessel segments with a
lumen diameter of 2 mm or more were included in the analysis.
Results
In all 100 patients, MDCT was successfully completed without complications.
The mean heart rate was 69 ± 13 beats/min (range, 42-115 beats/min).
Sixty of the patients examined had a heart rate of 70 beats/min or less, and
40 patients had a heart rate of more than 70 beats/min. The use of the
ß-blocker in the last 36 patients participating in our study
significantly decreased the heart rate during scanning (72 ± 14
beats/min without ß-blocker compared with 65 ± 10 beats/min with
ß-blocker, p < 0.01).
The optimal position of the imaging window for clear visualization of the
coronary arteries varied considerably from patient to patient
(Table 1). Although the mid to
late diastole cardiac phase frequently yielded the best image quality of left
main, left anterior descending, and left circumflex arteries, the optimal
position to clearly visualize the right coronary artery was more evenly
distributed over late systole, early diastole, and mid to late diastole. The
heart rate strongly influenced the optimal cardiac phase
(Table 2). The best imaging
window in patients with a heart rate of 70 beats/min or less was usually
during mid to late diastole. Because of the shorter duration of diastolic
relaxation, the optimal cardiac phase to visualize the coronary arteries in
patients with a heart rate of more than 70 beats/min was more frequently
during late systole and early diastole.
Of 400 coronary arteries studied, 285 (71%) could be evaluated for the
presence of occlusions or stenoses. In 39 patients (39%), all coronary
arteries were assessable by MDCT. One hundred and fifteen (29%) of 400
coronary arteries were judged to be unassessable because of motion artifacts
(84/115) or other reasons, such as severe calcification or incomplete
breath-holding by the patient (31/115). The right coronary artery was more
frequently affected by degraded image quality (48 patients) than were the
other arteries (left main, 8 patients; left anterior descending, 24 patients;
and left circumflex, 35 patients; p < 0.05). In the 60 patients
with a heart rate of 70 beats/min or less, only 31 (13%) of 240 coronary
arteries were affected by motion artifacts, whereas in the 40 patients with a
heart rate of more than 70 beats/min, 53 (33%) of 160 arteries were degraded
by motion artifacts (p < 0.001) (Figs.
1A,1B,1C,1D,2,3).
View larger version (142K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1A. —Cross-sectional multidetector CT (MDCT) of patients with
suspected coronary artery disease. Cross-sectional MDCT scan of 70-year-old
man shows proximal left anterior descending coronary artery (arrow)
without motion artifacts. Heart rate was 67 beats/min.
|
|
View larger version (182K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1B. —Cross-sectional multidetector CT (MDCT) of patients with
suspected coronary artery disease. Cross-sectional MDCT scan of 67-year-old
man shows mid right coronary artery (arrow) without motion artifacts.
Heart rate was 62 beats/min.
|
|
View larger version (149K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1C. —Cross-sectional multidetector CT (MDCT) of patients with
suspected coronary artery disease. Cross-sectional MDCT scan of 66-year-old
man shows proximal left anterior descending coronary artery. Visualization of
artery is affected by severe motion artifact (arrow). Heart rate was
89 beats/min.
|
|
View larger version (160K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1D. —Cross-sectional multidetector CT (MDCT) of patients with
suspected coronary artery disease. Cross-sectional MDCT scan of 74-year-old
man shows mid right coronary artery, visualization of which is severely
affected by motion artifact (arrow). Heart rate was 78 beats/min.
|
|
View larger version (179K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2. —Multiplanar multidetector reconstruction of 62-year-old woman
with suspected coronary artery disease shows clearly visualized right coronary
artery with no motion artifacts up to the distal segments. Heart rate was 59
beats/min.
|
|
View larger version (175K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3. —Multiplanar multidetector reconstruction of 51-year-old man
with suspected coronary artery disease shows right coronary artery. Motion
artifacts are visible (arrows). Heart rate was 72 beats/min.
|
|
MDCT revealed 51 (49%) of 104 high-grade stenoses (diameter reduction 
70%) and occlusions in the coronary arteries (including their side branches
with a diameter 2 mm). The overall sensitivity for stenosis detection in
assessable and nonassessable arteries decreased from 62% (36/58) in patients
with a heart rate of 70 beats/min or less to 33% (15/46) in patients with a
heart rate of more than 70 beats/min (p < 0.01)
(Table 3).
In the 285 assessable coronary arteries, all 11 occlusions and 40 of 45
high-grade stenoses were correctly identified on MDCT (Figs.
4A,4B
and
5A,5B).
The only lesion present in the left main artery, 30 of the 33 lesions present
in the left anterior descending artery (including diagonal branches with
diameters 2 mm), seven of nine lesions present in the left circumflex
artery (including marginal branches with diameters of 2 mm), and all 13
lesions present in the right coronary artery were correctly detected. Four of
five false-negative findings of stenosis were located in diagonal or marginal
branches. The mean reference diameter of these vessel segments was 2.5
± 0.5 mm. One false-negative finding of stenosis was in the left
circumflex artery, and, retrospectively, this error was found to be caused by
calcifications.
View larger version (96K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4A. —56-year-old man with coronary artery disease. Heart rate was
65 beats/min. Multiplanar multidetector (MDCT) reconstruction shows right
coronary artery (RCA) and left main artery and left anterior descending
coronary artery (LAD) with proximal high-grade stenosis (arrow).
|
|
View larger version (147K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4B. —56-year-old man with coronary artery disease. Heart rate was
65 beats/min. Invasive coronary angiogram confirms MDCT finding and shows
high-grade stenosis of the proximal left anterior descending coronary artery
(arrows).
|
|
View larger version (135K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 5B. —48-year-old man with coronary artery disease. Heart rate was
61 beats/min. Invasive coronary angiogram confirms MDCT finding and shows
severe stenosis of right coronary artery (arrow).
|
|
A total of 223 of the 229 assessable coronary arteries with no occlusion or
stenosis were correctly judged to be free of obstruction. In 26 vessels, a
high-grade lesion was incorrectly judged to be present on MDCT. The mean
(± SD) diameter reduction of these false-positive lesions was 31%
(± 22%). In seven vessels, a stenosis of intermediate degree (diameter
reduction of between 50% and 70%) was present. If only assessable coronary
arteries were considered, MDCT had a sensitivity of 91%, specificity of 89%,
positive predictive value of 66%, and negative predictive value of 98% for
detection of high-grade stenoses and occlusions.
Discussion
Our study expands on findings of prior reports
[9,
10,
12] that indicated ECG-gated
MDCT was highly accurate for the detection of coronary artery stenoses if the
image quality was sufficient. In our large group (100 patients), we found a
sensitivity of 91% and specificity of 89% in assessable vessel segments.
However, only 71% of the coronary arteries could be evaluated, and in only 39%
of the patients were all coronary arteries judged to be assessable. The high
number of unassessable coronary arteries was mostly due to motion artifacts.
As expected, the right coronary artery was most severely affected by motion on
MDCT, because this vessel displays the fastest velocity during the cardiac
cycle
[19,20,21,22].
We observed the strong influence that a patient's heart rate during
scanning had on image quality and accuracy for stenosis identification on
MDCT. Even though we selected the optimal reconstruction window for every
individual coronary artery, motion artifacts could not be prevented in imaging
of 84 (21%) of the 400 coronary arteries, and motion artifacts were
significantly more frequent in patients with heart rates of more than 70
beats/min than in patients with lower heart rates ( 70 beats/min).
Consequently, overall stenosis detection was more accurate in patients with
low heart rates. A higher heart rate leads to a shorter R
wave—to—R wave interval because of the shorter end diastolic
portion of the cardiac cycle. This portion of the cardiac cycle is crucial to
image acquisition at current temporal resolution of 250 msec.
In an earlier study, the best image quality was found in patients with
heart rates of less than 74.5 beats/min
[11]. The researchers used a
5-point scale to assess the image quality in each coronary segment. However,
their study was limited to the analysis of image quality and did not
investigate the detection of coronary artery stenoses.
Optimal positioning of the image reconstruction window seems to be crucial
to achieving optimal image quality. It is known that each of the coronary
arteries has a different motion pattern during the cardiac cycle
[19,
20,
22]. Our findings confirm the
results of a recently published study
[15]. Images of the right
coronary artery were most frequently of sufficient quality earlier in the
cardiac cycle than were those of the left anterior descending and left
circumflex arteries. In addition, a patient's heart rate during MDCT scanning
had a strong influence on which cardiac phase provided the clearest
visualization of the coronary arteries. Although in patients with lower heart
rates ( 70 beats/min) the optimal window position was usually found during
mid to late diastole, the best image quality in patients with a heart rate of
more than 70 beats/min was found during late systole and early diastole. The
large variation in the optimal position of the image reconstruction window and
the differences between the coronary arteries suggest that careful positioning
and selection of the image reconstruction window are crucial factors to
obtaining optimal image quality and diagnostic accuracy on MDCT.
In contrast to coronary angiography performed with electron-beam CT, MDCT
does not require a longer scanning duration for patients with lower heart
rates. Therefore, a useful approach might be to limit the use of MDCT for
coronary artery visualization to patients with lower heart rates or to use
pharmacologic interventions (e.g., ß-blockers) during the scanning to
enhance image quality and accuracy in the identification of coronary stenoses.
In addition, technical developments, such as a decreased gantry rotation time
and the simultaneous acquisition of more than four parallel slices may further
improve image quality and diagnostic accuracy.
We conclude that contrast-enhanced MDCT with retrospectively ECG-gated
image reconstruction is a promising tool in detecting coronary artery
stenoses. However, many coronary arteries are currently nonassessable because
of motion artifacts, and the accuracy of stenosis detection decreases as a
patient's heart rate increases.
References
- Achenbach S, Ulzheimer S, Baum U, et al. Noninvasive coronary
angiography by retrospectively ECG-gated multislice spiral CT.
Circulation
2000;102:2823
-2828[Abstract/Free Full Text]
- Kopp AF, Ohnesorge B, Flohr T, et al. Cardiac multidetector-row CT:
first clinical results of retrospectively ECG-gated spiral with optimized
temporal and spatial resolution [in German]. Rofo Fortschr Geb
Rontgenstr Neuen Bildgeb Verfahr
2000;172:429
-435[Medline]
- Ohnesorge B, Flohr T, Becker C. Cardiac imaging by means of
electrocardiographically gated multisection spiral CT: initial experience.
Radiology
2000;217:564
-571[Abstract/Free Full Text]
- Mochizuki T, Murase K, Koyama Y, Higashino H, Ikezoe J. Images in
cardiovascular medicine: LAD stenosis detected by subsecond spiral CT.
Circulation
1999;99:1523[Free Full Text]
- Knez A, Becker C, Ohnesorge B, Haberl R, Reiser M, Steinbeck G.
Noninvasive detection of coronary artery stenosis by multislice helical
computed tomography. Circulation
2000;101:221
-222[Free Full Text]
- Becker CR, Knez A, Ohnesorge B, Schoepf UJ, Reiser MF. Imaging of
noncalcified coronary plaques using helical CT with retrospective ECG gating.
AJR
2000;175:423
-424[Free Full Text]
- Knez A, Leber A, Becker CR, Becker A, Reiser MF, Haberl R.
Multirow-detector helical computed tomography for noninvasive angiography: a
new imaging modality for determination of severe coronary artery disease.
(abstr) Circulation
2000;102[suppl II]:411[Abstract/Free Full Text]
- Kopp A, Schroeder S, Kuettner A, Ohnesorge B, Georg C, Claussen CD.
Multidetector-row CT for noninvasive coronary angiography: results in 102
patients. (abstr) Radiology
2000;217(P):375
- Achenbach S, Giesler T, Ropers D, et al. Detection of coronary
artery stenoses by contrast-enhanced, retrospectively
electrocardiographically-gated, multislice spiral computed tomography.
Circulation
2001;103:2535
-2538[Abstract/Free Full Text]
- Nieman K, Oudkerk M, Rensing BJ, van Ooijen P, van Geuns RJ, de
Feyter PJ. Coronary angiography with multislice computed tomography.
Lancet
2001;357:599
-603[Medline]
- Hong C, Becker CR, Huber A, et al. ECG-gated reconstructed
multi-detector row CT coronary angiography: effect of varying trigger delay on
image quality. Radiology
2001;220:712
-717[Abstract/Free Full Text]
- Knez A, Becker CR, Leber A, et al. Usefulness of multislice spiral
computed tomography angiography for determination of coronary artery stenoses.
Am J Cardiol
2001;88:1191
-1194[Medline]
- Kachelriess M, Ulzheimer S, Kalender WA. ECG-correlated image
reconstruction from sub-second multi-slice spiral CT scans of the heart.
Med Phys
2000;27:1881
-1902[Medline]
- Ohnesorge B, Flohr T, Becker C, et al. Cardiac imaging with rapid,
retrospectively ECG synchronized multilevel spiral CT [in German].
Radiologe
2000;40:111
-117[Medline]
- Kopp AF, Schroeder S, Kuettner A, et al. Coronary arteries:
retrospectively ECG-gated multi-detector row CT angiography with selective
optimization of the image reconstruction window.
Radiology
2001;221:683
-688[Abstract/Free Full Text]
- Reddy G, Chernoff DM, Adams JR, et al. Coronary artery stenoses:
assessment with contrast-enhanced electron-beam CT and axial reconstructions.
Radiology
1998;208:167
-172[Abstract/Free Full Text]
- Reiber JHC, Serruys PW, Kooyman CJ, et al. Assessment of short-,
medium-, and long-term variations in arterial dimensions from
computer-assisted quantification of coronary cine-angiograms.
Circulation
1985;71:280
-288[Abstract/Free Full Text]
- Haase J, Di Mario C, Slager CJ, et al. In vivo validation of
on-line and off-line geometric coronary measurements using insertion of
stenosis phantom in porcine coronary arteries. Cathet Cardiovasc
Diagn 1992;27:16
-27[Medline]
- Achenbach S, Ropers D, Holle J, Muschiol G, Daniel WG, Moshage W.
In-plane coronary arterial motion velocity: measurement with electron-beam CT.
Radiology
2000;216:457
-463[Abstract/Free Full Text]
- Paulin S. Coronary angiography: a technical, anatomic and clinical
study. Acta Radiol
1964;233[suppl]:1
-215
- Hofman MB, Wickline SA, Lorenz CH. Quantification of in-plane
motion of the coronary arteries during the cardiac cycle: implications for
acquisition window duration for MR flow quantification. J Magn
Reson Imaging 1998;8:568
-576[Medline]
- Wang Y, Vidan E, Bergman GW. Cardiac motion of coronary arteries:
variability in the rest period and implications for coronary MR angiography.
Radiology
1999;213:751
-758[Abstract/Free Full Text]
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
Top
Abstract
Introduction
